UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Observations of lacustrine sedimentation at Lillooet Lake, British Columbia Gilbert, Robert 1973

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

Item Metadata

Download

Media
831-UBC_1974_A1 G54.pdf [ 25.31MB ]
Metadata
JSON: 831-1.0093262.json
JSON-LD: 831-1.0093262-ld.json
RDF/XML (Pretty): 831-1.0093262-rdf.xml
RDF/JSON: 831-1.0093262-rdf.json
Turtle: 831-1.0093262-turtle.txt
N-Triples: 831-1.0093262-rdf-ntriples.txt
Original Record: 831-1.0093262-source.json
Full Text
831-1.0093262-fulltext.txt
Citation
831-1.0093262.ris

Full Text

\ OBSERVATIONS  OF  LILLOOET  LACUSTRINE LAKE,  BR I T I S H  SEDIMENTATION C O L U M B IA  by ROBERT  GILBERT  M . A . , University of British Columbia, 1970  A  THESIS THE  SUBMITTED  IN  REQUIREMENTS DOCTOR  PARTIAL FOR  OF  THE  FULFILMENT DEGREE  OF  PHILOSOPHY  in the Department of Geography  We accept this thesis as conforming to the required standard  THE  UNIVERSITY  OF  BRITISH  December, 1973  COLUMBIA  OF  In  presenting  this  an advanced  degree  the  shall  I  Library  further  for  scholarly  by h i s of  agree  this  written  thesis  in partial  fulfilment  of  at  University  of  Columbia,  the  make  that  it  purposes  for  available  permission for  It  financial  is  gain  Department  Date  J^L  J?/  /?73  the  Columbia  shall  not  requirements  reference copying of  Head o f  understood that  permission.  The U n i v e r s i t y o f B r i t i s h Vancouver 8, Canada  for  extensive  may b e g r a n t e d b y  representatives. thesis  freely  British  the  be  I  agree  and  that  study.  this  thesis  my D e p a r t m e n t  copying or  for  or  publication  allowed without  my  III.  Abstract 2 Sediment from Lillooet River basin, 3000 km of alpine and subalpine landscape about 1 0 % of which is g l a c i e r i z e d , is largely deposited in Lillooet Lake. Dissolved sediment concentration in the streams flowing into Lillooet Lake is generally less than 100 mg I ^ . to over 3000 mg I \  Suspended sediment concentrations vary from 50 mg I ^  related closely to water discharge.  Coarse grained sediment  which moves as bed load is deposited as foreset beds at the delta l i p . sediments are distributed through the lake  (a) by dispersion and interflow above the  thermocline at low to moderate inflow rates and higher inflows.  The thermocline •  Suspended  effectively  (b) by interflow and underflow at limits  vertical sediment movement  during much of the melt season though during high flows the thermal structure of the lake may be temporarily destroyed by underflow and powerful interflow. The delta surface consists of rapidly shifting channels and sand bars.  Bed  load movement is sporadic and changing lake levels lead alternately to accumulation in backwater when the lake level rises or flushout as the lake level falls.  Coarse  sediment from bed load, deposited on the proximal, slope is redistributed by infrequent major slump events. at 120 m depth.  Mounds of slumped material cover the foreset slope to its base  Interflow and underflow give rise to a distinctive sedimentary sequence  of varved silts and clays interspaced with turbidites.  Thickness of the varves is related  to characteristics of the inflowing water and fine grained sediment.  Continuous records  of water temperature near the lake bottom indicate that frequent laminae within varves are due to intermittent underflow of suspended fluvial sediment.  Autumn floods result  in the deposition from suspension in the lake water of an anomalously thick, fine grained layer during the following winter.  Infrequent much smaller winter inflow peaks  result in underflow and the deposition of a lighter toned, coarser grained lamina in the winter deposit.  Both are useful stratigraphic markers.  Continuous seismic profile records indicate that as much as 200 m of sediment has been deposited in the lake.  Calculation based on present rates of accumulation of  iv. 20 - 50 mm yr  , depending on distance from the delta, indicates that reflecting  horizons in the sediment represent increased sediment input associated with volcanic activity 2400 years B.P.  a n d , less reliably, to Pleistocene g l a c i a t i o n .  Annual  long term (1913-1969) clastic sediment yield at Lillooet Lake is calculated as 1.1 x 6 3 —1 -1 10 m yr (specific denudation 0.28 mm yr ).  The advance rate of Lillooet  Delta indicates that y i e l d may have trippled after 1950 associated with river engineering, logging and agriculture.  V.  Preface Lillooet River V a l l e y in the southern Coast Mountains of British Columbia has become the site of a number of preliminary geomorphic and hydrologic studies. The region ranges from fertile valley bottom flood plain to high altitude icefields and large glaciers, and thus is suited to a wide variety of integrated investigations with varying emphases and approaches. The study of sedimentation in Lillooet Lake reported here was originally designed as a study of the physical processes active in a relatively large, high energy lake and of their effect on the sediment environment.  As work began however, it  became apparent that the mechanics of these processes could not be considered satisfactorily because: 1)  nothing was known of the sedimentary environment of this lake..  a few exceptions ,  With  no sufficiently similar lakes, which could be used as a basis for  experimental design, have been examined. 2)  the logistics and techniques of precise measurement of such transitory  phenomena as lake currents, particularly underflow, and the measurements of the mechanics of associated sediment movement were beyond the capacity of the writer and of the resources a v a i l a b l e . 3)  almost nothing was known of the sources of sediment or of the patterns  of sediment inflow at Lillooet Lake. 4)  as a corollary to 3), the processes, and particularly the interaction  between them, in a lake as large as Lillooet are highly complex, changing rapidly throughout the lake and through time. This then is a report of thesedimentary environment of Lillooet Lake, with particular reference to the morphology of the large delta at its head and to the patterns of sediment inflow and distribution in the lake.  Despite this reduction in  the initial objective, particularly with respect to the statement, development and dissertation of a thesis, this is the most complete study of current lacustrine sedimentation  vi. in a high energy mountain lake in North A m e r i c a .  As such it serves several functions  in addition to the obvious one of the description of a hitherto unknown environment: 1)  It provides a preliminary statement of the geomorphic environment of the  Lillooet River V a l l e y and of the patterns of sediment yield from i t .  From this it is hoped  that studies of the physical principles of landscape evolution in the region can b e g i n . 2)  It provides a comparison with other proglacial and mountain lakes and  confirms the dominance of certain processes in their sedimentology. 3)  It indicates the need for, and the direction of, research into the mechanisms  of individual processes.  It is from this research that genuine progress in the under-  standing of these complex environments w i l l be made.  Several points relating to the preparation of this work deserve comment: 1)  Discussion of the physics of lacustrine processes was avoided either in  review of other works since the author has reported on these elsewhere (Church and G i l b e r t , in press), or in the presentation of this thesis since insufficient information was gathered. 2)  A convention in geographic nomenclature was adopted for ease of  explanation, although the names are not officially recognized.  'North Lillooet Lake  refers to the portion of Lillooet Lake between the delta of Lillooet River and a bend and narrows in the lake 10.2 kilometres to the southeast. study was concentrated.  It was in this area that the  Lillooet Lake continues 12.5 kilometres south ('South  Lillooet Lake ) before draining into Tenasse Lake via a dredged channel and then into 1  Lillooet River and Harrison Lake.  'Bear Point , a name that appears on Palmer's map 1  (1859), divides 'North' and 'South' Lillooet Lake (figure 3.4). 3)  Frequent citations are made of a number of works in press.  With one  exception, that of Banerjee, these papers will form a Special Publication of the Society of Economic Paleontologists and Mineralogists on glacio-fluvial and g l a c i o -  1  vii.  lacustrine sedimentology now (December, 1973) being edited by B . C . and A . V . 4)  MacDonald  Jopling. Textural analysis, particularly of fine sediment, is used extensively in the  interpretation of the lake deposits. deposits was made.  However no examination of the mineralogy of these  Therefore, particles smaller than 2JJ  diameter are referred to as  clay or clay size without knowledge of whether they are clay minerals or fine fragments of other minerals.  The success of graduate research is usually due to many people besides the principal investigator, and in this case particularly so. Professors J . R .  M a c k a y , W . H . Mathews, J . W .  The author's committee,  Murray and S . O . Russell, have  carefully and critically examined the work and its presentation. appreciated by adoption.  Their comments are  In addition, Professor Mackay kindly allowed the use of  many items of his equipment, some of which proved essential to the work.  Other  material was supplied by the Water Survey of C a n a d a , and the Department of Geography, University of British Columbia. The author's research supervisor, Dr. H . O .  Slaymaker,  was very generous in his financial support through grants from the National Research Council of C a n a d a , and in his encouragement at all stages of the work.  He was very  helpful in the final preparation of the thesis. The people who worked in the field with the author are particularly appreciated. They range from fellow graduate students including John Ponton, Peter Lewis, and M i c h a e l Patterson to assistants David Whiting (1971) and David Maclean (1972) who often bore the brunt of daily field routine. F i n a l l y , sincere appreciation is extended to M r . Michael Church who has aided in every aspect of the project from substantial financial and material support, to field assistance, to the many long discussions which have guided the work and its author.  viii. Contents  Abstract  Mi.  Preface  v.  Contents  vi i i .  List of tables  xi.  List of figures  xii.  1.  Introduction 1.1.  1.2. 2.  1  Present understanding of sedimentation in proglacial and mountain lakes  1  The Lillooet Lake study  8  Description of water and sediment input to Lillooet Lake  11  2.1.  Introduction  11  2.2.  The role of climatic variables in determining inflow to Lillooet Lake  11  2.3.  Water inflow to Lillooet Lake  19  2.4.  Factors controlling sediment input to Lillooet Lake  32  2.4.1.  Basin morphometry  32  2.4.2.  Bedrock geology  33  2.4.3.  Processes governing sediment production  36  2.4.4.  Engineering works and land use changes affecting sediment production  2.5.  2.6.  39  Sediment supply at the delta  42  2.5.1 .  Dissolved sediment supply  42  2.5.2.  Suspended sediment discharge  45  2.5.3.  Bed load discharge  50  2.5.4.  Sediment inflow from other streams  57  Summary  58  ix. 3.  4.  5.  6.  Physiography of Lillooet Lake  61  3.1.  Subaerial morphometry of Lillooet Delta  61  3.2.  Lake floor physiography  68  3.3.  Summary  Process of sediment distribution in Lillooet Lake  78  4.1.  Suspended sediment concentration and density of lake water  83  4.2.  River water and the response of temperature and turbidity characteristics of Lillooet Lake water  85  4.2.1.  Formation and breakup of thermal structure in the lake  85  4.2.2.  Lake water conditions during the melt season  89  4.3.  Underflow  4.4.  Summary and discussion  96 104  Sediment of Lillooet Lake  107  5.1.  Analysis of texture of lake bottom sediments  107  5.2.  Strength of Lillooet Lake sediments and implication for slumping  117  5.3.  Mass properties of Lillooet Lake sediments  120  5.4.  Stratigraphy of Lillooet Lake sediments  125  5.4.1.  Varve thickness and relation to inflow  125  5.4.2.  Sedirrientological events not related to normal varve sedimentation  135  5.4.3.  Intravarve sedimentation  142  5.5.  Results of continuous seismic profiling in Lillooet Lake  145  5.6.  Rate of sediment accumulation in Lillooet Lake and yield from Lillooet River Basin  151  5.7  Summary  153  6.1.  Conclusions  156  6.2.  Further work  157  References  159  Appendices  168  2.1  Discharge records Green River at Pemberton  169  2.2  Suspended and dissolved sediment collection and analysis technique  178  2.3  Moment measures, Lillooet River bed material  180  5.1  Sampling devices used in the Lillooet Lake sediment study  181  5.2  Assessment of errors involved in sampling lake sediment  183  5.3  Moment measures, Lillooet Lake bottom sediments  186  5.4  Lillooet Lake sediments: mass properties and shear strength  187  XI .  List of tables  1.1  Observations of density and turbidity currents from selected lakes  2.1  Snow storage at Tenquille Lake and M c G i l l i v a r y Pass snow courses,  4, 5  1965  19  2.2  Summary of inflow from area tributary to North Lillooet Lake  21  2.3  Timing of daily maxima of water levels and air temperature at sites  2.4 2.5  in basins tributary to Lillooet Lake, July 20 to August 6, 1971  32  Summary of morphometric data:  32  Lillooet River V a l l e y  Summary of the concentration of dissolved material in several warm and hot springs in Lillooet River V a l l e y  2.6  36  Effect of dispersant on grain size determination - Lillooet Lake sediments  46  3.1  Rates of advance of Lillooet Delta  66  3.2  Summary of the subaqueous morphometry of Lillooet Lake  77  5.1  Recovery parameters of Lillooet Lake cores  125  5.2  Comparison of yarve thickness in Ekman and gravity core samples  128  A.2.1  Six hour mean discharge, Green River near Pemberton, 1971  171  A.2.2  Three hour mean discharge, Green River near Pemberton, 1972  175  A.2.3  Summary of moment measures,  180  A.5.1  Summary of moment measures, Lillooet Lake bottom sediments, 1970  186  A.5.2  Material density of Lillooet Lake sediments  189  A.5.3  Mass properties of Lillooet Lake sediments  190  A.5.4  Strength - Lillooet Lake sediments  192  Lillooet River bed material  XII .  List of figures  Frontispiece  Lillooet D e l t a , December 1968.  1.1  A conceptual model of sedimentation in high energy lakes.  2.1  Location of data gathering stations in Lillooet River area referred to in text.  2.2  ii. 9  12  Mean daily temperature and precipitation, Pemberton Meadows and A l t a Lake.  13  2.3  Mean daily temperature at Richardson's.  14  2.4  Maximum, mean, and minimum daily temperatures for stations in  2.5 2.6  Lillooet River watershed during the melt season 1965.  16  Examples of the effect of heavy fall rains at A l t a Lake and Pemberton Meadows on the discharge characteristics of Lillooet River.  17  Daily precipitation duration series A l t a Lake 1936-1965 and Pemberton Meadows 1931-1960.  2.7  18  Exceedance probabilities for approximately month end snow surveys at Tenquille Lake and M c G i l l i v r a y Pass.  20  2.8  Accumulation in metres of water equivalent on the surface of Place G l a c i e r , 1964-1965 from 0strem (1966, p. 104).  20  2.9  Drainage basins tributary to North Lillooet Lake .  22  2.10  Mean of mean daily discharge for Lillooet, G r e e n , and Birkenhead Rivers.  2.11 2.12  24  Coefficient of variation of mean daily discharge for Lillooet, G r e e n , and Birkenhead Rivers.  25  Mean daily discharge, Lillooet River and mean daily temperature and total daily precipitation Pemberton Meadows: a) 1940  26  b)  27  1948  2.13  Pearsonian Type III plot of recurrence series of mean daily discharges Lillooet River .  29  2.14  Duration series of Lillooet, G r e e n , and Birkenhead Rivers.  30  2.15  Time of maximum daily flow, lake water level or air temperature at sites in basins tributary to Lillooet Lake.  31  2.16  Hypsometry: Lillooet, Green and Birkenhead River basins.  34  2.17  Bedrock geology of basins tributary to North Lillooet Lake.  35  xiii. 2.18  Processes of sediment production and conveyance and resulting landforms in high energy alpine environments: a conceptual model.  2.19  Lillooet River valley showing the reclamation and river training carried out between 1946 and 1951.  2.20  37  40  Lillooet valley looking north northwest from the air over M c K e n z i e Cut.  41  2.21  The outlet of Lillooet Lake to Tenasse Lake.  41  2.22  Concentrations of dissolved sediments from Lillooet-Green River at Lillooet Delta,1971-1972.  43  2.23  Suspended sediment-discharge relation for Lillooet-Green River at Lillooet D e l t a , 1971-1972.  47  2.24  Relation between measured optical properties and suspended sediment concentrations in Lillooet-Green River at Lillooet Delta.  49  2.25  Concentrations of suspended sediment in Lillooet-Green River at Lillooet Delta during parts of the melt seasons 1971 and 1972.  2.26  51  Water density in the main distributary of Lillooet-Green River at Lillooet Delta during parts of the melt seasons 1971 and 1972 .  52  2.27  Mean daily  54  2.28  Texture of bed material of Lillooet-Green River, 1971-1972.  2.29  Spot observations of transmissivity and Secchi depth expressed as concentrations of suspended sediment at the mouths of Ore and  water level of Lillooet Lake, 1971-1973.  Birkenhead Creeks, 1972.  56  57  3.1  Selected vertical air photographs of Lillooet Delta  3.2 a  Low oblique photograph of Lillooet Delta , November 2 6 , 1969.  65  b  Bank erosion on a secondary distributory, September 23, 1969.  65  c  Sand storm in 15 m s ^ southwest w i n d , May 11, 1973.  65  d  Oblique photograph of Lillooet D e l t a , August 2, 1971.  65  3.3  Advance of Lillooet Delta between 1858 and 1969.  67  3.4  Bathymetry of North Lillooet Lake from corrected echograms.  69  3.5  Photograph looking northwest at the lip of Lillooet Delta across the mouth of the largest distributary during low water and sediment inflow on March 20, 1971 .  70  3.6  Echograms of Lillooet Lake  6 2 , 63  72-74  XIV.  3.7  M a p showing location of echo sounding profiles of figure 3 . 6 . , water level recorder, and permanent floating rafts.  75  4.1  Records during 1971 at raft 1 of a) temperature and b) transmissivity.  79  4.2  Records during 1972 at raft 1 of a) temperature and b) transmissivity.  80  4.3  Records during 1972 at raft 2 of a) temperature and b) transmissivity.  81, 82  4.4  Relation between measured optical properties and suspended sediment concentrations at the surface of Lillooet Lake.  4.5  Bathythermographs at raft 1 showing formation in spring and breakup in  84  fall of the thermal structure of Lillooet Lake.  87  Transmissivity at raft 1 during the winter 1971-1972 .  88  4.7  Bathythermographs in Lillooet Lake, 1970.  90  4.8  Relation between transmissivity at 15 m depth at rafts 1 and 2 in Lillooet Lake and mean daily concentration of suspended sediment in Lillooet-Green Rivers, 1972 .  92  Bathythermographs showing the destruction and reforming of the thermal structure of Lillooet Lake associated with three periods of strong underflow and interflow.  93  Settling times for particles of density 2.73 g ml ^ and equivalent fall diameters of 1 to 20 microns in still water with temperatures as recorded in Lillooet Lake.  95  4.6  4.9  4.10  4.11 a) Pattern of surface water movement July 2 3 , 1971, south end of North Lillooet Lake, b) transmissivity profiles at raft 1 and three sites shown in a). 4.12  97  Mean hourly water temperature approximately 1 - 2 m from the lake bottom at raft 1, and spot readings of temperature in Lillooet-Green River at the delta, July 11 to September 14, 1971 .  99  4.13  Traces of lake bottom temperature records on selected days in 1971 .  101  4.14  Evidence of a slump on the foreset slopes of Lillooet D e l t a , August 5, 1971: a) lake water level record showing large wave at 8:10 P . S . T . , b) tracing of lake bottom temperature record at raft 1 showing warm water pulse between 8:30 and 8:45 P . S . T .  103  5.1  Location of Ekman grab samples in Lillooet Lake for preliminary textural analysis.  108  5.2  Cumulative frequency distributions of Lillooet Lake surficial sediments.  109  5.3  Distribution of moment measures and per cent sand in Lillooet Lake sediments.  110-114  XV.  5.4  Scattergrams of moment measures of Lillooet Lake bottom sediments.  116  5.5  Shear strength and sensitivity of Lillooet Lake sediments.  118  5.6  Relation between depth of burial and undisturbed and remolded shear strength of pooled samples from Lillooet Lake.  119  5.7  Map showing bottom slopes in excess of those calculated to be stable.  121  5.8  Mass properties of Lillooet Lake sediments.  122  5.9  M a p showing location of samples for stratigraphic examination.  126  5.10  Thicknesses of varves from Lillooet Lake plotted against the year in which deposition occurred.  5.11  -  Mean varve thickness Lillooet Lake sediments.  127 129  2 5.12  Coefficient of determination, r , between varve thickness and Lillooet River inflow.  5.13  131  Relation between varve thickness and annual inflow in excess of specified mean daily flow for selected cores from Lillooet Lake.  132  5.14  Mean daily flow in Lillooet River for selected years.  134  5.15  Selected sections from cores from Lillooet Lake.  5.16  Thickness of sand layer deposited in Lillooet Lake in 1945.  139  5.17  Cumulative frequency distributions from the 1945 layer in core 75.  149  5.18  Ekman samples from Li Ilooet Lake.  143  5.19  Textural analysis of samples from the 1971 varve of sample 72.  146  5.20  Continuous seismic profiling records from Lillooet Lake.  5.21  Map showing location of C . S. P. records shown in figure 5 . 1 9 .  A.2.1  Relation  3 between mean monthly discharge (m  136, 137  147, 148 150  -1 s  ) of Lillooet  and Green Rivers , 1923-1951 .  170  A.2.2  Rating curve, Green River near Pemberton, 1971-1972.  170  A.5.2  Q u i c k direct shear apparatus.  188  Frontispiece: L i l l o o e t Delta, December, 1968. British C o l u m b i a Forest Service Photo 17140  1.  Introduction The processes of sediment distribution at the mouths of relatively high energy,  glacial sediment laden streams as found throughout the Coast Mountains of British Columbia remain relatively uninvestigated.  While literature on delta building and  deltaic sequence is extensive (see review works b y , for example, Welder, 1959; van Stratten; 1964; Shirley, 1966; U N E S C O , 1966; Russell, 1967; Coleman, 1969), emphasis has been placed on the marine deltas of the world's major rivers with different sediment and water input and different factors influencing sediment distribution.  This thesis reviews briefly the present understanding of high energy lacustrine  sedimentation and describes the results of observations at Lillooet Lake, British Columbia.  1.1.  Present understanding of sedimentation in proglacial and mountain lakes A number of studies of high energy deltaic environments in the Canadian  Cordillera have been reported including work at Lake Louise by Johnston (1922 b), at Lake C a v e l l by Kindle (1930), in Portland Canal (Bear River Delta) by Hanson (1934), at Garibaldi Lake by Mathews (1956), at Sunwapta Lake (Athabasca Glacier) by Mathews (1964), in Arrow Lake by Fulton and Pullen (1969), and more detailed studies of Fraser River Delta beginning with Johnston's work in 1922 (a) through Mathews and Shepard (1962) to recent studies (Mayers, 1969; Kellerhals and Murray, 1969; Tiffin e t a l . , 1971).  Exposed Pleistocene lacustrine sediments have received  some attention (Fulton, 1965; Shaw, in press) and one recently drained ice-dammed lake has been briefly studied (Hanson, 1932). Much is now understood about fluvial sediment distribution in reservoirs stemming from the American Southwest reservoir projects (Robinson, 1920; F i o c k , 1934), receiving most intensive study at Lake Mead (Grover and Howard, 1938; G o u l d , 1951; Smith et a L ,  1960).  This work firmly established the role of turbidity currents in  1  2 sediment distribution.  But factors affecting these processes vary greatly from the  British Columbia situation: 1)  Dissolved load of the Colorado River is very high particularly in the low  flow of autumn and winter when turbidity current flow in Lake Mead is most common, but is much lower in British Columbia streams. 2)  Deposition of flocculated sediment particles is highly significant in Lake  Mead but is much less important in fresh water lakes in British C o l u m b i a . 3)  Unlike local streams, the sediment load during spring melt is relatively  low, and flash floods on the semi-arid landscapes in late summer contribute the major sediment supply. A number of studies in environments closely related to the Coast Mountains are available but each is unique in setting or in methodology.  For example the work of  Axelsson (1967) in Sweden illustrates a delta with input typical of the mountain environment, but the emphasis of the study was on the subaerial sediment distribution and the lake depths were very much less than commonly encountered in British Columbia (cf. for example Arnborg, 1948; Dahlskog, 1966; Dahlskog, et a l . , 1972).  0strem  (in press) reports on the formation of varves in a Norwegian proglacial lake, the deglaciation of which has been carefully recorded, and shows that the number of varves is closely related to the number of years since deglaciation.  Gustavson's report on ice  contact Malaspina Lake (in press) demonstrates the role of highly turbid underflow in the formation of varved sediments. In most studies of high energy lacustrine environments/ both present and ancient, the very significant role of turbidity currents has been stressed. As early as 1885, Forel recognized that sediment charged river water may plunge beneath clearer lake water and flow down the lake bottom as a clearly defined flow.  Since then  3  the universal significance of turbidity currents^ has been recognized in both marine and lacustrine environments and the theory of stratified flow in fluids developed in fields as diverse as geology and meteorology and applied to turbidity currents.  Table  1 .1 summarizes some of the reported occurrences of density currents in fresh water lakes and reservoirs. In summaries of the mechanics of stratified flow (see especially Harleman, 1961) two points are generally stressed: 1)  Stratified underflow or interflow in a two-layer system is directly  analogous to open channel flow and thus the same laws govern its behaviour, except that in open channel flow, the density of the overlying fluid (air) is so small compared with the density of the stream water that it is usually ignored (Middleton, 1966). Also small is the frictional shear at the water-air interface when compared with bed and bank shear stresses.  In stratified flow both the density of the overlying fluid and  interfacial shear must be considered (see discussion in Church and G i l b e r t , in press). 2)  A point at which the analogy between stratified and open channel flow  breaks down is that of interfacial mixing between the fluids.  A discussion based on  the work of Keulegan (1949) and others is found in Church and Gilbert (in press). Significantly, most authors have noted that turbidity currents generally maintain themselves for long distances without appreciable mixing.  Density currents are often  weaker (the density difference with the surrounding fluid is small so the driving force of gravity (g ) is small: 1  g =g ( 1  -  j)  where p  is the density of the underflow  ( ) or the lake water (|) , so mixing may be more significant  (cf. Bell, 1947).  ^ In accordance with definitions usually accepted (Howard, 1953, p. 356, 357; Middleton, 1965, p. 248; 1966), the underflow events recorded in Lillooet Lake that form the focus of this work are turbidity currents. "True density currents are solutions or mixtures of gasses or liquids which differ in density from a main body of enclosing fluid by virtue of a difference in temperature or concentration of solute. Other density currents differ in density from an enclosing fluid because they contain suspended matter. . . . These currents constitute a special type of density current distinguished as 'turbidity' currents or 'suspension' currents" (Menard and Ludwick, 1951, pp. 2-3).  4  Table  1.1  Observations of density and turbidity currents from selected lakes Lake Lake Arthur, South A f r i c a  Author Bell (1942)  Notes V e l o c i t y of turbidity current 1 . 5 m s on average slope of 0.0028 for 9.6 km. Sediment in the currents was fine grained (72% less than 5 p , 6 0 % less than 2/j, "a few stray grains of sand . 1  Lake Constance (Rhine River)  Fore I (1885) Collet (1925) Bell (1942)  Large trench bordered by leeves leading from the mouth of Rhine River (4 km long, 600 m w i d e , 70 m deep). Density currents due partly to sediment load and partly to temperature differences. Counter circulation indicated by frequent debris line. Heavy rain and mud flow lead to high clay content suspended sediment discharge and thus turbidity current flow. V e l o c i t y is calculated as 0.07 m s " ' .  Echo Reservoir, Utah, U . S . A .  Redden (1938)  Elephant Butte Reservoir (Rio Grande River)  Fiock (1934)  Eril Emda, Algeria  Duquennois (1956)  Managed withdrawal of underflow at the base of the dam reduces sedimentation by 4 5 % . Underflow velocities up to 1 .2 m s recorded. Photocell at the base of the dam monitored 38 events in 2 years.  Lake Leman (Rhone River)  Fore I (1885)  A large trench bordered by leeves leading from the mouth of the Rhone River (6 km long, 500-800 m wide). Erosion by turbidity currents seen as mechanism of formation, of which coarse grained sediments are the principal agents. Counter circulation is indicated by presence of frequent debris line.  Bell (1942)  Bliss (1938)  Collet (1925) Bell (1942) Houbolt and Jonker (1968)  Heavy silt inflow during summer associated with torrential rains leads to turbidity currents of less than 1.8 m thickness that flow under the 53 km long reservoir in 2-5 days. Underflow was noted in 7 of 20 years of operation (15 events between 1931 and 1936). River water temperature was 6 ° C warmer than that of the reservoir. Sediment discharge at the dam varied from 30 - 100% of river sediment discharge during density current flow.  5 Table 1.1 continued Lake Mead (Colorado)  Grover and Howard (1938) Howard (1953) O ' B r i e n (1938) Rubey (1938) Bell (1942) Grould (1951) Lara and Sanders (1970)  Three turbidity currents between February 1 December 3 1 , 1935 (11 in first 7 years of operation) associated with high sediment inflow (250 x 1 0 ° ton d a y ) and small particle size ( 9 0 % finer than 20 fj). High salinity in summer leads to flocculation and rapid settling, thus no turbidity currents. 6 5 % of volume of sediment in Lake Mead is the result of turbidity current deposition. Thickness of the currents is estimated as 4 - 7 m and velocity as 0.006 - 0.65 m s . Underflow always occurred when river water density was greater than 1.011 but densities as low as 1.005 g ml sometimes produced underflow. Average bottom slope 0.00095 over 190 km. Grain size distribution of four turbidite samples was recorded as follows: - 1  1 5/u 5 -20p 20 - 5 0 j u  81% 17% 1.7%  77% 21% 2.1%  70% 27% 3.0%  77% 21% 2.3%  Overflow (spring and summer snow melt), interflow (August and September) and underflow (autumn and winter) all observed. O n l y occasionally do the turbidity currents travel full length of the lake arid w e l l up at the dam. Morris Dam (California)  Grover and Howard (1938)  Turbidity currents with warmer river water lead to lake overturn several days later making clear surface water turbid.  Sautet Reservoir (Drac River), France  N i z e r y and Bonin (1953)  High sediment yield from 840 km^ mountain basin. Underflow velocity calculated as 0.02 to 0.1 m s decreasing downslope, thickness as 2 - 20 m. Currents are not continuous to dam.  T.V.A.  Fry et a I, (1953)  Underflow in 17 of the reservoirs with depths varying from 24 - 88 m and lengths of 11 - 115 km is caused mainly by temperature difference. Reservoirs are strongly stratified, thus currents often occur as interflow.  Robinson (1920)  Turbidity currents noted often as upwelling of turbid water at the dam.  Reservoirs  Zuni Reservoir (Little Colorado River)  6 Since first recognition in the 1930's (see especially Kuenen, 1950; and Kuenen and M i g l i o r i n i , 1950) of the relation between turbidity currents and graded bedding and alternate coarse and fine beds in marine sedimentary rocks (particularly flysch deposits), the literature on the nature of currents and deposits has grown quickly (see Kuenen and Humbert, 1964) and the universality of the phenomenon established. Turbidites have been described in many sedimentary rocks throughout the world (see for example works edited by Bouma and Brower, 1964; and Lajoie, 1970) and are known to be significant in the sedimentology of nearly every present ocean and sea as well as in many fresh water lakes (Table 1.1).  The well known characteristics of turbidites  are summarized by Kuenen (1964) and in an extensive literature survey by Walker (1967, see especially his table 2, p. 32) in which he also reviews Bouma's useful division of the complete turbidite (1959) and relates this classification to the flow regimes of the turbidity currents (see also Walker, 1970).  Sanders (1965) would divide deposits into  those truly deposited from suspension in the turbidity current and those deposited from movement on the bed of material by processes in slumps, flowing grain layers, and moving viscous suspensions, not part of the turbulent movement in the current. the former would he distinguish as turbidites.  Only  While ancient turbidites have been  described largely in flysch sequences, turbidites in glacio-marine and glacio-lacustrine environments deposited in conjunction with the formation of varved sediments, which are the emphasis of this study, have received some attention in the literature. Some early workers (for example, Johnson, 1922 b; 1925;  K i n d l e , 1930; Antevs,  1951) suggested that deposition from suspension rather than turbidity current  deposition was the major process of sedimentation in proglacial lakes.  A t the time of  Kuenen's major studies of marine turbidity currents and associated deposition, he published a paper (Kuenen, 1951) on the role of turbidity currents in the formation of glacio-lacustrine varved sediments in which he developed the earlier ideas of de Geer (1912) on the significance of underflow in the formation of varved sediments.  7 He showed the importance of g l a c i a l sediment in causing sufficient density difference for turbidity current flow despite water temperature conditions and discussed the relation between structure of the varves and the nature of the currents.  The subsequent  findings of most workers studying modern and ancient glacio-lacustrine deposits support his conclusions.  Lajtai (1967) observed a variety of textural composition within  Pleistocene varves which he attributed to quasi-continuous underflow of varying power and sediment load combined with continuous normal sedimentation from suspension. Banerjee (1966) ascribed sedimentation within Carboniferous glacio-lacustrine varves or rythmites to turbidity currents.  Gustavson's study (in press) of ice marginal Malaspina  Lake indicates current structures within varves due to turbidity currents, the roles of surface flow from an overland stream and underflow from the ice margin, and confirms Kuenen's hypothesis (1951) of the much greater significance of suspended sediment than temperature differences in creating density differences.  Stratigraphic evidence  presented by Smith (1959), Ashley (in press) and Banerjee (in press) also support these suggestions.  Agterberg and Banerjee (1969, p. 647) in their statistical study of the  patterns of Pleistocene varve deposition succinctly summarize what is generally held to be the nature of varve formation: . . .a varve couplet has three genetically dissimilar parts (1, 2 a , and 2b): 1 . The silt (summer) part was deposited in a relatively short period by a turbidity current. Variations between successive layers may be strong and a single layer will show a strong decrease in thickness away from the source. When multiple graded units are present, this may mean successive turbidity currents generated in the same year or pulsations within a single turbidity current. 2. The clay (winter) layer consists of two parts: Part 2a deposited by the turbidity current after stagnation and Part 2b deposited by slow, continuous settling from suspension. As their studies showed and as would be expected, fine grained (winter) layers show less variation in thickness both spatially within a single layer and temporally between different layers than do the coarser grained summer layers.  8  Thus, current understanding of lacustrine sedimentation in high energy lakes with high inflow rates indicates that three processes are nearly always active although their relative importance varies from site to site: 1)  dispersion and eventual settling of fine sediment in lake water of nearly  equal density to river water.  Large scale circulation distributes this sediment more  or less uniformly through the lake, 2)  underflow of more dense river water as a turbidity current which carries  large quantities of coarser grained sediment to the most distal portions of the water body, 3)  deposition of the coarsest fraction at or near the river mouths.  Slopes  may be permanently stable when low rates of accumulation prevail but in many lakes slumping removes large volumes of material from the proximal slopes with corresponding major disturbance to the stratigraphy of the lake bottom sediments. A conceptual model of the interactions between these processes and the resulting sedimentary deposits is presented as figure 1.1.  This model was the basis  for the design and execution of the study reported in the following chapters.  1 .2.  The Lillooet Lake study A three year study carried out at Lillooet Lake in the South Coast Mountains  of British Columbia was designed to serve three functions: 1)  to provide a report on the sedimentary environment of a large and  hitherto uninvestigated lake in a region of British Columbia where geomorphic process studies are now just beginning, 2)  to develop techniques suitable for efficient investigation of moderate  and high energy lacustrine environments in mountain areas, 3) Figure 1.1 lakes.  to study specifically the mechanics of the processes of turbidity current (on page following)  A conceptual model of sedimentation in high energy  Sediment remains suspended in lake water  Settling according to  Stokes' Law  7 Fluvial deposition at distributary] mouths  E a>  Settling by vertical density current!  'Final'  Slumping on the delta  a> jr E o  in the l a k e  5  b e y o n d the  o  delta  jr  Erosion by density  in  Deposits  Deposits Slump generated density currents  >.  location  4> E "a <o  currents  River generated density currents  —  INITIATION  PROCESSES  OF  OF  DISTRIBUTION  DISTRIBUTION  RESULTS  OF  DISTRIBUTION  OUTPUT  10 flow and slumping and to relate these processes to the effects observed in the stratigraphy of the resulting deposits, thus providing tools for better interpretation of ancient sedimentary deposits, and the management of mountain reservoirs. The work which is reported in the following chapters consisted of: 1)  monitoring water and sediment inflow to the lake making use of long  term discharge records on the major streams, 2)  mapping the subaerial and subaqueous morphology of the large delta at the  head of Lillooet Lake with the aid of maps and photographs to determine historical changes in morphology, 3)  grab sampling and coring of lake bottom sediments to determine selected  mass properties related to slope stability and rate of compaction and to examine the stratigraphy of the surfacial lake bottom sediment, 4)  daily recording during the melt seasons 1971 and 1972 of water temperature  and turbidity (and thus, qualitatively, sediment content) from surface to lake bottom at one or more sites in the lake to determine patterns of sediment distribution in the lake water, 5)  continuous monitoring of water temperature at the lake bottom during the  melt season 1971 to record the passage of turbidity currents from the river, 6)  continuous seismic profiling of North Lillooet Lake to record the volumes  of sediment deposited and to attempt to reconstruct ancient sedimentological events in the lake.  11  2.  Description of water and sediment input to Lillooet Lake  2 . 1 . Introduction The sources and variations of water and sediment inflow to Lillooet Lake are highly significant in determining the processes of sediment distribution, and stratigraphy of the lake sediments. A n understanding of the physical properties of the basins tributary to the lake, and of the geomorphic, c l i m a t i c , and other processes determining the water and sediment input is therefore useful in understanding these processes of sedimentation.  Perhaps of equal importance, is the potential use of the information  of this research in the interpretation of ancient lacustrine environments ( e . g . Agterberg and Banerjee,  1969; Ashley, in press; Shaw, in press).  Although little is known of  the processes affecting input or their results in the Lillooet River area, their qualitative description in the following sections provides indication of the factors control ling the patterns of material supply to the delta system, and serves to indicate the direction of further work.  2.2.  The role of climatic variables in determining inflow at Lillooet Lake Mean daily temperature and precipitation at Pemberton Meadows (for  location of all stations see figure 2.1) for the standard period 1931-1960 are shown in figure 2 . 2 .  A l t a Lake precipitation records (figure 2.2) began in 1936 but temperature  data do not appear in the Monthly Records until 1952.  Temperature records kept  privately by M r . G . Richardson at Lillooet Lake from 1962 to the present and kindly made available to the writer, are summarized in figure 2 . 3 .  Short-term melt season  records of temperature and precipitation for higher elevations are collected at Place G l a c i e r (Glaciology Division, Inland Waters Directorate, Environment Canada) and as part of continued research in M i l l e r Creek Watershed ( H . O . Slaymaker),but it will be some time before representative data will be available.  12  Figure 2.1 L o c a t i o n of data gathering stations in Lillooet River area referred to in text.  13  Figure 2.2 Mean daily temperature and precipitation, Pemberton Meadows and A i . a Lake.  14  J  F  M  A  M  J  J  A  S  O  N  D  Figure 2.3 Mean daily temperature at Richardson's Data from maximum - minimum thermometer fastened to shaded northeast wall of unheated garage are not significantly different from those obtained a standard screen established at the site dur.ng two months in 1971 . i n  15 Unfortunately, all stations with long term meteorological data are located in the valley bottom, and do not represent the highly varied climatological regimens that determine the amount of water and its seasonal pattern of inflow to Lillooet Lake. This is especially significant since the events of hydrologic and sedimentological interest are largely controlled by processes in the alpine and subalpine regions of the basins.  Figure 2 . 4 , a plot of temperature during the melt season of 1965 at several  locations, indicates the variability of meteorological conditions throughout the drainage basins. Heavy early fall rains are very significant hydrologically and thus strongly affect sedimentological conditions in the lake.  Figure 2.5 illustrates the role of  sustained rain on normally low discharge in inflowing streams.  Compare these  precipitation events to the total daily precipitation duration series for Alta Lake and Pemberton Meadows (figure 2.6).  Rain storms are particularly significant in causing  high discharge when they are associated with the passing of the warm sector of cyclonic storms following heavy snow events.  Even these data records are inadequate to fully  understand the role of precipitation as a hydrologic control.  For example, mean  annual precipitation at Pemberton Meadows and A l t a Lake are 959 and 1372 mm respectively.  Mean annual runoff from Lillooet, Green and Birkenhead basins are  respectively 1880, 1760 and 1180 mm. Later in the autumn and winter, although heavy precipitation continues, most falls assnow and has little immediate effect on the hydrograph (see for example, the November 2 7 , 1948 event, figure 2.12).  Late winter storms, however, frequently  bring rain and warm temperatures which melt low elevation snowpacks, temporarily raising stream discharge (figure 2.12).  These events may have considerable  sedimentological significance in the lake (section 5 . 4 . 3 . ) . But by far the most significant hydrologic governor is the accumulation and melting of large snowpacks and the water and sediment contributions of the glaciers  16  20 Place Glacier 70  01 30r A l t a Lake  u 3 « Cl  £  0 30  T3  C  20  i  1  1  -y Au A vV IA v  10  Pemberton Meadows  1  1  i  July  1965  August  September  Figure 2.4 Mean daily temperatures f o r stations in L i l l o o e t River watershed during the melt season 1 9 6 5 .  17  800  600h  400h  200  60  PEMBER-  40  TON MEADOWS  20  0  U  Oct.19-31., 1 9 3 7  ft  Oct.  13-26,  1940  Sept. 2-11, 1956  "  A u g 3 0 - Sept. 1 1957  Figure 2.5 Examples of the effect of heavy fall rains at A l t a Lake and Pemberton Meadows on the discharge characteristics of Lillooet River (cf. recurrence series of Lillooet River daily flow values, figure 2 Each bar represents one day.  Daily Figure 2.6  precipitation (mm)  Daily precipitation duration series A l t a Lake 1936-1965 and Pemberton  Meadows 1931-1960.  19 in the stream basins tributary to the lake. documented.  Unfortunately, these are poorly  Data from two snow courses, both east of the area of heaviest snow  accumulation and glacierization (figure 2.1), are summarized in figure 2 . 7 . decrease in snowpack from west to east is clear.  The  Data for accumulation on the  surface of Place G l a c i e r during the winter of 1964-65 published by 0strem (1966a) are reproduced as figure 2 . 8 .  Accumulation at the snow courses in the same period  was near average (Table 2.1). Table 2.1 Snow storage at Tenquille Lake and M c G i l l i v r a y Pass snow courses, 1965 Approximate date, 1965  Accumulation in metres water equivalent Tenquille Lake M c G i l l i v r a y Pass  Feb. 1 Mar. 1 April 1 May 1 May 15 June 1  0.841 0.940 0.983 0.968 0.795  0.500 0.543 0.488 0.498  Note that accumulation above about 2100 metres a . s . l . is of the order of twice that on the Tenquille Lake course, and almost three times the accumulation at the same elevation at M c G i l l i v r a y pass.  These data indicate that perhaps as much as three  metres water equivalent are stored in places on the western boundaries of Lillooet and Green basins.  2 . 3 . Water inflow to Lillooet Lake Inflow from the three major basins tributary to North Lillooet Lake has been monitored by the Water Survey of Canada and its predecessors from as early as 1914 (Table 2.2).  Less than 10 per cent of the area providing water to the lake is  unrepresented by flow measurements.  O f this unmetered area, only Ure Creek  (figure 2.9) is actively building a delta in Lillooet Lake;  its sediment laden water  represents a minor contribution to underflow and sediment distribution patterns (section 2.5.4.).  20  F  M  A  M  F  M  A  M  Figure 2 . 7 Exceedance probabilities for approximately month end and mid month snow surveys (February to May) at Tenquille Lake (1953-1972) and M c G i l l i v r a y Pass (1952-1972). See hypsographic curves (figure 2.16) to compare station elevations in the basins.  '""0  2  Accumulation (m) Figure 2 . 8 Accumulation in metres of water equivalent on the surface of Place G l a c i e r , 1964-1965 from 0srrem (1966a, p. 104) . Data were collected in mid June 1965 when the snowpack in all but the accumulation area was isothermal so that considerable water was already lost particularly at lower elevations.  Table 2.2 Summary of inflow from area tributary to North Lillooet Lake  Station Name  Drainage area (km ) 2  Lillooet River near Pemberton  2100  Green River near Pemberton  Area in k m  2  Years of nearly continuous operation Begin Terminate  Years with missing data  N o . of years of complete record  1925 1927  48  126  a  Mean annual dis,charae  (rrr  14.5  1923  14  1914  1951'  1922  37  47.8  1946  1971  1946 1948 1949 1956 1971  21  23.7  850  Birkenhead River near Mount Currie  Distance from delta (km)  630  tributary to North Lillooet Lake with no measured inflow between Lillooet-Green gauges and delta: between Birkenhead gauge and delta: Ure Creek: Joffre Creek: Other small creeks: Total  Grand total of all area tributary to North Lillooet Lake:  135 46 84 63 37 ( 9 . 4 % of grand total)  365 km 3950 k m  2  °  years with ten or more consecutive missing values of discharge estimated to be greater than mean annual flow,  k  short term records are available during the melt seasons of 1971 and 1972 as well (Appendix 2 . 1 ) .  r  )  23  Discharges are recorded as mean daily flow interpreted from mean daily stage.  The means of mean daily flows for the three streams during their periods of  record are shown in figure 2 . 1 0 .  A l l three have two peaks of summer discharge,  the first in early June and the second in mid J u l y .  O n l y on Lillooet River is the  second peak larger, but on all three it corresponds very closely to the peak of mean daily temperature (figures 2.2 and 2.3).  Note that through June the mean daily  temperature remains nearly steady or even falls, interrupting a general rise to the peak in J u l y .  This may help account for the two peaks in summer discharge.  That  the second peak of Lillooet discharge is higher, is likely a reflection of the larger per cent of the basin that is snow and ice covered, and thus the availability of snow and particularly of glacier ice longer into the summer when higher temperatures are available for melt.  Further, highest mean daily temperature in alpine areas may  occur later than in the valleys as the snowpack becomes less effective in suppressing temperature as its size decreases through the summer. Although a number of workers ( e . g . 0 s t r e m e t a l . , 1967) have noted that rain may more seriously affect the discharge hydrograph than changes in temperature (radiation), its effect is not likely to be seen in summer long-term mean of these rivers since a) the mean precipitation time series is much less stable than temperature and b) precipitation in summer is much less. This is seen in an examination of the variability of mean daily discharge expressed as coefficients of variation for the three stations (figure 2.11).  The  greater relative variability of discharge in late winter is associated with irregular time and volume of low elevation snowmelt, in spring with the timing of the onset of major snowmelt, and in fall with heavy but irregular and short duration rain events. An indication of the relation of discharge to the climatic elements is seen in figure 2 . 1 2 .  Figure 2.10 Mean of mean daily discharge for Lillooet River (1923-1968), Green River (1914-1951) and Birkenhead River (1946-1969), for years of nearly complete record (Table 2 . 2 ) .  25  2  Ol  pi  1—1  I  I  I  I  I  1_J  I  I  1 I  I  I  I  I  I  1 I  1 I  J F M A M J J A S O N D  J F M A M J J A S O N D  Figure 2.11 Coefficients of variation of mean daily discharge (standard deviation * mean) for Lillooet River (1923-1968), Green River (1914-1951) and Birkenhead River (1946-1969).  26  Figure 2.12 Mean daily discharge, Lillooet River and mean daily temperature and total daily precipitation, Pemberton Meadows: a)  1940. The event of October 19-20 is the highest discharge on record.  27  Figure 2.12 b)  (cont'd)  1948, a year of unusually high early summer discharge due to below average temperature until late May and an unusually heavy snow accumulation.  28 The recurrence series (figure 2.13) and duration series (figure 2.14) show that the mean daily flows of Lillooet River can be distinguished into a least two groups. The modal mean daily discharge of 180 - 200 m  J  s  for Lillooet River separates the  low flows of winter and early spring from the summer melt flows, a distinction which is necessary for the formation of varves (section 5 . 4 . 1 . ) . O f the 14,976 data points in the recurrence series, the four largest values are 3 clearly anomalous.  Three of these (900, 716 and 668 m  -1 s  ) are the high flows  associated with the heavy autumn rains on snow of October 19-20, 1940 and of rain 3-1 alone on September 6, 1957.  (The other  (790 m  s  ) represents a delay of the  onset of major snowmelt until late in June in 1968 when the weather suddenly became very warm.) In order to investigate trend in discharge, the total annual volume of Lillooet 3-1 3-1 River greater than mean daily flows from 0 to 400 m s in increments of 10 m s was tested.  None of the 41 volume time series showed trend significant at 95 per cent 3-1  confidence.  That is, neither mean annual flow (volume greater than 0 m s  ) nor  any higher flow, showed a trend for the period of record 1929 through 1971 . A n indication of concentration time of water and sediment from the sources in the upper parts of the basins can be had from an examination of the times of peak water level (discharge) in diurnal fluctuations at a number of stations.  During the  period July 20 to August 6 , 1 9 7 1 , water level records were kept at M i l l e r Creek (McPherson and Slaymaker, 1972, p. 28) and during part of this time diurnal peaks were observed in Lillooet Lake levels.  These are summarized with other available  water level records and the air temperature record at Lillooet Delta in figure 2 . 1 5 . The mean time of each of these maxima and their lags behind earlier maxima are listed in Table 2 . 3 .  Although M i l l e r Creek station is approximately five kilometres  from the glacier supplying most of its water, and although there is a small lake at the toe of this g l a c i e r , while the Place station is a very short distance from the glacier  Lillooet River 1929-1968 Mean daily f l o w  800  5000 1  FtgftF©*2.18  Recurrence interval (days) Pesrsonian Type HI plot of recurrence series ©f mean daily discharges Lillooet River.  10000  20000  Discharge (rrr s"') Figure 2.14  Duration series of Lillooet (1923-1968),  Green (1914-1951) and Birkenhead (1946-1969) Rivers.  July  1971  August  Figure 2.15 Time of maximum daily flow, lake water l e v e l , or air temperature at sites in basins tributary to Lillooet Lake.  32 Table 2.3 Timing of daily maxima of water levels and air temperature at sites in basins tributary to Lillooet Lake, July 20 to August 6 , 1971 Mean time Mean lag behind earlier maxima (hours) of daily maxima Lillooet Temperature M i l l e r Place Green (P.S.T.) Ri ver A i r temperature M i l l e r Creek Place Creek Green River Lillooet River Lillooet Lake  15:30 20:10 21:45 23:35 04:30* 21:40*  0.0 4.6 6.2 8.1 13.0 30.2  0.0 1.6 3.5 8.4 25.6  0.0 1.8 6.8 23.9  0.0 4.9 22.1  0.0 17.2  * on the day following toe, M i l l e r Creek peak flow is 1 .6 hours earlier on the average.  This may reflect the  easterly aspect of the west end of M i l l e r Creek Basin and its glaciers.  2.4.  Factors controlling sediment input to Lillooet Lake  2 . 4 . 1 . Basin Morphometry:  Ponton (1972a, his Table I) summarizes the morphometric  characteristics of Green and Birkenhead River basins.  These may be compared with  similar statistics shown in Table 2 . 4 calculated in the same way for Lillooet Basin. Table 2.4 Summary of morphometric data: G l a c i e r i z e d area Strahler Streams:  Lillooet River V a l l e y (km ) (% of basin area) 2  order 1 2 3 4 5 6  Geometric mean bifurcation ratio Drainage density unglacierized area Circularity ratio Relief Mean elevation Hypsometric integral  number 731 153 26 8 3 1  km  -1  m m.a.s.  460 22 mean length (km) 1 .39 2.26 4.65 7.06 12.2 54.2 3.96 1.01 .416 2560 1577 .529  33 Hypsometry of the three basins is summarized in figure 2.16 (compare with topographic map, figure 2.9).  Ponton (1972a, b) indicates the significance of Pleistocene  glaciation in the hydrology of Green and Birkenhead basins; his conclusions would probably apply to tributaries in Lillooet basin as w e l l . 2 . 4 . 2 . Bedrock G e o l o g y : The following brief summary of the bedrock geology in the basins tributary to North Lillooet Lake is taken from Camsell (1918), Cairness (1925, 1927), Mathews (unpublished), Holland (1964), Nasmith, Methews and Rouse (1967), and Roddick and Hutchinson (1973).  Figure 2.17 summarizes the present knowledge of  the geology of the basins. The basins tributary to North Lillooet Lake are underlain by materials that vary in their potential as sources of sediment from highly erodible, thick deposits of Quaternary unconsolidated materials (5.7 per cent of the basin areas) to generally very resistant igneous intrusive rocks (65 per cent of the area).  The latter consist of  the massive, homogeneous granite, granodiorite and diorite of the northeastern margin of the presently exposed Coast Plutonic complex. 2.70.  Specific gravities vary from 2.65 to  Although these are probably the most resistant to weathering of the rocks in  the basins, they also underlie the major regions of glacierization along the west side of Lillooet and Green basins, so that their sediment production may be equal to or greater than that from the other rock types in the basins. Intermediate in resistance to erosion between the unconsolidated and igneous intrusive material are volcanic deposits (6.3 per cent) and sedimentary and volcanic materials, some of which are severely altered.  They consist mainly of thinly bedded  quartoze sediments with lesser amounts of limestone interspaced with basaltic flows. Resistance to weathering varies greatly from the highly resistant massive greenstones to closely jointed argi I lite and chert. A short lived period of volcanic activity about 2400 years B.P.  centered in  the vicinity of Plinth Mountain, was responsible for large beds of l a v a , pumice and  34  3000  Relative basin area (per cent) Figure 2.16 River basins.  Hypsometry of Lillooet, Green and Birkenhead  Figure 2.17 Bedrock geology of basins tributary to North Lillooet Lake summarized from Roddick and Hutchinson (1973) cast of 1230 W. latitude and Mathews (unpublished) west of 123°.  36 ash deposited and reworked particularly east of Meager and Plinth Mountains in the canyon of Lillooet River.  These materials are probably significant sediment sources  particularly when infrequent major mass movements on the mountain sides occur. Warm and hot springs are known at a number of locations along major structural trends and active faults in the valley:  near Meager-Li I looet River confluence, at  several locations along Lillooet Lake, and along Lillooet River north of Harrison Lake.  Undoubtedly, a number of small springs emerging under overburden or located  in remote parts of the basin are yet to be discovered.  The springs may be significant  in contributing to dissolved sediment yield (Table 2.5). Table 2.5 Summary of the concentration of dissolved material in several warm and hot springs in Lillooet River valley Location  Date  Concentration mg l "  34 km north of Harrison Lake *  394  16 km north of Harrison Lake *  1313  5 km north of Harrison Lake *  743  1 km from Lillooet Delta on northeast side Lillooet Lake  Temperature  1  48.8  D e c . 15, 1971  830  13.8  O c t . 15, 1972  907  14.6  * from El worthy (1926)  2 . 4 . 3 . Processes governing sediment production:  With the exception of preliminary  unpublished research by Slaymaker and the work of Ponton (1972a, b), almost nothing is known about geomorphic processes contributing sediment to Lillooet Lake: relative importance and their spatial and temporal variations.  their  As a first approximation,  the processes and landforms of Lillooet basin can be viewed as a conceptual model (figure 2.18) through which the unaltered rock mass passes to high altitude, mid  High altitude|  L o w altitude  Mid altitude! altered  altered  rock mass  rock mass  LANDFORMS  PROCESSES Unaltered rock mass  Glaciation  s  Solution  V  m  Frost action  V  m  Mass movement . without frost  V  Snow processes  V  Minor processes  m  m  Fluvial processes  V s . m  Yield  altered rock mass  very significant  High altitude altered mass M i d altitude altered mass  m  m  Low altitude altered mass  Talus Ice contact glacier debris Sol ifluction and nivation features -Slump, slide and avalanche features Alluvial fan Forest soil and soil creep vGlacio-fluvial features. F l o o d plain features Yield  . significant minor significance  Figure 2.18 Processes of sediment production and conveyance and resulting landforms in high energy alpine environments: a conceptual model.  CO  38  altitude and finally low altitude altered mass from which come flood plain deposits and ultimately sediment yield at the delta.  The concept of altitudes of altered rock  masses should not be considered too rigorously.  For example, glacio-fluvial  material can be found at high elevations but is usually the result of ice contact debris (altered mass 1) being reworked by streams. sequentially through each of the stages.  Note also that sediment need not pass  For example, the unaltered rock mass may  be subject to chemical solution, the solutes then passing directly to y i e l d . The transformations between each form are characterized by process sets P.| . . . . P^, each of which is considered to be a combination of the specific processes of g l a c i a t i o n , solution, frost a c t i o n , snow a c t i o n , fluvial processes and several minor processes (fire, w i n d , vegetation and temperature or water changes).  While extensive  studies have been carried out in other alpine areas (for example, Embleton and K i n g , 1968;  Slaymaker and McPherson (eds.), 1972), the results are only applicable in a  very qualitative way as in the model of figure 2 . 1 8 . However, it is welI established that glaciers are powerful agents of erosion. Published values of fine grained suspended sediment yield from a number of glaciers are summarized by Church and Gilbert (in press).  Most values are in the range 1.0  - 2 - 1 to 15 metric tons km  day  .  These figures do not include sediment yield by bed  load transport which may be as much as 90 per cent of the total sediment yield (Church, 1972; 0 s t r e m , in press) in glacial streams. -1 If we assume a conservative figure of 5 metric tons day  -2 km  through the  melt season as being the supply rate for glaciers in the Lillooet basin, then something like 0.3 x 10^ to 0.5 x 10^ metric tons of fine sediment production per year might be attributable to glacial a c t i o n .  Glaciers are considered effective principally  in eroding and transporting unaltered rock mass (P^, figure 2.18) at relatively high altitudes.  0strem (1966b) noted a lower limit of present glaciation in Lillooet valley  of 2300 to 2500 metres a . s . l . , although tongues of the largest glaciers reach as low as 1000 m a . s . l .  39 2 . 4 . 4 . Engineering works and land use changes affecting sediment production:  Since  first settlement in Pemberton valley between 1859 and 1962, river training to prevent flooding, and reclamation of valley bottom lands has been carried out.  These  practices have had a marked effect on sediment yield at the delta and on the processes of distribution in the lake. As early as 1911 it was recognized ( N e l s o n , unpublished) that sedimentation in the lower reaches of the river and the low gradient were the principal causes of flooding.  Although minor dyking and draining projects were carried out, it was not  until after World War II that extensive alterations were made to the floodplain. cutoffs were made above the Ryan-Lillooet confluence (figure 2.19).  Six  The largest,  M c K e n z i e Cut (figure 2.20), shortened the river by 3350 metres and doubled the gradient.  Flooding of back channels and low lying land was allowed in order to build  up the land surface.  The total reduction in river length of 5500 metres locally  steepened the gradient and allowed a much more efficient passage of sediment (Slaymaker and G i l b e r t , 1972). In 1952, after extensive surveys, dredging of the outlet of Lillooet Lake was completed thereby lowering the lake level by approximately 2.5 metres (figure 2.21). The sediment that was flushed out as a result of these changes had a marked effect on the rates of delta growth (section 3.1 .). 2 In a l l , some 80 km  of land were reclaimed for agriculture.  It is suspected,  however, that the land clearing itself has not greatly increased sediment y i e l d , because of the small area involved ( 2 . 3 % of the total area of Lillooet basin) and because virtually all the agriculture is carried out on the flat valley bottom land. Logging of the lower slopes of Lillooet basin has gone on at an increasing rate 2 since first settlement (for example, records indicate 0 . 8 km  clear cut in 1950-52 as  2 compared with 10.4 km  in 1967-69 (Slaymaker and G i l b e r t , 1972).  Although it has  been shown that cutting does not seriously affect sediment yield (Bates and Henry, 1928;  Wilson Cutoff 1950-51 tlovering Cutoff pre 1946  LILLOOET  Fowler Cutoff 1947  RIVER  Straightening  &  Reclamation  Fraser Cutoff 1947 Green Cutoff 1947 S£ Cutoff 1950 jMcKenzie 1947 Miller Cutoff ^enhead R.  0  sr  8  JO  Pemberton%=ft  Mt. Currie  Kilometres  Figure 2 . 1 9  Lillooet River valley showing the reclamation and river training carried out  between 1946 and 1951.  Figure 2.20 Lillooet Valley l o o k i n g north northwest f r o m the air over Mckenzie cut on the east and the confluences of R y a n and Miller Creeks f r o m the west  Figure 2.21 Outlet of Lillooet Lake to Tenasse Lake. (B. C. A i r Photograph B C 5 3 4 0 : 1 1 7 ) L o w e r i n g the lake involved dredging the fan of the stream entering f r o m the west and diverting its course to the south of the dredged section.  42 Lieberman and Hoover, 1951), removal-of the logs particularly by tractor skidding and the building of inadequately graded and drained logging roads may increase sediment y i e l d as much.as ten times (Tackle, 1962; Dyrness, 1965; Packer, 1965). It is suspected that forestry practice in Lillooet basin is not directed to minimizing sediment y i e l d increase, and that the high rates of erosion reported elsewhere may apply here.  2 . 5 . Sediment supply at the delta Samples of river water and bed material c o l l e c t e d in the channel of Lillooet River and in the main distributary on the delta were analyzed as detailed in Appendix 2 . 2 .  From these results a measure of sediment inflow to the delta was  obtained. 2 . 5 . 1 . Dissolved sediment supply: A number of writers (for e x a m p l e , C o r b e l , 1959; Durum et a ! . , 1960) have noted the significance of solution as an agent of denudation. However, less work has been done in alpine environments. Williams (1949) demonstrated that the low temperatures in alpine zones do not significantly reduce solution.  The  few published results a v a i l a b l e indicate a wide range of values from 26 metric tons year  - 1 - 2 - 1 - 2 km in Northern Sweden (Rapp, 1960a) to 135 metric tons year km in the  Upper Rhine V a l l e y ( J a c k l i , 1957).  Detailed calculations of rates of solution are not  possible with the data a v a i l a b l e (figure 2 . 2 2 ) .  However, i t w i l l be shown (section  4 . 2 . ) that these values, when compared with the rates of particulate material supply, are not significant in the density differences observed in Lillooet Lake and thus in the processes of sediment distribution by interflow and underflow. A n estimate of the rate of dissolved sediment y i e l d was attempted from the data.  During the two years of r e c o r d , concentrations of dissolved material varied  between approximately 20 and 45 mg I ^ with a mean of 3 0 . 0 mg I ^ during the melt season. As most workers have observed, low flow concentrations are apparently  J  43  100 1971  + • 50  [—  + + BO  c o  o  I  +  I  -J  I  _i  Figure 2.22 Concentrations of dissolved sediments from Lillooet-Green River at Lillooet Delta 1971-1972.  44  higher - say 50 to 70 mg I ' - a n d , although winter data are not available for Lillooet, a larger error in this estimate w i l l be far overshadowed by much smaller errors of summer concentration measures in the determination of total dissolved y i e l d , as the difference between volumes of winter and summer water inflow is so great. Thus an estimate for mean annual dissolved sediment inflow made from these figures and mean daily flow values is 0.19 x 10^ metric tons for Green and Lillooet Rivers - 1 - 2 (64 metric tons year  km  ).  This value is higher than some of those reported e l s e -  where especially when the nature of the bedrock is considered.  However, note that  the method of calculating dissolved solid concentration (Appendix 2.2) includes very fine particulate matter not able to settle out in several months, as well as genuine ionic solutes.  Note also that the dissolved material supply from hot springs in the  basins (Table 2.5) is very high and although they contribute only a small part of the water they may significantly raise dissolved sediment yields. Dissolved salts are known to  be significant in causing flocculation of fine  grained sediment flowing into oceans, but considerable flocculation also occurs in 'fresh water lakes. 1  Grover and Howard (1938) reported that median grain size  calculated by sedimentation techniques in Lake Mead water using only mechanical dispersion was 2 to 12 times greater than when samples were pretreated with chemical dispersants.  As standard procedure for sediment analysis by settling calls for the use  of chemical dispersants and distilled or demineralized water ( e . g . A S T M , 1963, p. 197), the probability for error in assessing effective grain size in situ is great. In order to test the role of flocculation in the presence of dissolved salts and to assess the errors involved in standard techniques of analysis for Lillooet Lake samples, the following simple test was performed: a large sample of fine sediment taken from the lake bottom eight kilometres from the delta edge (cf. section 5.4.) thoroughly mixed.  was  Sixteen samples of approximately 40 grams each were split from  the large sample and randomly assigned to four groups of four samples e a c h .  Hydrometer  analyses were performed on each sample in two blocks: the first consisting of two groups  45 of four using Lillooet Lake water and the second of two groups using distilled water. Standard techniques were followed closely except that samples in one group in each block were soaked in 125 ml of commercial dispersant ('Calgon') while the others were in untreated water according to their b l o c k . The results of analyses of variance summarized in Table 2 . 6 show that both the difference in treatment and the water used for the tests are highly significant in the determination of grain size and c l a y size content.  The clay size fraction shows the  effect more strongly than mean grain size as would be expected.  Although these  effects are not nearly as severe as those reported by Grover and Howard (1938), they do indicate the significance of f l o c c u l a t i o n in determining effective f a l l diameter of L i l l o o e t Lake sediments. 2 . 5 . 2 . Suspended sediment discharge:  In a relatively high energy, sand bed stream  such as Lillooet River the distinction between bed load and suspended load transport is somewhat arbitrary.  Suspended sediment for this study is defined as material caught  in a US D-49 integrating sampler ( S t i c h l i n g , 1969, p. 96) lowered from surface to bottom and raised a g a i n .  The samples are analyzed as detailed in Appendix 2 . 2 .  The concentration of suspended sediment shows a reasonable relation with the combined discharge of Lillooet and Green Rivers (figure 2.23) even though no account was taken of hysteresis between concentration and discharge on rising and f a l l i n g , stages.  However, during the period of sampling in the melt seasons of 1971 and 1972,  one major exception was noted.on the f a l l i n g stage from the year's peak flow (August 3-7, 1971 - the solid circles in figure 2 . 2 3 ) .  During this time a minor river training  project was being carried out 43 km upstream involving the cutoff of a meander and bank s t a b i l i z a t i o n .  Although no records e x i s t , agricultural and engineering  a c t i v i t i e s (section 2 . 4 . 4 . ) and natural events such as major slides in the lower v a l l e y of Meager Creek reported to have occurred in the mid 1930's must also have caused severe disruption to the relation observed in the two melt seasons (see also section  Table 2.6 Effect of dispersant on grain size determination - Lillooet Lake sediments  w Geometric Mean grain size  Per cent of sample in clay size range  None  Block  Treatment Dispersant  Treatment Dispersant  Block  None  Lake water  7.04 9.33 8.48 6.50  10.80 11.31 11.32 11.31  7.84  11.18  10.127 9.780 10.508 9.260  9.049 8.920 8.767 8.753  Mean  9.919  8.872  Distilled water  9.494 9.030 8.957 9.607  8.707 7.482 7.492 7.742  Distilled water  10.47 11.89 9.78 8.84  14.06 16.64 16.06 16.82  Mean  9.272  7.856  Mean  10.24  15.90  Lake water  Results of analysis of variance  Variance source Treatment Block TB interaction  F value 32.24** 14.66** 0.824 ns.  Variance source Treatment Block TB interaction  F value 64.75** 40.47** 4.247 ns.  47  Figure 2.23 Suspended sediment - discharge relation for LillooetGreen Rivers at Lillooet Delta, 1971-1972. Numbers by solid circles indicate dates in August 1971 of anomalously high concentrations.  48 5.4.2.)  so that an attempt to calculate long term sediment yield from these data  would no doubt be of questionable value.  A n estimate of yield based on rates of  sedimentation in the lake (section 5.6.) is more reliable. In conjunction with attempts to calculate the density structure of the lake water by recording water temperature and light transmissivity (section 4 . 2 . ) , observations were made, usually at least twice d a i l y , of the transmissivity of Lillooet River water at the mouth of the largest distributary.  The results of measurements with  a standard 0.30 m diameter solid white Secchi disc and a beam transmittance meter (Hydro Products Model 410-BR and 411) displaying per cent of incandescent light transmitted through a 0.10 m narrow path of water (see Jerlov, 1968, p. 48 ff) are shown in figure 2 . 2 4 .  Despite considerable scatter in the relation between Secchi  depth and transmissivity (figure 2.24 a) there are reasonable grounds to use the predictive relations of figure 2.24 b and c to estimate suspended sediment concentration in the river. Two complicating factors in these relations are not considered: 1) The grain size of the suspended sediments is equally as significant as the concentration in determining the relation with light attenuation (see for example, Lewis, 1970).  Although no data are a v a i l a b l e , it is suspected that the texture of  suspended sediments does not change greatly during the melt season when observations were made.  Nevertheless, this factor undoubtedly accounts for some of the scatter  in the plots.  Fitting Lewis' (1970) relation for concentration (c = 6.7 D^/SD) to  figure 2 . 2 4 b gives a grain diameter (D^) of 21 JJ, a figure in reasonable agreement with suspended sediment in other glacier fed streams (for example, Church, 1972). 2)  Since measured concentrations are from integrated samples (top to bottom)  Figure 2.24 (on page following) Relation between measured optical properties and suspended sediment concentrations in Lillooet-Green River at Lillooet Delta: a) Per cent transmission of incandescent light in a 0.10 m path length plotted against Secchi depth observations. b) and c) Predictive relation between Secchi depth (b), transmissivity (c) and measured suspended sediment concentrations.  Concentration (mg I"  1  o  * 1 0 " ) (= g I" ) 3  1  bo  a\  °  is)  Secchi depth - SD (m)  !u  <Ti  bo  4^  J  I  I  I  I  I  50  while Secchi depth readings are made in near-surface water, some scatter is probably related to the change in vertical concentration differences with different flow conditions.  Calculations of concentration of suspended sediment (see for example,  Henderson, 1966, p. 422 ff) at the water surface gives values between 0.3 and 0 . 7 of that near the bottom depending on grain s i z e , water temperature, concentration (which affects von Karman's constant), stream v e l o c i t y , and the friction factor. These relations and the discharge-concentration relation of figure 2.23 were used to calculate suspended sediment concentration variations in the melt season of 1971 and 1972 (figure 2.25). From water temperature and suspended sediment concentration, the density of river water can be estimated.  Temperatures were taken twice daily in early morning  just after peak flow (figure 2.15) and in late evening (see for example the 1971 data, figure 4.13).  When diurnal discharge fluctuations were occurring during much of the  clear, warm weather of the melt season, warmest water temperature was observed shortly after the peak flow in early morning and coolest in the evening about the time of lowest flow.  Temperature range between readings seldom exceeded 3 C ° , and  errors in interpolation are thought to be less than 0.5 C ° which would represent a density error of 0 . 4 x 10 ^ g m l  \  Values of density calculated at three hour  intervals are plotted in figure 2 . 2 6 . 2 . 5 . 3 . Bed load discharge:  Suspended sediment discharge is that calculated from  integrated water samples taken in the river.  The remaining, usually more coarse  grained material that is moved in permanent or occasional contact with the stream bed or as part of the bed forms of the stream was not measured directly for reqsons discussed below. The point of interest in terms of sediment yield to the lake is the delta l i p . During flood when the largest amounts of material are delivered to the delta, flow occurs on a front 1 .6 km wide (figure 3.1) consisting of shifting channels up to several  51  li I  I II  II  II  May  II  II  ii  !111 II  n U U " 10 20 June  II  II  June  II  II  'I' 30  III  "  U U 10 July  II— I  July  U U 20  llll 30  II  U U U LI—LU 10 20 30 August  II  H|l  II  II  II  II  II  II  II  August  Figure 2.25 Concentrations of suspended sediment in the main distributary of Lillooet-Green River at Lillooet Delta during parts of the melt seasons 1971 and 1972 based on the relation shown in figure 2.23 and three hour mean discharge data. Data were adjusted where Secchi depth, transmissivity and sample analysis indicated significant departure from the calculations according to the relations of figure 2.24 b and c.  52  10016  10 10016,  20 July  II, II  II  10 20 August  I! - !l  II  II  II  1972  10012  "E °10008 '<2 10004  <u O  10000;  pgOC  1  11  11  10  1  May  1  20  1  30  11  11  11  10  20 June  H "  30  II 'I II 10 20 July  30  II  II II n I' II 10 20 30 August  Figure 2.26 Water density in the main distributary of Lillooet-Green River at Lillooet Delta during parts of the melt seasons 1971 and 1972. Three hour mean values were calculated from the concentration data of figure 2.25 and twice daily temperature observations.  53 hundred metres wide and two metres deep, between which much of the surface is flooded.  Determining bed load discharge at the lip would require an extensive  sampling program involving logistic difficulties such as position finding, relocation of sample stations, and avoiding masses of floating and anchored debris which often block the channel mouths (figure 3.2 d). A t the delta lip stream velocity and slope are very variable both spatially across the delta front, and through time. in the river and lake.  Slope changes are the result of stage changes  During much of the melt season stage changes in Lillooet River  follow a diurnal pattern behind which Lillooet Lake stage changes lag by approximately 17 hours (figure 2 . 1 5 , Table 2.3).  Thus during part of the day sediment is deposited  in the lower reaches of the distributaries in backwater only to be flushed out later when the river is higher and the lake lower.  Occasionally the lake falls quickly  enough to create slopes large enough for shooting flow, antidunes and very rapid flush-out of sediment.  The short period of lake level record available (figure 2.27)  indicates that stage changes of up to 4 m occur in the lake between winter and summer. Thus during falling stages in late summer and into autumn much bed material in the lower reaches of the distrubutaries is washed into the lake so that the potential for oversteepening and slope failure (section 3.2.) on the proximal delta front may be greater at this time than during the height of the melt season when most bed load is brought to the delta.  These considerations make bed load measurement at the delta  lip very difficult. Measurement of bed load discharge in the single channel upstream from the distributaries avoids some of these difficulties.  However, no account can be taken of  material that passes this sampling point but that does not reach the delta lip:  material  which is deposited either temporarily or permanently on the delta surface or in lower energy distrubutary channels.  As w e l l , great practical difficulty was encountered  here in attempting to measure bed load.  Dunes of amplitude up to 0.8 m and stream  55  velocities in excess of 2 m s ^ were observed at the sampling site making the use of a V U V bed load sampler difficult.  In several tests, catches in the sampler varied  from less than 0.5 kg min ^ m (width) ^ to 17 kg min ^ m (width) ^ under the same conditions of flow.  Presumably the differences were due to the placing of the sampler  on top of or behind an individual dune (cf. for example, Stichling and Smith, 1968, P- 9). Many of the difficulties discussed above make the application of standard bed load discharge equations equally uncertain and orders of magnitude difference were obtained in calculations among several techniques.  The difficulties of maintaining  measurements of slope, depth and velocity in the lower reaches of the distributaries added to the uncertainty.  Since the results are of questionable validity they  are not  reported here. However, the results of map and air photograph measurements (section 3.1 .), lake bottom core sampling , and deep penetration seismic profiling (chapter 5) do permit a much more useful measure of the long term, average rate of sediment y i e l d to the lake without the estimates and errors involved in direct measurement (section 5.6.). Further, there is a clear distinction in texture between the sediments of bed load and bed material (figure 2.28 and Appendix 2.3) and those of the bottomset deposits (see for example, figure 5.2).  That is, on the basis of textural evidence,  coarse material is rarely transported far beyond the delta front so that the intra-varve structures (section 5 . 4 . 3 . ) are a reflection of short term (daily) inflow changes of fine grained, largely suspended sediment, for which some measure is a v a i l a b l e , rather than of bed load discharge.  Figure 2.28 (on page following) Texture of bed material from the main distributary and the channel above the distributary of Lillooet-Green River at Lillooet Delta, 1971-1972.  Per cent finer  Per  cent coarser  9S  57 2.5.4.  Sediment inflow from other streams:  Two other streams with appreciable  discharge into the north end of Lillooet Lake are Birkenhead and Ure Creeks. Observations of transmissivity and Secchi depths were made on streams in July and August 1972 and are plotted as concentration of suspended sediment in figure 2 . 2 9 using the relations for Lillooet River (figure 2 . 2 4 ) .  1000  1  Ftf + Ure Creek  c "I  • Birkenhead  500  . <_> e o o  -i-  +  10  +  1 5  July  +  20  +  25 1972  River  +.  5  10 August  ±1 15  Figure 2.29 Spot observations of transmissivity and Secchi depth expressed as concentrations of suspended sediment at the mouths of Ure and Birkenhead Creeks, 1972. ,-1 Concentrations in Birkenhead River seldom rise much above 100 mg I even during f l o o d .  The stream water was always clearer than lake v/ater and was not  observed to plunge beneath it down the deifa front at any time.  Thus, with the very  much lower discharge as compared with Lillooet River (figure 2 . 1 0 ) , the suspended sediment y i e l d from Birkenhead basin is of the order of 0 . 8 to 1.1 per cent of that of L i l l o o e t - G r e e n basin. Concentrations of suspended sediment in Ure Creek are lower than in L i l l o o e t Green River with one notable exception (July 3-4) during the period of record. no discharge data are a v a i l a b l e , it is believed that the flow characteristics are  While  58 similar to adjacent Green River.  The observed concentration values and estimated  flow from Green River records indicate that the suspended sediment yield from Ure Basin is 0.5 to 0 . 8 per cent that of Lillooet-Green basin.  Despite the fact that  concentrations are high enough for underflow to occur, the volumes of water and sediment involved are probably not very significant in determining the nature of the lake bottom deposits.  The effect of the very much coarser grained deposits of the fan  itself may not be so insignificant, however.  2.6.  Summary Lillooet River has the largest drainage basin in the southern portion of the  Coast Mountains of British Columbia. Above Lillooet Lake the basin is drained by 2 two principal streams:  Lillooet River (drainage area, 2100 km ) and Green River  2 (850 km ), the confluence of which is nine km from the lake.  Smaller streams,  Birkenhead, Joffre, and Ure increase the total drainage basin tributary to North Lillooet Lake to 3950 k m . 2  Climate is transitional between the west coast marine climate and a continental climate of the interior.  Mean winter temperature remains below freezing for four  months although short periods of melt occur at low elevations during most winters. In the alpine zone summer maxima seldom reach 2 0 ° C .  Precipitation reaches a  maximum in December of 150 - 200 mm in the valley bottom.  Much higher values  are estimated at higher elevations, particularly toward the west of the basin.  Summer  precipitation is much less, seldom exceeding 25 mm in August. The rugged topography of the basin is dominated (65% of basin area) by massive igneous intrusive plutons of granite, granodiorite and diorite, particularly in the northeast.  Post-Pleistocene volcanic activity in the region of Meager Mountain  dated at 2400 years B.P.  contributed large amounts of sediment, mainly in the form of  pumice and ash, to sediment yield and there is some evidence of continued tectonic activity in the basin.  59 2 Presently, glaciers and ice fields cover approximately 270 km primarily in the higher snow accumulation area along the west and north sides of the basins. Large, active valley glaciers extend to as slow as 1000 m a . s . I.  A n n u a l snow  accumulation in the shadowed eastern side of the basin where records are kept at 1700 m a .s. I. is of the order of 1 m water equivalent and is up to 2 m at higher elevations. It is estimated that 3 m or more may be average on the west side of the basin. Physiography and the climatic variables determine a hydrologic regime typical of the southeastern Coast Mountains.  Winter discharge is usually lowest in March just 3  before the onset of snowmelt, averaging 50 m  -1 s  from Lillooet-Green River. 3  -1  Beginning in A p r i l , inflow at Lillooet Lake increases to a mean maximum of 400 m s 3 by June although flow of 600 m  -1 s  is not uncommon.  In response to poor weather  characteristic of June, and to diminishing snow pack s i z e , averqge inflow decreases until July when a second peak, greater than the first only on Lillooet Reiver, ooaurs. From August inflow decreases steadily to the low values of winter.  However, discharge  is highly variable from year to year reflecting the strong climatic control and basin response.  Most noteworthy are the high discharge, short duration floods of autumn  frequently caused by warm rain in the warm sector of cyclonic storms following heavy snow fall at higher elevations along the storm's warm front. of October 19-20, 1940, which  Such an ev©nt was that  caused extensive flooding in the highest Plow in 50  3-1 years of record  - 1300 m  s  for Lillooet-Green River.  This hydrologic regimen in turn controls the sedimentological conditions at the mouth of Lillooet-Green River and throughout Lillooet Lake.  Si^imeht'transport, both  as bed and suspended load, is high even for similar mountain ^virbnments\  Although"  there are a number of warm and hot springs in the basin wrm q^s%o1Ved' ma'teri'd>l' : ;  contents as high as 1000 mg I \  :  dissolved load at Lillooet Lake is approximately 50 mg>  1 ^, an order of magnitude less than suspended load during the melt seadon.  Suspended  sediment which usually correlates well with discharge, reaches 1500 - 2000 mg I ^  60 during much of the melt season, although anomalous values in excess of 3000 mg I ^ have been observed during a river training project upstream.  In the melt season,  daily fluctuations in inflow cause peak sediment inflow from melt on the previous day at approximately 4:30 P . S . T . although time varies with meteorological conditions. Bed load discharge is unknown, but it is believed that most of the coarser material is deposited at the front of Lillooet Delta.  61 3. 3.1.  Physiography of Lillooet Lake Subaerial morphometry of Lillooet Delta 2 The subaerial surface of Lillooet Delta is a 1 .0 km  rapidly shifting braided distributaries.  triangular sand plain of  It slopes at approximately that of the lower  reaches of the river (0.0006) and is 1 - 2 m lower than the surface of the main flood plain of Lillooet River to the north. annually.  Thus it is subject to nearly complete inundation  The amount of surface flooded, and therefore susceptible to erosion/  deposition, is approximately a function of discharge, although as deposition continues, bars, particularly at the upstream end of the delta, are less completely flooded by a given discharge.  Channel patterns at various flows are shown in figure 3.1 .  Although local relief does not exceed 1 m except during low lake levels of winter, figure 3.1 shows that dominant distributaries have maintained themselves through most of the recorded history of the delta.  The main distributary usually  follows the valley side along the south side of the d e l t a , and the large linuoud bar indicated A in figure 3.1 c has grown and become stabilized by vegetation, principally Populus trichocarpa and Salix (sp.)  (figure 3.1 d).  Lowering of the lake in 1952 may  also have assisted in the quasi-permanent establishment of these bars.  Other bars can  also be traced through the photographs of figure 3 . 1 . Figure 3.1 (on pages following) as follows:  a b c d e  Selected vertical air photographs of Lillooet Delta  Date  Lillooet River 2 _] discharge (m s )  June 1 6 , 1 9 5 8 April 2 9 , 1960 July 2 2 , 1962 April 7, 1967 July 19, 1969  510 63 239 35 219  D  , . , Photograph number  BC2430:26 BC5012:53 BC4071:95 BC5225:118 BC5340:208  Compare b) with an air photograph taken July 3, 1960, in Church and Gilbert (in press) and d) with figure 3 in Gilbert (1972) taken September 13, 1967. Actual discharge is approximately 35 - 5 0 % greater than shown above due to unmetered addition from Green River. A l l photographs from British Columbia Government, Department of Lands, Forests and Water Resources.  62  63  64 Except in local scours, water depth seldom exceeds 1.5 m anywhere on the delta even during flood.  Dunes of amplitude 0.2 - 0.3 m are the dominant bed form  in the main distributaries although plain bed conditions or antidunes occur when the lake level falls sufficiently rapidly to cause oversteepening near the delta front (section 2 . 5 . 3 . ) .  During low flow when the lake level is down (figure 2.27),  channels become incised in the delta surface and erosion of the unconsolidated sands of the banks is rapid (figure 3 . 2 b ) .  The exposed surface is also susceptible to  significant wind erosion particularly in early spring when the sediment is dry (figure 3.2 c).  During much of the winter, however, snow cover prevents nearly  all but a small amount of bank erosion (frontispiece). Except on several bars near the upstream end of the delta, living vegetation is not significant in stabilizing the delta surface.  Equisetum which rapidly colonizes  the finer sediments following the melt season floods (figure 3.2 b), is destroyed or buried in the floods of the season following.  However, part of a large amount of  floating debris (figure 3.2 d) which is kept near the delta by counter circulation of lake water in the summer (section 4 . 2 . ) , comes to rest on the delta as the lake level falls.  Large logs and tree trunks which become embedded in the sediments  significantly retard bank erosion along the distributaries and wind wave erosion along the delta front during low water periods.  Figure 3.2  (on page following)  a)  Low oblique photograph of Lillooet Delta on November 2 6 , 1969, showing^ abandoned channels and debris armouring. Lillooet River discharge, 50 m s  b)  Bank erosion on secondary distributary September 2 3 , 1969. Lake level is approximately 3 m below summer maximum (figure 2.27). The bank in the background is that of the main floodplain of Lillooet River.  c)  Sandstorm in 15 m s ^ southeast w i n d , May 11, 1973.  d)  Oblique photograph of Lillooet D e l t a , August 2, 1971. River, 493 m s ' : Green River, 175 m s " . 3  1  3  1  Discharge of Lillooet  65  66  A n indication of the activity of the delta and of the large loads of sediment delivered to it can be obtained from measurements of the rate of delta front advance. In 1858 surveys were carried our by the Royal Engineers in conjunction with road building to the interior of British Columbia. The resulting map (Palmer, 1869) at 4 miles to one inch (1:253,440) shows the delta front at that time.  Detailed mapping  of the delta (1:4800) was carried out in November 1913 for a land reclamation report prepared by Cleveland and Cameron, Engineers (unpublished).  Provincial and  private agency air photography of Lillooet Delta began in 1948 and has been repeated at regular intervals since.  From these sources, the map shown as figure 3.3  was compiled. The area between successive fronts was measured and divided by the valley width to get the mean and mean annual frontal advance in each period shown in Table 3 . 1 .  The rapid increase after 1948 is probably associated with the steepening  of the river gradient when Lillooet Lake was lowered 2.5 m in 1952, and with the major dyking and straightening program carried out on the river in this period, as well as with logging and agricultural clearing in the watershed (section 2 . 4 . 4 . ) . Table 3.1 Rates of advance of Lillooet Delta  Period  1859-1913  1914-1948  54  35  1948-1953  1953-1969  Time  (years)  Mean frontal advance  (metres)  374  284  140  338  Mean annual advance  (metres)  7  8  30  21  4.7  16.1  Figure 3.3  Advance of Lillooet Delta between 1858 and 1969  68 3.2.  Lake floor physiography The morphometry of the subaqueous portion of Lillooet Dalta (figure 3.4),  which is not unlike other moderate energy lacustrine deltas (for example, Columbia River in Arrow Lake;  Fulton and Pullen, 1969), is described by Gilbert (1972, 1973)  and in McPherson and Slaymaker (1972, p. 20).  These reports are expanded upon in  the following paragraphs. From echo sounding surveys, the subaqueous delta deposits can be divided into three genetically distinct zones on the basis of slope and surface features. 1)  Classical foreset beds: at the distributary mouths is a sharp break of slope  from the channel bottoms to foreset beds dipping forward at 20 to 3 5 ° .  This transition  between the area of bed material transport and the avalanche zone of foreset bedding is particularly distinct in winter (figure 3 . 5 ,  cf. J o p l i n g , i960; A l l e n , 1966, 1968).  Within the limit of resolution, echograms indicate similar conditions during high flow and that this steeply dipping foreset zone extends only to 6 - 8 m depth. 2)  A n extensive zone of slumped material making up most of the delta front  may be subdivided into three sections. a)  A t approximately 8 m depth there is another sharp break of slope which  is not distinct on the echograms of figure 3.6 a . approximately 8 ° to 4 ° at 40 m depth.  Below, slope decreases from  The downslope profile is relatively smooth  and slightly concave upward (figure 3.6 a) but some irregularities can be seen in the cross sections shown in figure 3.6 b.  These may represent sublacustrine channels  created by underflow, but if so they are not as distinct or well formed as the channels with levees reported in Lake Mead by  Menard and  Ludwick  (1951),  in Lac Leman  by Collet (1925) and Houbolt and Jonker (1968) or as the gullies interpreted as slump features on large marine deltas such as Fraser River by Mathews and Shepard (1962)  Figure 3.4 (on page following) Bathymetry of North Lillooet Lake from corrected echograms. Horizontal control was by sextant to surveyed shore stations. Horizontal error 1 1 5 m, vertical error: l 0 . 4 m (less in shallow water except where slopes exceed 20°).  . fc<\  7o  Figure 3.5 Photograph l o o k i n g northwest at lip of Lillooet Delta across the m o u t h of the largest distributary during low water and sediment inflow on March 2 0 , 1971. L i l l o o e t River discharge, •3  -1  23 r r r s , lake surface elevation lip is approximately 0.60 m.  194 m a.s.l. Water depth at  71  and Mississippi River by Shepard (1956) and Walker and Massingill (1970).  Mass  property studies of Lillooet Lake bottom sediments indicate that this zone is unstable (section 5 . 2 . ) , and thus normally deposited o r slump deposited sediment cannot accumulate to any significant depth before failure-will occur. b) nearly 4 ° .  Between approximately 40 and 90 m depth slope varies from 3 ° to  This section is distinguished from the one above by a break of slope at  40 m depth and by mounds of material increasing in height from less than a metre in the upper portion to 4 - 5 m near the bottom.  There is little evidence of significant  filling between the mounds by normal sedimentation (figure 3.6 a).  Unlike the zone  above, the sediment of the floor of the lake can be clearly distinguished from the bedrock valley sides which are probably overlain by a thin layer of lacustrine sediments (figure 3.6 c).  The lake is slightly deeper at the contact between sides and  floor, particularly along the southwest side (figure 3.6 c) indicating that deposition occurs more rapidly along the centre of the lake (cf. section 5.4.1 .). c)  From 90 to 120 m depth the slope decreases from 2 . 4 ° to 1 . 4 ° .  Mounds of slumped sediment increase in height to 7 m in the distal portion, 2.1 km from the delta l i p .  The largest mounds are up to 200 m in length, and the echograms  indicate that considerable sedimentation has occurred between them, suggesting that they are older than those closer to the delta.  Figure 3.6 (on pages following) Echograms of Lillooet Lake (for location see figure 3.7). Note that these do not represent exactly bottom form as sound is received first from the point nearest the transducer on a surface normal to the sound path within the transmitted beam cone. This gives the parabolic traces of individual mounds of slumped material and increasing depth errors on steeper slopes (Krause, 1961; O f f i c e r , 1958). Sound velocity is assumed to be 1440 m (for water temperature of 8 ° C , and insignificant effect due to dissolved solids; Bark et a l . , 1964). a)  Longitudinal profile A - G of Lillooet Delta and Ure Creek. B , C , and D mark the crossing of sections (b), (c), and (d) on the following page.  b) and c) cross sections in the second zone of slumped material. d)  cross section D^ - D2 beyond the zone of major slumping.  e)  longitudinal profile H - O beyond Lillooet Delta in the deepest part of the lake.  73  Ol  Figure 3.7 Map showing location of echo sounding profiles of figure 3 . 6 , water level recorder, and permanently anchored rafts. Profiles A - D and C , - C were repeated at regular intervals during the melt seasons of 1971-1972 in an attempt to detect slumping of bottom sediment (section 4 . 4 . ) . 2  76 3)  The slumped material gives way abruptly at 120 m depth to nearly flat  lying sediment which extends to the south end of North Lillooet Lake.  This plain  slopes at 0 . 2 ° from the proximal edge to the constriction formed by Ure Creek (figure 3.4), beyond which slope increases to a maximum of 0 . 6 ° before reaching the flat floor in the deepest part of the lake.  Throughout this zone the sediments are  nearly flat lying from side to side and meet the valley sides abruptly (figure 3.6 d). A n exception is the fan delta of Ure Creek where the echograms of figure 3 . 6 a shows some slumped sediment at the base of the high angle delta. third zone is not featureless however.  The lake floor in this  O n the slightly greater slope beyond the Ure  Creek constriction are several low mounds less than 1 m high and 100 - 200 m long, terminated at the base of the slope by two series of mounds 2 - 3 m high (H - J , figure 3.6 d).  Subbottom sound penetration between these groups of mounds indicates  extensive deformation and folding of the upper layers of sediment (I,  figure 3.6 d).  Although there is much less resolution beneath the sediment surface in the slope area (H - I, figure 3.6 d) there is some indication of deformation here as w e l l .  The nearly  horizontal floor of the lake between J and N of figure 3.6 d consists of two flat surfaced zones where relatively deep sound penetration occurred (J - K, and M - N) separated by a slightly elevated zone of long low mounds.  Sound penetration between  K and L (figure 3.6 d) indicates that the original surface of the mounds was some metres below the present surface but that sediment has been deposited in layers of more or less uniform thickness, preserving the original topography with only partial filling of the hollows.  This z o n e , K - M , is probably the result of an extensive slump or  slumps that predates the core records discussed in section 5 . 4 . (cf. results of continuous seismic profiling, section 5 . 5 . ) .  In the flat surfaces areas (J - K, M - N) pinching  out and leansing of individual layers can be seen with some deformation on either side of the slump z o n e .  Cross section echograms in this basin area show similar features,  and that the slump zone (K - M) extends across most of the width of the lake floor, although it was not possible to detect from which side the material slipped.  77  From the lake floor the water shallows to 100 m depth at the constriction at the south end of North Lillooet Lake.  There is no evidence of sediment disturbance  on this slope.  3.3.  Summary The delta surface consists of a series of shifting, braided distributaries and  lingoid sand and gravel bars and is usually nearly completely flooded each year. Thus live vegetation is scarce and becomes established with difficulty.  However,  floating debris becomes lodged and buried to effectively armour the channels and delta front.  As lake stage falls 4 m from the level during melt season, in winter much  bank erosion and channel scour occur in the unconsolidated sands in the oversteepened reaches near the mouths, contributing large loads of coarse sediment to the foreset beds.  Smaller daily fluctuations during the melt season have a similar effect.  Delta has advanced into the lake on a 1.6 km front at 7 - 8 m yr  Lillooet  between 1858 and  1948, but at 20 - 40 m yr ^ since as a result of river engineering and logging activity in the basin. The subaqueous morphometry of Lillooet Delta is summarized in Table 3 . 2 . Table 3.2 Summary of the subaqueous morphometry of Lillooet Lake Morphometric zone 1)  Foreset beds consisting mainly of fluvial bed material avalanching from the distributary mouths  2)  Finer sediments form a relatively smooth surface to 20 m depth, beyond which slump deposits occur in irregular mounds increasing in height to 7 m in the deepest part of the zone  3)  Nearly flat lying fine grained sediments with some evidence of small slump deposits from the subaqueous valley sides with resulting disturbance and contortion of the lake floor sediment  Surface slope  Vertical extent  20-25°  near surface to6-8 m depth  4-8°  6-8 m to approx.40 m  3-4°  40-90 m depth  1.4-2.4° 0-0.5°  90-120 m depth 120-137 m (maximum depth in lake)  78 4.  Processes of sediment distribution in Li I Idoet Lake Findings reported in Chapter 2 indicate that large amounts of sediment are  delivered to Lillooet Lake each year and that the timing and nature of this sediment inflow, although concentrated in rhe melt season, are highly irregular.  In order to  relate the stratigraphy observed in samples collected from the lake floor (Chapter 5) to these patterns of inflow, observations were made of the processes of sediment distribution in the lake. 1)  They consisted of:  profiles, recorded usually d a i l y , of temperature (with a thermistor probe)  and transmissivity (measured as per cent incandescent light passing through a 0.10 m water path) at two raffs permanently anchored in the lake (figure 3.7) during the summers of 1971 (raft 1) and 1972 (rafts 1 and 2).  Measurements were made at 3.05  m vertical intervals and results are summarized in figures 4.1 - 4 . 3 . 2)  profiles of temperature, usually at raft 1, taken intermittently at intervals  of up to six weeks from June 1970 to October 1972. 3)  profiles of temperature at various points throughout North Lillooet Lake at  various times from 1970 to 1972 to observe 'instantaneous' patterns in the lake. 4)  profi les of transmissivity recorded at the same time as the temperatures of  2) and 3) after July 11, 1971. 5)  continuous record of temperature approximately 1 - 2 m from the lake  bottom at raft 1 in the period July 11 - September 14, 1971. 6)  echo soundings repeated at intervals of 1 - 2 weeks through the melt seasons  of 1971 and 1972 (figure 3.7) in an attempt to record movement of the slumped sediments of the delta front. (on the four pages following) Figure 4.1 Records during 1971 at raft 1 of a) temperature (isotherm interval, 1 ° C ) and b) transmissivity (isoline interval, 10%). Figure 4.2  Records during 1972 at raft 1 of a) temperature and b) transmissivity.  Figure 4 . 3  Records during 1972 at raft 2 of a) temperature and b) transmissivity.  I— 1  00K  1 0 061  ;—i  0 081  1 OOtl  1 0 091  1 O'OSI  n 1 0»1 O'OEl 'TS'b'W  1 OKI  .  1 O'OII  i  O'OOl  r006  83  These changes in measured lake water properties are the effects of the river water and sediment moving through and being dispersed in the lake.  As discussed  below, it is these observations that provide most of what has been learned of the distribution patterns.  4.1 . Suspended sediment concentration and density of lake water Transmissivity data, Secchi depth readings and near surface water samples analyzed for suspended sediment concentration as detailed in Appendix 2.2 are not sufficient to calculate the suspended sediment concentrations throughout the lake. The relations established between these variables at various points on the lake surface are shown in figure 4 . 4 .  The differences in relations between Lillooet River water  (figure 2.24) and those of figure 4.4 are a measure of the difference in effective grain size of the sediments.  While application of Lewis' (1970) relation to the data of  figure 2 . 2 4 b yielded an effective grain size of 21 p in the river (section 2 . 5 . 2 . ) , similar calculation for the lake gives an effective grain size of 1 . \j) at the lake surface (some of the scatter in figure 4 . 4 b is no doubt related to changes in grain size throughout the period of observation).  Even small currents, wind generated turbulence,  and Brownian effect become significant in retarding or preventing the slow settling of these fine grains. So, while the relations of figure 4 . 4 are reasonable for the lake surface water, they cannot be applied at depth because interflow and underflow bring coarser grained material into the lake (cf. stratigraphic data, section 5 . 4 . ) .  Thus, the patterns of  transmissivity (figures 4.1 - 4.3) are useful only in determining river water distribution in the lake.  Water sampling through the lake with sufficient intensity to adequately  monitor concentration changes would involve a minimum of 2000 samples during the melt season, which was beyond the capability of the project.  84  1/SD (rrf ) 1  30 50 70 Transmissivity (%) Figure 4 . 4 Relation between measured optical properties and suspended sediment concentrations at the surface of Lillooet Lake: a)  Per cent transmission of incandescent light in a 0.10 m path length plotted against Secchi depth observations.  b) and c)  Relation between Secchi depth (b), transmissivity (c)  and measured suspended sediment concentrations.  90  85  As density of lake water is partially dependent on suspended sediment, it could not be calculated accurately.  The maximum differences between densities in  Lillooet Lake and Lillooet River water due to suspended sediment were of the order -5 -1 300 x 10 g ml , approximately 50 times the density difference due to temperature between 7 and 8 ° C for example. -5 10x10  Bell (1947) reported that density differences of  -1 g ml  were sufficient for the maintenance of stratified flow.  Thus, errors  involved in an attempt to estimate concentrations (perhaps by modifying the relations of figure 4.4) in order to calculate density of lake water would be too great to warrant the attempt. 4.2.  River water and the response of temperature and turbidity characteristics of  Lillooet Lake water 9  3  The volume of water in North Lillooet Lake is 1 .49 x 10 metric measurement of figure 3.4).  m  (based on plani-  Mean inflow from Lillooet-Green River, Birkenhead  River, and the unmetered streams flowing into the lake (the latter estimated) are respectively 5.48 x 1 0 ° ,  0.747 x 1 0 ° , and 0.64 x 10° m^ year  ^ (total, 6.86 x 10°  3 - 1 m  year  ) which is sufficient to completely change North Lillooet Lake water 4 . 7 times  per year on the average.  Thus it would be expected that the lake water temperature  and suspended sediment concentration would reflect temporal patterns in inflow.  Since  Lillooet-Green River represents approximately 9 8 % of the fine grained sediment inflow (section 2 . 5 . 4 . ) and 8 6 % of the water inflow (section 2 . 3 . ) , its sediment and temperature patterns are compared with those observed in the lake. 4.2.1.  Formation and breakup of thermal structure in the lake:  A l l results indicate  that the thermocline in Lillooet Lake also represents a sediment (and probably water) mass transfer barrier, and thus the thermal structure, or lack of it, may be very significant in controlling the nature of the sediments on the lake floor.  Hutchinson  (1957, p. 427 ff), for example, discusses the annual thermal pattern of dimictic lakes  86  in the mid latitudes of the northern hemisphere.  The tripartite structure of summer  (epilimnion, thermocline or metalimnion, and hypolimnion) with its characteristic two counter circulation c e l l s , breaks up in autumn before the formation of a reverse thermal structure in winter.  During the near isothermal conditions of spring and autumn,  general overturn of the lake water is common. Lillooet Lake followed this pattern approximately, except for the disrupting effect on the summer pattern of the entry of Lillooet River water.  The sharpness and  depth of the thermocline at the end of the melt season is a function of the preceding inflow pattern and varies from year to year.  However, during the two years for which  records are available (figure 4.5 a and c ) , the thermal structure did not substantially change until mid to late O c t o b e r .  After the lake became nearly isothermal by late  November, bottom water cooled more or less at the same rate as near surface water until warming began in April of the year following (figure 4.5 b and d).  It is sus-  pected that the long narrow planimetric shape and northwest-southeast orientation of North Lillooet Lake, combined with the predominantly southeast winds of winter cyclonic storms permit a single, persistent circulation c e l l to be set up in winter, mixing and cooling the water nearly uniformly. Extensive ice formation on the lake surface during the winter is unusual. A local resident, M r . G . Richardson, reports that, between 1962 and 1973, only o n c e , in 1968-69 did the lake freeze over completely.  Ice thickness in North Lillooet Lake  was of the order of 50 mm, while in South Lillooet Lake which is much shallower, at least 200 - 300 mm of ice formed.  Near shore ice fringes are common in most winters  however. During the near isothermal conditions of winter the sediment content of the lake water responds much as water temperature.  In the winter of 1971-72, records were  kept irregularly of lake water transmissivity as plotted in figure 4 . 6 .  Following the  inflow peak of August 21-22, 1971 (figure 4 . 1 ) , lake water began to clear of suspended sediment.  After the lake water became nearly isothermal by October 30 (figure 4.5 c ) ,  87  Temperature ( ° C )  Figure 4.5 Bathythermographs at raft 1 showing formation in spring and breakup in fall of the thermal structure of Lillooet Lake. a)  autumn and winter 1970-71  b)  spring 1971  c)  autumn and winter 1971-72  d)  spring 1972  Figure 4.6  Transmissivity at raft 1 during the winter 1971-1972.  89 sediment concentrations also became constant with depth, supporting the hypothesis of general circulation of lake water proposed above. The relation between transmissivity and concentration for the lake surface water recorded in figure 4 . 4 c can be applied reasonably to the entire lake in winter when inflow of coarser grained sediment does not occur. mean concentration at raft 1 was approximately 40 mg I \ 1972 to 5 mg I \  During September 1971, decreasing by M a y 12,  If a l l of the difference were deposited at raft 1, then 1.4 mm of  winter layer (assuming bulk density of 1.7 g ml ^ - section 5.3.) would be seen in the sediment.  This is approximately what is observed in cores from many locations on  the lake bottom (section 5 . 4 . ) .  Further from the d e l t a , greater depth is offset by  lower sediment concentrations in the hypolimnion (figure 4.3);  thus the thickness of  the winter deposit might be expected to be much the same as at raft 1. With the onset of the melt season, the thermal structure begins to reform.  In  1970, a year of light inflow, a strong thermocline developed by mid June at about 40 m depth (figure 4 . 7 ) , and maintained itself relatively undisturbed by river inflow. As the melt season progressed, the thermal gradient became greater and the thermocline moved slightly downward in the lake (figure 4.7).  The same applies to  much of 1972. 4.2.2.  Lake water conditions during the melt season:  Temperature and sediment  content of Lillooet Lake respond rapidly to inflow beginning with the first major nival flood of each year.  For example, when inflow reached a crest on May 2 1 , 1972  (figure 2.25) suspended sediment concentration rose to 800 mg I ^.  Thermal structure  in the lake was just beginning to reform and the sediment content was the lowest of the year (figure 4.2 b).  By May 22 two distinct zones of turbid inflow can be  recognized: one of interflow centered 25 - 30 m below the surface and one of underflow  Figure 4 . 7 (on page following) Bathythermographs in Lillooet Lake, 1970. Observations on August 14, 31 and September 10-12 were made at a number of points in North Lillooet Lake.  91 in the lower 10 m of the hypolimnion. The inflow events of M a y 29 and June 9 , each larger than the one previous, had progressively greater effects, increasing both turbidity and the depth range through which the effect is seen.  However, as sediment  accumulates in the lake water, only the large inflow events such as those of July 12, 1972, and August 3-7, 1971 (figure 2.25) caused as large and widespread fluctuations as lesser events earlier in the melt season (figure 4.1 and 4 . 2 ) . The response of the lake, particularly the epilimnion, to inflow is rapid. Figure 4.8 shows that transmissivity at both rafts 1 and 2 followed faithfully the changes in inflow, and that fluctuations lagged slightly (variable periods up to 4-5 days) behind inflow, but that they corresponded very closely between rafts.  That is, apparently the  processes of dispersal into the lake take up to several days, but movement through the north section of Lillooet Lake is rapid thereafter.  The greater response earlier in the  melt season is partly a function of sediment content and thus density of the lake water and partly of the non-linearity (decreasing sensitivity at higher concentrations) of the transmissivity-concentration relation. During several periods of high inflow the thermal structure of the lake was disrupted.  Usually following a period of strong interflow in the epilimnion near the  delta, the thermocline was depressed a n d , as major underflow occurred, the hypolimnion was attacked from above and below and nearly destroyed (figure 4 . 9 a and b). to ten days thermal structure became re-established.  In one  The cold water necessary to  replenish the hypolimnion was most likely transported from the distal portion of North Lillooet Lake and displaced the warmer water upward.  Note for example, that the  event of August 4, 1972,(figure 4 . 9 c and 4.2) at raft 1 was not seen at raft 2 (figure 4.3). This may help explain the more rapid recovery of the hypolimnion than was observed in 1971. The August 4, 1972, event is anomalous in that, although it disrupted the lake water conditions at raft 1 for a short time - the only such major event in 1972 - it was not caused by a peak of inflow (figure 2.25).  Thus it possibly represented an internal  Z6  Temperature (°C)  Figure 4.9 Bathythermographs showing the destruction and reforming of the thermal structure of Lillooet Lake associated with three periods of strong underflow and interflow. Numbers refer to dates of the observations in July and August 1971 and 1972.  94  disturbance in the lake such as a major slump generated turbidity current (cf. section 4.3.). In both 1971 and 1972, although the epilimnion became turbid with inflowing sediment, the hypolimnion usually remained clearer.  This is particularly true farther  from the delta at raft 2 where very high transmissivities were recorded all through the melt season (figure 4.3 b). exceed 10 mg I \  Concentration of suspended sediment probably did not  A similar, but less pronounced condition was observed at raft 1.  Even in the unusually high inflow of August 3-7, 1971  (figure 2.25), the pattern is  seen (figure 4.1 b) as three water samples taken at 14:00 P . S . T . , August 4, 1971, at depths of 2 1 , 4 6 , and 75 m had concentrations of suspended sediment of 228, 122, and 49 mg I ^ respectively. From figures 4.1 - 4.3 and 4 . 8 it can be seen that the lake gradually clears of sediment after an inflow event but that the rate of clearing is usually not rapid enough to prevent accumulation of sediment in the lake water throughout the melt season.  This loss of suspended sediment occurs by  a) vertical transfer of sediment  to the lake floor, and b) by flush out to South Lillooet Lake and into Lillooet River beyond. a)  Figure 4.10 shows the settling times in still water of grains of various  diameter according to Stokes Law. 1  Mean summer temperatures vertically from  surface to bottom were used to calculate viscosity used in Stokes' equation.  The  difference in settling time between coarse and fine grains is made greater in that the fine grains are more susceptible to mixing currents and turbulence in the epilimnion. G i v e n that the water of the hypolimnion was generally less turbid than that of the epilimnion, it is suspected that the coarser suspended sediment (say, greater than 5 to ]0jJ  fall diameter) settles out quickly close to the delta while much of the very  fine sediment remains in the epilimnion to be flushed out to South Lillooet Lake and beyond.  95  Time (days) 0  2  4  6  8  10  12  14  16  Time (hours)  Figure 4.10 Settling times for particles of density 2.73 g m l " (mean material density of bottom sediments - Appendix 5.4) and equivalent fall diameters of 1 to 20 microns in still water with temperatures as recorded in Lillooet Lake.  96 b)  During the melt season a counter clockwise circulation sometimes occurs  in addition to a vertical circulation that a number of writers have reported (Bell, 1942;  Howard, 1953) associated with underflow.  This surface circulation brings  clearer, warmer water from South Lillooet Lake up the northeast side of the lake and sends more turbid, cooler water down the southwest side.  Observations on July 2 3 ,  1971, (figure 4.11) during high inflow from Lillooet-Green River indicate that transmissivity is nearly uniform at any depth above the thermocline throughout North Lillooet Lake;  that is, much of the sediment that moves into North Lillooet Lake  suspended in interflow passes through.  O n l y close to the boundary with the clear  water, where substantial mixing occurs, does the pattern change. Suspended sediment concentration in the upper few metres of the lake is often very much different than throughout the rest of the epilimnion.  The low concentration  of July 2 3 , 1971, at raft 1 (figure 4.1 b) for example, is probably associated with the counter clockwise circulation described above.  Further, a very strong thermal  gradient (and thus density gradient) prevails through most of the summer in these few metres (see for examples figures 4 . 8 and 4 . 9 ) .  This would form an even more effective  barrier than the thermocline preventing vertical transfer of sediment upward.  4 . 3 . Underflow Since the work of de Geer (1912), many workers have emphasized the role of turbidity currents in varve formation (Kuenen, 1951; Lajtai, 1967; Agterberg and Banerjee,  1969; Ashley, in press).  Although the findings of the previous section  suggest that interflow above the thermocline may be of equal or greater significance in determining the amount of sediment transported through the lake, underflow is no doubt more easily recognizable in the stratigraphic record.  In order to monitor  underflow events, water temperature was recorded continuously from July 11 to September 14, 1971, approximately 1 to 2 m from the lake floor at raft 1 .  Since the  probe was suspended from the surface, its distance from the bottom changed as the  Figure 4.11  a)  Pattern of surface water movement July 2 3 , 1971, south end of North Lillooet Lake,  b)  transmissivity profiles at raft 1 and three sites shown in a).  98 raft swung at anchor and as the lake level rose and fell (figure 2.27).  In place of  water velocity measure, or density or sediment concentration data, these records provide a measure of the timing of underflow events. Bottom temperatures and their fluctuations are a function of: 1)  currents within the hypolimnion generated by drag at the thermocline of  epilimnion currents, or currents due to circulation of water of changing density within the hypolimnion, 2) bottom.  turbidity or density currents bringing warmer river water along the lake  The magnitude of the temperature fluctuations from this cause depends on a  number of factors: a)  the initial temperature of the river water,  b) density difference from the surrounding water, velocity and temperature which determine the stability of the underflow according to Keulegan's (1949) criteria and thus the amount of mixing and cooling of the flow, c)  the rate of recovery of the lake water after the underflow event.  C l e a r l y , if residual water remains warm, the next event will appear much less pronounced on the temperature record, and d)  the response time of the thermistor probe and recorder.  Mean hourly temperatures, calculated from hourly maxima and minima, are plotted in figure 4 . 1 2 .  Although these means only approximate the hourly integrated  mean and hide short term fluctuations that characterize the underflow, examination of their plot is instructive.  Comparison with sediment concentration and density data  for Lillooet-Green River (figures 2.25 and 2.26) indicates that as an approximate 'rule of thumb , concentrations greater than 1000 mg I ^ and densities greater than 1  1 .0003 or 1 .0004 g ml ^ always produced underflow.  However, underflow occurred  at lower values as w e l l , and minor fluctuations occurred throughout the summer period.  Figure 4.12 Mean hourly water temperature approximately 1 - 2 m from the lake bottom at raft 1, and spot readings of temperature in Lillooet-Green River at the delta, July 11 to September 14, 1971 [  100  The first major period of underflow recorded was that of July 19 to 21 and probably 2 2 .  This was a response to the rapid increase of inflow on J u l y 18 and 19  after a 25 day period of below average inflow (figure 2.25 and 2.26).  In figure 4.1  note that the concentration of sediment throughout the lake water at raft 1 prior to July 20 was low, that the temperature of the lake was changing very slowly, and that interflow above the thermocline was pronounced on the 19th and days following. Figure 4.13 is a tracing of temperature records for selected days.  The event July 2 1 ,  although of shorter duration, appears to have been much more stable or of greater volume than those on the 19th and 20th.  A considerable portion of the inflow,  particularly early o n , was being drawn off as interflow. Many of the short duration events occur in the morning sometime after the period of highest inflow (figure 2.15).  Recovery of cold temperature following some  underflow events is rapid indicating that flow is not widespread across or down the delta front, so that cold hypolimnion water is close at hand and can spread back into the path of the current q u i c k l y .  Alternately, as the sediment is deposited from the  warm river water, its density may decrease to less than that of the colder water above so that this cooler water may displace it.  Whichever replacement process occurs  after the underflow, mixing may occur as the resulting temperature is sometimes warmer than that previous to the event (figure 4.12). Small, irregular fluctuations occurred between July 23 and 2 6 , a period of slightly lower inflow.  While underflow stopped, the hypolimnion was still being  charged with river sediment (figure 4.2 b).  In the following days, as inflow increased,  two periods of underflow were observed from July 27 to 30 and from August 1 to 2. The earlier consisted of short duration, single events lasting several hours, while the second was made of long, irregular events (figure 4.13).  From August 3 to 8,  although Lillooet Green discharge was falling rapidly, unusually high sediment Figure 4.13 (on page following) Traces of lake bottom temperature records on selected days in 1971. Breaks in the record of approximately 15 minutes occur when the probe was being used to record surface to bottom temperature profiles.  (Do)  9Jn|Djaduj8x  102  concentrations were recorded (figure 2.23) and underflow became quasi-continuous, although daily peaks within the period can be seen (figure 4.12).  During the height  of activity on August 4 to 7, underflow temperature was greater than that of the inflowing river.  During this time lake surface temperature ranged from 16 to 2 3 ° C ,  and surface water movement toward the delta maintained a large, nearly stationary debris l i n e .  It is likely that warm surface water mixed with the underflow downward  from the plunge point resulting in the higher temperatures. Following the decrease in the unusually high sediment inflow by August 8, short duration, daily underflows were observed on August 9 to 13, 16 and 18 to 19, although small fluctuations that may have represented underflow occurred throughout this period.  These events are recorded despite low inflow sediment concentrations,  and with only a few exceptions they occurred in the morning, even though only small fluctuations in inflow were observed in this period. The strong quasi-continuous event of August 21 and 23 was a response to the brief period of increased inflow.  Several periods of inflow greater than the usual  low values of late August and September may have produced weak, short duration events as well (figure 4.13). It has been assumed that underflow results from plunging river water.  However,  a number of short duration events may represent underflow generated by minor slumps on the classical foreset beds following the daily maximum of bedload input.  This  explanation is especially attractive during periods when underflow was recorded daily but inflow was lower than larger events that had not produced underflow.  The  major event of August 1972 discussed in the previous section is a particularly noteworthy example.  103 4.4.  Observations of slumping on Lillooet Delta Although the morphometry of the floor of Lillooet Lake suggests that slumping  is very significant in redistribution of surficial sediments throughout the lake and particularly on the delta front to 120 metres depth (section 3 . 2 . ) , attempts to monitor slump events directly were unsuccessful.  Echo sounding repeated at intervals through-  out 1971 and 1972 (figure 3.7), showed no distinguishable difference in the placing of slump mounds.  Thus, although minor avalanching of material in the limited zone of  classical foreset beds probably was common and may have been quasi-continuous during periods of high inflow or flush-out (see a brief review of the work of A l l e n and Jopling in Church and G i l b e r t , in press), major slumps are probably infrequent.  Time (P.S.T. Aug. 5, 1971) Figure 4.14 Evidence of a slump on the foreset slopes of Lillooet Delta, August 5, 1971: a)  lake water level record showing large wave at 8:10 P . S . T . ,  b)  tracing of lake bottom temperature record at raft 1 showing warm water pulse between 8:30 and 8:45 P . S . T .  However, indirect evidence does exist fora slump on August 5, 1971, the day of highest observed concentration of inflowing sediment from Lillooet River (figure 2.23). Figure 4.14 illustrates that, at approximately 08:10 - 08:20 P . S . T . , a major water  104 level disturbance was recorded on the lake although the surface was otherwise c a l m . A large wave was heard breaking on the southwest shore at about this time.  Short  period wind waves of measured amplitude 1.1m appear 28 to 30 mm high on the water level trace due to the damping effect of the small openings in the recorder well.  If the wave(s) that caused the disturbance of August 5 when the lake surface  was undisturbed by wind was of similar period, then its amplitude was of the order of 1.3m.  N o disturbances were observed on the traces of Green or Lillooet River stage,  indicating that the lake wave was not caused by a surge of inflowing water. Shortly thereafter (between 08:30 and 08:45 P . S . T . ) , lake water temperature approximately 1 m from the lake bottom at raft 1 rose by approximately 1 .4 C ° (figure 4.14 b).  Although fluctuations of this order or greater occurred throughout  this period of heavy sediment inflow (section 4 . 2 . 4 . ) , several factors make the event noteworthy: 1)  it is approximately twice the amplitude of events for several hours before  and after, 2)  it occurred shortly after the initial surface disturbance,  3)  it represents a sustained period of higher temperature sharply distinguished  from temperature before and after.  Other fluctuations were usually of short duration  (section 4 . 2 . 4 . ) . This circumstantial evidence leads to the suggestion that a major movement of sediment may have occurred on the morning of August 5, 1971, during high sediment inflow.  4.5.  Summary and discussion The capacity-inflow relation (Brune, 1953) of Lillooet Lake is 0.218 (annual)  and 0.43 (during the melt season).  Thus the rapid fluctuations of the large inflow of  sediment, nearly all of which is from Lillooet-Green River, dominate the sedimentary environment of Lillooet Lake.  105 Except during maximum flood, when concentration of inflowing suspended sediment exceeds 1500 mg I ^, the finest sediment passes into the lake as interflow at various depths in the epilimnion, where turbulence is sufficiently great and currents sufficiently strong to transport a high percentage of it out of North Lillooet Lake. Evidence for the latter observation is that a)  although the lag between inflow at the delta and observation at raft 1,  one kilometre away, was up to 4 - 5 days, lag between rafts 1 and 2 was seldom greater than 1 day (figure 4.8). b)  one set of observations indicates that the sediment content of the  epilimnion remains relatively unchanged even to the south end of North Lillooet Lake (figure 4.11), c)  during calm conditions, counter clockwise surface currents of velocity up  to 0.2 m s ^ (based on the rate of floating debris movement) have been observed in North Lillooet Lake but are most pronounced at the south end (figure 4.11) where 'clear' and turbid water meet, d)  the hypolimnion was nearly always clearer than the water of the  epilimnion above (see especially figure 4.3 b) indicating that settling times of fine grains are too long and turbulence too high to allow most, except the coarser sediment (say 10 - 20y(j) to fall through to the lake floor. A number of authors (Kuenen, 1951; A s h l e y , in press; Gustavson, in press) have suggested that, unlike nonglacial streams (Bell, 1947), the content of suspended sediment in glacial rivers is so great as to far overshadow density differences caused by different temperature between river and lake water or within the lake.  Although  Lillooet Lake is not p r o g l a c i a l , its thermal structure is such that distribution of sediment is strongly modified particularly during low and medium flow.  Thus the  stratigraphy and rate of accumulation of the bottom sediments may be more controlled by thermal structure in all but ice contact water bodies than workers, particularly those studying ancient sediments, have indicated.  106 During winter, on the other hand, the lake becomes nearly isothermal. Suspended sediment concentration also becomes uniform throughout the lake and decreases until onset of the next melt season.  Observations in one year indicate that  the amount of sediment lost from the lake water between the end of one melt season and the beginning of the next is approximately equal to the amount deposited in the dark winter layers of varves throughout the lake.  Thus, probably a lower per cent of  sediment is transported out of the lake in winter.  A single, large, slow moving  circulation c e l l , driven by persistent cyclonic winds, may be maintained during these winter conditions. During peak inflow, both interflow and underflow o c c u r , both sometimes sufficiently powerful to destroy temporarily the thermal structure of the lake near the delta.  In 1971 fluctuations in temperature 1 - 2 metres from the lake bottom at raft 1,  indicating the passage of significant quantities of warmer water, occurred 37 per cent of the time between July 11 and September 14.  Substantial periods of underflow such  as those of August 3-7 and August 20-22, 1971, were responses to high inflow. However, some short bursts of warm water (for examples, those of August 9-13) could not be related to inflow and may have been generated by minor slumps on the classical foreset beds. However, only one major slump was thought to have occurred in three years of observation on August 5, 1971, at the peak of sediment inflow.  107  5.  Sediment of Lillooet Lake Between 1970 and 1972 samples were collected in Lillooet Lake with an Ekman  grab sampler and a gravity coring device (Appendix 5.1) to determine the spatial variation of texture, mass properties, strength, and stratigraphy of the sediments. These characteristics are related to the inflow of water and sediment (Chapter 2), the morphometry of the lake floor (Chapter 3), and the patterns of sediment distribution in the lake water (Chapter 4), in the following sections.  5.1.  Analysis of texture of lake bottom sediments In June 1970, 50 Ekman samples were taken in the northwest part of the lake  (figure 5.1).  They were thoroughly mixed so that winter and summer deposits would be  included in hydrometer and sieve analysis.  Standard procedure was followed ( e . g .  A S T M , 1963) except that a deflocculating agent was not used.  However, it was  necessary to use distilled water in the analysis resulting in a small underestimate of grain size (section 2 . 5 . 1 . ) .  Two splits from each sample were run, the results combined,  and geometric moment measures c a l c u l a t e d , as summarized in figures 5.2 and 5.3 and Appendix 5 . 3 .  Comparison of the ogives of figure 2.28 and figure 5.2 (see also f i g u r e 6 ,  G i l b e r t , 1972) shows that virtually all particles smaller than sand size (62 swept out of the river channels in summer (Appendix 2.3).  ) are  However, little of the sand  is transported far from the delta (figure 5.3 e), suggesting that turbidity currents, although frequent (section 4 . 2 . 4 . ) are not particularly powerful.  Compare figures 5.2  and 5.3 e with figures 5 and 12 in Houbolt and Jonker (1968) which show 'mainly sand transported by turbidity currents to the deepest part of the lake.  1  O n the other hand,  the texture of delta front sediments in Lillooet Lake is coarser than Lake Mead turbidites (see for example, figure 8, G o u l d , 1951). Figure 5.2 (on page following figure 5.1) Lillooet Lake surficial bottom sediments.  Cumulative frequency distributions of  Figure 5.3 (on pages following) Distribution of moment measures in Lillooet Lake sediments: a) geometric mean in microns, b) geometric variance in microns (c) skewness, and (d) kurtosis, and (e) per cent sand.  108  Figure 5.1 Location of Ekman grab samples in Lillooet Lake for preliminary textural analysis.  601  110  111  I 13  kilometres  114  115 The morphometric zones described in section 3.2 can be divided on the basis of texture as w e l l , from the foresets as represented by samples 42 - 50 with mean grain size approximately 50 - 100 JJ,  to the slumped sediment zones (samples 22 - 4 0 ,  mean grain size 20 - 40//), to the area beyond the slumps with mean grain size less than 10 - 1 5 ^ (except off the mouth of Ure Creek).  Coarser grained sediments extend  farther into the lake on the northeast than on the southwest side although the main inflow occurs to the southwest.  Perhaps the bedrock spur that extends underwater from  the two small islands (figure 3.4) deflects underflow across the lake. V a r i a n c e , although a less strong indicator of sedimentation processes in Lillooet Lake, shows the samples of the delta surface and classical foreset beds are better sorted than more distal sediments where slumping and turbidity currents bring coarser sediments to mix with the fines deposited by settling. variance decreases.  Beyond the slump zone  Kurtosis also indicates this selective sorting (figure 5.4 d).  All  the lake bottom samples are negatively skewed, whereas the fluvial sediments are positively skewed.  Skewness decreases outward from the delta (figure 5.3 c) again  showing the influence of the coarser grains in the sediment close to the delta;  the  pattern of skewness matches approximately that for per cent sand shown in figure 5.3 e. The location of the mouths of the distributaries of Lillooet-Green River and of Birkenhead and Ure Creeks is reflected in the moment measures; sediment are coarser grained and better sorted in these higher energy environments.  This supports  morphometric evidence that turbidity currents do not occupy lake bottom channels as others have reported (Houbolt and Jonker, 1968; Menard and Ludwick, 1951) but spread unequally and intermittently across the lake bottom. Scattergrams of moment measures plotted in figure 5.4 summarize the above results: the finer sediments farther from the delta are generally less well sorted (figure 5.4 a and c) and less skewed to coarser grains (b).  Although the second through  fourth moments show some correlation, note that they are not independent of each other as standard deviation is used in the calculation of skew and kurtosis to make them non-dimensional.  116 Figure 5.4 Geometric mean (M) and geometric variance (V) in microns: skewness (S) and kurtosis (K) (normal distribution, 3.0) of logarithmic transformed results of textural analysis, Lillooet Lake, 1970.  117 5.2.  Strength of Lillooet Lake sediments and implications for slumping Using the quick direct shear apparatus described in Appendix 5 . 4 , unconsoli-  dated shear tests were performed on 17 cores at several points along their length. results are summarized in figure 5.5 and table A . 5 . 2 .  The  Most cores showed an increase  in 'undisturbed' and remolded strength with depth at least to 1 metre depth.  An  exception to this pattern was markedly lower strength in some cores at about 0.6 m depth where a large sand layer occurred, detectable only in dried cores (section 5.4.2.).  Sensitivity varied from 'low' (1 - 4) to 'sensitive' (4 - 8) (Terzaghi, 1955)  but no extra sensitive or quick (>16) sediments were found indicating that spontaneous liquefaction is not a major contributor to sediment movement. In order to assess the strength of sediment in terms of the shear forces generated by gravity on the sloping lake floor, the results were pooled and plotted in figure 5 . 6 . Simple regression of depth of burial against measured shear strength yields, for the 2  undisturbed samples  t = 2.74 + 4 . 6 9 Z  t - 8.92 + 1 3 . 4 Z with r = 0.60 and for the remolded sediments 2 -2 with r = 0.51 where r is the shear stress in g cm and Z is the  depth of burial in metres.  The considerable scatter is the result of combining  samples from different locations on the lake bottom and of the different texture and composition of the sediments in the column, as well as the measuring inaccuracy of the very simple instrument. A number of workers (Taylor, 1948; M o o r e , 1961)  where f- is the slope of the potential failure plane  stress due to gravity is £ = P^sin*  (assumed to be equal to that of the bottom) and P weight of the overlying sediment:  have shown that the shear  P  y  y  = ^-{p ~ y ° ) %  w  is the vertical pressure due to the c  o  s  * • This relation between  and Z is plotted in figure 5.6 assuming sediment density of 2.73 g ml ^ (Appendix 5.4) for various slopes from 1 ° to 2 0 ° .  N o w , if strength increases with depth faster than  shear stress, failure will never occur no matter how much sediment is deposited: Figure 5.5 sediments.  (on page following)  Shear strength and sensitivity of Lillooet Lake  118  ^5  0  0  »  i  •  40A  • L_  •  L  «  20  •  \  j.  h  IO  •  /  •  r  Sen s i tivity 0 5 I  \  _1_ L  30 A"  I  • I  r  \  I  r  ,  \  -  \  TT  r r r - r \  \ L  i  J  |— \  -  80A' \  "  '  U  V  I  •  J_ L 60B*  30A'O  O  ti  U  /  \  •  »  p  \: •  60 C'  _L L  / 70 B  I  r r\ / -L L  •  ' \  T  «  80B' L  50B'  60B'O  O 50A'  60 A '  1  -/80C I \ ' i I  Location of cores  O 40B'  \  50C' , r  •  I  \  \.  \ \  L  •  \  L  \  30 C'  T n\— *\  f  \  •  V  ,1  /  L  "I  r~5  \  I  ' \  I  /  /  r  J-  / X  20 r  \ \  30B  508' \  L  10 t  \  \ UNDISTURB  ' \  L_  \  \  5^0  •^400 *  60A  ,  0  \^EMOLDElK^  •  11 ' \ \ ^ .  2  J  \  x . L ^ L  1  strenqth (g cm" ) 10 20  0  \  h i  /  Shear  O, 70B  O  O 80B' O 80A"  \ \ \  \ II  1  \  \ X  j  119  0  0.2  0.4  0.6  0.8  1.0  Depth below sediment surface (m) Figure 5.6 Relation between depth of burial and undisturbed and remolded shear strength of pooled samples from Lillooet Lake. Lines 'IT and 'R' refer to the best fit linear regression relation for undisturbed ( + ) and remolded ( • ) strength. Relation between shear stress due to gravity and depth of burial shown for slopes from 1 ° to 2 0 ° .  120 that i s , the stress lines plotted i n figure 5.6 whose slopes are lower than-the-regression lines represent absolutely stable bottom slopes (within the scatter of the plots), while those with higher slopes intersect the regression lines at the point (depth) at which failure must o c c u r .  The c r i t i c a l bottom slopes (stress line p a r a l l e l to regression line)  are for undisturbed sediment 3°'28' and for remolded sediment 0 ° 5 3 ' . greater than these values are shown in figure 5 . 7 .  Bottom slopes  It is emphasized that these  c r i t i c a l values are approximate: 1)  the simple analysis assumes a large (infinite) extent of bottom sediment  and that no forces are acting except shear stress due to the down slope component of pressure, 2)  only the regression mean v a l u e , and not the effect of strong or weak  ( e . g . sandy) layers, is considered, 3)  only ' c r i t i c a l ' slopes are c a l c u l a t e d ; values slightly above c r i t i c a l w i l l  a l l o w considerable accumulation before failure occurs, 4)  given the rates of deposition observed the sediment is no doubt under-  consolidated and pore water pressure is greater than hydrostatic, making the sediment weaker (Terzaghi, 1956). The average degree of consolidation of Lillooet Lake prodelta sediments is probably of the order 80 - 9 0 % (Morgenstern, 1967).  5.3.  Mass properties of Lillooet Lake sediments Water content v/as determined at various depths in each of the cores shown  in figure 5 . 5 .  The pooled results are plotted as figure 5 . 8 .  A series of material  density measurements by pycnometer from sediments at various locations on the lake floor y i e l d e d a mean value of 2 . 7 3 g ml ^ with variance 2.91 x 10 ^ (Appendix 5.4) (compare with density of the igneous rock of Lillooet Basin reported by Roddick and Hutchinson, 1973, section 2 . 4 . 2 . ) .  This mean was applied as a constant with water  content to calculate v o i d r a t i o , porosity, and bulk density according to the equations of Appendix 5.4 (see for example T e r z a g h i , 1955). The results are listed in table A . 5 . 2 .  Figure 5.7  Map showing bottom slopes in excess of those calculated to be stable.  122  2.4 I—  2.0  1.6  Bulk density  T  1.2 —I  content  Figure 5.8  Mass properties of Lillooet Lake sediments,  123  and shown in figure 5 . 8 .  Compare these with bulk densities of approximately 1 .2 g  ml ^ in 'recently deposited sediment from turbidity currents and up to 1 .35 g ml ^ in 1  'compact sediment' beneath (burial depth approximately 0.5 m) in Lake Mead ( G o u l d , 1951, figure 5). Accurate assessment of consolidation of the sediments is difficult in that no consolidation tests were performed.  In a complex environment such as Lillooet Lake,  compressibility of sediment is complicated by a number of factors (cf. Terzaghi, 1955; Mathews and Shepard, 1962): 1)  the time since the sediment was deposited,  2)  whether the sediment is normally consolidated or preconsolidated (applies  to sediments in the slump zone where overburden has been removed), 3)  whether the sediment is undisturbed or remolded and its sensitivity,  4)  grain size and grain size distribution,  5)  shape of the particles (for example, mica plates which significantly  increase compressibility), 6)  material composition of the sediment (especially clay minerals),  7)  liquid and plastic limit of the sediment, and  8)  the presence or absence of gas bubbles in the sample at atmospheric  pressure. However, from the data of figure 5 . 8 , an approximation of the amount of compaction in the upper metre of sediment can be attempted.  The linear relation  between water content (w) and depth of burial in metres (Z) is w = 0.667 - 0.225 Z 2 with r  = 0.64 for O ^ Z < 1.  Scatter in the data does not warrant application of a  more sophisticated non-linear relation (Terzaghi, 1941).  From this equation the ratio  between water content (or void ratio assuming the density of water, yO depth Z and the water content at the sediment surface is given by  = 1 .0) at  124  w w Now, W  Thus,  o W  z  so  t /t  3.0  so  =  W  wz  sz  / W <sz  and V  so  (3.0-Z) =V  s z  (ewuming O /  so  = O /  s z  )  = 3.0/(3.0 - Z) for 0 * Z < 1 where W is the weight and V the volume of  sediment ( ) or water {^), g  t is the thickness of an element of sediment (for example,  a varve), and the subscripts o and z refer to the sediment surface (Z = 0) and depth Z = z metres.  For example,this approximation states that a varve at 1 metre depth is  compressed to 0.67 times its thickness when deposited. As the scatter in figure 5.8 indicates, and given the above list of factors affecting compressibility, this value w i l l vary from place to place in the lake.  When the correction was applied to varve  thicknesses from a number of locations in the lake (section 5 . 4 . 1 . ) , the relation between varve thickness and inflow as expressed by the coefficient of determination 2. (r^) was improved by a negligible amount. Nothing is known of consolidation at depths greater than one metre.  Void  ratio decreases according to the logarithm of applied pressure (Terzaghi, 1941) and therefore depth.  Borehole data from Fraser River Delta (Mathews and Shepard, 1962)  shows slight if any decrease in void ratio with depth between 20 and 100 m burial depth, although there is considerable scatter (their figure 12).  Although a value for  the compression index of 0.5 fits the Lillooet Lake data to 1 m depth approximately, its behaviour at depth may be considerably different even though the sediment only moderatly sensitive (Terzaghi, 1955, p. 568).  is  Thus in lieu of specific data, it  may be reasonable to assume that below about 1 .5 m depth, void ratio decreases slowly and irregularly from about 0.8 to 1 .0 to values of 0.5 to 0 . 8 at 60 to 80 m depth, the practical limit of resolution of continuous seismic profiling records (section 5 . 5 . ) .  125 5.4.  Stratigraphy of Lillooet Lake sediments Rhythmic layering which has been interpreted as varves appeared in most of  the 28 cores that were obtained for stratigraphic examination (figure 5.9).  However,  a wide variation instructures, both massive and intravarval, has been observed among cores and is discussed in the following paragraphs. 5.4.1.  Varve thickness and relation to inflow:  thickness are summarized in figure 5 . 1 0 .  Results of measurements of varve  A major difficulty in measuring thickness  from samples not in situ is that almost nothing is known of sampling disturbance particularly of differential packing and shrinking of layers with differing mass properties that occur at the time of sampling and as drying processes (see Appendix 5.2).  An  indication of these factors is seen in the penetration and shrinkage ratios given in table 5.1 .  The interpretation of penetration ratio is in dispute and no doubt varies  Table 5.1 Recovery parameters of Lillooet Lake cores Core # 20B 30A 30B 40B 40C 50A 50 B 50C 60A 60B 60C 70A 70B 71 75 80A 80B 80C 81 91 95  Length of dry core (mm) 300  530 530 955 870 705 800 700 745 830 915 810 795 700 705 950 865 950 730 875 810  Penetration ratio*  Settling and shrinkage  0.42  21 .0  0.45 0.39 0.69 0.49 0.70 0.49 0.54 0.52 0.49 0.70 0.53 0.54 0.55 0.51 0.53 0.46 0.48 0.52 0.57 0.51  10.2 12.2 3.5 4.8 12.5 7.4 11.4 11.3 10.8 12.0 14.7 8.0 18.0 17.3 13.7 6.1 15.3 21.0 21.7 18.2  length of core recovered * penetration of corer below the sediment surface * * length loss after drying as per cent of original length recovered.  Figure 5.9  Map showing location of samples for stratigraphic examination.  1970  1960 Figure 5.10 occurred.  1950  1940  1930  1920  1910  1900  1890  Thicknesses of varves from Lillooet Lake plotted against the year in which deposition  128 with the corer and the nature of the sediment (Appendix 5.2).  Thus, nothing is  known of how much compaction occurs on sampling or of the different rates of compaction that occur in sediments of different strength within the stratigraphic column.  Nor is it known if shrinkage is equal (at least between varves) along the  length of the core, as distinguishing layers are not seen until the core is partly dry. A n indication of the effect of the sampler can be obtained by comparing varve thicknesses recorded in gravity cores with those in 'undisturbed' Ekman samples taken at approximately the same location (Table 5.2).  With only 10 data pairs,  meaningful comparison is difficult especially given the scatter in the data of Table 5 . 2 , but it is suggested that disturbance, particularly compaction, is not much, if any, greater near the sediment surface in the gravity cores than in the Ekman samples.  In  the latter, penetration ratio was nearly 1.0 for all samples. Table 5.2 Comparison of varve thickness in Ekman and gravity core samples Ekman sample ^  Gravity core  Year  Ekman  52  50B  1970  19  21  0.90  72  71  1971 1970 1969  36 19 17  19 14 18  1.89 1.36 0.94  82  81  1969 1968 1967  15 18 14  11 15 24  1.36 1.20 0.58  92  91  1970 1969 1968  14 16 17  9 14 16  1.56 1.14 1.06  w  Varve thickness (mm) Gravity Ratio  Mean  1.20  In those samples a v a i l a b l e , it can be seen that mean varve thickness decreases outward from the delta, and towards the sides of the lake from the centre in the slump zone (figure 5.11).  Annual accumulation increases toward the delta u n t i l , in the  zone of bed load deposition on the classical foreset beds, it reaches a maximum of 5 - 8 m yr \  based on the rate of delta advance (section 3 . 1 . ) .  20B  Figure 5.11  50C  Mean varve thickness Lillooet Lake sediments.  130  To the writer's knowledge, only Granar (1956) has related sediment a c c u m u l a tion in varves (thickness) to inflowing water, and thus by implication to inflowing fine sediment.  Varve thickness has been related to recorded climatic variables  (Shostakovich , 1934) and a large literature discusses comparison among varve records and the use of varves in geochronology (see a brief review in Church and G i l b e r t , in press).  Almost 50 years of record of discharge in Lillooet River (section 2.3.)  provide  an excellent data base for comparing sediment accumulation with the rate of inflow. The varve records, with two exceptions - unfortunately the two longest records by far have patterns clearly comparable from core to core, especially the unusually thick varves of 1967, 1961 and 1958 (figure 5.10). The total variance in varve thickness can be separated into that due to variation in inflow and the remaining, unexplained variance, by calculation of the 2 Pearsonian coefficient of determination (r ) in this case, in a simple, linear regression model.  Inflow records for Lillooet River alone were used since none were available  for Green River during much of the period of interest (Table 2.2) and estimation from Lillooet data would not be accurate (Appendix 2.1).  In order to test the assumption  that mean daily inflow less than some value (a reference discharge, Q^) is not significant in the formation of varves, the annual inflow of Lillooet River greater than 3 reference discharges between 0 and 400 m  -1 s  was c a l c u l a t e d .  For example, for  = 200, only the volume of inflow on days with mean daily discharge greater than 3 -1 200 m s was considered in calculation of the annual total.  The total was then  2 used to calculate the coefficient of determination (r ) as shown in figure 5.12. Despite the possible sources of error - disturbance on coring, differential shrinkage, consolidation of sediment as burial depth and time increase, the use of Lillooet water inflow to indicate total sediment inflow, and sediment redeposition from infrequent, irregular, mass movement on the lake floor - the relation between inflow and varve thickness is remarkably good (see for example, the selected relations of figure 5.13).  0  100  200  300  400  0  100  200  300  Reference discharge (m-V ) 1  Figure 5.12 Coefficients of determination (r ) of varve thickness in mm and annual volume of Lillooet River inflow in m in excess of mean daily discharges from 0 to 400 s .  400  132  200 Core 4 0 B r2 = 0.81 100  y  Reference discharge, (Q ) 200 m3 -1 — \ r  / • -g20 E £30  s  Or = 310 m3 -1 s  2.0  0  3.0  2.0  4.0  Core 6IC r2 = 0.07  <u >  40  >  1  Core 91 — 1957  20  1  1  d  1 *  20  10  • • •  • •  3.0  4.0  including 1957 = 0.84 not including 1957  0 = 200 m3 -1 r  Q = 0 m3 s" r  r2 = 0.61  »m '  0 1  s  1  1  2.0  I  30  I  4.0  5.0  Total annual discharge o f Lillooet River (m3  x  1 0 ) greater than mean 9  daily reference discharge (m3 s-1) Figure 5.13 Relation between varve thickness and annual inflow in excess of specified mean daily flow for selected cores from Lillooet Lake.  133 O f the cores in which varves were measured, with the exception of core 9 5 , 2 those closest to and farthest from the delta yielded the highest values of r . 3  For these  -1  cores, volume of inflow greater than mean daily flow of 200 - 230 m s or 280 3-1 2 350 m s , depending on the c o r e , gave the highest values of r . However, values 2 of r  significantly greater at 9 5 % confidence (Snedecor and C o c h r a n , 1967, p. 185 ff.) 2 3-1 than r calculated with total annual inflow (reference discharge 0 m s ) occurred only on the following cores: 3  -1  30A, reference discharges 190 - 200 m s 40B, reference discharges 220 and 280 - 350 m^ s ^ 4 0 C , reference discharges 220 and 280 - 340  s ^  That i s , there is indication that there is a 'varve forming discharge' most significant in sedimentation close to the delta but also appearing in most other thickness-inflow relations except those for cores with very poor correlation with discharge. The first 2 3-1 peak in variance explanation (r ) occurs at about 200 m s mean daily flow; that is 3 -1 annual inflow on days with flow greater than 200 m correlated to varve thickness.  s  is generally most highly  From the recurrence series (figure 2.13) it can be seen  that approximately this discharge separates low flows from the higher flows of the melt 3 season (see also figure 2.12 and 5.14).  Note also that for inflow less than 200 m s  suspended sediment concentration is very low (50 - 100 mg I ^ - figure 2.23) as compared with concentrations up to an order of magnitude greater during the melt season.  This inflow generated underflow is not significant at the lower discharge 3  values (section 4 . 3 . ) .  The reason for the secondary peak at 280 - 350 m  -1 s  is not  clear. In those cores farther from the delta, one year, 1957, is anomalous in an otherwise reasonable relation between varve thickness and discharge (figure 5.13 d). In 1957 summer flow was unusually low but on September 6 the mean daily discharge 3 -1 rose to 716 m s , the second highest ever recorded (figure 5.14).  It is suspected  -1  134 600,  Figure 5.14  Mean daily flow in Lillooer River for selected  135 that sediment deposited on the foreset beds and in the lower reaches of the river during the relatively quiet summer was washed to the deeper parts of the lake in a single exceptionally powerful underflow event to form an anomalously thick varve. Two cores, 60C and 6 1 C , are unlike the others not only in that their varve thickness is poorly correlated with inflow but in a number of other factors as w e l l : 1)  they contain no layers of coarser material with the exception of one  small lamina in the 1948 varve of 6 0 C , 2)  most of the varves are simple dark-light couplets with few intravarval  laminae to distinguish them (figure 5.15 a) unlike the others which are much more complicated, 3)  rate of deposition is lower as varves are generally thinner than in other  cores even farther from the d e l t a , 4)  with the exception of the very recent deformation of the top of 6 1 C , no  other deformation, except minor load casts, can be seen in 88 years of record (section 5 . 4 . 2 . ) . The reasons for this anomaly are unclear. to areas nearby (figures 3.4 and 5.5).  Bottom topography is comparable  The lake floor around 6 0 C and 61C is  apparently protected from underflow and deposition from it. 5.4.2.  Sedimentological events not related to normal varve sedimentation:  Besides  normal varves seen in the core samples, two disrupting sedimentological characteristics, in places related, are also recorded: disturbed, folded or faulted sediments, and relatively thick beds of coarser grained sediment.  Although a small amount of  disturbance must occur at the time of sampling,many layers were preserved with only slight draw-down along the sides (for example, core 6 1 C , figure 5.15 a).  Thus one  can be relatively certain that major disturbances seen in the cores did not occur as the result of sampling procedure. Figure 5.15 (on pages following) Lillooet Lake.  Photographs of selected sections of cores from  136  137  138  138 a  O f a number of sand layers that occur in the record, one is largest and most persistent.  It can be reliably traced from cores 6 0 A , B, and 61 to core 95 but its  thickness and character vary throughout (figure 5.16).  In two cores  (71 and 75) the  varve record can be dated back through this sand layer (figure 5.10) to show that it occurred in 1945 (figure 5.15 c ) .  Nothing in the discharge record of that year was  3 unusual - mean discharge (124 m 3 summer flow reached 500 m  -1 s  ) was very close to the long term mean (126),  -1 s  , and no major inflow occurred before or after the  melt season. Although a minor late winter event gave rise to a double winter dark band (figure 5.15 c , c f . section 5 . 4 . 3 . below).  It was not until 1947 that the large  river training schemes were carried out in Lillooet River V a l l e y (figure 2.20).  Thus  the layer may represent deposition from a slump-triggered turbidity current or from anomalous inflow on one of the unmetered streams, for example, Ure Creek.  The  coarser grain size of the layer in core 60A and deformation of sediments beneath (figure 5.15 e) supports the hypothesis of origin from Ure Creek.  The pattern of  thickness (figure 5.16) suggests that the capacity of the current (or currents) decreased as it reached the nearly flat floor of the lake (core 7 5 , 8 0 A , B, C , and 81), and deposited there most of the load of sediment.  Unfortunately, cores closer to Lillooet  Delta than the 60 line are not sufficiently long to include this b e d . The sand layer in core 75 was divided into six samples each 15 mm thick (figure 5.15 c) for textural analysis by hydrometer.  The results are presented as  figure 5.17 which shows that the deposit is not a fining upward graded bed as might be expected in a turbidite.  Rather, coarser sediment at the top fines downward to a  minimum below mid depth.  G r a i n size increases again toward the bottom of the layer.  The grain size distributions are bimodal with coarse clay size and fine silt ( 3 - 8 j u ) representing 1 - 2 % of the sample weights (cf. figure 5.2) and 8 - 1 4 % finer than 1.5 ju .  This indicates that settling of very fine grains from suspension in the lake  water occurred continuously as deposition of coarser grains was occurring.  Thus the  139  0valu.es  2  5  10  20  50  100  200  Equivalent fall diameter (microns) Figure 5.17 in core 75.  Cumulative frequency distributions from the 1945 layer  140  bed was probably not deposited in a single event, but perhaps by a series of turbidity currents throughout the year.  This agrees with the anomalous grading, with the  observation from core 75 (figure 5.15 c) and others of very little 'normal' varve sedimentation in 1945, and with the observation of some bedding within this large coarser unit (for example, core 80B, figure 5.15 d).  With the exception of a basal  layer (sample 6 , core 75), none of the other beds within the sand layer can be traced from core to core. In some cores the sand layer is deposited on the sediments below with almost no disturbance to them ( e . g . core 80B);  in others a small amount of disturbance  extending a few millimetres into the sediment is seen ( e . g . core 75) while in a few others, disturbance usually in the form of a large fold or folds is recorded (for example, core 6 0 A , figure 5.15 e). In several of the longer cores from the flat lake floor, two other major sand layers can be seen deposited one or two years apart in the mid 1930's.  Lillooet River  was reported to have carried greater than usual loads of sediment from a large land slide or slides upstream during this period, but the relation of this to the coarser layers is unknown.  As only two cores are sufficiently long to record them, they  cannot be traced throughout the lake bottom.  In core 9 5 , folding of the sediments  below is associated with the deposition of the lower layer (figure 5.15 f). many other such layers occur in the stratigraphic record of Lillooet Lake.  Undoubtedly They may  be the reflecting horizons seen in the echograms of figure 3.6 e. In several cores, most notably 40B and C (figure 5.15 g ) , a number of coarse layers are seen in the 1958 varve. on record ( 5 x 1 0  9  In that year Lillooet River inflow was the highest  3 3 -1 m ), but it had remained low until a sudden rise to over 500 m s  beginning on May 6.  Unusually high flow was maintained without relief until  September, after which two major rain-induced events occurred (figure 5.13).  This  high inflow is related to the anomalously thick varve with its beds of coarser sediment.  141  Although the clearly distinguishable beds of coarser material become less distinct farther from the delta, the 1958 varve is thicker than others in the same record to the distal parts of the lake (figure 5.10).  Cores closer to the delta than the 40 line are  not long enough to contain the 1958 record. Small sand layers occur in the varves of a number of other cores (for example, in 1950, 1948, and 1967) and usually can be related to higher inflow in those years (figure 5.13). Besides deformation of the sediments associated with the deposition of superimposed coarser material, folding and disturbance are observed in several cores. Three such examples are from cores 5 0 C , 51C and 3 0 C . A series of very tight folds is observed in the 1956 layer of core 50C (figure 5.15 b).  Deformation occurred very  late in the year for only 5 mm of sediment were deposited on top beneath the dark winter layer.  Slumping of sediment associated with two autumn floods (figure 5.14)  may be responsible although the 1956 varve in other cores is undisturbed. The folded sediment appears to have been moved to the site of core 5 0 C , as a sharp (erosional ?) contact is seen with what appear to be normal summer deposits below. The upper 240 mm of core 60C are badly deformed (figure 5.15 h).  Core 6 1 C ,  taken in August 1971, 150 m from the location of 6 0 C , showed no disturbance at a l l . Either these deformed sediments extend or exert their influence over only short distances, or a slump event occurred in 1972 shortly before sampling took place.  That  normal sedimentation is not seen at the top of the core supports the latter hypothesis. Perhaps an earthquake on July 30, 1972, of magnitude 7.6 with epicentre 5 6 . 8 2 ° N , 1 3 5 . 6 8 ° W which caused a seiche on a number of British Columbia lakes ( N E I C , 1972), including Lillooet Lake, caused local slumping of sediment on the side of the lake near 6 0 C . O f the cores in the prodelta zone of slump deposits, only one, 30C (figure 5.15 i), shows any significant deformation.  In 30A and B varves were recorded back to 1958  and 1963 respectively indicating that the mounds are at least that o l d .  The disturbance  142  in the 1971 varve of 30C may be the result of the slump proposed in section 4 . 4 . Due to the sand content of the proximal sediments only one short core, 20B at raft 1, was successfully obtained.  It showed no disturbance in three years of varve record,  but the large proximal area cannot be represented by this one sample.  Thus the  disturbance in 30C might be large and more significant elsewhere on the prodelta slope. In none of the three most distal cores (figure 5.9) could varves be counted with certainty.  Core 111, from the slope off the mouth of Twin Creek contained, inter-  spaced with layers of fine sediment, a number of coarse sand beds with some gravel sized particles, the largest of which measured 55 mm (b axis).  These are interpreted  as slump or very powerful underflow deposits (although the largest pebbles may have been rafted by ice or floating debris).  However, the extent of the deposits is not  great as no sand layers were observed in core 101 taken from the base of the slope in the deepest part of the lake. None of the characteristics of the sediments could be related to human activities in Lillooet River V a l l e y upstream.  Although the rate of delta advance  increased more than threefold (section 3.1) apparently related to lowering of the lake, increased agriculture and logging, and river training, no unusual events - slumps or layers of coarser sediment - were recorded in the varves of 1947-1952, the years of greatest engineering activity in the basin (section 2 . 2 . 4 . ) .  5.4.3.  Intravarve sedimentation:  O f assistance in dating individual varves are  laminae within them that can be related to individual inflow events.  Examination of  these relations may help in the interpretation of sedimentary environments in similar ancient sediments.  Ekman samples shown in figure 5.18 display most clearly these intravarval  Figure 5.18  (on page following)  Ekman samples from Lillooet Lake.  (a)  Grab 4 2  (b) Grab 6 2  (c) Grab 72  144  144 a laminae.  A striking feature of the varves is that the winter layer often consists of  double or multiple dark bands.  Where the lighter toned, coarser grained lamina(e)  between is thick, an error may result by assigning one year to each dark-light couplet. These multiple winter laminae usually occur in years when late winter warm periods caused brief, usually small, inflow peaks. 3  For example in February and March 1972,  -1  inflow rose to 120 m s  .  This peak and similar peaks in 1968 and 1971 (figure 5.14)  are seen in the varve records as well (see especially sample 82, figure 5.18).  The  longer cores also have double winter layers in nearly all years with late winter inflow peaks.  However, double or multiple laminae also occur in a few years during which  no peaks were recorded.  During most winters lake water density is reduced significantly  when sediment settles out (section 4.2.1 .).  This and fine fluvial sediment accumulates  in the distributaries and on the proximal slopes.  It is suspected that even a small  increase in discharge is sufficient to entrain enough of this sediment to cause significant turbidity current flow.  As these winter inflow peaks are low compared with summer  flow, probably only the finer sediments (silt) are moved and thus most deposition does not occur until the turbidity current decays in the nearly level distal basin of the lake: that i s , samples from this area (72 - 92) show the effect more than the proximal samples (42 - 52). The winter layer of 1968-1969 is more complex than any other.  A t least 7  light toned laminae can be traced through the Ekman samples 72 - 92 but no corresponding late winter peaks occurred in 1969. This was the only winter on record during which the lake surface froze completely (section 4.2) but the significance of this event is difficult to relate to the multiple laminae seen in the winter layer. Generally the major autumn events generated by rain storms occur before winter deposition begins and are not as clearly distinguished by tone in the varve records as the late winter events.  However, in most years in which autumn floods  occurred, the dark, fine grained deposit of the winter is much thicker than average.  145  These thicker varves are seen in 1929, 1940, 1956, 1957, 1958, and 1967 (figure 5.15 and 5.18), all years in which autumn floods occurred (figure 5.14). exceptions however:  There are  for example, the winter layer of the 1937 varve in core 61C is  not thicker than others (figure 5.15 a) although a large flood occurred in that year (figure 5.14).  Apparently these floods bring large amounts of fine sediment into the  lake (perhaps more than would be indicated by the relation of figure 2.23 given the effect of hysteresis).  However, inflow usually falls rapidly to low autumn values  within a day or so of the flood so that this sediment may  be left in the lake water to  settle out through the winter, rather than being flushed out as occurs with sediment inflow during the melt season (section 4 . 2 . ) . The underflow events described in section 4 . 3 . can be related to laminae within the summer layer of the 1971 varve, particularly in the Ekman samples shown in figure 5.18.  However, only the two most powerful underflow events (August 4-8 and  August 21-22) show as deposits of significantly greater grain size (samples E and H, figure 5.19) A l l other samples analyzed from the 1971 varve, except the winter layer, had approximately the same grain s i z e .  5.5.  Results of continuous seismic profiling in Lillooet Lake O n August 19, 1972, continuous seismic profiles were made with a 'boomer'  sound source (Edgerton, 1964) in North Lillooet Lake.  Although some difficulty was  encountered in shielding the recording instrument from electrical noise made by three motor generators on the small floating platform (figure A . 5 . 1 ) and the outboard motor on the tow boat, the results presented in figure 5.20 give an indication of the subbottom topography of Lillooet Lake.  Figure 5.20 (on pages following figure 5.19) Lillooet Lake. For location see figure 5 . 2 1 . 1440 m s , in sediment, 2000 m s .  Continuous seismic profiles from Assumed velocity of sound in water,  146  0  Values  Equivalent fall diameter (microns) Figure 5.19 Texrural anal/sis of samples from rhe 1971 varve of sample 72 (see figure 5.18).  water_ depth (m) o o  J  L_\  L  lime  o  en  i  OJ  (s  ro  149 The velocity of sound in the lake bottom sediment is unknown. A review by Nafe and Drake (1964) lists the results of a number of studies mainly of marine sediments which indicate that the velocity of sound in near surface sediment (depth of burial less than 100 m) is between 1700 and 2300 m s ^.  V e l o c i t y is most strongly  related to porosity but grain size and carbonate content have also been shown to be significant in some circumstances (Sutton et a l , 1957).  In place of measurement, a  median of 2000 m s ^ is probably reasonable to apply as a first approximation to the Lillooet Lake sediments. In the proximal sediments, sound penetration and thus subbottom resolution are restricted by the coarser grain size of the material, the hummocky surface which scatters reflected sound, and the probable lack of coherent layering within the slumped sediments.  Farther from the d e l t a , at least three reflecting horizons can be  seen at depths in the deepest part of the lake of approximately 16, 4 2 , and 62 metres below the first reflecting horizon the water-sediment interface. can be traced as close to the delta as raft 1,  The second surface  where it lies 57 m below the present  surface but the others are obscured except in the distal basin.  Sedimentation between  the present and the second surface appears to have been relatively uniform a n d , except in the prodelta slump z o n e , there is little evidence of major slump deposits. Below the second, however, large mounds and banks can be seen , particularly in cross sections of figure 5.20.  Bedrock beneath the sediment is obscured or confused  in detail by side echos, lack of penetration, and noise.  However, the profiles do  indicate evidence of its surface according to the dashed lines of figure 5.20. Maximum sediment deposition therefore is of the order 180 - 120 m. The second and third surfaces maybe dated approximately and therefore related to geological and glacial history of Lillooet V a l l e y .  The mean thickness of  uncompressed varves in cores 8 1 , 9 1 , and 95 is 17.5 mm corrected according to the linear relation of figure 5 . 8 .  If it can be assumed that the void ratio at the sediment  Figure 5.21  Location of continuous seismic profiles of figure 5.20.  151  surface is approximately 2.0 and the mean void ratio between 0 and 16 m depth is 0 . 8 , then the annual accumulation in the sediment column is 7 mm yr ^.  Depth to the  second surface divided by this rate of accumulation yields an age of the second surface 3 of 2.3 x 10  years, assuming the rate has been constant in that period.  Similar c a l c u l a -  tion with the data at the location of core 30B, dates the second surface there at 2.2 x 3 -1 10 years B.P., (thickness 52 m, annual accumulation 24 mm yr ). This may be an underestimate as accumulation at this location must have increased as the delta advanced. Thus there is indication that the second surface is related to major volcanic eruption in 3 the vicinity of Meager Mountain dated at 2.44 x 10 (1967).  years B.P.  by Nasmith et a l .  Large amounts of volcanic ash and pumice were deposted east of these  mountains and Lillooet River must have rapidly cut through them to form its present canyon.  For a short period inflow may have been highly charged with this load, which  could be transported by underflow throughout the lake to be deposited as a coherent reflecting horizon.  Similar calculation applied to the third horizon yielded a very  3 approximate age of 7.4 - 11 x 10  years.  The wide range is due to the uncertainty  of the rate of deposition, undoubtedly different during the hypsithermal, and to the amount of compaction in the older sediments.  However, it is possible that this horizon  marks Pleistocene subglacial or proglacial sediment.  The lower horizon may represent  earlier events associated with glaciation but its origin is very uncertain. 5.6.  Rate of sediment accumulation in Lillooet Lake and yield from Lillooet River  Basin The rate of sediment accumulation can be estimated from the morphometric evidence of chapter 3, and from the varve thickness data in section 5 . 4 . 1 . Unfortunately, the continuous seismic profiles do not resolve subbottom structure near the delta.  It is not possible to extend the horizon estimated as 2400 years old since  even the lowest rate of delta advance indicates that the delta front would have been  152  17 km upstream at that time.  However, if it can be assumed that the prodelta  morphometry remains more or less the same as the delta advances (probably reasonable in view of the discussion of sediment strength in section 5 . 2 . ) , then the volume of sediment alone that has accumulated between 1913 and 1969 in the prodelta zone 7 (to the distal edge of the major slump mounds) was 5.3 x 10  3 m  from measurements on  figures 3 . 3 , 3 . 4 , 3 . 6 , and other echograms, assuming a void ratio of 0 . 8 .  The varve  thickness data of figure 5.11 indicates that accumulation of sediment alone in the rest 6 of the lake during the same period has been 6 . 3 x 1 0 time of deposition to have been 2.0 (figure 5.8).  3 m  assuming void ratio at the  That is, the annual accumulation of 6  solid material in Lillooet Lake is approximately l . O x 10  3 m .  In order to estimate the sediment inflow of Lillooet Lake, an indication of its trap efficiency is required.  N o data were kept at the outlet of Lillooet Lake of  sediment discharge but a rough estimate can be made from the work reported in section 4.2.  G i v e n observed circulation velocities and fall times shown in figure 4 . 1 0 ,  virtually all particles of diameter larger than 10JJ Lake.  w i l l be trapped in North Lillooet  While the relation of figure 4 . 8 c cannot be applied near the delta, it becomes  more nearly reasonable at the south end of North Lillooet Lake where all but the finest grained sediments have settled out.  From this, observed transmissivity, and  water inflow to North Lillooet Lake, approximately 0.17 x 10^ tons of sediment are estimated to have been flushed out of North Lillooet Lake in 1972. That is, the trap efficiency of North Lillooet Lake is of the order 93 per cent which falls in the range of data for reservoirs of similar capacity-inflow ratio presented by Brune (1953). Underflow is probably not significant in transporting sediment out of North Lillooet Lake because of the large extent of lake bottom of nearly zero slope and the rise of 40 m in the lake floor at its southern end.  While this value is from one year only,  substantial error here is small compared to the possible errors in accumulation measurement.  153 Thus It is indicated that the annual yield between 1913 and 1969 of all the 6 area tributary to North Lillooet Lake was l . l O x 10 0.28 mm yr ^).  3 m  -1 yr  (specific denudation,  If it is assumed that only Ure, Green and Lillooet Rivers are  contributing significant sediment, then the specific denudation in these basins was 0.36 mm yr ^ .  These values may overestimate the true rate of sediment production  considering that much of the rapid increase in delta advance rate between 1948 and 1969 is a measure of yield from the lower slopes, valley bottom, and river channel, probably related to human activity, rather than to substantially increased production of sediment from 'unaltered rock mass' (figure 2.18).  Further evidence for this  overestimate is obtained from measurements of the seismic records of figure 5.19. The annual accumulation of sediment alone in North Lillooet Lake beyond the slump  3  zone in the 2.4 x 10  years from the time the second reflecting horizon was estimated 4  to have been deposited, was 4.9 x 10 5 rate in this zone of 1 .1 x 10  3 m , that is, nearly one half of the present  3-1 m  yr  .  This is almost certainly an overstatement of  the difference considering the rate of delta advance and therefore more rapid sedimentation in this z o n e . 5.7.  Summary Textural data from fluvial and lake bottom sediments indicate that, during  the melt season, virtually all particles finer than 62  (silt and clay size) are  winnowed from the bed material in all but protected backwater areas on the delta surface.  However, the delta lip (specifically the small zone of classical foreset beds)  forms a distinct barrier between deltaic and lacustrine sediments. of the sediment beyond the foreset beds is sand size  Less than 5 0 %  - none is coarser than medium  sand - and within 1.0 - 1.5 km of the delta this value is less than 2 0 % .  Although  at the mouths of the principal distributaries and the mouth of Birkenhead and Ure Creeks, the sediment is coarser and better sorted, there is no evidence of underflow channels in which coarse sediment is carried to the deepest part of the lake.  154  However, underflow is important in distributing sediment throughout the lake and the resulting deposits are easily recognized in the stratigraphic column. A number of workers beginning with de G e e r in 1912 and later Kuenen (1951) have suggested that the main processes in the formation of varves is underflow.  However,  in thermally stratified lakes, which would include all but ice contact and immediately proglacial lakes, interflow into the less dense water of the epilimnion may be responsible for transporting water throughout the upper part of the lake and even into the outlet beyond.  In all but the most proximal section of Lillooet Lake, water of  the hypolimnion was usually very much clearer than that of the epilimnion indicating that little of this interflowing sediment is able to settle to bottom except close to the delta. During winter the thermal and suspended sediment structure of the lake breaks down, and density of the lake water is sufficiently reduced that even small increases in inflow are sufficient to cause underflow which brings coarser grained material to the deepest part of the lake to be deposited as a light toned lamina within the darker, very fine grained winter deposits.  The autumn floods, on the other hand, occur  before thermal structure is destroyed and before the lake is cleared of suspended sediment.  They bring into the lake a large quantity of fine sediment in a very short  time, sediment that cannot be flushed from the epilimnion as can summer inflow because discharge falls rapidly to low autumn values so quickly after the event. It is therefore deposited as an unusually thick winter  layer.  Thickness of the varves deposited in Lillooet Lake decreases outward from the d e l t a , although at least one section midway along the lake on the northwest side has thinner varves and is apparently not swept by underflow.  Varve thickness is well 2  correlated with inflow;  values of the coefficient of determination, r , as high as 3  0.89 were calculated.  M e l t season inflow greater than 200 m  -1 s  but less than the  3 -1 larger floods (400 m s ) produces highest correlation with varve thickness, and  155 thickness in cores from the deepest part of the lake where decaying underflow deposits sediment, are better related to inflow than thickness from those cores between. Strength measurements of lake bottom sediments with a simple quick shear box indicate that slopes as low as 1 . 6 ° are unstable in remolded sediment, although this represents an average in a relation with considerable scatter.  That is, the zone of  mounds corresponds with that of unstable sediments supporting the hypothesis that they are the result of slumping.  Little is known of the frequency and magnitude of slumps.  Some cores recovered from this area showed no disturbance in at least 10 years of varve record. Another showed disturbance in the 1971 varve possibly related to a large wave on the lake surface on August 5, 1971, thought to have been generated by a large movement of sediment on the lake floor.  156 6.1.  Conclusions The principal findings of this work are as follows: 1)  Snow storage and nival melt give rise to a highly seasonal inflow pattern  with mean summer runoff an order of magnitude greater than that of winter. 2)  The yield of fine sediment at Lillooet Lake is even more strongly seasonal  as sediment concentration varies approximately as the square of water discharge. 3)  Glaciers are a major contributor of fine sediment accounting for 6  approximately 35 - 4 0 % of the 1 .1 x 10  3 m  -1 yr  sediment yield at Lillooet Lake.  It is believed that the finest of this sediment is rock flour of clay size rather than clay minerals and soil colloids which would give different physical properties to the sediments being deposited in the lake (cf.. Terzaghi, 1955). 6 4)  Dissolved sediment yield is estimated as 0.19 x 10  1 5 % of the total y i e l d .  3 m  -1 yr  -approximately  Much of this sediment apparently remains in solution and is  passed through Lillooet Lake. 5)  Distribution of large loads of sediment in Lillooet Lake during the melt  season occurs by interflow and dispersion within the epilimnion or by underflow along the lake bottom.  Coarser material, particularly fluvial bed load is deposited in  classical foreset beds at the delta front where infrequent large slumps redistribute the unstable deposits. The zones of sedimentation are also energy levels in the delta system from the highest level of the proximal foreset beds where avalanching is quasicontinuous and input is high, through the zone of slumping where normal accumulation from settling and turbidity currents is disturbed by infrequent major slumping, to the most distal, nearly flat lying bottom sediments. 6)  The thermal stratification of Lillooet Lake strongly affects sediment  movement in the lake.  During periods of low to moderate inflow, much of the fine  sediment passes through the lake to the river beyond as interflow above the thermocline.  157 7)  Major underflow occurs during flood and the thermal structure of the lake  may be temporarily destroyed.  Deposition from these events can be recognized as  laminae within the summer layers of the varved sediments. 8)  Annual accumulation in varves is closely related to the amount of water  inflow to Lillooet Lake. 9)  Autumn floods frequently bring fine sediment to the lake, but it is not  flushed through as is summer inflow.  As the sediment content of the lake remains high  from summer inflow, underflow does not occur and this sediment is dispersed in the epilimnion and settles during the winter as an unusually thick dark band in the varve record and is useful as a dating marker. 10)  Similar floods occurring in late winter as the result of brief mild periods  produce a different result.  Apparently the lake water is sufficiently cleared of  sediment during the winter that even the relatively low concentration of suspended sediment in the inflow is sufficient to cause underflow and deposition of a lamina of lighter toned, coarser sediment within the winter layer of the varve.  Although a  distinct dating marker, this light layer may be mistaken for a separate varve.  6.2.  Further work Results of this study agree with those of most other work in high energy,  proglacial or mountain lakes: highly turbid underflow is the major agent transporting material beyond the delta front and the stratigraphy of the resulting deposits is distinctive and nearly universal.  Although the mechanics of stratified flow and  turbidity currents is established and laboratory studies have been carried out, no direct studies of the mechanics of sediment entrainment, transport and deposition by turbidity currents in proglacial lakes have been attempted.  Further investigation  should consist of the selection of a small active proglacial lake in which the velocity of underflow and associated lake circulation could be monitored with a network of current meters to determine:  158 1)  competence and capacity of the currents (cf. Harleman, 1961),  2)  rate of deposition and spatial distribution of sediments as the current  decays in the lake, 3)  bed shear stress which is significant in determining erosional and  depositional forms observed in ancient deposits (Shaw, in press; A s h l e y , in press; Banerjee, in press), 4)  interfacial  mixing and dilution and thus weakening of the current,  5)  the role of counter currents and general lake water circulation in  distributing sediment in the lake.  159 References Agterberg, F . P . , and Banerjee, I. 1969. Stocastic model for the deposition of varves in g l a c i a l lake Barlow-Ojibway, Ontario, Canada. C a n . J . Earth S c i . 6, p p . 625-652. Allen, J . R . L . 190.  deltas.  1966.  O n bed forms and paleocurrents.  Sedimentology, 6, pp. 153-  1968.  The diffusion of grains in the lee of ripples, dunes and sand  J . Sediment. Petrol. 38, pp. 621-632.  Anderson, F . E . and Hess, F.R. 1969. A n inexpensive coring platform for shallow estuarine waters. J . Sediment. Petrol. 39, pp. 1238-1242. Andrews, R.S., and Cepek, R.J. 1972. Mass physical and acoustic properties of sediments from the Monterey Submarine Fan (abstract). Trans. Amer. G e o physical U . , 5 3 , p. 1028. Antevs, E. 1925. Retreat of the last ice sheet in Eastern Canada. C a n a d a , M e m . 146. 1951.  G e o l . Survey  G l a c i a l clays in Steep Rock Lake, O n t a r i o , Canada.  Geol.  Soc. A m . Bull. 6 2 , p p . 1223-1262. Arnborg, L. 1948. The delta of Angermanalven. pp. 673-690.  Geografiska Annalar, 2 0 A ,  Ashley, G . M . in press. Rhythmic sedimentation in G l a c i a l Lake Hitchcock Massachusetts-Connecticut. Soc. Economic Paleontologists and Mineralogists Spec. Pub. ASTM  (Amer. Soc. for Testing and Materials) 1963. Tentative method for grainsize analysis of soils. Book of A S T M Standards. Chapter D422-63,pp. 194-205.  Axelsson, V . 1967. The Laitaure Delta: a study of deltaic morphology and processes. Geografiska Annaler, 4 9 , p p . 1-127. Banerjee, I. 1966. Turbidites in a glacial sequence: a study from the Talchir Formation, Raniganj C o a l f i e l d , India. J . G e o l . , 74, pp. 593-606. in press. Canada.  Sedimentology of Pleistocene glacial varves in O n t a r i o ,  G e o l . Surv. of C a n a d a , Bull. 226.  Bark, L . S . , Ganson, P.P., and Meister, A . 1964. in sea water. M a c M i l l a n C o . , 180 p. Bates, C . G . , and Henry, A . J . Wheel G a p , Colorado.  Tables of the velocity of sound  1928. Forest and streamflow experiment at Wagon Monthly Weather Review, Sup. 3, 79 p.  Bell, H.S. 1942. Density currents as agents for transporting sediments. 50, pp. 512-547.  J.  Geol.,  160 —•  1947. The effect of entrance mixing on the size of density currents  ;  in Shaver Lake.  Trans. A m e r . Geophys. U. 2 8 , p p . 780-791.  Bliss, J . H . 1938. Discussion following Grover and Howard (1938). C i v i l . Engineers, Trans., 103, pp. 747-751.  Amer.  Soc.  Bouma, A . H . 1959. Some data on turbidites from the Alpes-Maritimes (France). G e o l . en M i j n b o u w , 2 1 , pp. 223-227.  458 p.  1969.  Methods for the Study of Sedimentary Structures.  Bouma, A . H . , and Brouwer, A . (eds.) 1964. Turbidites. Sedimentology. Elsevier Publ. C o . , 3, 264 p. Brune, G . M . 1953. Trap efficiency of reservoirs. p p . 407-418.  Wiley C o . ,  Developments in  Trans. A m e r . G e o p h y s . U. 3 4 ,  Cairnes, C . E . 1925. Pemberton A r e a , Lillooet District, British C o l u m b i a . G e o l . Surv. C a n . Summ. Rept. , 1924, Part A , pp. 7 6 A - 9 9 A . 1927 Observations on Lillooet V a l l e y , British Columbia with particular reference to its geology and mineral deposits. C a n . M i n i n g J . 4 8 , pp. 140-144, 162-166. C a m s e l l , C . 1918. Reconnaissance along the Pacific G r e a t Eastern Railway between Squamish and L i l l o o e t . G e o l . Surv. Canada Summ. R e p t . , 1917, Part B, pp. 12-22. C h u r c h , M . 1972. Baffin Island sandurs: a study of arctic f l u v i a l processes. Surv. of C a n a d a , B u l l . 2 1 6 , 208 p.  Geol.  C h u r c h , M . A . , and G i l b e r t , R. in press. Proglacial f l u v i a l and lacustrine e n v i r o n ments, in G l a c i o f l u v i a i and G i a c i o l a c u s t r i n e Sedimentation. A . V . J o p l i n g and B . C . MauDcncId (eds.). Soc. Econ. Paleontologists and Mineralogists Spec. Publ. C l e v e l a n d and Cameron, Engineers, unpublished. Engineering reports to Pemberton drainage commissioners dated 7 July 1913 (3 p p . ) , 18 January 1915 (6 p . ) , 18 A p r i l 1914 (5 p.), 28 July 1914 (2 p . ) . C o l e m a n , J . M . 1969. Brahmaputra River; channel processes and sedimentation . Sedimentary G e o l . , 3, p p . 129-239. C o l l e t , L.W. Corbel, J .  1925. Les Lacs.  Paris, 320 p.  1959. Vitesse de I'erosion.  Z e i t s . fur Geomorphologie, 3, pp.. 1-28.  Dahlskog, S. 1966. Sedimentation and vegetation in a Lapland mountain d e l t a . Geografiska A n n a l e r , 4 8 A , pp. 86-101. Dahlskog, S . , Damberg, A . , Harden, P . P . , and L i l j e l u n d , L.E. 1972. The K v i k k j o k k |. D e l t a . Progress report on a muliidisciplinary research project on a boreal mountain-lake d e l t a . Uppsala Universitet Narurgeografiska Institutioned, U N G I Rapport 13, 78 p.  161 De G e e r , G . 1912. A geochronology of the last 12000 years. Congr. Stockholm, 1910. pp. 241-253.  Proc. Int. G e o l .  Duquennois, H. 1956. New methods of sediment control in reservoirs. 8, pp. 174-180.  Water Power,  Durum, W . H . , H e i d e l , S . G . , and Tison, L.J. 1960. World-wide runoff of dissolved solids. Int. Assoc. S c i . H y d r o l . , Publ, 5 1 , pp. 618-628. Dyrness, C . T . 1965. Soil surface condition following tractor and high lead logging in the Oregon Cascades. J . Forestry, 6 3 , pp. 272-275. Edgerton, H . E . , and Hayward, G . G . 1964. The Boomer sonar source for seismic profiling. J . Geophys. Res., 6 9 , pp. 3033-3042. Elworthy, R.T. 1926. Hot springs in Western Canada - their radio-active and chemical properties. C a n . Mines Branch Rept., 669, pp. 1-33. Embleton, C , and K i n g , C . A . M . 1968. G l a c i a l and Periglacial Geomorphology. St. Martins Press, New Y o r k , 608 p. Emery, K . O . , a n d D i e t z , R.S. 1941. Gravity coring instrument and mechanics of sediment coring. G e o l . Soc. A m . Bull., 5 2 , p p . 1685-1714. Emery, K . O . , and Hulsemann, J . open barrel gravity corers.  1964. Shortening of sediment cores collected in Sedimentology, 3, pp. 144-154.  Ericson, D.B., and W o l l i n , G . 1956. Correlation of six cores from the equatorial Atlantic and Caribbean. Deep Sea Res., 3, pp. 104-125. F i o c k , L.R. 1934. Records of silt carried by the Rio Grande and its accumulation in Elephant Bute Reservoir. Amer. Geophys. U. Trans. Part II, pp. 468-473. Forel, F . A . 1885. Les ravins sous-lacustres des fleuves glaciaires. A c a d . S c i . Paris, 101, pp. 725-728.  Comptes Rendus  Fry, A . S . , C h u r c h i l l , M . A . , and El den, R.A. 1953. Significant effects of density currents i n T V A ' s integrated reservoir and river system . Proc. M i n n . Int. Hydraul. C o n v . , Minneapolis, M i n n . , Sept. 1-4, 1954, pp. 335-354. Fulton, R.J. 1965. Silt deposition in iate-glacial lakes of southern British C o l u m b i a . Amer. J . S c i . , 263, pp. 533-570. Fulton, R.J., and Pullen, M . J . L . T . 1969. Sedimentation in Upper Arrow Lake, British Columbia. C a n . J . Earth S c i . , 6, pp. 785-790. G i l b e r t , R. 1972. Observations on sedimentation at Lillooet D e l t a , British Columbia . in Mountain Geomorphology ( O . Slaymaker and H . J . McPherson, eds.). Tantalus Press, pp. 187-194. 1973. Processes of underflow and sediment transport in a British Columbia mountain lake. Fluvial Processes and Sedimentation. Proc. Ninth Canadian Hydrology Symposium. N a t . Res. Council of C a n a d a , Edmonton.  162  G o u l d , H.R.  1951 .  Some quantitative aspects of Lake Mead turbidity currents.  Soc. E c o n . Paleontologists and Mineralogists  Special P u b . ,  2,  pp.  34-52.  G r o v e r , N . C . , and Howard, C . S . 1 9 3 8 . The passage of turbid water through Lake M e a d (with discussion). Amer. Soc. C i v i l E n g . , Trans., 1 0 3 , pp. 7 2 0 - 7 9 0 . G r a n a r , L. 1 9 5 6 . Dating of recent f l u v i a l sediments from the estuary of the Angerman River (the period 1 8 5 0 - 1 9 5 0 A . D.). G e o l . Foren. Stockholm, 7 8 , p p . 6 5 4 658.  Gustavson, T . C .  in press.  Sedimentation and physical limnology in proglacial  Malaspina Lake, southeastern A l a s k a . Mineralogists Special Pub!.  G u y , H.P.  Soc. Economic Paleontologists and  1 9 6 9 . Laboratory theory and methods for sediment analysis.  of Water-Resources Investigations of the U.S. 5 8 pp.  Techniques  G e o l . Surv. Book 5 , Chapter C I ,  Hanson, G . 1 9 3 2 . Varved clays of Tide Lake, British C o l u m b i a . Roy. Soc. of C a n a d a , Trans. S e r . , 3 , 2 6 , Sect. 4 , pp. 3 3 5 - 3 3 9 . —  1 9 3 4 . The Bear River Delta B . C . and its significance regarding Pleistocene and Recent g l a c i a t i o n . Roy. Soc. of C a n a d a , Trans., Ser. 3 , 2 8 , Sect. 4 , p p . 1 7 9 - 1 8 5 .  Harleman, D . R . F . 1 9 6 1 . Stratified flow. Chapter 2 6 in Streeter, V . L. ( e d . ) , Handbook of Fluid Dynamics. M c G r a w - H i l l Book C o . , New Y o r k , 2 1 p. Henderson, F . M . H o l l a n d , S.S.  1 9 6 6 . O p e n Channel F l o w .  M a c M i l l a n C o . , 5 2 2 p.  1 9 6 4 . Landforms of British Columbia : a physiographic o u t l i n e .  British Columbia Department of Mines and Petroleum Resources B u l l . 4 8 , 1 3 8 p.  Houbolt, J . J . H . C . , and Jonker, J . B . M . .  1 9 6 8 . Recent sediments in the eastern  part of the Lake of Geneva (Lac Le'man).  G e o l . en M i j n b o w , 4 7 , p p . 1 3 1 -  148.  Howard, C . S . 1 9 5 3 . Density currents in Lake M e a d . Proc. M i n n . Int. Hydraul. C o n v . , M i n n e a p o l i s , M i n n . , Sept. 1 - 4 , ' 1 9 5 3 , p p . 3 5 5 - 3 6 8 . Hutchinson, G . E . 1 9 5 7 . A Treatise on Limnology V o l . I Geography Physics and Chemistry. W i l e y and SOPS I n c . . New Y o r k , 1 0 1 5 p. Hvorslev, M . J . 1 9 4 9 . Subsurface exploration and sampling of soils for c i v i l engineering purposes. Waterways Experiment Station, U.S. Army Corps of Engineers, V i c k s b u r g , M i s s . 5 2 1 p. Isaacs, J . D . , and Brown, D . M . pp.  1 9 6 8 . 'Bootstrap'corer.  J . Sediment. P e t r o l . , 3 8 ,  159-162.  J i l c k l i , H. 1 9 5 7 . Gegenwartsgeologie des Bundnerischen Rheingebietes - ein Beitrag zur exogenen Dynamik A l p i n e r Gebirgslandschaften. Beitrdge zur G e o l o g i e der S c h w e i z , Geofechnische Serie, Lieferung 3 6 , p. 1 2 6 (not seen).  163 Jeriov, N . G .  1968. O p t i c a l Oceanography.  Elsevier Publ C o . , 1 9 4 p .  Johnston, W . A . 1922 a . The character of the stratification of the Recent delta of Fraser River, British C o l u m b i a , J . G e o l , 3 0 , p p . 115-129. —  1922 b.  Sedimentation in Lake Louise, A l b e r t a , C a n a d a .  Amer. J .  S c i . 5th Series 4 , pp. 376-386. J o p l i n g , A . V . 1960. A n experimental study on the mechanics of bedding. Thesis, Harvard U n i v . , Cambridge, M a s s . , 358 p.  Ph.D.  K e l l e r h a l s , P., and M u r r a y , J . W . 1969. Tidal flats of Boundary Bay, British C o l u m b i a . B u l l . Canadian Petroleum G e o l . , 1 7 , p p . 6 7 - 9 1 . Keulegan, G . H . 1949. Interfacial instability and mixing in stratified flows. of U.S. N a t i o n a l Bur. of Standards, 4 3 , p p . 487-500. Kindle, E.M.  1930. Sedimentation in a g l a c i a l l a k e .  J . Res.  J . G e o l . , 38, pp. 81-87.  Krause, D . C . 1962. Interpretation of echo sounding profiles. Int. Hydrographici 7 R e v . , 3 9 , p p . 65-123. Kuenen, Ph. H. 1950. Turbidity currents of high density. London, Part 8, p p . 44-52,  18th Int. G e o l . C o n g . ,  1951 . Mechanics of varve formation and the action of turbidity currents.  G e o l . Foren. Forhandl. 7 3 , p p . 69-84.  1964. Deep-sea sands and ancient turbidites. in Bouma, A . H . , and Brouwer, A . (eds.) Turbidites. Developments in Sedimentology. Elsevier . Pub. C o . , 3, p p . 3-33. Kuenen, Ph. H . , and Humbert, F.L. 1964. Bibliography of turbidity currents and turbidites. in Bouma, A . H . , and Brouwer, A . (eds.) Turbidites. D e v e l o p ments in Sedimentology. Elsevier Pub. C o . , 3, pp. 222-246. Kuenen, Ph. H . , and M i g ; i c r ! r , i , C . ' . 1950. Turbidity currents as a cause of graded bedding. J . G e o l . , 5 8 , pp. 91-127. Lajoie, J . (ed.) 1970. Flysch Sediirionroiogy in North A m e r i c a . Canada Special Paper, 7, 272 p. Lajfai, E.Z. 1967. The origin of some varves in Toronto, C a n a d a . S c i . , 4, p p . 633-639.  G e o l . Assoc. C a n . J . Earth  Lara, J . M . , and Sanders, j . l . 1970. The 1963-64 Lake M e a d Study. U . S . Interior Bureau of Reclamation, Rept. R E C - O C E - 7 0 - 2 1 , 172 p p .  Dept.  Lewis, C P . 1970. Estimation of suspended sediment concentrations in natural water bodies from Secchi disk measurements. M . A . Thesis, U n i v . of British Columbia 1970, 62 p. Lieberman, J . A ., and Hoover, M . D . 1951 . Streamflow frequency changes on Coweeta Experimental Watersheds. Amer. Geophys. U . , Trans., 3 2 , p p . 73-76.  164 Mathews, W . H . 1956. Physical limnology and sedimentation in a glacial lake. G e o l . Soc. A m . B u l l . , 6 7 , pp. 537-552. unpublished. 1964.  Geology of the Pemberton area.  1959, 4 p.  Sediment transport from Athabasca G l a c i e r , A l b e r t a .  Int.  Assoc. Scientific Hydrology, Pub. 6 5 , pp. 155-165. Mathews, W . H . , and Shepard, F.P. 1962. Sedimentation of Fraser River D e l t a , British Columbia. Amer. Assoc. Petroleum G e o l . , 4 6 , pp. 1416-1443. Mayers, I.R. 1969. A n analysis of the form and origin of the Fraser River Delta's subaqueous slump deposits. Amer. Geophys. U. Annual Meeting (abs.). McPherson, H . J . , a n d Slaymaker, H . O . Cordillera.  1972.  Geomorphology of the Canadian  Guide Book for Event Ca6 of the Int. G e o g . Congr.  Menard, H . W . , and Ludwick, J . C . of marine turbidity currents. Spec. Publ. 2, pp. 2-13.  1972, 85 p.  1951. Applications of hydraulics to the study Soc. Econ. Paleontologists and Mineralogists,  Middleton, G . V . 1965. Primary sedimentary structures and their hydrodynamic interpretation. Soc. Economic Mineralogists and Paleontologists, Spec. Pub. 12, 265 p. 1966.  Emperiments on density and turbidity currents.  of density currents.  II  Uniform Flow  C a n . J . Earth S c i . , 3, pp. 627-637.  Monney, N . T . 1971. Measurements of the engineering properties of marine sediments. M a r . T e c h . Soc. J . 5, pp. 21-30. Moore, D . G .  1961. Submarine slumps.  J . Sediment. Petrol. , 3 1 , pp. 343-357.  Morgenstern, N.R. 1967. Submarine slumping and the initiation of turbidity currents, in Marine Geotechnique, A . F . Richards (ed.). Univ. Illinois Press, pp. 189220. N a f e , J . E . , and Drake, C . L . 1964. Physical properties of marine sediments. S e a , Chapter 2 9 , V o l . 3 , M . N . Hill (ed.), pp. 794-815. Nasmith, H . , Mathev/s, W . H . , and Rouse, G . E .  1967.  The  Bridge River ash and some  other Recent ash beds in British C o l u m b i a . C a n . J . Earth S c i . 4, pp. 163-170. N E I C ( N a t . Earthquake Information Centre) 1972. Preliminary determination of epicentres, July 1972. U.S. Dept. Commerce, N a t . O c e a n i c and Atmospheric Administration. Nelson, J . D . unpublished. Report to the Howe Sound and Northern Development C o . , Vancouver, April 5, 1911, 1 p. N i z e r y , A . , and Bonnin, J . 1953. Observations systematiques de courants de densite dans un retenue hydroelectrique. Proc. M i n n . Int. Hydraulics Conv. Minneapolis, M i n n . , pp. 369-386.  165 O'Brien, M.P. 1938. Discussion following Grover and Howard (1938). C i v i l Engineers, Trans., 103, pp. 743-744.  Amer. Soc.  O f f i c e r , C . B . 1958. Introduction to the Theory of Sound Transmission with Applications to the O c e a n . M c G r a w - H i l l Book C o . , New York, 284 p. 0 s t r e m , G . 1966 a . Mass balance studies on glaciers in Western C a n a d a , 1965. G e o g . B u l l . , 8, pp. 81-107. ,  1966 b. and A l b e r t a .  The height of the glaciation limit in Southern British Columbia  Geografiska A n n a l e r , 4 8 A , p p . 126-138.  in press. Sediment transport in glacier streams. tologists and Mineralogists, Spec. Pub.  Soc. Econ. Paleon-  0 s t r e m , G . , Bridge, C . W . , and Rannie, W . F . 1967. Glacio-hydrology, discharge and sediment transport in the Decade G l a c i e r area, Baffin Island, N . W . T . Geografiska Annaler, 4 9 A , pp. 268-282. Packer, P.E. 1965. Forest treatment effects on water quality, in Sopper, W . E . , and L u l l , H.W. (eds.) Forest Hydrology. Pergamon Press, pp. 687-699. Palmer, H.S.  1859.  'Plan of route from New Westminster to Fountain in British  C o l u m b i a ' . Topographical Depot of the War O f f i c e (British). to one i n c h . 1 sheet.  Scale:  4 miles  Patton, K . T . , and G r i f f i n , G . T . 1969. A n analysis of marine corer dynamics. T e c h . Soc. J . , 3, pp. 27-40.  Mar.  Ponton, J . R . 1972 a. Hydraulic geometry of Green and Birkenhead Rivers: southwestern Coast Mountains, British Columbia. M . A . Thesis, 1972, Univ. of British C o l u m b i a , 77 p. 1972 b.  Hydraulic geometry in the Green and Birkenhead Basins,  British Columbia, in Slaymaker, O . , and McPherson, H . J . Geomorphology. Tantalus Press, pp. 151-160.  (eds.) Mountain  Rapp, A . 1960. Recent development of mountain slopes in Karkevagge and surrounding Northern Scandinavia. Geografiska Annaler, 4 2 A , pp. 65-200, 2 plates. Redden, R.E. 1938. Discussion following Grover and Howard (1938). C i v i l Engineers, Trans., 103, pp. 771-774.  Amer. Soc.  Richards, A . F . , and Parker, H.W. 1967. Surface coring for shear strength measurements. A m . Soc. C i v i l E n g . , Proc. of Conf. on C i v i l Engineering in the O c e a n s , pp. 445-489. Robinson, H . F . 1920. The silt problem of the Zuni Reservoir. Trans., 83, pp. 868-893. Roddick, J . A . , and Hutchinson, W . W . 1973. G e o l . Surv. Canada Paper Series.  A m . Soc. C i v i l Eng.  Geology of Pemberton E/2 Map-area.  166 Rosfelder, A . M . and Marshall, N . F . 1967. Obtaining large, undisturbed, and oriented samples in deep water, in Richards, A . F . (ed.) Marine Geotechnique. Univ. Illinois Press, pp. 243-263. Ross, D . A . , and Riedel, W . R . 1967. Comparison of upper parts of some piston cores with simultaneously collected open-barrel cores. Deep Sea Res., 14, pp. 285294. Rubey, W . W . 1938. Discussion following Grover and Howard (1938). C i v i l Engineers, Trans., 103, pp. 763-767. Russell, F.J.  1967.  River and Delta Morphology.  Amer. Soc.  Louisiana State U. Press, 55 p.  Sanders, J . E . 1965. Prinary sedimentary structures formed by turbidity currents and related resedimentation mechanics. Soc. Economic Paleontologists and Mineralogists, Spec. Pub., 12, pp. 192-219. Shaw, J . in press. Sedimentary successions in Pleistocene ice-marginal lakes. Economic Paleontologists and Mineralogists, Spec. Pub. Shepard, F.P. 1956. Marginal sediments of the Mississippi Delta. Petroleum Geologists, B u l l . , 4 0 , p p . 2537-2623. Shirley, M . L . Soc.  (ed.)  1966.  Deltas in their G e o l o g i c Framework.  Soc.  Amer. Assoc.  Houston  Geol.  Singler, J . C . 1971. The effect of varying the parameters of vane shear tests of marine sediments. M . S c . Thesis, N a v a l Postgraduate S c h o o l , Monterey, C a l i f . 1971, 73 pp. Shostakovich, V . B . 1936. Geschichtete Bodenablagerungen der Seen als Klima-Ann. M e t . Z . 5 3 , pp. 176-182. Slaymaker, O . , and Gilbert.. R. 19/2. Geomorphic process and land use changes in the Coast Mountains of British Columbia: a case study. Proc. Symposium International de Geomorphologie, Pt. 1, Processus Periglaciaires £dudies sur le Terrain. Liege-Caen, 1-9 JuiSlet, 1971, p p . 269-279. Slaymaker, O . , and McPherson, H . J . Tantalus Press, 274 p.  (eds.)  1972. Mountain Geomorphology,  Sly, P . G . 1969. Bottom sediment sampling. Proc. 12th Conference on Great Lakes Res., pp. 883-898. Smith, A . J . 1959. Structures in the stratified late-glacial clays of Windermere, England. J . Sediment. Petrology, 2 9 , pp. 447-453. Smith, W . O . , Vetter, C P . , Cummings, G . B . , e t a l . 1960. Comprehensive survey of sedimentation in Lake Mead 1948-1949. U . S . G e o l . Surv. Prof. Paper, 295. Snedecor, G . W . , Cochran, W . G . State Univ. Press.  1967.  Statistical methods, sixth edition, Iowa  167  S t i c h l i n g , W. 1969. Instrumentation and techniques in sediment surveying. in Instrumentation and Observations Techniques. Proc. H y d r o l . Symp. N o . 7, N a t . Res. C o u n c i l , C a n a d a , p p . 81-139. S t i c h l i n g , W . , and Smith, T . F . 1 9 6 8 . Sediment surveys in C a n a d a . Dept. Energy, Mines and Resources, Inland Waters Branch, T e c h . B u l l , 1 2 , 1 7 p . Sutton, G . , Berkhemer, H . , and N a f e , J . E . 1957. Physical analysis of deep sea sediments . Geophysics. 2 2 , pp. 779-812. T a c k l e , D. 1962. InfiItration in a Western Larch - Douglas-Fir stand following cutting and slash treatment. Intermountain Forest and Range Experiment Station, Res. N o t e N o . 8 9 , 7 p. Taylor, D . W .  1948. Fundamentals of Soil M e c h a n i c s , W i l e y and Sons, 700 p p .  Terzaghi, K. 1941. Undisturbed c l a y samples and undisturbed c l a y s . Engineers J . , 2 8 , p p . 211-231 . 1955.  sediments.  Boston S o c . C i v i l  Influence of g e o l o g i c a l factors on the engineering properties of  E c o n . G e o l . 50th Anniversary V o l . , p p . 557-618.  —- 1956. Varieties of submarine slope failures.  Series, 5 2 , 21 p.  Harvard Soil M e c h .  T i f f i n , D . L . , M u r r a y , J . W . , Mayers, I.R., and G a r r i s o n , R.E. 1971. Structure and origin of foreslope hills Fraser D e l t a , British C o l u m b i a . B u l l . Canadian Petroleum G e o l . , 19, p p . 589-609. UNESCO 1966. S c i e n t i f i c Problems of the Humid Tropical Zone Deltas and their Implications. Proc. of the Dacca S y m . , Paris. V a n Straaten, L . M . J . U. (ed.) Publishing C o . , 464 p.  1964. Deltaic and Shallow M a r i n e Deposits.  Elsevier  W a l k e r , R . G . 1967. Turbidite seaYmintcry structures and their relationship to proximal and distal depcj? tional environments. J . Sediment. Petrology, 3 7 , p p . 25-43 1970.  Review of vhe g^o-.netry and facies organization of turbidites and  turbidite-bearing basin;.  Geol.A:>soc. C a n a d a , Spec. P a p e r , 7, pp. 219-251.  Walker, J . R . ,  and M a s s i n g i l l , J . V .  Welder, F.A. River.  1959. Processes of delta ic sedimentation in the lower Mississippi Coastal Studies Inst., Louisiana State U n i v . , Contribution 5 9 - 7 .  Northeastern G u l f of M e x i c o .  1970. Slump features on the Mississippi F a n ,  G e o i . Soc. Amer. B u l l . , 8 1 , pp. 3101-3108.  W i l l i a m s , J . E . 1949. Chemical weathering at low temperatures. 3 9 , pp. 129-135.  G e o g . Review,  168  APPENDICES  169 Appendix 2.1  Discharge records, Green River at Pemberton  Water Survey of Canada discharge records for Green River at Pemberton, just upstream from its confluence with Lillooet River, were discontinued at the end of September, 1951.  Unfortunately, continuous records from Lillooet River cannot be  used to accurately estimate Green River discharge from 1951 to the present, as even mean monthly discharge values of the two rivers are not well correlated during the period records were kept at both stations (figure A . 2 . 1 ) . Therefore during the Lillooet Lake study in 1971 and 1972, observations were made of discharge to better evaluate water and sediment inflow.  A Leupold-Stevens  A-35 water level recorder on a bridge pier at the eastern boundary of Narin Falls Provincial Park recorded stage continuously from June 10 - December 2 1 , 1971, and April 19 - September 2 6 , 1972. A chain staff gauge fastened to the bridge pier provided level control. Discharge measurements were made by standard current metering techniques from a small boat downstream at a destroyed road bridge 50 m upstream from the first confluence with Lillooet River. section and the metering section.  Insignificant inflow occurs between the stage recording The rating curve is presented as figure A . 2 . 2 from 2  which the relation log J Q Q = 7.274 - 0.1646S with r  = 0.98 was used to calculate  6 or 3 hour mean discharge (Q) from corresponding staff gauge stage (S) to which the water level recorder was set. and A . 2 . 2 (1972).  These discharge values are listed in tables A . 2 . 1 (1971)  170  1  1  Figure A.2.1 Relation between mean m o n t h l y discharge (rr>3 s-1) o f L i l l o o e t and Green Rivers, 1923-1951 f  +  H-  ¥  +  +  ++  f  + ++ +  +  + J 4H-  200  L i l l o o e t River  400  Jo, AA-  A  Figure A . 2 . 2  f  Rating curve, Green River near Pemberton, 1971 - 19/2  2"  100  150 Discharge (m 3s-i)  200  171  Table A . 2 . 1 Six hour mean discharge, Green River near Pemberton, 1971 Discharge (m^ s ^) Date  Time P . S . T . 0000  JUNE 10 JUNE11 J0NE12 JUNE13 JUNE 14 JUNE15 JUNE16 JUNE17 JUNE 18 JUNE 19 JUNE20 JUNE21 JUNE22 JUNE23 JUNE24 JUNE25 JUNE26 JUNE27 JUNE28 JU029 JUNE30 JULY 1 JULY 2 JULY 3 JULY 4 JULY 5 JULY 6 JULY 7 JULY 8 JULY 9 JULY 10 JULY 11 JULY 12 JULY 13 JULY 14 JULY15 JULY 16 JULY 17 JULY 18 JULY 19 JULY20 JULY21 JULY22  0600  120.8 124.0 123.6 124.0 115.9 115.0 107.0 111.5 106. 2 1 0 8 . 6 9 9 . 2 101.5 100. 7 102.6 88.9 89.2 101.1 1 1 0 . 7 111.5 115.9 120.3 125.0 120.3 123.6 145. 4 1 6 3 . 0 231.0 254.9 214. 9 199.2 159.9 152.8 119.9 113.7 99. 9 97.7 93.0 97.0 96.2 99.2 92.3 93.3 89.2 93.0 99.2 96.6 89,5 94.4 9 5 . 9 106.2 93.3 97.7 87.9 87.9 78.7 79.6 8 8 . 9 104.6 101.5 110.7 129. 8 1 3 9 . 0 1 3 4 . 3 126.4 118.1 117.2 116.7 119.0 127. 4 128.3 137.4 134.8 145. 4 140.0 151.1 1 4 5 . 4 2 0 3 . 0 198.5 237.2 2 2 5 . 8 240. 8 2 3 0 . 1 236.3 227.5 229.2 219.0  1200  1800  Mean  120.3 121.2 121.6 117.6 1 1 4 . 5 1 1 9 . 9 108.2 102.2 110.3 107.0 1 0 2 . 6 107.0 103.0 9 7 . 3 103.8 96.2 93.0 97.5 96.2 90.9 97.6 86.5 84.9 87.4 105.0 102.2 104.8 113.7 1 1 2 . 4 113.4 116.7 109.9 118.0 119.0 1 1 8 . 5 1 2 0 . 3 153. 4 158. 1 1 5 5 . 0 261.7 235.4 245.7 177.1 1 6 1 . 7 188.3 136.4 1 2 5 . 9 1 4 3 . 8 105.0 9 9 . 2 109.4 93.0 89.5 95.0 93.3 89. 9 93.3 95.1 89.9 95.1 89.5 85.9 90.3 92.6 93.0 92.0 89.5 83.9 92.3 88.2 84.6 89.2 100.3 94.8 99.3 95.9 89.9 94.2 85.9 81. 1 85.7 78.4 75.5 78.1 99.6 94.8 96.9 107.0 1 1 3 . 7 1 0 8 . 2 139.5 128.3 134.2 115.0 1 0 8 . 6 121.1 107.0 1 0 0 . 7 1 1 0 . 7 109.9 1 0 5 . 0 112.6 115.9 1 1 1 . 5 1 2 0 . 8 120.3 1 1 7 . 2 127.4 123.6 1 2 1 . 7 132.7 132. 3 152. 8 145.4 177.8 2 0 1 . 5 195.2 200.0 210.1 218.3 2 0 2 . 3 202. 3 2 1 8 . 9 203.0 203.0 217.5 191. 1 1 9 1 . 8 2 0 7 . 8  172  JULY23 2 2 6 . 6 2 1 4 . 9 187.5 188.9 JULY24 2 1 8 . 2 2 0 6 . 9 1 8 0 . 5 179.2 JULY25 2 0 7 . 7 198.5 175.8 177. 1 JULY26 2 0 6 . 9 198.5 175.8 181.2 JULY27 2 1 2 . 5 199.2 177.1 1 8 4 . 7 JULY28 2 1 3 . 3 2 0 3 . 0 181.2 193. 3 JULY29 2 1 8 . 2 2 0 9 . 3 190.4 193. 3 JULY30 2 1 4 . 9 2 0 3 . 0 180.5 184. 0 JULY31 2 0 8 . 5 199.2 179.8 186. 1 AUG 1 2 1 1 . 7 2 0 3 . 8 1 8 6 . 8 184.0 AUG 2 1 9 5 . 5 177. 1 153.4141.6 AUG 3 1 4 7 . 7 140.6 129.3 125. 5 AUG 4 135.3 131.3 126.4 124.0 AUG 5 130.3 1 2 5 . 9 116.3 111.1 AUG 6 1 2 4 . 0 121.2 112.0 108.2 AUG 7 124.0 115.9 104.6 103.4 AUG 8 1 2 0 . 3 115.0 105.8 108.6 AUG 9 130.3 121.2 1 1 0 . 3 110.7 AUG 10 1 2 9 . 8 120„8 109.0 109.5 AUG 11 128.3 118.5 106.2 103. 8 AUG 12 118.1 110.3 99.9 99.2 AUG 13 116.7 108„6 98.4 98. 8 AUG 14 114.1 103.8 104.2 103.4 AUG 15 103.8 96.2 86.5 80. 8 AUG 16 8 6 . 9 81.1 77.2 79. 6 AUG 17 8 2 . 4 7 8,. 1 72.1 6 8 . 7 AUG 18 7 1 . 6 70.8 68.4 67.4 AUG 19 7 1 . 0 71. 3 68.9 71.0 AUG 20 8 5 . 2 102.6 116.7 131.3 AUG 21 1 3 4 , 8 135.9 148.8 155.7 AUG 22 183.3 159.9 136.4 119.4 AUG 23 1 1 3 , 7 99.9 94.1 8 6 . 2 AUG 24 8 4 . 3 82.4 83.3 87.2 AUG 25 7 9 . 3 79.0 73.0 77.2 AUG 26 7 0 . 0 68.9 64.9 74.4 AUG 27 6 3 . 2 67.9 64.9 76.4 AUG 28 6 3 . 4 69.5 78.7 74.9 AUG 29 7 1 . 3 8 5 . 5 74. 9 79.0 AUG 30 8 3 . 0 80. 8 92.3 84.6 AUG 31 9 0 . 6 79. 3 87.2 86.2 SEPT 1 7 3 . 5 69.5 70. 2 74.4 SEPT 2 6 4 . 4 68. 4 60. 8 64.4 SEPT 3 6 2 . 0 57.7 55.1 5 5 . 5 SEPT 4 5 7 . 5 5 9 . 0 84.6 64.4 SEPT 5 10 5.4 123-6 138.5 121.7 SEPT 6 116.3 100.7 80. 8 88.5 64. 4 SEPT 7 7 9 , 9 71.9 66.9 SEPT 8 65„9 7 0 . 5 8 7 . 2 100. 3 SEPT 9 9 1 . 6 66. 4 79.6 71.0 SEPT 10 68.1 6 3 . 9 6 5 . 4 142. 7 SEPT 1 1 2 0 0 .152.2 7 128.3 110.7 SEPT12 103.4 83. 0 95.5 83.3 SEPT 13 85o2 63.2 74.6 67.9 SEPT 14 6 3 . 9 52.9 59.7 55.5 SEPT15 0.0 0.0 0.0 0.0 SEPT 16 0.0 0.0 0.0 0.0 SEPT 17 0.0 0.0 0.0 0.0 SEPT18 0.0 0.0 0.0 0.0 SEPT 1 9 0.0 0.0 0.0 0.0 SEPT20 0.0 0.0 0.0 0.0  204. 5 196. 2 189. 8 190. 6 193. 4 197. 7 202. 8 195. 6 193. 4 196. 6 166. 9 135. 8 129. 3 120. 9 116. 4 112. 0 112. 4 118. 1 1 17.3 114. 2 106. 9 105. 7 106. 4 91. 8 81. 2 75. 3 69. 5 70. 6 109. 0 143. 8 149. 7 98. 5 84. 3 77. 1 69. 5 68. 1 71. 6 77. 7 85. 2 85. 8 71. 9 64. 5 57. 6 66. 4 122. 3 96. 6 70. 8 81. 0 77. 2 85. 0 148. 0 91. 3 72. 7 58.0 0. 0 0. 0 0. 0 0. 0 0. 0 0. 0  173 SEPT21 SEPT22 SEPT23 SEPT24 SEPT25 SEPT26 SEPT27 SEPT28 SEPT29  SEPT30 OCT 1 OCT 2  OCT OCT OCT OCT OCT  OCT OCT OCT OCT OCT OCT OCT OCT OCT OCT OCT OCT OCT OCT OCT OCT OCT OCT OCT OCT OCT OCT OCT OCT NOV NOV NOV NOV NOV NOV NOV NOV NOV NOV NOV NOV NOV NOV NOV NOV NOV NOV NOV  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 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19  0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0  35. 1 34. 2 32.8  33.8 100. 7 75.5 65.4 62.0 54. 9 45.6 46.3 47.2 47.9 80.8 58. 3 45.9 40.5 38.7 35. 2 33. 9 39. 0 36.5 35. 4 40.1 37.4 35.2 46. 5 41. 3 36. 7 35.0 33. 9 32. 7 32. 3 31.4 31, 1 32.2 31.3 30. 5 31.0 29.8 31. 0 35.2 43.6 44.1 39. 8 37.4 35.9 34.4 33. 2 31.9 31. 1  0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0  34.8 34.2 32.8  37.0 87.2 71.0 62.7 57.5  51.7 46.6 46.8 47.5 45.8 115.9 52. 9 44.1 39.2 36.9 34. 8 34.8 39.0 35.8 37.2 39,6 36,7 35.4 46. 1 39.5 34. 3 33.9 33. 3 32. 1 32. 1 31.0 3 1. 8 32.3 29.9 28.7 30.0 29.7 31. 2 40.5 44.9 42.9 38. 9 36.9 35.5 34.2 32. 9 31.7 31. 1  0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0  34.7 33. 5 32.8  46.5 79. 3 64.9 59.7 53. 3  47. 5 48.8 49.4 49.6 43.2 83.9 48.8 42.4 37.6 36.0 34.4 34.7 37.4 35.2 39. 6 38.3 35.9 37.0 43.7 37. 3 32.9 33. 7 33.0 31. 3 3 1.7 30.7 32. 2 32.1 28.3 28.8 29.9 29.7 31.8 43.4 45.3 41.8 38. 3 36.3 35.1 33.8 32. 5 31.4 31. 1  0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0  34.2 33.0 32.8  87.2 72. 7 61.3 58.8 50. 9  47.5 50.9 50.9 51.3 45.8 66.4 47. 5 41.2 36.2 35.5 34. 1 35.4 36.7 34.8 39.8 37.6 35.5 41.9 42. 4 36.3 34. 6 33. 5 32.8 31. 8 31.6 30.6 32. 2 31.4 29.8 29. 2 29.8 30.3 33. 4 43.1 44.9 40.5 37.9 36.0 34.8 33.4 32.2 31.2 31.0  0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0  34.7 33.7 32.8  51.1 85.0 68.2 61.6 55.9  50.4 48.0 48.3 48.9 45.7 86.7 51.9 43.4 38.4 36.8 34.6 34.7 38.1 35.6 38.0 38.9 36.4 37.4 44.7 38.6 34.6 34.0 33.3 32.0 31.9 30.9 31.8 32.0 29.8 29.3 30.2 29.9 31.9 40.6 44.7 42.3 38.7 36.7 35.3 34.0 32.7 31.6 31. 1  174  NOV NOV NOV NOV NOV NOV NOV NOV NOV NOV NOV DEC DEC DEC DEC DEC  20 21 22 23 24 25 26 27 28 29 30 1 2 3 4 5  31.7 42. 3 42. 1 38. 9 36. 9 35. 6 34. 1 33. 8 33.0 32. 2 31.2 30. 6 29. 7 29. 7 28. 8 28. 7  32.9 44. 1 41.5 38.0 36.7 35.0 34.3 33.7 32.9 31.8 30.5 30.4 29.7 29.4 27.4 28.6  35.4 43.7 40.7 37.6 36.3 34.6 34.1 33.4 32.8 31.6 29.4 29.8 29.9 29.0 27.8 27.8  38.6 43. 1 39.6 37.2 36.2 34. 3 34. 1 33.3 32.4 31.1 29.8 29. 7 29.8 28.9 28.4 28.2  34.6 43. 3 41.0 37.9 36.5 34.9 34. 1 33.5 32.8 31.7 30.2 30. 1 29.8 29. 3 28. 1 28.3  175  Table A . 2 . 2  Three hour mean discharge, Green River near Pemberton, 1972  Discharge (m Date  )  Time (Pacific Daylight) 0000  APR 19 APR 20 APR 21 APR 22 APR 23 APR 24 APR 25 APR 26 APR 27 APR 28 APR 29 APR 30 MAY 1 HAY 2 HAY 3 MAY 4 MAY 5 MAY 6 MAY 7 MAY 8 MAY 9 MAY 10 MAY 11 MAY 12 MAY 13 MAY 14 MAY 15 MAY 16 MAY 17 MAY 18 MAI 19 MAY 20 MAY 21 MAY 22 MAY'23 MAY 24 MAY 25 MAY 26 MAY 27 MAY 28 MAY 29 MAY 30 MAY 31 JON 1 JUN 2  s  35. 1 35.0 35.1 36.2 35.4 35.1 39.3 40.9 41.0 44.1 59.2 51.7 47.2 44.9 47.5 53.3 63.4 74.6 73.5 72.1 73.5 78.7 81.7 83.3 117.2 162.3 162.3 154.5 134.8 113.7 99.6 106.2 186.8 155.1 147.7 121.7 99.6 94.1 113.3 159.3 204.6 266.8 244.5 184.0 155.7  0300  0600  35.0 35.0 36.2 36.0 35.4 35.2 39.6 41.0 41.0 46.1 58. 1 50.7 46.3 44.9 48.4 54.5 65.9 76.9 72.7 72.7 74.9 79,6 83.6 85.5 125.9 169.3 165.5 154.0 133.3 111.5 99.6 115.9 197.7 152.8 151. 1 117.6 98.8 97.0 123.1 175.8 219.9 279.2 250.1 181.9 158.7  34.8 35.0 36.3 35.8 35.2 35.5 39.8 41.0 40.9 48.4 56. 8 49.9 45.8 44.7 49.0 55.5 67.9 78.1 71. 9 73.0 76.4 80.8 85.2 88.5 129.8 170.5 164.2 153.4 131.3 108.6 99.6 121.7 204.6 151.1 149. 9 113.7 98.4 99.2 129.8 178.5 221.5 281.3 244.5 178.5 158.1  0900  1200  1500  1800  34.8 34. 8 35.0 35.0 36.5 36.3 35.8 35.4 35.1 35.0 35.8 36.0 39.6 39.6 40.9 40.7 40.7 40.7 51.3 54.7 55.3 54. 3 49.2 48.4 45.1 44.9 44.6 44.4 49.2 49.4 56.2 56.4 68. 1 68. 1 77.8 76.9 71.0 70.2 72.7 72.4 77.2 77.5 80.8 80.5 85.2 84.6 91.9 94.4 131.8 1 3 0 . 3 166.7 159.9 162.3 1 5 7 . 5 150.5 146.0 128.8 1 2 5 . 0 105.8 103.0 98.8 96.6 124.5 1 2 5 . 0 1 9 3 . 3 182.6 148.2 1 4 5 . 4 146.0 141. 1 110.7 1 0 7 . 4 97.3 95.9 99.9 99.6 130.3 128.3 1 7 4 . 5 166.1 218.2 206.9 271.9 253.0 233.6 216.6 173.2 169.9 159.3 149.9  34.7 35.1 36.2 35.2 34.8 36.5 39.8 40.5 40.5 58.1 53. 3 47.5 44.6 44.6 49.4 56.4 67. 9 75.8 70.2 71.9 77.5 79.6 83.0 96.6 130.3 154.0 154.0 141.6 121.2 101.1 96.2 126.9 173.2 142.7 135.9 104.6 94.4 98.4 125.5 159.3 198.5 233.6 206.1 159.3 144.3  34.7 35.2 36.2 35.2 34.8 37.2 40.1 40.5 41.3 59.7 52. 9 47.5 44.6 45.3 50.1 58.1 69.5 75.2 70.5 71.9 77.8 79.3 82.4 97.7 136.9 153.4 152.8 137.9 118.5 99.9 96.6 131.8 166.1 141.1 129.8 102.2 93.0 98.8 125.9 159.9 206.1 224.1 193.3 154.5 140.0  2100  Mean  34.8 34.9 35.4 35.1 36.2 36.1 35.4 35.6 35.0 35.1 37.9 36.1 40.5 39.8 40.9 40.8 42.6 41.1 41.0 50.4 52.3 55.3 47.4 49.0 44.7 45.4 46.8 45.0 51.7 49.3 60.6 56.4 71. 9 67.8 74.4 76.2 71.3 71.4 72.4 72.4 78. 1 76.6 80.2 79.9 82.4 83.5 107.4 93.2 148.2 131.3 1 5 6 . 9 161.6 154.0 159.1 135.9 146.7 115.9 126.1 9 9 . 9 105.4 99.6 98.3 161.7 126.7 159.9 183.0 142.7 147.4 125.5 140.9 100.3 1 0 9 . 8 92.6 96.3 102.2 98.6 137.4 126.7 179.2 169.1 233.6 213.7 229.2 254.9 186.8 2 2 1 . 9 153.4 1 6 9 . 3 139.5 1 5 0 . 7  176  JOH JUH JUN JOH JOH JON JUN JUN JUN JUN JUN JUN JUN JUN JUN JUN JUN JUN JUN JUN JUN JUN JUN JUN JUN JUN JUN JUN JUL JUL JUL JUL JUL JUL JUL JUL JUL JUL JUL JUL JUL JUL JUL JUL JUL JUL JUL JUL JUL JUL JUL JUL JUL JUL JUL JUL JUL JUL JUL AUG  3 141.9 4 154.0 5 149.9 6 191.1 7 189.6 8 182.6 9 229.2 10 2 6 7 . 8 11 2 2 6 . 6 12 1 7 0 . 5 13 1 4 2 . 7 14 1 1 7 . 2 15 121.2 16 132.8 17 154.5 18 1 2 5 . 5 19 126.9 20 1 5 5 . 7 21 143.3 22 155.1 23 1 3 9 . 5 24 1 4 7 . 7 25 163.0 26 1 4 8 . 8 27 129.8 28 143.3 29 2 0 3 . 8 30 169.9 1 159.9 2 151.6 3 163.0 4 186.1 5 214.9 6 236.3 7 233.6 8 186.8 9 160.5 10 124.5 11 1 0 8 . 6 12 156.9 13 3 8 9 . 7 14 2 3 2 . 7 15 189.6 16 1 9 7 . 0 17 2 1 2 . 5 18 2 0 9 . 3 19 2 0 5 . 4 20 178.5 21 162.3 22 1 4 4 . 9 23 1 4 2 . 7 24 131.3 25 130.3 26 121.7 27 112.8 28 114.1 29 122.6 30 120.3 31 126.4 1 134.3  149.9 164.8 156.9 198.5 194.7 188.9 240.8 268.8 223.2 167.3 139.0 119.0 126.4 142.7 156.9 123.6 132.3 166.1 146.0 163.0 144.3 161.1 166.7 149.9 129.8 150.5 196.2 171.8 162.3 152.8 166.7 189.6 216.6 235.4 228.4 182.6 152.8 123.6 109.5 193.3 375.2 224.9 186.1 195.5 208.5 207.7 200.0 174.5 158.7 146.0 141.1 129.8 130.8 120.8 112.4 115.0 123.6 120.8 126.9 135.3  152.2 168.6 163.0 203.8 191.8 186.1 241.7 265.7 217.4 163.6 135.9 119.9 130.8 151.6 155.1 121.7 133.8 164.2 147.1 160.5 147.1 168.6 165.5 148.2 129,3 158.7 186.8 167.3 157.5 149.4 163.0 182.6 206.9 227.5 219.9 174.5 146.6 120.8 108.2 220.7 334.9 212.5 177.8 186.8 196.2 197.0 188.9 165.5 151.1 142.2 136.9 126.4 126.9 117.2 109.5 112.4 119.0 116.7 122.2 130.3  149.9 146.6 141.6 139.5 143.3 1 4 6 . 0 164.8 156.9 150.5 146.0 145.4 156.4 165.5 166.7 1 6 8 . 0 1 7 0 . 5 1 7 8 . 5 1 6 4 . 9 197.0 186. 1 1 7 3 . 8 171.8 176.5 187.3 186.1 1 7 7 . 8 1 6 9 . 9 167.3 1 7 1 . 8 181.1 181.2 172. 5 1 6 6 . 7 174.5 2 0 0 . 7 181.7 2 3 6 . 3 230.1 2 2 5 . 8 2 3 5 . 4 2 5 6 . 8 2 3 7 . 0 2 5 5 . 9 2 4 3 . 6 2 3 3 . 6 230.1 2 2 7 . 5 249.1 2 1 0 . 9 2 0 1 . 5 1 8 9 . 6 182.6 1 7 5 . 8 2 0 3 . 5 160.5 158. 1 155. 1 150.5 146.6 159.0 131.3 1 2 7 . 4 1 2 4 . 0 120.3 117.6 1 2 9 . 8 119.4 118.5 1 1 6 . 7 115.9 117.2 118.0 132.3 1 3 0 . 8 1 2 8 . 8 127.4 128.3 128.3 156.3 154.5 151.1 149.4 149.9 148.5 149.4 143.3 1 3 7 . 4 132.3 128.3 1 4 4 . 7 119.4 1 1 6 . 7 113.7 112.0 116.3 118.6 130.8 1 2 6 . 9 122.6 121.7 1 3 4 . 8 1 2 8 . 7 158.1 152. 8 147. 1 143.8 1 4 2 . 7 153.8 144.3 1 5 4 . 0 140.6 137. 4 135. 3 1 4 2 . 7 142.1 144.9 148.8 141.6 136.4 134.8 149.3 169.9 140.0 134.8 132.3 1 3 5 . 9 139.9 162.3 168.6 1 6 3 . 0 159.3 158.7 162.1 144.3 157.5 152. 8 149.4 147.7 158.1 128.3 140.0 135.3 131.8 130.3 141.1 171.2 128.3 126.9 128.3 1 3 4 . 8 129.5 179.2 183.3 1 9 4 . 0 2 0 5 . 4 2 0 8 . 5 176.8 159.9 170. 5 163.0 158. 7 163.0 177.6 151.1 153.4 146.0 144.3 151.1 158.0 142.7 143.8 137.9 135.9 1 4 2 . 7 148.9 155.1 137.4 132.3 133.3 147.1 143.3 177.1 1 4 7 . 7 143. 3 148.2 169.3 157.0 193.3 170.5 158.1 168.0 193.3 178.2 2 1 4 . 9 181. 9 179.8 195. 5 2 2 3 . 2 2 0 1 . 5 2 0 5 . 4 2 0 0 . 7 193.3 198.5 2 2 4 . 9 2 1 6 . 4 168.0 191. 1 182.6 181.2 185.4 2 0 3 . 4 139.5 166.1 163.6 162.3 161.7 170.7 117.6 134.3 128. 8 125.0 1 2 4 . 0 138.9 105.4 114.5 110.7 107.4 107.8 1 1 5 . 9 2 4 9 . 2 103.0 101.8 103.4 1 18. 1 107.3 301.2 2 7 1 . 9 3 0 4 . 6 3 7 2 . 4 3 7 9 . 5 2 6 8 . 5 199.2 277.1 2 5 8 . 8 2 4 7 . 3 2 3 9 . 0 3 0 2 . 9 169.3 1 8 8 . 9 1 7 7 . 8 177.1 1 8 4 . 7 1 9 9 . 7 175.8 160.5 156.9 167.3 187.5 174.4 182.6 168.6 167.3 184.0 2 0 7 . 7 185.3 184.7 171.2 166.1 177.1 197.0 188.9 177.1 177.1 1 7 3 . 8 182.6 196.2 191.1 155.7 168.0 160.5 170.5 175.1 180.7 144.3 1 4 7 . 7 141.6 145.4 1 5 5 . 7 158.1 137.4 137.9 133.8 133.3 137.9 144.9 129.8 132.3 128.3 129.3 140.0 137.6 121.7 125.0 120.3 120.8 1 2 6 . 9 130.4 122.2 117.2 114.1 115.9 1 2 5 . 0 122.7 112.8 117.6 114. 1 113.3 118.5 121.7 105.8 108.6 105.4 104.6 1 0 9 . 0 1 1 2 . 5 108.6 103.0 100.3 100.7 107.0 106.4 114.5 104.6 1 0 1 . 8 104.2 114.1 1 0 9 . 4 112.0 109.5 105.4 105.8 113.3 114.2 117.2 107.0 1 0 3 . 8 105.4 1 1 6 . 3 112.8 123.6 112. 4 108. 6 11 1.5 123. 6 118.6 1 1 7 . 6 1 1 3 . 3 113.3 120.3 123.5  177  AUG AUG AUG AUG AUG AUG AUG AUG AUG AUG AUG AUG AUG AUG AUG AUG AUG AUG AUG AUG AUG AUG AUG AUG AUG AUG AUG AUG AUG AUG SEP SEP SEP SEP SEP SEP SEP SEP SEP SEP SEP SEP SEP SEP SEP SEP SEP SEP SEP SEP SEP SEP SEP SEP SEP SEP  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 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  1 2 7 . 9 128.3 123.6 117.6 112.8 108.2 107.0 112.4 117.2 122.6 125.5 122,6 117.2 112.0 107.4 105.0 107.8 115.0 1 17.2 123. 1 121.2 116.7 1 1 1 . 5 1 0 7 . 0 103.8 103.8 113. 1 113.3 123.6 123.6 119.9 115.0 110.3 105.4 104.2 114.4 1 1 1 . 5 121.7 122.2 117.6 113.7 110.3 110.3 120.8 116.0 136.9 143.3 140.6 133.8 128. 8 125.0 127. 4 141.6 1 3 4 . 7 157.5 158.7 152.8 137.9 135.3 133. 3 132.3 140.6 143.6 151.6 151.6 147. 1 142.7 138.5 134.8 137. 4 147. 1 143.9 1 5 2 . 2 147. 1 139.0 129.8 122.2 116. 3 112.8 111.5 128.9 108.6 105.4 101. 8 9 9 . 6 100.3 99.9 97. 3 9 8 . 4 101.4 98.8 97.0 93.0 88.9 85.2 82. 1 79.6 79.0 87.9 79.0 78.4 76.9 74. 1 72. 1 70.2 69.7 73.7 69. 2 72. 1 73.0 71. 3 69.5 67. 1 68.8 67. 1 65.6 64.9 72. 1 74. 1 72.4 70.0 66. 1 65.9 68.4 67.9 69.6 72. 1 74.9 77.8 83.3 9 5 . 9 1 1 1 . 5 120. 3 115.9 94.0 1 0 9 . 0 104.6 98.4 94. 1 88.9 86. 2 85.5 94.0 85.5 85.5 83.3 80. 5 78. 1 75.8 73.5 74.9 78.0 72.7 79.0 78.4 75.5 72.7 68. 1 68. 1 73.5 70.2 73.2 79.6 80.2 78.7 76.9 74.9 85.9 75.8 78.6 76.9 91.9 9 1.2 8 9 . 5 88.9 93.3 86. 5 85.5 89.4 87.9 97.3 97.7 98. 1 95.5 88. 2 94. 1 93.8 91.6 88. 2 97.0 94.4 88.5 84.3 77.8 84. 9 80. 5 85.7 78.4 89.2 87.5 83. 3 79.9 74.9 84. 3 76.9 81.5 76.4 87.5 86.9 82. 7 78. 1 75.2 73.0 82. 1 79.9 73.8 87.5 85.5 79. 9 75.8 70.8 81.7 72.7 78.3 72.7 87.5 86.2 80.8 76.9 72.4 80.0 73. 5 76. 1 86.5 93. 3 9 1.2 8 5 . 9 81. 1 77. 8 76.4 87.5 84. 1 79.9 97.3 95.1 89.9 85.2 79.6 90.9 87.7 81.4 81.7 93.3 87.9 82.4 77.5 72. 1 78.6 74. 1 7 1 . 3 70. 2 75.8 73.8 69.5 65.9 61. 1 60. 1 63.2 66.6 63.2 69.7 67.4 63.4 60.4 57.0 65.9 62.7 58.8 58.8 64.6 61.8 58.6 70.0 64.1 70. ,5 6 8 . 4 59.5 59.9 73.2 69. 5 66. 1 63.9 63.2 77.5 69.4 74 ,9 66.9 78. 1 73.5 70.2 66.4 82. 1 73.9 81, ,7 67.6 71.3 76.4 73.8 71.0 72.7 75.2 84. 9 8 0 . 5 71.9 70.8 71.3 67. 1 63.9 59.2 56.8 63.9 73.8 61.5 57.9 56.0 54.9 52.9 50.7 50. 7 53.0 56.4 51.9 50. 3 52.7 52.7 52.3 57.0 53.8 52.1 53. 1 54.5 56.4 63.2 60. 6 57.7 50. 1 55.9 59.2 54. 1 51.9 50. 7 49.0 47.9 47.5 45. 3 44.6 46.9 49.8 46.5 44.7 45.4 4 5. 3 4 4 , 4 42. 4 41.9 43.7 44.7 43.2 4 1.8 46. 3 4 6 . 1 44, ,7 42.3 42. 1 43.7 43.2 43. 4 41.6 47.2 46.6 44, ,7 42.3 42.9 44.2 44.7 43.4 41.9 49.0 47.9 46, ,5 43.6 44. 1 45.8 47.4 44.9 42.9 50,7 48. 8 47, 5 4 5 . 4 44. 1 4 3 . 4 44. 1 46.6 48.4 52. 1 52.5 52, , 1 5 0 . 9 49.9 51.5 50.9 47.9 50. 5 50.9 48.6 47.5 44.6 43. 1 47.0 51.9 45.9 43.6 42.6 41.9 41.3 40. 1 39.9 4 1.3 4 1 . 3 42.8 40.5 42. 1 41. 3 40.7 39. 9 39.6 40.8 42.4 40.4 39.6 39.0 38.4 38.0 37.2 38.3 38. 1 39.5 37. 4 37.2 51.9 57.0 55.3 49.9 51.6 42. 1 47.5 57.9 51.3 47.9 47.5 47.5 47. 2 46.8 47.6 49.2 47.4 47.0 46. 1 44.7 43.4 41.9 41.0 43.5 46.6 42.8 41.5 40.4 39.9 39.5 38.4 37.7 39.2 40.7 38.9 38.0 37.3 36.7 36.2 35.8 35.6 36.4 37.6 36.0 35.8 35.4 35. 1 34.7 33.9 33.8 34.6 35.5 34.4 33.8 a  178  Appendix 2.2  Suspended and dissolved sediment collection and anal/sis technique  During the melt season sampling was carried out in the single channel of Lillooet River just upstream from the distributaries of its delta with a U . S .  D-49  sampler and during low water of autumn through spring at various locations near the river mouths with a U . S .  DH-48 sampler (Stichling, 1969).  samplers from surface to bottom and return.  Both are depth integrating  Two samples were taken each time.  Samples were retained until analysis in one pint (0.471 I) glass bottles in which they were taken.  Sample size varied between about 0.27 and 0.40 I.  Smaller  samples (insufficient for analysis) and those larger (involving the danger of water loss through the air escape tube of the sampler) were discarded at the time of sampling; the bottle was washed out carefully with river water, and reused immediately. discarded were samples containing obvious amounts of bed material.  Also  Samples were  capped with cardboard, returned to the laboratory within a week, weighed  (W^),  and stored in the dark for at least two months before analysis. Lake water samples were dipped by hand into a 0.91 I bottle. Suspended and dissolved analyses were carried out as follows: 1)  Samples were reweighed (WV,) and two 100 ml aliquots drawn off into two  tared evaporating dishes. 2)  A l l but approximately 50 ml of water were drawn off and discarded.  The  remaining water plus the settled suspended sediment were washed into a third dish after the bottle was reweighed 3)  (W^).  The three dishes were dried at 1 0 5 ° C until several hours after all traces of  water were gone from the evaporating dishes. 4)  Bottles were washed, dried, and reweighed  5)  Dissolved and suspended sediment concentrations were determined from the  following:  (W^).  179  D1  "  W  TD1  *  D2  "  W  TD2  "  W  TS  W  l  W  W  w  -  s  w VV  W1  V V  c  v  w  W2  =  C  D1  =  C  D2  =  C  D  =  D  C  =  V  "  W  S  T2  (g) (g)  -  W  / W  6  * io /v 6  D 2  m  +  C  D 2  (g)  4  (ml)  W2>  * io /v  D1  (g)  4  s  W  ( C  (g)  T3  ( W1  w  (mg •-')  c  (mg i-')  c  )/2  (mg ii-')  (w - w - w)  c  3  W  w  W  W  T1  -w -w  2  A  w  -  W  s  -  sc*  D  * 10"  s  4  (g)  6  (g)  c  1 o 6  /(  c W  wi)  (mg 1  where: c is the specific gravity of water in g ml  ' (1.00),  C is the concentration of dissolved (^) and suspended (^) sediment, is the correction for weight of dissolved sediment included in the suspended sediment aliquot (step 2), V  is the aliquot volume measured (^) corrected for evaporation  W  (Guy,  1969, p. 12),  (100 ml) and the aliquot volume  (^.),  is the weight of dissolved (p) or suspended (^) sediment in each aliquot,  W^^. is the corrected weight of the suspended sediment, Wj  is the tare weight, and Wyg  Wy^  are the weights of dissolved or suspended sediment plus tare,  is the weight of the water in the sample shortly after it was taken (^) and at the time of analysis ( ). 2  The difference (due to evaporation) was  usually 20 to 40 g , a significant amount.  180  Appendix 2.3  Table A . 2 . 3 Summary of moment measures - Lillooet River bed material  -7 -6 -5 -4 -3 -2 -1 0 12 15 16 19 22 24 26 30 33 38 41 43 48 51 52 55 58 61 64 67 70 73 76 79 82 86 89 94 97 100 103 106 109 112 113 114 115  Mean (JJ) 1159.0 525.7 464. 8 355. 3 1257. 4 755.4 434.0 809. 9 488.0 351. 4 732.5 352. 7 987. 7 546.6 1146.0 528.6 866. 6 228. 5 248. 8 863 0 1040.2 1149,2 1433.0 1249.5 755. 4 652.4 703. 3 635. 8 832. 2 1073. 8 789. 4 884.7 2631. 0 2284.5 1140.4 834, 4 1394.8 1286.0 640. 3 598. 3 591.6 841. 5 433. 2 778. 8 50 3 . 4 o  Variance (y) Skew 10.2 0.57 3.3 0.85 3.0 0.92 4.3 0.62 6.4 0.37 3.7 1.21 3.7 1.92 5.6 0.62 2.7 0.52 2.1 1.76 5.4 0.70 1.8 -0.30 3.7 0.91 3.0 1.03 5.8 0.43 2.6 1.06 4.4 0.68 2.0 -0.34 5.0 0.74 5.2 1.62 8.0 1.23 8.3 1.30 6.4 0.53 9.4 0.64 5. 1 1.48 2.8 1.28 3.3 1.38 3.5 1.69 3.8 0.70 5.7 1.00 4.8 0.83 1.14 6.5 12.3 0.07 5.6 -0.06 8.6 0.82 5.8 1.27 9.1 0.78 11.7 0.74 5.6 1.64 3.0 0.59 4.4 0.87 5.2 1.05 3.5 1. 19 4.2 0.87 2.2 0.87  Kurtosis 2.5 4.4 7.4 3.8 2.3 4.7 10.0 3. 1 4.4 12.4 3.7 5.3 4. 2 5.4 2.5 5.8 3.0 4.2 3.7 5.8 3.6 3.8 2.6 2.3 5.7 7.3 6. 1 7.9 4. 1 3.9 3.5 4. 1 1.6 2.3 2.6 4. 1 2.7 2.4 6.7 5.9 5.3 4.3 6. 1 4.3 7.5  %  Silt 0.0 0.0 0. 1 0.4 0.0 0.0 0.0 0.0 0.0 0.0 0. 1 0.0 0.0 0.0 0.0 0.0 0.0 0. 1 2. 1 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0. 1 0. 1 0. 1 0.0 0.0 0.0 0.0  181  Appendix 5.1  Sampling devices used in the Lillooet Lake sediment study  Two samplers were used in this study: a) mm).  a standard, weighted Ekman grab sampler, 6" x 6" x 9" (152 x 152 x 229  This sampler offers the advantage over most others of minimum sample disturbance  as the side walls are rigid (Sly,  1969).  Samples up to 150 mm long were obtained in  fine grained sediments but grains coarser than fine sand prevented sample retention. After removal from the sampler, the samples were stored in 105 mm diameter tubes (48 fluid ounce (1.4 I) 'tin cans', top and bottom removed), returned to the laboratory, and frozen for later analysis.  Some sagging of the unsupported sample occurred during  transfer to the storage tube resulting in down-bowed horizontal surfaces (see for example, figure 5.18). b)  a gravity corer of dimensions 1.97 m barrel length, 89 mm outside  diameter steel casing, 64 mm inside diameter plastic lining, 60 mm inside diameter nose cone and core catcher.  Design ratios (Hvorslev, 1949, pp. 105-108) were:  Inside clearance, C . = i Outside clearance, C  5.3% = 0  o Area or Kerf ration, C = 142% ' a Length-width ratio,  = 28.8  Although C , C , and C j values were well above those recommended by Hvorslev Q  (1949) and Richards and Parker (1967), they compare well with values given by Ross and Riedel (1967) and Rosfelder and Marshall (1967) for corers currently in use by many oceanographers. Provision was made for loading up to 45 kg on the guiding fins.  The corer  had no check valve but there appeared to be no core loss above the catcher. disturbance did occur in the upper centimetre of most cores.  However,  Nose cone cutting angle  was 25 and 3 0 ° (reasonable in view of the work of Patton and G r i f f i n , 1969, although not conforming to Hvorslev's, 1949, requirements) and the cone was flush with the  182  outside of the core barrel. site.  Cores taken for mass property analysis were extruded on  Cores for stratigraphic examination were extruded into trays on site, d r i e d , cut,  photographed and samples  returned to the laboratory in the liner for later work.  Both samplers were operated with a hand or gasoline driven winch from a floating platform much like that described by Anderson and Hess (1969).  183  Appendix 5.2  Assessment of errors involved in sampling lake sediment  Physical mass property determinations of sediment requires that the sample be as nearly undisturbed as possible. A completely undisturbed sample probably does not exist, but at least some of the sources of error can be evaluated or avoided. 1)  Errors resulting from sampler design: i)  passes.  wall thickness: The core tube must displace sediment into which it  Hvorslev (1949)  recommends that the area of kerf ratio be less than 1 0 % to  minimize disturbance, but many oceanographic corers in present use are over 1 0 0 % „ External sediment is likely to be forced into the core tube in the early stages of penetration when the vertical pressure due to the weight of the core in the tube is less than the pressure exerted below the corer by its weight.  This may result in lengthening  of the core near the top and a characteristic upward bulbous disturbance of the core (Hvorslev,  1949, p. 95).  During later stages of penetration, the sediments below the  corer may be bowed down, and weak or sensitive layers altered or completely squeezed out before entry into the core tube.  O n the other hand, this force may increase the  strength of the sediment below the sampler allowing a longer sample to be collected before serious disturbance occurs.  This may be the reason Ross and Riedel (1967) were  not able to find a relation between area ratio and shortening ratio. ii)  internal and external wall friction: Generally the nose cone is  made with outside and inside diameters slightly larger and smaller respectively than those of the core tube and liner above, in order to decrease friction.  Friction of the  outside wall adds to the force created by the frontal area of the core tube. friction is the principal cause of core shortening.  Inside  Shortening or penetration ratios  (core length + penetration length) are normally in the range 40 to 7 0 % .  When  inside friction is high or sediment weak the sides of the core may be dragged downward in parabolic arches (Hvorslev,  1949, figure 90).  Shortening is not usually due to  compaction of the sediment associated with water loss.  Various workers have studied  184  the manner in which shortening occurs.  Emery and Dietz (1941) in an early study  found that core shortening became progressively greater as a glass tube was forced into soft clay whose water content and strength were constant throughout, but that in most oceanic sediments core shortening was linear.  They suggested that in natural  sediment, the increase in strength of the sediment with depth offsets the increase in inside wall friction so that shortening rate remains nearly constant at least to several metres penetration.  Later work by Ericson and Wollin (1956) and Emery and Hulsman  (1964) confirmed that shortening was approximately linear but Richards and Parker (1967) suggested that, depending on corer design and sediment type, there was little shortening in the upper section of the core (a length of 10 to 20 inside diameters according to Hvorslev, 1949), but that beyond this length shortening occurred at an approximately linear rate. A large inside clearance ratio w i l l likely reduce the inside wall friction, but too large a clearance may allow the sediment to slump into the space along the walls of the liner, leading to increased disturbance and shortening. iii)  the angle on the cutting edge: previously thought to be significant  in determining penetration and disturbance was shown by Patton and G r i f f i n (1969) to be unrelated to these parameters. 2)  Errors resulting from sampling procedure and sediment characteristics:  Hvorslev (1949) determined that quick penetration in one continuous movement without rotation produced the longest, least deformed cores.  During withdrawal considerable  suction may develop at the bottom of the core tube which may cause the core to be drawn back out of the tube particularly if there is no check valve in the upper section of the core.  Isaacs and Brown (1968) suggest a tube to connect the bottom of the  corer to the free water above the sediment to equalize pressure during withdrawal. However, in Lillooet Lake cores, sediment was frequently retained in the nose cone below the tightly closed catcher indicating that suction during withdrawal was not significant.  185 Cores brought up from great depth may hydrostatic pressure.  For example, in cores brought from 130 metres depth the pore  water w i l l expand approximately 6 % . and weaken cores.  be deformed due to changes in  Gas coming out of solution may badly disrupt  Shocks and vibration due to rough handling may be significant  in altering the strength particularly of sensitive sediments.  Bacteria growth in warm,  light surface conditions may significantly weaken the sediment as well as destroying or masking structure . The sediment characteristics are important in determining the success of coring.  Coarse sand with almost no cohesion is most difficult to core and retain  successfully, whereas uniform fine sediments are usually cored most successfully. Sediments alternating between weak and strong layers often are most deformed by coring as the weak sediment is forced out while the stronger breaks up irregularly and will be included in a mass of remolded weak sediment.  Patton and G r i f f i n (1969)  pointed out that the characteristics of a given sediment that determine the performance of the corer and its success in obtaining relatively undisturbed samples are highly variable and somewhat difficult to determine.  They note that much more investigation  into corer performance is necessary before analytic assessment of the degree of disturbance and validity of mass property determination can be made.  186 Appendix 5.3 Table A . 5 . 1 Summary of moment measures # 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 35 36 37 38 39 40 41 42 44 45 46 47 48 49 50  Lillooet Lake bottom sediments,  Mean (jj) 7. 1 8. 5 8. 5 9.8 14. 4 22. 1 7. 1 11.0 10. 2 10. 8 10. 5 10.4 11.2 10.9 12.0 13.5 12.0 13.6 11. 3 8.4 22. 2 23.6 | 22.4 15.1 18.0 15.6 21. 4 16.3 22. 1 25.0 28. 2 25.6 21.3 20.6 29. 8 26.4 28. 5 36.4 38. 6 37.7 48.5 70.0 46. 4 52. 1 113. 1 41.4 76.2 90.0 60. 3 1  Variance 10.0 11.2 11.3 10.6 16.0 18.2 9.5 10.6 9.8 9.6 10.9 8.6 10.0 10.0 12.7 12.3 10.6 12.8 10.7 10.3 11.6 11.5 11.4 12.8 11.9 10.4 11.8 10.7 18.2 10.4 8.2 8.5 10.0 7.5 8.0 7.5 7.7 7.0 6.4 5.2 8.2 4.9 4.4 4.2 5.2 6.9 5.2 2.4 3.8  Skew -0.21 -0.00 -0.23 -0.38 -0.25 -0.52 - 0 . 19 -0.40 -0.43 -0.26 -0.53 -0.57 -0.43 -0.63 -0.62 -0.54 -0.56 -0.59 -0.63 -0.31 -0.99 -1.07 -0.96 -0.56 -0.70 -0.68 -0.75 -0.56 -0.52 -1.12 - 1 . 16 -1.22 -0.94 -0.90 -1.33 -1.18 -1.11 -0.64 -1.26 -1.39 -1.35 -1.57 -0.74 -0.68 - 1 . 34 -0.97 -0.84 -0.45 -1.85  1970  Kurtosis 2.7 2.9 2.7 2.8 2.4 2.4 2.7 3.0 3. 1 3.2 2.9 3.4 3. 1 3. 1 2.8 2.9 3.0 2.8 3. 1 2.8 3.4 3.7 3.6 3.0 2.9 3. 1 3. 1 3. 1 2.4 3.4 4.3 4. 3 3. 3 3.7 4. 3 4.2 4. 2 2.7 4.6 5.2 5.2 8.2 3.9 4.3 4.8 4.4 3.6 10.0 8.6  %  Sand  3.2 8.7 6.4 5.9 24.7 31.4 3. 1 9.0 5.9 9.2 6.7 4.6 8.0 5. 1 8.6 12.9 7.6 12.2 8.6 4.9 20.0 22. 1 22.4 16.9 19.7 10. 3 25. 1 14.8 31.3 23.5 23.5 20.9 17.8 10.3 25.0 16.9 24.9 36.0 35.0 29.2 51.3 71.2 42.2 48.8 83.7 40.6 69. 1 88.6 64.2  187 Appendix 5.4  Lillooet Lake sediments: mass properties and shear strength  Samples of Lillooet Lake sediments were analyzed for the basic sediment properties: (p^  material density water content void ratio porosity  (w = W v  (e = V  (n = V  bulk density ( p shear strength ( r ) sensitivity  v  /V  w  / W ) s'  / V a- p w) s / s ' f  = e/(l  = W/V  + e) )  = p  / e)  of 'undisturbed' ( ) and remolded ( ) sediment  (s = t^/  r  f )  where W is the sample weight - total (.),  of water ( ) and of sediment (W  = W\ - W  V is the sample volume with subscripts as for weight according to procedures outlined below. Material density:  From 13 of the cores preserved for stratigraphic analysis  (figure 5.5) samples were analyzed by standard pyconometric technique to give the results listed in table A . 5 . 2 . Water content:  As each of the 17 cores (figure 5.5) were extruded from  their liners, 150 ml samples were taken, weighed, oven dried for 12 hours at between 100 and 1 3 0 ° C , reweighed, and the water content calculated.  Since the material  density variance was so low (Table A . 5 . 2 ) , it was assumed that the mean value (2.73 g ml ^) could be used for all lake sediments in the calculation of the other mass properties.  Results are listed as table A . 5 . 3 .  Shear strength: was built (figure A . 5 . 2 ) .  Following Moore (1961)  a s i m p l e , unconfined shear box  Although this technique of producing shear failure does  not have some of the disadvantages involved with the use of the shear vane (Monney, 1971;  Singler, 1971; Andrews and C e p e k , 1972),  the failure surface is determined  by the apparatus and some disturbance occurs on coring (appendix 5.2), and on  ),  I-  63 m m  Plastic ring o f same inside diameter as the liner o f the gravity corer pulley 50 mm  shear:  ^surfacelight thread  N  N  co co  Water progressively added to container provides shear stress Figure A . 5 . 2  Q u i c k direct shear apparatus.  189 Table A . 5 . 2 Material density of Lillooet Lake sediments  Core #  Depth below sediment surface (m)  test* test* test* 61C 61C 50B' 50B' 50B' 41 41 61 71 75 75 81 95 101 111 121  Material _^ density (g ml )  0.1 0.4 0.7 0.1 0.5 0.1 0.2 0.4 0.0 0.7 0.0 0.0 0.0 0.6 0.0 0.0 0.0 0.0 0.0  2.71 2.74 2.72 2.73 2.76 2.70 2.71 2.72 2.74 2.74 2.74 2.75 2.73 2.75 2.73 2.74 2.71 2.74 2.76 Mean  2.73  •4 2.91 x 10"  Variance * test core taken near raft 2  transfer of the sediment to the shear box.  The latter disturbance was minimized by  making the inside diameter of the shear box rings the same as that of the core liner so that the sediment could be trimmed off at the desired depth as it was extruded and pushed directly into the box.  The two rings were separated slightly  began and friction of the pully was small enough to be ignored.  before the test  The weight of water  necessary to produce failure was measured with a spring balance. The sediment was thoroughly mixed and returned to the shear box for the remolded shear strength test.  Results of the work are summarized inTable A . 5 . 4 .  190  Table A . 5 . 3 Mass properties of Lillooet Lake sediments  Core ^  30A«  30B'  3 0 C  40A«  40B«  40C»  50A'  50B«  Depth of burial (m)  Bulk density (g ml-1)  Water content  Porosity  Void ratio  0.0 0.15 0.41 0.66  0.648 0.627 0. 587 0.472  1.77 1.71 1.61 1.29  0.639 0.631 0.616 0.563  1.54 1.60 1.70 2 . 12  0.0 0 . 18 0.46 0.66  0.657 0.550 0.590 0.531  1.80 1.50 1.61 1.45  0.642 0.601 0.617 0.592  1. 52 1.82 1.69 1.88  0.0 0 . 15 0.46 0.71  0.658 0.661 0.671 0.508  1.80 1.81 1.83 1.39  0.643 0.644 0.647 0.581  1.52 1.51 1.49 1.97  0.0 0 . 15 0.46 0.68 0.84  0.654 0.631 0.527 0.551 0.439  1.79 1.73 1.44 1.51 1.20  0.641 0.633 0.590 0.601 0.545  1.53 1.58 1. 90 1. 82 2.28  0.0 0.15 0,46 0,68 0.86  0.647 0.618 0.522 0.553 0.487  1.77 1.69 1.43 1.51 1.33  0.639 0. 628 0.588 0.602 0.571  1. 54 1. 62 1.92 1.81 2.05  0.0 0 . 15 0.41 0.68 0.89  0.700 0.630 0.558 0.592 0.493  1.91 1.72 1. 5 3 1.62 1.35  0.657 0.633 0.604 0.618 0.574  1. 43 1.59 1.79 1.69 2.03  0.0 0.15 0.46 0.74  0.548 0.550 0.570 0.430  1.50 1.50 1.56 1. 18  0.600 0.600 0.609 0.540  1.82 1.82 1.76 2.32  0.0 0.15 0.46 0.66 0.81  0. 546 0.585 0.503 0.452 0. 396  1.49 1.60 1.38 1.24 1.08  0.599 0.615 0.579 0.553 0.520  1.83 1.71 1.99 2.21 2.52  191  5 0 C  60ft«  60B«  60C  70&«  70B'  80ft'  80B-  80C  0.15 0.46 0.76  0.676 0.536 0.427  1.85 1.46 1.17  0.649 0.594 0.539  1.48 1.87 2.34  0.0 0.15 0.46 0.71 0.89  0.689 0.625 0.547 0.536 0.590  1.88 1.71 1.49 1.46 1.61  0.653 0.631 0.599 0.594 0.617  1.45 1.60 1.83 1.87 1.69  0.0 0.15 0.46 0.74  0.733 0.642 0.601 0.477  2.00 1.75 1.64 1.30  0.667 0.637 0.621 0.566  1.36 1.56 1.66 2.09  0.0 0 . 15 0.46 0.76  0.680 0.620 0.532 0.536  1.86 1.69 1.45 1.47  0.650 0.629 0.592 0.594  1.47 1.61 1.88 1.86  0.0 0.15 0.46 0.68 0.84  0.657 0.600 0.590 0.584 0.451  1.79 1.64 1.61 1.60 1.23  0.642 0.621 0.617 0.615 0.552  1.52 1.67 1.69 1.71 2.22  0 . 15 0.46 0.71 0.80  0.576 0.592 0.471 0.413  1.57 1.62 1.29 1.13  0.612 0.618 0.563 0.530  1.74 1.69 2.12 2.42  0.0 0 . 15 0.46 0.71 0.86  0.726 0.623 0.550 0.473 0.468  1.98 1.70 1.50 1.29 1.28  0.665 0.630 0.600 0.564 0.561  1.38 1.60 1.82 2.11 2.14  0.0 0 . 15 0.46 0.66 0.76  0.778 0.631 0.580 0.516 0.475  2.13 1.73 1.58 1.41 1.30  0.680 0.633 0.613 0.585 0.565  1.28 1.58 1.73 1.94 2.10  0.0 0 . 15 0.46 0.71 0.96  0.828 0.702 0.579 0.595 0.482  2.26 1.92 1.58 1.62 1.32  0.694 0.657 0.613 0.619 0.569  1.21 1.42 1.73 1.68 2.07  192  Table A . 5 . 4 Strength - Lillooet Lake sediments  _2 Core  #  ———  Depth of burial (m) 0.02  30B«  30C»  40A»  40B'  40C  50A •  50B«  50C«  60A«  Strength g cm Undist. Remolded  Sensitivity  Oo 3 0  0.61  7.57 1 0 . 10 14.36  2.27 2.52 5.68  3.33 4.01 2.5 3  0.05 0.30 0.61  6.94 10.89 19.41  2.52 2.84 4.57  2.75 3.83 4.25  0.08 0.33 0.63  8.39 8 . 36 11.52  1.42 2.84 4.73  5.9 1 2.94 2.44  0.05 0.30 0.61 0.79  10.26 11.99 11.05 20.20  3.63 4.79 4.57 8.99  2.83 2.50 2.42 2. 25  0.05 0.30 0.61 0.84  10.89 14.36 9.63 23„68  3.63 4.42 2.84 6 . 15  3.00 3.25 3.39 3.85  0.05 0 . 30 0.61 0.79  9„31 12,. 3 1 11.36 1 8 . 94  2.68 7. 26 5.36 5.05  3.4 7 1.70 2.12 3.75  0.05 0 . 30 0.61  10.89 13.73 18.78  1.57 3.31 4.42  6.94 4.15 4. 25  0.05 0 . 30 0.61 0.71  10.73 12.47 15.63 19.57  2.84 5.05 7.57 5.36  3.78 2.47 2.06 3.65  0.05 0.30 0.61  9.78 10.73 17.36  3.15 2.99 4.73  3.10 3.59 3.67  0.05 0.30 0 61 0.81  9.47 12.78 16.73 2 0 . 84  3.47 4.42 6.94 7.89  2.7 3 2.89 2.41 2.64  o  193  60B«  60C«  70A»  70B'  80A'  80B»  80C«  0.05 0.30 0.61  10.73 10.42 16.73  2.52 3.78 4.89  4.26 2.76 3.42  0.05 0.30 0.61  7.89 14.20 21.15  1.42 3.78 8.05  5.56 3.76 2.63  0.05 0.30 0.61 0.76  10.89 15.47 19.26 1 2 . 15  3 . 15 5.36 5.68 5.99  3.46 2.89 3. 39 2.03  0.05 0.30 0.61 0.76  8.34 18.63 1 2 . 15 23.99  3.78 5.21 2.68 6.31  2.34 3.58 4.53 3.80  0.05 0.30 0.61 0.76  12.63 1 6 . 10 19.73 25.26  3.31 5.99 5.84 5.68  3.82 2.69 3.38 4.45  0.05 0.30 0.61 0.71  1 0 . 10 1 6 . 10 22.73 17.52  3.47 5.99 4.42 6.31  2.91 2.69 5 . 14 2.78  0.05 0.30 0.61 0.89  9.78 1 5 . 30 17.50 23.00  3.47 5.99 7.20 8.20  2.82 2.55 2.43 2.80  

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0093262/manifest

Comment

Related Items