THE INFLUENCE OF GLACIER CHANGE ON SEDIMENT YIELD, PEYTO
BASIN, ALBERTA, CANADA

by

Theodore John Mlynowski
B.Sc., University o f Northern British Columbia, 2008

THESIS SUBMITTED IN PARTIAL FULFILLMENT OF
THE REQUIREMENTS FOR THE DEGREE OF
MASTER OF SCIENCE
IN
NATURAL RESOURCES AND ENVIRONMENTAL STUDIES
(GEOGRAPHY)

UNIVERSITY OF NORTHERN BRITISH COLUMBIA
September 2013
© Theodore John Mlynowski, 2013

UMI Number: 1525691

All rights reserved
INFORMATION TO ALL USERS
The quality of this reproduction is dependent upon the quality of the copy submitted.
In the unlikely event that the author did not send a complete manuscript
and there are missing pages, these will be noted. Also, if material had to be removed,
a note will indicate the deletion.

Di!ss0?t&iori Publishing
UMI 1525691
Published by ProQuest LLC 2014. Copyright in the Dissertation held by the Author.
Microform Edition © ProQuest LLC.
All rights reserved. This work is protected against
unauthorized copying under Title 17, United States Code.

ProQuest LLC
789 East Eisenhower Parkway
P.O. Box 1346
Ann Arbor, Ml 48106-1346

Abstract
The relation between sediment yield and glacier fluctuations at timescales less than a century
remains uncertain. The primary goal o f this study was to assess the influence o f glacier
activity on sediment yield within the Peyto Lake watershed. The research focused on a small
alpine watershed in the Rocky Mountains o f Alberta containing Peyto Glacier and the
proglacial Peyto Lake. Using photogrammetric methods 1 determined changes in length,
area, and volume of Peyto Glacier from a topographic survey map (1917) and 18 sets o f
aerial photographs (1947 - 2005). I also collected 18 sediment cores from Peyto Lake that
consists o f laminated, silt-clay couplets which can be shown through 137Cs activity to be
clastic varves. Varve thickness and sediment properties were combined to produce an annual
record (1917 - 2010) o f specific sediment yield (SSY) for the watershed. I then compared
the SSY record to dimensional changes o f Peyto Glacier as well as available mass balance
records, hydrometric records, and climate records over the study period (1917 - 2010). Over
the period 1917 - 2005, Peyto Glacier retreated 2198 ± 18 m, shrank 4.0 ± 0.9 km2, thinned
44 ± 31 m, and lost 581 ± 404 * 106 m3 water equivalent (w.e.). I measured an additional 85
± 4 x 106 m3 w.e. o f ice loss from thinning ice-cored moraines adjacent to the glacier. Over
the period 1917 - 2005 SSY averaged 446 ± 176 Mg km2 yr'1, which is among the highest
measured yields in the Canadian Cordillera; however, this value is relatively low for
glaciated basins worldwide. The SSY record has a poor relation to short-term dimensional
changes of Peyto Glacier, likely due to the complexity o f sediment transfers in proglacial
environments. Long-term trends in SSY are hypothesized to arise from increasing (1870 1940) and decreasing (1970 - 2010) glacier contribution to streamflow over the past century.

Table of contents
Abstract..............................................................................................................................................ii
Table o f contents..............................................................................................................................iii
List of tables..................................................................................................................................... vi
List o f figures.................................................................................................................................. vii
List of appendices.............................................................................................................................x
Acknowledgements..........................................................................................................................xi
1.

Introduction........................................................................................................................... 1
1.1

Research motivation.......................................................................................................1

1.2

Thesis objectives........................................................................................................... 2

1.3

Thesis outline.................................................................................................................3

2.

Literature review.................................................................................................................. 4
2.1

Sediment yield................................................................................................................4

2.2

Sediment yield and glaciers..........................................................................................6

2.3

History of Peyto Glacier................................................................................................8

2.4

Sedimentology within the Peyto Lake watershed.....................................................10

3.

Study area and methods.....................................................................................................12
3.1

Study area......................................................................................................................12

3.2

Methods for estimation o f glacier dimensions..........................................................15
3.2.1

Data sources and preparation..............................................................................15

3.2.1.1

Interprovincial Boundary Commission Survey m ap................................ 15

3.2.1.2

Aerial photographs...................................................................................... 16

3.2.1.3

Supplemental data (mass balance, discharge, clim ate)............................19

3.2.2

Measuring glacier surface and extents.............................................................. 21

3.2.2.1

Stereo model developm ent......................................................................... 21

3.2.2.2

Data collection from stereo models........................................................... 23

3.2.2.3

Stereo model quality control......................................................................24

3.2.2.4

Glacier change analysis.............................................................................. 26

3.2.2.5

Glacier volume change conversion........................................................... 28

Methods to estimate sediment yield.......................................................................... 30

3.3
3.3.1

Field sampling...................................................................................................... 30

3.3.2

Laboratory analysis............................................................................................. 31

3.3.2.1

Photography..................................................................................................31

3.3.2.2

Bulk-physical properties............................................................................. 32
iii

3.3.3

Core chronology.................................................................................................. 34

3.3.3.1

Cesium-137 analysis................................................................................... 34

3.3.3.2

Master chronology...................................................................................... 34

3.3.4

4.

Sediment yield calculations................................................................................ 36

3.3.4.1

Spatial sediment interpolations.................................................................. 36

3.3.4.2

Lake trap efficiency.................................................................................... 37

3.3.4.3

Sediment yield............................................................................................. 37

Results................................................................................................................................. 39
4.1

Glacier dimensions...................................................................................................... 39
4.1.1

Stereo model verification and bias removal......................................................39

4.1.2

Data completeness............................................................................................... 40

4.1.3

Changes to glacier length and area.................................................................... 42

4.1.4

Observed volume change................................................................................... 45

4.1.4.1

Peyto Glacier...............................................................................................45

4.1.2.2

Lateral moraines......................................................................................... 49

Sediment Analyses...................................................................................................... 50

4.2
4.2.1

Description o f sediments.....................................................................................50

4.2.2

Bulk-physical properties.....................................................................................51

4.2.2.1

Horizontal trends..................................................................................... 51

4.2.2.2

Vertical trends.............................................................................................53

4.2.3

Cesium- 137 (137C s )........................................................................................... 54

4.2.4

Varve thickness....................................................................................................55

4.2.5

Sediment yield calculations................................................................................ 58

4.3

Environmental controls on sediment yield................................................................61

5.

Discussion........................................................................................................................... 65
Changes to glacier length, area, and volume............................................................ 65

5.1
5.1.1

Length................................................................................................................... 65

5.1.2

A rea...................................................................................................................... 65

5.1.3

Volume..................................................................................

66

5.2

Sediment yield............................................................................................................. 68

5.3

Influence of glacier change on sediment yield......................................................... 71

5.4

Source o f errors and limitations o f study.................................................................. 76
5.4.1

Digital photogrammetry and measurements.....................................................76

5.4.2

Sediment yield......................................................................................................77
iv

6.

Conclusion and suggestions for further study

Literature cited..............................................................
Appendices....................................................................

List of tables
Table 3.1 Aerial photographs and metadata used in glacier analyses....................................... 18
Table 3.2 Pearson correlation coefficient (r) and significance (p) for checkpoint residuals
versus topographic predictors (Elevation, Easting, and Northing). The largest correlations are
bolded and the respective topographic biases were removed from the models........................25
Table 3.3 Information used to create stereo models and assess their accuracy and error........28
Table 4.1 Study periods categorized by the percent o f glacier surface measured in the stereo
models.............................................................................................................................................. 41
Table 4.2 Mean rate of length and area changes for Peyto Glacier............................................45
Table 4.3 Pearson correlation coefficient (r) for Mistaya River discharge versus net mass
balance and standardized varve thickness or SSYspiine. p < 0.05................................................ 64
Table A.l Measured (white), interpolated (light grey) and averaged (dark grey) varve
thicknesses (cm) for the period 1917-2010................................................................................ 92
Table A.2 Mean bulk-physical properties for samples cores for the period 1917 - 2010........95
Table B.l Specific Sediment Yield interpolated by Thiessen polygons and regularized spline.
.......................................................................................................................................................... 98
Table B.2 Core sample and SSY correlation table. Bolded values are significant (p < 0.05).
.......................................................................................................................................................... 99

List of figures
Figure 3.1 Peyto Lake watershed with the 2005 glacier extents................................................14
Figure 3.2 (A) Geo-rectified IBCS map #17 using 5th order polynomial transformation
model with GCPs (+ symbols). (B) Additional adjustment to map using a thin plate spline
(TPS) transformation model with additional GCPs (x symbols). Maps show the
transformation of the ridgeline discrepancy before the TPS transformation (red line) and after
the TPS transformation (green line).............................................................................................. 16
Figure 3.3 The amount o f glacier surface that could be measured each year based on
photographic coverage o f the glacier (solid black bars) and the amount o f glacier surface that
was measured (stripped bars).........................................................................................................24
Figure 3.4 Example o f checkpoints measured in the 1991 air photos (A) before bias removal
and (B) after bias removal..............................................................................................................25
Figure 3.5 Peyto Lake bathymetry (adapted from Chikita et al., 1996). The locations o f the
retrieved sediment core samples are shown. Note: core IO-Peyto(Ol) was taken from the
same location as core 10-Peyto(02). The small arrows indicate the primary inflow and
outflow of Peyto Creek...................................................................................................................30
Figure 3.6 Example o f wet (top) and partially dry (bottom) laminated sediment. The red and
blue lines on the dry sediments are measurements from the UTHSCSA Image Tool software.
The water-sediment interface is towards the right...................................................................... 32
Figure 4.1 Box plot showing the absolute elevation difference o f check points after bias
removal. Differences are between individual years and the 2005 reference data....................40
Figure 4.2 Changes in ice volume for Peyto Glacier categorized by the completeness of
surface points measured................................................................................................................. 42
Figure 4.3 Extents o f Peyto Glacier digitized from aerial photographs and the IBCS map
throughout the period 1917 - 2005................................................................................................ 44
Figure 4.4 Comparison of measured volume changes of w.e. (solid black bars) and measured
volume changes o f w.e. with seasonal melt adjustments (stripped bars)...................................46
Figure 4.5 Rates o f volume change for Peyto Glacier for periods with the most complete
data................................................................................................................................................... 47
Figure 4.6 Peyto Glacier elevation changes for 1917 - 1952, 1952 - 1977, and 1977 - 2005.48
Figure 4.7 Rate o f volume change for Peyto Glacier for periods with the second-most
complete data...................................................................................................................................49

Figure 4.8 Elevation change for the ice-cored moraines adjacent to Peyto Glacier from 1947
to 2005..............................................................................................................................................50
Figure 4.9 Spatial distribution of mean bulk-physical properties for Peyto Lake for the
period 1917-2010: (A) dry density, (B) wet density, (C) water content, and (D) organic
content. Mean values were calculated from the interpolated values between sub-sampled
locations down the core for every varve. The small arrows indicate the primary inflow and
outflow o f Peyto Creek...................................................................................................................52
Figure 4.10 Bulk-physical properties for core 1l-Peyto(N). Dry density (dark grey line) and
organic matter (light grey line) were sub-sampled at 2 cm intervals. Varve thickness was
interpolated (red line) near the top and measured (black line) downwards...............................54
Figure 4.11 Activity levels of Cesium-137 (137Cs) in lake sediments retrieved from core 10Peyto(02). Vertical error bars indicate counting errors (± 1 a) in 137Cs activity determination.
Sediment samples were taken from the centre o f identified varves and converted to calendar
years (AD)........................................................................................................................................55
Figure 4.12 Varve thicknesses measured from all collected cores. Proximal cores (left) were
collected up to 316 m from the delta and distal cores (right) were collected up to 2215 m
from the delta. Note: measurements from cores F, A, and R, were done between marker
beds, so varve thicknesses denote average rates.......................................................................... 56
Figure 4.13 Extended varve record from two cores. Vertical black lines delineate periods
1828 - 1916 and 1917 - 2003. Horizontal red lines show the average varve thickness for
those periods.................................................................................................................................... 56
Figure 4.14 Photographs of abnormally thick varves from core 1l-Peyto(O)..........................57
Figure 4.15 Boxplots of measured varve thicknesses for the study period (1917 - 2010). Red
line denotes an 8 year running mean. The period 2000-2010 contains less than 14
contributing cores............................................................................................................................58
Figure 4.16 Examples o f how sediment thickness is interpolated across the lake. (A) Peyto
Lake bathymetry highlighting various depths where the rate of sedimentation is assumed to
be 0 m yr'1. (B) Sediment thickness in 1970 was uniformly distributed within each Thiessen
polygon that surrounds a sediment core. (C) Sediment thickness in 1970 using regularized
spline interpolation (25 m x 25 m pixels). Sediment values incorporate areal variation in
sediment density and organic content........................................................................................... 59
Figure 4.17 Master chronology of varve thickness variation. The y-axis is reported as
standardized departures from a mean of zero...............................................................................60

Figure 4.18 Calculated specific sediment yield for the Peyto Lake watershed when sampled
sediment thickness was interpolated using Thiessen polygons and spline interpolations.
Sedimentation rates were assumed to equal 0 m yr'1 when depths were shallower than 0 m,
10 m, and 20 m ................................................................................................................................ 61
Figure 4.19 Specific sediment yield calculated by using a regularized spline interpolation
where sedimentation was assumed to be near 0 m yr'1 at 0 - 20 m o f depth. The grey
horizontal line denotes the mean and the black dashed line denotes an 8 year running mean.
61
Figure 5.1 Specific sediment yield as a function o f percent glacier cover for lakes in British
Columbia and Alberta (Graph modified from Hodder et al. (2006)). The various symbols
represent different tectonic belts....................................................................................................70
Figure 5.2 Specific sediment yield as a function o f drainage area (Graph modified from
Church and Slaymaker (1989)). Specific sediment yield for Peyto Lake (star) is shown on
the graph...........................................................................................................................................70
Figure 5.3 The relation o f specific sediment yield to basin area. For comparison, the Peyto
watershed (star) is plotted among other watersheds with glacier cover (graph modified from
H alletetal., 1996)........................................................................................................................... 71
Figure 5.4 Conceptual diagram of the proglacial alpine systems and their linkages to
lacustrine sediment (Hodder et al., 2007)..................................................................................... 72
Figure C.l 2005 AT aerial photograph showing the previous measured extents o f the Peyto
Lake delta..................................................................................................................................... 100
Figure C.2 The rate o f volume change that Peyto Glacier has undergone for the period 1917
- 2005 is categorized by data completeness o f > 50 %. Glacier changes highlighted in pink
are within bounds o f the error term............................................................................................100
Figure C.3 The rate of volume change that Peyto Glacier has undergone for the period 1917
- 2005 is categorized by data completeness o f > 20 %. Glacier changes highlighted in pink
are within bounds of the error term............................................................................................. 101

List of appendices
Appendix A: Ancillary core information.....................................................................................91
Appendix B: Specific sediment yield results and sediment core correlation........................... 97
Appendix C: Measured delta progradation and ancillary volume change results................100

x

Acknowledgements
The completion o f this thesis could not have been possible without a number o f caring people
who have supported me and believed in me throughout this challenging and rewarding
process. I would first like to thank my supervisor, Dr. Brian Menounos, for providing a
diverse project and the opportunity to be his student. It is obvious that you are passionate
about science, but your dedication was shown with sweat, blood, and tears when you dragged
my sediment cores up Heartbreak Hill. I am grateful for your mentorship whenever I needed
it, your high expectations, and your encouragement.
For the past three years my wife, Caroline Mlynowski, has made great sacrifices so
that I could pursue my Masters. I am forever indebt to you for your unquestionable support,
encouragement, and devotion.
The three field work expeditions at Peyto Lake were the hardest and most rewarding
part of my Masters. I am appreciative o f all my field helpers, Caroline, Brian, Rob Vogt,
Mary Samolczyk, and Dr. Joel Cubley who endured long, cold days on the ice, before
dragging heavy sediment cores back to the truck. I am particularly thankful to Caroline and
Rob for their willingness to even get out o f the truck when it was -31 °C.
There were many facets of my Masters that required additional help, knowledge, and
services. The guidance and help from Teresa Brewis and Christina Tennant saved me many
hours of technical difficulties with Vr Mapping software. I also express gratitude to Cardinal
Systems, LLC for providing a license and support for their Vr Mapping software. I thank
Bill Holmes from the Alberta Air Photo Distribution and Daniel Brown from National Air
Photo Library for providing additional air photographs o f Peyto Glacier. I am also grateful to
Lyssa Maurer for showing me the required methodology and techniques to play with mud. I

appreciate Matt Beedle’s quick and in-depth replies to my questions about glacier mass
balance. Finally, I would like to acknowledge the generosity o f Flett Research Ltd. for
processing additional sediment samples free o f charge.
This project was made possible through funding obtained from the University of
Northern British Columbia (UNBC), Geological Society o f America (GSA), and the National
Sciences and Engineering Research Council of Canada (NSERC).

1.

Introduction
1.1

Research motivation

Annually laminated (varved) lake sediments provide a detailed record of environmental
change. These lake sediment records are particularly important in many regions o f western
Canada where instrumental records are short or absent. In alpine environments, for example,
varved lake sediments have been used as proxies for glacier fluctuations (referred hereafter
as glacier activity) which have then been used to infer paleoclimatic conditions (Denton and
Karlen, 1973; Leonard, 1997).
Although past climate conditions have been inferred from lake sediment records, such
inferences must be approached with caution. The climate-glacier-lake sediment system is
complex and strong connections among climate, glaciers, and lakes are difficult to
demonstrate (Jansson et al., 2005; Hodder et al., 2007). Identifying a relation between just
two variables such as glacier activity and lake sedimentation is challenging enough (Leonard,
1997; Loso et al., 2004). Such difficulties may arise from glacier records being temporally
limited (e.g., Loso et al., 2004) or incomplete (e.g., Leonard, 1997).
Previous studies have considered how glacier activity influences sediment production
(e.g., Leonard, 1997), but exactly how this relationship changes through time is not clear. At
the centennial to millennial timescales, it is hypothesized that glacial erosion is proportional
to ice-cover, so long-term changes in glacier extent should be evident in the clastic sediment
record (Hallet et al., 1996; Leonard, 1997). At the annual to decadal timescales, the relation
o f ice cover and sedimentation has been examined (e.g., Loso et al., 2004). What has yet to
be understood, however, is the degree to which glacier fluctuations control sediment fluxes at
decadal timescales.

1

The proposed research will specifically investigate the relation between changes in
glacier dimensions and sediment yield at the decadal to centennial timescales. Geodetic data
for Peyto Glacier will be compared to annually laminated (varved) lake sediments from a
downstream proglacial lake. The results of the proposed research will aim to clarify the
relation between sediment yield and mountain glaciers. Insight into landscape evolution,
flood control, aquatic ecosystem health, and transport of environmental pollutants all require
insight into the production, distribution, and delivery o f fine sediments in montane
catchments.

1.2

Thesis objectives

The goal of this study is to assess the influence of glacier change on sediment yield within
the Peyto Lake watershed. The objectives o f this study are to:
(1)

use photogrammetry and geographic information systems (GIS) to determine the

dimensional (length, area, and volume) changes o f Peyto glacier from 1917 to 2005;
(2)

measure varved sediments and calculate sediment yield for the Peyto Lake

watershed for the period 1917-2010; and
(3)

assess whether sediment yield is fundamental related to the dimensional changes

o f Peyto Glacier.
I will specifically be able to assess the degree to which long term glacier fluctuations
influence proglacial lake sedimentation. To my knowledge, this will be the one of the few
studies that directly compares a detailed record o f sediment yield to a long record o f glacier
mass balance.

2

1.3

Thesis outline

I wrote and organized this thesis in a traditional format. Following this introductory chapter,
Chapter 2 reviews literature relevant to sediment yield, the relation o f glaciers and sediment
production, Peyto Glacier, and sedimentology within the Peyto Lake watershed. In Chapter
3 , 1 describe the methods used to determine the historical glacier dimensions o f Peyto Glacier
and calculate sediment yield from lake sediment core samples. Chapter 4 summarizes the
dimensional changes of Peyto Glacier, measured sediment yield and how they covary. In
Chapter 5 ,1 discuss the study’s major findings and describe sources of error and uncertainty.
Finally, Chapter 6 summarizes the major findings and provides suggestions for future
research.

3

2.

Literature review
2.1

Sediment yield

O f the sediments entrained in a river system, some o f the sediments are deposited within
sediment stores and the remaining sediments are transported by the river system as dissolved,
suspended, and bed-load fractions. Collectively the suspended and bed-loads fractions are
known as sediment yield which is described by Vanoni (2006) as the total sediment outflow
from a watershed that can be measured at a cross section o f reference over a specified period
o f time (expressed as mass per unit area per unit time).
Sediment yields depend on rates of primary erosion, changes in sediment storage, and
transportation capacity of the medium (e.g., river or glacier) within the watershed. Some
controlling factors on sediment production and sediment storage include local topography,
soil properties, climate, vegetation cover, catchment morphology, drainage network
characteristics and land use among others (Walling, 1994; Hovius, 1998). The relative
importance of each factor varies for each watersheds. For example, Restrepo et al. (2006)
found mean annual runoff explained 51% o f the variance in sediment yield for the
Magdalena watershed, Columbia. Hovius (1998), however, examined 97 watersheds from
around the world and found five environmental variables (specific runoff, drainage area,
maximum height, mean annual temperature, and annual temperature range) accounted for 49
% of the variance in 86 of the watersheds.
Quantifying the changes in sediment storage through time and space are also difficult
to determine. One method uses the sediment delivery ratio (SDR), defined as the proportion
o f sediment exported from a watershed relative to the proportion o f upland erosion (Walling,
1983). The SDR varies between 0 and 1 and is influenced by a range of environmental

4

factors that include: the nature, extent and location o f sediment sources; relief and slope
characteristics; the drainage pattern and channel conditions; vegetation cover; land use; and
soil texture (Walling, 1983). Specific sediment stores in the basin might include the base o f
the slope, in swales, on the flood plain, or within the channel (Walling, 1983). A common
view suggests that specific sediment yield declines as basin area increases (i.e., SDR
decreases) due to increased sediment storage on lower slopes and increased erosion in non­
vegetated areas at higher elevations (Walling, 1983; Ballantyne, 2002). Contrary to this
view, Dedkov and Moszherin (1992) demonstrate that a river system can have an inverse
relationship (i.e., SDR increases) when channel erosion is dominant. For example in British
Columbia, Church and Slaymaker (1989) found that secondary remobilization of Quaternary
sediments increased at all spatial scales up to 3 * 104 km2.
Sediment yield can be determined by a variety o f methods depending on the research
focus, time limits, and study area. Sediment yield is typically derived from measurements of
sediment load within a river; however, this method can be labour intensive, especially for
obtaining a dataset that spans a decade or more. Another method to calculate sediment yield
is to measure the mass o f sediment deposited on the lake bottom for a given period o f time.
Although it is possible to obtain long records of sediment yield through time (Menounos,
2006), the sediment entering and leaving the lake must be accounted for (Foster et al., 1990).
For instance, Owens and Slaymaker (1993) studied four lakes in the southern Coast
Mountains o f British Columbia and found 55 - 99 % of the sediment accumulation over the
last 2350 years originated from four external sources (aquatic production o f (1) organic
matter and (2) biogenic silica; (3) lake bank erosion; and (4) atmospheric dust derived from
outside the catchments) that do not contribute to denudation rates within the basin.

5

Additional variables such as sediment trap efficiency, re-suspension processes, sediment
density changes, and mixing processes o f the lake must also be determined (Foster et al.,
1990). Alternatively, sediment yield can be estimated by generalized equations (Slaymaker,
1977). Such equations tend to simplify the watersheds’ characteristics and do not account for
variability and stochastic processes (e.g., heavy rainfall). Therefore, Slaymaker (1977)
suggested equations that estimate sediment yield should not be used beyond the basin from
which the data were derived. Sediment yield models, however, allow the user to specify
variables that characterize a specific basin through space and time. The benefits o f using
equations and models are that they do not entail data collection in the field (unless for
validation); however, the validity and precision of each model can substantially vary for a
given watershed. The precision and predictive ability of sediment yield models thus depends
on the knowledge of basin’s characteristics and environmental controls.

2.2

Sediment yield and glaciers

Generally when glacierized catchments are compared to alpine catchments free o f
contemporary ice cover, glacierized catchments usually have a higher sediment yield
(Leonard, 1986; Hallet et al., 1996). This view is generally true in glacierized watersheds
because effective erosion rates and the mobilization o f sediments are dominated by glaciers.
Conversely, Hicks et al. (1990) found that precipitation rates influenced sediment yield to a
greater degree than percent glacier cover. The difficulties o f comparing erosion rates for
glacial and non-glacial processes are summarized by Harbor and Warburton (1993); they
highlight the reasons why erosion rates for glacierized and unglacierized basins are nearly
impossible to compare.

6

For glacierized watersheds, sediment yield is commonly related to percent glacier
cover. Globally, the relation between sediment yield and glacier cover has been shown to be
poor (Hallet et al., 1996; Hodder et al., 2007). There are a variety o f factors that contribute
to this scatter including regional climate and environmental differences (e.g., geology, slope,
aspect), as well as, localized changes to sediment storage. Mass wasting (e.g., Johnson and
Power, 1985) and fluvial erosion in the glacier's forefleld (e.g., Orwin and Smart, 2004), for
example, can elevate sediment yields, particularly during heavy rainfall events.
Alternatively, sediment yield may depend on the dynamic state of glaciers (i.e.,
advancing, retreating, and downwasting) rather than percent glacier cover. For the Tyndall
Glacier in southeast Alaska, for example, Koppes and Hallett (2006) found a strong, positive
correlation between glacier retreat rates and glacial sediment yields. The highest rates o f
sedimentation in Hector Lake, Alberta coincided with periods of glacier retreat or rapid
glacial advance rather than during times of maximum ice extent (Leonard, 1997).
Additionally, the thinning o f a glacier could further influence sediment production without
exhibiting significant retreats or advances.
Compounded with glacier dynamics, sediment yield may further be affected by the
rate at which sediment can be produced and evacuated from the sub-glacial environment
(Riihimaki et al., 2005). The rate that sediments can be produced through processes o f
plucking, abrasion, and crushing can be variable through time and space in the sub-glacial
environment. How sediments are evacuated from beneath the glacier is influenced by sub­
glacial water flow (Jansson et al., 2005; Bartholomaus et al., 2008). There is a non-linear
relationship between the rate at which sediments are evacuated and water availability.
Specifically, the increase of water availability to the sub-glacial environment can increase

7

water pressure beneath the glacier which effectively separates the glacier from its bed. Once
this separation has occurred, the flushing o f sediments can commence. However, as the
conduits enlarge, it takes increasingly larger amounts of water to maintain the same water
pressure needed to evacuate sediments. This non-linear relationship is evident at seasonal,
inter-seasonal, and sub-daily timescales (Jansson et al., 2005; Riihimaki et al., 2005).

2.3

History of Peyto Glacier

Detailed evidence of glacier activity is widespread in the Peyto Lake watershed and as a
result, Peyto Glacier is the focus of many research studies (Luckman, 2006). Average
aggregation o f the Peyto Creek delta suggests that Peyto Glacier retreated from Peyto Lake
about 13,000 calendar years before present (cal yr BP; Smith and Jol, 1997). Peyto Glacier
likely continued to retreat until 3000 cal yr BP to an elevation around 2125 m above sea level
(asl). In situ wood fragments found in a paleosol beneath till indicates Peyto Glacier
overrode a forest 3000 - 2800 l4C yr BP (formally known as the Peyto advance). It is
unknown whether the glacier continued to advance until 2500 14C yr BP or whether a second
advance occurred shortly thereafter. Two additional advances occurred at 1550 yr ,4C BP
and 820 14C yr BP, and Peyto Glacier respectively reached positions at least 1 km and 1.4 km
downvalley from the 1990 terminus extent (Luckman et al., 1993). The glacier attained its
Holocene maximum downvalley extent during the latter half o f the Little Ice Age (LIA), or
about 150 - 300 years ago (Luckman, 2006).
Heusser (1956) constrained the timing of the LIA advance to sometime between AD
1711 - 1863 by dating trees growing on trimlines on the lateral moraines on the east side of
the valley. Luckman (1996) investigated the area and confirmed the nineteenth century

8

minimum dates for the main moraine crest, but did not find evidence for an earlier trimline
described by Heusser (1956). Along the lower western lateral moraine, Luckman (1996)
found 11 standing and sheared tree snags along a sharply defined trimline in the forest. His
results indicate that Peyto Glacier achieved its maximum Holocene extent between AD 1837
and 1841. Just beyond the LIA limit, Luckman and Osborn (1979) identified a tephra layer
from the Mazama volcanic eruption suggesting Peyto Glacier has not been more extensive
than its LIA limit since ca. 7700 cal yr. BP.
Research on post LIA activity o f Peyto Glacier used dendrochronology and
photogrammetry to demarcate the position o f the glacier. Luckman (1996) collected an
extensive dendrochronological dataset for the Peyto watershed spanning ca. AD 1700 - 1990.
Watson and Luckman (2004) used dendroclimatology to reconstruct the summer, winter and
net mass balance record for Peyto glacier for the period AD 1673 - 1994. Generally,
reconstructed positive mass balance characterized the period AD 1673 - 1883 (+70 mm water
equivalent (w.e.) yr'1) followed by a period dominated by negative mass balance (-317 mm
w.e. yr'1). Notably, the periods AD 1695 - 1720, 1810 - 1825, and 1845 - 1880 reflect
intervals of pronounced positive mass balance. By using photographs, tree-ring data, and
measurements o f ice wastage, Bunger et al. (1967) mapped glacier retreat positions from AD
1897 to 1965. Wallace (1995) estimated the volumetric change for nearly the same period
(AD 1896 to 1966) as 108.9 x 107 m3. Holdsworth et al. (2006) calculated the loss o f ice
from AD 1966 to 1995 as 139 * 106 m3. Data from by Wallace (1995), Holdsworth et al.
(2006), and additional mass balance values in Demuth and Keller (2006) indicate that Peyto
Glacier has lost roughly 70% of its volume over the period AD 1896 - 1995. In terms of

9

aerial coverage, Peyto Glacier covered an estimated area o f 17.15 km2 in AD 1897,13.35
km2 in AD 1966, and 11.81 km2 in AD 1993 (Wallace, 1995; Demuth and Keller, 2006).

2.4

Sedimentology within the Peyto Lake watershed

Many studies within the watershed have focused on the dynamics o f sediment laden, glacier
discharge flowing into Peyto Lake. Vendl (1978) examined controls o f sedimentation in
Peyto Lake and found large pulses o f sediment travelled along the bottom topography o f the
lake before settling. Approximately 33 times more sediment is deposited in the proximal
basin (161 mg cm'2 d '') than in the distal basin (4.9 mg cm'2 d' *) o f the lake. Smith et al.
(1982) compared the sedimentation regime o f Peyto Lake to three other glacier-fed lakes
within close proximity to each other. O f these four lakes, Peyto Lake receives the highest
and most variable inflowing suspended sediment concentrations (SSC; mean: 720 mg f 1,
range: 138 - 2156 mg I"1). Binda (1984) investigated the suspended sediment and hydro­
chemical dynamics o f discharge from Peyto Glacier by measuring discharge, electrical
conductivity, temperature, and suspended sediment concentrations from Peyto Creek. From
his results, he hypothesized that the re-organization o f sub-glacial channels under changing
hydrostatic pressures and sediment availability was a major factor that influenced the output
o f suspended sediment from Peyto Glacier. The complex sediment dynamics o f Peyto Lake
are further documented by Chikita (1993). He found the sediment laden undercurrent pooled
in the proximal basin and would only ‘spill’ over the central sill into the distal basin when
SSC exceeded 80 mg F1at 6.5 °C. Chikita (1993) believed that most suspended sediment
originates from beneath the glacier because the sediment in and around the Peyto Creek
channel is dominated by gravel and sand.

10

Previous work has also considered sediments contained in Peyto Lake. Smith et al.
(1982) recovered 26 short, sediment cores from the lake. They found the sediment cores
contained horizontal laminations with a wide range o f thicknesses (< 1 mm to > 1 cm). They
thought the sediments were varved, but detailed analyses and independent dating were never
performed.

11

3.

Study area and methods
3.1

Study area

The Peyto Lake watershed (48.9 km2; 51°72' N, 116°52' W) is located 45 km north-northwest
of Lake Louise in Banff National Park, Alberta, Canada (Figure 3.1). Peyto Glacier flows
northeast and is one o f the outlets o f the Wapta Icefield. Peyto Glacier (~ 11.2 km2 in 2005)
and five other small cirque glaciers (1.5 km2) cover approximately 26% (12.7 km2) of the
watershed. Peyto Glacier lies between 2100 - 3200 m above sea level (asl), has a mean
equilibrium line altitude o f 2665 m asl, and a mean accumulation area ratio o f 0.40. Several
moraines adjacent to Peyto Glacier are ice-cored and cover an area of roughly 1.2 km2
(Ommanney, 1972).
Peyto Lake, the larger of two lakes in the watershed, covers an area o f 1.5 km at an
elevation of 1876 m asl and is situated 3 km downvalley from Peyto Glacier. Caldron Lake
(0.4 km2) is situated in a cirque 524 m (2400 m asl) above Peyto Lake and does not receive
glacier meltwater from Peyto Glacier. Peyto Lake consists of two deeper basins connected
by a relatively shallow sill (See Figure 3.5 for Peyto Lake bathymetry). The proximal basin
to Peyto Creek has a maximum depth of 43 m, whereas the deepest point in the distal basin is
49 m (Chikita et al., 1996). Peyto Lake receives an annual inflow o f approximately 45.3 x
106 m3; the lake has an approximate capacity o f 40.3 * 106 m3 which equates to a residence
time of roughly 11 months (Schuster and Young, 2006). Peyto Creek is dominated by glacier
and nival runoff where approximately 82% o f the runoff into Peyto Lake occurs from June to
August. The lake is nearly isothermal, varying from 5 °C to 7.5 °C in the summer, but the
distal basin may develop weak thermal stratification on warmer days. The isothermal lake
conditions, coupled with high SSC o f Peyto Creek cause underflow currents (i.e.,

12

hyperpycnal inflow) to dominate (Smith et al., 1982). Sediment-laden underflows account
for 61% o f the lake’s sedimentation compared to 32% and 7% for delta progradation and
overflows/interflows, respectively (Chikita, 1992).
Steep and mountainous alpine terrain (bedrock, talus, thin soils) represents about 53%
o f the watershed while forests cover about 17 % o f the catchment. Peyto and Cauldron lakes
represent the remaining 4% of the watershed. The underlying geology o f the basin is
primarily Precambrian and Cambrian argillite, limestone, and dolomite (Chikita, 1993).

13

r$unwapta
Mistaya
Lake Louise

\
Hydrometric gauge
VUx station
Water
UA extent
50m contour
Highway 93
Glacier
Ice-cored moraine
Forest

Dragonback
Glacier

0

05

1

i i

1
I

I

2 km
I

I

I

I

Figure 3.1 Peyto Lake watershed with the 2005 glacier extents.

14

3.2

Methods for estimation o f glacier dimensions

Length, area, and volumetric changes to Peyto Glacier were measured and calculated from a
historical map (1917) and 18 sets o f available air photos (1947 - 2005) using standard
photogrammetric methods (McGlone, 2004).

3.2.1
3.2.1.1

Data sources and preparation
Interprovincial Boundary Commission Survey map

Some of the earliest maps in the Rocky Mountains were produced during the Interprovincial
Boundary Commission Survey (IBCS) o f the Alberta-British Columbia border between 1903
and 1924. Surveyors took oblique, teirestrial photographs from mountain ridges and peaks
then applied photo-topographic methods to produce topographic maps. Although the main
objective of the survey was to define the crest of the Rocky Mountains separating the
provinces o f Alberta and British Columbia, the survey concurrently documented early extents
o f many glaciers in the Rocky Mountains (Interprovincial Boundary Commission, 1917;
Wheeler, 1920). A digital copy of IBCS map #17, which contains Peyto Glacier, was
obtained from Library and Archives Canada (LAC). Map #17 was surveyed in 1917 and has
a scale o f 1:62,500 with a 100 ft (30.48 m) contour interval. Thirty-three mountain peaks
and ridges on map #17 were used as ground control points (GCPs) and referenced to
previously orthorectified Landsat Thematic Mapper (TM) imagery, and a shaded relief model
derived from BC Terrain Resource Information Management (TRIM) digital elevation model
(Tennant and Menounos, 2013). The IBCS map was geo-rectified with PCI Geomatica
OrthoEngine v. 10.3 using the polynomial transformation model. The total root mean square
error (RMSE) for the Easting (.x) and Northing (y) coordinates is ± 15.18m. The geo­

15

rectified IBCS map had a significant ridgeline discrepancy that did not align with the actual
terrain resulting in erroneous glacier morphology (Figure 3.2). To fix the terrain
discrepancy, an additional 21 GCPs were collected and used in a thin plate spline (TPS)
transformation model. The TPS transformation model linearly stretched the terrain between
the GCPs yielding no RMSE values.

L?Sa6kc
0‘
% t> 0 -

X

Figure 3.2 (A) Geo-rectified IBCS map #17 using 5th order polynomial transformation model with GCPs (+
symbols). (B) Additional adjustment to map using a thin plate spline (TPS) transformation model with
additional GCPs (* symbols). Maps show the transformation o f the ridgeline discrepancy before the TPS
transformation (red line) and after the TPS transformation (green line).

3.2.1.2

Aerial photographs

For this study there are 18 sets of air photographs available from 1947 to 2005. Sequential
air photographs were acquired I -11 years apart; however, they were also acquired at

16

different times during the melt season which varied from 2 to 61 days apart. Aerial
photographs used in this study were scanned to a digital format at various resolutions from
the Canadian National Air Photo Library (NAPL; 1 2 - 1 4 pm), GeoBC within the British
Columbia government (10 -1 6 pm), and the Air Photo Distribution within the Alberta
government (20 pm; Table 3.1). The scale o f the air photographs ranged from 1:15,000 1:90,000. The 2001 and 2005 air photos exist as aerial triangulation (AT) scans rather than
‘regular’ air photos scans (1947 - 1997). The small area covered by Peyto Glacier enabled
the entire region to be captured in stereo in as few as two photographs for some years, but as
many as 13 for large-scale photographs.
Depending on the year, only a portion o f the accumulation zone was captured in the
photographs, but the terminus was captured for all years (Table 3.1). Poor contrast exists in
many o f the photographs, particularly in the uppermost elevations, which affected my ability
to make surface elevation models. Only the 1952, 1977, and 2005 air photographs had
excellent contrast in the accumulation zone and 100 % photographic coverage o f the glacier.

17

TatteTWVeriayjhotograghs^^
Days before
Acquisition
Year
last day of
date
melt
2005*
3-Aug-05
52
2001* 14-Sep-01
20
1997
5-Aug-97
50
1993
18-Sep-93
6
4-Sep-91
1991
20
1986
15-Aug-86
40
1982
30-Jul-12
56
1977
1-Aug-77
54
1973
24-Jul-73
62
1971
23-Sep-71
1
1967
25-Jul-67
61
1966 20-Aug-66
35
1955 24-Sep-55
0
1952
31-Jul-52
55
1951
6-Sep-51
18
1950
5-Aug-50
50
1949 31-Aug-49
24
1947 22-Sep-47
2

TSL (m asl)

2600
-

2540
-

2625
2490
2475
2500
2450
-

2625
2530
-

2525
2550
2470
2625
2640

Government
Source

Flight line

No. of
photos

Scale

BC
BC
BC
Federal
Federal
BC
BC
Federal
Federal
Federal
Federal
Federal
Federal
Alberta
Federal
Federal
Federal
Federal

15BCC05017
15BCB01035
30BCB97052
A27991
A27790
15BC86085
15BC82021
A24776
A23408
A22443
A20145
A19434
A15085
AS0162
A13321
A12816
A12313
A11120

8
3
13
6
2
2
2
5
7
2
9
6
3
5
5
12
5
2

1 30,000
1 35,000
1 15,000
1 50,000
1 50,000
1 60,000
1 60,000
1 70,000
1 40,000
1 90,000
1 30,000
1 40,000
1 35,000
1 40,000
1 35,000
1 35,000
1 40,000
1 40,000

♦Aerial triangulations scans
¥ The last air photo was acquired September 24th and approximates the last day o f melt.
TSL = Transient Snow Line

Glacier
cover in
photos (%)
100
94
99
100
100
100
87
100
97
100
72
99.5
49
100
100
98
92
56

Contrast in
accumulation
area
excellent
fair
poor
poor
fair
fair - good
fair - good
excellent
poor
good
excellent
fair
poor
excellent
poor
fair
excellent
good

3.2.1.3

Supplemental data (mass balance, discharge, climate)

Glacier mass balance refers to the difference between winter accumulation (snowfall) and
summer ablation (melting) integrated over the glacier area for a balance year, usually
expressed as an average depth in water equivalent (mm w.e.; Cogley et al., 2011). In 1965
the Canadian government selected Peyto Glacier as a representative glacier basin to be
monitored in the Rocky Mountains (Ommanney, 2002). The annual mass balance record for
Peyto Glacier is the longest and most complete in western Canada (1965 - 1990 and 1993 2010; Demuth and Keller, 2006). Unfortunately, both winter and summer mass balances
were only measured for the periods 1966 - 1990 and 1993 - 1995. The standard error for the
mass balance measurements were estimated by Demuth and Keller (2006) as 150 - 200 mm
w.e. In addition to glacier mass balance records, equilibrium line altitude (ELA) and
accumulation area ratio (AAR) measurements were provided. ELA is the elevation at which
snow accumulation equals snow ablation at the end o f summer; it is approximated by the
transient snow line (Cogley et al., 2011). AAR is the ratio o f the area of the accumulation
zone to the area o f the glacier, usually expressed as a percentage (Cogley et al., 2011). Net,
winter, and summer mass balances, ELA, and AAR records were obtained from World
Glacier Monitoring Service ('http://www.wgms.ch/index.htmn. Young (1981), and Demuth et
al. (2009).
Hydrometric records for the Peyto Lake watershed are limited. Environment Canada
had a seasonal hydrometric gauge (Gauge ID: 05DA008) on Peyto Creek from 1967 to 1977,
but these data are neither complete nor long enough to compare to annual sediment yield.
Instead, I used records from nearby hydrometric gauges on the Mistaya River (Gauged ID:
05DA007) and the Sunwapta River (Gauged ID: 07AA007). Peyto Creek is a tributary o f the

19

2

Mistaya River which covers an area o f 248 km . The Mistaya and Sunwapta stations are
approximately 53 km apart and more than 23 km from Peyto Glacier (See Figure 3.1). The
Mistaya gauge has seasonal and continuous hydrometric data for the periods 1950 - 1966 and
1967 - 2013, respectively. I created one contiguous record for the entire period (1950 - 2013)
by only using daily data between May and October. The Sunwapta River is an outlet of the
Columbia Icefields and is in the neighbouring basin north o f Mistaya River watershed. The
Sunwapta gauge has seasonal hydrometric data for 1949 - 1996 and 2005 - 2011. I used to
use the Sunwapta gauge to determine whether changes to the Mistaya River streamflow
occur at regional and/or local scales.
Representative climate records for the Peyto watershed are either temporally limited
or located too far away to be representative o f the watershed’s climate. The short,
unpublished climate record (1967 - 1977) near Peyto Glacier did not have as strong o f a
relation to the mass balance record as the climate recorded at other stations in the Canadian
Rocky Mountains (Letreguilly, 1988). Munro (2006) suggests the climate station was too
close to Peyto Glacier where local cooling from snow and ice reduced the variance in
temperature and relation to mass balance. The Improved Processes and Parameterization for
Prediction in Cold Regions (IP3) research network erected a climate station near Peyto
Glacier from 2002 to 2007, but the climate record is not o f sufficient length for a long-term
comparison; however, I use these data in a temperature index model to estimate ice melt for
years of aerial photography acquired early in the melt season (see section 3.2.3.6). Although
Lake Louise has the nearest climate station to Peyto Glacier, the meteorological tower was
moved numerous times and the data were deemed unreliable (Letreguilly, 1988). Even
though Jasper and Banff have more distant climate stations, their climate records have good

20

correlations with the mass balance records o f Peyto Glacier (e.g., Letreguilly, 1988; Watson
and Luckman, 2004; Munro, 2006). I use the homogenized surface air temperature data
(Vincent et al., 2012) and adjusted precipitation data (Mekis and Vicent, 2011) from Banff
and Jasper climate stations. Selected monthly climate averages from Banff and Jasper,
unlike data from ClimateWNA (Wang et al., 2012), have the strongest relation to the mass
balance record.

3.2.2

Measuring glacier surface and extents

In the next four sections, I describe how I created stereo models from 18 years o f aerial
photographs, digitized glacier extents and surfaces, conducted quality control, and calculated
changes to glacier length, area and volume.

3.2.2.1

Stereo model development

I digitized extents and surfaces of Peyto Glacier using the Vr Mapping Software suite
courtesy of Cardinal Systems, LLC. I specifically used the Vr AirTrig, Vr Two Orientation,
and the Vr Two software for the specific steps mentioned below. A slightly different
approach was undertaken depending whether the digital air photographs were ‘regular’ or
aerial trangulation (AT) scans. AT scans include aero-triangulation data, also known as the
exterior orientation file, which allows the user to create a three-dimensional (3-D) surface.
Regular air photographs, on the other hand, do not have aero-triangulation data associated
with their collection.
I followed a series of steps to create stereo models from the 2005 AT scans. In Vr
AirTrig, the photometry was provided by government issued camera calibration reports, then

21

the photograph fiducial marks were identified on the AT scans. Neighbouring photographs
were linked to each other with six common ‘tie points’ and adjacent flight lines were linked
with three common ‘pass points’. The aforementioned process is referred to as Inner
Orientation. In Vr Two Orientation, the exterior orientation file (i.e., aero-triangulation data)
was used to create 3-D stereo models o f the 2005 AT scans. Once the quality o f the 2005
stereo models was assured, 31 GCPs were measured on sparsely vegetated, stable surfaces
surrounding Peyto Glacier. GCPs were on stable bedrock features, free of snow/ice,
vegetation and loose debris, that are not expected to change from year to year (Kaab and
Vollmer, 2000; Schiefer and Gilbert, 2007). Locations o f GCPs were evenly distributed, in
terms of elevation and plane, on the terrain surrounding Peyto Glacier. The known 3-D
coordinates (i.e., Easting, Northing, and elevation) o f the GCPs were then used as reference
points in the creation of stereo models for regular air photos.
Like the 2005 AT scans, each year o f regular aerial photographs were imported into
Vr AirTrig for Interior Orientation. The same GCPs collected from the 2005 AT scans were
identified in the regular air photos and given the same coordinate (i.e., Easting, Northing, and
elevation). The collected GCPs were used to create an exterior orientation file and determine
the root means square error (RMSE) in the tie points, pass points, and GCPs. The lowest
RMSE value was always desired, but a RMSE < 1 m was considered suitable for this study.
With the newly created exterior orientation files, stereo models were built in Vr Two
Orientation. Once stereo models were created from sets o f AT scans and regular air
photographs, I used Vr Two to digitize and measure the extent and surface o f Peyto Glacier
for the 18 sets o f aerial photographs.

22

3.2.2.2

Data collection from stereo models

From the developed stereo models, 1 digitized and measured 3-D coordinates o f the glacier’s
surfaces and extents using Vr Two software. Unfortunately, measurements were hindered by
environmental variables. Late-lying snow in the accumulation zone, for example, obscured
the identification o f bare earth or ice. Visibility o f the glacier, along the eastern edge o f
Peyto’s terminus, was frequently obscured by shadows and steep terrain. The western region
in the accumulation zone and a medial moraine were also frequently obscured by debris
cover. These problematic features were compared in subsequent years to ensure continuity
and consistency in measurements. The extents of the ice-cored moraines were not easily
identified, so I delineated their extents by the observed deflation through subsequent years.
Glacier and moraine surfaces were digitized by measuring elevation points (mass
points) on a 100 x 100 m grid yielding roughly 1,400 possible points for a given year of
photography. I measured mass points within the 1917 IBCS glacier boundary for each year.
For all years, mass points were collected on the terminus; however, fresh snow, poor
photographic contrast, or missing glacier coverage in the air photos prevented mass points to
be collected in the accumulation zone (Figure 3.3; VanLooy and Forster, 2011). If a mass
point could not be digitized and measured with confidence, its elevation was estimated by
two different methods. First, if a missing mass point for a given year occurred between two
adjacent years with measured mass point data, the missing elevation values would be
interpolated (based on time) by the difference between the adjacent years. By using this
approach, I was able to constrain a likely elevation from known data which is not possible
with the second method. Many points were infilled by this sandwiching method, but if they

23

were not, the second method was employed. For this second method, instead o f estimating
the elevation o f an unknown mass point, the average elevation change between two
sequential years would be estimated from surrounding known mass points for a designated
elevation band. Average elevation changes of mass points on Peyto Glacier were calculated
by 100 m elevation bands (e.g., 3100 - 3200 m asl).
100 -r

c

a>

90 --

^

80

• s | 70
3> m
S? 8

60 - -

f §
8J

50

o
1 3 40
30
?'9
o o> 20

Q_

10

0

i

II ■ ■ If m
■ ■ ■ ■I
ii ■ II i

ii
I■ ■! ■
■ I■- i i if
■
■■ I I ■ I ■I ■! ii ii ■I n
I I ■ II ■ ■! i ii ■lip
Mill ■ii
i i i l l ■Ii
ii Ii ii IIII ii i
■ill ■iii iii iiif ■!
I
I ■ii
J
1

ii

1

1

1917 1947 1949 1950 1951 1952 1955 1966 1967 1971 1973 1977 1982 1986 1991 1993 1997 2001 2005
Y e ar (AD)

■ Photographic coverage of glacier

a G lacier surface m easured

Figure 3.3 The amount o f glacier surface that could be measured each year based on photographic coverage o f
the glacier (solid black bars) and the amount o f glacier surface that was measured (stripped bars).

3.2.2.3

Stereo model quality control

I assessed the precision o f stereo models by comparing stable surfaces o f a given year to the
2005 stereo models (i.e., reference year). I measured a series of points (checkpoints) for a
year, then I compared these to the same checkpoints measured from the 2005 photographs.
The elevation difference between checkpoints indicates a topographic bias between the two
stereo models. Checkpoints were measured on a 5 x 5 grid and spaced 10 m apart which I
collectively refer to as a ‘checkpatch’. Similar to GCPs, checkpatches were located on stable
bedrock features free of snow/ice, vegetation and loose debris. Ten checkpatches were
measured at a range o f elevations distributed around Peyto Glacier. The Wapta Icefield and

24

steep terrain surrounding Peyto Glacier limited the location and number o f checkpatches
measured. Depending on photography coverage for that year, 5 - 1 0 checkpatches were used.
I assessed spatial bias in the stereo models by plotting checkpoint residuals against
their respective Easting, Northing, and elevation coordinates (Figure 3.4). The linear model
with the highest, significant correlation indicated the strongest bias present in the stereo
model (Table 3.2). I used the linear model coefficient to remove the bias (Figure 3.4) from
the stereo models o f each year (Tennant and Menounos, 2013). Only one bias was removed
from the models as a method to prevent over-manipulation o f the measured data.

A

u

B

o

1
1m

o

2000

2200

2400

Etovatton (m)

2600

2800

3000

Elevation (m)

Figure 3.4 Example o f checkpoints measured in the 1991 air photos (A) before bias removal and (B) after bias
removal.

Table 3.2 Pearson correlation coefficient (r) and significance (p) for checkpoint residuals versus topographic
predictors (Elevation, Easting, and Northing). The largest correlations are bolded and the respective
topographic biases were removed from the models.

25

Year

Elevation
(r)

2001
1997
1993
1991
1986
1982
1977
1973
1971
1967
1966
1955
1952
1951
1950
1949
1947
1917
3.2.2.4

<P)
2.20E16
1.53E'15
9.99E'10
2.20E'18
2.82E12
9.13E'19
7.83E'01
6.17E-01
3.69E'07
6.22E13
2.20E"16
3.81 E-°s
2.20E16
7.54E-08
2.20E*16
2.20E16
4.64E15

-0.644
-0.555
-0.374
-0.904
0.496
-0.502
0.018
-0.034
0.331
0.569
-0.749
-0.330
0.662
0.349
0.689
-0.542
-0.628
-0.216 2.12 E 03

Easting
(r)

(P)

0.709
0.143
-0.229
0.248
-0.610
-0.075
-0.173
0.454
-0.097
0.302
0.390
0.173
0.217
0.483
-0.050
0.611
-0.494

2.20E18
5.99E02
2.65 E04
7.47E05
2.20E'18
2.62E01
9.46E*3
7.27E'13
1.44 E01
4.99E'05
7.72 E07
3.42E-02
1.04E-03
1.40E14
4.58E01
2.20E18
4.79E09

-0.191

6.67 E03

Northing

M
0.313
0.498
0.255
0.851
-0.447
0.445
-0.017
-0.169
-0.362
-0.717
0.611
0.072
-0.580
-0.400
-0.578
0.238
0.534
0.347

(P)

9.97 E05
2.45E12
4.64 E-05
2.20E16
5.37E10
2.54E12
8.05E01
1.09E02
2.27E"08
2.20E'18
2.20E"16
3.84E-01
2.20E'16
4.55E10
2.20E'16
3.09E-04
1.37E'10
4.82E-07

Glacier change analysis

I imported glacier extents and mass points datasets into ESRI ArcMap 10.1 where changes in
glacier length, area, and volume were calculated, and their respective errors assessed. Prior
to calculations, the identified model bias for each year was removed from the measured
glacier extents and mass point datasets. The entire extent o f Peyto Glacier was visible in the
aerial photographs for nine years, so the remaining years with incomplete glacier extents had
the missing boundaries replaced from the nearest year. Missing boundaries were always in
the accumulation zone where areal changes were negligible. The hydrological divide
between Peyto Glacier and the rest of the Wapta Icefield was demarcated with the hydrologic
analysis tool available in ArcMap.

26

Changes in glacier length, and area were calculated by differencing the sequential
year of data (e.g., 1997 minus 2001) where a receding glacier is denoted as negative and an
advancing glacier is denoted as positive. The lengths of Peyto Glacier were measured
perpendicular to the glacier terminus and parallel to the glacier’s flow line (Andreassen et al.,
2002). The length error term is the sum of the mean horizontal RMSExy (Easting (x) and
Northing (y) positional errors calculated in Vr Two for stereo model batch adjustments) and
one-half of the digital pixel resolution of each year o f data (Table 3.3). From 1991 to 1997
there was a large, thin region o f ‘dead ice’ that detached from Peyto Glacier, so the glacier
polygons were summed together accordingly. The area error term is equal to a buffer
surrounding the inside and outside of the glacier extent (Granshaw and Fountain, 2006;
Bolch et al., 2010). The buffer width is equal to the length error term.

Error change (E a)

for length or area between two consecutive years is thus:
(3.1)
where £, and E2denote the error terms for length or area for two consecutive years.
Changes between glacier volumes were calculated by differencing the sequential year
of mass point data and multiplying the difference o f each mass point by its representative
area (10000 m2; VanLooy and Forster, 2011). Elevation changes for points greater than three
standard deviations (± 3 a) of the mean were considered outliers and removed from the
analysis. The volume change error ( VEA) for a period is thus a function o f elevation and area
change:
(3.2)
where Daz is the standard deviation (± 1 o) o f the elevation change for a period calculated by
the checkpoints after bias removal; ZA is the mean elevation change for a period calculated by
27

the checkpoints after bias removal; A is the area of the largest extent used to calculated
volume change; and EA is the area error term of the largest extent.
Table 3.3 Information used to create stereo models and assess their accuracy and error._______________

Year
2005
2001
1997
1993
1991
1986
1982
1977
1973
1971
1967
1966
1955
1952
1951
1950
1949
1947

Pixel size
<m)
0.371
0.440
0.152
0.637
0.124
0.757
0.741
0.640
0.156
0.611
0.088
0.159
0.662
0.401
0.675
0.108
0.271
0.399

1917

-

No. of
GCPs

No. of
checkpatches

RMSE
x(m)

RMSE
y(m)

-

10
6
10
10
10
7
7
10
7
7
7
7
6
9
9
9
9
5
Mean RMSE
8

-

-

0.207
0.294
0.193
0.264
0.197
0.244
0.255
0.221
0.294
0.525
0.284
0.231
0.577
0.409
0.611
0.581
0.246
0.330
9.62

0.214
0.339
0.169
0.225
0.276
0.253
0.250
0.310
0.302
0.357
0.237
0.256
0.635
0.448
0.999
0.450
0.246
0.350
11.74

18
23
22
28
10
24
27
18
26
20
19
20
25
24
18
23
17
33*

RMSE
xy(m)

RMSE
z(m)
.

0.298
0.449
0.257
0.347
0.339
0.351
0.357
0.381
0.421
0.635
0.370
0.345
0.858
0.607
1.171
0.735
0.348
0.490
15.18

0.041
0.130
0.045
0.022
0.028
0.037
0.045
0.155
0.033
0.304
0.152
0.044
0.404
0.266
0.903
0.414
0.059
0.180

GCPs = Ground Control Points, RMSExyz = Root Mean Square Error on the Easting (jc), Northing (y) and
elevation (z) axes. * Mountain peaks were used as GCPs in the 1917 map, but not used in aerial photographs.

3.2.2.5

Glacier volume change conversion

I converted the measured changes of glacier volume to snow water equivalent (w.e.), so that I
could compensate for seasonal melt differences that resulted from the date aerial photographs
were acquired. The measured changes to glacier volume were converted to snow water
equivalent using an average density o f 760 kg m"3. I calculated average density using the
mean accumulation area ratio for Peyto glacier o f 0.40 where ice density in the ablation zone
was assumed 900 kg m'3 and ice density in the accumulation zone was assumed 550 kg m‘3
(Schiefer et al., 2007).
28

To adjust volumetric changes o f Peyto Glacier attributed to when aerial photographs
where acquired, I used a basic temperature index model given by Hock (2003):

/=i

= D D F ^ T +A t,
/=i

(3.3)

where M is the total amount o f ice or snow melt, T+ is the positive air temperatures, At is the
day for the period (n), and DDF is the degree-day factors (expressed as mm °C '' d '1). For
melt factors (or DDF), I used the values 2.32 mm °C '1 d '1 for snow and 5.57 mm °C '1 d '1 for
ice that were derived from the glacier mass balance records of Peyto Glacier (Shea et al.
2009). I distinguished different DDF surfaces (snow and ice) by the transient snow line
apparent in the aerial photographs. For the unknown mean daily temperature values, I
compared mean daily temperatures for August and September from Banff (1396 m asl) to
values recorded at the IP3 station (2240 m asl) near Peyto Glacier from 2002 to 2007 and
calculated a linear lapse rate o f 5.51°C km '1. From the Banff climate data, mean daily
temperatures were adjusted on the mass point grid (100 m * 100 m) across the surface o f
Peyto Glacier. For the study period, aerial photographs were acquired from as early as July
24th to as late as September 24th. Therefore, daily differential adjustments to glacier volume
were calculated and summed from the date aerial photographs were acquired until the enddate of the melt season approximated as September 24th.

29

3.3

Methods to estimate sediment yield
3.3.1

Field sampling

I retrieved core samples o f lake sediments taken from the bottom of Peyto Lake throughout
March 2010, February 2011, and April 2011. A 250 m sampling grid was selected to capture
the diversity of the lake's topography (i.e., basins and sills), and provide a large enough
sample size to conduct spatial analysis and statistics (Schiefer, 2004). In total, 20 percussion,
two Ekman box, and five gravity cores were collected. With coring instruments I collected
0.2 - 3.0 m long sediment cores along specified grid locations on the frozen lake surface
(Reasoner, 1993; Glew et al., 2001). Generally, the longer cores were collected in the deeper
basins or in close proximity to the Peyto Creek delta. I used Ekman box and gravity corers to
collect undisturbed sediments at the water-sediment interface (i.e. the upper 10 -1 5 cm of the
sediment surface; Hodder et al. 2006), which can be disturbed by the use o f a percussion
corer. The recovered sediment cores were transported to UNBC and stored in a refrigerator
at 4 °C until further laboratory analyses were conducted (Appendix A).

[I Cores used in analyses
o Cores collected - not used in analyses
Figure 3.5 Peyto Lake bathymetry (adapted from Chikita et al., 1996). The locations o f the retrieved sediment
core samples are shown. Note: core 10-Peyto(01) was taken from the same location as core 10-Peyto(02). The
small arrows indicate the primary inflow and outflow o f Peyto Creek.

30

3.3.2
3.3.2.1

Laboratory analysis
Photography

I split all cores lengthwise where the working halves were sub-sampled for bulk-physical
properties (see section 3.3.2.2) and the archive halves were photographed. I used a Nikon
D90 SLR camera with an 18 - 105 mm zoom lens to capture digital images that contained a
resolution of 12 megapixels. To capture maximum depth o f field and reduce radial lens
distortion, images were taken at the maximum aperture o f F-32 with a focal length of 50 mm.
Each image referenced core direction, sediment depth, date, times photographed, and a scale
bar. From the clearest images, I measured the length o f cores and silt-clay couplets with a
precision of 0.2 mm using a script program written to operate with UTHSCSA Image Tool
software V. 2.01 (Wilcox et al., 2002). The visible detail, color, and texture o f the core
samples changed considerably as they dried at room temperature, so cores were
photographed multiple times as they dried. Often the laminated sediments were not clearly
discemable until the sediments were allowed to dry out completely; unfortunately the drying
processes also caused cracks throughout the sediments (Figure 3.6). Both wet and dry cores
were measured as a method to quantify the size of cracks and sediment shrinkage when the
photographs were taken.

31

Figure 3.6 Example o f wet (top) and partially dry (bottom) laminated sediment. The red and blue lines on the
dry sediments are measurements from the UTHSCSA Image Tool software. The water-sediment interface is
towards the right.

3.3.2.2

Bulk-physical properties

I sub-sampled bulk-physical properties by standard methods outlined by Hakanson and
Jansson (2002). The sub-sampling interval varied by non-master and master core types.
Non-master cores were sub-sampled at 10 cm intervals for the primary purpose of
interpolating spatial variation (areal and depth) of sediment distribution within Peyto Lake.
The master core sites align with the longitudinal axis of the lake and were sub-sampled at 2
cm intervals as a way to determine the variation o f bulk physical properties at a higher

32

resolution. Sub-samples (2.0 ml) were taken with a 14 mm syringe along the longitudinal
axis of the core where minimal disturbance occurred. Sub-samples were weighed, ovendried at 105 °C for 12 to 24 hours, and then reweighed again at room temperature (Heiri et
al., 2001). Water content and dry bulk density were calculated as follows:
Percent water content =

Wet mass (g ) - Dry mass,n, (g)
Wet mass (g)

__
x100

~
,
Dry mass.n, (g )
Dry bulk density = --------------------- —,
Wet Volume (cm*)

_
(3.4)

(3.5)

where the subscription denotes the oven treatment temperature. Variations in water content
variation reflect the quality and character o f the deposits, sedimentation rates, degree of
compaction, organic content, and degree o f bioturbation (Menounos, 1997; Hakanson and
Jansson, 2002). The following loss-on-ignition (LOI) technique is a commonly used method
to estimate organic content of sediments (Heiri, 2001; Hakanson and Jansson, 2002). To
determine organic matter (OM) content, samples were heated in an oven to 550 °C for four
hours where the organic matter oxidized to carbon dioxide and ash. Once samples cooled to
room temperature, they were reweighed, so that OM content could be calculated:
OM LOI550 =

^D ry massm (g ) - Dry mass550 (g)^
Dry massl05 (g)

x 100

(3.6)

High-levels o f OM content may reflect periods of flooding, mass movement events, or
changes in primary productivity levels within the lake (Hakanson, 1995; Lacourse and
Gajewski, 2000; Heiri et al., 2001).

33

3.3.3
3.3.3.1

Core chronology
Cesium-13 7 analysis

An annual chronology o f sediment yield requires independent confirmation that silt-clay
couplets in the cores are varves. The isotope Cesium-137 ( I37Cs) is an artificial fallout
radionuclide that has two primary sources: (1) atmospheric nuclear weapons tests for the
period 1953 - 1963, and (2) the Chernobyl reactor fire of 1986 (Appleby, 2001). The
presence of 137Cs sediments allows researchers to verify the annual nature o f sediment
records (Lamoureux, 2001). From the North America 137Cs fallout record, peak nuclide
activity corresponds to 1963. 137Cs activity associated with the peak levels o f the 1986
Chernobyl reactor fire has not been detected in western Canada. Sub-samples were taken at
10 cm intervals down the entire length of core 10-Peyto(02). To constrain the dating of
1963, additional sub-samples were taken at assumed varve locations on both sides o f where I
counted the 1963 varve should be located (a total o f 21 sub-samples). Sub-samples yielding
>5 grams dry weight were taken with a 14 mm diameter syringe and freeze-dried for 24
hours before being sent to Flett Research Ltd., Winnipeg, Canada, for 137Cs activity
measurements by gamma spectrometry using 19% and 25% efficient HPGe detectors.

3.3.3.2

Master chronology

I produced a master varve chronology by merging individual varve records obtained from the
sediment cores. A varve chronology enables incorrectly identified varves and missing
sediment layers in cores to be identified (Lamoureux, 2001). A complete chronology is
needed for all cores for an accurate sediment yield to be calculated. From the 137Cs
radionuclide dating results described in section 4.2.3,1 was able to create a chronology for

34

core 10-Peyto(02) and convert it to calendar years (AD). I produced a varve chronology for
the remaining cores by matching unique sediment layers (marker beds) among cores.
Although the high sedimentation rate o f Peyto Lake resulted in large varves that could easily
be linked among multiple cores, there were only three key marker beds identified among all
cores: two light coloured beds deposited in 1834 and 1846; and an unusually thick layer
deposited in 1983. In July o f 1983, a rainfall-triggered flood destroyed the hydrometric
monitoring station on Peyto Creek (Schuster and Young, 2006). Although, small flood and
mudflows had previously been observed, nothing approached the magnitude o f the 1983
event described by Johnson and Power (1985). According to their study, large amounts of
precipitation eroded an ice-cored moraine and the subsequent gravel (6000 m3) was deposited
down valley.
Once varve chronologies were created for all cores, I created a master chronology
following the methods in Desloges (1994) which standardizes varve thickness:
V -V
VSI = -J
,
0"v

(3.7)

where VS, is the standardized thickness for year i, V is the measured varve thickness, V is
the mean varve thickness for the core, and a v is the standard deviation. The standardized
values for each core were averaged for that year which resulted in a master chronology of
varve thickness variation. This chronology, in particular, is important for highlighting a
common signal among cores and identifying rates when varve thicknesses were above or
below average.

35

3.3.4

Sediment yield calculations

3.3.4.1

Spatial sediment interpolations

In most cores, the sediment laminations near the water-sediment interface were disturbed,
unidentifiable, or missing. The unknown sediment thickness for a single core was estimated
by a ratio determined by a neighbouring core with the highest correlation o f varve thickness
variation. For instance, if core X has 75 % o f the sedimentation as core Z and they have a
significant correlation with each other, the missing sediment thickness for core X is equal to
75% of core Z for the same region.
For Peyto Lake, annual variation in sediment thickness could not be explained by
environmental variables (e.g., distance to the delta and lake depth) like that found by Schiefer
(2004) for Green Lake, BC. Instead o f using a multiple linear regression to interpolate
sediment thickness, I used two methods o f interpolation: Thiessen polygons and regularized
spline. These two methods o f interpolation provide a range of possible sediment yield values
that cannot be independently confirmed by a simple technique. The first method uses
Thiessen polygons as representative areas around each core sample (e.g., Figure 4.16; Evans
1997). This method assumes that sedimentation occurs at all lake depths equally within each
polygon. In reality, however, a lake environment dominated by underflow and influenced by
environmental variables, such as lake depth and wind, limits sedimentation above a certain
depth (Smith and Ashley, 1985). Therefore, I also conducted a first-order sensitivity analysis
to estimate the limits o f where sedimentation could have occurred in Peyto Lake when lake
depths were deeper than 0 m, 10 m, and 20 m (covering 100 %, 86 %, and 73 % o f the lake
area, respectively). The second method o f interpolation uses a regularized spline which
estimates values using a mathematical function to minimize overall surface curvature (e.g.,

36

Figure 4.16). Even though the regularized spline method does not encompass bathymetry, it
produces a smoothed surface that is similar to the expected sedimentation pattern. For both
methods, the interpolated surface passes exactly through the sample points.

3.3.4.2

Lake trap efficiency

Verstraeten and Poesen (2000) reviewed methodologies to estimate the percent o f sediment
remaining in a lake after flowing into it, also known as the lake’s trap efficiency (TE). They
provide a modification of the Brune (1953) TE curve that has a focus on lakes dominated by
different sediment textures. Specific to fine-grained sediments (silt and clay), TE is defined
as:
(3.8)
(formula originally cited in Harbor et al., 1997)
where C is volumetric capacity of the lake and I is the annual inflow to the lake.

3.3.4.3

Sediment yield

Calculation of sediment yield (SY) from multiple cores follows a similar equation to that
used by Evans (1997):

(3.9)

where Zx is sediment thickness (measured or interpolated) o f a unit in core /; A x is the
representative area (Thiessen polygon or raster pixel); Dx is the mean dry weight/wet
volume; L\ is the mean percentage LOI value; and TE is the estimated trap efficiency of the
lake (see section 3.3.4.2.).
37

Mean diatom and authigenie carbonate concentrations are two additional variables
that can affect SY values, but they are not easily measured. The diatom concentration in
Peyto Lake was not measured; however, the concentration is expected to be negligible as
shown in nearby lakes (Hickman and Reasoner, 1998; Hobbs et al., 2011). For watersheds
dominated by carbonate bedrock, like Peyto basin, authigenie carbonate content determined
by LOI methods is not possible (Evans, 1997).
To compare sediment yield among different sized watersheds, sediment yield is
divided by watershed area which is then defined as specific sediment yield (SSY):
SSY = ------)
Watershed area (km )

(3.11

38

4.

Results

The results are categorized into three sections regarding glacier dimensions, sediment
analyses, and environmental controls on sediment yield.

4.1

Glacier dimensions

In the following three sections, I assessed the quality of the stereo models and determined
environmental biases (i.e., Easting, Northing, or elevation); I quantified data availability by
photographic coverage and measurable glacier surface; and I determined changes to glacier
length, area and volume.

4.1.1

Stereo model verification and bias removal

Identified biases from the photographic based stereo models and 1917 IBCS map were
quantified and removed. The mean horizontal (xy) and vertical (z) RMSE calculated for the
18 years of aerial photography were respectively 0.49 m and 0.18 m (Table 3.3). Mean
checkpatch elevation difference before bias removal was 2.67 ± 2.95 m. There were
significant linear biases in all stereo models. Elevation biases were present in majority o f
models, but Easting biases were present in 1949, 1951,1973,1977, 1986, and 2001;
Northing biases were present in 1967 and 1971 (Table 3.3). The linear model coefficients
(Table 3.2) were used to adjust the stereo models, and following adjustment the mean
checkpatch elevation difference became -0.00 ± 2.28 m. The newer aerial photographs (1966
- 2001) were the most accurate as the elevation differences for the checkpatches were
roughly ± 5 m. In comparison, older aerial photographs (i.e., 1947 - 1955) produced less

39

accurate corrected models where the elevation difference for the checkpatches often
exceeded ± 5 m (Figure 4.1).
The mean horizontal RMSEjty for the 1917 IBCS map was 15.8 m. There were no
vertical adjustments to the planar map. The mean checkpatch elevation difference before
bias removal was 29.01 ± 37.67 m. The Northing coordinate was the only significant bias
that was present in the IBCS map. After bias removal, the mean checkpatch elevation
difference was 0 .0 1 ± 35.03 m (Q25 = -29.81 m, Q50 = -7.44 m, Q75 = 12.25 m).

t -------- 1------------1----------1---------- 1---------- 1---------- 1---------- 1--------- 1-----------1---------- 1---------- 1---------- 1---------- 1---------- 1---------- 1---------- r

1947

1949

1950 1951

1952

1955

1966

1967 1971

1973

1977

1982

1986

1991

1993

1997 2001

C alendar Year

Figure 4.1 Box plot showing the absolute elevation difference o f check points after bias removal. Differences
are between individual years and the 2005 reference data.

4.1.2

D ata completeness

The amount of glacier surface from which elevation could be extracted varied due to
photographic coverage, poor contrast in the accumulation zone, cloud cover or a combination
o f these factors (Figure 3.3). Unknown surface elevations were estimated; however, the
accuracy of infilled values was not independently confirmed. The quality o f the stereo
40

models was independently assessed by checkpatches (see section 4.1.1.), but the assessment
did not consider the percent o f the glacier’s surface that was available to be measured. Here,
the calculated changes to glacier volume are categorized by ‘completeness’ that reflect the
amount of glacier surface measured and my confidence in the quality of the data. Defined
periods were specified by years containing: > 99 % o f the surface measured; > 70 % o f the
surface measured; > 50 % of the surface measured; and > 20 % of the surface measured
(Table 4.3). The measured glacier surface from the 1947 stereo models resulted in erroneous
results, so they are omitted from the remaining results.

>99%
1977-2005
1952-1977
1917-1952
-

> 70 %
2001-2005
1993-2001
1991-1993
1986-1991
1977-1986
1967-1977
1966-1967
1952-1966
1950-1952
1949-1950
1917-1949

-

-

-

-

-

-

>50%
2001-2005
1997-2001
1993-1997
1991-1993
1986-1991
1982-1986
1977-1982
1971-1977
1967-1971
1966-1967
1952-1966
1950-1952
1949-1950
1917-1949

-

-

-

-

-

-

-

-

-

>20%
2001-2005
1997-2001
1993-1997
1991-1993
1986-1991
1982-1986
1977-1982
1973-1977
1971-1973
1971-1967
1967-1966
1955-1966
1955-1952
1952-1951
1950-1951
1949-1950
1917-1949

Categorizing the data by ‘completeness’ does two things simultaneously: it breaks the
data into smaller periods and smaller volume changes. For instance, the average periods are
29.3, 8.0, 6.3, and 5.2 years for each category (> 99, > 70, > 50, and > 20 %, respectively).
As the completeness o f the data decreases (i.e., > 99 to >20 %), trends in the data emerge
(Figure 4.2). That is, as the completeness o f the data decreases, the total volume change o f

41

glacial ice and total ice (glacier and ice-cored moraines) increases respectively by 9 % and 5
%. The volume change o f the lateral ice-cored moraines, in contrast, decreases by 22 % as
the completeness decreases. The most complete data have fewer values (n=3) available to
compare to sediment yield which is not ideal for finding a correlation. The least complete
data have more values (n=l 7) to compare to sediment yield, but limits the utility of these data
to compare to the sediment yield record.

> 99 %
_

Completeness
> 70 %
> 50 %

> 20 %

0
■ Dirty Ice
a Clean be

5 -100
”e
x -200

‘t o

^

-300

a>

j=
(0 -400
<■> -500
a>

i

o

-600

> -700
-800
Figure 4.2 Changes in ice volume for Peyto Glacier categorized by the completeness o f surface points
measured.

4.1.3

Changes to glacier length and area

For the period 1917 - 2005, Peyto Glacier experienced a net retreat (Figure 4.3). During this
period, Peyto Glacier retreated a total distance o f -2198 ± 18 m along its flow line which
yields an average recession rate of -25 ± 0 m yr'1 (Table 4.1). Periods with an above average
retreat rate include: 1947 - 1949,1950 - 1951, 1955 - 1966, 1967- 1971, and 1973 - 1982.
In 1917, the area covered by Peyto Glacier was 15.2 ± 0.5 km2. By 2005, the glacier
shrank to -11.2 ± 0.02 km2 which corresponds to a retreat rate o f -0.046 ± 0.005 km2 yr'1.
Peyto Glacier grew during the periods: 1949 - 1950, 1951 - 1952, and 1971 - 1973 (Table
42

4.1). These periods of positive growth are likely spurious and attributed to: (1) patches of
snow along the perimeter of the glacier that were measured in the more recent set of aerial
photographs, but not in the older set; and (2) differences in the time o f year when the aerial
photographs were acquired (Table 3.1). Generally, the majority o f glacier ice melted in
regions below the ELA; however, glacier ice was also observed to have melted in peripheral
areas of the accumulation zone and in a central location where a rock island began to emerge
(ca. 1966). Periods with the largest change in glacier coverage include: 1947 - 1949,1950 1951, 1966 - 1971, 1973 - 1977, and 1997 - 2001.

43

Ice-cored moraine
50rn contour \
LIAextent

Figure 4.3 Extents o f Peyto Glacier digitized from aerial photographs and the IBCS map throughout the period
1917-2005.

44

2001 -2005
1997-2001
1993- 1997
1991 - 1993
1986- 1991
1982- 1986
1977- 1982
1973 - 1977
1971 - 1973
1967- 1971
1966- 1967
1955 - 1966
1952- 1955
1951 - 1952
1950- 1951
1949- 1950
1947- 1949
1917-1947

-21.5
-18.9
-15.2
-21.4
-14.3
-18.5
-31.8
-28.5
-18.1
-41.7
-16.5
-42.3
-7.5
-10.3
-119.5
-10.0
-26.0
-20.5

1917-2005

4.1.4
4.1.4.1

±
±
±
±
±
±

±

0.2
0.2
0.2
0.4
0.2
0.3
0.2
0.2
0.4
0.2
0.8
0.1
0.4
1.4
1.5
1.5
0.5
0.6

-3.83
-12.50
-2.15
-6.52
-6.12
-4.19
-4.80
-7.84
10.65
-7.05
-12.85
-4.33
-3.18
8.66
-11.55
15.71
-13.71
-3.98

-25.0 ±

0.2

-4.55

±

±
±
±
±
±
±
±
±

±
±

X

A Length (m yr'1)

S»

Period

>

Table 4.2 Mean rate^oftengthjm djire^

10
±
±
±
±
±

±

yr’1)
0.72
0.64
0.70
1.23
0.58
0.88
0.68
0.71
1.49
0.89
3.07
0.26
1.41
4.76
5.52
5.36
1.66
1.60

±

0.55

±
db
±
±

±

±
±
±
±
±
±

Observed volume change
Peyto Glacier

Changes to glacier volume were measured and converted to water equivalent (w.e.). I then
applied the temperature index model to adjust the volumetric changes of Peyto Glacier
attributed to when aerial photographs were acquired (Figure 4.4). The importance of
applying a temperature index model to volume change was particularly important for the
periods 1949 - 1950, 1966 - 1967, and 1971 - 1973. For these periods, the volume changes
for Peyto Glacier were positive before the temperature index model was applied, but became
negative after applying the temperature index model.

45

120
_

110

■ M easured Volume C hange

100

B M easured Volume C hange with seaso n al adjustm ents

1

70

&

60

; so
|> 40

I

30

-30

I

-40
-50

Period (year AD)

Figure 4.4 Comparison o f measured volume changes o f w.e. (solid black bars) and measured volume changes o f
w.e. with seasonal melt adjustments (stripped bars).

Changes to glacier volume were categorized by data completeness ranging from most
complete to least complete: > 99 %, > 70 %, > 50 %, and > 20 %. The error terms for the
least complete data (> 20 % and > 50%) generally made interpretation inconclusive
(Appendix C), so the remaining results focus on the most complete data sets (>70% and
>99%).
The most complete data show that Peyto Glacier and the ice-cored moraines lost -666
± 406 x 106 m3 w.e. over the past 88 years (Figure 4.5). Eighty seven percent o f this volume
change (-581 ± 404 x 106 m3 w.e.) originated from the clean ice o f Peyto Glacier. If I divide
the latter volume by the average area o f the glacier (13.2 km2), the glacier thinned a total of
-44.0 ± 30.6 m w.e. which corresponds to a rate o f -0.50 ± 0.35 m w.e. yr'1. The large error
term arises from the low accuracy o f the 1917 IBCS map. For the period 1917 - 1952, errors
in the 1917 IBCS map is evident when compared to the periods 1952 - 1977 and 1977 - 2005
(Figure 4.6). The eastern portion of the ablation zone, for example, gained volume and

46

regions in the accumulation zone lost unlikely amounts o f volume. The highest rate of
volume loss for the glacier (-10.9 ±2.1 * 106 m3 w.e. yr'1) occurred during the period 1952 1977.

1917-2005

1952-1977

1917-1952

1977-2005

Period

Figure 4.5 Rates o f volume change for Peyto Glacier for periods with the most complete data.

The > 70% dataset has similar results; however, the total volume change (695 ± 406 x
106 m3 w.e.) increased by 4 % when compared to the most complete data set (Figure 4.7).
The periods with the highest rates o f volume loss include 1952 - 1966 and 1991 - 1993. The
period 1950 - 1952 was the only epoch in which the glacier gained volume (95 ± 78 * 106 m3
w.e. yr'1).

47

1952-1977

LIA extent

'^ N j f

{PUBl Ice-cored moraine

I

.. ] Newest ice surface (50 m contour)

Elevation Change (m)

H I 101-120
H I 81 - 100
H H l 6 1 -8 0
I

" 'I 41 -60

I

| 21 -4 0

I'""..I 1-20
j

I - 19-0

1

1 -39 - -20

j

j -59 - -40

1

| -79--60

—

-99 - -80

H

-119--100

HI -139- -120
HI -157- -140

Figure 4.6 Peyto Glacier elevation changes for 1917 - 1952, 1952 - 1977, and 1977 - 2005.

48

O)

§ -20
•5 -30
® -40

at
a. -so -60
-70 -

P eriod

Figure 4.7 Rate o f volume change for Peyto Glacier for periods with the second-most complete data.

4.1.2.2

Lateral moraines

Approximately 13% (-85 ± 4 * 106 m3w.e.) of the total glacier volume change originates
from thinning o f the ice-cored moraine (Figure 4.8). For the periods 1955 - 1966, 1966 1967, and 1973 - 1977 deflation o f the ice-cored moraine respectively accounted for 18, 18,
and 67 % of the observed ice loss. These elevated contributions could be attributed to the
remobilization o f freely available sediment on the ice-cored moraine (e.g., Bennett et al.,
2000). For unknown reasons the ice-cored moraines experienced volume growth for the
periods 1951 - 1952 and 1993 - 1997. Although localized increase in sediment storage could
explain this growth, it is more likely explained from differences between the stereo models
rather than an increase in ice volume.

49

LIAextent
■1947 /
2005 /

Ice-cored moraine

Elevation change (m)
— t 21 - 40
1-20

'" ^ x ^ d j y o -

/,

j

1-1 9 -0

I

j -39 - -20
-59--40

• V

7 9 --6 0

/

-84 - -60
0

0.25

0.5

li. ..I— A— 1— 1

1 km
I,

I

,1

I

Figure 4.8 Elevation change for the ice-cored moraines adjacent to Peyto Glacier from 1947 to 2005.

4.2

Sediment Analyses
4.2.1

Description of sediments

O f 27 sediment core samples collected, I processed 18 percussion cores and one Ekman-box
core. The remaining nine cores were not processed because the laminations were disturbed,
unidentifiable, or too thin to measure. In most cores, clear sediment laminations were not
visible until the sediments were allowed to partially dry which caused cracks to develop. For
50

the length o f the cores (0.2 - 2.9 m), cracks accounted for an average of 4.5 ± 1.8 % of their
length. Additionally, sediment cores either had missing laminations near the water-sediment
interface, disturbed laminations from barrel distortion, or indistinct laminations from too low
o f sedimentation.
The annual nature o f the laminated silt-clay couplets was independently confirmed by
137Cs radionuclide dating as varves (see section 4.2.3). A ‘varve’ was originally described by
De Geer (1912) as a complete annual cycle consisting o f a coarse-grained, pale summer layer
below a fine-grained, dark winter layer. Varves in Peyto Lake are relatively thick (> 1 cm)
as a result of the sediment laden underflows in the summer months, and low sediment input
in the winter months. Visually, the varves typically consist of a thick, light-coloured silt
layer and a very thin, dark-coloured clay cap (Figure 3.6). According to Ashley (1975), the
varves in Peyto Lake belong to Type III classification where the clay cap is less than half the
total varve thickness. In water depths deeper than 30 m, recovered varves are thick with
sharp boundaries. Some of the thicker varves (> 2 cm) contain multiple graded beds which
are interpreted to arise from intra-annual variation in sedimentation rates (e.g., Figure 4.14;
Gilbert, 2003). Closer to the inflow o f Peyto Creek, recovered cores contain indistinctlylaminated sands, gravels, and occasionally cobbles. Varves in this region could not be
distinguished as a result o f the high sedimentation and the dewatering structures present
(Zolitschka, 2007).

4.2.2
4.2.2.1

Bulk-physical properties
Horizontal trends

Bulk-physical properties of the Peyto Lake sediments are spatially variable (Figure 4.9).
The mean water content within the sediments is 37.5 ± 7.2 %. The lowest water content
51

(-29% ) occurs for sediments in proximity to the delta and water content progressively
increases to 52% in the distal and shallower regions. The water content o f Peyto Lake
sediments is inversely related to dry density (1^= 0.97, p < 0.01, n = 17). Water content is
inversely related to density (Menounos, 1997; Hakanson and Jansson, 2002). The coarser
sediments near the Peyto delta are denser (-1.5 g cm'3) than the silts and clays found in distal
settings (< 1.0 g cm'3). The mean dry and wet densities o f lake sediments were 1.27 ± 0.24 g
cm'3 and 2.01 ± 0.18 g cm'3, respectively. The mean organic content was relatively low at
1.9 ± 0.5 %. The spatial distribution o f organic content is moderately correlated (r2 = 0.45, p
< 0.01, n = 17) to water content and moderately correlated (r2= 0.35, p < 0.01, n = 17) to dry
density.

A)Dry DansJty (g o n 4 )

O 0809-1.000
0 toot >1200
^
1.201-1.400
l i p 1.401-1.000

pv

B)vtet Oanatty (g cm ’)
© 1925-1900
^ 1J01-2.000
@ 2.001-2.200

_

1

dr

\(/

J }

Itf

C) Water Content (%)
© 29*30
0 31-35
(£) 39-40
©

« -«

mu

51-66

N

- ^

QV-.

2.201-2.400

D)Organio Content (%}

© 1.2-1.#
0 16-2.0
@ 2-1-2*
/ J y ik ^

Qv-_

2,6' 30. _

01

250
1 500m
J

Figure 4.9 Spatial distribution o f mean bulk-physical properties for Peyto Lake for the period 1917-2010: (A)
dry density, (B) wet density, (C) water content, and (D) organic content. Mean values were calculated from the
interpolated values between sub-sampled locations down the core for every varve. The small arrows indicate the
primary inflow and outflow o f Peyto Creek.

52

4.2.2.2

Vertical trends

Variation o f physical properties also changes with depth (Figure 4.10). Water content in the
proximal portion o f Peyto Lake, for instance, tends to decrease with depth. Core 11 -Peyto
(N), sampled from the proximal basin, has the highest water content (36.3 %) near the watersediment interface and progressively decreases at a rate o f 2.9 % m '1. Such variation can
arise from variable sedimentation rates, quality and character o f the deposits, degree o f
compaction, and bioturbation (Hakanson and Jansson, 2002). For the same core, dry density
is the lowest near the water-sediment interface at 1.22 g cm'3 and increases to 1.39 g cm'3at
152.5 cm of depth. The general trend of organic content remains relatively low and constant
at 1.56 ± 0.27% .

53

Dry Density (g cm'3)
1.00

1.25

I

I

1.50

1.75

I_______ I

o -i

S

o

CM

o _
fo>

o>
in
CO
Ol

h>
h- Q
01

o _

<

EE

o

=

w 8

a>
Q

“

^® £to
CO 0)

p-

o -

=

=_

in

o> oCO
CO
01

CO
CO
01

<

I"01

© -J

0 1 2 3 4 5
Varve thickness (cm)

0.0 0.5 1.0 1.5 2.0 2.5
Organic Matter (% LOI 550aC)

Figure 4.10 Bulk-physical properties for core 1 l-Peyto(N). Dry density (dark grey line) and organic matter
(light grey line) were sub-sampled at 2 cm intervals. Varve thickness was interpolated (red line) near the top
and measured (black line) downwards.

4.2.3

Cesium - 137 (137Cs)

The maximum 137Cs activity level occurs at 45.5 cm of depth in core 10-Peyto(02) and
independently confirms the varved nature o f the sediment record (Figure 4.11).
54

Counted varves (Calendar years AD)
05

...
CO CO 050XJX7) 0)0)

«MD<0«O<D<O

g

0>0>0)050505

I 1 I I I I I It I I I I 1I

—I_

sz
te> .3*

T5

b» «
2
Q
Q CN
i-

Q_

D

l

j

i—I—

i—
70

60

50

— I—

40

20

Sedim ent depth (cm)

Figure 4.11 Activity levels o f Cesium-137 ( 137Cs) in lake sediments retrieved from core 10-Peyto(02). Vertical
error bars indicate counting errors (± 1 a) in 137Cs activity determination. Sediment samples were taken from the
centre o f identified varves and converted to calendar years (AD).

4.2.4

Varve thickness

Varve thickness generally decreases with distance from the inflow o f Peyto Creek, a common
pattern found in proglacial lakes (Figure 4.12; Smith and Ashley, 1985; Menounos et al.,
2005). The median varve thickness decreases from 2.12 cm (core = I, n=50) near the delta to
0.12 cm in the lake’s most distal setting. A multiple linear regression model revealed that 66
% of the variation in mean varve thickness for the period 1917-2010 can be explained by
distance from the point o f inflow and lake depth. For individual years, however, the relation
is far more complex and could not be explained by the two variables alone. Other
environmental variables such as climate (Zolitschka, 2007), extreme discharge (Sander et al.
2002), change in sediment storage, and/or the complex dynamics o f Peyto Lake (Chikita et
al., 1996) might reflect the inconsistent patterns o f varve thickness from year to year.

55

Proximal+

-► Distal

Basln

u
00>
>
a>

Shelf

Basin

<o

I (N
£
O
K

L

N

M

O

05

c

03

D

P

F

B

G

01

02

A

R

Core
Figure 4.12 Varve thicknesses measured from all collected cores. Proximal cores (left) were collected up to 316
m from the delta and distal cores (right) were collected up to 2215 m from the delta. Note: measurements from
cores F, A, and R, were done between marker beds, so varve thicknesses denote average rates.

For the period 1917 - 2010 the median varve thickness for all cores was 1.20 cm
(Figure 4.13). The extended varve record measured from two cores (02 and O) revealed that
varves deposited during the period 1917 - 2003 are 207 % thicker and more variable (1.14 ±
0.37 cm) and than varves deposited during the previous period 1830 - 1916 (0.55 ± 0.11 cm).
5
Cora: 10-P*yto(02)
Cow: 11-P«yto<0)
4

B
■8. 3

■k

t
!»

>

1

0
1830

1840

1850

1860

1870

1880

1890

1900

1910

1920

1930

1940

1950

1980

1970

1980

1990

2000

2010

Calendar YMr (AD)

Figure 4.13 Extended varve record from two cores. Vertical black lines delineate periods 1828 - 1916 and 1917
- 2003. Horizontal red lines show the average varve thickness for those periods.

56

Median varve thickness exeeded 2.0 cm in six years (1941, 1942, 1961, 1970, 1983,
and 1998). The thick varve o f 1983 likely arose from sediments delivered during an extreme
precipitation event which elevated streamflow and caused mass wasting (Johnson and Power,
1985). The thick varves of 1970 and 1998 coincide with years o f strong negative mass
balance o f Peyto Glacier. The nature o f the thick varves shown for 1941, 1942, and 1961
cannot be associated with other records; however, the microstratigraphy suggests they
formed during conditions similar to that o f the 1998 varve (see Figure 4.14).

1942

1970

.
1941
4.

1961

1 .

1998

23

.

5

6.

cm
1983

Figure 4.14 Photographs o f abnormally thick varves from core 1 l-Peyto(O).

57

The inter-annual variation o f varve thickness shows a distinct temporal trend (Figure
4.15). The extended varve record shows there was a distinct shift in sedimentation in 1870
(Figure 4.13). From 1870, varves slowly and progressively became thicker to a maximum
thickness in 1942. Between 1943 and 1954, varves thinned but thickened again during the
period 1954 - 1957. After 1958, varves slowly and progressively became thinner for the
following 42 years. Over the period 2000 - 2010, varves thickened from 2000 until 2004 and
thinned thereafter.

'------- < n v i v m m i n 'TTT n rrrr r rr rrrf t n t m 11 i m i i i 11 11 r n 'Trrr r rr r r r n 11 i i i n ii 11 »i i n r i i i i i i i 11 11i i i

1

1917 1921 1925 1929 1933 1937 1941 1945 1949 1953 1957 1961 1965 1969 1973 1977 1981 1985 1989 1993 1997 2001 2005 2009

Calendar Year (AO)

Figure 4.15 Boxplots o f measured varve thicknesses for the study period (1917 - 2010). Red line denotes an 8
year running mean. The period 2000-2010 contains less than 14 contributing cores.

4.2.5

Sediment yield calculations

I used two interpolation methods to estimate the areal extent of varve thickness. The spatial
distribution o f sediment thickness for both interpolation methods substantially differ, but
estimates of regularized spline interpolation are more realistic than those using the Thiessen
polygon method (Figure 4.16). Not surprisingly, standardized varve thickness (Figure 4.17)

58

is correlated with SSY estimated with Thiessen polygons (r = 0.94, p < 0.001) and
regularized splines (r = 0.93, p < 0.001; Appendix B).

Sedim ent thickness (m)

O

H

Core locations

>20m
>10 m

I

Iso b a th s (m )

] >0m

—-

I

I 0-002

M

0.02 - 0 04
0.04 - 006

0, 10, 20, 30

0.06 - 0.08

3 5 ,4 0

0.08 - 0.1

0

0 25

4 1 ,4 2 ,4 3 ,4 4 ,4 6 ,4 8

0 . 1 0 - 0.12

*—

km

0.5

Figure 4.16 Examples o f how sediment thickness is interpolated across the lake. (A) Peyto Lake bathymetry
highlighting various depths where the rate o f sedimentation is assumed to be 0 m yr'1. (B) Sediment thickness
in 1970 was uniformly distributed within each Thiessen polygon that surrounds a sediment core. (C) Sediment
thickness in 1970 using regularized spline interpolation (25 m * 25 m pixels). Sediment values incorporate
areal variation in sediment density and organic content.

59

o
<N

tO
<0 «■)
c
u

jc

i5

P
ir>

o

o
o

1*0 u>
k. O

(0 '
T3
C ©

o

<N
T

1915

1920

1925

1930

T

T

1935

1940

1945

1950

1955

1960

1965

1970

1975

T

1980

1985

1990

1995

2000

2005

2010

C alen dar Year

Figure 4.17 Master chronology o f varve thickness variation. The y-axis is reported as standardized departures
from a mean o f zero.

The sensitivity analysis reveals the magnitudes o f the observed trends differed among
the approaches, but the general trends remained the same (Figure 4.18). Comparably,
SSYspiine, SSYThiessen-20. SSYjhiessen-io were respectively calculated as 30 %, 20 %, and 10 %
lower than the SSYihiessen-o interpolation (Appendix B). Provided all of these approaches
yield similar trends in sediment yield, I elected to simplify the following discussion by only
considering data from the SSYspijne interpolation (Figure 4.19). For the period 1917 - 2010,
SSYSpiine averaged 446 ±176 (la ) Mg km'2 yr"1.

60

*[_ S2 ~| Thiessen- Om —
>.
_ Thiessen - 10m —
<? o
Thiessen - 20 m —
E ^ “
Spline - 20 m —

u> ~

j

it N \ /,

-------------- 1-------------- j-------------- 1-------------- 1-------------- 1---------------1
— ,--------------------,------------------- ,-------------- 1
1920

1930

1940

1950

1960

1970

1980

1990

2000

2010

C alendar Year

Figure 4.18 Calculated specific sediment yield for the Peyto Lake watershed when sampled sediment thickness
was interpolated using Thiessen polygons and spline interpolations. Sedimentation rates were assumed to equal
0 m yr'1 when depths were shallower than 0 m, 10 m, and 20 m.

"D

CL ®
in

1920

1930

1940

1950

1960

1970

1980

1990

2000

2010

Calendar Year
Figure 4.19 Specific sediment yield calculated by using a regularized spline interpolation where sedimentation
was assumed to be near 0 m yr'1 at 0 - 20 m o f depth. The grey horizontal line denotes the mean and the black
dashed line denotes an 8 year running mean.

4.3

Environmental controls on sediment yield

There are many interacting and complex environmental processes that can influence the
sedimentation in a lake (Hodder et al., 2007). Prior to exploring the cause (i.e., environment)

61

and effect (i.e., sediment yield) relation, the interactions among environmental variables are
presented.
Glacier mass balance is closely linked to meteorological conditions that occur at the
glacier surface (Rasmussen and Conway, 2004). Generally, the mass balance record for
Peyto Glacier was most closely linked to the average temperatures for Jasper. The average
maximum summer temperature (June, July, August) from Jasper was significantly correlated
with summer mass balance (r = -0.68, p < 0.001, n = 27). Similarly, the average minimum
temperatures (May, June, July) were significantly correlated to net mass balance (r = -0.73, p
< 0.001, n = 43) and winter mass balance (r = -0.78, p < 0.001, n = 27), respectively.
The hydrometric record of Peyto Creek is relatively short (8 years in length) and
obtaining a strong relation to long SSY record (94 years in length) was not ideal. Instead, I
used the longer hydrometric records from the nearby stations on the Mistaya and the
Sunwapta rivers to reflect regionally representative hydrologic conditions in the area. For the
month o f August, Peyto Creek discharge is highly correlated to Mistaya River discharge (r =
0.89, p = 0.003, n = 8) and the Sunwapta River discharge (r = 0.90, p = 0.002, n = 8). Mean
August discharge for Peyto Creek is also correlated to the mean maximum temperature for
August (r = 0.85, p < 0.007, n = 8). Mistaya and Sunwapta rivers are highly correlated for
mean August discharge (r = 0.78, p <0.01, n = 51), but this correlation decreases for mean
maximum daily discharge (r = 0.40, p <0.01, n = 52). Net, summer, and winter mass balance
records are uncorrelated with discharge from all three hydrometric stations.
SSY is not correlated to glacier mass balance or discharge, but is weakly correlated
with climate. The mean August maximum temperature for Jasper had a weak correlation with

62

SSY (r = 0.37, p =0.001, n = 75) and standardized varve measurements (r = -0.43, p < 0.001,
n = 75).
The varve records from several individual cores are correlated with hydro-climatic
variables. The varve record from Core N, for example, was weakly correlated to winter mass
balance (r = -0.30, p =0.05, n = 28) and net mass balance (r = -0.47, p =0.001, n = 42). The
standardized variation of varve thickness was also weakly correlated with net mass balance (r
= -0.36, p =0.02, n = 44), ELA (r = 0.38, p =0.01, n = 44), and AAR (r = -0.42, p =0.001, n =
43).
Changes in glacier dimensions were calculated for periods when aerial photographs
were available. For these periods, SSY has no relation to changes in length, area, volume,
and net mass balance o f the glacier.
The correlations among glacier mass balances, discharge and climatic variables
limited the possible number o f variables used in a multiple linear regression. Individually,
net mass balance or discharge had no relation to SSY; together the two variables had a weak
relation with SSY. SSY had a moderate correlation (r = 0.45, p = 0.009, n = 44) with net
mass balance and mean discharge for July and August (Q ja ) from the Mistaya River.
Similarly, net mass balance and mean Q j A had the strongest correlation (r = 0.52, p = 0.002, n
= 44) with the standardized varve record. The relation improved slightly (r = 0.53, p = 0.001,
n = 44) when the net mass balance (bn) was converted to total volume change w.e. (i.e., bn *
glacier area). There were slightly lower significant correlations when discharge was
averaged over different months (Table 4.2). Sediment cores C and 03 were the only cores
that had significantly weak correlations with discharge and net mass balance. I also

63

examined the potential for temporal lags among the systems, but no improved relations were
found.
Table 4.3 Pearson correlation coefficient (r) for Mistaya River discharge versus net mass balance and
standardized varve thickness or SSYmim,. p < 0.05._____________________________

Months of mean
discharge*
JJA
JA
A
JAS
MJJASO

Standardized
Varves (r)

SSYtpiim
(r)

0.510
0.520
0.514
0.506
0.512

0.431
0.455
0.451
0.439
0.450

*M = May, J = June, J = July, A = August, S = September, O = October

64

5.

Discussion
5.1

Changes to glacier length, area, and volume
5.1.1

Length

Over the period 1917 - 2005 Peyto Glacier receded -2198 ± 18m which corresponds to a
recession rate o f -24.98 ± 0.20 m yr'1. Previously, changes to the length o f Peyto Glacier
were also reported and calculated from photographs (Thorington, 1934; Kingman, 1938),
botanical dating of the moraines (Field and Heusser, 1954), annual field surveys
(Ommanney, 1972) or employing a combination o f these methods (Meek, 1948; Brunger et
al., 1967; Henoch, 1971). Ommanney (1972) reported that between 1917 and 1962, Peyto
Glacier receded -1141 m which is similar to my estimate (1125 ± 18 m) for the same period.
In comparison to regional glaciers, termini on the Columbia Icefield receded at a
slower rate o f -12.8 ± 0.4 m yr'1 for nearly the same period (1919 - 2009; Tennant and
Menounos, 2013). Similarly, Castle Creek Glacier in the Cariboo Mountains receded at the
lower rate of -14.3 m yr'1 between 1959 and 2009 (Beedle et al., 2009). For the periods 1917
- 1949 and 1919 - 1948, Peyto and Athabasca glaciers receded at similar rates o f -20.5 ± 0.6
m yr"1 and -20 m yr'1 (Kite and Reid, 1977), respectively. Also, Denton (1975) reported
Saskatchewan Glacier to have receded -25 m yr'1 (1953 - 1963), whereas Peyto retreated
-42.3 ±0.1 m yr'1 from 1955 to 1966. Ommanney (1972) recorded the highest rate o f retreat
(-74.7 m yr'1) from 1956 - 1960 when the glacier receded into a deep, narrow gorge. Due to
topographic conditions near the snout, government surveys of terminus change were
suspended after 1962 (Ommanney, 1972).

5.1.2

Area

65

From 1917 to 2005 Peyto Glacier lost -4.0 ± 0.9 km2 which corresponds to a shrinkage rate
o f -0.045 ± 0.010 km2 yr'1. Previous studies calculated shrinkage rates o f -0.036 km2 yr'1
from 1896 to 1917, -0.066 km2 yr"1 from 1917 to 1966 (Wallace, 1995), and -0.047 km2 yr"1
from 1966 to 1989 (Glenday, 1991). From 1917 to 1966, my estimate of shrinkage rate
(-0.022 ± 0.018 km2 yr'1) is 67 % smaller than the estimate o f Wallace (1995). Such a large
discrepancy may be explained by Wallace’s use o f the original 1917 glacier boundary map
where he did not correct for the substantial topographical error. My results, however, are
comparable with Glenday (1991) for the period 1966 - 1989.
When compared to other glaciers in the Canadian Rocky Mountains, Peyto Glacier
shrank at a moderate rate. DeBeer and Sharp (2007) report that glaciers lost -15% o f their
area during the period 1951 - 2001, which was slightly less than what I calculated for Peyto
(-17.8 ± 0.3 %) for the same period. The Columbia Icefields lost -33.5 ± 6.1% of its area
during the period 1919 - 2009 (Tennant et al., 2012) which is slightly more than what I
calculated for Peyto Glacier for the period 1917 - 2005 (-26.4 ± 5 .8 %).

5.1.3

Volume

My results generally agree with previous calculations of volumetric changes for Peyto
Glacier. They indicate that Peyto Glacier, excluding the ice-cored moraines, lost -515 ± 404
x 106 m3 w.e. from 1917 to 1991 with an annual volume change o f -5.9 ± 4.6 * 106 m3 w.e
yr"1. The combined results from Wallace (1995) for 1917 - 1966 and Glenday (1991) for
1966 - 1989 equate to a total loss o f -854.9 x 106 m3. If I convert their volume change to
•5

snow water equivalent using a mean ice density o f 760 kg m ", their volume change equates
to -649.8 x 106 m3 w.e. which is an average rate o f -9.0 x 106 m3 w.e yr'1. The large

66

discrepancy between our results is probably from different 1917 extents (15.2 km2vs. 16.0
km2) being used. If I account for different 1917 extents being used by dividing volume by
average area, my results show Peyto Glacier thinned an average o f -0.51 ± 0.40 m w.e. yr'1
(1917 - 1991) which is within error o f Wallace and Glenday’s calculations (-0.64 m w.e. y r'1:
1917 - 1989). From the mass balance record (1966 - 2007), Peyto Glacier thinned an average
rate of -0.63 m w.e. yr'1 (Demuth and Keller, 2006), which is comparable to my calculations
(-0.55 ± 0.04 m w.e. yr'1: 1966 - 2005). My results indicate that Peyto Glacier thinned -44 ±
31 m w.e. on average from 1917 to 2005 (-0.50 ± 0.35 m w.e. yr'1). Glaciers o f the Columbia
Icefield (Tennant and Menounos, 2013) for the period 1919 - 2009 thinned by a comparable
magnitude (-49 ± 25 m w.e.) and at a similar rate (-0.6 ± 0.3 m w.e. yr'1). For the southern
Rocky Mountains (approximate ice cover is 849 km2) from 1985 to 1999, Schiefer et al.
(2007) calculated an average thinning rate o f -0.64 ± 0.15 m w.e. y r'1.
In this study, rates of volume change calculated for the periods 1917 - 1949, 1949 1950, 1950 - 1952, and 1966 - 1967 have large uncertainties. The error term for the period
1917 - 1949 is relatively large because o f the uncertainty o f the 1917 IBCS map. I expect
periods less than three years to have large errors, but the reason for the extremely large error
for 1949 - 1950 is associated with the quality of the stereo models. For the period 1950 1952, Peyto Glacier could have gained volume as suggested by an increase in area and the
minimal retreat from 1950 to 1951. The primary concern for the period 1950 - 1952 is the
abnormally large rate o f volume change which is far larger than any other period. Although
the error term for the period 1966 - 1967 is large, it is within error o f the mass balance record
of 1967 (i.e., +10 ± 200 mm w.e.).

67

Many studies identify the ice-cored moraines adjacent to Peyto Glacier (e.g., 0strem
and Arnold, 1970; Ommanney, 1972; Nakawo and Young, 1982; Binda, 1984; Johnson and
Power, 1985; Glenday, 1991), but few studies quantify its contribution to meltwater in the
basin. Using sequential LiDAR surveys over a 23 month period, Hopkinson and Demuth
(2006) calculated that the ice-cored moraines contributed to 6% o f the total glacier runoff
which is about half of my estimate (11 ± 1 %) for the period 1977 - 2005. Like Hopkinson
and Demuth (2006), I found no evidence that the downwasting is due to mass movement
events with the exception of the mass wasting events of 1983 (Johnson and Power, 1985).

5.2

Sediment yield

Due to complex dynamics of sediment transport in Peyto Lake, I was unable to interpolate
sedimentation using a spatial statistical model (e.g. Schiefer, 2004). Instead, I performed two
methods of interpolation and assessed the reliability o f these methods using a first-order
sensitivity analysis. My results indicate that interpolation methods affect the quantity of
estimated sediment accumulation in the lake for a given year, but that inter-annual trends in
sediment yield remained insensitive to the type of interpolation methods used.
Additional sediment cores would have been useful to determine: (1) the depth at
which annual sedimentation rate was close to zero; and (2) the rate and extent o f delta
progradation (e.g., Appendix C). In this study, I assumed that negligible sedimentation
occurred in water depths shallower than 20 m. The rate and extent of delta progradation was
also not determined because few samples were taken in proximity to the Peyto Lake delta,
and correlation among these cores was difficult. Sedimentation rates near the delta are
highly variable where previous research shows a fivefold reduction in sedimentation within a

68

100 m distance (Vendl, 1978). In addition, aerial photography used in this study indicates
Peyto Creek laterally migrated over the delta’s surface which complicates areal
sedimentation patterns in the proximal basin.
SSC have been studied in the watershed (e.g., Vendl, 1978; Smith et al., 1982; Binda,
1984; Chikita, 1992, 1993; Chikita et al., 1996); only Vendl (1978) reported annual values.
For the summer o f 1976, Vendl (1978) calculated the total suspended sediment flowing into
Peyto Lake as 38000 Mg (approximately 778 Mg km'2 yr'1). For the same year, I calculated
sediment yield to be on the same order of magnitude: 529-789 Mg km'2 y r'1 depending on
the method of interpolation. My calculations are probably underestimated because they do
not consider the coarser particles (e.g., gravel) deposited in the delta region. The rough
agreement between our values suggests that my SSY calculations are representative of the
Peyto Lake watershed.
For the period 1917 - 2010, SSY averaged 446 ± 176 Mg km*2 yr"1 which is among
the highest SSY values in the Canadian Cordillera when plotted by glacier cover (Figure 5.1).
Hodder et al. (2006) calculated average SSY for a number o f glacierized watersheds in the
Canadian Cordillera and suggested that glacier cover is a good predictor for SSY for a given
tectonic belt and climate region. My results, however, show that SSY for Peyto watershed
does not fit this relation. SSY for the Peyto Lake basin is among the highest fluvial based
estimates o f SSY for British Columbia when plotted by drainage area (Figure 5.2). Church
and Slaymaker (1989) only report four undisturbed watersheds that were above the inferred
pattern for British Columbian rivers; all o f these outliers are from glacierized watershed.
Although SSY for the Peyto watershed is among the highest documented values in western
Canada, it is not particularly high for glaciated watersheds elsewhere (Figure 5.3). Hallet et

69

al. (1996) showed glaciated basins have a higher and distinct SSY when compared to non­
glaciated basins around the world. They found yields typically increased with basin size,
from cirque glaciers to fast-moving glaciers in Alaska, presumably due to higher effective
erosion rates o f the latter. In short, SSY for the Peyto watershed are among the highest in
western Canada, yet relatively low among other glaciated watersheds around the world.

i ------------- r

400
2
a>
‘5k

Peyto
r*‘

T

A Meziadin

1

T

A Bowser

300

I?
E
rE200

A Llewellyn (pre 1950)

CD

Stave

o<D
a.
(A

o

D Sunwapta
Chilko

O

100

A

—Lillooet/Harrison

A Llewellyn (post 1950)

Hector

O
Moose

i

0

N ostetuko

A Lillooet

for lakes in the Foreland belt (n = 5):
(SSYl = 23+3.57*[%glacier cover] P- = 0.98

Chephren
am son

i

I

60

40

20

0

_L
80

glacier cover (%)
Figure 5.1 Specific sediment yield as a function o f percent glacier cover for lakes in British Columbia and
Alberta (Graph modified from Hodder et al. (2006)). The various symbols represent different tectonic belts.

Drainage area (hectares)

10

100

s '* ? .

0.01

0 001
00'

100

Luano

1.00G.00C

D rem age a n e firm 2)

Figure 5.2 Specific sediment yield as a function o f drainage area (Graph modified from Church and Slaymaker
(1989)). Specific sediment yield for Peyto Lake (star) is shown on the graph.

70

10’

E
10co
5
2
<D
>“
_
10
c
a>
E
TJ
a)
co
o 10;
i£
O
0)
a.

i

•f1
*

n

JS
Mf1
f

♦T

CO

10 ‘
10 '

(+ it

_X
* *5§».

f
±

' 1
'
101

i

r

€Tfl....
.
A.

i

f

„ *
___________

10

10;

10'

Basin Area (km2)
Figure 5.3 The relation o f specific sediment yield to basin area. For comparison, the Peyto watershed (star) is
plotted among other watersheds with glacier cover (graph modified from Hallet et al., 1996).

5.3

Influence o f glacier change on sediment yield

Many environmental variables control lake sedimentation as shown by the conceptual
diagram by Hodder et al. (2007; Figure 5.4). The diagram highlights the complex
interactions among six basic environmental systems. In this study, I explicitly investigated
how the glacier system (i.e., dimensional changes) influenced the lacustrine system (i.e.,
sediment yield). To account for confounding effects from the climate and fluvial systems, I
used available climate and hydrometric records from the region. The results from this study
could not clearly link the SSY record to the glacier, climate, or fluvial systems at annual to
decadal timescales. It is possible that the combination o f all three systems influenced the
SSY record to some degree, but the majority o f SSY variance remains unexplained.

71

Meltwater
Seasonal ablation
Flow capacity
Hysteresis
JOkdhiaups

Regional Climate

Sediment

LaOude
Elevation
ConfinentaMy
Ocean currents
Atmospheric pressure cells
Sea surface temperature
Sea surface pressure
PrevaMng wand

Enosian
Substrate type
Storage
Release timing
Glacier history
Non glacial sources

-Ice

-Local Climate

Mass balance
AdvanceAetraat /
Surging
fr

Insolation
Cloud cover
Temperature
Rain
Snowfall and melt

Glacier
System

IMicrometeorology

Slope processes
Weathering
Watershed size
F>ara0ac»l slug migration

r-Geoiogy

Ltthoiogy
Structure

Geologic &
Geomorphic
System

Climate
System

I

Lacustrine
System

Terrestrial
Biological
System
Fluvial
System

Human activity

Water Discharge

Animal activity

Temperature
Predptalion
Water storage
Persistence

2

&
«

1§
55

s
co

CO

CO

Sediment Discharge
Sedknent rating curve
Flaw capacity/competence
Rare everts
Sedknent erosion
Storage

II
> |

Deforestation
River engineering

Linear Statistical Models

L Sediment Dispersal
Density
CorioSs effect
Wild
Basin geometry
Trap efficiency
Local events
infrequent events
Flocculation

Geomorphic

Grazing
Burrowing

Vegetation
Type and density
Fir®

Slope stability

Lacustrine
Sediment
■■■■a

L Varve thickness
■-Bulk density
t-Mass accumulation rate
Figure 5.4 Conceptual diagram o f the proglacial alpine systems and their linkages to lacustrine sediment
(Hodder et al., 2007).

Based on the results presented in chapter 4 , 1 surmise that Peyto Lake SSY is related
to the volume o f glacial runoff for the following reasons. First, years of high SSY occur

72

during years o f strong negative mass balance or high mean temperatures (e.g., 1929, 1934,
1941, 1970, and 1998). The SSY peak o f 1970, for example, coincides with a drought year
where glacier melt contributed up to 82% o f the streamflow (Young, 1991). Second, about
71% of the Peyto Creek discharge originates from Peyto Glacier (Schuster and Young,
2006). Peyto Creek discharge was also shown to be related to related to SSC (r=0.61) and
suspended sediment load (r = 0.88; Vendl, 1978). Third, 27 % o f the annual variance in SSY
was attributed to net mass balance and Mistaya River discharge records. Although the
relation is weak, it is significant and suggests that glacier runoff is linked to SSY, but in a
complex manner.
The long-term trend in SSY appears to mirror long-term changes in glacier runoff. I
hypothesize that discharge contribution from Peyto Glacier reached its peak in mid-20th
century and the resulting trend is evident in the SSY record. As a glacier recedes and ice
melts, its potential contribution to water supply diminishes over time (Moore et al., 2009;
Nolin et al., 2010). Peyto Glacier has continuously retreated since the end o f the Little Ice
Age; the glacier has lost over 70 % o f its volume since 1896 (Demuth and Keller, 1996). As
glacier volume declined there was likely a similar decline in glacier runoff. The extended
varve record indicates that varves thickened from 1870 to about 1940. If SSY and the
magnitude of glacier melt are related, then maximum glacier runoff from Peyto Glacier
occurred at about 1940. For the period 1940 - 1980 SSY remained relatively level and high
which corresponds to the period (1952 - 1977) with the highest rate of glacier change.
For the proceeding period 1977 - 2005, two additional factors may explain trends in
SSY. First, the Pacific Decadal Oscillation (PDO) shift in 1976 caused glaciers throughout
western North America, including Peyto Glacier, to lose more mass (Luckman, 1998; Moore

73

et al., 2009). In spite o f the strong years o f negative net mass balance, the rate o f volume
change (Figure 4.5) for Peyto Glacier declined confirming that peak flow was likely reached
prior to 1976. Second, SSY concurrently declined after 1980, which may be likewise related
to the long-term decrease in glacier volume. For the period 1977 - 2005, however, my results
indicate the rate of sediment delivery to Peyto Lake was considerably less than the rates for
the periods 1917 - 1952 and 1952 - 1977. This decrease in SSY could be explained by
several factors including sediment exhaustion from sedimentary stores adjacent to or beneath
the glacier (long-term hysteresis), and/or a decrease in the rates of primary erosion. Basal
erosion rates, for instance, would be expected to decline as the glacier area in contact with
bedrock declines (Hallet et al., 1996).
Declining contribution of glacial meltwater to streamflow has been studied elsewhere
in the Canadian Rocky Mountains. Research by Demuth and Pietroniro (2003), for example,
found the mean and minimum discharge records o f the Mistaya River have been declining
since 1950 in spite o f a modest increase in precipitation and temperature. They also suggest
the ability of glacier cover to regulate streamflow in the Mistaya basin may have been in
decline since the mid-1900s. In the neighbouring Bow River watershed, Hopkinson and
Young (1998) found the ability of the receding Bow Glacier (ca. 1998) to enhance Bow
River flow would have been reduced by 2% had the same climate as 1970 occurred. The
summer of 1970 was abnormally hot and dry which resulted in an exceptionally high
streamflow contribution by glaciers.
This study specifically examined the relation between SSY and dimensional changes
o f Peyto Glacier, but other factors besides glacier change are known to influence SSY in
glaciated catchments. Changes in sediment storage, both beneath the glacier and in the

74

watershed, likely influenced changes in the SSY record. Seasonal hysteresis between
discharge and SSY are common beneath glaciers (Riihimaki et al., 2005). Binda (1994), for
example, hypothesized that observed hysteresis between SSC and discharge for Peyto Creek
was due to the re-organization of sub-glacial channels and sediment availability. As the melt
season progresses, sediment exhaustion likewise increases (Gumell et al., 1994). These
temporal changes in sediment availability also occur in proglacial regions and streams. In
addition to the sediment supplied by the glacier, sediment could also derive from local
channel and hill slope processes. Orwin and Smart (2004) found up to 80 % o f the total
suspended sediment yield came from the proglacial area below Small Glacier, BC.
Stochastic mass movement events within the watershed would also increase the SSY. Based
on the photographic record from the basin, I did not find evidence for notable erosive
processes within the Peyto watershed. Finally, intermittent storage areas for sediment
between Peyto Glacier and Peyto Lake could also induce a time lag that exists at diurnal,
seasonal, and annual timescales.
In summary, a number o f factors confound the relation between SSY and dimensional
changes of Peyto Glacier. I hypothesize that glacier runoff is the primary influence on the
SSY trend which climaxed in 1941, but remained high from 1940 to 1980. Elevated runoff
would have increased the potential to erode and transport sediments beneath and adjacent to
the glacier. My results also suggest the rate of sediment delivery to Peyto Lake, relative to
volume change, has declined in recent decades possibly resulting from sediment exhaustion
or a decline in primary erosion rates.

75

5.4

Source of errors and limitations of study
5.4.1

Digital photogrammetry and measurements

Photogrammetric accuracy largely depends on the quality and coverage o f existing aerial
photographs. In essence, lower coverage or fewer measurements translates to less accurate
data, more estimated data, and therefore less reliable data. Incomplete coverage required
spatial and temporal interpolation, typically for areas in the accumulation zone o f Peyto
Glacier. Low photographic contrast in the accumulation zone and in shadows inhibited my
ability to perceive depth and hence measure surface elevation change (Vanlooy and Forster,
2011). Determining elevations in areas of steep relief around Peyto Glacier was also difficult
and probably led to larger errors in comparison to measurements made in areas o f flatter
terrain (Fox and Nuttall, 1997; Schiefer and Gilbert, 2007).
Additional sources of error include corrections required for the timing o f the aerial
photography and the lack of independent estimates o f glacier elevations. The surface
elevation o f the glacier continuously changes as snow accumulates or melts throughout the
year, so it is important to compare elevation points collected in the same season (Vanlooy et
al., 2006). As the time between consecutive sets of aerial photography increases, the
magnitude of this error term decreases (Zemp et al., 2013). For my study, consecutive
photos were acquired 14-61 days apart for periods less than two years during the melt
season. By using a temperature index model, volume adjustments were made to compensate
for different acquisition dates; however, this correction does not correct for changes in length
or area. Finally I used checkpoints on stable, unvegetated terrain with high photographic
contrast to assess possible elevation biases in my stereo models for a given year. This error
analysis was conducted in regions o f high contrast which allows improved photogrammetric

76

height measurements; however, this is not entirely representative o f measurements done on
the glacier surface where low contrast is prominent (Fox and Nuttall, 1997). Independent
measurements of surface elevation for the glacier would allow me to assess the reliability o f
my elevation measurements for Peyto Glacier.

5.4.2

Sediment yield

Reliability o f sediment yield estimated from varved sediment records depends on the quality
of the lake sediment samples and the interpretation o f varved sediments by the researcher.
Zolitschka (2007) outlines four types o f systematic errors associated with varve
chronologies: technical problems, depositional events, sedimentation rates, and varve
preservation. Incomplete core recovery and core barrel distortion are technical errors
inherent in sediment core recovery (Glew et al., 2001). Depositional events (e.g., turbidites)
within localized areas of the lake can influence varve thickness. Both low (< 2 mm) and high
(> 20 mm) sedimentation rates can make varve identification difficult (Zolitschka, 2007).
Changes in lake level or autochthonous productivity can influence the preservation o f varves.
In addition, interpretive errors from counting extra varves (Type A error) or not counting
varves (Type B error) can arise if the chronology cannot be confirmed independently. A
relative error (Type C error) can arise if Type A and B errors are consistent among core
chronologies (Lamoureux, 2001).
In this study, sediment yield calculations were subject to the quality o f sediment
samples, depositional events, interpretation of varve boundaries, and interpolation of
sediment thickness across Peyto Lake. I was able to distinguish the water-sediment interface
on one core; however, other cores either had the upper sediments missing or the upper

77

sediments lacked sharp varve boundaries. All cores had some degree of barrel distortion
ranging from minor “U-bending” to severe cracking and bending from pressurized freezing.
The lack of water-sediment interface inhibited the measurements o f more recent varve
formations and the sediment distortion likely altered measurements of varves. Depositional
events were widespread in some cores (e.g., H, I, and J) nearest the delta which prevented the
measurement o f the varves and correlation with other cores. Low sedimentation prevented
correlation among cores and measurements within the shallow regions o f the lake, but years
o f high sedimentation increased the likelihood of Type A errors in the varve chronology. For
most years, correlation among cores was done with multiple varves reducing the likelihood
o f Type A and Type B errors. Independent confirmation o f the 1983 flood event and
identifying 1963 peak associated with 137Cs radionuclide dating suggests that my varve
chronology is reliable. However, my chronology is still subject to Type C errors.
Interpolations o f the varve measurements and chronologies were conducted to calculate a
range of sediment yields. Two methods o f interpolation were conducted which resulted in a
very similar pattern over time suggesting that interpolation methods has a low influence on
the temporal pattern but a larger influence on the magnitude of sediment yield.

78

6.

Conclusion and suggestions for further study

The primary goal o f this study was to assess the influence o f glacier activity on sediment
yield within the Peyto Lake watershed. I used photogrammetry to determine dimensional
changes (length, area, and volume) o f Peyto Glacier. From 1917 to 2005 Peyto Glacier
retreated -2198 ± 18 m, shrank -4.0 ± 0.9 km2, thinned -44 ± 31 m, and lost a volume o f -5 81
± 404 x 106 m3 w.e. From 1917 to 2010, SSY averaged 446 ±176 Mg km2 yr'1 which is
among the highest in the Canadian Cordillera, but relatively low for glaciated basins.
Measured dimensional changes to Peyto Glacier, along with supplemental mass balance,
climate, and hydrometric records, were compared to the SSY record. Dimensional changes to
Peyto Glacier were uncorrelated to SSY. About 27% o f the SSY variance could be explained
by net mass balance and Mistaya River discharge. Glacier change likely influenced SSY at
the inter-annual to inter-decadal scales, but the majority o f the variance remains unexplained.
The work presented in this thesis provides a SSY record for the Peyto Lake watershed
and expands the known record of length, area, and volume changes for Peyto Glacier. The
Peyto Lake watershed was an ideal site to examine the influence o f glacier activity on SSY
for a number of reasons: (1) there exists 18 sets o f air photos and a historical map that dates
to 1917; (2) Peyto Glacier has been the focus o f many types of studies; (3) the history of
Peyto glacier has been well-preserved on the landscape; (4) sediment transport from the
glacier is directly coupled to the lake basin; (5) Peyto Glacier has the longest mass balance
record in western Canada; (6) there are nearby climate and hydrometric stations; and (7)
Peyto Lake sediments are varved. I was, however, not able to identify primary
environmental controls o f SSY and this outcome highlights the complexity o f sediment
transfers in glacierized basins. Additional research could build on this study and use sediment

79

budget framework to understand sediment transfers in the Peyto Lake catchment. A sediment
budget study could identify the rate that sediments can be produced and evacuated from the
sub-glacial environment of Peyto Glacier. Additional work should also quantify long-term
changes in sediment storage between the glacier’s snout and Peyto Lake.

80

Literature cited
Andreassen, L.M., Elvehey, H. and Kj0 llmen, B. 2002. Using aerial photography to study
glacier changes in Norway. Annals o f Glaciology, 34: 343-348.
Appleby, P.G. 2001. Chronostratigraphy techniques in recent sediments. In: Last, W.M. and
Smol, J.P. (eds). Tracking Environmental Change Using Lake Sediments Volume 1: basin
analysis, coring, and chronological techniques. Kluwer Academic Publishers, Dordrecht,
The Netherlands, p. 171-203.
Ashley, G.M. 1975. Rhythmic sedimentation in glacial lake Hitchcock, MassachusettsConnecticut. In: Jopling, A.V. and McDonald, B.C. (eds). Glaciofluvial and
Glaciolacustrine sedimentation. Society o f Economic Palaeontologist and Mineralogists.
Special Publication, 23: 304-320.
Ballantyne, C.K. 2002. Paraglacial geomorphology. Quaternary Science Reviews, 21: 19352017.
Bartholomaus, T.C., Anderson, R.S., and Anderson, S.P. 2008. Response of glacier basal
motion to transient water storage. Nature Geoscience, 1: 33-37. doi:10.1038/ngeo.2007.52
Beedle, M.J., Menounos, B., Luckman, B.H. and Wheate, R., 2009. Annual push moraines as
climate proxy. Geophysical Research Letters, 36. L20501. DOL10.1029/2009GL039533.
Bennett, M.R., Huddart, D., Glasser, N.F., and Hambrey, M.J. 2000. Resedimentation of
debris on an ice-cored lateral moraine in the high-Arctic (kongsvegen, Svalbard).
Geomorphology, 35: 21-40.
Binda, G.G. 1984. Fluvioglacial sediment and hydrochemical dynamics, Peyto Glacier,
Alberta. MA thesis. Ottawa, ON, University of Ottawa. 139 pp.
Bolch, T., Menounos, B. and Wheate, R., 2010. Landsat-based inventory o f glaciers in
western Canada, 1985-2005. Remote Sensing o f Environment, 114: 127-137.
DOI: 10.1016/j .rse.2009.08.015.
Brune, G.M. 1953. Trap efficiency of reservoirs. Transactions o f the American Geophysical
Union, 34: 407-18.
Brunger, A.G., Nelson, J.G., and Ashwell, I.Y. 1967. Recession o f the Hector and Peyto
Glaciers: further studies in the Drummond glacier, Red Deer Valley area, Alberta.
Canadian Geographer, XI (1): 35-48.
Chikita, K.A. 1992. The role o f Sediment-Laden Underflows in Lake Sedimentation:
Glacier-Fed Peyto Lake, Canada. Journal o f the Faculty o f Science, Hokkaido University.
Series VII (Geophysics), 9: 211-224.

81

Chikita, K.A. 1993. Acomparative study on suspended sediment discharge initiated by snowor glacier-melting. IAHS Publ., 217: 257-267.
Chikita, K.A., Smith, N.D., Yonemitsu, N., and Perez-Arlucea, M. 1996. Dynamics of
sediment-laden underflows passing over a subaqueous sill: glacier-fed Peyto Lake,
Alberta, Canada. Sedimentology, 43: 865-875.
Church, M. and Slaymaker, O. 1989. Disequilibrium of Holocene sediment yield in glaciated
British Columbia. Nature, 337: 452-454.
Cogley, J.G., Hock, R., Rasmussen, L.A., Arendt, A.A., Bauder, A., Braithwaite, R.J.
Jansson, P., Kaser, G., Moller, M., Nicholson, L. and Zemp, M. 2011. Glossary o f Glacier
Mass Balance and Related Terms, IHP-VII Technical Documents in Hydrology No. 86,
LACS Contribution No. 2, UNESCO-IHP, Paris. 114 pp.
DeBeer, C.M. and Sharp, M.J. 2007. Recent changes in glacier area and volume within the
sourthem Canadian Cordillera. Annals o f Glaciology, 46: 215-221
De Geer, G. 1912. A geochronology o f the last 12000 years. Congr. Geol. Int. Stockholm
1910, C.R.p. 241-253.
Dedkov, A.P. and Moszherin, V.T. 1992. Erosion and sediment yield in mountain areas o f
the world. In: Walling, D.E., Davies, T.R. and Hasholt, B.(eds). Erosion, Debris Flows
and Environment in Mountain Regions. IAHS Publ. 209: 29-36.
Demuth, M.N. and Keller, R. 2006. An assessment of the mass balance o f Peyto Glacier
(1966-1995) and its relation to recent and past century climatic variability. In: Demuth,
M.N., Munro, D.S., and Young, G.J. (eds). Peyto Glacier: One Century o f Science. NHRI
Science Report No. 8. p. 82-134.
Demuth, M.N., and Pietroniro, A. 2003. The Impact o f Climate Change on the Glaciers o f the
Canadian Rocky Mountain Eastern Slopes and Implications for Water Resource-related
Adaptation in the Canadian prairies, “Phase 1”— Headwaters o f the North Saskatchewan
River Basin. Report to the Climate Change Action Fund—Prairie Adaptation Research
Collaborative, PARC. Project P55. Geological Survey o f Canada Open File 4322.
Demuth, M.N., Sekerka, J. and Bertollo, S. 2009. Glacier mass balance observations for
Peyto Glacier, Alberta, Canada. Spatially Referenced Data Set contribution to the
National Glacier-Climate Observing System, State and Evolution of Canada's Glaciers,
http://pathwavs.geosemantica.net/collections/documents folderview.aspx?coll=0456508e048a-48dl -965e-b22efe0e79b9. Geological Survey o f Canada.
Denton, G.H., 1975. Glaciers o f the Canadian Rocky Mountains. In W.O. Field (ed.).
Mountain Glaciers o f the Northern Hemisphere. Hanover, NH: United States Army Cold
Regions Research and Engineering Laboratory, p. 603-653.

82

Denton, G.H. and Karlen, W. 1973. Holocene climatic variations-the pattern and possible
cause. Quaternary Research, 3: 155-174.
Desloges, J.R. 1994. Varve deposition and the sediment yield record at three small lakes of
the southern Canadian cordillera. Artie and Alpine Research, 25: 130-140.
Evans, M. 1997. Holocene sediment yield and geomorphic sensitivity in alpine landscapes,
Cathedral Lakes, British Columbia. Phd Dissertation. UBC, Vancouver, 295 pp.
Field, W.O., and Heusser, C.J. 1954. Glacier and botanical studies in the Canadian Rockies,
1953. Canadian Alpine Journal, 37: 127-140.
Foster, I.D. L., Dearing, J.A., Grew, R. and Orend, K. 1990. The sedimentary data base: an
appraisal o f lake and reservoir sediment based studies o f sediment yield. In: Walling,
D.E., Yair, A. And Berkowicz, S. (eds). Erosion, Transport and Deposition Processes.
IAHS Publ., 189: 19-43.
Fox, A.J. and Nuttall, A.M. 1997. Photogrammetry as a research tool for glaciology.
Photogrammatric Record, 15: 725-737.
Gilbert, R. 2003. Varves. In: Middleton, G.V. (ed.) Encyclopaedia o f Sediments and
Sedimentary Rocks. Springer, Dordrecht, The Netherlands, p.764-765.
Glenday, P.J. 1991. Determination o f morphologic and volumetric change, Peyto Glacier,
Alberta, 1966-1989. MA Thesis. Waterloo, ON, Wilfrid Laurier University. 130 pp.
Glew, J.R., Smol., J.P. and Last, W.M. 2001. In: Last, W.M. and Smol, J.P. (eds). Tracking
Environmental Change Using Lake Sediments Volume 1: basin analysis, coring, and
chronological techniques. Kluwer Academic Publishers, Dordrecht, The Netherlands, p.
73-105.
Granshaw, F.D. and Fountain, A.G. 2006. Glacier change (1958-1998) in the North Cascades
National Park Complex, Washington, USA. Journal o f Glaciology, 52: 251-256.
Gumell, A. M., Hodson, A., Clark, M.J., Bogen, J., Hagen, J.O., and Tranter, M. 1994. Water
and sediment discharge from glacier basins; an arctic and alpine comparison. In: Olive,
L.J., Loughran, R.J., and Kesby, J.A. (eds) Variability in stream erosion and sediment
transport. IAHS Publication, 224: 325-334
Hallet, B., Hunter, L. and Bogen, J. 1996. Rates o f erosion and sediment evacuation by
glaciers: A review of field data and their implications. Global and Planetary Change, 12:
213-235.
H&nkanson, L. 1995. Models to predict organic content o f lake sediments. Ecological
Modelling, 82: 233-245.

83

Hankanson, L. and Jansson, M. 2002. Principles o f Lake Sedimentology. The Blackburn
Press, New Jersey, USA. 316 pp.
Harbor, J. and Warburton, J. 1993. Relative rates o f glacial and nonglaical erosion in alpine
environments. Arctic and Alpine Research, 25: 1-7.
Harbor, J., Bhaduri, B., Angelakis, L. and Snyder, J. 1997. Sediment basins, using modified
stormwater management basins and sediment basins to reduce water pollution from
construction sites in Ohio. Kent, OH: Department of Geology, Kent State University.
Heiri, O., Lotter, A.F., and Lemcke, G. 2001. Loss on ignition as a method for estimating
organic and carbonate content in sediments: reproducibility and comparability o f results.
Journal o f Paleolimnology, 25: 101-110.
Henoch, W.E.S. 1971. Estimate of Glaciers Secular (1948-1966) Volumetric Change and its
contribution to the discharge in the Upper North Saskatchewan River basin. Journal o f
Hydrology, 12: 145-160.
Heusser, C.J. 1956. Postglacial environments in the Canadian Rocky Mountains. Ecological
Monographs, 26: 253-302.
Hicks, D.M., McSaveney, M.J., and Chinn, T.J.H. 1990. Sedimentation in proglacial Ivory
Lake, Southern Alps, New Zealand. Arctic and Alpine Research, 22: 26-42.
Hickmann, M. and Reasoner, M.A. 1998. Late Quaternary diatom response to vegetation and
climate change in a subalpine lake in Banff National Park, Alberta. Journal o f
Paleolimnology, 20: 253-265.
Hobbs, W.O, Vinebrooke, R.D. and Wolfe, A.P. 2011. Biogeochemical responses o f two
alpine lakes to climate change and atmospheric deposition, Jasper and Banff National
Parks, Canadian Rocky Mountains. Canadian Journal o f Fisheries and Aquatic Sciences,
68: 1480-1494.
Hock, R. 2003. Temperature index melt modelling in mountain areas. Journal o f Hydrology,
282: 104-115.
Hodder, K.R., Desloges, J.R., and Gilbert, R. 2006. Pattern and timing o f sediment infill at
glacier-fed Mud Lake: implications for late glacial and Holocene environments in the
Monashee Mountain region of British Columbia, Canada. The Holocene, 16: 705-716.
Hodder, K.R., Gilbert, R., and Desloges, J.R., 2007. Glaciolacustrine varved sediment as an
alpine hydroclimatic proxy. Journal o f Paleolimnology, 38: 365-394. DOI
10.1007/sl 0933-006-9083-9

84

Holdsworth, G., Demuth, M.N., and Beck, T.M.H. 2006. Radar Measurements o f Ice
Thickness on Peyto Glacier, Alberta- Geophysical and Climatic Implications. In: Demuth,
M.N., Munro, D.S., and Young, G.J. (eds). Peyto Glacier: One Century o f Science. NHRI
Science Report No. 8. p. 59-79.
Hopkinson, C. and Demuth, M. N. 2006. Using airbone lidar to assess the influence o f
glacier downwasting on water resources in the Canadian Rockey Mountains. Canadian
Journal o f Remote Sensing, 32: 212-222.
Hopkinson, C. and Young, G.J. 1998. The effect o f glacier wastage on the flow o f the Bow
River at Banff, Alberta, 1951-1993. Hydrological Processes, 12: 1745-1762.
Hovius, N., 1998. Controls on sediment supply by large rivers. In: Relative Role ofEustasy,
Climate, and Tectonism in Continental Rocks. Society o f Sedimentary Geology. Special
Publ., 59: 3-16.
Jansson, P., Rosqvist, G. and Schneider, T. 2005. Glacier Fluctuations, Suspended sediment
flux and glacio-lacustrine sediments. Geografiska Annaler, 87A: 37-50.
Johnson, P.G. and Power, J.M. 1985. Flood and landslide events, Peyto Glacier terminus,
Alberta, Canada, 11-14 July 1983. Journal o f Glaciology, 31: 86-91.
Interprovincial Boundary Commission. 1917. Report o f the Commission Appointed to Delimit
the Boundary between the Provinces o f Alberta and British Columbia: From 1913-1916,
Part I. Ottawa, Ontario, Office o f the Surveyor General.
Kaab, A. and Vollmer, M. 2000. Surface geometry, thickness changes and flow fields on
creeping mountain permafrost: automatic extraction by digital image analysis. Permafrost
Periglacial Processes, 11: 315-326.
Kingman, H.S. 1938. Peyto Glacier, in Glacier Observations, 1936 and 1937 by
A.A.McCoubrey. Canadian Alpine Journal, 25: 113-116.
Kite, G.W. and Reid, I.A., 1977. Volumetric change o f the Athabasca Glacier over the last
100 years. Journal o f Hydrology, 32: 279-294.
Koppes, M. and Hallet, B. 2006. Erosion rates during rapid deglaciation in Icy Bay, Alaska.
Journal o f Geophysical Research 111: F02023, doi:10.1029/2005JF000349.
Lacourse, T. and Gajewski, K. 2000. Late Quaternary vegetation history o f Sulphur Lake,
southwestern Yukon territory, Canada. Arctic, 53: 27-35.
Lamoureux, S. 2001. Varve Chronology Techniques. In: Last, W.M. and Smol, J.P. (eds).
Tracking Environmental Change Using Lake Sediments Volume 1: basin analysis, coring,
and chronological techniques. Kluwer Academic Publishers, Dordrecht, The Netherlands,
p. 247-260.
85

Leonard, E.M. 1986. Varve studies at Hector Lake, Alberta, Canada, and the relationship
between glacial activity and sedimentation. Quaternary Research, 25: 199-214.
Leonard, E.M. 1997. The relationship between glacial activity and sediment production:
evidence from a 4450-year varve record o f neoglacial sedimentation in Hector Lake,
Alberta, Canada. Journal o f Paleolimnology, 17: 319-330.
Letreguilly, A. 1988. Relation between the mass balance o f western Canadian mountain
glaciers and meteorological data. Journal o f Glaciology, 34: 11-18.
Loso, M.G., Anderson, R.S. And Anderson, S.P. 2004. Post-Little Ice Age record o f Coarse
and fine clastic sedimentation in an Alaskan proglacial lake. Geology, 32: 1065-1068.
doi:10.1130/G20839.1
Luckman, B.H., 1996. Dendroglaciology at Peyto Glacier, Alberta. In Tree, Rings,
Environment, and Humanity, Dean, J.S., Meko, D.S., and Swetnam, T.W. (eds).
Radiocarbon, Tucson, Arizona, p.679-688.
Luckman, B.H., 1998. Landscape and climate change in the central Canadian Rockies during
the 20th century. The Canadian Geographer, 42: 319-336.
Luckman, B.H. 2006. The neoglacial history o f Peyto Glacier. In: Demuth, M.N., Munro,
D.S., and Young, G.J. (eds). Peyto Glacier: One Century o f Science. NHRI Science
Report No. 8. p. 25-58.
Luckman, B.H. and Osborn, G.D. 1979. Holocene glacier fluctuations in the middle
Canadian Rocky Mountains. Quaternary Research, 11: 52-77.
Luckman, B.H., Holdsworth, G. and Osborn, G. 1993. Neoglacial Glacier Fluctuations in the
Canadian Rockies. Quaternary Research, 39: 144-153.
McGlone, J. (ed). .2004. Manual o f Photogrammetry, 5th ed. American Society for
Photogrammetry and Remote Sensing, Bethesda, MD, USA. 1151 pp.
Meek. V. 1948. Glacier Observations in the Canadian Cordillera. Canadian Geographical
Journal, 37: 190-209.
Mekis, E. and Vincent, L.A. 2011. An overview o f the second generation adjusted daily
precipitation dataset for trend analysis in Canada. Atmosphere-Ocean, 49: 163-177.
Menounos, B. 1997. The water content of lake sediments and its relationship to other
physical parameters: an alpine case study. The Holocene, 7: 207-212.
Menounos, B. 2006. Anomalous early 20th century sedimentation in proglacial Green Lake,
British Columbia, Canada. Canadian Journal o f Earth Science, 43: 671-678.
86

Menounos, B., Clague, J. Gilbert, R., and Slaymaker, O., 2005. Environmental reconstruction
from a varve network in the southern Coast Mountains, British Columbia, Canada. The
Holocene, 15: 1163-1171.
Moore, R.D., Fleming, S.W., Menounos, B., Wheate, R., Fountain, A., Stahl, K., Holm, K.,
and Jakob, M. 2009. Glacier change in western North America: influences on hydrology,
geomorphic hazards and water quality. Hydrological Processes, 23: 42-61.
Munro, D.S. 2006. Linking the weather to glacier hydrology and mass balance at Peyto
Glacier. In: Demuth, M.N., Munro, D.S., and Young, G.J. (eds). Peyto Glacier: One
Century o f Science. NHRI Science Report No. 8. p. 134-176.
Nakawo, M., and Young, G.J. 1982. Estimate o f glacier ablation under a debris layer from
surface temperature and meteorological variables. Journal o f Glaciology, 28: 29-34.
Nolin, A.W., Phillippe, J., Jefferson, A., and Lewis, S.L. 2010. Present-day and future
contributions o f glacier runoff to summertime flows in a Pacific Northwest watershed:
Implications for water resources. Water Resources Research 46, W 12509,
doi: 10.1029/2009WR008968.
Ommanney, C.S.L. 1972. Glacier Surveys by District Personnel o f the water survey o f
Canada: 2.Peyto Glacier. Department o f environment, Inland waters Branch, Glacier
Inventory Note No. 7. 20 pp.
Ommanney, C.S.L. 2002. Glaciers o f the Canadian Rockies. In Glaciers o f North America Glaciers o f Canada - Glaciers o f the Canadian Rockies. U.S. Geological Survey
Professional Paper 1386-J-l.
Orwin, J. F. and Smart, C. C. 2004. Short-term spatial and temporal patterns o f suspended
sediment transfer in proglacial channels Small River Glacier, Canada. Hydrological
Processes, 18: 1521-1542.
Ostrem, G., and Arnold, K. 1970. Ice-cored moraines in southern British Columbia and
Alberta. G eogra/M a Annaler 52A: 120-128. doi: 10.2307/520605.
Owens, P. and Slaymaker, O. 1993. Lacustrine Sediment Budgets in the Coast Mountains of
British Columbia, Canada. In: McManus, J. and Duck, R.W. (eds). Geomorphology and
Sedimentology o f Lakes and Reservoirs. John Wiley and Sons Publishers, Toronto,
Canada, p. 105-123.
Rasmussen, L.A., and Conway, H. 2004. Climate and Glacier Variability in Western North
America. Journal o f Climate, 17: 1804 - 1815.
Reasoner, M.A. 1993. Equipment and procedure improvements for a lightweight,
inexpensive, percussion core sampling system. Journal o f Paleolimnology, 8: 273-281.
87

Restrepo, J.D., Kjerfve, B., Hermelin, M. and Restrepo, J.C. 2006. Factors controlling
sediment yield in a major South American drainage basin: the Magdalena River,
Columbia. Journal o f Hydrology, 316: 213-232.
Riihimaki, C.A., MacGregor, K.R., Anderson, R.S., Anderson, S.P., and Loso, M.G. 2005.
Sediment evacuation and glacial erosion rates at a small alpine glacier. Journal o f
Geophysical Research 110: F03003, doi:10.1029/2004JF000189.
Sander, M., Bengtsson, L., Holmquist, B., Wohlfarth, B. and Cato, I. 2002: The relationship
between annual varve thickness and maximum annual discharge (1909-1971). Journal o f
Hydrology, 263: 23-25.
Schiefer, E.K. 2004. Contemporary Sedimentation Patterns within Green Lake, Southern
Coast Moutains, British Columbia. Phd dissertation. UBC, Vancouver, 206 pp.
Schiefer, E. and Gilbert, R. 2007. Reconstructing long-term morphometric change in a
proglacial landscape using historical aerial photography and automated DEM Generation.
Geomorphology, 88: 167-178.
Schiefer, E., Menounos, B. and Wheate, R. 2007. Recent volume loss of British Columbia
glaciers, Canada. Geophysical Research Letters, L16503, loi:10.1029/2007GL030780.
Schuster, C.J. and Young, G.J. 2006. The derivation o f runoff from the Peyto Glacier
catchment. In: Demuth, M.N., Munro, D.S., and Young, G.J. (eds). Peyto Glacier: One
Century o f Science. NHRI Science Report No. 8. p. 226-254.
Shea, J.M., Moore, R.D., and Stahl, K. 2009. Derivation o f melt factors from glacier massbalance records in western Canada. Journal o f Glaciology 55: 123-130.
Slaymaker, O. 1977. Estimation o f sediment yield in temperate alpine environments. IAHS
Publ., 122: 109-117.
Smith, N.D. and Ashley, G.M. 1985. Proglacial Lacustrine Environment. In: G.M. Ashley,
J.Shaw and N.D. Smith (eds.) Glacial Sedimentary Environments. Society of Economic
Paleontologists and Mineralogists, Short Course 16, Tulsa, OK, p. 135-216.
Smith, D.G., and Jol, H.M. 1997. Radar structure o f a Gilbert-type delta, Peyto Lake, Banff
National Park, Canada. Sedimentary Geology, 113: 195-209.
Smith, N.D., Vendl, M.A. and Kennedy, S.K. 1982. Comparison o f sedimentation regimes in
four glacier-fed lakes of western Alberta. In: Davidson-Amott, R., Nickling, W. and
Fahey, B.D. (eds.) Research in Glacio, Glacio-Fluvial and Glacio-Lacustrine Systems,
Proceedings o f the 6th Guelph Symposium on Geomorphology, 1980. Geobooks,
Norwich, p. 203-238.

88

Tennant, C., and Menounos, B. 2013. Glacier change o f the Columbia Icefield, Canadian
Rocky Mountains, 1919-2009. Journal o f Glaciology, 59: 671-686.
Tennant, C. Menounos, B., Wheate, R., and Clague, J. 2012. Glacier Change in the Canadian
Rocky Mountains, 1919 to 2006. The Cryosphere, 6: 1541-1552.
Thorington, J.M. 1934. Peyto Glacier, in Glacial Observations in the Canadian Cordillera
1933-1934 by A.O. Wheeler. Canadian Alpine Journal, 22: 172-198.
Vanlooy, J.A. and Forster, R.R. 2011. Use of historical elevation data to calculate surfaceelevation and volume changes o f Ha-iltzuk Icefield, southwest British Columbia, Canada,
1970 to mid-1980s. Annals o f Glaciology, 52: 109-115.
VanLooy, J.A., Forster, R.R. and Ford, A. 2006. Accelerating thinning of Kenai Peninsula
glaciers, Alaska. Geophysical Research Letters 33, L21307, dio:10.1029/2006GL028060
Vanoni, V. A. 2006. Sedimentation Engineering. American Society o f Civil Engineers, New
York, USA, 418 pp.
Vendl, M.A. 1978. Sedimentation in Glacier-fed Peyto Lake, Alberta. MSc. thesis. Chicago,
Illinois, University o f Illinois at Chicago Circle. 71 pp.
Verstraeten, G. and Poesen, J. 2000. Estimating trap efficiency o f small reservoirs and ponds:
methods and implications for the assessment o f sediment yield. Progress in Physical
Geography, 24: 219-251.
Vincent, L.A., Wang, X.L., Milewska, E.J., Wan, H., Yang, F. and Swail, V. 2012. A Second
Generation of Homogenized Canadian Monthly Surface Air Temperature for Climate
Trend Analysis. Journal o f Geophysical Research 117, D 18110,
doi: 10.1029/2012JD017859.
Wallace, A.L. 1995. The volumetric change o f the Peyto Glacier, Alberta, Canada 18961966. MA Thesis. Waterloo, ON, Wilfred Laurier University, 140 pp.
Walling, D.E. 1983. The sediment delivery problem. Journal o f Hydrology, 65: 209-237.
Walling, D.E. 1994. Measuring sediment yield from river basin. In: Lai, R. (ed.), Soil
Erosion Research Methods. Soil and Water Conservation Society, Ankeny, IA, p. 39-80.
Wang, T., Hamann, A., Spittlehouse, D.L. and Murdock, T.Q. 2012. ClimateWNA - Highresolution spatial climate data for western North America. Journal o f Applied
Meteorology and Climatology, 5 1: 16-29.
Watson, E. and Luckman, B.H. 2004. Tree-ring-based mass-balance estimates for the past
300 years at Peyto Glacier, Alberta, Canada. Quaternary Research, 62: 9-18.

89

Wheeler, A.O. 1920. The application o f photography to the mapping of the Canadian Rocky
Mountains. Canadian Alpine Journal, 2: 76-87.
Wilcox, C.D., Dove, S.B., McDavid, W.D., and Greer, D.B. 2002. UTHSCSA Image Tool
Version 2.01: Freeware software available from the Department of Dental Diagnostic
Science at the University o f Texas Health Science Center at San Antonio:
http://compdent.uthscsa.edu/dig/itdesc.html or search for “Image Tool 3.0”
Young, G.J. 1981. The mass balance o f Peyto Glacier, Alberta. Canada, 1965 to 1978. Arctic
and Alpine Research, 13: 307-318.
Young, G.J. 1991. Hydrological interactions in the Mistaya basin, Alberta, Canada. Snow,
Hydrology and Forests in High Alpine Areas. Proceedings of the Vienna Symposium,
August 1991, LAHS Publications. No. 2005: p. 237-244.
Zemp, M., Thibert, E., Huss, M., Stumm, D., Rolstad Denby, C., Nuth, C., Nussbaumer,
S.U., Moholdt, G., Mercer, A., Mayer, C., Joerg, P.C., Jansson, P. Hynek, B., Fischer, A.,
Escher-Vetter, H., Elvehoy, H., and Andreassen, L.M. 2013. Uncertainties and re-analysis
of glacier mass balance measurements. The Cryosphere Discussions, 7: 789-839.
Zolitschka, B. 2007. Varved lake sediments, In: S.A. Elias (ed.), Encyclopaedia o f
Quaternary Science. Elsevier, Amsterdam, The Netherlands, p. 3105-3114.

90

Appendices
Appendix A: Ancillary core information

Table A T ^ IeasuredJ'w hite^Jnte^ lolatec^^

Core Sample
2010

2009
2008
2007 0.96
2006 1.09
2005 0.81
2004 1.44
2003 1.61 0.59
2002 1.55 0.35
2001 0.65 0.24
2000 0.38 0.17
1999 0.52 0.24
1998 2.49 1.79
1997 1.32 1.16
1996 0.42 0.48
1995 0.52 0.52
1994 0.82
1993 0.49
1992 0.68
1991 0.75
1990 0.75 0.58
1989 0.59 0.67
1988 0.75 0.74
1987 0.59 1.06
1986
0.88
1985 mmm o.gg
1984 | H | 1.48
1983 H f l 3.91
1982 B H 1.17

V©
to

0.69
0.09
0.41
0.24
0.75
0.37
0.28
0.15
0.22
0.37
0.45
0.93
0.41
0.45
0.41
0.24
0.47
0.34
0.26
0.39
0.34
0.24
1.30
1.54
0.65
0.67
1.21
0.78
0.78
0.74
0.82
0.95

VO

u>

5.29
4.29
1.83
2.07
2.05
4.18
7.67
4.46
2.78
1.29
6.00
3.70
1.60
2.09
3.50
2.44
2.36
3.24
1.72
2.17
1.25
1.90
1.27
1.09
0.86
6.59
4.82
0.88
1.54
0.68
1.21
3.16

1.93
2.41
1.57
2.33
1.95
1.69
2.26
1.63
1.63
1.44
2.36
3.22
1.99
1.54
2.75
2.34
2.90
2.31
1.69
3.03
3.05
2.33
3.17
3.58
0.78
3.15
2.64
1.28
2.45
2.36
3.77
3.02

0.95
0.87
1.31
1.12
1.76
1.48
1.62
1.17
1.45
0.91
3.21
4.02
3.70
2.35
3.20
3.03
2.35
2.31
1.69
2.90
3.20
2.35
2.87
3.13
1.33
2.43
2.21
0.82
1.77
1.75
3.01
2.72

0.79
1.32
1.45
1.62
1.96
0.74
5.44
1.85
2.64
1.36
3.30
2.73
3.21
2.42
2.89
2.82
2.84
2.22
1.65
2.91
2.95
1.67
4.12
1.84
1.90
1.99
2.19
0.31
0.52
0.84
3.05
2.53

2.02
1.91
1.89
1.64
1.76
1.34
1.21
0.57
1.13
0.53
2.29
2.83
1.55
1.32
1.79
1.38
1.98
1.80
0.83
1.89
1.70
0.87
2.10
2.06
0.81
1.80
1.80
0.61
1.48
1.63
2.33
2.16

1.70
1.64
1.34
0.73
0.67
0.76
1.23
1.41
1.30
1.32
1.72
1.63
0.92
0.97
2.02
1.74
2.09
1.08
0.62
0.88
3.07
1.97
1.64
1.80
0.60
1.38
1.24
0.67
0.87
0.76
1.40
1.80

0.70
0.40
0.42
0.42
0.35
0.64
0.35
0.35
0.22
0.22
0.64
0.30
0.29
0.48
0.43
0.48
0.21
0.46
0.57
0.66
1.21
0.42
0.55
0.62
0.37
0.37
0.42
0.18
0.22
0.20
0.42
0.42

1949
1948
1947
1946
1945
1944
1943
1942
1941
1940
1939
1938
1937
1936
1935
1934
1933
1932
1931
1930
1929
1928
1927
1926
1925
1924
1923
1922
1921
1920

VO

0.98
1.59
0.83
0.76
0.63
1.35
0.66
2.73
1.90
1.45
0.79
1.40
1.32
0.91
0.62
1.26
0.56
2.02
0.97
1.28
1.07
0.60
0.68
0.68
0.50
0.93
0.70
0.33
0.64
0.60

0.89
1.45
1.24
0.86
0.70
1.04
0.87
1.74
0.97
1.04
0.52
0.76
0.77
0.58
0.47
1.47
0.72
1.09
0.70
0.77
0.92
0.46
0.48
0.48
0.51
0.53
0.45
0.29
0.37
0.53

2.28
3.17
1.97
3.10
1.88
2.70
1.97
4.13
3.35
4.18
3.85
5.03
2.59
2.13
1.06
5.89
2.18
3.23
1.59
1.68
2.97
1.55
1.73
1.49
1.49
1.35
1.44
0.82
1.46
1.91

1.12 0.75
0.50 0.33
0.41 0.28
Mean 0.98 1.07 0.92

so

0.46 1.33
1.91
0.87 0.78
1.11
0.37 0.59
0.95
2.31 0.38 1.14 2.13

1.35
0.52
0.61
1.94 0.31

2 . 5 3 H H 1.66 1.77
0.49 m i 0.79 0.98
0.81 0.52
0.83 2.60 2.04 2.05

■

°-42mm.

0.90 1.52 1.00
0.71 1.06 0.56
0.61 0.66 0.62
1.79 1.62 1.26

0.71
0.29
0.26
0.55 0.11

Jfobl^^^MeaiHjulk^)h^sicayjro|j>ertiesfbi^

Core Sample
01
02
03
05
A
B
C
D
F
G
I
K
L
M
N
0
P
R
Mean
SD

Type* Length* Cracks* Dry density Organic content Wet density Water content
(gem-3)
(E.P)
(cm)
(8 cm0)
(%)
{%)
(%)
1.63
E
50.1
19
6.8
0.81
1.8
P
1.63
50.3
232
0.0
0.81
1.8
6.8
1.91
41.5
P
235
1.12
2.2
2.11
32.3
P
221
5.9
1.43
1.6
3.7
2.23
36.2
P
133
1.42
2.8
38.7
P
4.6
2.4
1.95
292
1.20
2.07
P
218
1.5
1.43
1.7
30.9
1.41
2.15
34.5
P
191
5.8
1.9
P
4.5
1.93
40.1
210
1.15
2.1
1.24
2.00
38.0
P
3.1
2.0
208
P
2.20
0.0
1.55
1.4
29.9
153
29.4
P
3.3
1.56
2.20
225
1.2
P
2.18
29.8
234
1.6
1.53
1.3
3.4
1.4
2.11
32.7
P
1.42
216
P
205
3.6
2.01
34.0
1.33
1.6
1.96
37.9
P
205
4.8
1.22
1.6
P
144
6.4
1.69
52.2
0.81
2.9
P
6.7
0.92
2.4
1.79
49.0
82
2.01
200
3.9
1.27
37.5
1.9
2.2
0.18
48
0.24
0.5
7.2

•Length o f wet sediment within each core
¥ Measured in the sharpest photos and refers to the amount o f crack space that developed along the entire length o f the sediment.
$ E = Ekman core, P = Percussion core

vo

On

Appendix B: Specific sediment yield results and sediment core correlation.

97

Table B .l

: Sediment

Year
Thiessen - 0 m
Thiessen -10 m
Thiessen - 20 m
Spline - 20 m
Year
Thiessen - 0 m
Thiessen -10 m
Thiessen - 20 m
Spline - 20 m
Year
Thiessen - 0 m
Thiessen -10 m
Thiessen - 20 m
Spline - 20 m
Year
Thiessen - 0 m
Thiessen -10 m
Thiessen - 20 m
Spline - 20 m
Year
Thiessen - 0 m
Thiessen -10 m
Thiessen - 20 m
Spline - 20 m
Year
Thiessen - 0 m
Thiessen -10 m
Thiessen - 20 m
Spline - 20 m

'O

00

2010 2009 2008 2007 2006 2005 2004 2003 2002
257
380 466 674 570 418 798 966 943
230 341
417 603 511
856
380 725 871
202
300 367 532 454 343 652 773 765
181
268 324 474 412 302 531 634 618
1993 1992 1991 1990 1989 1988 1987 1986 1985
377 548 387 526 419 542 593 502 394
339 489 348 470 372 490 534 450 355
298 431
307 415 326 439 474 397 315
248
357 261
383 308 404 417 382 285
1976 1975 1974 1973 1972 1971 1970 1969 1968
789 1004 620 593 423 1092 1226 667 606
709 895 554 532 380 982 1115 609 548
629 780 486 470 337 868 1003 549 489
529 685 431
424 313 742 874 497 433
1959 1958 1957 1956 1955 1954 1953 1952 1951
779 812 338 912 788 301
491
525 709
706 736 305 818 710 270 440 476 639
632 663 271
724 631
240 390 428 572
587 611
232 645 558 220 370 401
558
1942 1941 1940 1939 1938 1937 1936 1935 1934
1099 1259 1041
539 576 869 682 344 1068
989 1128 937 490 527 782 614 310 973
879 999 826 439 476 690 542 276 881
776 928 682 372 396 568 443 252 837
1925 1924 1923 1922 1921 1920 1919 1918 1917
394 369 364 222 308 507 522 302 247
352 330 327 200 277 453 467 271
221
311
291
290
178 244 398 412 240
193
307 264 257
205 368 359 206 173
161

2001 2000 1999 1998 1997 1996 1995
453 352 552 1314 745 320 382
408 315 494 1202 672 287 342
363 277 432 1080 597 252 300
326 239 365 843 508 220 252
1984 1983 1982 1981 1980 1979 1978
489 980 470 835 748 575 600
518 538
445 888 423 748 669
401
796 375 657 587 461 476
547 517 396 441
353 684 321
1967 1966 1965 1964 1963 1962 1961
878 679 819 736 457 743 815
664 410 671
736
793 613 741
707 547 663 592 363 600 659
596
643 511
614 530 335 561
1950 1949 1948 1947 1946 1945 1944
811
546 829 526 694 588 952
729 490 743 472 626 525 858
646 434 659 420 554 462 761
397 646 400 461
423 690
581
1933 1932 1931 1930 1929 1928 1927
427 919 499 431 1175 450 439
386 827 447 386 1044 402 393
344 732 394 340 906 352 347
316 639 350 316 808 326 325
SD
Mean
630 251
567 227
503 203
446 176

1994
525
478
430
360
1977
652
589
526
470
1960
542
488
435
404
1943
762
683
604
552
1926
327
295
265
243

T ableB ^^C oresam p lean d S S Y co rrelatio n table^

SSY
Std.
P
0
N
M
L
K
1
G
D
C
B
05
03
02
01
SSYspline
SSYT hiessen 20

0.94
0.26
0.73
0.75
0.69
0.75
0.69
0.60
0.32
0.77
0.87
0.51
0.75
0.70
0.68
0.96
0.99
1.00

SSY
0.93
0.25
0.71
0.73
0.69
0.75
0.78
0.54
0.31
0.76
0.87
0.54
0.72
0.70
0.68
0.95
1.00

Std. = Standardized varve thickness

01
0.95
-0.10
0.89
0.88
0.59
0.81
0.41
-0.24
NA
0.93
0.92
NA
0.85
0.65
0.65
1.00

02
0.76
0.30
0.51
0.63
0.30
0.50
0.57
0.19
0.48
0.75
0.75
0.32
0.47
0.70
1.00

03
0.79
0.18
0.65
0.65
0.39
0.59
0.59
0.16
0.24
0.69
0.70
0.32
0.59
1.00

05
B
0.79 0.58
0.09 0.17
0.69 0.54
0.70 0.52
0.37 0.16
0.73 0.56
0.45 0.53
0.24 -0.36
0.49 0.25
0.53 0.39
0.69 0.57
0.48 1.00
1.00

C
0.90
0.25
0.71
0.70
0.49
0.74
0.68
0.25
0.33
0.79
1.00

D
G
I
0.81 0.47 0.32
0.28 0.36 -0.01
0.63 0.23 0.18
0.62 0.53 0.06
0.41 -0.02 0.35
0.61 0.42 0.01
0.59 0.33 0.12
0.21 -0.26 1.00
0.20 1.00
1.00

K
0.72
0.23
0.48
0.59
0.40
0.62
1.00

L
0.84
0.22
0.69
0.79
0.41
1.00

M
0.55
0.13
0.46
0.28
1.00

0
p
Std.
N
0.85 0.82 0.36 1.00
0.29 0.20 1.00
0.67 1.00
1.00

Appendix C: Measured delta progradation and ancillary volume change results.

Figure C .l 2005 AT aerial photograph showing the previous measured extents o f the Peyto Lake delta.

100

40

o -40 ■
-60
-80

*

-!c

*

%

Period

Figure C.2 The rate o f volume change that Peyto Glacier has undergone for the period 1917 - 2005 is
categorized by data completeness o f > 50 %. Glacier changes highlighted in pink are within bounds o f the error
term.

100

200
>. 150

£
e

100

K

Period

Figure C.3 The rate o f volume change that Peyto Glacier has undergone for the period 1917 - 2005 is
categorized by data completeness o f > 20 %. Glacier changes highlighted in pink are within bounds o f the error
term.

101