Analysis of Glacial Streams on Mount Rainier: Temporal and Spatial Trends in Glacial Meltwater

Item

Title

Eng Analysis of Glacial Streams on Mount Rainier: Temporal and Spatial Trends in Glacial Meltwater

Date

2017

Creator

Eng Wilmes, Kristin

Subject

Eng Environmental Studies

extracted text

ANALYSIS OF GLACIAL STREAMS ON MOUNT RAINIER:
TEMPORAL AND SPATIAL TRENDS IN GLACIAL MELTWATER

by
Kristin Wilmes

A Thesis
Submitted in partial fulfillment
of the requirements for the degree
Master of Environmental Studies
The Evergreen State College
June 2017

© 2017 by Kristin Wilmes. All rights reserved.

This Thesis for the Master of Environmental Studies Degree
by
Kristin Wilmes

has been approved for
The Evergreen State College
by

________________________
{Erin Martin, Ph. D.}
Member of the Faculty

________________________
Date

ABSTRACT
Analysis of Glacial Streams on Mount Rainier:
Temporal and Spatial Trends in Glacial Meltwater

Kristin Wilmes

Organic carbon (OC) is stored in glacial environments. During the melt season, some of
this OC is exported to downstream aquatic ecosystems in the form of dissolved organic
carbon (DOC). The labile quality of OC found in glacial environments can stimulate
heterotrophic communities at the base of downstream food webs. The current temporal
trends of DOC export in glacial meltwater in Washington are unknown, even though
Washington is the second most glaciated state in the country. Here, data collected from
glacial meltwater on Mount Rainier displays the highest DOC concentrations occurring
early during spring snowmelt (1.5–4.9 mg L-1), with levels decreasing after mid-June and
staying consistently low for the rest of the melt season (0.5-1.3 mg L-1). Simple linear
regression analysis suggests that DOC concentrations are dependent upon snow depth
and the amount of snowmelt, being generally significantly positively correlated (p = <
0.05) with DOC concentrations. This indicates that snowmelt is likely a mobilizer of
DOC from glacial environments during the spring when snow depths are high. This also
likely indicates that snowmelt mobilizes any terrestrial DOC deposited on the snow
surface. High concentrations of sulphate, reflecting the composition of underlying rocks
and minerals, generally occur around the same time as high DOC concentrations suggest
that some of the meltwater in the glacial streams originates beneath the glacier. Given the
association with DOC concentrations, this may suggest DOC is also mobilized from the
drainage system beneath the glacier. As such, the spring appears to be a time of high
DOC export from glaciers. With projections of decreasing snow pack, earlier snowmelt
and disappearing glacial ice, sources of glacially-derived DOC will decrease and may
display peak concentrations at different times. These changes could result in changes to
the structure of glacial streams on Mount Rainier.

Table of Contents
1. Introduction………………………………………………………………………..1
2. Literature Review………………………………………………………………….5
2.1 Glacial Melt Season Drainage Dynamics……………………………………..6
2.2 Dissolved Organic Carbon in Glacial Environments………………………….9
2.3 Glaciers Local to the Pacific Northwest……………………………………..15
2.4 Dissolved Organic Carbon in the Food Web………………………………...18
2.5 Using Major Ions to Understand Chemical Weathering Processes in the
Subglacial Environment……………………………………………………...20
3. Methods…………………………………………………………………………..27
3.1 Research Design……………………………………………………………...27
3.2 Sampling Locations………………………………………………………….29
3.3 Geology of Mount Rainier…………………………………………………...32
3.4 Field Methodology…………………………………………………………...33
3.5 Lab Methodology…………………………………………………………….35
3.6 Statistical Analysis Methods…………………………………………………38
4. Results……………………………………………………………………………38
4.1 Dissolved Organic Carbon Analysis…………………………………………38
4.2 Other Parameters: pH, alkalinity and pCO2…………………………………47
4.3 Major Ion Analysis…………………………………………………………..50
5. Discussion………………………………………………………………………..62
5.1 Temporal and spatial patterns of DOC export……………………………….62
5.2 Chemical weathering patterns in the subglacial environment……………….67

iv

5.3 The effects of glacial area on DOC and major ion export: The implications of
continued glacial recession and climate change………………………………....73
6. Conclusion……………………………………………………………………….76

v

List of Figures
Figure 1: Figure displaying the evolving subglacial drainage system for an alpine
(temperate) glacier. A) Snow melt dominates early in the melt season. B) Snow
melt retreats up the ice and the exposed ice develops crevasses and moulins
rerouting meltwater to the base of the glacier. C) The continued retreat of the
snow line affects the development of the drainage system beneath the ice. It
transitions from a distributed system to a channelized system as the snow line
retreats. (as cited in Brown, 2002)………………………………………………...8
Figure 2: Glacier DOC concentrations. a–e, Average concentration of DOC in
cryoconites (a), surface (b), englacial (c) and basal ice (d) and in glacier melt
water (e) for the AIS, GIS and MGL. Red dots indicate the single values for each
entity (Hood et al., 2015)………………………………………………………...10

Figure 3: Reconstruction of glaciers showing the glacial recession on Mount Rainier from
1896 to 1994 (Hekkers & Thorneycroft, 2011)………………………………….16
Figure 4: Zoomed out view of the glaciers the feed the rivers where the sampling
locations are located. Emmons Glacier is about 4.3 mi2 in area and the Inter
Glacier situated next to Emmons Glacier is about 0.3 mi2 in area. The White River
comes from the glacial melt off Emmons Glacier and the Inter Fork River comes
from the melt waters of the Inter Glacier (Hekkers & Thorneycroft, 2011)……..30
Figure 5: Zoomed-in view of the field area and sites on Mount Rainier (adapted from
USGS, 2016)……………………………………………………………………………..31
Figure 6: DOC concentrations in the upper and lower White River sites over the glacial
melt season……………………………………………………………………….39
Figure 7: DOC concentrations at the upper and lower Inter Fork White River sites over
the glacial melt season…………………………………………………………...40
Figure 8: DOC concentrations in the two non-glacial creek sites over the glacial melt
season…………………………………………………………………………….40
Figure 9: Relationship between DOC (mg L-1) and snow depth (inches) at the upper Inter
Fork White River site…………………………………………………………….43
Figure 10: Relationship between DOC (mg L-1) and snow depth (inches) at the lower
Inter Fork White River site………………………………………………………43
Figure 11: Relationship between DOC (mg L-1) and total 7-day snowmelt (inches) at the
lower White River site…………………………………………………………...44
Figure 12: Relationship between DOC (mg L-1) and total 7-day snowmelt (inches) at the

upper I
v

Figure 13: Relationship between DOC (mg L-1) and snowmelt (inches) at the lower Inter
Fork White River site…………………………………………………………….45
Figure 14: Relationship between DOC (mg L-1) and velocity (m/sec) at the upper Inter
Fork White River site…………………………………………………………… 45
Figure 15: Relationship between DOC (mg L-1) and velocity (m/sec) at the upper White
River site…………………………………………………………………………46
Figure 16: Relationship between DOC (mg L-1) and daily precipitation (cm) at the upper
White River site………………………………………………………………….47
Figure 17: Alkalinity values (µeq L-1) for all glacial and non-glacial streams throughout
the field season…………………………………………………………………...50
Figure 18: Temporal trends of the major ions (µeq L-1) in the upper White River……...52
Figure 19: Temporal trends of the major ions (µeq L-1) in the lower White River……...52
Figure 20: Temporal trends of the major ions (µeq L-1) in the upper Inter Fork White
River……………………………………………………………………………...53
Figure 21: Temporal trends of the major ions (µeq L-1) in the lower Inter Fork White
River……………………………………………………………………………...53
Figure 22: Temporal trends of the major ions (µeq L-1) in the non-glacial creek 1. Major
ion concentrations were much higher initially in the non-glacial creek, so y-axis
has a different scale than all the other graphs……………………………………54
Figure 23: Temporal trends of the major ions (µeq L-1) in the non-glacial creek 2……..54
Figure 24: HCO3- concentrations (µeq L-1) in the glacial and non-glacial streams over the
field season……………………………………………………………………….55
Figure 25: Concentrations of SO42- (µeq L-1) for the glacial and non-glacial streams over
the field season…………………………………………………………………...56
Figure 26: Visual comparison of SO42- (µeq L-1) and DOC (mg L-1) at both White River
sites over the field season………………………………………………………..57
Figure 27: Visual comparison of SO42- (µeq L-1) and DOC (mg L-1) at both Inter Fork
River sites over the field season…………………………………………………57
Figure 28: C-ratio values for the White River, the Inter Fork White River and the nonglacial creeks over the field season………………………………………………59
Figure 29: The comparisons of HCO3_ (µeq L-1) and SO42- (µeq L-1) for the upper White
River site…………………………………………………………………………60
vi

Figure 30: The comparisons of HCO3_ (µeq L-1) and SO42- (µeq L-1) for the lower White
River site…………………………………………………………………………60
Figure 31: The comparisons of HCO3_ (µeq L-1) and SO42- (µeq L-1) for the upper Inter
Fork White River site…………………………………………………………….61
Figure 32: The comparisons of HCO3_ (µeq L-1) and SO42- (µeq L-1) for the lower Inter
Fork White River site…………………………………………………………….61

vii

List of Tables
Table 1: Data for snow depth, snowmelt, precipitation and temperature (NOAA,
2016)……………………………………………………………………………..41
Table 2: Correlation results of DOC concentration with multiple parameters. R-value
based on correlation analysis using Spearman’s ρ due to the data being not
normally distributed. Significant p values reported……………………………...42
Table 3: pCO2 values at the glacial and non-glacial stream sites over the field season…48
Table 4: pH values for the glacial and non-glacial sites over the field season…………..48
Table 5: Alkalinity values (µeq/L) for the glacial and non-glacial streams over the field
season…………………………………………………………………………….49
Table 6: Average major ion concentrations (µeq/L) for all glacial and non-glacial stream
sites………………………………………………………………………………51

viii

Acknowledgements
I would like to thank my thesis advisor, Dr. Erin Martin, for providing me with
encouragement and guidance through the thesis process. I would also like to thank the
Science Support Center staff for providing me with equipment to complete this project. A
special thanks to Jenna Nelson who was my Science Instruction Technician and helped
me with many different aspects of my laboratory work, Sina Hill who helped me with
laboratory equipment, and Kaile Adney who always made sure I had all the filters
necessary for this project! Another big thank you to the Evergreen State College for the
Student Foundation Activity Grant, as well as the Mazama’s Club for providing me with
a grant that helped me fund this project. One last huge thank you to my friends and
family who supported me through this long project, especially my husband Tal, who
helped me with all the field work!

ix

1. Introduction
Worldwide, glaciers contribute a surprisingly large source of carbon to
downstream environments, around 1.04 ± 0.18 teragrams (Tg) of dissolved organic
carbon (DOC) per year to freshwater and marine ecosystems (Hood, Battin, Fellman,
O’Neel & Spencer, 2015). Mountain glaciers contribute the largest amount of DOC to
these aquatic environments (Hood et al., 2015). Since glaciers are sensitive to
environmental changes, they are great indicators of climatic changes. Over the last
century, the mountain glaciers in Washington have consistently receded. With
Washington being the second most glaciated state in the country, these mountain glaciers
are an important source of freshwater and DOC to local streams and rivers (Hekkers &
Thorneycroft, 2011). Mount Rainier’s glaciers alone contribute meltwater to six major
rivers in the region (National Park Service [NPS], 2016).
Recent research has shown that the DOC from glacial environments is more labile
in nature than other forms of carbon (Hood et al., 2009; Hood et al., 2015). The labile
nature found in some of this glacially derived DOC is a product of microbial
communities living on the glacial surface and beneath the glacier (Anesio, Hodson, Fritz,
Psenner & Sattler, 2009; Bhatia et al., 2013; Sharp et al., 1999; Skidmore, Fought &
Sharp, 2000; Wadham et al., 2004; Wadham et al., 2010). Microbial communities found
in holes on the glacial surface perform photosynthesis resulting in the accumulation of
organic carbon (OC) (Anesio et al., 2010; Anesio et al., 2009), while OC found beneath
the glacial ice is from soils and vegetation that were overridden by the ice during the last
ice age (Bhatia et al., 2013; Wadham et al., 2010). Microbial communities have been
discovered living beneath glaciers that can mobilize and metabolize this source of ancient
1

OC. This can not only degrade the OC and make it more labile, but it can produce CO2
beneath the ice that can drive chemical weathering processes (Barker, Sharp, Fitzsimons
& Turner, 2006; Lawson et al., 2104). The more labile or bioavailable nature of glacially
derived DOC stimulates heterotrophic communities in downstream ecosystems, which
influences organisms in the rest of the aquatic food web (Hood & Scott, 2008; Hood et
al., 2009; Hood et al., 2015; Fellman et al., 2015).
Warmer atmospheric temperatures caused by climate change are resulting in a
decreasing snowpack during the winter months and earlier snowmelt in the spring
(International Panel on Climate Change [IPCC], 2014). With projections of glacial
recession continuing, the current temporal trends of DOC export will shift in time and
quantity. This shift in timing and quantity of glacially derived DOC will likely affect the
structure of the local aquatic food webs that rely on this flux of labile OC (Hood &Scott,
2008; Hood et al., 2009; Hood et al., 2015) and freshwater during the dry summer months
when other mountainous ephemeral streams dry up and no longer provide any DOC to
the larger river systems.
In Washington, there are numerous glaciers in the Cascade and Olympic
Mountains. These glaciers feed many important rivers in the Pacific Northwest, but
similar to other regions, almost all the glaciers have been receding for decades (NPS,
2016). These rivers are home to many important aquatic species, like salmon, which the
Washington Department of Fish and Wildlife [WDFW] (2017) have listed as a federally
threatened species. A study completed by Fellman et al. (2015) tracked glacially derived
OC through different trophic levels in an Alaskan stream, from macroinvertebrates to
fish. A shift in timing in the OC export through glacial meltwater would potentially have
2

effects on the local aquatic food web, including native fish species in Washington’s rivers
and streams.
As glaciers recede the amount of water being exported from the glacial
environment changes. Studies have shown that initially there may be an increase in
meltwater, but eventually glacial recession leads to a decrease in meltwater export as the
glacier gets smaller (Fountain & Tangborne, 1985). This change in meltwater may affect
the amount of DOC exported from glacial environments. Another consideration is that as
glaciers recede terrestrial succession follows the retreating ice, so in the coming decades
there will likely be a shift from glacially-derived DOC to terrestrial DOC sources
(Fellman et al., 2014; Hood et al., 2008). To address the question of how glacial recession
may affect DOC export from glacial environments, this study examined the DOC export
from the meltwater of one of the largest glaciers on Mount Rainer, Emmons Glacier, and
one of the smallest glaciers, Inter Glacier. By examining data from these two differing
glaciers it may be possible to understand changes that could occur as glaciers in the
region continue to recede.
Presently, there are multiple studies that have examined the temporal trends of
OC export from larger ice sheets, like the Greenland Ice Sheet and the Antarctic Ice
Sheet, and in mountainous glaciers in Alaska, the European Alps and the Himalayan
Mountain range (Barker et al., 2006; Bhatia et al., 2013; Hood et al., 2008; Hood et al.,
2009; Lawson et al., 2014; Spencer et al., 2014; Wadham et al., 2010).). There have been
no such studies completed on the DOC export from glacial meltwater in the Pacific
Northwest region. Data from these other glaciated regions cannot be extrapolated to the
Pacific Northwest because they do not have the same climate and lithology which can
3

result in differing DOC trends. Without any data on the current temporal trends of DOC
export to proglacial streams, there will be no way to understand how these trends will
change with continued glacial recession or if it will affect local aquatic ecosystems.
The purpose of this study is to gather data on the current temporal and spatial
trends in DOC and major ion export in glacial meltwater from Emmons Glacier and Inter
Glacier on Mount Rainier. Major ion export is important because it can elucidate flow
paths by which meltwater is leaving the glacier, and can therefore clarify pathways of
carbon export within the glacier. This research will provide local natural resource
managers and policy makers with information to use to help protect our natural resources
as climate change continues to affect the environment and local ecosystems. This study
examines how DOC and major ion concentrations in streams draining glaciers change
over the summer meltwater period. To my knowledge this study provides a first estimate
of the potential delivery of carbon to downstream environments that originate from
glacial environments in the state of Washington. The study will also provide an
understanding of how the glacial area at the streams’ headwaters influences the DOC and
major ion concentrations. By looking at streams with different glacial sizes at the
headwaters, insight into how DOC and major ions concentrations may change as glaciers
recede can be concluded. To answer these questions, this study examined DOC and major
ion concentrations from two glacial streams on Mount Rainier, as well as two non-glacial
streams for comparison

4

2. Literature Review
The purpose of this literature review is to provide background information as to
why studying the export of dissolved organic carbon (DOC) and major ions from bulk
glacial meltwater on Mount Rainier is crucial for local and global comprehension of how
DOC influences food web structures in proglacial aquatic ecosystems. Previous studies
have examined the export of glacial organic carbon (OC) from mountain glaciers in
Alaska, India, Norway, Sweden and the European Alps, as well as on the Antarctic Ice
Sheet (AIS) and the Greenland Ice Sheet (GIS) (Barker et al., 2006; Bhatia et al., 2013;
Hood et al., 2008; Hood et al., 2009; Lawson et al., 2014; Spencer et al., 2014; Wadham
et al., 2010). These studies all occur in regions displaying a variety of differing climates,
latitudes and lithologies. The varying results from these studies presents evidence that
DOC and major ion export from glacial environments can differ depending on the
geology of the region, the climate of the region, the time in the glacial melt season, and
the size of glacial coverage. Studies have shown how this glacially-derived OC is
integrated through food webs in aquatic ecosystems (Fellman et al., 2015) and how it can
affect these ecosystems (Slemmons, Saros and Simon, 2013). To my knowledge there
have been no studies examining the export of DOC from bulk glacial meltwater in the
Pacific Northwest. The research highlighted in this literature review gives an overview of
the methods used to examine DOC export in glacial meltwater and the methods used to
gain insight into the unique properties of this glacially derived OC. This literature review
also covers the methods used to determine chemical weathering patterns and drainage
system dynamics beneath the ice by utilizing the major ion concentrations.

5

This research project will add a new data set from a region previously unstudied
to the larger pool of data on carbon export from glacial meltwater. The data gathered can
be beneficial on a local level, giving natural resource managers an understanding of the
current temporal and spatial trends of glacial DOC and major ion export. With this
knowledge, hopefully they can properly protect our local natural resources as these
export trends will likely shift with the projected changing snowpack, precipitation,
temperature and ultimately the receding glaciers.

2.1 Glacial Melt Season Drainage Dynamics
Glaciers have different temperature regimes allocating them into three categories:
1) temperate (warm-based) glaciers, 2) subpolar (polythermal) glaciers and 3) polar
(cold-based glaciers) (as cited in Brown, 2002, p.857-858). The glaciers found on Mount
Rainier are temperate (warm-based) glaciers, so this section focuses on the melt season
drainage dynamics of this specific type of glacier. The significance of warm-based
glaciers is they have water between the glacial ice and the bedrock because the glacial ice
is at the pressure melting point (Bennet & Glasser, 1996) compared with cold-based
glaciers that are frozen to the bedrock because the glacial ice is below the melting point
(Tranter et al., 1996).
As the melt season begins in Washington, typically early to late May depending
on the year, snow melt dominates stream runoff in the alpine regions. At this stage in the
melt season, most the water in the streams beyond the glacier comes from snow melt
instead of glacial ice melt (see figure 1, part a) (Collins, 1979; Fountain & Walder, 1998).
Before any ice is exposed, the snowmelt percolates through the snow and firn (granulated

6

snow that has not yet been compacted into glacial ice) until it reaches the ice below and
where it is either stored in the firn until later in the melt season or it is exported down the
ice to proglacial (beyond the ice) streams (Fountain & Walder, 1998) During this early
stage of the melt season the drainage system beneath alpine glaciers in considered to be a
distributed system, with low passages and cavities resulting in slower flowing water
(Brown, 2002; Fountain & Walder, 1998).
As the snowline retreats up the glacial ice, snow and ice melt initially runoff the
top of the glacial ice to proglacial streams. As more ice is exposed, moulins and crevasses
begin to form in the glacier. Once these moulins and crevasses are established the melt
water travels to the base of the glacier through these routes (see figure 1, part b) (Brown,
2002; Fountain & Walder, 1998). This influx of water from the surface of the glacier to
the base of the glacier is generally a quick process which causes the drainage system to
evolve into a more channelized system compromised of larger channels and faster
flowing water (Brown, 2002; Fountain & Walder, 1998).
During the transition from the distributed system to the channelized system the
drainage system is often closed to the atmosphere since the channels fill up quickly with
water and become pressurized before larger channels are cut into the sediment and ice
under the glacier (Fountain & Walder, 1998). The channelized drainage system continues
to develop beneath the ice following the retreating snow line. As more glacial ice is
exposed more crevasses and moulins form in the glacier. This allows more meltwater to
travel to the drainage system beneath the ice (see figure 1, part c) (Brown, 2002; Fountain
& Walder, 1998).

7

Figure 1: Figure displaying the evolving subglacial drainage system for an alpine
(temperate) glacier. A) Snow melt dominates early in the melt season. B) Snow melt
retreats up the ice and the exposed ice develops crevasses and moulins rerouting
meltwater to the base of the glacier. C) The continued retreat of the snow line affects the
development of the drainage system beneath the ice. It transitions from a distributed
system to a channelized system as the snow line retreats. (as cited in Brown, 2002).

8

2.2 Dissolved Organic Carbon in Glacial Environments
Mountain glaciers and ice sheets play a significant role in the global carbon cycle.
These frozen settings contain organic carbon (OC) in the supraglacial (the glacial
surface), englacial (situated in the glacier) and subglacial (beneath the glacial ice)
environments. The export of this OC to proglacial (anything immediately beyond the
terminus of the mountain glacier) aquatic ecosystems occurs through snowmelt and
glacial meltwater over the summer months (Lawson et al., 2014; Spencer et al., 2014).
Mountain glaciers and ice sheets also export major ions to proglacial streams as chemical
weathering occurs in the subglacial drainage system resulting in the ions accumulating in
the meltwater (Brown, 2002).
Glaciers and ice sheets globally store around 70% of the Earth’s freshwater. With
climate change causing an increase in atmospheric temperatures, glaciers are receding at
increased rates and supplying more freshwater to downstream environments (Hood et al.,
2015, Slemmons et al., 2013). The recession of mountain glaciers is occurring at faster
rates than the recession of larger ice sheets, so they provide important insight into how
the timing and magnitude of OC export to proglacial stream environments will occur
(Hood et al., 2015).
A study done by Hood et al. (2015) calculated the amount of DOC currently
stored and then released from mountain glaciers and ice sheets across the world (see
figure 2). The study used data on the amount of DOC concentrated in englacial ice core
samples from multiple mountain glacier (MGL) sites, as well as the Greenland Ice Sheet
(GIS) and the Antarctic Ice Sheet (AIS) to calculate the global mass of DOC stored in
glacial environments (Hood et al., 2015). The research team found that the global

9

estimate of dissolved organic carbon (DOC) in glaciers is 4.48 ± 2.79 petagrams of
carbon (PgC) (Hood et al., 2015). This is in comparison to the northern arctic permafrost
regions that store around 1672 PgC in the frozen soil (Tarnocai et al., 2009). Hood et al.
(2015) went on to calculate the amount of DOC stored in different glacial regions, with
93% (4.19 ± 2.78 Pg C) of the glacial DOC stored in the AIS, 5% (.22 ± .06 Pg C) stored
in the GIS and 2% (.07 ± .01 Pg C) stored in mountain glaciers. As glacier and ice sheets
continue to melt this OC will be released to proglacial stream and marine environments.

Figure 2: Glacier DOC concentrations. a–e, Average concentration of DOC in
cryoconites (a), surface (b), englacial (c) and basal ice (d) and in glacier melt water (e)
for the AIS, GIS and MGL. Red dots indicate the single values for each entity (Hood et
al., 2015).

Although the AIS and GIS store significantly more DOC than mountain glaciers,
the mountain glaciers are receding at faster rates than the large ice sheets, which results in
them exporting a higher amount of DOC to proglacial ecosystems (see figure 2). Hood et
al., (2015) estimated that globally the annual export of DOC from glacial meltwater is

10

around 1.04 ± 0.18 teragrams of carbon (TgC), which is less than the carbon that evades
rivers, streams, lakes and reservoirs every year but it is still significant. According to
Raymond et al., (2013) around 1.8 ± .25 PgC evades from streams and rivers every year,
while 0.32 ± .26 to .52 PgC comes from lakes and reservoirs annually. When broken up
into glacial region, MGL export 56% (0.58 ± 0.07 TgC) of the DOC to proglacial aquatic
ecosystems annually, compared to the GIS that exports 0.22 ± 0.04 TgC yr–1 and the AIS
that exports 0.24 ± 0.16 TgC yr–1 (Hood et al., 2015). Mountain glaciers across the world
come from different regions with different geology and climates, so they may have
varying concentrations of DOC exported through glacial meltwater.
Currently, mountain glaciers experience a higher glacial mass loss per year than
larger ice sheets due to glacial recession occurring at quicker rates in the mountain
glaciers (Hood et al., 2015). One study found that globally, around 0.14 TgC of the DOC
exported every year to proglacial aquatic ecosystems was the result of glacial mass loss
(Hood et al., 2015). Since the glacial mass loss occurring in mountain glaciers does not
represent a steady flux over time, this means that as glaciers continue to recede, there will
be a change in the amount of DOC exported to proglacial environments (Hood et al.,
2015). The timing of glacial melt will also change as climate change continues to cause
changes in weather patterns, snowmelt, and ultimately glacial recession. The amount of
snowpack is likely to diminish with rising atmospheric temperatures, leading to earlier
snowmelt or no snow, and a change in glacial melt discharge (Department of Ecology
[DOE], 2007; Grandshaw & Fountain, 2006; NPS, 2016). This highlights the importance
in understanding the role mountain glaciers play in the global carbon cycle, since they are
currently contributing more DOC to aquatic environments than the larger ice sheets.

11

Over the last one hundred years, the air temperature in alpine regions has
increased by an average of 0.3˚C per decade resulting in continual glacial recession and
making mountainous glaciers an important signifier of climate change (Slemmons et al.,
2013). Studies have shown that glacial coverage is a factor influencing the timing,
magnitude and source of OC and major ion export, so comparing mountain glaciers of
different sizes may provide critical insight into how this export will change as glaciers
continue to recede (Fellman, Hood, Spencer, Stubbins & Raymond, 2014; Hood and
Scott, 2008). Results from studies comparing watersheds with differing glacial coverage
showed that decreasing glacial coverage caused a decrease in the glacially derived DOC
concentrations but an increase in terrestrial DOC concentrations (Fellman et al., 2014;
Hood & Scott, 2008). This is because vegetative succession progresses as the ice retreats
in alpine regions.
In non-glacial streams and rivers, OC typically comes from terrestrial sources,
like plant or soil material, but many glacial streams begin above the tree line so the
stream is receiving less terrestrial sources of OC (Fellman et al., 2015; Hood et al., 2009).
Instead, these glacial stream ecosystems rely on the delivery of OC from the snow and
glacial meltwater. Recent studies have shown that the OC from glacial environments is
typically more labile, or bioavailable, to aquatic ecosystems in comparison to OC from
terrestrial sources (Barker et al., 2006; Bhatia et al., 2013; Hood et al., 2009; Lawson et
al., 2014; Spencer et al., 2014). A study done by Hood et al., (2009) found that
watersheds with large glacial coverage displayed an increase in the bioavailability of the
carbon. This means that the OC is readily available to heterotrophic communities that
make up the base of the aquatic food web. Export of glacial DOC can stimulate the

12

heterotrophic communities, which then can fuel production in throughout the rest of the
food web (Bhatia et al., 2013; Fellman et al., 2015; Hood et al, 2009; Hood et al., 2015).
OC on the surface of these glaciers and ice sheets comes from cryoconite holes,
which harbor microbial communities (Anesio et al., 2009; Anesio et al., 2010;
Porazinska, Fountain, Nylen, Tranter & Virginia, 2004). These cryoconite holes form due
to the deposition of organic matter onto the glacial surface. The sun warms this organic
matter, which is darker than the ice, causing it to melt into the ice. The organic matter
deposited on the ice surface comes from terrestrial sources, such as dust and soil
deposition, as well as from aerosol particles from anthropogenic combustion products
(Anesio et al., 2009; Bhatia et al., 2013; Hood et al., 2009; Hood et al., 2015). The
cryoconite holes fill up with water as the melt season begins, which can create an
autotrophic system that is home to an extremely diverse microbial community (Anesio et
al., 2009). These microbial communities consist of different assemblages of viruses,
algae, and bacteria (Anesio et al., 2009; Anesio et al., 2010; Porazinska et al., 2004;
Stibal & Tranter, 2007). The microbial communities photosynthesize and grow within the
cryoconite holes. (Anesio et al., 2009; Stibal & Tranter, 2007). This in situ primary
production results in organic carbon accumulating in these cryoconite holes faster than
decomposition from respiration can occur (Anesio et al., 2009; Stibal & Tranter, 2007).
Organic carbon can be flushed from these cryoconite holes early in the melt
season when snowmelt is the dominate source of the meltwater. A study by Anesio et al.
(2009) used ∆14C signatures to track the DOC fixed in cryoconite holes. The research
team found that around 10% of this DOC from these holes is exported to proglacial
environments early in the season as the snowmelt flushes out some of the carbon from

13

these autotrophic environments (Anesio et al., 2009). As the melt season continues and
the melt regime shifts away from snowmelt, the DOC from the cryoconite holes or from
atmospheric deposition can be delivered to the subglacial environment as ice melts on the
glacial surface. This ice melt can release frozen DOC from the surface ice and can flush
DOC from the cryoconite holes as the water travels across the ice and then to the
subglacial environment through moulins and crevasses in the ice (Anesio et al., 2009;
Bhatia et al., 2013; Brown, 2002; Fountain & Walder, 1998; Lawson et al., 2014).
The delivery of DOC to the subglacial environment from the surface through
cracks and crevasses in the ice is not the only source of OC to the subglacial
environment. During the last ice age, soils and vegetation were overridden by glaciers
trapping it beneath the ice (Sharp et al., 1999; Skidmore et al., 2000). Recent studies have
found evidence that there can be microbial communities in these rock-water-ice
interfaces utilizing OC sources and affecting chemical weathering processes occurring
beneath the ice and (Bhatia et al., 2013; Hood et al., 2009; Skidmore et al., 2000; Sharp et
al., 1999; Wadham et al., 2010). Researchers have discovered that these microbial
communities can fix inorganic carbon through the process of chemolithoautotrophy as
well as metabolize the organic carbon found in the subglacial environment (Lawson et
al., 2014; Skidmore et al., 2000). These microbial processes produce a more labile form
of OC that is then exported in the glacial meltwater (Lawson et al., 2014). The carbon
products of chemolithoautotrophy and primary production represent a younger source of
carbon, while the carbon from soils and vegetation beneath the ice represent an older
source of carbon (Lawson et al., 2014; Skidmore et al., 2000). By looking at the ∆14C
signatures of the DOC, the age of the carbon sources in glacial environments can be

14

determined. The export of this bioavailable OC occurs as meltwater drains to proglacial
aquatic environments from the subglacial environment.

2.3 Glaciers Local to the Pacific Northwest
The Pacific Northwest has mountain glaciers throughout the Cascade Mountain
range and in the Olympic Mountains, with ~ 449 km2 of glaciers and perennial snow and
ice in Washington (Hekkers & Thorneycroft, 2011). This makes Washington the second
most glaciated state in the country and these glaciers feed many of the local streams and
rivers (Hekkers & Thorneycroft, 2011; NPS, 2016). Mount Rainier has 27 glaciers, and it
is the most ice-covered mountain outside of Alaska (NPS, 2016; U.S Geological Survey
[USGS], 2014). The glaciers on Mount Rainier feed the headwaters for six major rivers in
the region suggesting the potential importance of DOC export from glacial meltwater to
food webs downstream (NPS, 2016).
On Mount Rainier, the glaciers have been receding for at least the last 70 years,
with ~ 17% mass ice loss on the northern side of the mountain and ~ 26% mass ice loss
on the southern side of the mountain (see figure 3) (Hekkers & Thorneycroft, 2011;
Nylen, 2001). This mass ice loss from glaciers over large time scales initially causes an
increase in meltwater discharge to proglacial environments, but as glaciers continue to
get smaller over time the discharge eventually decreases. This changing discharge due to
glacial recession demonstrates how shrinking glaciers will cause a change in the timing
and magnitude of this meltwater. A study done by Granshaw & Fountain (2006) on
glacial recession in the Northern Cascade Mountains found that glacial mass loss in the
months of August through September resulted in a 0.1% to 6.0% increase in stream flow.

15

Figure 3: Reconstruction of glaciers showing the glacial recession on Mount Rainier from
1896 to 1994 (Hekkers & Thorneycroft, 2011).

According to the Department of Ecology [DOE] (2007), snowpack in about 73%
of the Northern Cascades has decreased and the snowmelt is occurring at earlier times of
the year, resulting in stream flows peaking at earlier times in the spring. The reduction of
this snowpack is contributing to the increase in stream flow later in the summer months
because it is causing the glacial ice surface to be exposed to solar radiation for longer

16

time-periods. The increase in the late summer months does not result in a huge discharge
peak like the spring snowmelt peak, but it still represents an important change since it
affects DOC export. In the summer of 2015, Mount Rainier’s glacial melt was higher
than ever previously observed in the early summer months (NPS, 2016).
Increases in meltwater will ultimately decrease and eventually disappear if trends
in mountain glacial recession continue. As of 2007, at least 53 of the glaciers in the North
Cascades have disappeared (DOE, 2007) and 82 glaciers have disappeared from the
Olympic Mountains since 1980 (Riedel & Larrabee, 2015). As these alpine environments
lose glacial coverage, there will be a shift in the DOC introduced to the aquatic
ecosystems from glacial sources to terrestrial sources. This occurs as the bare ground left
after the glaciers retreat is succeeded by vegetation. Since the DOC exported from glacial
environments has been shown to be more labile in nature than DOC from terrestrial
sources, this shift in OC sources could have an impact on primary production at the base
of the food web (Hood et al., 2009; Hood et al., 2015; Hood & Scott, 2008; Slemmons et
al., 2013). This change in primary production could ultimately cause a shift in food webs
downstream (Bhatia et al., 2014; Fellman et al., 2015; Hood et al., 2009; Hood et al.,
2015). This highlights the importance in gathering a data set for the current spatial and
temporal trends in DOC export from meltwater on Mount Rainier. The shift in DOC
sources may lead to shifts in our local aquatic food webs, which boast important local
species like salmon. Examining data from glaciers with different glacial area may give an
insight into the changes that could occur with glacial recession.

17

2.4 Dissolved Organic Carbon in the Food Web
Hood et al. (2009) conducted a study to understand the labile nature of DOC from
glacial meltwater and what this indicates for proglacial aquatic ecosystems. This is
possible because glacially derived DOC has a unique isotopic signature in comparison to
DOC from non-glacial streams, where it is δ13C enriched and ∆14C depleted (Fellman et
al. 2015; Hood et al., 2009). δ13C is useful for understanding the source of carbon,
because different carbon sources can have differing δ13C values. ∆14C is used to
understand the age of the carbon. New evidence has revealed that microbial communities
in the subglacial environment can survive on the ancient carbon stored beneath the ice
and the alteration of the DOC gives the carbon the enriched δ13C signature (Hood et al.,
2009; Skidmore et al., 2000). This enriched δ13C signature signifies the carbon is from
autocthonous material found in aquatic ecosystems, meaning the carbon is produced from
within the glacial environment (microbial processes) rather than being input from a
terrestrial source (Hood et al., 2009). In the study by Hood et al. (2009) the researchers
used δ13C and Δ14C isotopes to calculate the percentage of bioavailable carbon in
watersheds with differing glacial coverage. The study found that in the Gulf of Alaska,
the watersheds with a higher glacial coverage had more labile matter in the meltwater and
the glacially-derived C was responsible for the older ∆14C age (Hood et al., 2009). This
once again highlights the importance of comparing DOC export data between watershed
with varying glacial coverage.
A study by Fellman et al. (2015) went further in understanding the role of this
glacial DOC in food webs. Using δ13C, Δ14C and δ15N (stable nitrogen isotope used to
track food sources through food webs) values, the research team tracked the isotopic

18

signatures of sources (biofilm, particulate organic matter, leaf litter) and consumers
(macroinvertebrates, Fish: Coho Salmon and Dolly Vardon). The study found that in a
glacial stream in Juneau, Alaska ~ 51% of the carbon in consumers diets was from glacial
OC sources in the upper part of the river, while downstream ~ 34% of the carbon in
consumers diets was from glacial OC sources (Fellman et al., 2015). These glacial OC
sources for consumers are in the form of biofilm that integrates this labile glacial OC into
its structure giving it a 13C enriched and 14C depleted signature (Fellman et al., 2015).
The results showing that different fish species in Alaska rely on food sources that
assimilate this glacial DOC into their structure demonstrates the importance of OC export
from glacial meltwater to proglacial aquatic environments. In Washington, salmon are an
important species for multiple reasons and for different groups of people. There are
currently 624 populations of salmon in Washington’s streams and rivers (WDFW, 2016)
These salmon not only rely on the structure of the current food web which may be
influenced by glacial DOC export in summer months, but they also rely on consistent
cooler stream temperatures, high stream flows and high dissolved oxygen needed for
spawning that the glacial melt provides (Slemmons et al., 2013). Glacial recession will
affect all these properties, as well as shifting the timing and magnitude of glacial melt in
coming years which will have negative impacts on salmon species. Salmon populations
in Washington have already been declining over the past century due to loss or
destruction of habitat, land use, dams, overfishing and hatcheries, yet salmon are still a
very important part of Washington’s past and present. According to Washington’s
Department of Fish and Wildlife (2008), “the non-treaty commercial fishery in
Washington waters also contributes an estimated $38 million in net economic values (net

19

income or profits) and recreational fisheries generate an estimated $424 million in net
economic values.” Salmon are important to local Native American communities who
have historically relied on these fish for sustainability and spiritual reasons.
A change in river food web structures could be damaging to multiple fish
populations who are already struggling with population decline due to these other issues.
By gathering data to form a better understanding of the local DOC export to these
proglacial ecosystems, wildlife managers can make informed decisions about how to
protect different species in these rivers, like salmon, steelhead and trout.

2.5 Using Major Ions to Understand Chemical Weathering Processes in the
Subglacial Environment
To gather a general understanding of the chemical weathering occurring in the
subglacial environment, this thesis project gathered data on major ion export over the
summer months. If microbial communities are present in the subglacial environment and
they are oxidizing organic carbon for energy, this could affect the chemical weathering
processes and the major ions being assimilated into the meltwater (Bhatia et al., 2013;
Wadham et al., 2010). The major ions of interest used here to understand the general
chemical weathering patterns beneath the glacial ice are Na+, Mg2+, Ca2+, K+, HCO3−and
SO42- (Bhatia et al., 2013; Brown, 2002; Raiswell, 1984; Wadham et al., 2010). These
ions are important to look at because Ca2+ and Mg2+ are the generally the first cations
weathered into glacial meltwaters, followed by Na+ and K+, while HCO3−and SO42- are
the principal anions (Brown, 2002; Raiswell, 1984).

20

By looking at the associations between different ions in glacial meltwater, a more
detailed comprehension of the chemical weathering processes occurring can be
understood. This information can then be compared between the streams with differing
glacial coverage to assess whether chemical weathering and major ion export may change
as glaciers recede. According to Raiswell, (1984) acid hydrolysis, which is the most
important process in weathering rock minerals, is driven by free protons (H+) in the
subglacial environment. This process can be understood by looking at the chemical
weathering processes of three minerals commonly found in bedrock beneath glaciers,
carbonates, silicates and aluminosilicates (Raiswell, 1984).
1) Carbonates: CaCO3(s) (calcite) + H+(aq)  Ca2+(aq) + HCO3-(aq)
2) Silicates: Mg2SiO4(s) (forsterite) + 4H+(aq)  2Mg2+(aq) + H4SiO4(aq)
(dissolved silica)
3) Aluminosilicates: 2NaAlSi3O8(s) (albite) + 2H+(aq) + 9H2O(l)  2Na+(aq) +
4H4SiO4(aq) + Al2Si2O5(OH)4(s) (kaolinite)
The availability of the protons beneath the ice controls the rate of ion dissolution
into the subglacial waters. The weathering processes in equations 1, 2, and 3 show the
general weathering processes of protons (H+) on these common minerals, but there are
two separate processes that provide these protons (H+) that drive chemical weathering.
The two main processes are carbonate dissolution and sulfide oxidation (Bhatia et al.,
2013; Brown, 2002; Raiswell, 1984). The chemical equations for these two processes that
result in free protons (H+) are the dissolution and dissociation of atmospheric CO2 which
involves carbonate reactions (carbonate dissolution) (equation 4 and 5) and sulfide
oxidation (equation 6) (Brown et al., 2002; Raiswell, 1984).
21

4)

CO2 (g) + H2O(aq)  CO2(aq)  H2CO3(aq)  H+(aq) + HCO3-(aq)  2H+(aq) +
CO3-(aq)

5)

CaCO3 (s) + H2CO3-(aq)  Ca2+ (aq) + 2HCO3-(aq)

6)

4FeS2 (s) + 15O2 (aq) + 14H2O (l)  4Fe(OH)3 (s) +8SO42-(aq) +16H+ (aq)

As the equations above indicate, O2 and CO2 are needed for carbonate dissolution
to occur beneath the ice. This process is common not only in glacial environments, but
also in non-glacial streams and rivers because carbonates are easily weathered into water
by carbonic acid (H2CO3) attack (see equation 5) (Berner & Berner, 2012). Some
constraints on ion dissolution into meltwater include the residence time of the water in
subglacial environment, access the subglacial water has to weathered sediment (rock
flour), the supply of O2 and CO2 in the drainage system, and the lithology of bedrock. The
supply of O2 and CO2 in the subglacial environment is often limited by access of the
meltwaters to the atmosphere. The drainage system can be closed to the atmosphere
under certain conditions, like when the drainage tunnels are filled completely with
meltwater. In a closed-system the only supply of O2 or CO2 is from gas bubbles in the
ice, englacial voids or microbial activity that oxidizes OC (Brown, 2002).
Sulfide oxidation is not as common as carbonate dissolution in non-glacial
streams. In glacial environments, the glacier moves over the rocks grinding it into rock
flour. This rock flour commonly contains pyrite minerals (FeS2) and since the rock has
been ground up it has fresh pyrite surfaces present (Brown, 2002; Raiswell, 1984). The
fresh pyrite surfaces are reactive with water, resulting in the quick assimilation of
sulphate (SO42-) into meltwater (see equation 6) (Brown, 2002; Bhatia et al., 2013;
Raiswell, 1984; Wadham et al., 2010). In oxygen-rich waters at 0 °C, sulphate
22

concentrations can quickly reach concentrations as high as ~ 400 µeq L-1 from sulfide
reactions (Bhatia et al., 2013; Sharp et al., 1999). Non-glacial streams do not have rock
flour present allowing for the quick production of sulphate, so they commonly have lower
concentrations of sulphate. Studies have also indicated that snowmelt has little sulphate
present, so sulphate in glacial meltwater is mostly due to pyrite weathering beneath the
glacier (Brown, 2002, Mitchell et al., 2013). Because of this, sulphate is a metric that can
be used as a tracer indicating meltwater in glacial streams likely came from the subglacial
drainage system.
Microbial communities in the subglacial environment have also been shown to
mediate sulfide oxidation, which is the reduction of sulfur compounds and this can result
in additional sulphate in glacial meltwater (Bhatia et al., 2013; Mitchell et al., 2013;
Wadham et al., 2010). Since sulphate concentrations should not get over ~ 400 µeq L-1
from sulfide reactions involving pyrite in oxygen-rich waters at 0 °C, sulphate
concentrations over this amount may imply that microbial communities could be
performing sulfide oxidation producing these high sulphate concentrations (as cited in
Bhatia et al., 2013 p.340).
By comparing the proportions of HCO3−and SO42- in the meltwater a general
understanding of processes occurring in the subglacial drainage system can be concluded.
Data on HCO3−and SO42- can help distinguish the source of the protons (H+) that drive the
acid hydrolysis reactions (Bhatia et al., 2013; Brown, 2002; Hasnain, 1999a; Raiswell,
1984; Wadham et al., 2010). The C-ratio [HCO3-/ (HCO3- + SO42-)] or the S-ratio [SO42−/
(SO42−+ HCO3-)] provide an understanding of the chemical weathering occurring beneath
this ice (Brown, 2002; Hasnain, 1999a; Singh et al., 2014). For this project, the C-ratio
23

was utilized to understand the general chemical weathering processes. If the C-ratio
produces a ratio ~ 1 then carbonate reactions dominate in the subglacial environment
(Brown, 2002; Hasnain, 1999a). This mean the dissolution and dissociation of CO2
produces the protons needed for chemical weathering. The equation for these reactions
are (Brown, 2002):
7) CaCO3(s) (calcite) + CO2(aq) + H2O(aq) = Ca2+(aq) + HCO3-(aq)
8) CaAl2Si2O8(s) (Ca-Feldspar) + 2CO2(aq) +2H2O(aq) =
Ca2+(aq) + 2HCO3- + H2Al2Si2O8(s) (weathered feldspar)

A C-ratio of 1 indicates that the drainage system is open to the atmosphere, because the
atmospheric CO2 is needed to produce protons, although CO2 can also come from bubbles
in the ice or from the oxidation of carbon by microbial communities beneath the ice
(Bhatia et al., 2013; Brown, 2002; Wadham et al., 2010). If the C-ratio is ~ 0.5 then
coupled sulfide oxidation and carbonate dissolution are the predominate hydrolysis
processes occurring and pyrite oxidation produces some of the protons for chemical
weathering (Brown, 2002; Hasnain, 1999a; Singh et al., 2014). The equation for this
coupled reaction is (Brown, 2002):
9) 4FeS2(s) + 16CaCO3(s) + 15O2(aq) + 14H2O(aq) = 16Ca2+(aq) + 16HCO3-(aq) +
8SO42-(aq) + 4Fe(OH)3(s)
The S-ratio [SO42−/ (SO42−+ HCO3-)] is an alternative ratio that researchers have
used to give similar insight into the which reactions are occurring beneath the glaciers
(Brown, 2002). An S-ratio of 0.5 signifies there is the coupled sulfide oxidation and

24

carbonate dissolution, where an S-ration of 0.0 suggests the weathering is just from
carbonate reactions (Brown, 2002).
Alternately data on HCO3- and SO42- can be compared to gain insight into the
possible presence of microbial communities beneath the ice. When meltwater flows
through subglacial flour in the subglacial drainage system, carbonate hydrolysis occurs
quickly due to the high reactivity of the minerals in the rock flour (Bhatia et al., 2013;
Wadham et al., 2010). The carbonate hydrolysis reaction results in the production of
HCO3- apart from SO42- (Bhatia et al., 2013; Tranter et al., 2002; Wadham et al., 2010).
By utilizing the theoretical solubility of calcite in meltwater at 0°C, this reaction should
result in ~ 220 µeq L-1 of HCO3-. Plots comparing HCO3- versus SO42- with y-intercepts
greater than 220 µeq L-1 indicates there must be an additional source of CO2 to account
for the additional HCO3- in the meltwaters (Bhatia et al., 2013; Brown, 2002; Wadham et
al., 2010).
This CO2 source could be from contact with the atmosphere if it is an open
drainage system. If the drainage system is closed and there is no contact with atmospheric
CO2, studies have recently shown it can be a result of the oxidation and fermentation of
OC by microbial communities beneath the ice (Bhatia et al., 2013; Brown, 2002;
Wadham et al., 2010). These ratios and ion comparisons can change over a glacial melt
season reflecting the evolution from snowmelt to meltwater from the subglacial drainage
system. These ratios allowed Wadham et al. (2010) to determine that microbial
communities were present beneath the glacial ice in Antarctica, Greenland, Norway,
Switzerland and Svalbard. The research team found high intercepts of HCO3- in borehole

25

samples of meltwater from glaciers in Norway and Switzerland, indicating microbes were
oxidizing subglacial OC and creating CO2 (Wadham et al., 2010).
A study by Brown (2002) discusses using pCO2 values, along with the major ion
data to get an understanding into if the glacial drainage system is open or closed to
atmospheric gases. A drainage system open to the atmosphere would display pCO2 values
equal to atmospheric values, while a low or high pCO2 value would indicate that the
system is closed to the atmosphere (Brown, 2002; Hasnain, 1999a; Raiswell, 1984). Low
pCO2 value occur when carbonate reactions cause a more rapid use of protons in the
subglacial environment than can be resupplied by CO2 diffusion in these water-ice-rock
interfaces (Brown, 2002). This often occurs when there is a large amount of new rock
flour present and a large volume of meltwater (as cited in Brown, 2002 p.865). High
pCO2 values occur when hydrolysis provides more protons than there are protons being
consumed in the subglacial environment (Brown, 2002; Hasnain, 1999a). This can
happen when protons are provided to meltwater through sulphide oxidation or additional
snowmelt input (Brown, 2002; Hasnain, 1999a). The neutralizing of acidity by carbonates
or the freezing of the meltwater can also contribute to the high pCO2 values. It is
common for meltwater to transition from open to closed drainage systems or from high to
low pCO2 conditions throughout the summer as the drainage system evolves (Brown,
2002; Hasnain, 1999a). By combining pCO2 data with the major ion data this study can
gain more insight into the processes occurring beneath the ice and the main source rocks
contributing to weathering.
This literature review has discussed the studies previously done on DOC and
major ion export from glacial meltwater. It has discussed many methods used to
26

understand the properties of this glacial DOC and its importance to aquatic ecosystems
downstream. It has also highlighted the need for similar research in the Pacific Northwest
for two main reasons. To add to the larger body of knowledge on glaciers role in the
global carbon cycle and so local wildlife and natural resource managers can properly
protect our resources as glaciers continue to recede and ultimately disappear. My research
project will give local insight into the temporal and spatial patterns of DOC and major
ion export. This project looked at glacial meltwater from Emmons Glacier, which has the
largest area on Mount Rainier (11.1 km2), and meltwater from the Inter Fork Glacier,
which is one of the smallest glaciers on the mountain (0.8 km2) (NPS, 2016; USGS,
2016). Data from two non-glacial streams was also gathered, so a comparison with these
should provide interesting insight into how DOC and major ion export varies depending
on glacial size at the headwaters. This project presents an opportunity to study current
temporal and spatial trends in local DOC export and study how the timing and
concentration of DOC could change as glacial recession continues over time (NPS,
2016).

3. Methods
3.1 Research Design
This research project aimed to perform an analysis of multiple water parameters
on glacial headwater streams on Mt. Rainier. The water parameters measured on sight
from these streams included temperature, dissolved oxygen, conductivity, velocity, pH,
and pCO2. Stream samples were collected to analyze back in the lab for alkalinity, major
ions (SO42-, HCO3-, Na+, K+, Mg2+ and Ca2+) and dissolved organic carbon (DOC). The
27

samples were gathered throughout the months of May to September in 2016 in order to
observe any changes that may occur to the water parameters over the glacial melting
season.
The glacial rivers come from two significantly different sized glaciers. The White
River comes from the Emmons Glacier, which is the largest glaciers on Mount Rainier
with an area of 11.1 km2 (NPS, 2016). The Inter Fork White River comes from the glacial
melt waters of the Inter Glacier, which has an area of about 0.8 km2 (USGS, 2016). The
two non-glacial streams used for reference came from precipitation events and the waters
of nearby lakes. Since the non-glacial streams come from the drainage waters of lakes
rather than being dominantly precipitation driven, they do not represent perfect reference
streams but do provide interesting references since they are in previously glaciated
regions. Often as glaciers melt, they leave behind chunks of ice that can melt into lakes,
so although not ideal reference streams they can still give some insight into how an
evolving glacial landscape can alter stream parameters.
The first non-glacial stream comes from the waters of Shadow Lake, but this
stream dried up by the end of June so another non-glacial stream nearby was used for
continued sample collection. The second non-glacial stream came from the waters of
Frozen Lake. The difference in glacial area at the headwaters will provide for comparison
between glacial and non-glacial streams and provide understanding in how this glacial
coverage may influence the water parameters. On the glacially fed streams, there were
two sampling locations, one upstream and one downstream. A comparison of these two
locations should provide insight into how distance from the glacier may affect the water
parameters as well as any influence from merging streams.
28

3.2 Sampling Locations
All the glacial stream sites and non-glacial stream sites are located on the Eastern
side of Mount Rainier in Washington State (see figure 4 and 5). These sites were chosen
due to accessibility and the feasibility of collecting samples from multiple locations.
There are two separate sampling sites on the glacially-derived streams, the White River
and the Inter Fork White River, which eventually merge into one river. There is one
sampling site on the two non-glacial streams nearby for reference, and these streams
eventually merge into the Inter Fork River. On the White River, one sampling site (Upper
White River) is located close to the terminus of Emmons glacier. The other White River
site is located downstream (Lower White River), which is situated in an area past the
merging of the Inter Fork White River with the White River. On the Inter Fork White
River, one sampling site (Upper Inter Fork) is located upstream near the terminus of the
Inter Glacier. The second sampling site on the Inter Fork White River (Lower Inter Fork)
is located downstream from the merging of the non-glacial creek with the Inter Fork
White River. Both non-glacial streams are located on the northern side of the Inter Fork
White River and they both eventually merge into this stream. All the stream sampling
locations are located between 4,000 and 4,500 feet in elevation (NPS, 2016).

29

Figure 4: Zoomed out view of the glaciers the feed the rivers where the sampling
locations are located. Emmons Glacier is about 4.3 mi2 in area and the Inter Glacier
situated next to Emmons Glacier is about 0.3 mi2 in area. The White River comes from
the glacial melt off Emmons Glacier and the Inter Fork River comes from the melt waters
of the Inter Glacier (Hekkers & Thorneycroft, 2011).

30

Figure 5: Zoomed-in view of the field area and sites on Mount Rainier (adapted from
USGS, 2016).
= Field sites along the White River, which comes from the Emmons glacier that has
the largest glacial coverage
= Field sites along the Inter Fork River, which comes from that Inter Fork glacier that
has the smaller glacial coverage
= Field sites on the non-glacial streams that receive no glacial melt (0% glacial
coverage)

The forested areas in the White River region on Mount Rainier consist mostly of
subalpine fir (Abies lasiocarpa), mountain hemlock (Tsuga mertensiana), Alaskan yellow
cedar (Cupressus nootkatensis), whitebark pine (Pinus albicaulis), and Engelmann spruce
(Picea engelmannii) (NPS, 2007). The White River area has a rich understory common
for wet mountainous environments, including a variety of species of mushroom, ferns,
mosses and wildflowers. (NPS, 2016). The White River begins near the tree line at the
terminus of the Emmons Glacier and travels into the tree line. The sampling locations for
this project were located near or slightly within the tree line. There was 0% canopy cover
present in the White River sampling locations, which minimizes sources of possible
31

terrestrial carbon to the river system. The Inter Fork River begins at a much higher
elevation at the terminus of the Inter Glacier, but its headwaters are still near the tree line.
This is due to glacial recession from the Inter glacier leading to terrestrial succession at
these higher latitudes. The Inter Fork River is mostly in the forest with both sampling
locations along the river having ~ 30% to 40% canopy cover. The two non-glacial creeks
were both located at lower elevations in the forest, and they ~ 70% tree canopy cover or
more.
3.3 Geology of Mount Rainier
Mount Rainier is the tallest stratovolcano in the Cascade Mountain Range
reaching 14,410 feet (NPS, 2005). This impressive height is one of the reasons that
Mount Rainier has so many glaciers. With so much ice present, the mountain is the
largest single peak glacial system in the country (NPS, 2005). The mountain began
forming 500,000 years ago through successive lava and pyroclastic flows that left behind
mainly andesitic and basaltic rocks (USGS, 2014). Erosion of these rocks by the glaciers
leaves behind moraines and glacial till. Mudflows and debris flows have occurred in the
past moving around vast amounts of material, while glacial outbursts move large amounts
of water and sediments into the valleys below. In 1963, there were multiple rockslides
from Little Tahoma Peak onto Emmons Glacier, which resulted in rock debris covering
the lower part of the glacier. The rock cover on the ice creates insulation keeping the
glacier cooler than the other glaciers on Mount Rainier. This has resulted in less recession
occurring at the Emmons Glacier in comparison to the other glaciers, but it has still been
receding since the late 1800’s (Hekkers & Thorneycroft, 2011; NPS, 2005). The last
volcanic activity at Mount Rainier occurred between 1820 and 1850 based on tephra

32

deposits (USGS, 2015). Currently, the volcanic status of Mount Rainier is active (USGS,
2014). Emmons Glacier is located on andesite flows that are around 180-280,000 years
old and the Inter Glacier sit on top of andesite lava flows that are around 450-500,000
years old (USGS, 2014). Research has indicated the presence of pyrite minerals in
hydrothermal alterations in the rocks on Mount Rainier (John, Sisson, Breit, Rye &
Vallance, (2008).
3.4 Field Methodology
In the field, the glacial and non-glacial streams were measured for multiple water
parameters. The pH was recorded on site using an Oakton pH probe. At least five
measurements of pH were taken over a ten-minute period.
A YSI Pro 2030 meter, which was calibrated for freshwater, was used to measure
dissolved oxygen, conductivity and temperature of the streams. At least three
measurements each of dissolved oxygen, conductivity and temperature were recorded.
An Eosense GPsensor for measuring pCO2 was used in the streams. The pCO2
probe was turned on for 30 minutes before use. The pCO2 probe was not always available
in the field due to poor weather conditions so some pCO2 rates were calculated using the
temperature and alkalinity data gathered. The equation used to calculate pCO2 is
(Wanninkhof, 1992; Weiss, 1974):
pCO2 = [H2CO3]/KH
Where KH = e^(-58.0931+90.5069(100/T)+22.294ln(T/100))

33

The alkalinity data was used to determine the concentration of carbonic acid [H2CO3] in
the water. This is because the equation for alkalinity is:
Alkalinity = [HCO3-] + 2*[CO32-] + [OH-] - [H+]

Carbonic acid was calculated from the temperature-dependent equilibrium constant
equation for the dissociation of carbonic acid to bicarbonate [HCO3-] (USGS, 2012):
H2CO3 = ([H+][HCO3-])/K1
Where K1 = ([HCO3-][H+])/[H2CO3]
K2 = ([CO3]*[H+])/[HCO3-]
pH = -log[H+]  [H+] = log^pH
A Model 2100 Series Current Velocity “Swoffer” Meter was utilized to gather
velocity data for the Inter Fork River and the non-glacial streams. Discharge of the White
River was too high to safely use the swoffer meter or gather depth measurements. Due to
this, it was not possible to get accurate velocity measurements, so the float method was
used to gather data on the velocity of the White River. The float method uses an object
that floats in the water to determine how fast the water is moving. For this project, an
orange was floated down a 50 foot stretch of the White River and the time it traveled in
the water was recorded. The float method was repeated three times. Then the equation V
= Q/A was used to calculate velocity where V = velocity (travel distance/travel time), Q
= discharge (velocity * area) and A = cross sectional area (width * average depth).
Field observations were recorded at each site, including weather, canopy cover,
substrate, and if there was the presence or absence of primary production or
34

macroinvertebrates in the streams. Data on precipitation, temperature and snow depth
were gathered from NOAA’s NOWData sets (2016) from the Paradise weather station on
Mount Rainier. There is no weather station near the White River where the sampling
locations were but the Paradise weather station is nearby and located at a similar
elevation. This data was extrapolated to all the sampling locations since they are in close
proximity to one another. Snowmelt data was calculated from snow depth data by
subtracting snow depths. This study used the total snowmelt that occurred over 7 days
prior to the date of field work to see if DOC concentrations were dependent on the
snowmelt over the prior week.

3.5 Lab Methodology
Filtration and Preservation
Water samples for laboratory analysis (dissolved organic carbon, major ions,
alkalinity) were gathered in one liter acid washed (10% or 1.2 molar HCl for 12 hours)
polypropylene bottles in the field. These samples were kept on ice until they arrived back
at the laboratory for filtration and analysis or storage. All water samples brought back
from the field were immediately filtered in the laboratory at The Evergreen State College.
Dissolved Organic Carbon (DOC) Analysis
Water samples analyzed for DOC were filtered through 0.07 µm glass fiber
filters that had been combusted at 450˚C for 4.5 hours. The water samples were then
stored in 25 mL glass vials that had been combusted at 500˚C for 4.5 hours. These glass

35

vials were frozen until analysis (~28 days). The organic carbon analysis was completed
by the Analytical Service Center at the University of Washington using the TOC/TN
analyzer. DOC methods were gathered from the U.S.G.S methods (2002). Analytical
error was less than 10% based on replicate samples.
Alkalinity Methods
Alkalinity samples were filtered through 0.45 µm cellulose acetate filters and
stored in 250 mL acid washed polypropylene bottles and kept around 4˚C for
preservation until analysis (within 24 hours) (USGS, 2012). To analyze the water samples
for alkalinity, a sulfuric acid titrant solution of 0.2 N, along with a standard solution of
0.01639 N Na2CO3 were created. To determine the normality of the sulfuric acid solution,
100 mL of the Na2CO3 solution was titrated with the sulfuric acid solution using a 0.2 mL
Gilmont micrometer burette following methods described in the USGS protocols (2012).
Titrations were repeated three times for accuracy.
Once the normality of the sulfuric acid solution was calculated, the sulfuric acid
was then used to titrate the water samples from the field using 100 mL water samples
following the USGS (2012) protocol for alkalinity. An Orion pH electrode was used for
pH measurements. Each water sample was titrated three times for accuracy. The Gran
method was used for calculating the alkalinity of the water samples
Dionex IC25 Methods
Water samples to be analyzed for major anions were run on the Dionex Ion
Chromatograph 25. They were filtered through 0.45 µm cellulose acetate filters and
stored in 250 mL acid washed polypropylene bottles and kept around 4˚C for
36

preservation until analysis (within 1 month) (U.S EPA, 1993). Sulfate standards were
created by drying K2SO4 at 105˚C for 30 minutes and then cooling it in a desiccator.
Standards of 1 ppm, 5 ppm and 10 ppm and a quality control of 7 ppm were created to
run on the IC25. The quality control was prepared from a different stock of K2SO4 to
insure accuracy. Concentrations for the standards were determined based off previous
studies findings of SO42- in glacial melt water. The standards, quality control, blanks and
stream water samples were all filtered through 0.2 µm filters on the IC25. The standard
curve was always 0.9998 or greater and the quality control was ± 15% of what the quality
control concentration should be based on calculations. The major anion (SO42-) was
determined by ion chromatography on the Dionex IC 25 using an anion trap column
(ACT-3, 4mm), a guard column (AG17, 4 mm) and an analytical column (AS17, 4 mm).
The injected water samples were 25 mL and each sample ran for 14 minutes through the
Dionex IC 25 with a flow rate of 1 mL/sec. The eluent was helium gas set at a
concentration of 10 mM. Analytical error was less than 10% based on replicate samples.
PerkinElmer Elan DRC-e ICP-MS Methods
Water samples to be analyzed for major cations were run on the PerkinElmer Elan
DRC-e ICP-MS. They were filtered through 0.45 µm cellulose acetate filters into 250 mL
acid washed polypropylene bottles, acidified using nitric acid (HNO3) and kept around
4˚C for preservation until analysis (within 6 month) (U.S EPA, 1994). Calibration
standards of 100 ppb, 500 ppb, 1,000 ppb, 1,500 ppb and 2,500 ppb were created using
Mg2+, K+, Na+ and Ca2+. A quality control of 700 ppb was made using a different stock of
the analytes and an internal standard of Scandium (Sc) was created. Each standard, the
quality control and the blank were prepared in 50 mL acid washed tubes with a 1% HNO3
37

matrix and1 ppm of the internal standard. When ran, the standards had a R2 value of
0.9998 or greater and the quality control was ± 10% of what the quality control
concentration should be based on prior calculations. The samples were prepared in 15 mL
acid washed tubes with a 1% HNO3 matrix and1 ppm of the internal standard. The
samples were run after the standards and quality control were ran to insure accuracy. The
major cations (Mg2+, K+, Na+, Ca2+) were determined by inductively coupled plasmamass spectrometry (ICP-MS) on the PerkinElmer using Elan instrument software. A
quantitative summary analysis was performed with the machine in dual detector mode
using an auto-analyzer. The ICP-MS was in peak processing mode and DRC mode
throughout the analysis. Analytical error is generally less than 10% based on replicate
samples.
3.6 Statistical Analysis Methods
The data analysis was completed using excel and JMP Pro12. Excel was used to
plot the temporal data gathered on DOC and major ions throughout the melt season.
Excel and JMP Pro12 were also used to perform simple linear regression analysis on the
data. JMP Pro12 was used to perform correlation analysis on DOC concentrations against
multiple parameters. Spearman’s ρ was used because some of the data did not have a
normal distribution.

4. Results
4.1 Dissolved Organic Carbon Analysis
Dissolved organic carbon concentrations had peak values in early to mid-June and
then decreased over time for all sites (see Figure 6 & 7). Concentrations in the upper
38

White River (UWR) site ranged from 0.1 to 1.5 mg L-1, while the lower White River
(LWR) site had values that ranged from 0.7 to 4.9 mg L-1. The upper Inter Fork White
River (UIF) site had DOC concentrations ranging from 0.7 to 3.3 mg L-1 and the lower
Inter Fork White River (LIF) site had DOC concentrations of 0.8 to 3.4 mg L-1. The first
non-glacial creek (NGC1) had DOC concentrations of 1.5 to 7.0 mg L-1. The DOC
concentrations in the NGC1 showed a different pattern compared with the glacial streams
displaying two high peaks between late May and mid-July. After mid-July, this nonglacial creek dried up, so there is no data past this date for that creek. The second nonglacial creek (NGC2) had DOC concentrations ranging from 2.0 to 4.0 mg L-1.

White River DOC Concentrations
7
6

DOC mg/L

5
4
LWR

3

UWR
2
1
0
5.21.16

6.4.16

6.18.16

7.9.16

7.23.16

8.12.16

8.27.16

9.10.16

Date of Collection

Figure 6: DOC concentrations in the upper and lower White River sites over the glacial
melt season.

39

Inter Fork White River DOC Concentrations
7
6

DOC mg/L

5
4
UIF

3

LIF
2
1
0
5.21.16

6.4.16

6.18.16

7.9.16

7.23.16

8.12.16

8.27.16

9.10.16

Date of Collection

Figure 7: DOC concentrations at the upper and lower Inter Fork White River sites over
the glacial melt season.

Non-Glacial Creek DOC Concentrations
7
6

DOC mg/L

5
4
NGC1

3

NGC2
2
1
0
5.21.16

6.4.16

6.18.16

7.9.16

7.23.16

8.12.16

8.27.16

9.10.16

Date of Collection

Figure 8: DOC concentrations in the two non-glacial creek sites over the glacial melt
season.

40

To gain insight to what parameters may be controlling DOC export, DOC
concentrations were plotted against data on temperature, precipitation, snow depth,
snowmelt, and velocity (see table 1). Simple linear regression was used to understand if
DOC was dependent on any specific parameter. A correlation analysis was also
performed to understand positive and negative relationships between the data (see table
2). The data used for temperature, precipitation and snow depth all came from NOAA’S
NOWData sets (2016), and this data is displayed in Table 1 for the timepoints that were
sampled. Velocity data came from measurements made in the field using the swoffer
meter or the float method. Snow depth, snow melt and velocity displayed significant
results for some sites when plotted against DOC concentrations, while temperature and
precipitation did not display significant results (see table 2).
Climatological Data from Paradise Rainier, WA
5.21.16 6.4.16
6.18.16
7.9.16
7.23.16
8.13.16
8.27.16
9.10.16
High Temperature (°C)
5
18.9
8.9
10
10
23.3
21.1
16.1
Low Temperature (°C)
0.6
3.3
1.1
2.8
4.4
13.9
10.6
8.9
Average Temperature (°C)
2.8
11.1
5
6.4
7.2
18.6
15.8
12.5
Precipitation (cm)
0.58
0
1.3
1.4
0
0
0
0
Snow Depth (inches)
100
70
42
0
0
0
0
0
Total 7-Day Snowmelt (inches)
10
25
20
10
0
0
0
0

Table 1: Data for snow depth, snowmelt, precipitation and temperature (NOAA, 2016).

Velocity
Snow Depth
Snowmelt
1 Day Precipitation
7-Day Precipitation
1 Day Temperature
7-Day Temperature

R Values from Correlation Analysis
Upper White River Lower White River Upper Inter Fork
Lower Inter Fork
-0.79 (p = 0.04)
-0.33
-0.68
-0.43
-0.02
0.79 (p = 0.02)
0.08 (p = 0.03)
0.76 (p = 0.05)
-0.13
0.76
0.81 (p = 0.03)
0.77 (p = 0.04)
0.78 (p=0.04)
0.33
0.37
0.51
0.47
0.003
0.02
0.34
0.5
0.01
0.09
0.3
0.16
0.1
0.26
0.55

41

Table 2: Correlation results of DOC concentration with multiple parameters. R-value
based on correlation analysis using Spearman’s ρ due to the data not being normally
distributed.
Snow depths throughout the field season ranged from 0 to 100 inches (NOAA,
2016). The deepest snow level during the data collection period occurred in mid-May and
then the snow depth decreased over time until July. After July, the snow depth stayed at 0
inches for the remaining part of the field season. Based on simple linear regression, the
relationships between snow depth and DOC concentration were significant at the UIF
with a p = < 0.001, R2 = 0.97, followed by the LIF with a p = 0.0427, R2 = 0.59. (see
figures 9 & 10). Significant relationships between snowmelt and DOC were found at the
UIF with a p = 0.0031, R2 = 0.85, followed by the LWR with a p = 0.0047, R2 = 0.76,
and the LIF with a p = 0.0298, R2 = 0.64 (see figures 11, 12, & 13).

Upper Inter Fork River
4

p-value = <.001

3.5

R² = 0.9747

DOC (mg/L)

3
2.5
2
1.5
1
0.5
0
0

10

20

30

40

50

60

70

80

Snow Depth (inches)

Figure 9: Relationship between DOC (mg L-1) and snow depth (inches) at the
upper Inter Fork White River site.

42

Lower Inter Fork River
4

p-value = 0.0427

3.5

DOC (mg/L)

3

R² = 0.5934

2.5
2
1.5
1
0.5
0
0

10

20

30

40

50

60

70

80

Snow Depth (inches)

Figure 10: Relationship between DOC (mg L-1) and snow depth (inches) at the
lower Inter Fork White River site.

Lower White River
6

DOC (mg/L)

5

p-value = 0.0047

4

R² = 0.7615

==

3
2
1
0
0

5

10

15

20

25

30

Snowmelt (inches)

Figure 11: Relationship between DOC (mg L-1) and total 7-day snowmelt (inches)
at the lower White River site.

43

Upper Inter Fork River
4
3.5

p-value = 0.0031

DOC (mg/L)

3

R² = 0.8504

2.5
2
1.5
1
0.5
0
0

5

10

15

20

25

30

Snowmelt (inches)

Figure 12: Relationship between DOC (mg L-1) and total 7-day snowmelt (inches)
at the upper Inter Fork White River site.

Lower Inter Fork River

4
3.5

p-value = 0.0298

DOC (mg/L)

3

R² = 0.6441

2.5
2

1.5
1
0.5
0
0

5

10

15

20

25

30

Snowmelt (inches)

Figure 13: Relationship between DOC (mg L-1) and snowmelt (inches) at the
lower Inter Fork White River site.

Velocity values for the White River ranged from 0.4 to 0.7 m/sec while the Inter
Fork White River had velocity values ranging from 0.4 to 1.2 m/sec. The non-glacial
creeks had velocities ranging from 0 to 0.1 m/sec. Based on simple linear regressions of
velocity against DOC, the UIF site displayed significant results with a p = 0.0113, R2 =
0.75, as well as the UWR site with a p = 0.0168, R2 = 0.71 (see figure 14 & 15).
44

DOC (mg/L)

Upper Inter Fork River
4
3.5
3
2.5
2
1.5
1
0.5
0

p-value= 0.0113
R² = 0.7507

0

0.2

0.4

0.6

0.8

1

1.2

1.4

Velocity (m/sec)

Figure 14: Relationship between DOC (mg L-1) and velocity (m/sec) at the upper
Inter Fork White River site.

Upper White River

DOC (mg/L)

9.0
8.0

p-value = 0.0168

7.0

R² = 0.7133

6.0
5.0
4.0
3.0
2.0
1.0
0.0
0

0.2

0.4

0.6

0.8

1

1.2

1.4

1.6

Velocity (m/sec)

Figure 15: Relationship between DOC (mg L-1) and velocity (m/sec) at the upper White
River site.

Average temperatures on Mount Rainier ranged from -1.4 to 19 ± 17.3 °C throughout
the field season (see table 1) (NOAA, 2016). Relationships between DOC concentrations
and the daily average temperature from the day of sampling were not significant based on
simple linear regression. Simple linear regressions were also performed on DOC
45

concentrations against a 7-day average temperature. The relationships between the 7-day
average temperature and DOC were not significant (see table 2).
The daily precipitation totals varied from 0 to 2.4 cm during the sample collection
period (see table 1) (NOAA, 2016). Based on simple linear regression, the relationship
between DOC concentrations and the daily precipitation totals were significant for the
UWR site displaying with a p = 0.047, R2 = 0.58 (see figure 16). Simple linear
regressions were also done using a 7-day precipitation total instead of the daily
precipitation total to see if precipitation over time affected DOC export. The relationships
between the 7-day precipitation totals and DOC were not significant (see table 2).

Upper White River

DOC (mg/L)

1.6
1.4

p = 0.047

1.2

R² = 0.5774

1
0.8
0.6
0.4
0.2
0
0

0.1

0.2

0.3

0.4

0.5

0.6

Daily Precipitation (cm)

Figure 16: Relationship between DOC (mg L-1) and daily precipitation (cm) at the upper
White River site.

4.2 Other Parameters: pH, alkalinity and pCO2
46

The pCO2 values in the glacial streams were generally low, with some values less
than atmospheric equilibrium values, and some higher (see table 3). Values at the UWR
site ranged from 199.2 to 293.1 ppm with an average of 247.2 ± 15.5 ppm. The LWR site
had values ranging from 214.8 to 594.9 ppm with an average value of 321.9 ± 45.4 ppm.
The Inter Fork White River sites had much more variable pCO2 values, but overall
displayed higher values than the White River. The UIF values ranging from 289.5 to
618.9 ppm with an average of 454.4 ± 52.7 ppm and the LIF site had values ranging from
266.3 to 833.7 ppm with an average value of 494.4 ± 81.1 ppm. The NGC1 had pCO2
values ranging from 259.3 to 637.7 ppm with an average of 434.8 ± 89.8 ppm and the
NGC2 had pCO2 values ranging from 607.1 to 753.9 ppm with an average of 655.4 ±
43.6 ppm.

Site/Date
LWR
UWR
LIF
UIF
NGC1
NGC2

5.21.16

6.4.16
369
429

259.3

214.8
288.6
833.7
604
610.9

pCO2 µatm or ppm
6.18.16
7.9.16
7.23.16
8.12.16
8.27.16
9.10.16
310.9
310.9
594.9
250.8
255.1
268.6
257.2
293.1
225.8
199.2
261.1
205.6
266.3
316.5
450
784
321.3
554.0
289.5
402.5
618.9
420.5
516.4
328.9
281.5
384.6
637.7
607.1
753.9
586.8
673.8

Table 3: pCO2 values at the glacial and non-glacial stream sites over the field season.
pH values that were used to calculate the pCO2 values were gathered in the field
(see table 4). Values for pH at all the sites was fairly consistent throughout the summer
months, with the NGC2 displaying higher pH values than any other sites. At the UWR
site the pH values ranged from 7.1 to 7.3 and at the LWR site pH ranged from 6.9 to 7.3.
Values ranged from 7.0 to 7.3 at the UIF site and they ranged from 6.9 to 7.3 at the LIF
site. The NCG1 displayed values ranging from 6.9 to 7.4 while the NGC2 had pH values
ranging from 7.6 to 7.7.
47

Site/Date
LWR
UWR
LIF
UIF
NGC1
NGC2

5.21.16

6.4.16
7.3
7.3
7.3
7.3
7.3

6.18.16
7.1
7.1
7.0
7.1
6.9

pH Values
7.9.16
7.23.16
7.2
7.3
7.3
7.2
7.3
7.2
7.2
7.1
7.3
7.4

8.12.16
6.9
7.2
7.1
7.0
7.2
7.7

8.27.16

9.10.16

7.3
7.3
6.9
7.2

7.2
7.1
7.2
7.1

7.3
7.3
7.1
7.3

7.6

7.7

7.6

Table 4: pH values for the glacial and non-glacial sites over the field season.
Alkalinity values in both glacial streams were generally consistent throughout the
summer, although both White River sites displayed a large dip in alkalinity values during
early June while the Inter Fork River displayed a gradual decrease in alkalinity during
that time (see figure 17, table 5). The alkalinity in the UWR ranged from 91.9 to 122.7
µeq L-1, while the LWR site had values ranging from 68.4 to 243.2 µeq L-1. The UIF has
alkalinity values ranging from 139.2 to 205.2 µeq L-1 and the LIF had values ranging
from 132.4 to 256.9 µeq L-1. The first non-glacial creek had alkalinity values in the same
range as the glacial streams, but the alkalinity in NGC1 increased over time, while the
glacial streams were generally decreasing over time. The alkalinity in NGC1 ranged from
134.1 to 275.4 µeq L-1. The second glacial stream had alkalinity levels higher than any of
the other glacial or non-glacial sites with values ranging from 669.7 to 718.8 µeq L-1.

Site/Date
LWR
UWR
LIF
UIF
NGC1
NGC2

Alkalinity (ueq/L)
5.21.16
6.4.16
6.18.16
7.9.16
7.23.16
8.12.16
8.27.16
9.10.16
243.2
68.4
159.6
166.8
126.1
137.8
127.6
146.5
91.9
178.1
138.2
95.7
109.5
110.5
122.7
256.9
224.1
145.0
140.9
132.4
152.1
143.3
180.7
205.2
143.1
148.5
139.2
146.8
160.3
182.2
161.3
134.1
177.3
243.7
275.4
680.9
669.7
718.8
682.6

Table 5: Alkalinity values (µeq/L) for the glacial and non-glacial streams over the field
season.

48

Alkalinity in Stream Water
800
700

Alkalinity (µeq/L)

600

LWR
UWR

500

LIF

400

UIF

300

NGC1

200

NGC2

100
0
5.21.16

6.4.16

6.18.16

7.9.16

7.23.16 8.12.16 8.27.16 9.10.16

Date of Collection

Figure 17: Alkalinity values (µeq L-1) for all glacial and non-glacial streams throughout
the field season.

4.3 Major Ion Analysis
The White River sites, the Inter Fork White River sites and the non-glacial creek
all showed similar trends in the relative amounts of major ions in the stream water (see
figures 18-23). At all the sites throughout the summer the general trend for the major
cation concentrations was Ca2+ > Na+ > Mg2+ > K+. The general trend for the major anion
concentrations was HCO3- ≥ SO42-. The K+, HCO3-, and SO42- concentrations were very
similar throughout the summer in all the White River and Inter Fork White River Sites
(see table 6).
At the upper and lower White River sites, all the major ions generally have the
highest concentrations early in the melt season and then the concentrations generally
49

decrease slightly over the summer months. The Ca2+ and Na+ are much more variable
throughout the summer months in comparison to the other major ions (K+, HCO3-, and
SO42-) which sustain lower and steadier concentrations throughout the summer (see figure
18 & 19). At the upper and lower Inter Fork White River sites, Ca2+, Mg2+ and Na+
initially have high concentrations, begin to decrease but then have peak concentrations
again in late July. Similar to the White River, K+, HCO3-, and SO42- sustain lower but
steadier concentrations over the field season (see figure 20 & 21). The first NGC sampled
began with peak concentrations in May with especially high concentrations of Ca2+, Mg2+
and Na+. The concentrations of K+, HCO3-, and SO42- are low similar to the other rivers.
All the ion concentrations from the NGC1 decreased quickly through mid-July and the
stream had dried up by late July (see figure 22). At the second NGC that was used after
the first creek dried up displayed concentration values similar to the glacial creek ion
concentrations with the exception of HCO3- that had really high concentrations (see
figure 23).

UWR

Average Major Ion Concnetrations (µeq/L)
LWR
UIF
LIF
NGC1

NGC2

Ca2+ 973.7 ± 106.4 887.5 ± 32.6 940.1 ± 67.4 1008.4 ± 102 885.5 ± 386.1 998.8 ± 192.4
Mg2+ 352.5 ± 39.5 339.3 ± 11.4 353.8 ± 26.3 349.7 ± 29.7 309.8 ± 135.8 372.1 ± 89.1
K+
115.6 ± 16.8 128.5 ± 11.4 139.8 ± 19.6 125.9 ± 12.7 126.5 ± 63.3 121.2 ± 34.7
Na+ 719.6 ± 79.1 557.9 ± 93.2 546.8 ± 107 724.2 ± 159.8 920.4 ± 533.5 895.2 ± 140.9
SO42- 135.4 ± 17.6 124.6 ± 14.3 106.1 ± 12 91.5 ± 12.5
28.2 ± 10.6 61.6 ± 2.4
HCO3- 120.6 ± 12.1 146.6 ± 18.5 160.4 ± 9.9 171.5 ± 17.2 197.9 ± 48.2 684.6 ± 12

Table 6: Average major ion concentrations (µeq/L) for all glacial and non-glacial stream
sites.

50

Upper White River
1800

Major Ions (µeq /L)

1600
1400
1200

Ca

1000

Mg

800

K

600

Na

400

SO4

200

HCO3

0
5.21.16

6.4.16

6.18.16

7.9.16

7.23.16 8.12.16 8.27.16 9.10.16

Date of Collection

Figure 18: Temporal trends of the major ions (µeq L-1) in the upper White River.

Lower White River
1800

Major Ions (µeq/L)

1600
1400

Ca

1200

Mg

1000

K

800

Na

600

SO4

400

HCO3

200
0
5.21.16

6.4.16

6.18.16

7.9.16

7.23.16

8.12.16

8.27.16

9.10.16

Date of Collection

Figure 19: Temporal trends of the major ions (µeq L-1) in the lower White River.

51

Upper Inter Fork River

1800

Mojor Ions (µeq/L)

1600
1400

Ca

1200

Mg

1000

K

800

Na

600

SO4

400

HCO3

200
0
5.21.16

6.4.16

6.18.16

7.9.16

7.23.16

8.12.16

8.27.16

9.10.16

Date of Collection

Figure 20: Temporal trends of the major ions (µeq L-1) in the upper Inter Fork White
River.

Lower Inter Fork River
1800
1600

Major Ions (µeq/L)

1400

Ca

1200

Mg

1000

K

800

Na

600

SO4

400

HCO3

200
0
5.21.16

6.4.16

6.18.16

7.9.16

7.23.16

8.12.16

8.27.16

9.10.16

Date of Collection

Figure 21: Temporal trends of the major ions (µeq L-1) in the lower Inter Fork White
River.

52

Non-Glacial Creek 1
3000

Major Ions (µeq/L)

2500
2000

Ca
Mg

1500

K
Na

1000

SO4
HCO3

500
0
5.21.16

6.4.16

6.18.16

7.9.16

7.23.16

Date ofCollection

Figure 22: Temporal trends of the major ions (µeq L-1) in the non-glacial creek 1. Major
ion concentrations were much higher initially in the non-glacial creek, so y-axis has a
different scale than all the other graphs.

Non-Glacial Creek 2

1800
1600

Major Ions (µeq/L)

1400
1200

Ca

1000

Mg
K

800

Na

600

SO4
400

HCO3

200
0
5.21.16

6.4.16

6.18.16

7.9.16

7.23.16

8.12.16

8.27.16

9.10.16

Date of Collection

Figure 23: Temporal trends of the major ions (µeq L-1) in the non-glacial creek 2.

53

The trends of HCO3- and SO42- are important to look at using a smaller scale since
they are important in the chemical weathering processes beneath the glacial ice. The
HCO3- concentrations in the glacial streams start with low concentrations and in general
slowly decrease over the summer months (see figure 24). In contrast, NGC1 starts with
low concentrations that increase over time. NGC2 displays higher concentrations of
HCO3- than any of the other glacial or non-glacial creeks. Both glacial streams and the
NGC1 have HCO3- concentrations ranging from 68.3 to 274.8 µeq L-1, while NGC2 has
HCO3- concentrations ranging from 668.8 to 714.7 µeq L-1.

Bicarbonate Concentrations
800
700

HCO3- (µeq/L)

600
LWR

500

UWR

400

LIF

300

UIF

200

NGC1

100

NGC2

0
5.21.16 6.4.16 6.18.16 7.9.16 7.23.16 8.12.16 8.27.16 9.10.16

Date of Collection

Figure 24: HCO3- concentrations (µeq L-1) in the glacial and non-glacial streams
over the field season.

Using a smaller scale to look at SO42- concentrations over time the differences
between the glacial creeks and the non-glacial creek concentrations can be observed. The
SO42- concentrations in all the glacial creek sites peaked in mid-June with the highest
concentrations ranging from 139.8 to 219.5 µeq L-1 (see figure 25). The highest
54

concentration of 219.5 µeq L-1 is from the UWR river site which is located closest to the
terminus of Emmon’s Glacier. The peak in sulfate levels in the glacial streams generally
coincides with the peak DOC concentrations (see figure 26 and 27). In contrast, the SO42concentrations in NGC1 started low and dropped lower when the glacial creeks were
experiencing the peak concentrations. The non-glacial creeks had SO42- concentrations
ranging from 10.1 to 64.8 µeq L-1. There were no significant relationships between DOC
and sulfate when a simple linear regression was ran comparing the two variables.

250

Sulfate Concentrations in Stream Water

Sulfate (µeq/L)

200
UWR
LWR

150

UIF
100

LIF
NGC1

50

NGC2

0
5.21.16 6.4.16 6.18.16 7.9.16 7.23.16 8.13.16 8.27.16 9.10.16
Date of Collection

NOWData sets
Figure 25: Concentrations of SO42- (µeq L-1) for the glacial and non-glacial
streams over the field season.

55

6

10

5

8

4

6

3

4

2

2

1

Sulfate (mg/L)

12

0

DOC (mg/L)

White River

UWR Sulfate
LWR Sulfate
UWR DOC
LWR DOC

0
5.21.16

6.18.16

7.23.16

8.27.16

Date of Collection

Figure 26: Visual comparison of SO42- (µeq L-1) and DOC (mg L-1) at both White
River sites over the field season.

8

4

7

3.5

6

3

5

2.5

4

2

3

1.5

2

1

1

0.5

0

0
5.21.16

6.18.16

7.23.16

DOC (mg/L)

Sulfate (mg/L)

Inter Fork River

UIF Sulfate
LIF Sulfate
UIF DOC
LIF DOC

8.27.16

Date of Collection

Figure 27: Visual comparison of SO42- (µeq L-1) and DOC (mg L-1) at both Inter
Fork River sites over the field season.

56

A general understanding of the chemical weathering occurring beneath the
glacial ice can be inferred using associations of the major ions in proglacial stream water
following procedures similar to those found in Wadham et al., 2010 and Bhatia et al.,
2013. This method is typically conducted on borehole meltwaters. That said, this data
may not be completely accurate since the samples were gathered after the water exits the
subglacial environment and mixes with englacial and supraglacial waters as well as
coming in contact with the atmosphere. Research comparing borehole waters and
proglacial stream waters has shown that ion concentrations in boreholes are much higher
than that of the proglacial streams but proglacial stream water samples can still be used to
get an overall idea of the chemical weathering patterns occurring subglacially (Brown,
2002).
One ion association, the C-ratio ratio [HCO3-/(HCO3- + SO42-)] (see figure 28),
can provide an understanding of the main processes driving chemical weathering beneath
the ice which are carbonate dissolution and sulfide oxidation. The UWR had a generally
consistent C-ratio throughout the summer, ranging from 0.37 to 0.54. The LWR
displayed similar trends to the UWR, with a C-ratio ranging from 0.37 to 0.61 as the
summer progressed. The UIF site had a C-ratio ranging from 0.49 to 0.77, with the
highest C-ratio’s occurring in late spring. The LIF site showed similar trends to the UIF
site with a C-ratio ranging from 0.51 to 0.86 and the highest values occurring in late
spring. The first non-glacial creek had the highest values for the C-ratio, ranging from
0.73 to 0.96, while the second non-glacial creek had values ranging from 0.91 to 1.

57

C-Ratio
1.2
1

C-ratio

LWR
0.8

UWR

0.6

LIF
UIF

0.4

NGC1
NGC2

0.2
0
5.21.16 6.4.16 6.18.16 7.9.16 7.23.16 8.12.16 8.27.16 9.10.16
Date of Collection

Figure 28: C-ratio values for the White River, the Inter Fork White River and the nonglacial creeks over the field season.
The last ion association utilized was plotting SO4- against HCO3- which has been
used in prior research to gather evidence of microbial activity in the subglacial
environment (see figure x). The regression equation of the association between SO4- and
HCO3- should result in a y-intercept of 220 µeq L-1 based on the theoretical solubility of
calcite in meltwater at 0°C (Raiswell, 1984). Y-intercepts larger than this indicates that
there must be an additional source of CO2 to account for the higher concentrations of
HCO3- if the drainage system is closed to the atmosphere (Bhatia et al., 2013; Raiswell,
1984; Wadham et al., 2010). The association of SO4- and HCO3- in the UWR resulted in a
p = 0.001, R2 = 0.86 and a y-intercept of 34.6 µeq L-1 (see figure 29). The LWR had a p =
0.03, R2 = 0.57 and a y-intercept of 25.4 µeq L-1 (see figure 30). while the LIF had a p =
0.03, R2 = 0.57 with a y-intercept of 266.1 µeq L-1(see figure 32). The results of the plot
for SO4- against HCO3- at the UIF site were not significant (see figure 31).

58

Upper White River
200

y = 0.6348x + 34.597
R² = 0.8573

180

HCO3- (µeq/L)

160
140

p-value = 0.001

120
100
80
60
40
20
0
0

50

100

150

200

250

Sulfate (µeq/L)

Figure 29: The comparisons of HCO3_ (µeq L-1) and SO42- (µeq L-1) for the upper
White River site.

Lower White River
300

p = 0.03

HCO3- (µeq/L)

250

y = 0.9723x + 25.404
R² = 0.5684

200
150
100
50
0
0

50

100

150

200

250

Sulfate (µeq/L)

Figure 30: The comparisons of HCO3_ (µeq L-1) and SO42- (µeq L-1) for the lower
White River site.

59

Upper Inter Fork River

250

HCO3- (µeq/L)

200
150
100
y = -0.3874x + 201.47
R² = 0.2186

50
0
0

20

40

60

80

100

120

140

160

Sulfate (µeq/L)

Figure 31: The comparisons of HCO3_ (µeq L-1) and SO42- (µeq L-1) for the upper
Inter Fork White River site.

Lower Inter Fork River
300

HCO3- (µeq/L)

250
200
150
y = -1.0331x + 266.08
R² = 0.5662

100
50

p = 0.03

0
0

20

40

60

80

100

120

140

160

Sulfate (µeq/L)

Figure 32: The comparisons of HCO3_ (µeq L-1) and SO42- (µeq L-1) for the lower
Inter Fork White River site.

60

5. Discussion
5.1 Temporal and spatial patterns of DOC export
Current DOC export from the two glacial streams studied on Mount Rainier show
similar temporal patterns over the melt season. DOC concentrations displayed the most
variability during May and June, with the peak concentrations (1.5 to 4.9 mg L-1)
occurring in early to mid-June. After the peak in late spring, DOC concentrations in all
the glacial streams decreased and leveled off to < 1 mg L-1 in July. The DOC
concentrations stayed low through mid-September when the melt season ends and snow
begins to accumulate for the winter. This DOC data is similar to results found in other
studies, which generally found that DOC concentrations in glacial meltwater ranges from
about 0.1 to 4.0 mg L-1 with peak concentrations occurring early in the melt season
(Bhatia et al., 2013; Fellman et al., 2014; Hood et al., 2009; Lafreniere & Sharp, 2004).
The UWR and UIF sites are likely to have the most influence from glacial
meltwater since these sites are located very close to the glaciers’ terminus. These sites are
less likely to have DOC inputs from other sources, like vegetation, soils, non-glacial
streams or groundwater intrusion (Fellman et al., 2014; Hood & Scott, 2008). The LWR
had the highest DOC concentration of all the glacial stream sites with a concentration of
4.9 mg L-1 C. The LWR site is located the furthest downstream from either glacial
terminus. Multiple non-glacial streams tributaries, and the Inter Fork White River have
merged into the White River at this location. The LWR site is also below the tree line, so
it seems that DOC from any terrestrial sources are contributing to this higher
concentration. The UWR site, which is close to the terminus of Emmon’s Glacier, has

61

much lower DOC values during the peak export in comparison to the LWR, with a
concentration of 1.5 mg L-1 C. The lower peak DOC concentration is more representative
of the proglacial waters that are most influenced by the glacial environment and least
influenced by terrestrial carbon sources, which is expected.
The UIF and LIF sites had much more similar values during the peak export, with
concentrations of 3.3 mg L-1 C and 3.4 mg L-1 C respectively. These sites were both
located in areas that were in the tree line, so were not only receiving inputs of carbon
from the glacial environment but from terrestrial sources as well. The UIF site should
have less influence from terrestrial sources, because it is closer to the Inter Glacier’s
terminus.
By July, DOC concentrations at all the glacial stream sites decreased and
displayed similar concentrations throughout the rest of the summer. All sites had
generally low concentrations, ranging from 0.6 mg L-1 C to 1.0 mg L-1 C. The non-glacial
creek showed a much flashier pattern of DOC export, with high peaks observed in midMay and in mid-June. By mid-July this non-glacial stream dried up. This indicates the
importance of the glacial meltwater during the warm, dry months of summer. The glacial
streams provide a consistent supply of cool, freshwater and DOC to downstream aquatic
ecosystems. In contrast, the non-glacial streams that rely on precipitation often dry up
and no longer provide freshwater or DOC to downstream aquatic environments. With
recent studies showing that the DOC from glacial environments is more labile in quality
compared to other DOC sources (Hood et al., 2009; Hood et al., 2015) this DOC input
from glacial meltwater is important to proglacial aquatic ecosystems, even if it is being
exported in small concentrations. A study completed by Fellman et al., (2015)
62

demonstrated that glacially derived DOC is utilized throughout the food web from
biofilm to invertebrates and different fish species. This illustrates how important this
source of DOC is to these aquatic environments.
At the proglacial stream sites on Mount Rainier, peak DOC concentrations
occurred during the period of snow melt and lower concentrations occurred later in the
season when glacial melt dominated. One theory is that the peak DOC concentrations in
the glacial streams during spring is partly due to the flushing of open cryoconite holes on
the surface of the glacier as the snow melts. Based on the significant results of simple
linear regression plots of DOC concentrations against snowmelt (p = < 0.05) this seems
like a possible theory. When there is still snow on top of the glacial ice, the snow melt
permeates through the remaining snow and firn cover until it reaches the ice surface
where it either travels across the ice and exits from the supraglacial environment into a
proglacial environment or it travels to the subglacial environment through crevasses in
the ice (Fountain & Walder, 1998; Lafreniere & Sharp, 2004). This process of the melt
water traveling across the ice surface and exiting the supraglacial environment is one that
could flush any exposed cryoconite holes early in the melt season. This snowmelt can
also wash terrestrial carbon into streams from plants and soils, so this may also be a
source of DOC resulting in the peak DOC concentrations. Since the UWR site and the
UIF site are closer to the terminus of the glaciers the terrestrial carbon sources should not
be as influential as at the lower glacial river sites.
A second theory is based on the coinciding peaks of the DOC and SO42concentrations in the glacial streams during late spring. Studies have demonstrated that
snow and supraglacial water generally have very low levels or no SO42- present compared
63

to subglacial waters that become enriched with SO42- (Anderson, Longacre and Krall,
2003; Mitchell et al., 2013). The acquisition of SO42- into meltwater occurs when water in
the subglacial environment comes in contact with the rock flour generated by the
movement of the glacier (Brown, 2002; Mitchell et al., 2013; Raiswell, 1984). This rock
flour contains fresh surfaces of the mineral pyrite and these fresh mineral surfaces
quickly react with O2 and H2O, which results in the addition of SO42- into the subglacial
meltwater (Brown, 2002; Raiswell, 1984).
Since there is a peak in SO42- concentrations in late spring, it seems to indicate
that meltwater may be originating from the subglacial environment to produce this peak.
This could be the point in the melt season when enough of the glacier’s ice surface is
exposed and crevasses begin to form, causing more meltwater to flow to the subglacial
environment, which begins the evolution of the drainage system from a distributed, slow
flowing system to a channelized, quicker flowing drainage system (Fountain & Walder,
1998). As this drainage system transitions due to the influx of meltwater in the subglacial
environment, the flow in the subglacial drainage system increases, which could result in
the peak SO42- concentrations. If the ice on Emmon’s glacier at lower elevations was
snow free during this time of the year it would provide conditions for the evolution of the
drainage system as described above. Since DOC concentrations peak around the same
time as the SO42- concentrations, that could indicate that some of the DOC is from carbon
pools beneath the glacier.
Based on the significant p-values (p = < 0.05) for the plots of velocity against
DOC concentrations at the UWR site and the UIF site, it seems higher DOC
concentrations occur with lower velocities of the stream water. Velocities in the White
64

River and Inter Fork White River were the highest mid-June through the end of August
when the DOC concentrations were the lowest. This could indicate that the higher
velocities may be diluting the DOC and sulfate concentrations in the meltwater. By
gathering future data on meltwater from bore holes drilled through the glacial ice, a more
accurate understanding of these concentrations during the summer months is possible.
The Inter Fork White River sites are more complicated than the White River sites
because they are located in the forest and have a tree canopy over the stream. This means
that there will be carbon from terrestrial sources influencing the DOC concentrations in
these stream samples. To tease out the contribution of each different source of carbon in
glacial streams, future research will need to focus on δ 13C, ∆14C analyses and
radiocarbon dating. This allows researchers to understand what portion of the DOC is
likely from glacially derived sources versus terrestrial sources so a more intricate data set
of glacial DOC export over the melt season can be gathered. All the glacial stream sites
display a similar temporal pattern, so although data on the DOC sources can get more
detailed with future research, this current data still shows the general trend of DOC
export from glacial meltwater. When compared to the temporal pattern of DOC
concentrations in the non-glacial creeks, the glacial streams display a different DOC
export pattern.
To understand what factors were most important in controlling DOC export from
glacial environments, the relationships between DOC export and multiple parameters
were examined. Plots of DOC concentration against snow depth, snowmelt, and velocity
all had significant p-values as well as relatively high R2 values, although these values
varied from site to site. DOC concentrations plotted against temperature and precipitation
65

had less significant p-values overall compared to the other parameters. One explanation
for the varying p-values is the lower sites on the White River and Inter Fork White River
are influenced by the mixing of non-glacial streams and terrestrial DOC inputs. These
confounding factors could obscure the data since the influence of glacial melt on DOC
export would be diluted by other sources of DOC. Overall, it seems that all parameters
have an influence on the DOC export in the glacial bulk meltwaters, with snow depth,
snowmelt and velocity having the strongest influence on DOC export.

5.2 Chemical weathering patterns in the subglacial environment
The major ions that are exported by glacial melt water during the summer months
gives can provide a general insight into the chemical weathering that is occurring in the
subglacial environment. They also can provide evidence for the presence or absence
microbial communities beneath the ice. The cation concentrations found in the White
River and the Inter Fork White River were generally high compared to other studies done
on glacial meltwater (Bhatia et al., 2013; Brown, 2002; Singh et al., 2014). The anion
concentrations were generally on the low end in comparison to other studies done on
glacial meltwater (Brown, 2002; Singh et al., 2014; Wadham et al., 2010). These ion
concentrations are reasonable based on the composition of the andesitic lava beneath the
glaciers. Andesite is composed of ~50% SiO2 (Schuman, 1993) The major mineral
assemblages include plagioclase, pyroxenes, hornblende, with possible minor mineral
assemblages of biotite, magnetite, apatite, zircon, ilmenite and garnet (Schuman,1993).
Researcher have also found pyrite minerals present in volcanic rock on Mount Rainier,

66

which provides a source of sulphate for chemical weathering processes (John et al.,
2008).
Using the associations of different ion concentrations in the meltwater, a general
understanding of the chemical weathering processes occurring in the subglacial drainage
system of the Emmon’s Glacier and the Inter Glacier can be gathered. Here I discuss the
results of these major ion concentrations from the upper White River site and the upper
Inter Fork site. These sites were located closest to the glacial terminus, so the results from
these sites represent the bulk glacial meltwaters before they are complexed by other water
sources or weathering regimes.
Upper White River
Acid hydrolysis is the most important weathering reaction occurring beneath the
ice (Brown, 2002; Raiswell, 1984). This reaction occurs as protons (H+) drive acid
hydrolysis and weather ions out of the rocks or rock flour, which is crushed up rock
particles from glacial erosion (Sharp, Richards & Tranter, 1998). The dissolution and
dissociation of atmospheric CO2 (carbonate dissolution), and sulfide oxidation are the
main reactions that produce H+ protons in the subglacial environments (Brown, 2002;
Raiswell, 1984). In general, silicates, aluminosilicates and carbonates dominate in the
rocks found beneath glaciers, with accessory pyrite minerals sometimes present that
represent the source of sulphate in meltwater (Raiswell,1984). By using the C-ratio
[HCO3-/(HCO3- + SO42-)] the dominate source of protons driving the acid hydrolysis can
be determined (Brown, 2002; Raiswell, 1984).

67

The C-ratio calculated for the White River sites revealed a value around 0.5
throughout the entire glacial melt season. This value indicates that there is coupled
sulphate oxidation and carbonate dissolution occurring in the subglacial environment of
Emmons glacier. The C-ratio indicates that the source of the protons beneath Emmon’s
Glacier are derived from the sulfide oxidation and the dissolution and dissociation of
atmospheric CO2.
pCO2 values for the UWR are consistently below 300 ppm throughout the field
season. This could indicate the subglacial drainage system was out of contact with the
atmosphere (~409 ppm) (NOAA, 2017) and although my sites were in contact with the
atmosphere, equilibration of the CO2 had not occurred yet (Brown, 2002). The low pCO2
values are a common occurrence when a fresh source of water, like snowmelt, comes in
contact with a large amount of newly ground rock flour (Brown, 2002). The pCO2 values
at the LWR site are below 400 ppm except for a peak from early July to early August that
reaches 595 ppm. High pCO2 values can also indicate a closed drainage system that is
receiving protons faster than they are used for chemical weathering, which could be the
result of snowmelt or an increase in sulfide oxidation (Brown, 2002). The LWR site is
further downstream from the glaciers terminus than the UWR site, so it has likely been
more heavily influenced by contact with the atmosphere. Looking at the UWR site which
is closer to the terminus of the glacier, the pCO2 levels provide evidence for a closed
drainage system beneath the glacier.
By plotting HCO3- versus SO42- the resulting regression line can give insight into
any potential presence of microbial communities beneath the ice. When subglacial flour
comes in contact with meltwater, carbonate hydrolysis occurs due to the high reactivity of
68

the rock flour (Bhatia et al., 2013; Wadham et al., 2010). The carbonate dissolution
reaction results in the production of HCO3- at a rate quicker than SO42- is produced, so
there HCO3- concentrations will increase in the meltwater before SO42- concentrations
begin increasing (Bhatia et al., 2013; Tranter et al., 2002; Wadham et al., 2010). Based on
the theoretical solubility of calcite in meltwater at 0°C, this reaction should result in
around 220 µeq L-1 of HCO3- in the water. Y-intercepts larger than this indicates that if
the subglacial drainage system is closed to the atmosphere, there must be an additional
source of CO2 to account for the higher HCO3- concentrations in the meltwaters. (Bhatia
et al., 2013; Brown, 2002; Wadham et al., 2010). Some studies reason that y-intercepts
higher than 220 µeq L-1 suggest microbial communities are present in the subglacial
environment that oxidize organic carbon generating additional CO2, which produces more
protons that drive the chemical weathering processes (Bhatia et al., 2013; Tranter et al.,
2002; Wadham et al., 2010). The plot of HCO3- versus SO42- from the UWR resulted in a
regression line with a y-intercept < 220 µeq L-1. This data does not provide evidence to
suggest that there are microbial communities producing additional CO2 beneath Emmon’s
Glacier.
These results of the plots are likely skewed due to the fact that they were analyzed
from bulk meltwater samples, which have come in contact with atmospheric CO2
although at some sites it seemed equilibration of the CO2 had not occurred yet. Future
research should focus on gathering meltwater samples from boreholes in the glacier. This
would provide more accurate data on the subject since boreholes allow researchers to
gather water samples that are out of direct contact with the atmosphere. Studies done
comparing borehole waters to bulk meltwaters indicate there is a substantial difference in
69

data depending on where you gather the glacial waters from (Brown, 2002; Bhatia et al.,
2013; Tranter et al., 2002; Wadham et al., 2010). Although there is no evidence from this
data suggesting there are microbial communities in the subglacial environment of
Emmons Glacier, there are still consistent concentrations of DOC supplied to the White
River throughout the melt season. Although the sources of the DOC cannot be fully
understood without further research, some of the DOC is likely from the subglacial
environment where soils and vegetation were overrun by the ice during the last ice age.
Even if microbial communities are not altering this carbon source, the subglacial
environment is likely still be a source of DOC to downstream ecosystems.
Upper Inter Fork White River
The C-ratio in the Inter Fork White River samples was around 0.8 early in the
season and then evolved to around a 0.55 in mid-June to July and stayed fairly consistent
throughout the rest of the summer months. These values indicate that carbonate reactions,
or the dissolution and dissociation of atmospheric CO2, are the dominate source of
protons beneath the Inter Glacier early in the melt season. The chemical weathering
reactions shift from being dominated by carbonate dissolution to coupled sulfide
oxidation and carbonate dissolution by mid-June and this trend continues for the rest of
the melt season.
pCO2 values for the UIF had much greater variation than the UWR. The values
ranged from 289.5 to 618.9 ppm. The LIF displayed values similar to the UIF with values
ranging from 266.3 to 833.7 ppm. Both sites displayed the lowest pCO2 values in midJune similar to all the other glacial and non-glacial sites. With the UIF closer to the Inter

70

Glacier’s terminus, it displays the trends most influenced by the glacial meltwater.
Overall the UIF displays values lower than the values of the LIF site, likely because the
water at the LIF site had more of an influence from contact with the atmosphere and has
had time for the CO2 in the stream to reach equilibrium.
The plot of HCO3- versus SO42- from the UIF site resulted in y-intercept < 220
µeq L-1. This result is similar to the results obtained for the UWR and they do not provide
evidence for microbial activity in the subglacial environment. Again, these results are
likely skewed due to the water samples being in contact with atmospheric CO2. The UIF
site may also have slightly differing results than the UWR site because it is located in an
area with tree cover. Overall, this data does not provide evidence for microbial activity in
the subglacial environment of the Inter Glacier.
There is still a steady flux of DOC to the Inter Fork River throughout the summer
with a peak in the late spring, indicating there is a source of DOC in this area consistently
providing DOC to the stream. The DOC concentrations from the bulk meltwaters of the
Inter Fork White River does displays trends similar to the DOC concentrations from the
White River, but does not display trends similar to the non-glacial creek. This suggests
the data gathered for the glacial streams are displaying the general trends of DOC
exported from glacial meltwater. As mentioned previously, future research could gather a
more detailed understanding of the proportions of glacially derived carbon sources and
terrestrial carbon sources by analyzing the δ13C, ∆14 C and doing radiocarbon dating. The
results from the UIF and UWR demonstrate that DOC export occurs throughout the
glacial melt season to proglacial streams, whether this carbon source is glacially derived
or from terrestrial sources.
71

5.3 The effects of glacial area on DOC and major ion export: The implications of
continued glacial recession and climate change
This study gives insight into how DOC concentrations and chemical weathering
regimes differ depending on the glacial area at the headwaters. Emmon’s Glacier is the
largest glacier on Mount Rainer (11.1 km2), the Inter Glacier is one of the smallest
glaciers (0.8 km2) and the non-glacial creek is fed by lake waters and precipitation.
Although stream water that is fed by lake waters may not be the ideal in some
circumstances as a reference, it can be more effective when looking at changes between
streams with differing glacial area at the headwaters. This is because as glaciers recede,
they often leave behind large chunks of ice, which melt into lakes in these alpine
environments. By comparing data between these three sites it can give insight into the
changes that may occur as glaciers in the region continue to recede. I only used the data
from the UWR and the UIF to compare to one another and with the non-glacial creek
since they represent the purest glacial bulk meltwater.
The DOC peak concentrations were lowest in the White River (Emmon’s
Glacier), with higher peak concentrations in the Inter Fork White River (Inter Glacier)
and the highest concentrations in the non-glacial creek. Prior research suggests that DOC
sources may change as glacial recession results in less glacial area for cryoconite holes to
form and less subglacial area for microbial communities to mobilize the organic carbon
left beneath the ice (Fellman et al., 2014; Hood & Scott 2008). Instead more terrestrial
sources of organic carbon will be input into the streams and rivers as terrestrial

72

succession follows glacial retreat, which results in higher concentrations of DOC. Not
only do the sources of carbon shift, but the timing of the peak DOC export will shift as
factors like snowmelt and precipitation patterns shift in timing (Hood & Scott, 2008;
Hood et al., 2009). Models on snowmelt and precipitation done by the IPCC (2014)
indicate that snowmelt will happen earlier in the season and precipitation will shift from
snow to rain earlier in the season. The likely change in the timing of peak DOC
concentrations could have an effect on the heterotrophic communities that rely on this
DOC source (Fellman et al., 2015; Hood & Scott, 2008). An indication of this possible
shift in the peak DOC concentrations is the non-glacial creek displayed the highest DOC
concentrations earlier in the spring than the glacial streams. The current timing of the
peak DOC export in the study, will likely occur earlier in the calendar year in coming
decades since snowmelt and snow depth seemed to play a role in DOC export. The
temporal trends of DOC export from meltwater needs to be monitored in the coming
decades to really gather firm evidence of how these patterns may shift.
Since the DOC from glacial environments has been found to be more labile in
nature than DOC from other sources, this shift in available labile DOC to the downstream
aquatic environments could result in a shift in the structure of the aquatic food web
(Hood & Scott, 2008; Hood et al., 2009,). If glaciers end up disappearing altogether, this
would result in a shift from stream environments receiving predominately bioavailable
glacially derived DOC to receiving predominately terrestrial DOC, which would likely
result in a shift in the local aquatic food webs (Fellman et al., 2014). That is because the
bioavailable glacially derived DOC stimulates production in the heterotrophic
communities which make up the base of food webs (Fellman et al., 2014; Hood & Scott,
73

2008). Without any input of glacial DOC, the heterotrophic communities would likely
change in structure with effects rippling through the aquatic ecosystem.
Not only will the recession of glaciers result in a decreasing export of glaciallyderived DOC to downstream environments, it will also result in a decreasing supply of
fresh, cool water to these streams and rivers. During the hot, dry summer months, glacial
meltwater supplies streams with a constant supply of water. The non-glacial streams and
creeks that are reliant on precipitation events or lake water often dry up part way through
the summer, like what occurred to NGC1 in this study. There was a notable difference
between the discharge of the White River, the Inter Fork White River and the non-glacial
creek based on physical observation in the field. When glaciers are receding, the
discharge has been shown to initially increase (Fountain & Tangborn, 1985) and then it
begins to decrease until it becomes a non-glacial stream reliant on precipitation. This
change in water input would likely have an impact on downstream aquatic ecosystems. In
the Pacific Northwest, salmon species rely on high flows to reach their spawning grounds
and cool temperatures to induce spawning. Without the freshwater input from glaciers,
species like the salmon would be negatively impacted by the disappearing glaciers.
Looking at the C-ratio, it seems likely that as the glaciers recede a shift in
weathering regimes may occur. The C-ratio in the spring at the UWR (Emmon’s Glacier)
was ~ 0.5, while the C-ratio at the UIF (Inter Glacier) and the NGC1 was ~0.8. This
indicates there could possibly be a shift from coupled sulfide oxidation and carbonate
dissolution producing protons for chemical weathering to primarily carbonate dissolution
producing the protons for chemical weathering as glacier size decreases. This could be
likely due to the glaciers covering less rock surface as they recede causing less chemical
74

weathering from the rock-water-ice contact. Instead, succession of vegetation in the area
where glacial ice once covered rock will develop more soils for the vegetation and result
in a different chemical weathering regime more indicative of the soils properties. This
could also be due to differences in the underlying geology. Overall, this data
demonstrates that there are differences in the concentrations of DOC and major ions in
streams with differing glacial area at the headwaters. By monitoring this data into the
future, a more concise understanding of the changes that may occur with receding
glaciers can be gathered.

7. Conclusion
Glacier meltwaters from Mount Rainier are an important source of DOC to local
proglacial environments. Both glaciers studied on Mount Rainier showed similar
temporal patterns of DOC export compared to nearby non-glacial creeks indicating
glaciers have unique properties that influence this export. By comparing glacial streams
with different glacial coverage at the headwaters, this study provides insight into how
DOC and major ion export may change as glaciers recede. Taking into consideration
future climate scenarios it seems that not only will the DOC concentrations change as
glaciers recede, but the timing of the peak DOC concentrations will likely shift in time
and occur earlier in the season. Changes in DOC export from glacial environments are
liable to cause changes to the aquatic food web. This could be problematic for important
river organisms, including salmon and trout which are an economically and traditionally
important species in the Pacific Northwest. Future research using stable isotope analysis

75

and radiocarbon analysis would allow for a more detailed understanding of the DOC
sources and ages found in glacial watersheds on Mount Rainier. These techniques can
also be used to track glacially derived carbon through the food web. Understanding if
important fish species in our region are in any way reliant on this glacial carbon is crucial
in keeping these fish populations strong.
When ancient DOC in the subglacial environment is mobilized it can stimulate
heterotrophic communities in the glacial streams (Hood et al., 2009; Lawson et al., 2014;
Spencer et al., 2014). Finding evidence of microbes in the subglacial environment by
using these ion associations may suggest that they are helping mobilize some of the
ancient DOC. The associations of the major ions found in the bulk meltwaters on Mount
Rainier did not provide evidence for microbial communities beneath the ice. A more
detailed study is needed in order to understand if there are microbial communities in the
subglacial drainage system

`.

The chemical weathering pattern in the subglacial environments indicates that the
current chemical weathering regime beneath Emmon’s Glacier and Inter Glacier is
generally coupled carbonate dissolution and sulfide oxidation. The slight differences in
the C-ratio early in the melt season imply that the source of the meltwaters may come
from different regions beneath the ice, so microbial communities may be present in some
regions beneath the ice but not others warranting the need for more research. A study by
Wadham et al. (2010) indicated that harder bed rock (high silica content) beneath glaciers
provide less substrate for microbes than soft bedrock (sedimentary rocks). This may be
the case on Mount Rainier since the glaciers are located on top of Andesite lava flows
which are high in silica content. Future research should use borehole investigations to get
76

a more accurate idea of the DOC and major ion concentrations in the subglacial
environments. The bore hole investigations drill a hole through the glacier to gather a
water sample from beneath the ice. By doing this, complicating factors like exposure to
atmospheric CO2 can be eliminated.
This study provides data on the current spatial and temporal trends of DOC and
major ion export in glacial streams on Mount Rainier. This data can provide insight into
the local glacial chemical weathering regimes and the changes that may occur to these
patterns as climate change causes the glaciers to recede. This data is beneficial for any
researchers interested in continuing to monitor carbon export from glaciers in the Pacific
Northwest. It would also be valuable to natural resource managers or policy makers to
use when making decisions about protecting our natural resources, like salmon or trout.
Understanding glacial dynamics is complicated, yet important on a global and local scale.
Ice sheets and mountain glaciers are contributing labile DOC to proglacial coastal and
freshwater environments. As these ice sheets and mountain glaciers recede proglacial
aquatic environments will all likely undergo shifts to the local food webs. This study has
provided a first glimpse into the local trends in glacial carbon cycling and discussed the
implications of the recession of the glaciers on Mount Rainier which will continue as the
climate continues to change.

77

References

Anderson, S. P., Longacre, S. A., & Kraal, E. R. (2003). Patterns of water chemistry
and discharge in glacier-fed Kennicott River, Alaska: Evidence for subglacial
water storage cycles. Chemical Geology, 202, 291-312.
Anesio, A. M., Hodson, A.J., Fritz, A., Psenner, R., & Sattler, B. (2009). High
microbial activity on glaciers: importance to the global carbon cycle. Global
Change, 15 (4), 955-960.
Anesio, A. M., Sattler, B., Foreman, C., Telling, J., Hodson, A., Tranter, M., &
Psenne, R. (2010). Carbon fluxes through bacterial communities on glacier
surfaces. Annals of Glaciology, 51 (56), 32-40.
Barker, J. D., Sharp, M. J., Fitzsimons, S. J., & Turner, R. J. (2006). Abundance and
dynamics of dissolved organic carbon in glacier systems. Arctic, Antarctic,
and Alpine Research, 38(2), 163–172.
Bennett, M., & Glasser, N.F. (1996). Glacial geology: ice sheets and landforms.
Chichester, England: John Wiley.
Berner, E. K., & Berner, R. A. (2012). Global environment: Water, air and
geochemical cycles. Princeton, New Jersey: Princeton University Press,
Bhatia, M. P., Das, S. B., Xu, L., Charette, M. A., Wadham, J. L., & Kujawinski, E.
B. (2013). Organic carbon export from the Greenland ice sheet. Geochimica et
Cosmochimica Acta , 109, 329-344.
Brown, Giles H. (2002). Glacier meltwater hydrochemistry. Applied Geochemistry.
17, 855-883.
Collins, D. (1979). Hydrochemistry of meltwaters draining from an alpine glacier.
Arctic and Alpine Research, 11(3). DOI: 10.2307/1550419.
Fellman, J. B., Hood, E., Spencer, R. G. M., Stubbins, A., & Raymond, P. A. (2014).
Watershed glacier coverage influences dissolved organic matter
biogeochemistry in coastal watersheds of southeast Alaska. Ecosystems. DOI
10.1007/s10021-014-9777-1.
Fellman, J. B., Hood, E., Raymond, P. A., Hudson, J., Bozeman, M., & Arimitsu, M.
(2015). Evidence for the assimilation of ancient glacier organic carbon in a
proglacial food web. Limnology and Oceanography 1-11. Doi:
10.1002/lno.10088.
78

Fountain, A. G., & Tangborn, W. V. (1985). The effect of glaciers on streamflow
variation. Water Resources Research. 21(4), 579-586.
Fountain, A. G., & Walder, J. S. (1998). Water flow through temperate glaciers.
Reviews of Geophyasics, 36(3), 299-328.
Gleick, P. H. (1996) Water resources. In S. H. Schneider (Ed). Encyclopedia of
Climate and Weather (pp. 817-823). New York: Oxford University Press.
Granshaw, F. D., & Fountain, A. G. (2006). Glacier change (1958–1998) in the north
cascades national park complex, Washington, USA. Journal of Glaciology,
52(177), 251-256.
Hasnain, S.I. (1999a). Himalayan glaciers: hydrology and hydrochemistry. New
Delhi: Allied Publishers (P) Ltd.
Hasnain, S. I., & Thayyen, R. J. (1999b). Controls on the major-ion chemistry of the
Dokriani glacier. Journal of Glaciology, 45(149), 87–92.
Hekkers, M., & Thorneycroft, K. (2011). Glaciers of Washington: Glaciers of the
American West. Retrieved from http://glaciers.us/glaciers-washington.
Hindshaw, R. S., Tipper, E. T., Reynolds, B. C., Lemarchand, E., Wiederhold, J. G.,
Magnusson, J., & Bourdon, B. (2011). Hydrological control of stream water
chemistry in a glacial catchment (Damma Glacier, Switzerland). Chemical
Geology, 285(1–4), 215–230. https://doi.org/10.1016/j.chemgeo.2011.04.012
Hood, E., & Scott, D. (2008). Riverine organic matter and nutrients in southeast
Alaska affected by glacial coverage. Nature Geoscience, 1(9), 583–587.
https://doi.org/10.1038/ngeo280
Hood, E., Fellman, J., Spencer, R. G., Hernes, P. J., Edwards, R., & D’Amore, D.
(2009). Glaciers as a Source of Ancient and Labile Organic Matter to the
Marine Environment. Nature, 462, 1044-1047. doi:10.1038/nature08580
Hood, E., Battin, T. J., Fellman, J., O'Neel, S., & Spencer, R. J. (2015). Storage and
release of organic carbon from glaciers and ice sheets. Nature Geosciences, 16. DOI: 10.1038/NGEO2331
IPCC, 2014: Climate Change 2014: Impacts, Adaptation, and Vulnerability. Part B:
Regional Aspects. Contribution of Working Group II to the Fifth Assessment
Report of the Intergovernmental Panel on Climate Change [Barros, V. R., C.
B. Field, D. J. Dokken, M. D. Mastrandrea, K. J. Mach, T. E. Bilir, M.
Chatterjee, K. L. Ebi, Y. O. Estrada, R. C. Genova, B. Girma, E. S. Kissel, A.
79

N. Levy, S. MacCracken, P.R. Mastrandrea, & L.L. White (eds.)]. Cambridge,
United Kingdom and New York, NY, USA: Cambridge University Press, pp.
688.
John, D. A., Sisson, T. W., Breit, G. N., Rye, R. O. & Vallance, J. W. (2008).
Characteristics, extent and origin of hydrothermal alteration at Mount Rainier
Volcano, Cascades Arc, USA: Implications for debris-flow hazards and
mineral deposits. Journal of Volcanology and Geothermal Research, 175,
289–314. doi:10.1016/j.jvolgeores.2008.04.004.
Lafreniere, M.J., & Sharp, M. J. (2004). The concentration and fluorescence of
dissolved organic carbon (DOC) in glacial and nonglacial catchments:
interpreting hydrological flow routing and DOC sources. Arctic, Antarctic,
and Alpine Research, 36,(2), 156–165.
Lawson, E. C., Wadham, J. L., Tranter, M., Stibal, M., Lis, G. P., Butler, C.E. H.,
Laybourn-Parry, J., Nienow, P., Chandler, D., & Dewsbury, P. (2014).
Greenland ice sheet exports labile organic carbon to the Arctic oceans.
Biogeoscience, 11, 4015-4028.
Lundquist, J., & Roche, J. (2009). Climate change and water supply in western
national parks. Park Science, 26(1), 31-34.
Mitchell, A.C., Lafrenière, M. J., Skidmore, M. L., & Boyd, E. S. (2013). Influence of
bedrock mineral composition on microbial diversity in a subglacial
environment. Geology, 41(8), 855–858.
Mote, P. W., & Salathe Jr, E. P. (2010). Future climate in the pacific northwest.
Climatic Change, Springer. DOI 10.1007/s10584-010-9848-z.
National Oceanic and Atmospheric Administration [NOAA]. (2016). NOWData NOAA Online Weather Data. Retrieved from
http://w2.weather.gov/climate/xmacis.php?wfo=sew.
National Park Service (NPS). (2005).Mount Rainier national park: geologic resource
evaluation report. Natural Resource Report. National Parks Service. U.S.
Department of the Interior. Retrieved from
https://www.nature.nps.gov/geology/inventory/publications/reports/mora_gre
_rpt_view.pdf.
National Park Service (NPS). (2007). Mount Rainier: the forest communities of
Mount Rainier national park. Park History. National Parks Service. U.S.
Department of the Interior. Retrieved from
https://www.nps.gov/parkhistory/online_books/science/19/chap5.htm
80

National Park Service (NPS). (2016) Glaciers. Glacial Features. National Parks
Service. U.S. Department of the Interior. Retrieved from
http://www.nps.gov/noca/learn/nature/glaciers.htm.
Nylen, T. H. (2001). Spatial and temporal variations of glaciers (1913-1994) on Mt.
Rainier and the relation with climate. Retrieved from Portland State
University.
Porazinska, D. L, Fountain, A. G, Nylen, T. H., Tranter, M., Virginia, R. A., & Wall,
D. H. The biodiversity and biogeochemistry of cryoconite holes from
McMurdo Dry Valley glaciers, Antarctica Arctic, Antarctic, and Alpine
Research, 36(1), 84-91.
Raiswell, R. (1984). Chemical models of solute acquisition in glacial melt waters.
Journal of Glaciology, 30(104), 49-57.
Raymond, P. A., Hartmann, J., Lauerwald, R., Sobek, S., Mcdonald, C., & Hoover,
M. (2013). Global carbon dioxide emissions from inland waters. Nature, 503
(7476), 355-359.
Riedel, J., & Larrabee, M. A. (2011). North cascades national park complex glacier
mass balance monitoring annual report, water year 2009: North coast and
cascades network. National Park Service, Fort Collins, Colorado.
Schuman, W. (1993). Handbook of rocks, minerals & gemstones. Boston & New
York: HarperCollins Publishers & Houghton Mifflin Company.
Sharp, M, Richards, K. S., & Tranter, M. (1998). Glacier Hydrology and
Hydrochemistry. Chichester, England: John Wiley & Sons LTD.
Sharp, M., Parkes, J., Cragg, B., Fairchild, I. J., Lamb, H., & Tranter, M. (1999).
Widespread bacterial populations at glacier beds and their relationship to rock
weathering and carbon cycling. Geology, 27, 107-110.
Singer, G. A. (2012). Biogeochemically diverse organic matter in Alpine glaciers and
its downstream fate. Nature Geosciences. 5, 710–714.
Singh,V. B., Ramanathan, A., Pottackkai, J. G., Sharma, P., Linda, A., Azam, M. F.,
& Chatterjee, C. (2014). Chemical characterization of meltwater draining
from Gangotri Glacier, Garhwal Himalaya, India. Journal of Earth Sciences,
121(3), 625-636.

81

Spencer, R. J., Vermilyea, A., Fellman, J., Raymond, P., Stubbins, A., & Scott, D.
(2014). Seasonal variability of organic matter composition in an Alaskan
glacier outflow: insight into glacier carbon sources. Environmental Research
Letters, 9, 1-7.
Slemmons, K. E. H., Saros, J. E., & Simon, K. (2013). The influence of glacial
meltwater on alpine aquatic ecosystems: a review. Environmental Science:
Processes & Impacts, 15(10). https://doi.org/10.1039/c3em00243h
Skidmore, M. L., Foght, J. M., & Sharp, M. J. (2000). Microbial life beneath a high
Arctic glacier. Applied and Environmental Microbiology, 66(8), 3214–3220.
Stewart, I. T., Cayan, D. R & Dettinger, M. D. (2005). Changes toward earlier
streamflow timing across western north america. Journal of Climate, 18,
1136-1155.
Stibal, M., & Tranter, M. (2007). Laboratory investigation of inorganic carbon uptake
by cryoconite debris from Werenskioldbreen, Svalbard. Journal of
Geophysical Research Atmospheres, 112(G4).
Tarnocai, C., Canadell, J. G., Schuur, E. A. G., Kuhry, P., Mazhitova, G., & Zimov,
S. (2009). Soil organic carbon pools in the northern circumpolar permafrost
region. Global biogeochemical cycles, 23(GB2023).
doi:10.1029/2008GB003327,
Tranter, M., Brown, G. H., Hodson, A., & Gurnell, A.M. (1996). Hydrochemistry as
an indicator of subglacial drainage system structure: a comparison of Alpine
and Sub-Polar environments. Hydrological Processes, 10, 541–556.
Tranter, M., Sharp, M. J., Lamb, H. R., Brown, G. H., Hubbard, B. P., & Willis, I. C.
(2002). Geochemical weathering at the bed of Haut Glacier d’Arolla,
Switzerland—a new model. Hydrological Processes, 16, 959–993. DOI:
10.1002/hyp.309.
U.S Environmental Protection Agency (EPA). (1983). Sample preservation. Methods
for chemical analysis of water and wastes. Cincinnati, Ohio, U.S EPA.
Retrieved from http://sisbl.uga.edu/epatab1.html.
U.S Environmental Protection Agency (EPA). (1994). Method 200.8, Revision 5.4:
Determination of trace elements in waters and wastes by inductively coupled
plasma – mass spectrometry.

82

United States Geological Survey (USGS). (2014). Glaciers Help to Shape Mount
Rainier. Volcano Hazards Program: Mount Rainier. Retrievedfrom
https://volcanoes.usgs.gov/volcanoes/mount_rainier/geo_hist_glaciers.html.
United States Geologic Society (USGS). (2014). Geology and History Summary for
Mount Rainier. Mount Rainer. Retrieved from
https://volcanoes.usgs.gov/volcanoes/mount_rainier/geo_hist_summary.html.
United States Geologic Society (USGS). (2016). Ice, snow, and glaciers: the water
cycle. The Water Cycle - USGS Water Science School. Retrieved from
https://water.usgs.gov/edu/watercycleice.html.
Wadham, J. L., S. Bottrell, M. Tranter, and R. Raiswell (2004), Stable isotope
evidence for microbial sulphate reduction at the bed of a polythermal high
Arctic glacier, Earth Planet. Science Letters, 219, 341–355.
Wadham, J. L., Tranter, M., Skidmore, M., Hodson, A. J., Priscu, J., Lyons, W. B.,
Sharp, M., Wynn, P., & Jackson, M. (2010). Biogeochemical weathering
under ice: size matters. Global Biogeochemical Cycles, 24 (3025).
doi:10.1029/2009GB003688.
Wanninkhof, R. (1992), Relationship between wind speed and gas exchange over the
ocean, J. Geophysical Research, 97, 7373–7382, doi:10.1029/92JC00188.
Washington Department of Ecology (DOE). (2007). Facts about Washington’s
retreating glaciers and declining snow pack. Ecology’s Water Resource
Program. Retrieved from
https://fortress.wa.gov/ecy/publications/publications/0711016.pdf.
Washington Department of Fish and Wildlife (WDFW). (2008). Economic Analysis
of the Non-Treaty Commercial and Recreational Fisheries in Washington
State. Publications. Retrieved from http://wdfw.wa.gov/publications/00464/.
Washington Department of Fish and Wildlife (WDFW). (2016). Hatcheries.
Conservation. Retrieved from
http://wdfw.wa.gov/hatcheries/mass_marking.html.
Washington Department of Fish and Wildlife (WDFW). (2016). Recreational Salmon
Fishing. Fishing and Shellfishing. Retrieved from
http://wdfw.wa.gov/fishing/salmon/species.html.
Washington Department of Fish and Wildlife (WDFW). (2017). Conservation.
Species of Concern. Retrieved from
http://wdfw.wa.gov/conservation/endangered/All/#.
83

Weiss (1974). Carbon dioxide in water and seawater: The solubility of a non-ideal
gas. Marine Chemistry, 2, 203–215, doi:10.1016/0304-4203(74)90015-2.

84

85

86

Media

Wilmes_KMESthesis2017.pdf