-
Title
-
Eng
Effects of Climate and Land Use on Terrestrial Dissolved Organic Carbon Loading From Glacially Fed and Lowland Puget Sound River Basins
-
Date
-
2018
-
Creator
-
Eng
Cracknell, Paula
-
Subject
-
Eng
Environmental Studies
-
extracted text
-
EFFECTS OF CLIMATE AND LAND USE
ON TERRESTRIAL
DISSOLVED ORGANIC
CARBON LOADING FROM
GLACIALLY FED AND LOWLAND
PUGET SOUND RIVER BASINS
by
Paula Cracknell
A Thesis
Submitted in partial fulfillment
of the requirements for the degree
Master of Environmental Studies
The Evergreen State College
December 2017
©2017 by Paula Cracknell. All rights reserved.
This Thesis for the Master of Environmental Studies Degree
by
Paula Cracknell
has been approved for
The Evergreen State College
by
________________________
Erin Martin PhD
Member of the Faculty
________________________
Date
ABSTRACT
Effects of Climate and Land use on Terrestrial Dissolved Organic Carbon Loading from
Glacially Fed and Lowland Puget Sound Rivers
Paula Cracknell
This study aims to compare how climate change and dominant land cover affect
the quantity of dissolved organic carbon (DOC) loading into the South Puget Sound
estuary in Washington State. DOC concentrations and flow measurement data spanning
ten years (1999 to 2008) were obtained from publicly available data collected by
Washington State Department of Ecology’s ambient monitoring stations. Monitoring
stations located at the mouth of seven rivers found within four watersheds were used in
this study. Watersheds were delineated using ArcMap version 10.5 digital elevation
model, and were identified as the Puyallup, Nisqually, Deschutes and South Hood Canal
watersheds. All drain into the Puget Sound, which include areas that have chronically low
dissolved oxygen. DOC loading data were compared to climate data (precipitation and
temperature) and land use data (percentage of land cover) for correlation analysis. Results
indicate that temperature and precipitation are weakly correlated with DOC loading once
DOC loading data is normalized to account for watershed size. However, within
individual watersheds, a rising trend of DOC loading correlated with increasing
precipitation is apparent. Percentage of wetlands and developed land cover classification
are positively correlated with increased DOC loading across all watersheds. Further study
into how climate and land cover classification impact DOC loading into the Puget Sound
region is recommended on a larger scale and over longer timescales throughout each
watershed to assist with Puget Sound ecosystem recovery efforts.
Table of Contents
Table of Contents
1.0 Introduction…………………………………………………………………………..1
2.0 Literature Review ........................................................................................................5
2.1 Introduction ........................................................................................................5
2.2 Dissolved Organic Carbon .................................................................................7
2.3 Carbon Cycling in Rivers……………......…………………………………….7
2.4 Environmental Health Risks of Excess DOC..……………………………......9
2.5 The Puget Sound Ecosystem……………………………………….....……...10
2.6 Dissolved Oxygen and Nutrient Loading………………......……….……….11
2.7 Methods for Analyzing DOC Export from Rivers……………………..……14
2.8 Climate Change……...………………………………………………………16
2.9 DOC and Climate Change in the South Puget Sound……………......………17
3.0 Conclusion……………………………………………………….…………..20
3.0 Methods………………………...………………………………………...…….……21
3.1 Study Area and Sampling Sites……………………………………….……..21
3.1.2 Nisqually Watershed…………………………………………..…….…….22
3.1.3 Puyallup Watershed……………………………………………..….….….24
3.1.4 Deschutes Watershed……………………………………………………..25
3.1.5 South Hood Canal Watershed……………………………………...….…..26
3.2 Watershed Delineations…………………………………..…………………28
3.3 Land Cover Values………………………..………………………………..29
3.4 Climate Data………………………………………………………………...30
3.5 Loading Values for DOC……………………………………………….…..31
3.6 Statistical Analyses: Climate Data and DOC…………………….….…………..31
3.6.1 Climate Data…………...………………………………………………….31
iv
3.6.2 Land Cover Correlations………….......……………………………………32
4.0 Results………………………………………...........……….…………………….…33
4.1 Temperature Comparisons……………………………………..…………….33
4.2 Precipitation Comparisons……………………………….......…………...…34
4.3 Temperature and DOC………………………………….....……………...…36
4.3.1 Temperature and DOC Normalized by Basin Size………….……37
4.4 Precipitation and DOC……………………………...………………………37
4.4.1 Precipitation and DOC Normalized by Basin Size.................……39
4.5 Land Cover and DOC Correlations………………..……………………….39
5.0 Discussion………………………………………..……………………………...…45
5.1 Temperature and DOC………….……………...…………………………..46
5.2 The Importance of Precipitation on DOC Loading………………….……..47
5.3 Land Cover and DOC……………………………………..……………….49
6.0 Conclusion………………………………………………………………….……..51
7.0 Bibliography……...……………………………………………………………….52
v
List of Figures
Figure 1. Map of 303(d) Waters…………………………………………………....……3
Figure 2. Ambient Monitoring Stations……………………………………….…….......21
Figure 3. Nisqually Watershed……………………………………………….…....……23
Figure 4. USGS Lidar of Mount Rainier Glaciers…………………………..…..………23
Figure 5. Puyallup Watershed……………………………………….………..……..…..25
Figure 6. Deschutes Watershed………………………………………………….………26
Figure 7. South Hood Canal Watershed………………………………….…….….….…27
Figure 8. Location in Washington State…........................................................................28
Figure 9a. Mean Monthly Surface Temperatures……………………………….…...…..33
Figure 9b. Mean Annual Surface Temperatures……………………………….…….…..34
Figure 10. Sum Annual Precipitation…...........................................................................35
Figure 11. Average Annual DOC Loading………………………………………….…..35
Figure 12. Linear Regression of Temperature and DOC..……………….……….……..36
Figure 13. Linear Regression of Normalized Temperature and DOC………….………..37
Figure 14. Total Precipitation……………………………………………………………38
Figure 15. Linear Regression Total Precipitation and DOC....…………………………..38
Figure 16. Linear Regression of Normalized Precipitation and DOC………………...…39
Figure 17. NLCD South Hood Canal………...…………………………………………..41
Figure 18. NLCD Nisqually……………………………………………………………...42
Figure 19. NLCD Puyallup………………………………………………………………43
Figure 20. NLCD Deschutes……………………………………………………………..44
Figure 21. Linear Regression Highlighting El Niño……………………………………..49
vi
List of Tables
Table 1. Location of Ambient Monitoring Stations………………………………….…22
Table 2. NLCD Reclassifications………………………………………………….……30
Table 3. Spearman’s ρ Rank Correlation Results……………………………………....40
Table 4. NLCD Percentages per Watershed……………………………………………40
vii
Acknowledgements
I would like to thank Teizeen Mohamedali at the Northwest region of Washington
Department of Ecology for use of her MLR derived data on DOC loading from long term
ambient monitoring stations for the rivers used in this study. Additionally, I would like to
thank my advisor, Dr. Erin Martin for her unending patience and guidance as my thesis
advisor, and for nudging me in the right direction when I needed it most. Most of all, I
am ever grateful for the love and patience of my children, Eva and Lilly who inspire me
to be a better person and never stop learning. I am thankful every day by the drive to do
better that they ignite in me.
I also would like to thank all the MES faculty who provided a sound board for my
many…many ideas for thesis research, and my cohort for all the laughs, blowing off
steam and supporting me in attaining my academic goals.
viii
1. INTRODUCTION
The long-term effects of climate change on coastal inland areas are expected to be
substantial. Rising sea levels and salt water intrusion threaten the quality and quantity of
inland freshwater systems, as well as the survival of estuarine biota (Marcogliese, 2016;
Sutter et al., 2015). By the year 2100 surface air temperatures are predicted to increase
globally between 1.4°C and 5.8°C, (IPCC, 2007), attendant to this increase in
temperature, changes in hydrologic patterns due to increased precipitation will transport
more terrestrial organic matter and pollutants into aquatic environments (Evans et al.,
2005; Tian et al., 2013).
Under climate change scenarios (IPCC, 2007) for the Pacific Northwest, warmer
ambient temperatures and increased winter precipitation is expected. Wetter, warmer
climate could lead to longer growing seasons, creating a larger pool of terrestrially
derived carbon transported to aquatic systems through runoff (Mote et al., 2003; IPCC,
2014; Littell et al., 2014). In addition, relationships between total carbon in aquatic
systems, rising temperatures and increased gross primary productivity (GPP) have been
observed in temperate wetlands (Ritson et al., 2014; Dinsmore et al., 2010) and temperate
river systems in the United States (Butman et al., 2016).
Warmer and wetter climate may elevate terrestrially derived Dissolved Organic
Carbon (DOC) loading from freshwater streams into estuarine environments (Bianchi,
2011; Dodds, 2006; Ritzon et al., 2014; Dinsmore et al., 2010). In the south Puget Sound
in Washington State, increased DOC export from fresh to marine waters could also
increase transport and settling of dissolved metals from terrestrial to aquatic
environments (Guggenberger et al., 1994 et al., Fichot and Benner, 2014; Kellerman et
1
al., 2014), cause brownification of drinking water systems (Fasching et al., 2014) and
exacerbate ocean acidification and harmful algal blooms in the marine environment, all
posing a significant environmental health risk for humans and aquatic life alike (Olson,
2017, Wallace et al., 2014).
Latitudes 40° North and above typically experience shorter growing seasons and
nutrient limited soils in higher elevations where growth potential is significantly limited,
resulting in lower nutrient and DOC concentrations in aquatic systems (Hood et al., 2006;
Wise et al., 2011) However, recent research (Evans et al., 2012; Tian et al., 2013) has
reported an increase of DOC concentrations from terrestrial origin into freshwater
catchments across northern latitudes, (25° to 55° N) and increased organic carbon
concentrations in northern streams and rivers, leading to greater CO2 respiration into the
atmosphere (Butman and Raymond, 2011). Carbon inputs from terrestrial sources to
freshwater and estuarine environments can adversely impact aquatic ecosystem
functioning. Increased DOC in freshwater systems is associated with transport of
pathogens and metals, (Winterdahl et al., 2014; Hallegraeff, 1993) and promotes growth
of bacteria and algae species that in excess can limit availability of dissolved oxygen in
aquatic systems and become toxic (Wallace et al., 2014; Van de Waal et al., 2009).
Water quality regulation standards enforced by the EPA require states to create
criteria for how each water body is used. Examples of designations for water bodies are
swimming beaches, public drinking water supplies or shellfish protection areas. Numeric
water quality criteria are used to describe the condition of water quality, and, to protect
against disease and loss of biodiversity (Ecology, 2015). In Washington State, water
quality criteria are categorized by numbers 1-5, the higher the number the more impaired
2
the aquatic system is. Category 5 waters are listed as 303(d) impaired water bodies, and
require that local and state governments and citizens to develop working plans towards
improving water quality. Much of the Puget Sound and greater Salish Sea do not meet
EPA water quality standards for many parameters ranging from temperature to high metal
concentrations and presence of toxic substances (Ecology, 2015).
In addition, many areas throughout the Sound do not meet EPA requirements for
dissolved oxygen levels and are listed as category 303(d) waterbodies (figure 1). Low
dissolved oxygen (DO) is of concern throughout the Puget Sound. Low DO stresses
marine life and leads to periodic fish kills in the Puget Sound (Deepe et al., 2013; Spietz
et al., 2015). Dissolved oxygen levels can be further degraded when algae multiply and
senesce during algal blooms.
Figure 1. Map of waterbodies listed on EPA and Ecology’s 303(d) list of impaired waterbodies for
dissolved oxygen. Red squares indicate category 5 waters for dissolved oxygen
3
Recent nutrient loading models (Pelletier et al., 2017; Roberts et al., 2014;
Mohamedali et al., 2011) identify the South Puget Sound as a region with increased
sensitivity to carbon flux (Pelletier et al., 2017) and dissolved oxygen levels
(Mohamedali et al., 2011). DOC is transported to aquatic systems from decaying plant
and animal matter, and soil erosion. As DOC is released throughout an aquatic system it
provides substrate for phytoplankton, and nutritional energy to photosynthesizing
cyanobacteria (Kirchman et al., 1991; Znachor and Nedoma, 2009; Larsson and
Hagström,1979; Carney et al., 2016) which encourages both their rapid reproduction and
death. Rapid growth and senescence of phytoplankton creates a physical barrier on the
surface of water, preventing any mixing with air, effectively reducing subsurface
dissolved oxygen and increasing water temperatures during periods of intense growth
(Roberts et al., 2014).
DOC flux is intrinsically tied to climate. Temperature and hydrology changes
associated with climate change alter terrestrial processes in soils, wetlands and vegetation
growth and senescence (Evans et al., 2005). Climate change could be the driver of
widespread DOC increase in northern aquatic systems (Tian et al., 2013). Live
vegetation, leaf litter, root exudation and soil erosion contribute significantly to DOC in
terrestrial streams through leachate (France et al., 1996; Roberts and Bilby, 2009). Land
processes such as decomposition of soil and plant matter are well explored mechanisms
for DOC flux in aquatic systems throughout the Puget Sound region (Awale et al., 2017;
Gray et al., 2016; Roberts et al., 2009; Yano et al., 2004; Qualls et al., 1991).
4
Understanding how changes in climate, precipitation and dominant land type impact
DOC flux in rivers draining into the Puget Sound remains poorly understood.
Investigating the effect of temperature and precipitation on DOC loading over time
from rivers draining into areas of the South Puget Sound with chronically low dissolved
oxygen will help to understand how, if at all, DOC flux changes in response to ambient
climate conditions. The objective of this study is to investigate whether temperature and
precipitation affects DOC loading from four South Puget Sound watersheds, South Hood
Canal, Deschutes, Nisqually and Puyallup. The South Hood Canal and Deschutes
watersheds are small lowland catchments (705.37 km2 and 420 km2 respectively), while
the Puyallup and Nisqually are glacially fed rivers with larger basin sizes (2455 km2 and
1399 km2 respectively). Annual averages of DOC loading from each river over a period
of ten years will be analyzed to identify patterns between DOC loading, climate and land
cover, and to evaluate whether DOC loads are increasing over time.
While thirty-year climate cycles are typically analyzed for assessing climate related
changes in the environment, (IPCC, 2014; Ault et al., 2014) the analysis of decadal
temperature, precipitation and DOC loading values can illustrate a finer scale variance of
climactically driven changes within single watersheds (Ault et al., 2014; Luo et al.,
2013).
2.0 Literature Review
2.1 Introduction
By the year 2100 surface air temperatures are predicted to increase globally between
1.4°C and 5.8°C, (IPCC, 2007). Warmer ambient temperatures in the Pacific Northwest
5
will result in longer growing seasons, increased soil and surface water temperatures, and
increased GPP (Mote and Salathe, 2010; Leung and Wigmosta, 1999; Tian et al., 2013;
Williams et al., 2014). Changes in annual precipitation, especially associated with
prolonged dry periods and sudden severe storm events, put watersheds at increased risk
of erosion and scouring, and increase the transport of pollutants (Khir-Eldien and Zahran,
2017; Yasarer et al., 2017). Longer growing seasons and transport of decaying plant
matter and soils may elevate DOC availability and loading from freshwater streams into
estuarine environments. Terrestrial DOC export from fresh to marine waters provide
substrate for photosynthesizing bacteria and algae to bind to (Wear et al., 2015; Carney et
al., 2016) and upon senescence these micro-organisms release more even more DOC
(Lampert, 1978; Khangaonkar et al., 2012). In the South Puget Sound in Washington
State, uncontrolled aquatic algae growth limit dissolved oxygen concentrations (Winter et
al., 1975; Speitz et al., 2015; Schiebly et al., 2015) and in areas of the sound with
chronically low dissolved oxygen (Fig.1), changes in DOC flux could adversely affect
water quality and contribute to acidification in the Puget Sound (Pelltier et al., 2017).
In addition, rivers are a major source of CO2. DOC is respired back to the atmosphere
as inorganic Carbon, CO2 (Butman and Raymond, 2011, Fasching et al., 2014). DOC
export from rivers is understudied in the South Puget Sound region, and, as organic
carbon is intrinsically tied with CO2 respiration (Butman and Raymond, 2011) and ocean
acidification, (Pelltier et al., 2017) understanding DOC export under changing climate
can help inform restoration decisions in the South Puget Sound.
6
2.2 Dissolved Organic Carbon
Dissolved Organic Carbon is defined as aqueous organic material that passes
through a 0.7 µm filter in fresh water, and a 0.2 µm in the marine environment
(Kirchman et al., 1991). DOC solutes are part of the dissolved aqueous matrix and
provide substrate (Arandia-Gorostidi et al., 2017) for free floating primary producing
microorganisms such as cyanobacteria and diatoms (Bittar et al., 2015; Sarmento et al.,
2016). DOC is comprised of decaying organic matter and organic rich material that
provide the molecular substrate for other metals and compounds to bind to (Thingstad et
al., 1997; Lange et al., 2016).
DOC has an integral role in healthy aquatic ecosystem functioning and carbon
cycling. Through microbial respiration, much of the DOC in rivers is degassed into the
atmosphere as CO2, (Butman & Raymond, 2011; Raymond & Bauer, 2011; Smith et al.,
2017), or the carbon can be buried and stored in ocean sediment (Fasching et al., 2014).
Although DOC is a naturally occurring and a necessary carbon species for supporting
ecosystem wide food webs, excess DOC in aquatic systems can alter ecosystem
functioning and environmental health (Harvey et al., 2002; Hansel and Carlson, 2014;
Dhillon et al., 2013).
2.3 Carbon Cycling in Rivers
Carbon is stored on land in sediment and rock, vegetation, soils and living
organisms. Eventually, terrestrial carbon is eroded by wind and precipitation, and much
of the terrestrially derived carbon is transported to rivers and oceans as runoff (Fasching
7
et al., 2014; Raymond et al., 2013). Rivers flush carbon from land to the ocean, and by
doing so, rivers play a significant role in the global carbon cycle (Butman et al., 2016).
DOC comprises half of the total organic carbon in rivers, the other half is
particulate organic carbon, larger particles of aqueous organic material that do not pass
through a 0.7 µm filter (Kirchman et al., 1991). Total organic carbon export is often
quantified together when accounting for riverine carbon flux, and DOC is of interest
because it is directly available to microorganisms, facilitating CO 2 fluxes to the
atmosphere (Butman et al., 2015).
Rivers in the United States transport 100 teragrams (TgC) of carbon per year to
the ocean (Butman et al., 2015). However, much is unknown about terrestrial sources of
carbon transport to rivers and how it is changing over time in response to intensified
agricultural land use, and development. Butman et al., (2015) estimates that carbon flux
from land to rivers increase by 0.1 to 0.2 pentagrams of carbon per year due to human
disturbances such as deforestation and agricultural use, and most of the DOC in rivers is
from young carbon leached from soils and vegetation. Younger DOC in rivers result from
run off and recent land based disturbances, (Butman et al., 2015; Butman et al., 2016) and
increased precipitation attendant to climate change, could increase the rate at which
terrestrial DOC is transported to aquatic systems (Tian et al., 2013; Evans et al., 2012).
Additionally, higher DOC concentration in rivers is a concern under climate
change models. Butman and Raymond (2011) estimated the amount of carbon degassed
from streams in the United States is high, especially between latitudes of 25°N to 50°N.
In their study, (2015) they estimate that 97 ± 32 Tg of CO2 are degassed from rivers into
8
the atmosphere each year. Further, their results show that CO2 evasion from streams is
positively correlated with precipitation, emitting more CO2 from terrestrial origin into the
atmosphere from regions in the United States with high snowpack and rainfall. In the
Pacific Northwest, climate models predict dry, hot summers followed by warm and wet
winters. Increased precipitation following long dry events could result in even more DOC
runoff to rivers and potentially cause a positive feedback loop for CO2 evasion to the
atmosphere (Salathé et al., 2014; Butman and Raymond, 2011).
2.4 Environmental Health Risks of Excess DOC
Elevated DOC leads to freshwater brownification and acidification, harming
aquatic organisms (Jones and Lennon, 2015). In systems that utilize surface waters for
drinking water, elevated DOC reduces the effectiveness of chlorination in the treatment
process, increasing risks of bacterial contamination and the formation of carcinogenic
organochlorine compounds (McBean et al., 2010; Evans et al., 2005). The darkening of
water associated with brownification limits light penetration causing early senescence of
aquatic vegetation, leading to decomposition and transport of decaying matter and
sediment erosion. This process can increase the likelihood of low dissolved oxygen and
higher stream temperatures through limiting light penetration and microbial respiration in
both freshwater and marine environments (Fasching et al., 2014; Pelltier et al., 2017). In
addition, DOC particles bond with metals toxic to aquatic life, increasing their solubility
in freshwater and molecular weight (McBean et al., 2010; Evans et al., 2005).
Research on DOC loading into coastal and estuarine systems is sparse (Evans et
al., 2005; Tian et al., 2013; Mattson et al., 2015), and of these studies, none attempt to
9
quantify riverine DOC loading into the Puget Sound Washington State. Increased
occurrence of algal blooms and prolonged periods of low DO have been reported in the
Puget Sound (Pelltier et al., 2017; Ahmed et al., 2017; Hallock, 2009), and increased
nutrient loading including DOC has been reported in some areas (Pelltier et al.,
Mohamedali et al., 2011). However, how DOC loading from Puget Sound rivers is
changing is unexplored, warranting further investigation into how much DOC is loading
into the Puget Sound ecosystem, how this loading is correlated with climate and land use
and if loading is increasing over time (see section 3).
2.5 The Puget Sound Ecosystem
The Puget Sound is a unique inland estuary nestled into the southernmost reach of
the Salish Sea within Washington State. With 2,500 miles of shoreline, the Puget Sound
is the second largest estuarine sea in the United States. The deep canals and finger like
topography of these waters were created by the retreat of the Cordillerian ice sheet during
the Last Glacial Maximum (Crandell et al., 1958). The glacial carving of the regions
bathymetry gives the Salish Sea marine connectivity to the Pacific Ocean through the
Strait of Juan de Fuca and the Strait of Georgia. The straits greatly influence marine
circulation in the northern Salish Sea, and to a lesser extent, the Puget Sound
(Mohamedali et al., 2011; Sutherland, 2011). Tidal circulation patterns in the Puget
Sound have slow regeneration times, (replacing older, oxygen depleted water with fresh
seawater) and can have periods of prolonged residence times. Mohamedali (2011)
modeled regeneration of seawater in South Puget Sound, and found that in some areas
(Hood Canal), regeneration can take 282 days to as long as 292 days in the Tacoma
Basin. The slower tidal exchange with the Pacific Ocean make the Puget Sound
10
especially sensitive to terrestrial pollutants (Feely et al., 2010; MacCready, 1999), land
use changes in the near shore environment, and, to nutrient loading from the 10,000
rivers and streams draining into the Puget Sound (Mohamedali et al., 2011; Roberts &
Bilby, 2009).
Slow water circulation in the Puget Sound combined with rapid urban sprawl in
upstream watersheds have a significant impact on water quality (Moore et al., 2011).
Washington State Department of Ecology has determined that much of the Sound fails to
meet federal water quality standards under federal clean water act (Ecology, 2003;
Mohamedali et al., 2011).
Many reaches of the Puget Sound and its adjacent watersheds fail to meet water
quality criteria and frequently experience hypoxic events due to elevated temperatures
and low dissolved oxygen (Mohamedali et al., 2011). In addition, toxic compounds are
ubiquitous in this ecosystem and accumulate in organisms from all trophic levels in the
marine food web (Norton et al., 2011). Impaired waters are a hazard to human health and
the aquatic food web. Given the sensitivity of the Puget Sound, it is imperative that
scientific research can drive solutions to improve water quality.
2.6 Dissolved Oxygen and Nutrient Loading
Identifying the mechanisms of pollution transport to the Puget Sound is key for
developing strategies to mitigate impaired water quality. Many studies in the region have
examined land use impacts, climate change and water temperature as contributing factors
to impaired water quality. These studies have focused on the biological impacts of
impaired waters, and its effects on salmonid populations, or shellfish biotoxins (Moore et
11
al., 2011; Naiman et al, 1992; May, 1997). Other studies have focused on the interaction
between terrestrially derived toxics and algal blooms associated with anthropogenic
sources (Norton et al., 2011; Feely et al., 2010). While these issues are all intrinsically
connected, research on loading of compounds from the surface waters to the Puget Sound
are few.
Many areas in the Puget Sound fail to meet water quality standards for dissolved
oxygen. Levels of DO in many reaches of the Sound are chronically low, with some areas
regularly measuring at less than1 mg/L (Khangaonkar et al., 2012). Mohamedali et al.,
(2011) and Ahmed et al., (2014) identified nutrient loading from anthropogenic sources
to be a major contributor to low DO in the South Puget Sound. Further, in areas with high
anthropogenic nutrient loading, addition of nutrients reduced DO in already stressed
regions of the sound by as much as 0.4 mg/L. Pelltier et al., (2017) also found that inputs
of carbon increase DO sensitivity in the Puget Sound. Pelltier et al., (2017) found that
increased ocean acidification also correlates with areas of the Puget Sound with
chronically low DO.
Analyzing water quality data to determine loading into adjacent water bodies is
typical when calculating total pollutant loads in a watershed. While loading of DOC has
not been quantified in the South Puget Sound region, Mohamedali et al., (2011)
developed a strategy for quantifying dissolved inorganic nitrogen (DIN) loading into the
Puget Sound as part of the Department of Ecology’s efforts to identify mechanisms that
contribute to low dissolved oxygen throughout the Salish Sea and Puget Sound. Daily
nutrient concentrations were developed from historical ecological sampling efforts and
12
long-term monitoring stations to assess changes throughout time. Loading was calculated
using the following equation:
Daily Load= (predicted daily concentration) X (daily streamflow)
Equation 1. Calculation for estimating loading from rivers into the Puget Sound. Concentration is measured
from mg/L and streamflow in cubic feet per second
Mohamedali et al. (2011) reports that DIN loading from rivers did not necessarily
coincide with the size of the watershed or the size of the river. Some smaller rivers load
more DIN into the Puget Sound than larger rivers. In addition, the seasonal loading of
DIN into the Sound is not flow dependent. High levels of DIN were loading into the
Puget Sound regardless of seasonal flow. DIN loading from rivers was correlated with
watersheds with high development and agricultural intensity.
Mohamedali (2011) concluded that analyzing loading data on an annual scale in
this study was more appropriate than a seasonal scale. Puget Sound watersheds are
diverse, the headwaters of some rivers originate in mountainous regions and experience
pulse flows during the spring snowmelt, while the headwaters of other rivers are at lower
elevation and receive little or no flow associated with spring snow melt, rather,
experience high flow events during storms in the fall and winter (Babson et al., 2006;
Cuo et al., 2009). The diversity of South Puget Sound watersheds deems an annual
loading analysis of DOC to be an appropriate temporal scale to analyze changes over
time when looking at the Puget Sound as a whole.
Like nutrient loading, DOC can reduce DO in an aquatic system. Brownification
associated with excess DOC limits light penetration and leads to local warming effects
13
and hypoxia (Turner et al., 2008). Analyzing DOC loading from rivers in the southern
Puget Sound on an annual time scale can help to understand how much DOC is entering
the system from the mouths of rivers, and what environmental factors are correlated with
DOC loading. Gaining a temporal and spatial understanding on DOC loading in the Puget
Sound is imperative for improving water quality.
Dissolved Organic Carbon Export
2.7 Methods for Analyzing DOC Export from Rivers
Local research on riverine transport of DOC in the Puget Sound is currently not
present in the literature, however, many studies analyzed DOC export globally and found
varied results. Correlations of DOC export with land type and soil type have conflicting
results. Curtis (1998) and Xenopoulos et al., (2003) sampled lakes in temperate regions,
and found that watersheds with isolated wetlands not connected to flowing water had
consistently higher concentrations of DOC export, while samples taken from clear
montane lakes with poor soil composition and wetlands with connectivity to flowing
water had lower concentration of DOC. The spatial correlation of elevated DOC and
isolated wetland proximity indicate that areas with a variety of vegetation and humic soils
have higher concentration of DOC runoff. The exception to this, Xenopoulous found, is
coniferous lands with isolated wetlands have lower DOC concentration in nearby lakes
than areas without coniferous forest. A possible mechanism for this is that forests use
available carbon and store it in the vascular structure of vegetation (Xenopoulos et al.,
2003; Michalzik et al., 2001).
14
The hydrography of lakes differs from rivers flowing to the sea (Wetzel, 2001).
Huntington & Aiken (2013) argue that in a riverine system, particularly larger rivers, the
diversity of the landscape can make it difficult to correlate variables such as land type, as
discharge and flow direction may change several times and flow through a variety of land
types before the river reaches the ocean.
Huntington & Aiken (2013) sampled monthly for DOC export along the entirety
of the Penobscot River. DOC concentrations were analyzed using a correlation analysis
for the variables of season, slope, area and distance to wetlands. Land values were
obtained by calculating percentages in ArcGIS software. Statistical analysis identified
that watershed slope, season (summer) and forest type (coniferous or hardwood) were
negatively correlated with high DOC concentrations, while DOC concentration and
percentage of watershed covered by isolated wetlands was positively correlated.
Wetlands are naturally high in DOC and decaying vegetation that can load into rivers
through the water table or in runoff.
Mattsson et al., (2015) identified that hydrology and precipitation are changing
due to the warming climate. Total organic carbon (TOC) export was analyzed in thirty
river basins flowing into the Baltic Sea. Land type and percentage were calculated using
GIS software to obtain percentages of agricultural, peat, and forest catchments. Ten years
of historical TOC and discharge data were obtained from the Finnish ambient monitoring
network and were multiplied by flow to develop loading values of DOC. Mean
temperature and precipitation during the study period were quantified and analyzed
against the land type variables using the Spearmans rank correlation analysis. Results
indicate that land type had no correlation with TOC, rather, increased flows associated
15
with precipitation drove elevated TOC in the 30 basins sampled. The carbon species
sampled was different, as TOC includes particulate and dissolved organic carbon,
however the lack of correlation with peat and wetlands indicates that abiotic variation
associated with climate change are affecting carbon in this boreal region.
Boreal regions are experiencing rapid environmental stresses due to climate
change (Mattsson et al., 2015). Melting permafrost saturates groundwater and increases
flooding events, while at the same time, larger and more frequent forest fires are
occurring (Terrier et al., 2013; IPCC, 2014). Unstable soils and increased vegetation will
increase DOC in this area that is already saturated with DOC. Understanding mechanisms
and impacts of a changing climate in different ecosystems helps to understand how
abiotic conditions can impact DOC loading globally.
2.8 Climate Change
Aquatic systems are sensitive to changes in temperature. Increased precipitation
can overwhelm lowland areas and lead to flooding and significant erosion of stream
channels. Likewise extended dry periods lower the water table leading to drought
conditions that increase erosion in streams and rivers (Raymond & Oh, 2007). Increasing
atmospheric CO2 concentrations are amplifying climate variability, drought, and frequent
storm events, especially in the northern latitudes (Tian et al., 2013; Butman & Raymond,
2011). Rivers are a major source of CO2, and much of the DOC in aquatic systems is
respired, and then degassed back into the atmosphere before reaching the oceans (Butman
& Raymond, 2011).
16
Evans (2005) analyzed fifteen years of DOC measurements from a long-term data
set of ambient monitoring collected by the UK Acid Waters Monitoring Network
(AWMN). Historical data of DOC concentrations (mg/L) were analyzed using the
Seasonal Kendall test to identify trends in DOC concentration. Results indicated that
DOC concentration had increased at all sites. The magnitude of the trend was calculated
using the Sen slope estimation method and revealed that DOC concentrations had
increased 91% over the fifteen-year study period. The rate of increase was found to be
proportional among all sites in the AWMN. A stepwise regression was performed to
identify statistical relationships between variables and potential mechanisms for the
increase. The variables analyzed were year and time of year, temperature, and chemical
variables associated with acid rain. Increasing temperature was identified as the
mechanism for increasing DOC on an annual scale, because warmer temperatures
increase the length of the growing season for both terrestrial and aquatic plants and
animals, and, the rate of senescence and regeneration of organic matter (Evans et al.,
2005).
Warmer temperatures associated with climate change are impacting the ecological
chemistry of systems worldwide. Temperature and precipitation have increased globally
during the 21st century, (Hayhoe et al., 2007; Balch et al., 2012) and are expected to
continue rising. Hydraulic alteration, intensified solar radiation and pulse flows from
increased precipitation are some of the climate change effects that aquatic systems are
experiencing (Butman & Raymond, 2011; Raymond & Oh, 2007).
DOC export in river systems is a result of the balance between primary
production, decomposition of organic matter and terrestrial inputs. The quantity of DOC
17
in aquatic systems is impacted by changing climate (Tian et al., 2013). Most studies
analyzing changes in riverine DOC are localized to one area or a single river residing in
one climate zone. Tian et al., (2013) analyzed fourteen years of DOC and flow data
available from the USGS National Stream Quality Network (NSQN) from seven major
watersheds in differing climactic zones throughout the United States. The variables of
surface temperature, precipitation, and land cover types were used to predict DOC
concentrations using a series of simple linear regressions. Precipitation had no linear
correlation across sites and land type had some correlation as all sites shared similar
percentages of wetlands. The strongest variable across all climate zones was temperature.
Seasonal variation in temperature showed no correlation, however annual temperatures
across sites was positively correlated and was concluded to be the mechanism for
elevated DOC across seven climate zones revealing much about how increasing
temperatures associated with climate change are impacting rivers regardless of the
location of the river on a geodedic gradient.
Raymond and Oh (2007) found that temperature was not statistically correlated
with DOC export across three watersheds in the agricultural hub of the United States.
Using seven years of publicly available USGS DOC data, researchers used LOADEST
modelling tools to assess how temperature, precipitation, discharge and land use
correlated with DOC and export of other nutrients. Their results indicated that
temperature had little correlation with DOC export, rather discharge associated with
increased precipitation determined how much DOC was in the river systems at a given
time. This study took place in watersheds with altered hydrology dominated by
agriculture. This variable was not included in the model, however other studies have
18
found that agricultural runoff during storm events is a major source of non-point source
pollution, including DOC (Veum et al., 2009; Howarth et al., 2009).
The process of farming moves a lot of organic material. Farm machinery turn
soils over and planting crops yields high qualities of decaying organic matter after
harvest is complete. Physically altering land for farming practices results in nutrient rich
soils that run off during storm events both into surface water, and ground water (Veum et
al., 2009).
Bauer et al. reports that temperature, and flooding associated with storms is
responsible for DOC loading from rivers into the coastal environment. Bauer et al.,
identifies land management as the primary driver of the increased DOC load in rivers and
the coastal environment. Coastal areas generally have high populations, and agriculture is
usually near rivers or streams for ease of irrigation. Researchers argue that human
disturbance, by moving more soil and altering natural hydrology is the leading cause for
annual increase of DOC in freshwater systems, rather than abiotic conditions.
2.9 DOC and Climate Change in the South Puget Sound
Rising temperature, reduced snowpack and increased precipitation attendant to
climate change scenarios in Washington State, will negatively impact water quality (Mote
and Salathè, 2010). Rising sea levels are expected to inundate low lying areas and near
shore aquifers, (Mote and Salathè, 2010) and already have in islands throughout the
Puget Sound (Banas et al., 2015). Increased nutrient and carbon loading are increasing
the rate of ocean acidification, and can further reduce dissolved oxygen levels in the
Puget Sound, (Pelltier et al., 2017; Mohamedali et al., 2011) while dryer summers reduce
19
biodiversity and soil moisture in forest stands, increasing the risk of wildfire and erosion,
leading to scouring of organic matter during the increased cold season precipitation under
future climate scenarios (IPCC, 2014; Mote and Salathè, 2010; Pelltier et al., 2017;
Khangaonkar, 2012; Terrier et al., 2013; Mote et al., 2013).
3.0 Conclusion
DOC flux in northern freshwater systems is increasing (Evans et al., 2008;
Butman and Raymond, 2011; Tian et al., 2013). Freshwater systems have been altered
considerably by humans, and coastal areas are increasingly experiencing hypoxic events
due to human disturbance (Turner et al., 2008; Bauer et al., 2013). Climate change is a
major challenge for limiting DOC export into the coastal environment, and how it will
affect biota over time is not yet clear. However, there is a gap in the literature involving
DOC loading into the Puget Sound estuary, and measuring variables of land use and
temperature are important for identifying how elevated DOC enters aquatic systems.
The studies described in this literature review reveal that there are conflicting
conclusions in what mechanisms explain changes in DOC loading to estuarine
environments. However, the time scale in which to analyze DOC data suggests that
annual time scales reveal the most about how loading is changing over time. In addition,
the conclusions of these studies indicate that temperature, precipitation and land use are
important variables that are positively correlated with elevated DOC levels in rivers.
Whether climate or land use can illuminate how the South Puget Sound region is
experiencing changes in DOC loading, and if changes in DOC loading are happening in
the region is unknown and will be explored in this study.
20
3.0Methods
3.1 Study area and Sampling Sites
Four watersheds located in the South Puget Sound of Washington State were
chosen due to their discharge into marine areas with chronically low dissolved oxygen
levels. Two of the watersheds (Puyallup and Nisqually) are glacially fed from Mount
Rainier with steep elevation changes, and two, (Deschutes and South Hood Canal) are in
lowland forested and agricultural areas, rich in wetland and prairie soils. All DOC and
discharge measurements were collected from Washington State Department of Ecology
long term ambient monitoring water quality stations located near the mouth of each
watershed river (figure 2). Coordinates for each station are in Table 1.
21
Figure 2. Locations of Ecology long term ambient monitoring stations where DOC
concentrations and flow were used for this study.
River
Nisqually
Puyallup
Deschutes
Skokomish
Dosewallips
Duckabush
Hamma-Hamma
Station Number
11A70
10A050
13A050
16A070
16D070
16C070
16B070
Latitude
47.0620
47.2136
47.0151
47.3098
47.6901
47.649
47.5501
Longitude
-122.6964
-122.3414
-122.9026
-123.1771
-122.8991
-122.9349
-123.0529
Table 1. Station names and location for Ecology Long Term Ambient Freshwater Monitoring station where
discharge and DOC samples were extracted for this study
3.1.2 Nisqually Watershed
22
The Nisqually Watershed includes the Nisqually River. The headwaters of the
Nisqually River originate from the Nisqually Glacier on the southern side of Mount
Rainer (Figures 3 and 4). The elevation of the river’s headwaters is 1,466 meters, and the
river quickly descends to sea level (0 meters) in the South Puget Sound. The total length
of the Nisqually River is 130 km, and the watershed basin covers 1,970 km2. Average
discharge is 1,460 cubic feet per second (cfs) (Ahmed et al., 2014).
The terrain of Nisqually watershed varied. The headwaters are rocky with sparse
vegetation, and quickly descend through forested regions, and grassland foothills with
varied agricultural uses. Much of the Nisqually River passes through Joint Base LewisMcChord and the Nisqually Indian Reservation to where it empties in Nisqually Reach in
the South Puget Sound at the Billy Frank Jr. National Wildlife refuge north of Olympia
Washington (Figure 3).
Figure 3. The Nisqually River Watershed and location of the Nisqually Ecology Ambient Monitoring
Station
23
Figure 4. USGS Lidar image of Mount Rainier glaciers. Arrows depict glacial headwaters of the Nisqually
Carbon, White and Puyallup Rivers. Image Source National Park Service
3.1.3 Puyallup Watershed
The Puyallup Watershed includes the Puyallup River, as well as the Carbon and
White Rivers, which drain into the Puyallup before the rivers empty into the Puget
Sound. The headwaters of the Puyallup River and its main tributaries are glacially fed
from Mount Rainier. The Puyallup River begins in two forks, The North and South
Puyallup Rivers. The north fork originates from the Puyallup Glacier, and the south fork
from the Tahoma Glacier on Mount Rainier (figure 4). The two streams flow through
Mount Rainier National Park until they join to form the mainstem Puyallup River.
24
Two major rivers with glacially fed headwaters (the Carbon and Emmons Glaciers) join
the mainstem Puyallup River further downstream. The Carbon River and the White
River, (figure 5) which originate from the Carbon and Emmons Glaciers on Mount
Rainier (figure 4).
The Puyallup River and its tributaries originate at an elevation of 1,539 meters
and descend to sea level (0 meters) to the where the river drains to Commencement Bay
in Tacoma, South Puget Sound. The total length of the Puyallup, Carbon and White
Rivers combined is 241km, and the Puyallup watershed basin covers 2455 km2. Average
discharge for the Puyallup River near the mouth is 3,313 cfs (Ahmed et al., 2014).
Much of the Puyallup basin is forested, with agricultural land dispersed
throughout. The river passes through mountainous regions, national forest and privately
owned forested areas, and the Puyallup Indian Reservation. The Puyallup drains to a
heavily urbanized area in Commencement Bay in the city of Tacoma.
Figure 5. The Puyallup Watershed and location of the Puyallup Ecology Ambient Monitoring Station
25
3.1.4 Deschutes Watershed
The Deschutes watershed includes the Deschutes River. The headwaters of the
Deschutes River are in lowland hills of the Gifford Pinchot National Forest, in Lewis
County Washington. The total length of the Deschutes River is 80 km, and the watershed
basin covers 420 km2. Average annual discharge for the Deschutes River is 396 cfu.
(Ahmed et al., 2014).
The terrain of the Deschutes watershed is varied. The headwaters are in national
forest lands, and the river courses through agricultural valleys until it reaches the heavily
urbanized South Puget Sound where the river drains to Budd Bay, in the city of Olympia
(figure 6).
26
Figure 6. The Deschutes watershed and location of the Deschutes Ecology Ambient Monitoring Station
3.1.5 South Hood Canal Watershed
The South Hood Canal Watershed includes the Skokomish, Dosewallips,
Duckabush and Hamma-Hamma Rivers (figure 7). The headwaters of all rivers in the
watershed are in the Olympic Mountains, in the Olympic National Park. The Skokomish
River is the largest River draining into Hood Canal, Puget Sound.
The Skokomish River including the North and South Fork is 69 km long, with
average discharge of 1,210 cfu, the Dosewallips River is 45.8 km long, with average
discharge of 446 cfs, the Duckabush River is 42.30 km long with average discharge of
416 cfs, and the Hamma Hamma is 27.73 km long with average discharge of 364 cfs
(Ahmed et al., 2014; O’ Connor, 1980). The basin size of the South Hood Canal
Watershed, including the four rivers is 705.37 km2.
The terrain of the South Hood Canal Watershed is varied. All four rivers in this
watershed originate in rocky mountainous areas and descend through forested areas rich
in wetlands and relatively little development.
27
Figure 7. The South Hood Canal Watershed and locations of Ecology Ambient Monitoring Stations
28
Figure 8. Study area locations in relation to Washington State and the United States
3.2 Watershed Delineations
Watershed delineations were made using the National Hydrography Dataset
(2016) and Arc GIS version 10.5 digital elevation model (DEM). Arc DEM determines
how water flows across the landscape to draw watershed boundaries. For the South Hood
Canal Watershed, the Skokomish, Hamma Hamma, Dosewallips and Duckabush Rivers
fall into the same DEM watershed, so loading values from all four of these rivers were
summed to obtain the total loading value for the entire watershed (Figure 9). For the
Puyallup River, the White and Carbon Rivers flow into the Puyallup, which eventually
drains into the Puget Sound (Figure 11). These rivers were also included in the DEM,
29
although DOC loads for these rivers were not individually quantified. Their contributions
to the total DOC load is measured at a single location, the monitoring station at the mouth
of the Puyallup River (figure 2). For the Deschutes, (Figure 12) the only river within the
delineation boundary was the Deschutes River, likewise for the DEM of the Nisqually
River (Figure 10).
3.3 Land Cover Values
Land cover types listed as percentage of each land cover class were obtained from
the 2006 and 2001 National Land Cover Datasets (NLCD) available from the US EPA.
Land cover was extracted for delineated watersheds using Arc GIS spatial analyst tools,
and pixels were converted to percentages using the calculate geometry tool. Study areas
contained thirteen different NLCD landcover types, which were merged into six major
land types: Agricultural, Developed, Wetland, Forested, Rocky Slope and Glacial (Table
2). Apart from the glacial land cover class, land cover classes for agricultural, developed
and forested land classes were merged to eliminate values with less than 0.01% of land
cover, or to generalize land type.
Differences in land cover percentages between 2001 and 2006 NLCD layers were
quantified and averaged to capture any changes in land cover during the study period.
30
NLCD land use/land cover type
Developed- Open Space
Developed- Low Intensity
Developed- Medium Intensity
Barren Land
Deciduous Forest
Evergreen Forest
Mixed Forest
Glacial
Pasture/ Hay
Cultivated Crops
Palustrine Forested Wetland
Palustrine Scrub/Shrub Wetland
Reclassification used in this study
Developed
Developed
Developed
Rocky Slope
Forest
Forest
Forest
Glacial
Agriculture
Agriculture
Wetland
Wetland
Table 2. NLCD Land Cover Reclassifications used in this study
3.4 Climate Data
Climate data, (temperature and precipitation) were obtained from the National
Climatic Data Center (NCDC, 2011). Data were from 1999 to 2008 and were selected
from different spatial scales throughout each watershed. Multiple weather stations were
averaged within each watershed to capture the overall temperature and precipitation
trends each year for the entire watershed. All weather stations that fell into the delineated
watershed boundary were included and quantified.
Values were obtained by searching for the weather station coordinates on the
NCDC mapping tool. Daily temperature and precipitation averages for the entire ten-year
study period were obtained and averaged to assess monthly and annual averages for the
entire watershed. All precipitation data was summed annually. Temperature data was
obtained in units of Celsius (°C), and precipitation was measured in millimeters (mm).
31
3.5 Loading Values for DOC
All loading values used in this study were obtained from the South Puget Sound
Dissolved Oxygen Study by Mohamedali et al., (2011) and Ahmed et al., (2014). In these
studies, monthly DOC concentrations (mg/L) and daily flow measurements were
obtained from long term ambient monitoring stations at the mouth of each river (figure
2). Loading values are the sum of concentration X flow and are used to obtain daily
values (kg/day) of DOC loading into the South Puget Sound. Daily loading values were
summed to obtain single annual values for each year (1999-2008) used in this study and
are from publicly available on Ecology’s Environmental Information Management (EIM)
database that is accessible via Ecology’s website.
3.6 Statistical Analyses of Climate Data and DOC
3.6.1 Climate Data
Data was assessed for normality and a linear regression analysis was performed
using JMP SAS 13.0 statistical software to identify significant variables associated with
annual DOC loading. For each watershed, DOC loading was the dependent y-axis
variable, and temperature and precipitation were the independent x- axis variables.
Variables were deemed significant when p value was less than 0.05. A full factorial
ANOVA analysis was also used to assess changes within each individual watershed.
Watershed size was log transformed to assess normality. Once it was established
that it was normally distributed, relationships between watershed size and other variables
were assessed. Linear regression analysis was also performed on the normalized data to
32
assess whether DOC loading is increasing over time, and if DOC loading is correlated
with temperature and precipitation.
3.6.2 Land Cover Correlations
Land cover percentages and correlations with DOC loading were not normally
distributed. Land cover data was analyzed using non-parametric methods. A Spearman’s
Rho rank correlation was used to illustrate relationships between land cover percentages
and DOC loading. The Spearman’s rank correlation compares the rank correlation
coefficient (r) between two variables X and Y. The values of the variables are converted
into ranks, then r is calculated as:
Equation 2. Spearman’s Rho rank correlation equation (Yue et al., 2002)
Where Ri is the rank of the X value (land cover percentage) and Si is the rank of the
Y value (DOC loading) and R and S are the means of the Ri and Si values (SAS Institute,
2015; Wang and Yin, 1996). This non-parametric correlation analysis works well with
data that are not normally distributed and when comparing means across small sample
sizes. Percent land cover was not distributed among all watersheds, especially compared
to loading in Kg/day.
33
4.0 Results
4.1 Temperature Comparisons
Surface air temperatures were relatively uniform across all four watersheds
studied. Averaged monthly temperature observations followed seasonal patterns (figure
9a), and varied between the high of 19.53°C and the low of 1.62°C. Average annual
temperature in the Puyallup River watershed was lowest during all months compared to
the Nisqually, Deschutes and South Hood Canal watersheds. Temperature was
consistently higher in the Deschutes watershed during the study period.
Averaged mean annual temperatures 1999-2008 (figure 9b) show in greater detail
the temperature differences between each watershed without seasonal interference.
Glacially fed rivers, (Puyallup and Nisqually) had generally lower temperatures
(Nisqually ~1°C lower; Puyallup ~3°C lower) than the Deschutes and South Hood Canal
Watersheds.
Mean Monthly Temperature °C
25.00
20.00
Puyallup
15.00
Nisqually
Deschutes
10.00
Skokomish
5.00
0.00
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
Figure 9a. Mean monthly surface temperatures averaged over the period 1999-2008
34
Mean Annual Temperature °C
14.00
12.00
Skokomish
10.00
8.00
Puyallup
Deschutes
Nisqually
6.00
4.00
2.00
Linear (Skokomish)
Linear (Puyallup)
Linear (Deschutes)
Linear (Nisqually)
0.00
Figure 9b. Mean annual surface temperature averaged over the period 1999-2008
4.2 Precipitation Comparisons
Precipitation patterns were varied among the four watersheds. Precipitation data
from the Deschutes watershed showed an insignificant, yet declining trend over the tenyear period, while the South Hood Canal, Puyallup and Nisqually Watersheds had
variable (rising and falling) trends within each watershed (Figure 10). The Nisqually and
Puyallup are glacially fed rivers, so much of the precipitation in the eastern reach of the
watersheds is from snowpack, while the South Hood Canal Watershed received very little
to no snow during the ten-year study period, and precipitation was almost solely in the
form of rain.
35
Sum Annual Precipitation (mm)
12000
10000
Skokomish
Puyallup
8000
Deschutes
Nisqually
6000
Linear (Skokomish)
4000
Linear (Puyallup)
Linear (Deschutes)
2000
Linear (Nisqually)
0
1998 2000 2002 2004 2006 2008 2010
Figure 10. Sum precipitation over the period 1999-2008 and over all four watersheds
All four watersheds show a spike in DOC loading during the 2006-2007 El Niño
year event. Flooding was reported along all four rivers during this particularly wet El
Niño. DOC loading showed an increasing trend during the ten-year period, (Figure 11)
with a slight dip in trend following the 2006-2007 El Niño.
Annual DOC Load Kg/day
25000
20000
Skokomish
15000
Nisqually
Deschutes
10000
Puyallup
5000
0
1999 2000 2001 2002 2003 2004 2005 2006 2007 2008
Figure 11. Average DOC loading per year in all four watersheds. Spikes in trendlines correspond
to El Niño events Southern Oscillation events
36
4.3 Temperature and DOC
Figure 12 shows regression analysis of mean surface temperatures against mean
annual DOC loading for all watersheds. In the combined analysis of all four watershed
loading values, warmer mean surface temperatures are negatively correlated with DOC
loading (R2= 0.69, p < 0.001).
Mean annual surface temperatures among all four sites was highly variable,
temperature ranged from 6°C to almost 12°C, and loading values were also much higher
in rivers with larger drainage basins (Puyallup drainage basin 2455 Km2, Nisqually 1339
Km2, South Hood Canal 705.37 Km2, Deschutes 420 Km2).
The strong correlation between cooler temperatures and DOC loading could be
due to regional differences, as well as differences in basin size. Statistical relationships
between temperature and DOC loading among all four watersheds is insignificant.
◊ = Puyallup
▼= Nisqually
● = South
Hood
Canal
* = Deschutes
Figure 12. Mean annual temperature and annual DOC loading in all four watersheds
37
4.3.1 Temperature and DOC Normalized by Basin Size
Normalized data yielded different results. Normalizing loading by watershed size
removes the possibility that basin size is skewing statistical results. Results of linear
regression using normalized DOC data and temperature showed no statistical significance
among the four watersheds (p=0.8829). Results from the full factorial ANOVA analysis
(F= 4.7781, p < 0.0584) suggest that changes in DOC loading trends as a function of
rising temperature are not present within each individual watershed.
◊ = Puyallup
2
▼= Nisqually
● = South Hood
Canal
* = Deschutes
Figure 13. Normalized annual DOC loading and mean annual temperature in all four watersheds
4.4 Precipitation and DOC
Total precipitation ranged from 116 mm per year as the lowest precipitation data
point in the Deschutes watershed, to 434 mm per year in the Puyallup watershed.
Precipitation remained consistent during the ten-year study period except for the
Deschutes watershed (figure 14). A significant drop in precipitation occurred in 2004 and
38
precipitation levels remained lower than pre-2004 levels for the duration of the study.
Linear regression results indicate that when all four watersheds are considered together,
there is a strong positive correlation between total annual precipitation and mean annual
DOC loading (R2= 0.57, p < 0.001).
Precipitation (mm)
Total Precipitation 1999-2008
500
450
400
350
300
250
200
150
100
50
0
Deschutes
Nisqually
Skokomish
Puyallup
1999 2000 2001 2002 2003 2004 2005 2006 2007 2008
Year
Figure 14. Total annual precipitation at all four watersheds 1999-2008
◊ = Puyallup
▼= Nisqually
● = South
Hood Canal
* = Deschutes
Figure 15. Total annual precipitation and DOC loading in all four watersheds
39
4.4.1 Precipitation and DOC Normalized by Basin Size
Normalizing watershed size and DOC loading by area (km2) yielded different
results with linear regression analysis and total precipitation between 1999-2008.
Normalized loading data were statistically insignificant (p= 0.0740). Individual
watersheds however, did show significant correlation between total precipitation and
DOC loading. Results from full factorial ANOVA analysis (F= 6.84, p= 0.0293) indicate
that within individual watersheds increasing DOC loads are positively correlated to
precipitation. The Deschutes watershed particularly stands out, as unlike the nonnormalized data, the Deschutes had a very strong correlation between increased
precipitation and loading.
◊ = Puyallup
2
▼= Nisqually
● = South
Hood Canal
* = Deschutes
Figure 16. Total annual precipitation at all four watersheds 1999-2008
4.5 Land Cover and DOC Correlations
Percent land cover and DOC loading showed strong correlations for developed
land cover (rs= 1.0; p=0.001) and wetland land cover (rs= 1.0; p=0.001). All other land
40
cover types are weakly correlated with DOC loading. Table 3 presents the results of the
Spearman’s Rho (ρ) analysis, as well as the p values associated with each land cover type
per watershed, and figures 17-20 depict the GIS results of the NLCD land cover types.
Land cover types were variable across all watersheds. Forest land cover was the
most dominant land type, (> 50% in all watersheds) while percentages of all other land
cover types varied (Table 4).
Variable
% Wetland
% Forest
% Agriculture
% Developed
% Glacial
% Rocky Slope
Spearman's ρ
1.00
-0.40
-0.20
1.00
-0.20
-0.80
p value
<0.0001
0.6000
0.8000
<0.0001
0.8000
0.2000
Table 3. Results from Spearman’s ρ correlation analysis
NLCD Land
Cover Type
Wetland
Forest
Agricultural
Developed
Glacial
Rocky Slope
Nisqually
Puyallup
Deschutes Skokomish
Percentage Percentage Percentage Percentage
1.08
55.65
35.4
4.13
0.09
3.65
1.82
67.16
5.74
21.51
3.09
0.69
3.71
42.77
27.69
25.54
0
0.29
1.66
69.78
3.56
17.69
1.77
5.54
Table 4. NLCD land cover types and percentage results
41
Figure 17. NLCD Land Cover results for the South Hood Canal Watershed
42
Figure 18. NLCD Land Cover results for the Nisqually River Watershed
43
Figure 19. NLCD Land Cover results for the Puyallup River Watershed and tributaries
44
Figure 20: NLCD Land Cover results for the Deschutes River Watershed
45
5.0 Discussion
Differences in mean annual DOC loading across all four watersheds is primarily
driven by the total amount of precipitation and size of watershed. Furthermore, wetland
and developed land cover class percentages are positively correlated with DOC loading in
all watersheds studied. Temperature was negatively correlated with DOC loading when
the data from all four watersheds was combined in this South Puget Sound study and did
not compare well with other studies in the literature (Tian et al., 2013; Evans et al., 2005)
where positive correlations between increasing temperatures and DOC loading were
found. The linear regression analysis of data not normalized by watershed size show that
watersheds with cooler temperatures have higher DOC loading while watersheds with
warmer annual temperatures have less. However, the negative correlation between
temperature and DOC loading us a spurious correlation, due to the observation that the
coolest watersheds are physically larger than the warmer watersheds. This will be
described in more detail below.
Once DOC loading is normalized by the size of the watershed, the relationship
between DOC and temperature is statistically insignificant across all watersheds. Major
differences within each watershed however, do show an increase of DOC loading as a
function of increasing temperature evident by an increasing trend line, however this
increase was not statistically significant using either Linear Regression or Factorial
ANOVA analyses. This was especially evident in the Deschutes watershed (Figure 13),
which is the smallest watershed, yet data points on the linear regression graph showed
increased loading correlated to warmer temperatures more so than the other watersheds.
46
The Deschutes watershed has the highest percentage of agricultural land class and
developed land cover compared to the other watersheds. It is possible that intensified
agricultural land use upstream of the heavily developed land near the mouth of the river
are sources of DOC in the Deschutes watershed during warmer years. Another possible
mechanism is that increased ambient temperatures could help facilitate riparian and
aquatic vegetation growth, and lengthen the growing season for aquatic algae, which
become DOC during decomposition.
5.1 Temperature and DOC
Unlike previous studies, (Tian et al., 2013; Evans et al., 2005) cooler temperatures
were more closely correlated with DOC loading, and warmer temperatures were not.
There are several mechanisms that could explain these results. The basin size of each
watershed is drastically different, as well as average flow throughout the seasons. The
Puyallup watershed had the coldest temperatures yet the largest DOC load and basin size
of all watersheds studied. The origin of the Puyallup, Carbon and White Rivers are from
glacial sources and the river experiences increased flow (cfs) during the spring snowmelt,
as well as increased precipitation during the fall wet season towards the mouth of the
river. Normalizing the data however shows that these results are due to the increase in
flow rather than the actual increase of DOC related to temperature.
Normalized data reinforce the weak correlation between DOC loading and
warmer temperature. Normalizing data show that DOC loading is staying constant
regardless of watershed size across all four watersheds, however DOC loading is
increasing as a function of temperature within each individual watershed, yet not enough
to yield statistically significant results. Analyzing these data on a longer time scale would
47
be prudent for detecting any significant relationships between DOC loading and
temperature within individual watersheds.
These results also highlight the difficulty in analyzing water quality parameters
for comparison across watersheds. Future studies should aim to examine watersheds that
are similar in size, flow and source of headwaters to further improve insight into how
temperature is affecting DOC loading in rivers draining into the Puget Sound.
Temperatures and climatic patterns are changing in the region because of climate change
(Wilhere et al., 2017; Gergel et al., 2017). Assessing the impact of climate change on
DOC loading is unclear from this study.
5.2 The importance of precipitation on DOC
This study highlights the strong effect that precipitation has on DOC loading in the
south Puget Sound. Basin size did not appear to be a confounding factor in the analysis of
how increased precipitation affects DOC loading, as is evident by analyzing normalized
data. The South Hood Canal basin is one of the smaller lowland watersheds analyzed in
this study (705.37 km2) and the Puyallup River basin the largest (2455 km2), yet annual
precipitation for the South Hood Canal watershed was comparable to annual precipitation
in the Puyallup.
Additionally, the regression analysis of DOC loading and precipitation (figures 15
and 16) showed an increasing trend in DOC load as precipitation increased across all
watersheds, regardless of basin size. Flooding is a normal occurrence in the South Hood
Canal Watershed along all the rivers that were included in the DEM delineation. Flooding
events likely result in larger quantities of DOC entering the rivers from upstream sources
48
within landscape, rather than from the river channel itself. Nonetheless, precipitation and
DOC loading are positively correlated among all sampling sites.
DOC loading also increased considerably during El Niño years (Figure 21). DOC
loading spiked during and following El Niño events, indicating that discharge is driving
DOC loading into the South Puget Sound. During intense precipitation events water
levels rise rapidly as terrestrial runoff inputs reach rivers at high volumes and increase
river discharge considerably (Budde, 2015). The input of terrestrial DOC into rivers
during El Niño events yields higher loads of DOC delivered to the Puget Sound. These
results indicate that storm events are particularly important for the transport of DOC to
the marine waters of the South Puget Sound.
Figure 21. Annual Precipitation and normalized DOC data with El Niño years circled
49
5.3 Land Cover and DOC
Annual measurements of normalized DOC loading in all four watersheds
demonstrated that wetland and developed land classes are correlated with increased
DOC loading. Other studies, (Laudon et al. 2011; Andersson and Nyberg, 2008)
discovered that wetland influence on DOC loading was substantial, that just 10% of
wetland land cover can explain as much as 50% of DOC concentrations throughout the
watershed. Likewise, Tian et al., (2013) quantified the effect of wetlands on DOC in
large and small watersheds and found a strong correlation in smaller watersheds, but
no difference in larger watersheds.
The results of this study are surprising, as all watersheds have such a small
percentage of wetland land cover compared to basin size. Results indicate that DOC
loading is positively correlated with wetland land cover, yet wetlands comprise less
than 4% of land cover in all basins studied (Table 4). Wetlands are rich in organic
carbon, and connectivity to major rivers throughout each watershed could result in
higher delivery of DOC to rivers, especially during El Niño events.
The positive correlation between DOC loading and developed land cover is also
surprising, as studies in the literature (Veeum et al., 2013; Oh et al., 2013; Stockmann
et al., 2013) showed positive correlations between agricultural land use and DOC
loads, however no studies were found that could explain mechanisms for how
developed land cover and higher DOC loads are correlated. Data collection throughout
the stream, including sampling locations throughout entire watersheds could help to
50
investigate and understand the mechanisms behind land cover correlations and DOC
loading.
Analyzing land cover as a percentage perhaps confounded the results of the
Spearman’s ρ calculation, and further studies should attempt to quantify the effect of
wetlands by statistically analyzing the number of wetlands throughout the watershed
rather than percentage. Additionally, a more representative sampling, by calculating
DOC loading throughout the watershed, and not only at the mouth, could yield more
definitive results in how land cover effects DOC loading throughout each watershed.
The sampling locations, at the mouth of each river are in developed areas.
Development and impervious surfaces could impact how much DOC is at the mouth of
each river, and how much is distributed further upstream.
These results also highlight the difficulty in analyzing water quality parameters for
comparison across watersheds. A larger sample size, analyzing more watersheds with
comparable basin sizes, flow and headwaters could have provided more insight into how
temperature is affecting DOC loading in rivers draining into the Puget Sound.
Temperatures and climatic patterns are changing in the region because of climate change
(Wilhere et al., 2017; Gergel et al., 2017). Assessing the impact of climate change on
DOC loading is unclear from this study.
6.0 Conclusion
This study is the first to analyze climate effects and land cover on DOC loading in
the South Puget Sound. Implementing methods used in other studies (Tian et al., 2013;
Mohamedali et al., 2011) yielded different results relative to that found in this study.
51
Basin size was variable throughout each watershed as well as timing of pulse flows from
snowmelt and seasonal precipitation was highly variable. Refining the methods for future
studies is recommended.
Precipitation is positively correlated with DOC loading, as results from the South
Hood Canal and the Puyallup Watersheds are comparable despite differences in each
basin. Increased precipitation appears to be a major contributing factor to DOC loading in
the four watersheds studied. Increased precipitation is expected to be attendant to climate
change in the South Puget Sound region, therefore further studied into how terrestrial
sourced DOC affects water quality in the tributary regions of these rivers is imperative
for water resource planning.
52
7.0 Bibliography
Ahmed, A., Pelletier, G., & Roberts, M. (2017). South Puget Sound flushing times and
residual flows. Estuarine, Coastal and Shelf Science, 187, 9-21.
Andersson, J. O., & Nyberg, L. (2008). Spatial variation of wetlands and flux of
dissolved organic carbon in boreal headwater streams. Hydrological
Processes, 22(12), 1965-1975.
Arandia-Gorostidi, N., Weber, P. K., Alonso-Sáez, L., Morán, X. A. G., & Mayali, X.
(2017). Elevated temperature increases carbon and nitrogen fluxes between
phytoplankton and heterotrophic bacteria through physical attachment. The ISME
journal, 11(3), 641.
Ault, T. R., Cole, J. E., Overpeck, J. T., Pederson, G. T., & Meko, D. M. (2014).
Assessing the risk of persistent drought using climate model simulations and
paleoclimate data. Journal of Climate, 27(20), 7529-7549.
Awale, R., Emeson, M. A., & Machado, S. (2017). Soil Organic Carbon Pools as Early
Indicators for Soil Organic Matter Stock Changes under Different Tillage
Practices in Inland Pacific Northwest. Frontiers in Ecology and Evolution, 5, 96.
Babson, A. L., Kawase, M., & MacCready, P. (2006). Seasonal and interannual
variability in the circulation of Puget Sound, Washington: a box model
study. Atmosphere-Ocean, 44(1), 29-45.
Balch, W. M., Drapeau, D. T., Bowler, B. C., & Huntington, T. G. (2012). Step-changes
in the physical, chemical and biological characteristics of the Gulf of Maine, as
documented by the GNATS time series. Marine Ecology Progress Series, 450,
11-35.
Banas, N. S., Conway-Cranos, L., Sutherland, D. A., MacCready, P., Kiffney, P., &
Plummer, M. (2015). Patterns of river influence and connectivity among
subbasins of Puget Sound, with application to bacterial and nutrient
loading. Estuaries and Coasts, 38(3), 735-753.
Bauer, J. E., Cai, W. J., Raymond, P. A., Bianchi, T. S., Hopkinson, C. S., & Regnier, P.
A. (2013). The changing carbon cycle of the coastal ocean. Nature, 504(7478),
61-70.
Bernstein, L., Bosch, P., Canziani, O., Chen, Z., Christ, R., & Riahi, K. (2008). IPCC,
2007: climate change 2007: synthesis report.
Bianchi, T. S. (2011). The role of terrestrially derived organic carbon in the coastal
ocean: A changing paradigm and the priming effect. Proceedings of the National
Academy of Sciences, 108(49), 19473-19481.
53
Bianchi, T. S., Garcia‐Tigreros, F., Yvon‐Lewis, S. A., Shields, M., Mills, H. J., Butman,
D., ... & Walker, N. (2013). Enhanced transfer of terrestrially derived carbon to
the atmosphere in a flooding event. Geophysical Research Letters, 40(1), 116122.
Bittar, T. B., Vieira, A. A., Stubbins, A., & Mopper, K. (2015). Competition between
photochemical and biological degradation of dissolved organic matter from the
cyanobacteria Microcystis aeruginosa. Limnology and Oceanography, 60(4),
1172-1194.
Budde, M. (2015). Interactive comment on “Impact of two different types of El Niño
events on runoff over the conterminous United States” by T. Tang et al.
Butman, D., Stackpoole, S., Stets, E., McDonald, C. P., Clow, D. W., & Striegl, R. G.
(2016). Aquatic carbon cycling in the conterminous United States and
implications for terrestrial carbon accounting. Proceedings of the National
Academy of Sciences, 113(1), 58-63.
Butman, D. E., Wilson, H. F., Barnes, R. T., Xenopoulos, M. A., & Raymond, P. A.
(2014). Increased mobilization of aged carbon to rivers by human disturbance.
Nat. Geosci. 8, 112–116.
Butman, D., & Raymond, P. A. (2011). Significant efflux of carbon dioxide from streams
and rivers in the United States. Nature Geoscience, 4(12), 839-842.
Butman, D., Stackpoole, S., Stets, E., McDonald, C. P., Clow, D. W., & Striegl, R. G.
(2016). Aquatic carbon cycling in the conterminous United States and
implications for terrestrial carbon accounting. Proceedings of the National
Academy of Sciences, 113(1), 58-63.
Carney, R. L., Seymour, J. R., Westhorpe, D., & Mitrovic, S. M. (2016). Lotic
bacterioplankton and phytoplankton community changes under dissolved organiccarbon amendment: evidence for competition for nutrients. Marine and
Freshwater Research, 67(9), 1362-1373.
Chapter 173-201A WAC, Water Quality Standards For Surface Waters Of The State Of
Washington. (2003). Washington State Department of Ecology, Olympia, WA.
Retrieved from:
https://fortress.wa.gov/ecy/publications/SummaryPages/173201A.html
Crandell, D. R., Mullineaux, D. R., & Waldron, H. H. (1958). PLEISTOCENE
SEQUENCE IN SOUTHEASTERN PART or THE PUGET SOUND
LOWLAND, WASHINGTON.
Cuo, L., Lettenmaier, D. P., Alberti, M., & Richey, J. E. (2009). Effects of a century of
land cover and climate change on the hydrology of the Puget Sound
basin. Hydrological Processes, 23(6), 907-933.
54
Curtis, P. J. (1998). Climatic and hydrologic control of DOM concentration and quality in
lakes. In Aquatic humic substances (pp. 93-105). Springer Berlin Heidelberg.
Deppe, R. W., Thomson, J., Polagye, B., & Krembs, C. (2013, September). Hypoxic
intrusions to Puget Sound from the ocean. In Oceans-San Diego, 2013 (pp. 1-9).
IEEE.
Dhillon, G. S., & Inamdar, S. (2013). Extreme storms and changes in particulate and
dissolved organic carbon in runoff: Entering uncharted waters? Geophysical
research letters, 40(7), 1322-1327.
Dinsmore, K. J., Billett, M. F., Skiba, U. M., Rees, R. M., Drewer, J., & Helfter, C.
(2010). Role of the aquatic pathway in the carbon and greenhouse gas budgets of
a peatland catchment. Global Change Biology, 16(10), 2750-2762.
Dodds, W. K. (2006). Eutrophication and trophic state in rivers and streams. Limnology
and Oceanography, 51(1part2), 671-680.
Evans, C. D., Jones, T. G., Burden, A., Ostle, N., Zieliński, P., Cooper, M. D., ... &
Freeman, C. (2012). Acidity controls on dissolved organic carbon mobility in
organic soils. Global Change Biology, 18(11), 3317-3331.
Evans, C. D., Monteith, D. T., & Cooper, D. M. (2005). Long-term increases in surface
water dissolved organic carbon: observations, possible causes and environmental
impacts. Environmental Pollution, 137(1), 55-71.
Fasching, C., Behounek, B., Singer, G. A., & Battin, T. J. (2014). Microbial degradation
of terrigenous dissolved organic matter and potential consequences for carbon
cycling in brown-water streams. Scientific reports, 4, 4981.
Feely, R. A., Alin, S. R., Newton, J., Sabine, C. L., Warner, M., Devol, A., Krembs, C &
Maloy, C. (2010). The combined effects of ocean acidification, mixing, and
respiration on pH and carbonate saturation in an urbanized estuary. Estuarine,
Coastal and Shelf Science, 88(4), 442-449.
Fichot, C. G., & Benner, R. (2014). The fate of terrigenous dissolved organic carbon in a
river‐influenced ocean margin. Global Biogeochemical Cycles, 28(3), 300-318.
France, R., Culbert, H., & Peters, R. (1996). Decreased carbon and nutrient input to
boreal lakes from particulate organic matter following riparian clearcutting. Environmental Management, 20(4), 579-583.
Fry, J., Xian, G., Jin, S., Dewitz, J., Homer, C., Yang, L., Barnes, C., Herold, N., and
Wickham, J., 2011. Completion of the 2006 National Land Cover Database for
the Conterminous United States, PE&RS, Vol. 77(9):858-864.
Gergel, D. R., Nijssen, B., Abatzoglou, J. T., Lettenmaier, D. P., & Stumbaugh, M. R.
(2017). Effects of climate change on snowpack and fire potential in the western
USA. Climatic Change, 141(2), 287-299.
55
Gray, A. N., Whittier, T. R., & Harmon, M. E. (2016). Carbon stocks and accumulation
rates in Pacific Northwest forests: role of stand age, plant community, and
productivity. Ecosphere, 7(1).
Guggenberger, G., Glaser, B., & Zech, W. (1994). Heavy metal binding by hydrophobic
and hydrophilic dissolved organic carbon fractions in a spodosol A and B horizon.
Water, Air, & Soil Pollution, 72(1), 111-127.
Hallegraeff, G. M. (1993). A review of harmful algal blooms and their apparent global
increase. Phycologia, 32(2), 79-99.
Hallock, D., 2009. River and Stream Water Quality Monitoring Report, Water Year 2008.
Washington State Department of Ecology, Olympia, WA. Publication No. 09-03041. www.ecy.wa.gov/biblio/0903041.html
Harvey, C. F., Swartz, C. H., Badruzzaman, A. B. M., Keon-Blute, N., Yu, W., Ali, M.
A., ... & Oates, P. M. (2002). Arsenic mobility and groundwater extraction in
Bangladesh. Science, 298(5598), 1602-1606. Hansell, D. A., & Carlson, C. A.
(Eds.). (2014). Biogeochemistry of marine dissolved organic matter. Academic
Press.
Hasler, C. T., Butman, D., Jeffrey, J. D., & Suski, C. D. (2016). Freshwater biota and
rising pCO2?. Ecology letters, 19(1), 98-108.
Hayhoe, K., Wake, C. P., Huntington, T. G., Luo, L., Schwartz, M. D., Sheffield, J., ... &
Troy, T. J. (2007). Past and future changes in climate and hydrological indicators
in the US Northeast. Climate Dynamics, 28(4), 381-407.
Homer, C., Dewitz, J., Fry, J., Coan, M., Hossain, N., Larson, C., Herold, N., McKerrow,
A., VanDriel, J.N., and Wickham, J. 2007. Completion of the 2001 National Land
Cover Database for the Conterminous United States. Photogrammetric
Engineering and Remote Sensing, Vol. 73, No. 4, pp 337-341.
Hood, E., Gooseff, M. N., & Johnson, S. L. (2006). Changes in the character of stream
water dissolved organic carbon during flushing in three small watersheds,
Oregon. Journal of Geophysical Research: Biogeosciences, 111(G1).
Hotchkiss, E. R., Hall Jr, R. O., Sponseller, R. A., Butman, D., Klaminder, J., Laudon,
H., ... & Karlsson, J. (2015). Sources of and processes controlling CO2 emissions
change with the size of streams and rivers. Nature Geoscience, 8(9), 696-699.
Howarth, R. W., Fruci, J. R., & Sherman, D. (1991). Inputs of sediment and carbon to an
estuarine ecosystem: Influence of land use. Ecological applications, 1(1), 27-39.
Huang, W., & Chen, R. F. (2009). Sources and transformations of chromophoric
dissolved organic matter in the Neponset River Watershed. Journal of
Geophysical Research: Biogeosciences, 114(G4).
56
Intergovernmental Panel on Climate Change. (2014). Climate Change 2014–Impacts,
Adaptation and Vulnerability: Regional Aspects. Cambridge University Press.
Johannessen, S. C., Potentier, G., Wright, C. A., Masson, D., & Macdonald, R. W.
(2008). Water column organic carbon in a Pacific marginal sea (Strait of Georgia,
Canada). Marine environmental research, 66, S49-S61.
Jones, S. E., & Lennon, J. T. (2015). A test of the subsidy–stability hypothesis: the
effects of terrestrial carbon in aquatic ecosystems. Ecology, 96(6), 1550-1560.
Kellerman, A. M., Dittmar, T., Kothawala, D. N., & Tranvik, L. J. (2014).
Chemodiversity of dissolved organic matter in lakes driven by climate and
hydrology. Nature communications, 5, 3804.
Khangaonkar, T., Sackmann, B., Long, W., Mohamedali, T., & Roberts, M. (2012).
Simulation of annual biogeochemical cycles of nutrient balance, phytoplankton
bloom (s), and DO in Puget Sound using an unstructured grid model. Ocean
Dynamics, 62(9), 1353-1379.
Khir-Eldien, K., & Zahran, S. A. (2017). Climate Changes Vulnerability and Adaptive
Capacity. The Nile River, 567-595.
Kirchman, D. L., Suzuki, Y., Garside, C., & Ducklow, H. W. (1991). High turnover rates
of dissolved organic carbon during a spring phytoplankton
bloom. Nature, 352(6336), 612-614.
Lampert, W. (1978). Release of dissolved organic carbon by grazing
zooplankton. Limnology and Oceanography, 23(4), 831-834.
Lange, M., & Gleixner, G. (2016, April). Plant diversity induces a shift of DOC
concentration over time-results from long term and large scale experiment.
In EGU General Assembly Conference Abstracts (Vol. 18, p. 8882).
Larsson, U., & Hagström, A. (1979). Phytoplankton exudate release as an energy source
for the growth of pelagic bacteria. Marine biology, 52(3), 199-206.
Leung, L. R., & Wigmosta, M. S. (1999). Potential climate change impacts on mountain
watersheds in the Pacific Northwest. JAWRA Journal of the American Water
Resources Association, 35(6), 1463-1471.
Littell, J. S., Mauger, G. S., Salathe, E. P., Hamlet, A. F., Lee, S. Y., Stumbaugh, M. R.,
... & Mantua, N. J. (2014). Uncertainty and extreme events in future climate and
hydrologic projections for the Pacific Northwest: providing a basis for
vulnerability and core/corridor assessments. Climate Impacts Group.
Luo, Y., Ficklin, D. L., Liu, X., & Zhang, M. (2013). Assessment of climate change
impacts on hydrology and water quality with a watershed modeling approach.
Science of the Total Environment, 450, 72-82.
57
MacCready, P. (1999). Estuarine adjustment to changes in river flow and tidal
mixing. Journal of Physical Oceanography, 29(4), 708-726.
Marcogliese, D. J. (2016). The distribution and abundance of parasites in aquatic
ecosystems in a changing climate: more than just temperature. Integrative and
comparative biology, 56(4), 611-619.
Mattsson, T., Kortelainen, P., Räike, A., Lepistö, A., & Thomas, D. N. (2015). Spatial
and temporal variability of organic C and N concentrations and export from 30
boreal rivers induced by land use and climate. Science of the Total
Environment, 508, 145-154.
May, C. W. (1997) The cumulative effects of urbanization on Puget Sound lowland
ecoregion. Puget Sound Research 1998. University of Washington.
McBean, E., Zhu, Z., & Zeng, W. (2010). Modeling formation and control of disinfection
byproducts in chlorinated drinking waters. Water Science and Technology: Water
Supply, 10(5), 730-739.
McDonald, C. P., Stets, E. G., Striegl, R. G., & Butman, D. (2013). Inorganic carbon
loading as a primary driver of dissolved carbon dioxide concentrations in the
lakes and reservoirs of the contiguous United States. Global Biogeochemical
Cycles, 27(2), 285-295.
Michalzik, B., Kalbitz, K., Park, J. H., Solinger, S., & Matzner, E. (2001). Fluxes and
concentrations of dissolved organic carbon and nitrogen–a synthesis for temperate
forests. Biogeochemistry, 52(2), 173-205.
Mohamedali, T., Roberts, M., Sackmann, B., & Kolosseus, A. (2011). Puget Sound
dissolved oxygen model nutrient load summary for 1999–2008. Washington State
Department of Ecology, Olympia, WA.
Moore, S. K., Mantua, N. J., & Salathé, E. P. (2011). Past trends and future scenarios for
environmental conditions favoring the accumulation of paralytic shellfish toxins
in Puget Sound shellfish. Harmful Algae, 10(5), 521-52.
Mote, P. W., & Salathe, E. P. (2010). Future climate in the Pacific Northwest. Climatic
Change, 102(1-2), 29-50.
Mote, P. W., Parson, E. A., Hamlet, A. F., Keeton, W. S., Lettenmaier, D., Mantua, N., ...
& Snover, A. K. (2003). Preparing for climatic change: the water, salmon, and
forests of the Pacific Northwest. Climatic change, 61(1), 45-88.
Naiman, R.J. Beechie, T.J. Benda, L.E. Berg, D.R. Bison, P.A. MacDonald, L.H.
O’Conner, M.D. … and Steel, E.A. (1992) Fundamental Elements of Ecologically
Healthy Watersheds in the Pacific Northwest Ecoregion. Watershed Management:
Balancing Sustainability with Environmental Change. 127-188.
Norton, D., D. Serdar, J. Colton, R. Jack, and Lester, D. (2011). Control of Toxic
Chemicals in Puget Sound: Assessment of Selected Toxic Chemicals in the Puget
58
Sound Basin, 2007-2011. Washington State Department of Ecology, Olympia,
WA.
Olson, M. B., Wuori, T. A., Love, B. A., & Strom, S. L. (2017). Ocean acidification
effects on haploid and diploid Emiliania huxleyi strains: Why changes in cell size
matter. Journal of Experimental Marine Biology and Ecology, 488, 72-82.
Pelletier, G., Bianucci, L., Long, W., Khangaonkar, T., Mohamedali, T., Ahmed, A., &
Figueroa-Kaminsky, C. (2017). Salish Sea Model: Ocean Acidification Module
and the Response to Regional Anthropogenic Nutrient Sources. Washington State
Department of Ecology, Olympia, Wa.
Qualls, R. G., & Haines, B. L. (1991). Geochemistry of dissolved organic nutrients in
water percolating through a forest ecosystem. Soil Science Society of America
Journal, 55(4), 1112-1123.
Raymond, P. A., & Bauer, J. E. (2000). Bacterial consumption of DOC during transport
through a temperate estuary. Aquatic Microbial Ecology, 22(1), 1-12.
Raymond, P. A., & Oh, N. H. (2007). An empirical study of climatic controls on riverine
C export from three major US watersheds. Global biogeochemical cycles, 21(2).
Raymond, P. A., Hartmann, J., Lauerwald, R., Sobek, S., McDonald, C., Hoover, M., ...
& Kortelainen, P. (2013). Global carbon dioxide emissions from inland
waters. Nature, 503(7476), 355-359.
Ritson, J. P., Graham, N. J. D., Templeton, M. R., Clark, J. M., Gough, R., & Freeman,
C. (2014). The impact of climate change on the treatability of dissolved organic
matter (DOM) in upland water supplies: a UK perspective. Science of the Total
Environment, 473, 714-730.
Roberts, M. (2014). Puget Sound and the Straits Dissolved Oxygen Assessment: Impacts
of Current and Future Human Nitrogen Sources and Climate Change Through
2070. Washington State Department of Ecology, Environmental Assessment
Program.
Roberts, M. L., & Bilby, R. E. (2009). Urbanization alters litterfall rates and nutrient
inputs to small Puget Lowland streams. Journal of the North American
Benthological Society, 28(4), 941-954.
SAS Institute Inc. 2015. JMP® 12 Multivariate Methods. Cary, NC: SAS Institute Inc.
Salathé Jr, E. P., Hamlet, A. F., Mass, C. F., Lee, S. Y., Stumbaugh, M., & Steed, R.
(2014). Estimates of twenty-first-century flood risk in the Pacific Northwest based
on regional climate model simulations. Journal of Hydrometeorology, 15(5),
1881-1899.
Sarmento, H., Morana, C., & Gasol, J. M. (2016). Bacterioplankton niche partitioning in
the use of phytoplankton-derived dissolved organic carbon: quantity is more
important than quality. The ISME journal, 10(11), 2582-2592.
59
Sheibley, R. W., Konrad, C. P., & Black, R. W. (2015). Nutrient attenuation in rivers and
streams, Puget Sound Basin, Washington (No. 2015-5074). US Geological
Survey.
Smith, M. W., Herfort, L., Fortunato, C. S., Crump, B. C., & Simon, H. M. (2017).
Microbial players and processes involved in phytoplankton bloom utilization in
the water column of a fast‐flowing, river‐dominated estuary. MicrobiologyOpen.
Spietz, R. L., Williams, C. M., Rocap, G., & Horner-Devine, M. C. (2015). A dissolved
oxygen threshold for shifts in bacterial community structure in a seasonally
hypoxic estuary. PloS one, 10(8), e0135731.
Stubbins, A., Hood, E., Raymond, P. A., Aiken, G. R., Sleighter, R. L., Hernes, P. J., ... &
Abdulla, H. A. (2012). Anthropogenic aerosols as a source of ancient dissolved
organic matter in glaciers. Nature Geoscience, 5(3), 198-201.
Sutherland, D. A., MacCready, P., Banas, N. S., & Smedstad, L. F. (2011). A model
study of the Salish Sea estuarine circulation. Journal of Physical
Oceanography, 41(6), 1125-1143.
Sutter, L. A., Chambers, R. M., & Perry, J. E. (2015). Seawater intrusion mediates
species transition in low salinity, tidal marsh vegetation. Aquatic Botany, 122, 3239.
Terrier, A., Girardin, M. P., Périé, C., Legendre, P., & Bergeron, Y. (2013). Potential
changes in forest composition could reduce impacts of climate change on boreal
wildfires. Ecological Applications, 23(1), 21-35.
Thingstad, T. F., HagstrÖm, Å., & Rassoulzadegan, F. (1997). Accumulation of
degradable DOC in surface waters: Is it caused by a malfunctioning
microbialloop?. Limnology and Oceanography, 42(2), 398-404.
Tian, Y. Q., Yu, Q., Feig, A. D., Ye, C., & Blunden, A. (2013). Effects of climate and
land-surface processes on terrestrial dissolved organic carbon export to major US
coastal rivers. Ecological engineering, 54, 192-201.
Turner, R. E., Rabalais, N. N., & Justic, D. (2008). Gulf of Mexico hypoxia: Alternate
states and a legacy. Environmental Science & Technology, 42(7), 2323-2327.
Van de Waal, D. B., Verspagen, J. M., Lürling, M., Van Donk, E., Visser, P. M., &
Huisman, J. (2009). The ecological stoichiometry of toxins produced by harmful
cyanobacteria: an experimental test of the carbon‐nutrient balance
hypothesis. Ecology letters, 12(12), 1326-1335.
Veum, K. S., Goyne, K. W., Motavalli, P. P., & Udawatta, R. P. (2009). Runoff and
dissolved organic carbon loss from a paired-watershed study of three adjacent
agricultural watersheds. Agriculture, Ecosystems & Environment, 130(3), 115122.
60
Wallace, R. B., Baumann, H., Grear, J. S., Aller, R. C., & Gobler, C. J. (2014). Coastal
ocean acidification: The other eutrophication problem. Estuarine, Coastal and
Shelf Science, 148, 1-13.
Wang, X., & Yin, Z. Y. (1997). Using GIS to assess the relationship between land use
and water quality at a watershed level. Environment International, 23(1), 103114.
Wear, E. K., Carlson, C. A., James, A. K., Brzezinski, M. A., Windecker, L. A., &
Nelson, C. E. (2015). Synchronous shifts in dissolved organic carbon
bioavailability and bacterial community responses over the course of an
upwelling‐driven phytoplankton bloom. Limnology and Oceanography, 60(2),
657-677.
Wetzel, R. G. (2001). Limnology: lake and river ecosystems. Gulf Professional
Publishing.
Wilhere, G. F., Atha, J. B., Quinn, T., Tohver, I., & Helbrecht, L. (2017). Incorporating
climate change into culvert design in Washington State, USA. Ecological
Engineering, 104, 67-79.
Windecker, L., Brzezinski, M. A., Wear, E., Carlson, C. A., & Passow, U. (2016,
February). Revising the release of fixed carbon in coastal phytoplankton: the role
of transparent exopolymer particles (TEP). In American Geophysical Union,
Ocean Sciences Meeting 2016, abstract# EC43A-04.
Winter, D. F., Banse, K., & Anderson, G. C. (1975). The dynamics of phytoplankton
blooms in puget sound a fjord in the northwestern united states. Marine
Biology, 29(2), 139-176.
Winterdahl, M., Bishop, K., & Erlandsson, M. (2014). Acidification, Dissolved Organic
Carbon (DOC) and Climate Change. In Global Environmental Change (pp. 281287). Springer Netherlands.
Wise, D. R., & Johnson, H. M. (2011). Surface‐Water Nutrient Conditions and Sources in
the United States Pacific Northwest. JAWRA Journal of the American Water
Resources Association, 47(5), 1110-1135.
Xenopoulos, M. A., Lodge, D. M., Frentress, J., Kreps, T. A., Bridgham, S. D.,
Grossman, E., & Jackson, C. J. (2003). Regional comparisons of watershed
determinants of dissolved organic carbon in temperate lakes from the Upper Great
Lakes region and selected regions globally. Limnology and Oceanography, 48(6),
2321-2334.
Yano, Y., Lajtha, K., Sollins, P., & Caldwell, B. A. (2004). Chemical and seasonal
controls on the dynamics of dissolved organic matter in a coniferous old-growth
stand in the Pacific Northwest, USA. Biogeochemistry, 71(2), 197-223.
61
Yasarer, L. M., Bingner, R. L., Garbrecht, J. D., Locke, M. A., Lizotte, R. E., Momm, H.
G., & Busteed, P. R. (2017). Climate Change Impacts on Runoff, Sediment, and
Nutrient Loads in an Agricultural Watershedin the Lower Mississippi River
Basin. Applied Engineering in Agriculture, 33(3), 379.
Yue, S., Pilon, P., & Cavadias, G. (2002). Power of the Mann–Kendall and Spearman's
rho tests for detecting monotonic trends in hydrological series. Journal of
hydrology, 259(1), 254-271.
Znachor, P., & Nedoma, J. (2009). Importance of dissolved organic carbon for
phytoplankton nutrition in a eutrophic reservoir. Journal of plankton research,
32(3), 367-376.
62
