The Evergreen State College Institutional Repository

Harris_FTMESThesis2017.pdf

Media

Part of The Influence of Land Use on Lowland Streams in the Puget Sound: A Case Studey from Carpenter Creek

extracted text (extracttext:extracted_text)

THE INFLUENCE OF LAND USE
ON LOWLAND STREAMS IN THE PUGET SOUND:
A CASE STUDY FROM CARPENTER CREEK

by
Fawn Trey Harris

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

©2017 by Fawn Trey Harris. All rights reserved.

This Thesis for the Master of Environmental Studies Degree
by
Fawn Trey Harris

has been approved for
The Evergreen State College
by

________________________
Richard Bigley, PhD
Member of the Faculty

________________________
Date

ABSTRACT
The influence of land use on lowland streams in Puget Sound:
A case study from Carpenter Creek
Fawn Trey Harris
Human activities and land use, including urbanization has degraded water quality and
habitats throughout the world. The small streams of the Puget Sound are increasingly
threatened by urbanization as the population in the Puget Sound expands. This research
investigates land use and spatial and temporal variation in water chemistry relationships
within Carpenter Creek, in the Puget Sound Basin of Washington State.
The physiography of stream sampling sites were found to be homogenous in several
landscape metrics, and therefore this study was unable to discern specific relationships
between water chemistry parameters and landscape metrics. An ANOVA revealed
significant seasonal differences in several water chemistry parameters, providing a
baseline for seasonal variation in water chemistry parameters for Puget Sound lowland
streams.
An ANOVA and Tukey’s Post Hoc test found significant differences in fecal coliform
contamination between seasons at each site, with the significant increases occurring
during the summer season. Fecal coliform means for the summer season exceeds the
maximum concentration for Washington State surface waters in salmon bearing streams.
This trend was observed during the dry season which contradicts current literature and
implies a possible underrepresented pollution pathway. The growing rates of urbanization
within the Puget Sound lowland, pose significant threats to endangered and threatened
salmonid species which rely on these habitats to complete their life history strategies.
Understanding how these land uses affect water quality can provide important
information for habitat managers.

Table of Contents
List of Figures .................................................................................................................................... i
List of Tables ................................................................................................................................... iii
Acknowledgements......................................................................................................................... iv
Introduction ..................................................................................................................................... 1
Literature Review ............................................................................................................................. 6
Land use change and water quality ............................................................................................. 6
Mechanisms of land use influence on water quality ................................................................... 7
Land use-water quality relationships and climate change ........................................................ 14
Current literature examining land use and water quality ......................................................... 15
Limitations ................................................................................................................................. 18
Methods ......................................................................................................................................... 23
Study Site ............................................................................................................................... 23
Water Quality Data Collection ............................................................................................... 25
Geographical Information Systems Data Collection .............................................................. 27
Statistical Analysis ...................................................................................................................... 30
Spatial Analysis........................................................................................................................... 34
Discussion ...................................................................................................................................... 37
References ..................................................................................................................................... 43
Appendices..................................................................................................................................... 48
Appendix 1 ................................................................................................................................. 48
Appendix 2. ................................................................................................................................ 48
Appendix 3. ................................................................................................................................ 49

i

List of Figures
Figure 1 Conceptual model of anthropogenic stream stressors.
......................................................................................................................................................... 2
Figure 2 Changes in the ratio of groundwater, interflow, evatranspiration, and surface runoff
based on land cover in Western Washington.
......................................................................................................................................................... 4
Figure 3 Water cycle within an urbanized watershed.
....................................................................................................................................................... 11
Figure 4 Relationship between watershed urbanization (%TIA) and biological integrity in Puget
Sound lowland (PSL) streams.
....................................................................................................................................................... 13
Figure 5 Annual Precipitation for Kingston, Washington.
....................................................................................................................................................... 23
Figure 6 Aerial map of Carpenter Creek and surrounding wetlands located on the Kitsap
Peninsula in Kingston, Washington.
. ...................................................................................................................................................... 24
Figure 7 Elevation map of Carpenter Creek monitoring locations and surrounding areas.
....................................................................................................................................................... 29
Figure 8 Line graph illustrating fecal coliform trends in Carpenter Creek from 2001-2016.
. ...................................................................................................................................................... 32
Figure 9 Bargraphs illustrating total discharge (cfs), dissolved oxygen (mg/L),and pH means and
standard deviations by site.
. ...................................................................................................................................................... 33
Figure 10 Scatter plot of discharge (cfs) by Site and Season
....................................................................................................................................................... 33
Figure 11 Scatter plot of fecal coliform (colonies/100mL) by Site and Season
....................................................................................................................................................... 33
Figure 12 Map of land use types within a 200-m buffer of Carpenter creek.
....................................................................................................................................................... 35
Figure 13 Map of percent of imperviousness within the Carpenter Creek basin for the years 2001
and 2006
....................................................................................................................................................... 37

ii

List of Tables
Table 1 Table of ANOVA and Kruskal-Wallis results for dissolved oxygen, temperature, pH,
discharge, and fecal coliform by Site and Season.
....................................................................................................................................................... 31
Table 2 Table of parameter means and standard error for each monitoring site by season.
....................................................................................................................................................... 32
Table 3 Geographical Information Systems data sources, data files, and landscape metrics used
in analysis
....................................................................................................................................................... 36

iii

Acknowledgements
I would like to take this opportunity to sincerely thank my thesis advisor, Richard Bigley PhD, for
providing continued support, feedback, and laughs throughout my entire thesis process.
Richard, it has been a pleasure to work with in the classroom, in the field, and throughout this
entire project. Thank you for sharing your knowledge with me over the last two years. I can’t
thank you enough for your contribution to my education during my time in the MES program.
To all the amazing MES faculty, who have supported my education and encouraged me along
the way, I am forever grateful. There were times where some of you challenged me to my full
potential and inspired me to work even harder than I thought possible (Peter Dorman).
To my MES cohort, you are all amazing and thank you for your continued efforts to make this
world a better place. Thank you for your encouragement and support throughout this journey.
To the Stillwaters Environmental Center, namely Joleen Palmer, Naomi Maasberg, and Jenise
Bauman, thank you for giving me the opportunity to utilize the organization’s data, computers,
resources and software to complete my thesis project. Thank you also for your continued
encouragement, generosity, compassion, and everything you do for the community. I hope that
this research might provide some useful information for future restoration actions in Carpenter
Creek.
To my partner, Sean Harris, and children Shun-la-ta Smith and Katana Harris, thank you for being
so supportive of my studies throughout this time. While I can’t ever get back the hours I was
away for school or work, I hope that I will be able to contribute something special to the world
that will have made it all worth it.
And to close, a sincere and absolute thank you to my family, dearest friends, and my community
who have stood behind me and have supported every step of my education. I could not have
done any of this without the support and love of several amazing people.

iv

Introduction
Our waterways and their biotic communities are in peril due to increasing
urbanization and anthropogenic land use change. There is a substantial body of literature
documenting that watershed urbanization is associated with substantial alterations in flow
patterns, channel morphology, water quality, and biotic communities (Sun and Lockaby
2012, Ding et al. 2016, Bunn and Arthington 2002, Booth 2004, May 1997, Lenat and
Crawford 1994, Poff et al. 1992). The small streams of the Puget Sound have been
especially affected by land use alteration and urbanization (May 1997, Morley and Karr
2002, Booth et al. 2005). However, the relationships between urbanization and water
quality in these streams are relatively understudied. As urban landscapes continue to
expand, it will be increasingly important to understand how these urban water systems
function. With forecasts of increasing population and changes in precipitation resulting
from climate change, the relationships between land use and water quality will likely take
on increasing importance and need further exploration.
Human land use change can affect the physical and chemical properties of nearby
aquatic habitat, by altering the groundwater exchange with surface water (Sun and
Lockaby 2012, Allan and Castillo 2007, Hayashi and Rodenberry 2002). Land use
change is often associated with the replacement of natural vegetation with impervious
surfaces, reduction in transpiration and filtration within the drainage basin, and increases
in urban runoff frequency and intensity (Sun and Lockaby 2012, Allan and Castillo
2007). There are several mechanisms through which land use can affect water quality
including: sedimentation, nutrient enrichment, contaminant pollution, hydrological
alteration, and riparian clearing/canopy opening (Allan 2004). Urbanization exacerbates
these mechanisms and plays a significant role in decreasing the water quality of urban
1

streams. Exploring the relationships that influence these mechanisms will prove
beneficial to the restoration of degraded urban streams.

Figure 1 Conceptual model of anthropogenic stream stressors. From Booth et al. 2004.

Poff et al. (1997) describes how flow regime influences water quality, energy
sources, physical habitat, and biotic interactions and therefore influences the ecological
integrity of the system. Water quality and quantity is extremely important to the life
history strategies of many aquatic organisms. Several aquatic species use water chemistry
indicators to trigger different life cycle events. For example, streamflow plays a
significant role in the life history strategies of many fishes (Booth 2004), with life history
strategies linked directly to flow regimes (e.g. phenology of reproduction, spawning
behavior, larval survival) (Welcomme 1985; Junk et al. 1989; Copp 1989, 1990; Sparks
1995, Humphries et al. 1999 as referenced by Bunn and Arthington 2004). Therefore, any
alteration in the natural flow regime could dramatically impact the health and survival of
some fish species. Salmon species are especially susceptible to changes or fluctuations in
2

certain water quality parameters such as temperature (Carter 2008). Over the last several
decades, salmonid populations in the Pacific Northwest have been declining due to the
cumulative effects of land use practices, agriculture and urbanization (May 1997).
Studying these land use practices and their relationships to water quality parameters is an
important step in beginning to restore these salmon populations and their habitats.
The most common effect of urbanization within a river catchment is alteration of
the hydrological flows (Sun and Caudwell 2015, Sun and Lockaby 2014). Urban areas
within river catchments can alter the river’s hydrology through sediment deposition,
erosion, alteration in evapotranspiration, alteration of interflow, and groundwater
permeability (Figure 2). Impervious surfaces and loss of natural ground cover
significantly alter the landscape, and disrupt natural hydrologic interactions. The amount
of imperious surface in urban areas is a primary contributor to the alteration of flow
characteristics in urban streams. Roads, rooftops, and lawns are all impervious surfaces
and increase the transportation of pollutants into water systems.
Structures constructed in-stream (e.g., dams, culverts, weirs, ladders) and along
riparian habitats effect the flow of water and the movement of energy within the system.
In the Puget Sound, anthropogenic changes to in-stream habitat have created a habitat
which is very different from the habitat where the aquatic species evolved (May 1997),
hereby making it more and more difficult for endangered and threatened species to
recover. The increasing urbanization within Puget Sound lowland streams, will continue
to alter the natural flow regime and the biological integrity of the stream.

3

Figure 2 Changes in the ratio of groundwater, interflow, evapotranspiration, and surface runoff based on land cover in
Western Washington. From Sheldon et al. 2005)

Nutrient enrichment and pollutant contamination are also common in urban
streams. It is well documented that nutrient enrichment is associated with agricultural
land use, human activities, and waste systems. Urban areas contribute to physical and
chemical pollutant contamination which severely stresses the biological integrity of
surrounding stream systems. Road way runoff and runoff from impervious surfaces are
the primary pathways for nutrients (e.g. nitrogen, phosphorous, organic material) and
pollutants (e.g. inorganic compounds, heavy metals, bacteria) to contaminate a stream
system. Studies have shown that Washington State has higher highway runoff rates than
the national average (Herrera Consultants 2007). Although research has shown that the
land use type is a direct contributor to nutrient and chemical pollution (Neilsen et al.

4

2012) there is little research exploring these relationships in Pacific Northwest lowland
streams.
With forecasts of a changing precipitation and its seasonality with climate change
(Pachauri et al. 2014), understanding land use and water quality relationships will be
extremely important to providing the best available science to policy makers and natural
resource managers. The future of Puget Sound stream systems and salmonid populations
depend on a thorough understanding of the mechanisms affecting water quality in
lowland streams. This research intends to; build on current literature about land use/water
quality relationships in small order streams, provide information about urbanization
effects on water quality in urban streams, provide insight to current restoration managers
actively working on the site, and build upon existing literature on Puget Sound lowland
streams.

5

Literature Review
Land use change and water quality

Growing urbanization in the Pacific Northwest threatens the health of many urban
watersheds and the health of the associated biological communities. Human population
growth in the Pacific Northwest carries a host of environmental threats; including
imperviousness, urban sprawl, contaminants, pollution pathways and the modification of
stream and riparian habitats. Urbanization can have several effects on watersheds such as;
the alteration of natural hydrology, loss of riparian habitat, and the alteration of water
chemistry. In the Puget Sound lowland ecoregion, small streams and associated wetlands
are the ecosystems most effected by urbanization (May 1997). Many of these ecosystems
are critical habitat for migrating, rearing, and spawning salmonids (May 1997), as well as
with important habitat for several other Pacific Northwest species. Understanding how
different land uses affect water chemistry parameters can help better inform natural
resource managers and urban planners to prepare for environmental resilience when
facing climate change.
Most land use changes have the potential to affect the structure and function of
surrounding aquatic ecosystems, floodplains, and watersheds (Ecology 2015, Allan and
Castillo 2007, Poff et al. 1997, Naiman et al. 1999, Allan and Castillo 2007). Land use
changes, rather than in-stream structures, are the primary cause of alteration in the natural
flow regime due to changes in sediment delivery, decreases in soil infiltration, and
increased runoff (Poff et al. 1997). Land use changes not only negatively alter the natural
hydrology and water chemistry, but they can also alter the life processes of several
aquatic organisms. Some organisms, such as salmonid species, are extremely susceptible
to changes in water quality.
6

Stream systems are largely affected by disturbances within their watersheds,
whether or not the disturbance occurs in close proximity of the stream. There are several
mechanisms through which land use can affect stream habitat and water quality. Allan
(2004) has grouped these mechanisms into the following categories: sedimentation,
nutrient enrichment, contaminant pollution, hydrological alteration, riparian
clearing/canopy opening and loss of woody debris. Of the water quality parameters
effected by land use, this research plans to examine fecal coliform, nutrient cycling,
temperature and streamflow. The water quality parameters selected for this study are
significant to aquatic biota, ecosystem functioning, and human health. Understanding
how these parameters are affected by different land uses can assist planners and scientists
in preparing these local systems for ecological resilience in the face of climate change.

Mechanisms of land use influence on water quality
Sedimentation

Sedimentation is a natural occurrence in aquatic systems, but increases or
decreases in the natural sediment load can affect the physical attributes of the system and
the lifecycles of the aquatic biota. The movement of sediment through an aquatic system
is essential to the formation of channel morphology and creation of substrate.
Sedimentation is critical for the healthy ecological functioning of aquatic and terrestrial
communities near stream and river systems (Naiman et al. 1992). Land use change poses
significant threats to a stream’s natural flow regime and associated biota due to alteration
in the sediment flows and transport (Poff et al. 1997). Urbanization and land use change
pose significant threats to the natural flow regime and the water quality within an aquatic
system.
7

Sediment movement and accumulation plays a large role in the presence and
concentration of certain pollutants in surface water. Many pollutants tend to bind with
fine particulate material, organic matter, and sediments; allowing for easy transport of the
pollutants throughout an aquatic system (May 2009). Because of their low solubility in
water, Polycyclic Aromatic Hydrocarbons (PAHs) and many organic compounds are
transported to water systems during high intensity rainfall events and other storm water
events (Herrera 2007). Similar to organic compounds, bacteria and nutrient
concentrations are also commonly associated with sediments due to their ability to bind
to sediments (Herrera, 2007). Alteration and removal of natural vegetation communities
and forest stands also increases the sedimentation to the watershed (Sun and Caudwell
2015) and reduces soil infiltration (Poff et al. 1997). Increases in pollutants due to
growing urbanization and alteration of sedimentation regimes, poses significant threats to
the biodiversity within urban stream systems.
Sedimentation plays a large role in the structure and function of stream systems,
and growing urbanization increases the risk of sedimentation instability and the
likelihood of pollutant transport. Sediment is often associated with pollution due to the
ability to work as a transport service for organic materials. The concentration of sediment
in the water column and size of sediments can have direct effects on water quality
parameters including; lowering the dissolved oxygen content, increasing temperature,
increasing turbidity, altering the nutrient cycle and streamflow (Sun and Cauldwell 2015,
Sun and Lockaby 2012, May 1997, Poff et al. 1997). Urban imperviousness increases
sedimentation deposits and subsequent pollution loads, creating a significant need for

8

growing the wealth of knowledge pertaining to urban land use-water quality
relationships.
Nutrient Enrichment

There are several environmental and anthropogenic pathways in which nutrients
can enter aquatic systems. Environmental nutrient pathways include atmospheric
deposition, organic matter decomposition, groundwater, and soil inputs (Allan and
Castillo 2007). Anthropogenic land use change can create additional nutrient pathways
through the modification of riparian habitats, extension of impervious surface, clear
cutting, stream modification and urban development (Sun and Lockaby 2012, Allan and
Castillo 2007, Booth 2004). Nitrogen in watershed catchments has been positively
correlated with agriculture and urban lands, and negatively correlated with increasing
forest cover (Allan and Castillo 2007). Increasing the abundance and density of
agriculture and urban areas is expected to increase the nutrient concentrations of nearby
streams and watersheds.
Nutrients are an increasing concern to aquatic ecosystems, due to their ability to
stimulate plant and algal growth. Eutrophication and hypoxia can occur when the
dissolved oxygen content in water has been compromised due to excessive algal growth.
Nutrients are usually measured in ammonium nitrate, nitrate, nitrite, total Kjedahl
nitrogen, and total nitrogen (Herrera Consultants 2007). Many aquatic organisms are
adversely affected by high nutrient concentrations and low dissolved oxygen levels. In
recent years, the Puget Sound has seen significant increases in the frequency and severity
of hypoxic events along shorelines and estuaries (Ecology 2017). As the climate in the
Puget Sound continues to change, it will be become increasingly important that scientists
understand the sources of pollution and their effects on surrounding watersheds.
9

Understanding the influences of urbanization and land use on nutrient pollution can assist
urban planners and restoration scientists to prepare for resilience in a time of climatic
change.
Contaminant Pollution

Runoff from impervious surfaces is a large pollution pathway, serving as a
transport vector for agricultural fertilizers, atmospheric deposition and nitrogen from car
exhaust (Herrera Consultants 2007). Impervious surfaces are described here as,
“anthropogenic land use change that have resulted in impermeable land cover (e.g.
rooftops, road ways, and lawns)”. Impervious surfaces allow for chemicals and nutrients
to gather and be washed into surrounding watersheds with rainfall and first flush events
(Allan and Castillo 2007, Allan 2004). Increasing urbanization in watersheds often has
negative effects on freshwater biota through the anthropogenic development of natural
habitat into impervious surface and the introduction of pollution pathways. As land cover
increases to greater than 50% imperviousness, the biological integrity and water
chemistry of streams becomes highly compromised (May 1997). The higher than
average nutrient levels and growing urbanization in western Washington indicate a need
for thorough understanding of these local systems. The unique geomorphology of the
Puget Sound region, implies an urgent need for more research on these relationships in
Puget Sound lowland streams.
In addition to runoff containing high levels of nutrients, organic compounds, and
metals; it also serves as a pathway for bacterial infections to enter water systems.
Measuring fecal coliform bacteria can help assess the risk of a water body to a
bacteriological contamination. Fecal coliform bacteria are not directly harmful to humans
but is used an indicator of potential fecal contamination (Herrera Consultants 2007).
10

Urban streams tend to have elevated levels of fecal coliform and e. coli. bacteria
compared to their non-urban counterparts (Sun and Lockaby 2015). High levels of fecal
coliform have been positively correlated with the presence of pathogens which are known
to cause human illness (KCHD 2015). In Kitsap County, point and non-point sources of
fecal coliform bacteria come from failing septic systems, combined sewer systems,
agricultural waste, food waste and storm water drainage pipes (KCHD 2015). Runoff and
chemical contaminants are a serious problem in Puget Sound urban streams, with little
research available about the different land use variables and their effects on different
water quality parameters.

Figure 3 Water cycle within an urbanized watershed. From Sun and Lockaby (2012).

11

Hydrologic Alteration

Land use change within and near streams can have severe effects on a stream’s
natural flow regime. The primary land use activities that can affect the natural flow
regime include timber harvest, agriculture, urbanization, and livestock (Poff et al. 1997).
Many watersheds in the Puget Sound region have experienced many, if not all, of these
land use activities throughout the last several hundred years. Any alteration to these
stream conditions can potentially alter the stream’s hydrogeology causing severe longterm effects such as erosion, deposition, and flooding. Alteration of the hydrological
regime can also affect the ability of aquatic organisms to establish or may affect the
overall composition of existing habitat.
Streamflow is essential to many of the stream’s characteristics and plays an
important role in the life history strategies of several aquatic organisms. Streamflow is
variable and differs widely based on location and several other contributing factors. The
quantity, timing, and temporal patterns of streamflow are extremely variable and
influence the physical, biological, and chemical conditions of the stream (Allan and
Castillo 2007). Flow is ultimately derived from precipitation and influenced by climate,
geography, soil type, topography, and vegetation (Poff et al. 1997). Climate change and
projected urbanization increases are expected to have several effects on streamflow of
Puget Sound urban streams. Morley and Karr (2002) found significantly lower biological
integrity in urban streams that experienced a high degree of flow flashiness. Analyzing
the relationships between land use and streamflow in low order urban streams could be
essential in preparing these systems for ecological resilience for climate change.

12

Figure 4 Relationship between watershed urbanization (%TIA) and biological integrity in Puget Sound lowland (PSL)
streams. The benthic index of biotic integrity (B-IBI) and the abundance ratio of juvenile coho salmon to cutthroat
trout were used as indices of biological integrity. From May 1997.

Riparian Clearing/Canopy Opening

Healthy and intact riparian areas are extremely important to the structure and
function of all stream and river systems. The quality of the riparian habitat is essential for
the healthy functioning of a stream ecosystem (Wang 2001). The alteration of riparian
habitat has the ability to alter stream temperature, quantity and character of dissolved
organic carbon reaching the stream, sedimentation, shade availability and bank stability
(Allan 2004). Any alteration in the riparian corridor has great potential to negatively
impact stream water chemistry.

13

Many land use changes near streams are associated with riparian clearing and
canopy opening. Clearing riparian vegetation and opening the canopy creates a pathway
for light energy to reach the stream, subsequently reducing shade and increasing
temperature (Allan 2004). Increases in temperature may influence the growth of harmful
algal blooms. In addition to the temperature effects, riparian clearing is linked to erosion
and sedimentation which can alter dissolved oxygen content and the concentration of
pollutants (Allan 2004). Removing native riparian plant species may also effect the
riparian zone’s ability to filter harmful contaminants and pollutants from storm water
runoff before entering the stream.
Although riparian areas effect water quality, several studies have found that the
influence of land use extends much further than the riparian area (Utz et al 2016, Nielsen
et al. 2012, Pratt and Cheng 2012, Morley and Karr 2002). Booth et al. (2014) found that
percent urbanization in the watershed is not the most important factor influencing
temperature in urban streams of Puget Sound, but that many factors such as canopy
cover, contribute to the overall stream temperatures. The work of Booth et al. (2014) and
others illustrate the need for more research on land use effects on water quality.
Understanding land use effects at multiple scales may help better inform water resource
managers on the scale and intensity of land use-water quality relationships.

Land use-water quality relationships and climate change

Urban development in the Puget Sound has increased drastically in the last several
decades with populations continuing to grow. Growing urbanization, coupled with
climate change, is expected to further alter the water chemistry from baseline values,
posing threats to aquatic habitats and organisms. Changes in climate (precipitation and
14

temperature) are expected to adversely impact surface water chemistry (Murdock et al.
2000). Increased risk of drought and changes in precipitation patterns are expected to
increase the effects of urbanization on water quality (Sun and Caudwell 2015). Due to
these expected outcomes, it is increasingly important for water quality and land use
relationships to be studied, especially in areas with unique topography such as the Puget
Sound.

Current literature examining land use and water quality

Previous applications
While many of the mechanisms of how water quality might be affected by land
use have been investigated, the specific relationships that exist between land use and
water quality parameters remain understudied. The existing literature on land use and its
effect on water quality parameters have been limited in temporal and spatial scope.
Relationships between agricultural land use and water quality are perhaps the most
studied of all land use patterns. The effects of urbanization on water quality has also been
well documented in recent years. However, many researchers describe a need for more
research investigating these relationships at multiple spatial and temporal scales, as well
as in diverse geomorphic conditions.
Agricultural land use and its relationship to water quality parameters has been
commonly studied and has been attributed to increased nutrient levels in nearby
watersheds (Lenat and Crawford 1994, Allan and Castillo 2007, Nielsen et al. 2012, Ding
et al. 2016). Nitrogen and phosphorous concentrations have been associated with
fertilizers and animal waste from agricultural areas (Allan and Castillo 2007). The
development of agriculture land in watersheds provides many non-point pollution sources
15

of nutrients to the aquatic habitats (Allan 2004). Ding et al. (2016) found that poor, water
quality was associated with high patch densities of cropland, orchards, and agriculture.
Agricultural areas are also associated with poor lentic habitat quality (Allan 2004) and
lower invertebrate species diversity (Lenat and Crawford 1994). Although there have
been a few studies examining the relationships between agricultural land use and water
quality, most of these studies are limited to a specific geographic region or their sampling
methods are limited.
Urbanization, usually defined as areas of impervious surfaces, has been linked to
many different effects on water quality. The health of a stream has been directly
correlated to the percentage of impervious surfaces within a catchment (Alberti et al.
2007, Morley and Karr 2002, Arnold and Gibbons 1996). Impervious surfaces influence
the frequency and amount of pollution reaching aquatic systems (Utz et al. n.d.). Utz et
al. (n.d.) describes that impervious surface is a fundamental landscape stressor. Storm
water pollution is of interest in urban areas, and some studies have noted urbanization’s
contributions to storm water pollution. Storm water is often linked to increased nutrients,
sedimentation, and heavy metals in surrounding watersheds (McCarthy et al. 2008, May
1999). Many aspects of urbanization influence the flow and movement of pollutants into
a water system through the creation of new pollution pathways and creating new sources
of pollution.
Past and present land uses have been found to affect stream health (Maloney and
Weller 2011). Yu et al. (2013) performed statistical and spatial analyses with water
quality and land use data collected from Shenzhen watershed in China, and found strong
correlations between increasing urbanization coverage and decreasing water quality. The

16

Puget Sound is a mosaic of small streams and rivers, increasing urbanization in these
areas is expected to have direct effects on the water quality in nearby watersheds.
Understanding how these relationships effect different types of streams with different
geomorphic conditions is important to the future of water quality issues.
Some research from the Puget Sound region has looked at the biological integrity
of urban streams by sampling for benthic invertebrates (Morley and Karr 2002, Alberti et
al. 2007). Both studies used benthic invertebrate sampling and the Benthic Index of
Biological Integrity (B-IBI) to characterize the effect of land use on stream water quality.
Trends in this research found that biological integrity decreases as urbanization increases,
but water chemistry parameters were not sampled in these studies. Morley and Karr
(2002) did measure stream flow, and found B-IBI to be correlated with flow fluctuation.
Although these studies are useful in understanding the extent of urbanization effects on
stream biota in the Pacific Northwest, the results do not provide information regarding
water quality and land use relationships.
Many studies in the existing literature utilize Geographic Information Science
(GIS) and water quality data to spatially analyze land use and water quality relationships.
Tu et al. (2011) studied the relationship between impervious surface density and
percentage of land use type to water quality parameters. The results of Tu et al. (2011)
found that the impact of land use on water quality differed between parameter and the
level of urbanization in the watershed. Because of the various environmental factors that
can influence water quality, site specific information is important to local agencies and
scientists.

17

River and stream systems are very diverse in biological, chemical, and physical
characteristics. Because each stream and watershed has several unique attributes, it is
often difficult to extrapolate data from one stream and apply it to another, even if several
physical factors are similar (e.g. catchment size, drainage areas, land use type). With this
respect, it may also be difficult to compare a stream to a reference stream due to the
unique watershed characteristics of individual streams. Researchers suggest that some of
the variability in water quality measurements is based on the geomorphic features of a
particular watershed (Yu et al. 2013, Ding et al. 2016). Due to the range of geomorphic
variability and its influence on land use-water quality relationships, it is important to
study these relationships over a wide range of geomorphic types and features. This
research looks to explore the land use-water quality relationships in Puget Sound lowland
streams, a very understudied system, the Kitsap Peninsula.

Limitations

Although land use and water quality relationships have been of interest to many
researchers for the past several decades, there are several limitations to the existing
studies. Many studies are restricted to a specific geographical region or place in time,
which may or may not be relevant to the Pacific Northwest. In the existing literature,
several studies lack large sets of water quality sampling data or use short or snapshot
sampling events. Much of the time, the lack of sampling data is directly related to the
lack of funding for monitoring. Unfortunately, this can lead to the failure to recognize
trends in the data (Type I Error). Large datasets over a long temporal scale are the most
efficient way to study land use-water quality relationships.

18

A recent study by Ding et al. (2016), examined land-use and water quality
parameters from multiple spatial locations but from only one point in time. Another
recent study in Shenzhen, China used statistical and spatial analyses to examine land use
and water quality (Yu et al. 2013), but the study was again limited in the amount of data
collected. The data collected for Shenzhen was limited to a two-year period, which can
lead to analysis errors by failing to recognize seasonal and climatic relationships. Two
years is a relatively small amount of time to notice trends in data, especially when water
quality can be affected by seasonal climate changes.
Another limitation in the current literature is the origin of the data and the method
of collection used by several studies. To promote accuracy and precision, the water
quality collection methods and analysis should be the same at each sampling location. A
study conducted by Pratt and Chang (2012), used water quality data collected by multiple
agencies, each using a different collection methods and protocols. Haidary et al. (2013)
examined differences in water quality parameters between different land uses but used 24
different wetlands for the study, all of which could have different pollution pathways and
pollution influences. Each wetland has its own unique geomorphology which influences
certain water quality parameters either directly or indirectly. Comparing land use-water
quality relationships from multiple spatial locations is not as accurate as comparing land
use-water quality relationships from the same spatial location over a period of time. The
most effective way to assess relationships at the land-water interface, is to monitor
parameters over a long temporal range in the same spatial location.
One study, Zhou et al. (2012), collected an impressive 18 years of data to examine
land use and water quality data for the Dongjiang River. While this study provides useful

19

insight into the scale of land use patterns and their relationships to water quality, the
researchers averaged and clumped the water quality data into time periods lasting several
years. In an effort to mitigate the effect of precipitation on flow and water quality, the
researchers choose to examine parameters collected during the dry season only (Zhou et
al. 2012). Analyzing parameters collected during the dry season and negating those
collected during the wet season fails to identify several relationships that might exist
seasonally, or even worse, may fail to identify much larger data trends. Zhou et al. 2012
grouped data into multiple year periods, completely failing to examine the relationships
that might exist with season or year. There are several factors that must be accounted for
when attempting to analyze parameters influenced by environmental trends. Considering
that water quality is influenced by biological, chemical, environmental, and physical
factors; it is pertinent that these mechanisms be investigated in their entirety.
Another limitation in the current literature, is the lack of existing research on low
order streams. Low order streams-first and second order streams- play significant roles in
aquatic ecosystems yet their importance is often overlooked in research. Although small
in size, low order streams provide the same ecosystem services as larger streams and
rivers. Low order streams are highly susceptible to land-use disturbances because they
are highly interconnected to the surrounding landscape (Freeman et al. 2007 as
referenced in Ding et al. 2016). Although low order streams make up a significant portion
of the world’s rivers and streams, there is very little existing research on low order
streams (Ding et al. 2016). Currently, the only research on land use and water quality in
low order streams was conducted in the monsoon ecosystems of China. Ding et al. (2016)
discovered that water quality in low order streams is most affected by land use

20

configuration. While Ding et al. (2016) found significant trends in their research, there is
still a substantial need for more research on land use and water quality relationships in
low order streams.
Of the reviewed research, only a few studies were found investigating water
quality and/or land use and in the Puget Sound lowland streams (Luce et al. 2014,
Shandas and Alberti 2008, Alberti et al. 2007). However, these studies were all from
different geographical locations around the Puget Sound, with none from the Kitsap
Peninsula. For example, Shandas and Alberti (2008), examined water quality and land
use metrics in eight watersheds in the Puget Sound, and found that upland riparian habitat
had a large influence on water quality. However, this study was conducted in streams on
the western slopes of the Cascade mountains, which has very different topography than
the Kitsap Peninsula. Luce et al. (2014), studied stream temperate variability in the
Pacific Northwest, but again the sampling locations consisted of many mountainous
streams and relatively few lowland streams. The locations of these streams have different
environmental influences than streams on the Kitsap Peninsula and therefore the
relationships between land use and water quality are possibly very different between
geographical location. Many of the streams and rivers in Washington State originate from
melting snowpack and glaciers, and the literature available reflects this, with little
available research on groundwater fed streams in the Puget Sound.
The literature review found one study from the Kitsap Peninsula investigating
stream water chemistry, however this study had very different parameters than those
being examined here. Researchers from Stanford University, conducted a study
investigating if herbicides, pharmaceuticals and personal care products had pathways to

21

surface water other than septic systems. Dougherty et al. (2010) sampled creeks and
groundwater in the Liberty Bay Watershed on the Kitsap Peninsula, and found that the
specified compounds were being released into the environment by sources other than
waste water. While this research is not investigating the same relationships, this study
provides important information about the water chemistry influences to Kitsap Peninsula
groundwater fed streams. In the Puget Sound, shorelines and areas with less than 50 feet
above sea level have generally been found to be groundwater recharge aquifers (Vacarro
et al. 1998 as referenced by Dougherty et al. 2010). It is possible that some water
pollution sources in Puget Sound lowland streams are due to contamination of the
groundwater recharge aquifers, perhaps this research may provide some insight into these
mechanisms.
A review of the current literature suggests that there is a significant and pressing
need for more research on land-use and water quality relationships, especially on the
topographically and geographically unique, Kitsap Peninsula. While many researchers
have found notable relationships existing between land use variables and water quality
trends, many of these same researchers have expressed the spatial and temporal variance
in the data and the need for more research. In the Puget Sound, much of the existing
literature on land use and water quality is focused around storm water runoff and its
effects on the salmonid populations. This literature review did not reveal any studies
examining land use-water quality relationships Puget Sound lowland streams, with
almost no research on Kitsap County low order streams. The current research describes
that many variables influence water quality and therefore site specific information is
often the most useful in understanding these relationships. This research will build upon

22

existing literature and explore land use effects on water quality in an urban, Puget Sound
lowland stream.

Methods
Study Site

Carpenter Creek is a 2.9-mile stream located within the Foulweather
Bluff/Appletree Cove watershed in Kingston, Washington. The creek is a groundwater
fed stream, originating at approximately 280 feet above sea level. Carpenter Creek is a
second-order stream, joined by Trillium Creek approximately 210 m downstream from
Carpenter Lake. The elevation of the stream decreases steadily from the headwaters to the
mouth of the stream, with the largest declines near the headwaters. The primary soil type
of the Carpenter Creek drainage basin was found to be advanced outwash, a highly
permeable soil. The creek empties into Appletree Cove before draining into the Puget
Sound. The total watershed encompasses an area of 1,886 acres (KPHD 2014) of the
Kitsap Peninsula.

Figure 5 Annual Precipitation for Kingston, Washington. Data from
http://www.idcide.com/weather/wa/kingston.htm

23

Figure 6 Aerial map of Carpenter Creek and the surrounding wetlands. Located on the Kitsap Peninsula in Kingston,
Washington. Stillwaters Environmental Center Carpenter Creek water quality monitoring sites are illustrated.

24

Carpenter Creek and the Carpenter Creek estuary provides important habitat to
migrating and rearing salmonids and forage fish. There are several salmonid species
which utilize Carpenter Creek including: coho (Oncorhynchus kisutch), chum
(Oncorhynchus keta), cutthroat trout (Oncorhynchus clarkii), and chinook
(Oncorhynchus tshawytscha) (Azerrad 2012). Chinook is currently listed as a threatened
species in the Puget Sound (UFWS 2015). Carpenter Creek falls under the WAC 173201A-600, where any surface waters not listed on table 602-Carpenter is not listed- shall
be, “protected for the designated uses of: Salmonid spawning, rearing, and migration. . .”
and is also recognized and protected for uses as “Core summer salmonid habitat; and
extraordinary primary contact recreation” (KPHD 2014). The estuary provides a haven
for aquatic species, serving as the last functioning estuary on the east side of the Kitsap
Peninsula before leaving the waters of the Puget Sound.

Water Quality Data Collection

Water quality data was obtained from the Stillwaters Environmental Center in
Kingston, Washington. Stillwaters sampled water quality monthly, at three sampling
locations in Carpenter Creek, from 2001-2016 (Figure 6). Site 1 is located just north of
the Carpenter Creek saltmarsh at (N 47°47’54.6966, W -122°30’48.6354) and is
surrounded by mixed conifer stands of predominately, Douglas fir (Pseudostuga
menzessii) and Western red cedar (Thuja plicata). Site 2 is located approximately 400 m
north of Site 1 at (N 47°48.9958’, W -122°30’45.4824) and is surrounded by similar
mixed conifer stands, and is in close proximity of a small housing development. Site 3 is
located upstream of Site 1 and Site 2 at (N 47°48’34.816, W -122°31’15.132) just north
of Bond Road, the main highway into Kingston. Site 3 is also surrounded by mixed
25

conifer stands. The sampling locations are all located within the City of Kingston’s Urban
Growth Area and have varying levels of urbanization.
Several water chemistry parameters were sampled at each sampling location.
However, this study only used data from the following water quality parameters:
temperature, pH, dissolved oxygen, discharge, nitrate, phosphate, and fecal coliform.
These parameters were chosen based upon the ability to effect in-stream biota. All
parameters, except fecal coliform, were recorded in the field by a team of trained
volunteers utilizing established protocols. Grab water samples are taken for fecal
coliform and sent to an accredited laboratory. Field data was recorded onto field
datasheets and later transcribed into a digital format. For quality assurance purposes, the
digital data was cross checked with field datasheets for inaccuracies.
Grab water samples were collection using WA State Department of Ecology
Method EAP030 for Fecal Coliform and sent to Kitsap County Health Department for
analysis. The grab water samples were taken in sterile bottles provided by the lab. Water
samples were taken before field measurements to avoid contamination. After collection,
the bottle is stored in a cool, dark cooler before transport to the lab. A second water
sample, using the same protocols, was collected for field analysis of nitrate and
phosphate. Phosphates and nitrates were measured using LaMotte water test kits,
Phosphate Model VM12-Code 4408 and Nitrate Model NCR-Code 3110 (LaMotte,
Chestertown, MD).
Digital multiparameter meters were used to obtain data for dissolved oxygen, pH
and discharge (stream flow). Dissolved oxygen was measured using a YSI 200 Model
Dissolved Oxygen meter (YSI Incorporated, Yellow Springs, OH). PH was measured

26

using a multi-parameter meter, Hanna Model HI98129 (Hanna Instruments, Woonsocket,
RI). Stream flow was measured using a current velocity meter, Swoffer Meter Model
2100-B (Swoffer Incorporated, Federal Way, WA). Digital readings from the meters were
recorded on field datasheets.

Geographical Information Systems Data Collection

Geological and land use data for Carpenter Creek and the surrounding watershed
were collected from local and national Geographical Information Systems (GIS) digital
databases. Digital Elevation Maps (DEM) were obtained from the Kitsap County GIS
database and downloaded into ArcGIS. Elevation data was extracted from the DEM
raster datasets to create slope and elevation contour layers using ArcGIS Spatial Analyst.
The mean, minimum, and maximum elevation and slope for each sampling location and
the surrounding riparian area was calculated from this data. The slope and elevation did
not differ vary significantly between sites.
Geologic data was obtained from the United States Geological Survey (USGS)
GIS datasets. The USGS National Soils Database (NSD) GIS geodatabase was used to
obtain soil classifications for the Foulweather Bluff/Appletree Cove watershed. The
USGS National Hydrology Database (NHD) GIS geodatabase was used to obtain
hydrological and wetland data for the study watershed. GIS data layers for the watershed
were created from the geodatabases.
Land use data was obtained from the Coastal Change Analysis Program (C-CAP)
regional land cover database and Kitsap County’s Comprehensive Plan landcover data.
C-CAP is a nationally standardized database of 25 land use classifications at different
time intervals, at 30-m resolution. C-CAP datasets from 2001 and 2006 were obtained for
27

use in this study. These datasets were compared to Kitsap County’s Comprehensive Plan
dataset to account for the most recent land uses located within the watershed. Land use
classifications from both datasets were simplified and grouped into four categories:
forested, urban, residential, and wetland. Although Kingston has served as a place of
agriculture in the past, the existing land uses no longer reflect an agricultural
classification.
To depict the most accurate data on the extent of urbanization within the
watershed, road and Census data were obtained from the Kitsap County GIS data
website. Road line layers were used to calculate road density within drainage basin and
the stream buffer area. The number of single family taxlots (SFT) were obtained from
Census TRACT 2010 data from the Washington State Geospatial Portal. Single family
taxlots were used to measure the extent of imperviousness within the stream buffer area.
The year-built dates for the SFT were used to calculate the amount of urban growth
within the stream since 2001 (the beginning of the water sampling data). The C-CAP
dataset was also used to examine the amount of imperviousness throughout the watershed
and riparian habitat.

28

Figure 7 Elevation map of Carpenter Creek monitoring locations and surrounding areas.

29

Results
Statistical Analysis

The water chemistry data was analyzed for missing data and outliers; extreme
outliers were omitted from the data analysis. The water quality datasets provided by the
Stillwaters Environmental Center were mostly complete with very few missing data
points, missing data points were replaced with season parameter means. Most of the
missing data were due to equipment malfunction or the inability to access the sampling
site. A fourth sampling location was omitted from this study due to a difference in the
length of sampling time. Monitoring did not start at the fourth site until 2004, while
monitoring at the first three sites started in 2001. Several of the water chemistry
parameters did not follow a normal distribution and were logged transformed before data
analysis.
All statistical analyses were performed using R 3.1 software. A One-Way
Analysis of Variance (ANOVA) or Kruskal-Wallis test were used to look for differences
in water chemistry parameter means between Site, Season, Site and Season, and Percent
Land Use Type. Tukey’s Post-Hoc was performed to determine where the differences
occurred. Significant differences were found in mean dissolved oxygen between Site and
Season (F(6,537) =4.14, p<0.05). Differences in dissolved oxygen means were found
between every site during every season (p<0.05). Significant differences were found in
mean stream temperature between Site and Season (F (6,338) =11.45, p<0.05). Differences
in temperature means between each sampling site were found during the Fall and Spring
seasons (p<0.05), there was no difference in temperature between site during Summer
and Winter seasons. Significant differences were also found in mean pH between Season
F(2,537) =10.08, p<0.05and Site (F (6,537) =47.57, p<0.05). Significant differences in pH
30

were found between Sites 2 and Site 3 during all seasons (p<0.05), and differences
between Site 1 and Site 3 during the winter season (p<0.05). There are significant
differences in discharge means between Site and Season (F (6,537) =3.39, p<0.05).
Differences in discharge means were found between all Sites during the winter season
(p<0.05). Significant differences were found in fecal coliform means between all sites
during the summer Season (F (6,537) =2.15, p<0.05). The summer season fecal coliform
means were all above the Washington State maximum for salmon-bearing streams
(Figure 8).
Table 1 Table of ANOVA and Kruskal-Wallis results for dissolved oxygen, temperature, pH, discharge, and fecal
coliform by Site and Season. Statistically significant results are denoted with **. There are significant differences
between dissolved oxygen by site and season (F (2,3,6) = 4.136, p=0.000451), temperature by site and season (F (2,3,6)
=11.454, p<0.05), pH by site (F (2) = 2.505, p<0.05), pH by season (F (3) =10.083, p <0.05) discharge by site and season (F
(2,3,6) =3.394,p<0.05), and fecal coliform by season (F (3) =46.627, p<0.05).

31

Table 2 Table of parameter means and standard error for each monitoring site by season. Parameters measured are
dissolved oxygen (mg/L), temperature (C ), pH, nitrate (ppm), phosphate (ppm), discharge (cfs), and fecal coliform
(colonies/100mL), respectively. A Tukey’s Post Hoc test was performed and statistically significant differences are
donated by *. There are significant differences in dissolved oxygen, temperature, and fecal coliform means between all
sites and seasons. There are significant differences in pH between all sites during winter months, and pH means are
significantly different at Site 2 during all seasons. There are significant differences in discharge means between Site 1
and Site 3 during the winter season. There are no significant differences in nitrate and phosphate means between site
and season.
Site
1
1
1
1
2
2
2
2
3
3
3
3

Season
Fall
Spring
Summer
Winter
Fall
Spring
Summer
Winter
Fall
Spring
Summer
Winter

DO mn
9.06*
10.04*
8.73*
1.68*
7.59*
8.9*
7.82*
9.0*
10.36*
10.85*
9.63*
12.42*

±std
0.14
0.14
0.10
0.17
0.14
0.20
0.16
0.23
0.14
0.20
0.14
0.28

Temp
9.74*
9.99*
13.91*
4.91*
8.11*
11.77*
14.65*
NA
10.85*
9.47*
14.25*
5.53*

±std
0.42
0.37
0.23
0.22
0.42
0.34
0.21
0.25
0.38
0.31
0.21
0.26

pH mn ±std NO3 mn
7.1 0.1 0.32
7.2 0.1 0.27
7.2 0.1 0.34
6.8* 0.1 0.37
6.9* 0.1 0.31
6.9* 0.2 0.27
7.0* 0.1 0.31
6.5* 0.1 0.46
7.5 0.1 0.34
7.6 0.1 0.33
7.5 0.1 0.85
7.4* 0.1 0.38

±std PO43- mn
0.07
1.17
0.04
1.00
0.06
0.99
0.07
1.10
0.05
1.00
0.02
0.99
0.03
1.14
0.16
1.14
0.05
1.00
0.03
1.00
0.51
1.05
0.06
0.95

±std Discharge mn
±std FC mean ±std
0.12 1.31
0.40
na
16.65
0.04 2.31
0.31 48.8* 12.69
0.04 0.59
0.08 141.04* 19.24
0.12 3.8*
0.54
na
4.41
0.02 1.04
0.35
na
49.67
0.05 2.64
0.43
na
22.86
0.06
na
0.12 267.18* 49.14
0.13
na
0.64
na
6.24
0.01 0.79
0.21 144.4* 47.03
0.02 1.68
0.38 70.29* 21.95
0.03 0.55
0.09 267.19* 49.14
0.03 1.17*
0.17 22.45* 5.97

Figure 8 Line graph illustrating fecal coliform trends in Carpenter Creek from 2001-2016. Several measurements
exceed Washington State maximum criterion for fecal coliform (50 colonies/100 mL).

32

Figure 9 Bar graphs illustrating total discharge (cfs), dissolved oxygen (mg/L),and pH means and standard deviations by site. An ANOVA
was used to determine if there were significant differences in parameter means between site. A Tukeys Post Hoc test was used to look
where the differences occurred, statistically significant differences are denoted by differences in lettering.

Figure 10 Scatter plot of Dissolved oxygen (mg/L) by Site and Season.

Figure 11 Scatter plot of fecal coliform (colonies/100mL) by Site and Season. There is a significant increase in fecal coliforms during
the summer months at all sites.

33

Spatial Analysis

ArcGIS (ESRI 2015) was used to determine land use composition, percent land
use and surface geology characteristics of the study creek and surrounding watershed.
Elevation and slope were extracted from 30-m resolution DEMs provided by Kitsap
County GIS databases. Geologic and hydrological data were obtained from the USGS
GIS databases. Shapefiles and raster files were projected on NAD 1983 StatePlane
Washington State North FIPS geographic coordinate system with a Lambert Conformal
Conic projection. Land use shapefiles were provided by Kitsap County and the C-CAP
project, these data were used to categorize land use types within the watershed and
buffered stream section.
Using ArcGIS geoprocessing tools, a 200-m buffer was created around Carpenter
Creek and its tributary. The buffer was created to understand the cumulative effects of
riparian land use on water chemistry parameters within a stream reach. For each water
sampling location, stream sections were delineated from the buffer, extending from the
sampling location upstream to the headwaters. Each stream section represents the
cumulative drainage area of the corresponding upstream sections. The average slope of
each sampling site was found to be about 3.1%. While the sites differed in elevation, the
difference was insignificant, with all site elevations between 10-40 feet above sea level.

34

Figure 12 Map of land use types within a 200-m buffer of Carpenter creek. Stream sections are denoted by outlines.

35

Land use composition layer files were clipped to stream section layer files to
calculate the percent composition of land use within each stream section. Land use
shapefiles were compared with SFTs to account for residential land use on a small scale,
and were found to align with SFTs with no adjustments needed. There were no
significant differences between land use composition percentage between sites. Each site
was found to have less than 20% forested land, with the majority land use urban or
residential. SFT shapefiles were used to calculate the percent of residential and urban
land use growth during the sampling period. It was found that urbanization has increased
by 11% within the 200-m buffer of the stream during the sampling period. Road
shapefiles were used to create road density maps, with road densities high within 200-m
of all sampling sites. C-CAP 2001 and 2006 impervious data was used to create maps
illustrating the growth of imperviousness within the watershed. Impervious surface raster
data was only available for years 2001 and 2006, therefore impervious surface
percentages were calculated using land use cover datasets.

Table 3 Geographical Information Systems data sources, data files, and landscape metrics used in analysis
Source
Data
Kitsap County
Transportation Data

Landscape Metrics
Street Density
Population Density
(count/m2), Single Family
Kitsap County
Population
Taxlots (SFT)
Surface Geology Soil Types, %
USGS
National Soils Database
permability, % impermability
Total Stream Length (m),
USGS
National Hydrology Database Stream Density (km/m2)
% Imperviousness, %
Forested, % Urban, %
C-CAP
C-CAP 2001, C-CAP 2006
Residental, % Wetland
Elevation, Mean Elevation,
Min Elevation, Max Elevation,
Mean Slope, Min Slope, Max
Kitsap County
DEM
Slope, % Slope
Forested, Urban, Residental,
Ecology
Comprehensive Land Use Plan Wetland

36

Figure 13 Map of percent of imperviousness within the Carpenter Creek basin for the years 2001 and 2006. The
amount of imperviousness has increased rapidly within the five-year period.

Discussion
Statistical analyses showed significant differences in several water chemistry
parameters between site and season for the 15 years of available data. Seasonal variation
in water chemistry parameters is largely attributed to seasonal variation in precipitation
events and human land use activities. Differences in parameters between site could be
influenced by the physical characteristics of each site or a variety of geochemical
properties that may be site specific, such as groundwater infiltration.
Dissolved oxygen (DO) was the most affected parameter, with significant
differences in DO between every site and every season. DO is directly related to
37

temperature through water solubility, therefore it is expected that differences in DO are
directly related to differences in temperature. Temperature was found to be significantly
different between all sites during the fall and spring seasons. The differences in DO and
temperature means between season can be directly related to seasonal variation and
precipitation influence. Significant differences were found in discharge and fecal
coliform between all sites during the winter and summer months, respectively. The
differences in discharge during the winter months can related to the different site
geography and the associated flow at each location. It is hypothesized, that the high
concentrations of fecal coliform at all sites during the summer season are related to land
use practices including the use of fertilizers, and lower flows. These results are novel,
and should encourage increased examination of the seasonal variation of urbanized Puget
Sound lowland streams.
The percentage of land use did not differ significantly between site and therefore
made model analysis unreliable. Multiple regression models did not show any
relationships between land use category percentage and water chemistry measurements
within the study scale. in the scope of the study examined land use-water quality
relationships at the riparian buffer scale, and therefore the scale was restricted to a 200-m
buffer around Carpenter Creek and its tributaries. Because this study was restricted in
scope, the relationships between land use and water chemistry in lowland streams may
not have been fully explained. The relatively small size of Carpenter Creek and relatively
uniform geological and land use characterization of the surrounding riparian areas, limit
the ability of statistical and spatial analysis to establish relationships between land use
and water chemistry variables. Although the data obtained this study may not be

38

sufficient in understanding land use and water chemistry relationships in Puget Sound
lowland streams, it still provides valuable information about the characteristics of water
parameters in these streams.
The data gathered and analyzed in this case study helps build upon existing
literature on the water chemistry of small, urban, Puget Sound lowland streams. Although
direct relationships between land use variables and water quality were not established in
this research, trends in water quality chemistry between seasons in small urban lowland
streams were identified. The trends identified in this study can be used to providing
researchers with background information on urban lowland streams in Puget Sound. This
research compliments existing literature (McCarthy et al. 2008, May 2009, Maloney and
Weller 2011) which illustrates that urbanization influences the flow and movement of
pollutions through a water system.
Although direct relationships between urbanization and other land uses were not
established, the long history of fecal coliform contamination throughout the stream,
displays a potential relationship between urbanization and increasing fecal coliforms.
Interestingly, studies investigating fecal coliform concentrations in runoff in western
Washington found high concentration loads associated with periods of high precipitation
(Herrera Consultants 2007), whereas the data in this study show the highest fecal
contaminations occur within Carpenter Creek during the summer season, where
precipitation is at its lowest. This contradicts the current literature, and suggests that the
high fecal coliform concentrations in Carpenter Creek are highly influenced by some
mechanism other than precipitation. However, the factors affecting bacterial
concentrations in roadway runoff in Western Washington have not been fully

39

investigated (Herrera Consultants 2007). A study conducted by Kelsey et al. (2004) found
that high concentrations of fecal coliforms were often found near septic systems. This
study did not consider the location of septic systems. Given current literature, the high
concentrations of fecal coliform during the summer season are probably directly related
to human activity and this research could help provide insight on how to mitigate this
water pollution.
The main limitation of this study, was the inability to complete some spatial
analysis due to a lack of variation in geographical, geological, and land use data. The lack
of significant differences between percent land use, soil type, precipitation, and elevation
between sampling locations, made it difficult to run statistical models. The selected
sampling sites were too physically similar to properly assess variation between site and
make significant inferences about land use influence on the water chemistry. To mitigate
this problem in future studies, researchers should use more sampling locations with
different land use percentages and compare them. Perhaps, more sampling points along
the stream would increase the variation between sampling location and may increase
statistical and spatial analyses. In addition, urbanization grew 11% within the stream
buffer system during the sampling period, this rapid urbanization of riparian areas could
heighten the difficultly in assessing relationships between water chemistry and land use
patterns.
Although this study may have limitations to describing land use and water quality
relationships, the findings of this study help provide a significant amount of background
information on urbanized Puget Sound lowland streams. This study also helps highlight
the need for more research on bacteria pollution pathways in the Puget Sound and

40

groundwater pollution pathways. Future studies should be conducted on land use
variables and Puget Sound lowland stream water chemistry. It is possible that several of
the water chemistry parameters are affected by groundwater processes and may not be
recognizable through surface water measurements.

Conclusion
As urbanization continues to grow within our watersheds, it may become more
and more difficult to analyze the relationships between land use and water chemistry.
Although this study did not reveal a land use relationships to water quality, over 70% of
the study sites are classified as ‘residential’ or ‘urbanized’ land use. Therefore, some
trends in water quality may be related to land use, but they remained unnoticed due to
homogeneity among sample sites. Past studies have emphasized the difficulty in
analyzing land use relationships due to complex hydrological processes and the influence
of past land use practices (Maloney and Weeler 2011). Continued anthropogenic
alteration of the landscape, makes it extremely difficult to assess how different land use
changes are influencing water quality. Stream systems, large and small, are influenced by
several variables and are constantly in motion. Therefore, it may be difficult to study any
true relationships existing between land use and water quality, unless the land use and
activities around the stream stay static for the entire sampling period.
While land use and water chemistry relationships were not defined in this study,
this research provides valuable information on water chemistry in highly urbanized Puget
Sound lowland streams. This research found significant trends in fecal coliform
concentrations and dissolved oxygen. Fecal coliform trends were significantly higher
during summer seasons (dry season), which contradicts past research which found fecal

41

coliform trends highest during high precipitation events. Trends seen in fecal coliform
concentrations in Carpenter Creek exceed the state maximum for salmon bearing waters
and imply a need for local planners to address this issue. This research implicates the
need for further research into land use and water quality relationships in Puget Sound
lowland streams, particularly groundwater fed streams of the Kitsap Peninsula, as well as
a significant need to address the issue of bacterial pollution in Carpenter Creek.
Urbanization within watersheds has been found to have negative effects on the
biological integrity of a stream system. Even though relationships between land use
factors and specific water chemistry parameters remain largely unidentified, this research
has identified water chemistry parameters that may be of interest for investigation in
future studies. This study also investigated the seasonal water quality in Puget Sound
lowland streams. In addition, the results of this research implicate that highly urbanized
lowland streams may be especially susceptible to bacterial pollution. This may become
especially important with climate change and the increased risk of pathogen transmission.
Future studies should investigate these relationships further to establish a thorough
understanding of these complex Puget Sound lowland streams.

42

References
Alberti, M., Booth, D., Hill, K., Coburn, B., Avolio, C., Coe, S., & Spirandelli, D. (2007).
The impact of urban patterns on aquatic ecosystems: An empirical analysis in Puget
lowland sub-basins. Landscape and Urban Planning. 80, 345-361.
Allan, J. D. (2004). Landscapes and Riverscapes: The Influence of Land Use on Stream
Ecosystems. Annual Review of Ecology, Evolution & Systematics, 35(1), 257–284.
https://doi.org/10.1146/annurev.ecolsys.35.120202.110122
Allan, J.D. and Castillo, M.M. (2007). Stream Ecology: Structure and Function of
Running Waters. 2nd Edition. Springer Publishing. Dordrecht, The Netherlands.
Arnold Jr., C. L., & Gibbons, C. J. (1996). Impervious surface coverage. Journal of the
American Planning Association, 62(2), 243.
Azerrad, J. (2012). Washington Department of Fish and Wildlife. A Newsletter for
Washington’s Professional Planning Community. Retrieved from
http://wdfw.wa.gov/publications/01401/
Booth, D.B. (2005). Challenges and prospects for restoring urban streams: a perspective
from the Pacific Northwest of North America. Journal of North American Benthological
Society 24(3), 724–737. http://dx.doi.org/10.1899/04-025.1
Booth, D. B., Kraseski, K. A., & Rhett Jackson, C. (2014). Local-scale and watershedscale determinants of summertime urban stream temperatures. Hydrological Processes,
28(4), 2427–2438. https://doi.org/10.1002/hyp.9810
Bunn, S.E., and Arthington, A.H (2002). Basic Principles and Ecological Consequences
of Altered Flow Regimes for Aquatic Biodiversity. Environmental Management 30(4),
492–507. DOI: 10.1007/s00267-002-2737-0
Carter, K. (2008). Effects of Temperature, Dissolved Oxygen/Total Dissolved Gas,
Ammonia, and pH on Salmonids: Implications for California's North Coast TMDLs.
California Regional Water Quality Control Board, North Coast Region.
Ding, J. Yaun, J. Liu, Q. Zhaojiang, H. Liao, J. Lan, F. and Peng, Q. (2016) The
influence of land use pattern on water quality in low order streams of the Dongjiag River
Basin, China: A multiscale analysis. Science of the Total Environment. 552, 205-216.
Dougherty, J.A., Swarzenski, P.W., Dinicola, R.S. & Reinhard, M. (2010). Occurrence of
herbicides, pharmaceuticals, and personal cares products in surface water and ground

43

water around Liberty Bay, Puget Sound, Washington. Journal of Environmental Quality .
39, 1173–1180. doi:10.2134/jeq2009.0189
Ecology (2015). Washington’s Water Quality Management Plan to Control Nonpoint
Pollution. Washington State Department of Ecology.
https://fortress.wa.gov/ecy/publications/documents/1510015.pdf
Ecology (2017). Map of occurrences of hypoxia in Puget Sound. Washington Department
of Ecology. Retrieved from
http://www.ecy.wa.gov/programs/eap/Nitrogen/PugetSoundHypoxiaMap.html
ESRI (2015) ArcGIS. www.ESRI.com
Haidary, A., Amiri, B. J., Adamowski, J., Fohrer, N., & Nakane, K. (2013). Assessing the
Impacts of Four Land Use Types on the Water Quality of Wetlands in Japan. Water
Resources Management, 27(7), 2217–2229. https://doi.org/10.1007/s11269-013-0284-5
Hayashi, M., & Rosenberry, D. O. (2002). Effects of ground water exchange on the
hydrology and ecology of surface water. Ground water, 40(3), 309-316.
Herrera Environmental Consultants (2007). Untreated highway runoff in Western
Washington. Prepared for the Washington State Department of Transportation.
Hunsacker, C. T., & Levine, D. A. (1995). Hierarchical Approaches to the Study of
Water Quality in Rivers. BioScience, 45(3), 193–203.
Kelsey, H., Porter, D.E., Scott, G., Neet, M., & White, D. (2004) Using geographic
information systems and regression analysis to evaluate relationships between land use
and fecal coliform bacterial pollution. Journal of Experimental Marine Biology and
Ecology 298(2), 197-209.
Kitsap Public Health District (KCHD)(2014). Annual Water Quality Report. Water
Pollution Identification & Correction Program. Retrieved from
http://www.kitsapcountyhealth.com/environment/files/reports/2014/Intro_withCover.pdf
Lenat, D.R., & Crawford, J.K. (1994). Effects of land use on water quality and aquatic
biota of three North Carolina Peidmont streams. Hydrobiologica. 29, 185-199.
Luce, C., Staab, B., Kramer, M., Wenger, S., Isaak, D., & McConnell, C. (2014).
Sensitivity of summer stream temperature to climate variability in the Pacific Northwest.
Water Resources Research. DOI: 10.1002/2013WR014329
McCarthy, S.G. Incardona, J.P. & Scholz, N. L. (2008) Coastal storms, toxic runoff, and
the sustainable conservation of fish and fisheries. American Fisheries Society Symposium
64, 7–27.
44

Maloney, Kelly O., and Donald E. Weller. (2011) "Anthropogenic disturbance and
streams: land use and land‐use change affect stream ecosystems via multiple pathways."
Freshwater Biology 56(3), 611-626.
May, C. W. (1997). The cumulative effects of urbanization on Puget Sound lowland
ecoregion. Puget Sound Research, University of Washington.
May, C.W. (1999). Stormwater toxicity: Literature review report. Battel Pacific
Northwest Laboratories. Sequim, Washington.
Morley, S.A. & Karr, J.R. (2002). Assessing and Restoring the Health of Urban Streams
in the Puget Sound Basin. Conservation Biology, 16(6), 1498–1509.
Murdock, P.S., Baron, J.S., & Miller, T. (2000). Potential effects of climate change on
surface water quality in North America. American Water Resources Association, 36(2),
347-366.
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.
Naiman, R. J. Elliott, S.R. Hatfield, J.M. & O’Keefe, T.C. (1999) Biophysical
interactions and the structure and dynamics of riverine ecosystems: the importance of
biotic feedbacks. Hydrobiologia, 410, 79-86.
Nielsen, A., Trolle, D., Søndergaard, M., Lauridsen, T. L., Bjerring, R., Olesen, J. E., &
Jeppesen, E. (2012). Watershed land use effects on lake water quality in Denmark.
Ecological Applications, 22(4), 1187–1200. https://doi.org/10.1890/11-1831.1
Lenat, D. R., & Crawford, J. K. (1994). Effects of land use on water quality and aquatic
biota of three North Carolina Piedmont streams. Hydrobiologia, 294(3), 185–199.
https://doi.org/10.1007/BF00021291
Pachauri, R. K., Allen, M. R., Barros, V. R., Broome, J., Cramer, W., Christ, R., ... &
Dubash, N. K. (2014). Climate change 2014: synthesis report. Contribution of Working
Groups I, II and III to the fifth assessment report of the Intergovernmental Panel on
Climate Change (p. 151). IPCC.
Poff, L.N., Allan, D.J., Bain, M.B., Karr, J.R., Prestegard, K.L., Richter, B.D., Sparks,
R.E. & Stromberg, J.C. (1997). The Natural Flow Regime. BioScience, 47(11), 769-784.
Pratt, B., & Chang, H. (2012). Effects of land cover, topography, and built structure on
seasonal water quality at multiple spatial scales. Journal of Hazardous Materials, 209–
210, 48–58. https://doi.org/10.1016/j.jhazmat.2011.12.068
45

Shandas, V. & Alberti, M. (2009). Exploring the role of vegetation fragmentation on
aquatic conditions: Linking upland with riparian areas in Puget Sound lowland streams.
Landscape and Urban Planning, 90, 66-75.
Sheldon, D., Hurby, T., Johnson, P., Harper, K., McMillian, A., Granger, T., Stanley, S.,
& Stockdale, D. (2005) Wetlands in Washington State: Volume 1 A synthesis of science.
Washington State Department of Ecology. [PDF] Retrieved from
http://www.ecy.wa.gov/biblio/0506006.html
Sliva, L., & Dudley Williams, D. (2001). Buffer Zone versus Whole Catchment
Approaches to Studying Land Use Impact on River Water Quality. Water Research,
35(14), 3462–3472. https://doi.org/10.1016/S0043-1354(01)00062-8
Sun, G., & Caudwell, P. (2015). Impacts of urbanization on stream water quality and
quantity in the United States. Water Resources IMPACT, 17(1), 17-20.
Sun, G., & Lockaby, B.G. (2012). Water quality and quantity and Urban-Rural interface.
Urban-Rural Linking people and Nature, 29-48.
Sweeney, B. W., & Newbold, J. D. (2014). Streamside Forest Buffer Width Needed to
Protect Stream Water Quality, Habitat, and Organisms: A Literature Review. JAWRA
Journal of the American Water Resources Association, 50(3), 560–584.
https://doi.org/10.1111/jawr.12203
Tong, S. T. Y., & Chen, W. (2002). Modeling the relationship between land use and
surface water quality. Journal of Environmental Management, 66(4), 377–393.
https://doi.org/10.1006/jema.2002.0593
Tu, J. (2011). Spatially varying relationships between land use and water quality across
an urbanization gradient explored by geographically weighted regression. Applied
Geography, 31(1), 376–392. https://doi.org/10.1016/j.apgeog.2010.08.001
U.S. Fish & Wildlife Service (2015). Environmental Conservation Online System.
Retrieved from http://ecos.fws.gov/tess_public/reports/species-listed-by-statereport?state=WA&status=listed
Utz, R.M., Hopkins, K.G. Beesley, L. Booth, D.B., Hawley, R.J., Baker, M.E., Freeman,
M.C. & Jones, K.L. (n.d.) Ecological resistance in urban streams: the role of natural and
legacy attributes. Society for Freshwater Science., 35.
http://www.jstor.org/stable/10.1086/684839
Wang, X. (2001). Integrating water quality management and land use planning. Journal
of Environmental Management, 61, 25–36. DOI:10.1006/jema.2000.0395

46

Washington State Legislature. WAC 173-201A-600. Use Designations-Fresh Waters.
Filed on 11/20/06; Effective on 12/21/06. Retrieved from
http://apps.leg.wa.gov/WAC/default.aspx?cite=173-201A-600
Yu, D., Shi, P., Liu, Y., & Xun, B. (2013). Detecting land use-water quality relationships
from the viewpoint of ecological restoration in an urban area. Ecological Engineering,
53, 205–216. https://doi.org/10.1016/j.ecoleng.2012.12.045
Zhou, T., Wu, J., & Peng, S. (2012). Assessing the effects of landscape pattern on river
water quality at multiple scales: A case study of the Dongjiang River watershed, China.
Ecological Indicators, 23, 166–175. https://doi.org/10.1016/j.ecolind.2012.03.013

47

Appendices
Appendix 1. Pebble size frequency in a stream reach near the headwaters of Carpenter Creek
in Kingston, WA.

Appendix 2. Characterization of bankfull and wetted widths from headwaters to the mouth of
Carpenter Creek in Kingston, WA.

48

Appendix 3. Water chemistry data for Carpenter Creek (2001-2016). Data provided by
Stillwaters Environmental Center.

49

50