University of Southern California Dissertations and Theses

Using the Digital Shoreline Analysis System (DSAS) to analyze changes in shoreline position caused by seawalls along a section of Oregon's coast

(USC Thesis Other)

Using the Digital Shoreline Analysis System (DSAS) to analyze changes in shoreline position caused by seawalls along a section of Oregon's coast

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content

Using the Digital Shoreline Analysis System (DSAS) to Analyze Changes in Shoreline Position
Caused by Seawalls Along a Section of Oregon’s Coast

by

Melina Bennett

A Thesis Presented to the
FACULTY OF THE USC DORNSIFE COLLEGE OF LETTERS, ARTS AND SCIENCES
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirements for the Degree
MASTER OF SCIENCE
(GEOGRAPHIC INFORMATION SCIENCE AND TECHNOLOGY)

August 2021

Copyright © 2021 Melina Bennett
ii

Dedication

For
My Parents,
Everyone who believed in me,
And Mr. Bowie – just keep swimming

iii

Acknowledgements
I would like to thank my advisor, Dr. Fleming, for his help and guidance with this project
as well as his patience and understanding when life interfered with this project. I would also like
to thank my committee members, Dr. Wilson and Dr. Marx, as well as Dr. Bernstein for their
feedback and guidance.

vi
Table of Contents
Dedication ....................................................................................................................................... ii
Acknowledgements ........................................................................................................................ iii
List of Figures .............................................................................................................................. viii
List of Tables .................................................................................................................................. x
List of Abbreviations ..................................................................................................................... xi
Abstract ........................................................................................................................................ xiii
Chapter 1 Introduction .................................................................................................................... 1
1.1. Coastal Processes and Protection Structures.......................................................................1
1.1.1. Coastal Processes .......................................................................................................1
1.1.2. Erosion Mitigation Structures ....................................................................................6
1.2. Study Area ..........................................................................................................................8
1.3. Objectives .........................................................................................................................11
Chapter 2 Related Work................................................................................................................ 13
2.1. Shoreline Protection Structures.........................................................................................13
2.2. Shoreline Monitoring Within the Lincoln Littoral Cell ....................................................15
2.2.1. SPSs and Shoreline Change Analysis ......................................................................15
2.2.2. Extreme Storms and Wave Variability ....................................................................17
2.2.3. Coastal Policy ..........................................................................................................19
2.3. Shoreline Extraction and Analysis ....................................................................................19
2.3.1. Proxy-Based Shoreline Extraction ...........................................................................20
2.3.2. Datum-Based Shoreline Extraction..........................................................................21
2.3.3. Combining Proxy- and Datum-Based Shorelines ....................................................22
2.3.4. Shoreline Change Analysis ......................................................................................24
Chapter 3 Methodology ................................................................................................................ 27
vii
3.1. Data Acquisition and Shoreline Digitization ....................................................................27
3.1.1. LiDAR Derived DEMs ............................................................................................28
3.1.2. NAIP Imagery ..........................................................................................................29
3.2. Calculating Shoreline Bias and Uncertainties...................................................................30
3.2.1. Proxy-Datum Bias ....................................................................................................30
3.2.2. HWL Shoreline Uncertainty ....................................................................................31
3.2.3. MHW Shoreline Uncertainty ...................................................................................32
3.2.4. Proxy-Datum Bias Uncertainties .............................................................................32
3.3. Digital Shoreline Analysis System (DSAS) .....................................................................34
3.3.1. Baseline, Shorelines, and Transects .........................................................................35
3.3.2. Complex Shoreline Data ..........................................................................................36
3.3.3. Shoreline Change Statistics......................................................................................37
Chapter 4 Results .......................................................................................................................... 39
4.1. Shoreline Digitization .......................................................................................................39
4.2. Shoreline Change ..............................................................................................................42
4.2.1. Overall Shoreline Change ........................................................................................42
4.2.2. Shoreline Change with Respect to Seawalls ............................................................48
Chapter 5 Discussion and Conclusions ......................................................................................... 52
5.1. Limitations and Future Improvements ..............................................................................52
5.2. Analysis of Shoreline Change Rates .................................................................................53
5.3. Coastal Policy ...................................................................................................................55
5.4. Conclusions .......................................................................................................................57
References ..................................................................................................................................... 59
Appendix A DSAS Summary Report for 1997 - 2016 ................................................................. 64
Appendix B DSAS Summary Report for 1997 – 2014 ................................................................. 67
viii
List of Figures
Figure 1. Cross shore profile depicting the difference in tidal datums in relation to the shore ...... 3
Figure 2. Diagram showing the typical equilibrium in sediment transport between summer
and winter ................................................................................................................................ 5
Figure 3. Occurrence of El Niño and La Niña from 1990 to present, the clored peaks in the
graph show when moderate and extreme El Niño and La Niña events have occurred ........... 6
Figure 4. Installed seawall in Oregon ............................................................................................. 7
Figure 5. Newly installed riprap in Lincoln County, Oregon ......................................................... 8
Figure 6. Location of the LLC relative to the Oregon coast. Red lines represent the locations
of seawalls. ............................................................................................................................ 10
Figure 7. Sediment size distribution along the LLC ..................................................................... 11
Figure 8. Change in shoreline position near a channel mouth of the Nile River ......................... 14
Figure 9. Difference in HWL position interpreted by three different analysists .......................... 23
Figure 10. Comparison of rate change statistic methods using shoreline position data ............... 25
Figure 11. DSAS workflow from the DSAS User Guide ............................................................. 34
Figure 12. Lines generated from NAIP imagery with NDWI classification showing where
shadows and wave swash are mis-classified. ........................................................................ 40
Figure 13. 2016 NAIP imagery (a) and NAIP imagery with NDWI classification applied with
a 0.09 threshold (b) and 0.1 threshold (c). ............................................................................. 41
Figure 14. Change rates in meters per year for shorelines between 1997-2016 calculated
by WLR (left) and EPR (right). ............................................................................................. 45
Figure 15. The EPR for the 1997-2016 shorelines. ...................................................................... 46
Figure 16. The WLR for the 1997-2016 shorelines. ..................................................................... 47
ix
Figure 17. Comparison of NSM in meters. ................................................................................... 49
Figure 18. A closer look at the EPR for the 1997-2016 shorelines. ............................................. 50
Figure 19. A closer look at the EPR for the 1997-2014 shorelines. ............................................. 51
Figure 20. Graph showing the cool and warm phases of the PDO ............................................... 54
Figure 21. Seawall collapse at the southern end of the LLC leaving three houses dangerously
exposed to further cliff .......................................................................................................... 56
x
List of Tables
Table 1. The date, source, resolution, and shoreline proxy of the data used in this study............ 28
Table 2. DSAS baseline geodatabase field requirements ............................................................. 35
Table 3. DSAS shoreline geodatabase field requirements ............................................................ 36
Table 4. DSAS proxy-datum bias table field requirements .......................................................... 37
Table 5. The average shoreline changes rates and percent erosion and accretion for each
statistic from the DSAS summary report. ............................................................................. 43
Table 6. Count of the occurrence of erosion or accretion at the beginning and end of the
seawall groups. ...................................................................................................................... 48

xi
List of Abbreviations
DEM Digital Elevation Model

DSAS Digital Shoreline Analysis System

EPR End Point Rate

GIS Geographic Information System

HWL High Waterline

IPCC Intergovernmental Panel on Climate Change

LiDAR Light Detection and Ranging

LLC Lincoln Littoral Cell

LRR Linear Regression Rate

MHHW Mean Higher High Water

MHW Mean High Water

MIR Middle Infrared

MNDWI Modified Normalized Difference Water Index

MSL Mean Sea Level

NAIP National Agriculture Imagery Program

NAVD88 North American Vertical Datum of 1988

NDWI Normalized Difference Water Index

NIR Near-infrared

NOAA National Oceanic and Atmospheric Administration

NOS National Ocean Service

NSM Net Shoreline Movement

PDO Pacific Decadal Oscillation

xii
SCE Shoreline Change Envelope

SLR Sea Level Rise

SPS Shoreline Protection Structure

SWIR Short-wave Infrared

USGS United States Geological Survey

WLR Weighted Linear Regression Rate
xiii
Abstract
The Lincoln Littoral Cell (LLC) contains a 24 km stretch of coastline along Oregon’s
central coast. Nearly 50% of the LLC’s coastline has been armored with shoreline protection
structures (SPSs), mainly riprap and seawalls. SPSs are constructed to reduce damage to coastal
developments caused by breaking waves, flooding, and sediment erosion. Although the SPSs are
meant to protect the coast from erosion, they can ultimately cause erosion adjacent to the
structure or further down shore. Future projections and models show an increase in frequency of
large storm systems that generate larger than average waves and water levels, resulting in
increased erosion and flooding. This project utilizes the USGS’s ArcGIS add-on, Digital
Shoreline Analysis System (DSAS), to analyze digitized shoreline positions from 1997 to 2016.
Visual analysis shows that while on average the shoreline is accreting at a rate of 0.32 m/yr,
there is localized erosion adjacent to 53% of the SPSs. Future policies regarding the placement
and build of SPSs should take into consideration the long-term negative effects of these
structures.

1
Chapter 1 Introduction
Shoreline protection structures are generally constructed to protect buildings and other
structures from damage caused by breaking waves. But as sea levels rise and the frequency of
storm surges increase, it is becoming more apparent that these structures are interfering with the
natural coastal processes that govern beach morphology. This interference can lead to an increase
in erosion rates further along the shoreline. This study used shorelines extracted from aerial
imagery and Light Detection and Ranging (LiDAR) derived digital elevation models (DEMs) to
analyze changes in shoreline position along a heavily armored coastline in central Oregon.
Coastal Processes and Protection Structures
Coastal erosion due to sea level rise is a growing threat to coastal communities. The
Intergovernmental Panel on Climate Change (IPCC) estimates that sea levels could rise
anywhere from 0.28 to 0.98 m in the next 80 years depending on location (IPCC 2013). With
over half a million people living along Oregon’s coast, an increase in sea level will expose more
people and structures to coastal hazards including saltwater inundation, flooding, and damages to
buildings caused by sediment erosion (NOAA 2021c). Because of this, erosion mitigation efforts
in the form of shoreline protection structures (SPSs) have been put into place to protect coastal
developments from erosion. Man-made SPSs are a relatively short-term solution and can cause
erosion adjacent to the SPSs themselves.
1.1.1. Coastal Processes
The simplified definition of the shoreline is the boundary between water and land.
Although this seems simple enough, there are many factors that influence the position of this
boundary such as water level and sediment transport (Boak and Turner 2005). Because the
2
shoreline cannot be easily defined, shoreline proxies are often used in shoreline analysis. There
are two types of shoreline proxies, visible and datum based. Visible proxies include alongshore
features that are easily seen and include features such as a vegetation line, the top or base of a
cliff, erosion scarp and the wet/dry sediment boundary or high-water line (HWL) (Boak and
Turner 2005). Datum based shorelines use an averaged high or low water-level at a given
location. These datums are calculated by averaging the value of high- or low-water levels
recorded by a water gauge over a 19-year period known as the tidal datum epoch. NOAA
calculates tidal datums, which includes data such as mean high water (MHW), mean sea level
(MSL), and others, for each epoch. The current epoch uses data collected between 1983 and
2001. The next tidal datums for the epoch spanning from 2002 to 2020 are set to be released in
2025 (NOAA 2021f). Standard water level datums are calculated for MHW, mean higher high
water (MHHW), mean low water (MLW), and mean lower low water (MLLW). Figure 1 shows
the position of the tidal datums relative to the shore.
A littoral cell is defined by the natural cycle of fluctuating sediment loss and deposition.
Specifically, a littoral cell is a segment of coastline that begins at a point where sediment is
introduced and ends where sediment is deposited (Davidson-Arnott 2010; van Rijn 2011).
Sediment introduced by rivers and cliff erosion are common sources of input into the sediment
budget, while sediment that is transported and deposited offshore onto the continental shelf
removes sediment from the sediment budget (Davidson-Arnott 2010). Littoral cells are not a
closed system and sediment can be exchanged between adjacent cells (Anderson et al. 2018;
Davidson-Arnott 2010). Seasonal variabilities, influenced by tides and storm surges, redistribute
the sediment from the beach to nearshore sediment stores and vice versa (van Rijn 2011).
3

Coastal erosion affects the coastal zone by permanently removing sediment from the
sediment budget. There are many natural and anthropogenic factors that contribute to the onset
of coastal erosion including large storm systems, damming of rivers, and coastal development.
One of the greatest influences is sea level rise (SLR). As sea levels rise, areas at higher
elevations are increasingly exposed to powerful waves and flooding during large storm systems
(van Rajin 2011).
There are several processes that influence changes to regional sea level. Some of these
include an increase in the temperature of ocean waters causing thermal expansion, melting of
glaciers and ice sheets, and vertical movement of the Earth’s crust due to tectonic or volcanic
activity and isostatic adjustment caused by the melting of glaciers or sediment loading (Cazenave
Figure 1. Cross shore profile depicting the difference in tidal datums in relation to the shore
(Gill at al. 2001).
4
and Cozannet 2014). The Earth has a natural cycle of warming and cooling that regulates these
factors. Even though SLR due to warming trends are naturally part of this cycle, the rate at which
this is occurring is increasing (IPCC 2013). This increased rate is concerning because this will
expose more people living along to coast to hazards such as flooding and damage to
infrastructure and cultural assets (Reguero et al 2018).
Coastal erosion and SLR are just two factors that influence shoreline variability.
Shoreline variability refers to the cycles of progradation and retrogradation of the shoreline that
can occur temporally from tens of years to centuries (Stive et al. 2002). There are many other
natural and anthropogenic factors that contribute to the shoreline variability. Some of these
include wave and tide conditions, availability of sediment, geologic setting, shore nourishment,
coastal structures, and natural resource extraction (Stive et al. 2002).
One of the natural influences on shoreline variability along the Oregon Coast are the
occurrences of El Niño and La Niña events. The El Niño phenomenon occurs when the
temperature of equatorial surface water in the Pacific rise above average, or below average for
La Niña (NOAA 2021a). El Niño events culminate in larger than average storm surges, wave
energy, and an increase in flooding events, all of which cause higher than average erosion rates
(Barnard et al. 2015). Shorelines located between headlands generally have little to no erosion,
but during El Niño events there is a greater rate of erosion that occurs at the southernmost
sections and along the north side of inlets as shown in Figure 2. An increase in water level is
another characteristic of El Niño events. During major El Niño events, water levels can increase
by up to 0.4 m (Komar, Allan, and Ruggiero 2011). El Niño events occur every 3 - 5 years with
major events occurring every 20 years, shown in Figure 3 (NOAA 2021a, Cai et al. 2014). The
last notable El Niño event occurred over the winter between 2015 and 2016. Under current
5
anthropogenic warming trends, extreme El Niño events are projected to double from one in every
20 years to one in every 10 by 2090 (Cai et al. 2014).

At a larger time scale, the Pacific Decadal Oscillation (PDO) is characterized by monthly
averaged sea surface temperature anomalies in the North Pacific. The PDO is influenced by
multiple oceanic processes from the northern and tropic regions of the Pacific, such as El Niño
and others (Newman et al. 2016). Warm and cool phases of the PDO alternate every 20 years and
because the PDO reflects multiple processes, it does not have as apparent characteristics like
large storm systems with El Niño.
Figure 2. Diagram showing the typical equilibrium in sediment transport between summer and
winter (a) and where greater wave energies during El Niño years causes erosion in headland
bound shorelines (b) (Ruggiero et al. 2013).
6

1.1.2. Erosion Mitigation Structures
Man-made structures, such as seawalls, are commonly constructed to mitigate the effects
of coastal erosion. Although these structures are used to inhibit erosion in specific areas, research
shows that they interfere with the sediment transport within a littoral cell. This can prevent
sediment from entering and moving within the system as well as causing greater erosion adjacent
to the structure and/or further down the shoreline (Kraus and McDougal 1996; van Rijn 2011).
Seawalls are structures that run parallel to the shoreline and are generally constructed
using concrete. They are designed to prevent landward retreat of the shoreline by reflecting or
dissipating the energy of breaking waves (Kraus and McDougal 1996). Ripraps are a type of
Figure 3. Occurrence of El Niño and La Niña from 1990 to present, the clored peaks in the graph show
when moderate and extreme El Niño and La Niña events have occurred. (Trenberth 2020).
7
seawall consisting of large chunks of rock or concrete that are piled along the shoreline to
dissipate the energy of breaking waves (Hiller, Lia, and Aberle 2019). Examples of a seawall and
riprap installed along the Oregon Coast are given in Figures 4 and 5 respectively. The length and
location of where the seawall is constructed determines the effectiveness of the structure
(Weggel 1988). Seawalls constructed close to the active shoreline have a greater influence on
sediment transport by preventing the sediment behind the wall from entering into the system and
causing erosion at either the base of the wall and/or where the wall ends (Weggel 1988). Because
placement of these hard structures is critical to their effectiveness and varies depending on the
beach type, any negative effects might not be noticeable in the short term (Pilkey and Wright
1988). The dynamic nature of the sea level and shoreline position prevent these structures from
permanently occupying an ideal location to inhibit coastal erosion. This fluctuation can
ultimately lead to unintended alterations in shoreline morphology.
Figure 4. Installed seawall in Oregon (DLCD 2021).
8

Study Area
The Oregon coast lies on top of the Cascadia Subduction Zone where the Juan de Fuca
oceanic plate is subducting under the North American continental plate. The Cascadia
Subduction Zone was thought to be inactive due to the lack of seismic activity, but it is now
believed that the two plates are locked together (Komar and Shih 1993). As a result of the strain
due to the lock, the southern Oregon coast is being uplifted at a rate faster than eustatic SLR
(Komar and Shih 1993). The northern and central sections of the Oregon coast are being uplifted
at a slower rate than eustatic SLR resulting in a regional SLR of 1 to 2 mm/yr (Ruggiero et al
2013).
Figure 5. Newly installed riprap in Lincoln County, Oregon (Kauffman 2018).
9
Located along Oregon’s central coast, the Lincoln Littoral Cell (LLC) stretches 24 km
from Cape Foulweather north to Cascade Head. The LLC contains the most developed section of
Oregon’s coast. Roughly 50% of the shoreline is armored with 244 SPSs built to protect these
developments (Gardner, 2015; Good 1992). Riprap and seawalls are the two types of SPSs
constructed along the shoreline (Good 1992). The extent of the constructed SPSs is shown in
Figure 6.
The majority of the beaches within the LLC are backed by cliffs comprised of Pleistocene
sands and gravel (Shih and Komar 1994). The sediments on the southern beaches are fine-
grained and increase in size north towards Siletz Spit and then gradually decrease in size further
north to Cascade Head, shown in Figure 7 (Komar and Shih 1993). This is of importance because
sediment size influences erosion rates. Dissipative beaches that consist of finer grained sand, like
that along Lincoln City, have a smaller slope resulting in lower energy wave swash (Komar and
Shih 1993). Reflective beaches consist of gravel and have a steeper profile resulting in higher
energy wave swash (Komar and Shih 1993). There are two small rivers that drain into the ocean
within the LLC, the Salmon and Siletz Rivers. The Salmon River contributes a minor amount of
sediment because of its small size whereas the Siletz River is likely a sediment store with marine
sediment being deposited within the Siletz estuary. The main source of sediment within the LLC
is a result of cliff erosion (Shih and Komar 1994). With the majority of the beaches in the LLC
backed by cliffs, the 244 SPSs are preventing an estimated 39% of the sediment supply from
entering the sediment budget (Good 1992).
10

Figure 6. Location of the LLC relative to the Oregon coast. Red lines
represent the locations of seawalls.
11

Objectives
The LLC, located along Oregon’s Central Coast, is heavily armored with seawalls and
riprap. These structures were constructed to protect the ground under and around infrastructure
from eroding by dissipating and reflecting wave energy. But when water levels and wave energy
are greater than average, like during El Niño events, erosion can occur at the base, either end of
the structure, and further down the shore. The rate of major El Niño events are increasing and in
conjunction with SLR, the structures will be exposed to larger waves and flooding more
frequently. This will eventually lead to damaging the structure and infrastructure it was meant to
Figure 7. Sediment size distribution along the LLC (Shih and Komar
1994).
12
protect. Understanding exactly how the shore morphology responds to the installed SPSs can
help coastal planners and engineers with future planning and assessments on previously installed
structures. The goal of this study is to determine if shoreline change analysis can be used to
detect erosion caused by seawalls along the LLC.
13
Chapter 2 Related Work
This chapter reviews literature related to the effectiveness of SPSs, discusses previous shoreline
monitoring studies within the LLC, and compares shoreline change analysis methods.
Shoreline Protection Structures
The various methods for coastal erosion mitigation have been reviewed by van Rijn 2011
and Pranzini, Wetzel, and Williams 2015. They found that groins and seawalls are the most
common shoreline protection structures and both structures cause greater erosion elsewhere in
the littoral cell. Because of this, most countries are moving towards erosion mitigation methods
that do a better job of conserving the surrounding environments. This includes taking into
consideration the sediment transport within the littoral cell because different shoreline protection
structures work differently depending on the shoreline type (Pranzini, Wetzel, and Williams
2015).
Darwish et al. (2017) compare shorelines along the Nile Delta from 1945 to 2015. In the
late 1960s a dam was built upstream, significantly reducing the sediment output of the Nile
River. This drop in the sediment supply resulted in erosion focused immediately down-drift of
channel mouths. To remedy this, seawalls were constructed in the early 2000s (Darwish et al.
2017). The construction of the dam and seawalls has had a large impact on the shorelines; Figure
8 shows how these SPSs altered the shoreline position, and that erosion is focused adjacent to the
structures (Darwish etl al. 2017).
14

Along a section of the Northern Tuscany Coast in Italy, shoreline protection structures
have been used to mitigate erosion caused by rising sea levels, reduced sediment output from
three nearby rivers, and the construction of harbors (Pranzini et al. 2018). In this study, Pranzini
et al. (2018) follow the construction of groins, detached breakwaters, and seawalls along the
coast. They map the changes to the shoreline position from 1878 to 2017 using historical maps,
ortho-rectified aerial photographs, and satellite imagery. The majority of these structures were
constructed in response to increased erosion down-drift of an older structure. The use of these
hard structures is popular because they provide almost instantaneous results compared to long-
term methods like sediment bypassing and beach nourishment (Pranzini et al. 2018). This is one
Figure 8. Change in shoreline position near a channel mouth of the Nile River showing where
erosion is focus after the construction of two seawalls on either side of the channel mouth
(Darwish et al. 2017).
15
of the only studies that specifically looks at the effectiveness of SPSs for the lifespan of the
structure and discusses the benefits of switching to a more sustainable erosion mitigation
method.
Although seawalls are known to have adverse effects on shoreline morphology, not much
is known about how shorelines with seawalls will react to SLR. Beuzen et al. (2018) created a
model to test how shorelines with seawalls will change due to SLR. The model projected the
shoreline profiles with dissipative and reflective seawall structures to profiles without seawalls to
compare the profile response to SLR. They found that with an increase in water level, shorelines
with seawalls (dissipative and reflective) eroded similar volumes of sediment as shorelines with
no protection but that the erosion was focused to areas adjacent to the seawall.
Shoreline Monitoring Within the Lincoln Littoral Cell
The majority of the shorelines within the LLC are heavily armored with riprap and
seawalls. Because of the high wave energy and shoreline variability along Oregon’s coast,
previous studies have assessed the LLC’s shoreline position, erosion hazards, wave climate, and
current coastal policies.
2.2.1. SPSs and Shoreline Change Analysis
A thorough account of the LLC’s SPSs and the regulations governing the alterations and
construction of new structures is provided in Good’s doctoral thesis (Good 1992). Good created
a spatial database that included information on when the structures were built, sediment supply,
erosion rates, and Oregon’s policies on SPSs. The main goal of the thesis was to assess the
implementation of shoreline protection policies. To quantify this, he looked at the effectiveness
of the SPSs. Before the implementation of Goal 18, new development did not take future
16
geologic and oceanographic hazards into consideration and built structures in less-than-ideal
locations that eventually required SPSs. Assuming that most of the sediment supplied to the
beach is from cliff erosion and using erosion rates, Good calculated that the existing SPSs
prevent about 39% of the annual sediment supply from entering the sediment budget. This is a
good argument on the probability that SPSs will have a negative effect on the surrounding
shoreline within the LLC but there is no direct spatial analysis testing this claim.
In a similar concept to Good’s analysis, Priest (1999) used erosion rates to predict areas
that would be affected by coastal erosion. The shorelines of the LLC were assessed for erosion
hazards to project property damage that could occur within a 60-year time frame. Shorelines with
SPSs were assumed to have the same erosion rate as adjacent unprotected shorelines unless the
SPSs met a set of criteria. The criteria required the structures to be at least 150 m in bluff-backed
shorelines or 300 m for dune-backed shorelines, have been standing undamaged for at least 10
years, and could not have an active landslide located behind it (Priest 1999). There is not much
discussion on the SPSs that met the above criteria or if there were any erosion hazards that were
a direct result of existing SPSs.
Allan, Komar, and Priest (2003) used shoreline position data from littoral cells north and
south of the LLC to determine any long-term trends in shoreline position. Although this study
did not include the LLC specifically, the authors were able to conclude that long-term shoreline
change for the entire coast of Oregon was negligible. They determined that the short-term
shoreline variability caused by El Niño and La Niña events should be the main consideration
when assessing erosion hazards and setback distances for constructing new developments.
More recently, the United States Geologic Survey (USGS) used historical aerial
photographs, National Ocean Service (NOS) topographic sheets (T-sheets), and LiDAR to
17
calculate the short- and long-term shoreline change rates along the coastlines of Oregon and
Washington (Ruggiero et al. 2013). Shorelines were digitized from the source material and then
analyzed in USGS’s ArcGIS add-on Digital Shoreline Analysis System (DSAS). Short-term
change rates are determined by calculating the endpoint, which is the change in shoreline
position divided by the time between the data sets. Long-term change rates required at least four
years of data and is determined by the slope of the linear regression. Both short- and long-term
change rates are calculated because some shorelines do not change in a linear trend and linear
regression cannot account for these non-linear trends. Short- and long-term change rates for the
LLC were determined to be -0.3 ± 0.1 m and 0.1 ± 0.5 m respectively. The short-term change
rate was attributed to the possibility that sediment supply has been interrupted due to the
presence of SPSs (Ruggiero et al. 2013). The goal of this study was to determine shoreline
change rates of large sections of coastline along the pacific northwest, making the scope of this
study too large to be able to examine how SPSs specifically affect the shoreline within the LLC.
This project uses a similar methodology as this study to calculate shoreline change rates.
2.2.2. Extreme Storms and Wave Variability
Over the winters of 1997-98 and 1998-99 four extreme storms, classified by generating
deep water significant wave heights greater than 10 m, occurred in the Eastern North Pacific
(Allan and Komar 2002). The 1997-98 winter was classified as an El Niño with one extreme
storm in November that had a storm surge of 0.41 m bringing the total water level to 0.81 m.
There were three extreme storms the following La Niña winter. The largest storm surge recorded
from the three storms was 0.61 m, because water levels are lower during La Niña events
compared to El Niño, the total water level reached 0.82 m (Allan and Komar 2002). Alan and
Komar (2002) looked into the deep-water wave heights for the previous 25 years and concluded
18
that large storms like this are occurring more frequently with increasing strength. Allan and
Komar (2006) revisited their 2002 study to add to and update their predictions. They found that
the general trend in increased wave height continues. This time they noted the connection to the
PDO and that it was shifting to a predominantly La Niña phase suggesting a decrease in overall
erosion.
More recent studies that created models on deep-water wave height disagree with Allan
and Komar’s (2006) prediction of increasing wave heights. Erikson et al. (2015) used two
different climate models to simulate surface winds and create wave height predictions. They
found that predicted wave heights should decrease along the Oregon coast but note that among
other similar studies there are some contradictions to these findings. They recommend a future
study that takes into account different climate projections with more iterations to capture a wider
range of conditions that influence wave height.
Trends in shoreline variability have been identified in conjunction with events such as El
Niño, but such events do not account for trends that occur at larger time scales. Anderson et al.
(2018) created a model of wave energy off of the Oregon Coast to try to identify any
relationships between shoreline variability and multidecadal oceanic phenomena. They were able
to identify a direct correlation between the modeled wave energy and the PDO index. This was
observed in Pacific City, Oregon (located in the littoral cell just north of the LLC) between 1970
where riprap were installed due to high erosion rates and 1984 when the riprap were completely
buried by sand (Allan, Komar, and Priest 2003; Anderson et al. 2018). Anderson et al.’s (2018)
model showed a decrease in wave energy coinciding with a cool phase of the PDO in the 1970s
and an increase in wave energy in the 1980s during a warm phase of the PDO. This correlation
19
supports Allan and Komar’s (2006) observations that shifts in the PDO result in either increased
El Niño or La Niña events which directly correspond to erosion rates.
2.2.3. Coastal Policy
In 1967 the Beach Bill was passed requiring that the beaches along Oregon’s entire coast
are accessible to the public for free. The Oregon Statewide Planning Goal 18, Beaches and
Dunes (OAR 660-015-0010) was created to help ensure that the Beach Bill is maintained. Goal
18 provides details on where developments can be constructed along the coast. Additionally,
Goal 18 requires SPSs to have a permit and outlines the requirements for a tax lot to be eligible
for a SPSs. One of the requirements is that the development on the lot has to have been built
before January 1, 1977. Gardner (2015) assessed these policies and all oceanfront tax lots along
the Oregon Coast to provide ideas on how to improve the policies and permitting regulations. At
the time of the report, tax lots were only assessed based on adjacent structures, SPSs under 50 ft
did not require a geologic report, and little to no plan was in place to address situations where
development might become compromised due to erosion. To address this, Gardner (2015)
recommends that all permits require a geologic and hazard assessment, impacts of the SPSs be
assessed for the entire littoral cell, and that there should be set design standards for constructing
new SPSs and repairing old ones. This report highlights the fact that the current coastal policies
do not take into account the future impacts of SLR, climate change, and the installed SPSs.
Shoreline Extraction and Analysis
There are several different methods when it comes to digitizing shoreline position
depending on the source data. These sources can include historical shorelines from NOS T-
20
Sheets, georeferenced aerial and satellite imagery, LiDAR point clouds and DEMs. Once the
shorelines are extracted, positional rate changes are able to be calculated and analyzed.
2.3.1. Proxy-Based Shoreline Extraction
To determine shoreline position, the boundary between the ocean and land needs to be
identified and this can be accomplished by classifying each pixel as either land or water. This
process can be automated by using different spectral bands to calculate an index and specifying a
threshold that determines the values that represent land or water (Fisher, Flood, and Danaher
2016). There are many indices that have been created to distinguish between land and water with
two the most commonly used ones being Normalized Difference Water Index (NDWI) and
Modified Normalized Difference Water Index (MNDWI).
NDWI was introduced in 1996 by S. K. McFeeters and uses the green and near infrared
(NIR) spectral bands to identify pixels representing water. This index is based off of the
Normalized Difference Vegetation Index which uses the red and NIR bands to enhance
vegetation features. NDWI enhances water features due to the fact that water has a high green
and low NIR reflectance compared to vegetation and soil features (McFeeters 1996). The
equation returns either a positive or negative value, with positive values indicating a greater
green reflectance and therefor representing water (McFeeters 1996).
One of the issues with NDWI is that urban areas can have a NIR reflectance value that is
close to its green reflectance, and this can result in urban areas having a NDWI value that is
positive, falsely classifying it as water (Xu 2006). To address this Xu (2006) created the
MNDWI that uses middle infrared (MIR) instead of NIR. Urban areas reflect more MIR than
NIR resulting in values that will always be negative (Xu 2006).
21
Both NDWI and MNDWI only use two spectral bands to calculate the difference in
reflectance values and depending on what other features, such as shadows, are in the image can
result in misclassification (Feyisa et al. 2014). In a study using NDWI to identify pixels
representing swimming pools in a residential neighborhood, McFeeters (2013) addressed the
misclassification due to shadows by changing the threshold from 0 to 0.3. Multiple studies have
compared the different water indices and concluded that the varying accuracies of each index is
influenced by cloud cover, beach type, and landcover features (Fisher, Flood, and Danaher 2016;
Kelly and Gontz 2018).
Once the water index has been calculated, the resulting raster images have two values
that ideally denote either water or land. The shoreline position can then be extracted by
converting the rasters to vectors. The resulting vectors will most likely need to be cleaned up to
remove any lines that do not represent the shoreline (Sunder, Ramsankaran, and Ramakrishnan
2017).
2.3.2. Datum-Based Shoreline Extraction
Two methods for extracting shorelines from LiDAR data are the Profile and Contour
methods. The Profile method utilizes LiDAR point cloud data and was first described by
Stockdon et al. (2002). This method looks at a cross-shore profile and points that fall within one
meter of either side of the profile. A datum-based shoreline position, such as MHW, is selected
and the points ± 0.5 m are plotted with a regression line. The horizontal shoreline position and
foreshore beach slope are determined by the slope of the regression line. This is repeated every
20 m along the shoreline.
The Contour method uses a DEM and adds a contour line along a specified elevation
representing the shoreline position and is described in detail by Harris et al. (2006). The contour
22
spatial analysist tool in either ArcGIS Pro or ArcMap can be used to add a contour line along a
datum-based shoreline elevation, generally MHW. This elevation is used as the base contour
value and the contour interval is set to a large arbitrary number. The contour feature class is
edited to only include lines that represent the shoreline and remove any loops or breaks.
Although both of these methods use LiDAR data to extract datum-based shorelines, they
do not necessarily place the shoreline in the same position. Ferris et al. (2018) compared these
two methods to determine which one was more accurate. Shorelines were generated from the
same LiDAR data using both methods. They concluded that there was an insignificant difference
between the shoreline positions generated by the two methods. Comparing the methods, the
Profile method requires programming knowledge, and can be time consuming but uses linear
regression to calculate the shoreline position and slope. The Contour method is relatively quicker
and easier to create but cannot extract shorelines where water levels are above MHW.
2.3.3. Combining Proxy- and Datum-Based Shorelines
NOS T-Sheets are historical records of shorelines that were mapped through surveys and
aerial photographs using the HWL. Because these records date back to the 1800’s, the HWL is
still one of the most commonly used shoreline proxies (Moore, Ruggiero, and List 2006,
Ruggiero et al 2013). Although the HWL is frequently used, there are uncertainties associated
with its position relative to the recorded high-water level such as short-term fluctuations in wave
energy (Ruggiero et al. 2013). This is reflected as an over-estimation of the shoreline. Another
uncertainty is how the HWL is interpreted and digitized for shoreline analysis. Both Moore,
Ruggiero, and List (2006) and Ruggiero et al. (2006) had multiple analysists draw the HWL
from an aerial photograph to demonstrate the variability in HWL interpretation. The resulting
offset between these shorelines from Moore, Ruggiero, and List (2006) are shown in Figure 9.
23
Just as the HWL overestimates the recorded high water-level, MHW datum-based shorelines
underestimate the high water-level. This is because tidal datums are an average of high tide
measurements at one location and fail to take into consideration alongshore variations in wave
conditions and shore morphology (Ruggiero et al. 2013).

Ruggiero and List (2009) recognized the offset of the HWL and MHW shorelines as a
bias. They determined that the bias was a function of the water-level, foreshore beach slope, and
the significant offshore wave heigh and period. Because all of these factors are easily measured
Figure 9. Difference in HWL position interpreted by three different analysists
(Moore, Ruggiero, and List 2006).
24
or estimated they were able to create a proxy datum bias equation. Although these measurements
are easily obtainable for current data, most historic data is not recorded at the same spatial and
temporal resolution and therefore have uncertainties associated with them. Ruggiero and List
(2009) used the proxy datum bias equation to derive equations to determine the long-term
estimates for these factors as well as the uncertainty in the proxy datum bias equation. These
equations allow for HWL and MHW shorelines to be used in the same analysis and are defined
and discussed further in Section 3.2.
2.3.4. Shoreline Change Analysis
Shoreline change analysis calculates the rate at which the shoreline position moves over a
period of time. This rate change can be calculated using various statistical methods including end
point rate (EPR), linear regression (LLR), and weighted linear regression (WLR). Genz et al.
(2007) compare nine statistical methods that have been used in shoreline change analysis to
determine which method best represents the change rates for shorelines in Maui, Hawaii. Figure
10 shows the methods compared and the difference in how the regression line is fitted for each
method using the same shoreline data (Genz et al. 2007). They concluded that all methods tested,
except EPR, average of rates (AOR), and middle description length (MDL), are acceptable for
shoreline change analysis. Specifically, if uncertainties are known, weighted least squares (WLS)
and reweighted weighted least squares (RWLS) provide the best results. If uncertainties are
unknown, the best method would be least absolute deviation (LAD), but ordinary least squares
(OLR), reweighted least squares (RLS), and jackknifing (JK) could be used (Genz et al 2007).
Each statistical method is explained in detail in Genz et al. (2007). This study uses the EPR,
OLR (LRR), and WLS (WLR) as described in Section 3.3.3.
25
Figure 10. Comparison of rate change statistic methods using shoreline position data (Genz et al. 2007). The EPR, OLS, and WLS
are calculated by DSAS with the latter two referred to as LRR and WLR respectively.
26
Shoreline positions can be calculated from either a transect from baseline method or
change polygon method. The change polygon method overlays two shoreline vectors from
different years and creates polygons from where the shorelines intersect. These polygons
represent erosion or accretion and are added together to get the total area of change. When
multiple shorelines are analyzed, change rates for each shoreline are calculated in relation to a
baseline. These rates can then be plotted with a regression line to determine the overall rate
change (Smith and Cromley 2012).
Transect from baseline methods involve creating transects perpendicular to a baseline and
calculating change rates from where the shorelines intersect the transects. Different change rate
statistics can then be calculated using two or more of the shorelines. DSAS uses the transect
from baseline method and is used in this study. DSAS calculates multiple change rate statistics
including EPR, LRR, WLR for each transect (Hemmelstoss, 2018).
Albuquerque et al. (2013) compared the end point rate generated by DSAS and the
change polygon method. They used satellite imagery and aerial photographs from 2007 to 2011
to extract shoreline vectors. The resulting statistics from DSAS and the change polygon method
linear regression rates were 12.31 m/yr with an R
2
of 13% and 1.76 m/yr with an R
2
of 94%
respectively. They concluded that the linear regression rate DSAS reports reflects seasonal
variability instead of permanent erosion and accretion. Additionally, Smith and Cromley (2012)
found that different rates would be calculated depending on the shoreline morphology and where
the baseline is placed. The advantage to DSAS compared to the change polygon method is that
proxy-based and datum-based shorelines can be used in the same analysis and rates are
calculated for every transect. This means that a greater selection of data over a larger time period
can be used.
27
Chapter 3 Methodology
This chapter discusses the data and methodology for shoreline change analysis used in this study.
The DSAS ArcMap add-on is used to calculate change rates from shoreline positions that were
extracted from LiDAR derived DEMs and NAIP imagery.
Data Acquisition and Shoreline Digitization
Shoreline positions are digitized from LiDAR derived DEMs and NAIP imagery, and are
the main inputs used in DSAS to calculate shoreline change rates. This study uses both MHW
and HWL proxies for shoreline position. The source, date, resolution, and shoreline proxy used
for each data set is provided in Table 1. The MHW level calculated for South Beach, Oregon was
used for the LLC’s MHW value as this is the closest water gauge to the study area, roughly 15
miles south of the study area, and was downloaded from NOAA’s Tides & Currents website
(NOAA 2021f). For the current epoch, the MHW at South Beach, Oregon is 2.097 m relative to
the North American Vertical Datum of 1988 (NAVD88).
A shapefile containing the location, permit, and number of repairs of SPSs along
Oregon’s coast was retrieved through Oregon’s ArcGIS data server (Oregon Parks and
Recreation Department 2021). The shapefile was originally created by the Oregon Coastal
Management Program to assess the policies regarding SPSs. The database is now managed and
updated by Oregon Parks and Recreation Department.

28
Table 1. The date, source, resolution, and shoreline proxy of the data used in this study.
Dates Sources Resolution (m) Shoreline Proxy
October 17, 1997 NASA/NOAA/USGS 1 MHW
April 27, 1998 NASA/NOAA/USGS 1 MHW
September 9, 2002 NASA/USGS 1 MHW
June 23, 2009 NAIP 1 HWL
2010
U.S. Army Corps of Engineers
(USACE)
1 -
June 15, 2012 NAIP 1 HWL
July 1, 2014 NAIP 1 HWL
August 20, 2014
U.S. Army Corps of Engineers
(USACE)
1 MHW
April 28, 2016 USGS 0.5 MHW
June 25, 2016 NAIP 1 -

3.1.1. LiDAR Derived DEMs
LiDAR datasets for the Oregon coast are available for download from NOAA’s Data
Access Viewer (NOAA 2021b). Each dataset has been turned into a digital elevation model
(DEM) by the source and is available as a TIFF file. All DEMs were downloaded with the NAD
1983 StatePlane Oregon North horizontal and NAVD88 vertical coordinate systems with both
vertical and linear units in feet. The various sources for each dataset are outlined in Table 1. The
TIFF files were downloaded as compressed files and were extracted using 7-ZIP file manager (7-
zip.org). Shorelines were extracted from the DEMs following similar methodologies to Harris et
al. 2006 and Farris et al. 2018. The 2010 DEM data was not utilized in this study because it did
not have enough spatial coverage to accurately represent the shoreline throughout the study area.
29
A polyline representing the shoreline position for each dataset was created by using the
Create Contour tool in ArcGIS Pro. The MHW level was used for the baseline with a contour
interval of 100 m and a z factor of 1. These polylines were edited to remove line segments that
did not represent the shoreline. The shorelines were then smoothed using the Smooth Line tool
with the PEAK algorithm and a 10 m tolerance in ArcGIS Pro (Farris et al. 2018).
3.1.2. NAIP Imagery
NAIP imagery was downloaded through USGS’s Earth Explorer as compressed TIFF
files (USGS 2021). For every year available four images were downloaded to cover the study
area. NAIP image specifications are outlined in Table 1. The 7-Zip file manager was again used
to extract the compressed files. The TIFFs were imported into TerrSet and converted to IDRISI
raster files (.rst) which separated the files into the four spectral bands (red, green, blue, and NIR).
The Mosaic tool with average overlap was used to combine the different images into one. A
mosaic was created of the green and NIR bands for each year. The normalized difference water
index (NDWI) was calculated using Equation (1) and the Overlay tool. NDWI was applied to
increase the values of reflected green wavelengths from the surface of the water while decreasing
the reflected values of NIR wavelengths (McFeeters 1996). The NDWI images were then
exported to TIFF files.

𝑁𝑁𝑁𝑁𝑁𝑁𝑁𝑁 =
𝐺𝐺𝐺𝐺𝐺𝐺𝐺𝐺𝐺𝐺 − 𝑁𝑁𝑁𝑁𝑁𝑁 𝐺𝐺𝐺𝐺𝐺𝐺𝐺𝐺𝐺𝐺 + 𝑁𝑁 𝑁𝑁𝑁𝑁

(1)

In ArcGIS Pro, the NDWI NAIP images were added to the working file geodatabase. The
raster images were then reclassified into two classes to identify each pixel as representing either
water or land. A threshold value was applied to better represent the HWL, the boundary between
visibly wet and dry sand. The threshold value was determined by visually comparing the
30
reclassed image to the color composite image. Due to differences in lighting at the time the
image was captured, each year required a different threshold value. The 2016 NAIP imagery was
ultimately not used in this study, this is discussed in further detail in Chapter 4. Reclassifying the
raster image also converts the raster data to an integer dataset which is required to convert it
from a raster to a vector dataset. The vector dataset is composed of polylines that depict the
boundaries between the two classes. The polylines are then edited to only include lines
representing the HWL.
Calculating Shoreline Bias and Uncertainties
There are uncertainties associated with the accuracy of how data are collected and
processed. These uncertainties vary with every step from the collection of LiDAR and aerial
imagery to georeferencing and processing, as well as natural variations caused by tide level,
waves, and beach slope due to seasonal sediment transport. The uncertainties for each
measurement are calculated and are used by the DSAS program to calculate the errors and
uncertainties for shoreline change rate statistics. The uncertainties are also used by DSAS when
both MHW and HWL shorelines are used in the same shoreline position change calculations.
3.2.1. Proxy-Datum Bias
The offset between MHW and HWL shorelines can vary greatly depending on the slope
of the beach and wave runup. Ruggiero and List (2009) determined that this difference between
the HWL and MHW positions can be calculated using the tide level (ZT), MHW level (ZMHW),
beach slope (tan β), offshore significant wave height (Ho), and offshore wavelength (Lo) in
Equation (2). The offshore wavelength can be calculated using linear theory (Equation 3), where
g is the acceleration due to gravity and T is the offshore dominant wave period. Ruggiero and
31
List (2009) determined that if we assume that the HWLs are formed from MHW the offset
between the two essentially cancel each other out and give us Equation (4). Average offshore
wavelength and wave heights were calculated by NOAA’s National Data Buoy Center (NDBC)
using data collected by buoy 46050, which is 20 nautical miles offshore of Newport, Oregon
(NOAA 2021e).

𝐵𝐵 𝐵𝐵𝐵𝐵𝐵𝐵 = ( 𝑋𝑋 𝐻𝐻𝐻𝐻 𝐻𝐻 − 𝑋𝑋 𝑀𝑀 𝐻𝐻 𝐻𝐻 )
=
� 𝑍𝑍 𝑇𝑇 + 1.1(0.35tan β( 𝐻𝐻 𝑜𝑜 𝐿𝐿 𝑜𝑜 )
�
1
2
� �
+
[𝐻𝐻 𝑜𝑜 𝐿𝐿 𝑜𝑜 � 0.563tan β
2
+ 0.004 � ]
1
2
�
2
� − 𝑍𝑍 𝑀𝑀 𝐻𝐻 𝐻𝐻 tan β

(2)
𝐿𝐿 𝑜𝑜 = �
𝑔𝑔 2 𝜋𝜋 � 𝑇𝑇 2

(3)
𝐵𝐵 𝐵𝐵𝐵𝐵𝐵𝐵 = ( 𝑋𝑋 𝐻𝐻𝐻𝐻 𝐻𝐻 − 𝑋𝑋 𝑀𝑀 𝐻𝐻 𝐻𝐻 )
=
1.1(0.35tan β( 𝐻𝐻 𝑜𝑜 𝐿𝐿 𝑜𝑜 )
�
1
2
� �
+
� 𝐻𝐻 𝑜𝑜 𝐿𝐿 𝑜𝑜 � 0.563tan β
2
+ 0.004 � �
1
2
�
2
tan β

(4)

3.2.2. HWL Shoreline Uncertainty
Uncertainties associated with the digitization of HWL shorelines from NAIP images are
determined by two components. The first is the uncertainty of the accuracy of georeferencing the
images (Ug). Starting in 2009, NAIP imagery had to adhere to an “absolute accuracy
specification” which requires all georeferenced points to be within 6 m of true ground with a
95% confidence level (U.S. Department of Agriculture 2017). The second uncertainty is the
uncertainty of the HWL position when the images were taken (Upd). This is the same calculation
as the proxy-datum bias in Equation (4) (Ruggiero et al. 2012). The total uncertainty in HWL
shoreline positions (UHWL) are then calculated using:
32

𝑈𝑈 𝐻𝐻𝐻𝐻 𝐻𝐻 = � 𝑈𝑈 𝑔𝑔 2
+ 𝑈𝑈 𝑝𝑝 𝑝𝑝 2
(5)

3.2.3. MHW Shoreline Uncertainty
As with the HWL shoreline uncertainty, the MHW shoreline uncertainty is calculated
using two sources associated with data collection. The first uncertainty is the reported horizontal
accuracies for each LiDAR derived DEM (Ul). The second uncertainty is the variance of
foreshore slope values (Us). The MHW shoreline position uncertainty (UMWL) is calculated
using:

3.2.4. Proxy-Datum Bias Uncertainties
The terms used to calculate the proxy-datum bias, significant wave height, dominant
wave period, foreshore beach slope, and water level, have corresponding uncertainties. Ruggiero
and List (2009) have derived four equations to compute the proxy-datum bias uncertainty using
the above mentioned terms, as follows:
𝜕𝜕 𝐵𝐵𝐵𝐵 𝐵𝐵 𝐵𝐵 𝜕𝜕 𝑍𝑍 𝑇𝑇 =
1
tan 𝛽𝛽
(7)
𝑈𝑈 𝑀𝑀 𝐻𝐻 𝐻𝐻 =
�
𝑈𝑈 𝑙𝑙 2
+ 𝑈𝑈 𝑠𝑠 2

(6)
33
𝜕𝜕 𝐵𝐵 𝐵𝐵𝐵𝐵𝐵𝐵
𝜕𝜕 tan 𝛽𝛽 =
0.31
[ 𝐻𝐻 0
𝐿𝐿 0
(0.56 tan
2
𝛽𝛽 + 0.004)]
1/ 2
𝐻𝐻 0
𝐿𝐿 0
+
− 𝑍𝑍 𝑇𝑇 − 0.55[ 𝐻𝐻 0
𝐿𝐿 0
(0.56 tan
2
𝛽𝛽 + 0.004)]
1/ 2
− 𝑍𝑍 𝑀𝑀 𝐻𝐻 𝐻𝐻 tan
2
𝛽𝛽
(8)
𝜕𝜕 𝐵𝐵 𝐵𝐵𝐵𝐵𝐵𝐵
𝜕𝜕 𝐻𝐻 0
= �
0.19 tan 𝛽𝛽 ( 𝐻𝐻 0
𝐿𝐿 0
)
1/ 2
𝐿𝐿 0
+
0.28
𝐻𝐻 0
𝐿𝐿 0
(0.56 tan
2
𝛽𝛽 + 0.004)
�
1/ 2
×
𝐿𝐿 0
(56 tan
2
𝛽𝛽 + 0.004)
tan 𝛽𝛽
(9)
𝜕𝜕 𝐵𝐵 𝐵𝐵𝐵𝐵𝐵𝐵
𝜕𝜕 𝐿𝐿 0
= �
0.19 tan 𝛽𝛽 ( 𝐻𝐻 0
𝐿𝐿 0
)
1/ 2
𝐻𝐻 0
+
0.28
𝐻𝐻 0
𝐿𝐿 0
(0.56 tan
2
𝛽𝛽 + 0.004)
�
1/ 2
×
𝐻𝐻 0
(56 tan
2
𝛽𝛽 + 0.004)
tan 𝛽𝛽
(10)

The uncertainties of each term, δH, δL, δtan β, and δZt, are calculated following the methods
used by Ruggiero et al. (2012). Wave height, wavelength, and beach slope uncertainties are
calculated by finding the difference between the 95- and 50-percent exceedance statistics for
each term. The assumption that the HWL is at the same elevation as MHW and generated from a
previous high tide has the uncertainty of not knowing the water level that left the HWL. This
uncertainty is calculated by subtracting MHW from mean higher high water (MHHW). These
uncertainties and the derived uncertainty Equations (7- 10) are used in Equation (11) below to
calculate the overall proxy-datum bias uncertainty.

𝛿𝛿 𝐵𝐵𝐵𝐵𝐵𝐵 𝑠𝑠 = � �
𝜕𝜕 𝐵𝐵 𝐵𝐵𝐵𝐵𝐵𝐵
𝜕𝜕 𝑍𝑍 𝑇𝑇 δZ
𝑇𝑇 �
2
+ �
𝜕𝜕 𝐵𝐵 𝐵𝐵𝐵𝐵𝐵𝐵
𝜕𝜕 tan 𝛽𝛽 δ tan 𝛽𝛽 �
2
+ �
𝜕𝜕 𝐵𝐵 𝐵𝐵𝐵𝐵𝐵𝐵
𝜕𝜕 𝐻𝐻 0
δH
0
�
2
+ �
𝜕𝜕 𝐵𝐵 𝐵𝐵𝐵𝐵𝐵𝐵
𝜕𝜕 𝐿𝐿 0
δL
0
�
2
� (11)
34
Digital Shoreline Analysis System (DSAS)
DSAS is an add-on tool in ArcMAP created by USGS to determine the rate of shoreline
change. DSAS requires all input data to be stored in a personal geodatabase. At least two
digitized shorelines and a user defined baseline are required to calculate shoreline change
statistics. DSAS uses the baseline to create perpendicular transect lines that are used in the
shoreline change calculations. The DSAS workflow is outlined in Figure 11.

Figure 11. DSAS workflow from the DSAS User Guide (Himmelstoss et al. 2018). Step 7,
shoreline forecasting, was not used in this study.
35
3.3.1. Baseline, Shorelines, and Transects
The DSAS user guide (Himmelstoss et al. 2018) recommends creating a baseline by
adding a new feature class to manually draw a line or use a buffer of an existing shoreline. For
this project, a buffer 80 m to the left of the 2016 DEM shoreline is used as the baseline. The
required attribute fields for the baseline are given in Table 2. The ID field identifies segments of
the baseline in order from the beginning to the end of the line. DSAS uses this field to place
transect lines in sequence if the baseline is segmented.

Table 2. DSAS baseline geodatabase field requirements (Himmelstoss et al. 2018).
Field Name Data Type Attribute addition DSAS Requirement
OBJECTID Object identifier Autogenerated Required
SHAPE Geometry Autogenerated Required
SHAPE_Length Double Autogenerated Required
ID Long Integer User-Created Required

As required by DSAS, digitized shorelines were merged into a single feature class. A
DSAS_date field with the date the LiDAR and NAIP imagery were captured was added to the
attribute table as an identifier for each shoreline before the merge. The date field is required by
DSAS, the other required fields for the shoreline feature class are given in Table 3. The
DSAS_uncy is the shoreline position uncertainty. This is calculated using Equation (5) and (6)
for NAIP and DEM derived shoreline positions, respectively. The DSAS_type identifies the
specific shoreline as either MHW or HWL. This field is optional and is only used when the two
different shoreline proxies are used in the same change rate analysis.

36
Table 3. DSAS shoreline geodatabase field requirements (Himmelstoss et al. 2018).
Field Name Data Type
Attribute
Addition
DSAS
Requirement
OBJECTID Object identifier Autogenerated Required
SHAPE Geometry Autogenerated Required
SHAPE_Length Double Autogenerated Required
DATE_(DSAS_date) Text (Length=10 OR Length20) User-Created Required
UNCERTAINTY
(DSAS_uncy)
Any numeric field User-Created Required
SHORELINE_TYPE
(DSAS_Type)
Text User-Created Optional

Transects are automatically drawn by DSAS perpendicular to the baseline. Transect
spacing and length can be specified by the user. A transect spacing of 50 m was specified with
the option to clip transects at the furthest shoreline extent for this project. After the transects
were cast, they were manually inspected and edited to make sure they crossed all the shorelines
only once.
3.3.2. Complex Shoreline Data
DSAS allows for both MHW and HWL proxy shorelines to be used in the same shoreline
change analysis. To do this, DSAS uses a user defined proxy-datum bias correction (Equation 4)
to correct for the difference between the MHW and HWL shorelines. To be able to use both
shoreline proxy types, the MHW shorelines need to be converted into a calibrated route, which
adds an M-value to each vertex. The M-values are edited to have unique numbers that identify
their locations along the shoreline. This ID links the vertices along the MHW shoreline to the
37
proxy-datum bias table. This table contains the information DSAS uses for the proxy-datum bias
correction and associated uncertainty calculations, required fields are given in Table 4. The
UNCY field is the MHW shoreline uncertainty (UMHW) at each point calculated from Equation
(6). The bias field is the proxy-datum bias using Equation (4), and the UNCYB is the uncertainty
in the proxy-datum bias from Equation (11).

Table 4. DSAS proxy-datum bias table field requirements (Himmelstoss et al. 2018)
Field Name Data Type Field Description
ID Long Integer
Cross-shoreline LiDAR profile identifier stored as
the M-value at each vertex in the calibrated
shoreline route. This links the LiDAR shoreline
with the uncertainty table.
UNCY Any numeric field
Positional uncertainty associated with natural
influences over the shoreline position (wind,
waves, tides) as well as measurement uncertainties
associated with the collection of the LiDAR data.
BIAS Any numeric field
Proxy-datum bias value describing the
unidirectional horizontal offset between the MHW
elevation of the LiDAR data and the HWL
shoreline position.
UNCYB Any numeric field Uncertainty of the proxy-datum bas value.

3.3.3. Shoreline Change Statistics
The five statistical values calculated through DSAS are: (1) NSM; (2) SCE; (3) EPR; (4)
LRR; and (5) WLR. These statistics are calculated for every transect with rates reported in
meters per year. NSM returns the distance in meters between the oldest and youngest shorelines.
SCE is the distance in meters between the two furthest shorelines independent of time. The EPR
calculates the net shoreline movement rate by dividing NSM by the elapsed time between the
38
oldest and youngest shorelines. Both NSM and EPR only look at the oldest and youngest
shorelines and do not take into account any other shorelines.
LRR is calculated by plotting the distance of the shorelines from the baseline against
time. A least-squares regression line is fitted to the data, the slope of this line is the shoreline
change rate. Like LRR, WLR is calculated using a best fit line. With WLR, the shoreline points
are given weights based on their position uncertainty. Points with a smaller uncertainty have a
higher weight than those with large uncertainty when determining the placement of the best fit
line. Unlike EPR, LRR and WLR use every data point to determine shoreline change rate. Users
can specify a confidence level that DSAS uses to calculate the uncertainty of the LRR and WLR.
DSAS also calculates the R
2
statistic, for both LRR and WLR, that represents how well the best
fit line represents all of the data points (Himmelstoss et al. 2018).

39
Chapter 4 Results
Chapter 4 provides the results from the outlined methods in Chapter 3 for shoreline digitization
and shoreline change analysis through DSAS, using proxy- and datum-based shorelines.
Shoreline Digitization
Shorelines created from the LiDAR derived DEMs were fairly simple and did not need
much editing. When the contour lines were generated, contour lines were also placed where the
DEM layer ended. These lines were edited to be removed as they did not represent the shoreline.
The shorelines created from the NAIP imagery were initially a lot messier due to
classifying the pixels using NDWI. The polylines generated from the NDWI classified images
went through extensive editing to make sure they represented the HWL. Although a threshold
was used to help better classify the pixels, many pixels were mis-classified. Pixels containing
areas of shadows or roads were classified as water. Figure 12 shows where shadows interfered
with the detection of the HWL.
Another factor that influenced the effectiveness of the NDWI classification was the tide
level at the time the image was taken. The 2016 NAIP imagery was taken at high tide where the
HWL was not discernable. These images were ultimately not used as the NDWI classification
did not accurately represent the HWL. Figure 13 shows a section of 2016 NAIP imagery
compared with the NDWI classification applied with a 0.09 and 0.1 threshold. The shoreline
could have been extracted from the imagery using an alternative method but this was beyond the
scope of this project.

40

Figure 12. Lines generated from NAIP imagery with NDWI classification showing where
shadows and wave swash are mis-classified.
41

Figure 13. 2016 NAIP imagery (a) and NAIP imagery with NDWI classification applied with a 0.09
threshold (b) and 0.1 threshold (c).
42
Shoreline Change
After running the shoreline change analysis statistics, DSAS generates two new feature
classes. The first new feature class is a copy of the transect feature class with the addition of
shoreline change statistics. The second contains points that hold position records where the
transect lines cross the shorelines. In addition to these feature classes, a text document is
generated containing a summary report that provides the averages for each of the calculated
statistics.
4.2.1. Overall Shoreline Change
DSAS statistics were generated for two periods, 1997-2016 and 1997-2014. The
shoreline for 2016 was generated using LiDAR data collected on April 28, 2016. The period
from roughly July 2015 through April 2016 was considered a very strong El Niño (see Figure 3).
Powerful waves generated by large storm systems during El Niño events influence shoreline
position. Because of this, the DSAS statistics were run for the shorelines from 1997 to 2014 to
remove any variations that might have been caused by these larger than average storm surges.
The average shoreline change rates along with percent erosion and accretion from the
DSAS summary report are given for both the 1997-2016 and 1997-2014 shoreline date groups in
Table 5. The full summary reports for both date groups are provided in Appendix A and
Appendix B.
The shoreline change analysis for 1997-2016 shows that overall, there is a general trend
of accretion within the LLC. The average NSM between 1997 and 2016 is 5.18 m with an
average EPR of 0.32 m/yr. The average LRR is 1.21 m/yr while the average WRR is 1.06 m/yr.
43
Although the average shoreline rate shows that accretion is the dominant trend, there are sections
of the shoreline that are eroding.

Table 5. The average shoreline changes rates and percent erosion and accretion for each statistic
from the DSAS summary report for shorelines between 1997-2016 and 1997-2014.

Date Group SCE (m) NSM (m) EPR (m/yr) LRR (m/yr) WLR (m/yr)
1997-2016
Average 78.88 5.18
0.32 +/-
0.58
1.21 +/-
0.58
1.06 +/- 0.6
Percent
erosion
- 42.82% 42.82% 29.17% 28.94%
Percent
accretion
- 57.18% 57.18% 70.83% 71.06%
Maximum
erosion
- -57.67 -3.55 -3.15 -3.07
Maximum
accretion
- 168.47 12.38 10.57 10.21
1997-2014
Average 76.74 19.71 1.2 +/-0.6 1.86 +/- 0.7
1.42 +/-
0.73
Percent
erosion
- 24.94% 22.76% 22.76% 25.91%
Percent
accretion
- 75.06% 75.06% 77.24% 74.09%

Maximum
erosion
- -43.49 -2.7 -2.94 -2.84

Maximum
accretion
- 202.59 12.03 16.91 14.13
44
The EPR calculates the position change from the youngest shoreline to the oldest,
whereas the WLR plots all shoreline positions giving the points with smaller uncertainty values
more weight and uses the slope of a best fit line to calculate the rate of change. The change rate
value calculated by EPR and WLR are notably different, 0.32 m/yr and 1.06 m/yr respectively
for 1997-2016. The differences between these two rate calculations are compared in Figure 14.
Figures 15 and 16 compare the EPR and WLR rate graphs spatially to the shoreline. The WLR
rates show that the northern section of the shoreline is accreting at a faster rate than when
calculated by the EPR.
Like the 1997-2016 shoreline change rates, the 1997-2014 change rates show that there
was an overall trend of accretion throughout the study area. With the 2016 shoreline removed,
the average shoreline rate change increased. The EPR, LRR, and WLR rates differed by 0.88,
0.65, and 0.36 m respectively, while the NSM increased from 5.18 m to 19.71 m. The difference
in NSM from the shoreline groups show that the 2016 shoreline was located further inland
compared to the 2014 shoreline indicating an erosional period and agreeing with the trend of
erosion during El Niño events.
Both shoreline groups show that erosion is more frequent in the southern section of the
study area. This follows the general pattern where erosion is more prominent in the southernmost
section of the coastline when it is bound on either side by headlands, and that erosion occurs at
greater rates during El Niño warm phases.

45

Figure 14. Change rates in meters per year for shorelines between 1997-2016 calculated
by WLR (left) and EPR (right).
46

Figure 15. The EPR for the 1997-2016 shorelines.
47
Figure 16. The WLR for the 1997-2016 shorelines.
48
4.2.2. Shoreline Change with Respect to Seawalls
To better view the shoreline changes with respect to seawall position, the seawalls have
been sectioned into 19 groups and numbered from south to north. Figure 17 shows the location
of each seawall group and the NSM for both date groups. The NSM for the 1997-2016 shorelines
show that there is some correlation between seawall placement and areas of erosion and
accretion. The 1997-2014 shorelines show less of a correlation with the majority of erosion
occurring between shoreline groups 1 and 5. Table 6 lists the counts of erosion and accretion
occurring at the beginning and ends of the seawall groups. From this data, there is no discernible
pattern to where erosion and accretion are occurring in relation to seawalls.

Table 6. Count of the occurrence of erosion or accretion at the beginning and end of the
seawall groups.
Year Group
Change Rate
Statistic
Beginning End
Erosion Accretion Erosion Accretion
1997-2016
EPR 10 9 8 11
WLR 4 15 5 14
1997-2014
EPR 3 16 5 14
WLR 8 11 10 9

Shoreline rate change associated with SPSs is most apparent when looking at the EPR for
the 1997-2016 shorelines. Figure 18 takes a closer look at the 1997- 2016 EPR. In Figure 18 (c)
and (d) shoreline retreat is occurring directly at the ends of all the shoreline groups except 10, 13
and 19. Figure 19 shows the EPR for the 1997- 2014 shorelines, here shoreline change is not
occurring at the beginning and ends of the SPSs like in the 1997-2016 change rate map. Figure
19 (c) shows that for the most part, the shoreline is retreating only in front of the SPSs.
49

Figure 17. Comparison of NSM in meters from 1997-2014 (left) and 1997-2016
(right).
50

Figure 18. A closer look at the EPR for the 1997-2016 shorelines.
51

Figure 19. A closer look at the EPR for the 1997-2014 shorelines.
52
Chapter 5 Discussion and Conclusions
This chapter discusses the findings from the shoreline change analysis as well as the methods
used and how these findings and methods can be applied to future research and coastal policy.
Limitations and Future Improvements
NAIP imagery was chosen for this study due to its spatial and temporal availability,
having a 1 m resolution, as well as being free-of-charge to the public. An advantage to NAIP
imagery, compared to other aerial imagery, is that a CNIR sensor is used. CNIR sensors capture
the red, blue, green, and NIR wavelengths. Because each wavelength is recorded separately, the
NDWI was able to be calculated using the green and NIR bands. As discussed in Chapter 2, new
indices have been created to improve upon NDWI and better delineate pixels representing water
versus land. Some of these indices utilize middle infrared (MIR) wavelengths that are not
recorded by the CNIR sensors. Using satellite imagery that captures MIR in conjunction with
one of the updated water indices might have provided more accurate results and made shoreline
digitization slightly easier. The next available highest spatial resolution and free-of-charge data
for the study area was imagery from the Landsat satellites, but due to the 30 m resolution and a
mean daily tidal range of 1.9 m along the LLC shore, these images were not used.
The use of NDWI to digitize the shoreline was not optimal for the data and study area
used in this study. The placement of the HWL was highly variable based on when the image was
taken and the tide level at the time, as shown in Figures 12 and 13. Using a different method,
such as manual delineation, might provide more accurate and consistent shoreline placement.
Shoreline change analysis shows erosion and accretion through the seaward or landward
movement of the shoreline. Although this provides a good indication of where erosion and
53
accretion are occurring, it does not provide information on how the morphology of the shore is
changing. In addition to the methods used in this study, future analysis could use the DEMs to
analyze the sediment volume change to better visualize where erosion and accretion are
occurring with respect to the SPSs. This could provide more insight into the long-term effects of
the structures.
Analysis of Shoreline Change Rates
This study found that overall, the majority of the shoreline is accreting. Depending on the
statistical method used, the average change rates range from 0.32 to 1.42 m/yr. These rates vary
greatly compared to the short-term change rate calculated by Ruggiero et al. (2013) of -0.3 m/yr.
The difference between this study and that of Ruggiero et al. (2013) are the time periods
evaluated. Ruggiero et al. (2013) used shorelines from the 1960s to 2002 whereas this study
looked at shorelines from 1997 to 2016.
This difference can be explained by the decadal coastal variability described by Anderson
et al. (2018). In 2002 the PDO index was at the end of a cool phase. The latest warm PDO phase
started in 2014 and peaked in April 2016, which is when the LiDAR data was collected for both
the 2014 and 2016 shorelines used in this study. According to Anderson et al. (2018), warm PDO
phases are characterized by an increase in wave power along the Oregon coast compared to cool
phases. This results in erosion being focused at the southern ends of littoral cells and accretion at
the northern ends. This pattern of erosion and accretion is reversed during cool phases and can be
seen when the change rates calculated in this study are compared with those of Ruggiero et al.
(2013) and the PDO index shown in Figure 20.

54

As discussed in Chapters 1 and 2, it is known that SPSs can cause disruptions in the
natural cycle of sediment transport within a littoral cell. Specifically for seawalls, these
disruptions appear as sediment being trapped behind the wall and erosion occurring at the base
and ends of the wall. For the 1997-2016 shorelines the end point rate (EPR) showed that 47% of
the 19 seawall segments had erosion occurring at either end whereas the weighted linear
regression (WLR) showed only 21% had erosion. The EPR and WLR change rates for the 1997-
2014 shorelines were opposite with 24% and 42% respectively. Both the EPR and WLR for both
shoreline groups indicate that in general, the seawalls within the study area are serving their
purpose and protecting structures from shoreline retreat. Although the data show that accretion is
the prominent trend throughout the study area, they also highlight areas where erosion is
occurring adjacent to seawalls. This is made even more apparent when looking at the DEM
subtractions showing the change in elevation around the seawalls.
The difference in the 1997-2016 and 1997-2014 data show that the DSAS rates are
influenced by seasonal variability. This seasonal shoreline variability makes it difficult to
Figure 20. Graph showing the cool and warm phases of the PDO (NOAA 2021d).
55
analyze any long-term effects caused by seawalls. However, by comparing the two date groups
and identifying the patterns of seasonal variability, these results demonstrated that shoreline
change analysis can be used to identify areas of erosion caused by seawalls. This can be of
importance for coastal planners and managers when making decisions, such as setback distances,
that consider the variability and effects of 100-year storms.
Both the EPR and WLR change rates were compared in this study. Their calculated rate
changes for both date groups differed by 0.74 and 0.22 for 1997-2016 and 1997-2014
respectively. Although Genz et al. (2007) found WLR to be the most accurate statistic to
calculate rate changes with uncertainties, both Allen et al. (2003) and Ruggiero et al. (2013)
determined that WLR should only be used for long-term studies. This is because WLR assumes a
linear trend in shoreline change and can be influenced by outliers like shorelines recently
affected by a large storm system. EPR is used for short-term rate changes calculated by Ruggiero
et al. (2013) because it only uses the oldest and most recent shorelines and therefor does not
assume a linear trend. Seasonal variability is reflected in EPR as shown by the comparison of the
1997-2014 and 1997-2016 change rates in this study. In contrast, the 1997-2016 WLR more
closely resembled the 1997-2014 EPR.
Coastal Policy
Currently, Oregon has policies in place to help protect the beach and keep it accessible to
the public. Goal 18 is one of these policies that defines where SPSs can be constructed, and the
requirements needed for homeowners to acquire a permit to build one. Gardner (2015) assessed
these policies and discussed how they could be improved. One of these improvements suggests
that policies need to address future climate change and adaptation planning. Goal 18 only
specifies that SPSs need a permit if they are to be constructed beyond the vegetation line, which
56
defines the boundary between the state recreation area and private land (DLDC 2021; ORS 2019)
To bypass this, homeowners have been building SPSs behind the vegetation line to prepare for
future shoreline retreat; once these structures are exposed or collapse, it is the homeowner’s
responsibility to clean up the debris (Gardner 2015). This is becoming a more common
occurrence. In March of this year, an unpermitted seawall collapsed onto the beach leaving three
houses exposed to future erosion (Figure 21) (Brock 2021). This property is located at the
southernmost end of the LLC where current erosion rates are high. It has been nearly 20 years
since the last shift of the PDO. If the 20- to 25-year trend continues, the PDO should enter into a
warm phase soon. This implies that there will be an increase in El Niño events and ultimately an
increase in erosion, specifically at the southern end of the LLC.

Figure 21. Seawall collapse at the southern end of the LLC leaving three houses dangerously
exposed to further cliff (Brock 2021).
57
Two other improvements to policies Gardner (2015) highlights are specifying the
structure design and requiring impact reports of a structure for the entire littoral cell instead of
just to adjacent properties. Hard structures, such as seawalls and riprap, are currently the only
erosion mitigation efforts in place within the LLC. Gardner (2015) suggests that policy makers
look into alternate mitigation efforts such as vegetation stabilization, cobble revetments, and
beach nourishment. Once more is known about trends in shoreline variability associated with
long-term events, such as the PDO, Stive et al. (2002) proposes that beach nourishment can be
placed more effectively and efficiently.
This study shows that future research using similar short-term shoreline change analysis
can help by providing the necessary information needed to fill in the gaps of current coastal
policies in Oregon. With the correlation between the PDO and shoreline variability, shoreline
change analysis can focus on past events that are similar to current or future conditions. This can
then be used to improve policy relating to emergency situations like the SPSs collapse this year.
By analyzing the entire shoreline within a littoral cell over different periods of time and wave
conditions, a better understanding of how SPSs affect the entire littoral cell can be made.
Knowing how the shore changes under varying conditions and how SPSs influences this, will
lead to improved decision making and safer mitigation efforts.
Conclusions
Identifying the connection between the PDO and shoreline variability along the Oregon
Coast is a major step in better understanding coastal processes and how they affect the shore.
Most of the studies reviewed in Section 2.2 emphasize that future hazard assessments, policy
decisions, and SPSs construction take into consideration the centennial and decadal influences on
shoreline variability (Allan and Komar 2006; Barnard et al. 2015; Ruggiero et al 2013; Stive
58
2002). Future research can build upon studies like this one, that use short-term shoreline change
analysis to observe the affects SPSs have on the surrounding shore, to help preserve the coast
and coastal communities.
This study found that short-term shoreline change analysis can be used to identify
seasonal variability, specifically the variability between an average and El Niño years, and was
able to identify variability due to the PDO when compared to previous shoreline change analysis.
To improve upon this study and to get a better picture of shoreline variability, EPR analysis
should be used to compare shorelines that have been influenced by known factors, such as El
Niño or La Niña years. Comparing these could help identify trends in shoreline change directly
related the influencing factors and could potentially help identify other factors and trends in
shoreline change. Additionally, volumetric change analysis using LiDAR data would be
beneficial in the assessment of how SPSs affect shore morphology. Volume change analysis
would be able to show where sediment is gained and lost in relation to SPSs, help identify a
littoral cell’s sediment budget, and trends in sediment transport. There are many unknowns about
the processes that influence coastal morphology and how their interactions with SPSs will affect
the morphology. Hopefully, the methods and ideas presented in this study can be used to help
improve the way SPSs and associated erosion hazards are assessed and addressed.
59
References
Allan, J. C., Komar, P. D., Priest, G. R. 2003. “Shoreline Variability on the High-Energy Oregon
Coast and its Usefulness in Erosion-Hazard Assessments.” Journal of Coastal Research
Special Issue (38): 83-105.
Allan, J. C., Komar, P. D. 2006. “Climate Controls on US West Coast Erosion Processes.”
Journal of Coastal Research 22 (3): 511-529.
Anderson, D., Ruggiero, P., Antolinez, J. A. A., Mendez, F. J., Allan, J. 2018. “A Climate Index
Optimized for Longshore Sediment Transport Reveals Interannual and Multidecadal
Littoral Cell Rotations.” Journal of Geophysical Research: Earth Surface 123: 1958-
1981.
Barnard, P. L., Short, A. D., Harley, M. D., Splinter, K. D., Vitousek, S., Turner, I. L., Allan, J.,
et al. 2015. “Coastal Vulnerability Across the Pacific Dominated by El Niño/Southern
Oscillation.” Nature Geoscience 8: 801-808.
Boak, E. H., Turner, I. L. 2005. “Shoreline Definition and Detection: A Review.” Journal of
Coastal Research 21 (4): 688-703.
Brock, M. 2021 “Cliffside homes on Oregon Coast threatened by seawall collapse.” Oregon
Live. Accessed April 30, 2021. https://www.oregonlive.com/news/2021/03/cliffside-
homes-on-oregon-coast-threatened-by-seawall-collapse.html
Cai, W., Borlace, S., Lengaigne, M., van Rensch, P., Collins, M., Vecchi, G., Timmermann, A.,
et al. 2014. “Increasing Frequency of Extreme El Niño Events Due to Greenhouse
Warming.” Nature Climate Change 4: 111-116.
Cazenave, Anny, and Gonéri Le Cozannet. 2014. “Sea Level Rise and Its Coastal Impacts.”
Earth’s Future 2 (2): 15–34.
Darwish, K., Smith, S. E., Torab, M., Monsef, H., Hussein, O. “Geomorphological Changes
along the Nile Delta Coastline between 1945 and 2015 Detected Using Satellite Remote
Sensing and GIS.” Journal of Coastal Research 33 (4): 786-794.
Department of Land Conservation and Development Commission (DLDC). 2021. “Oregon’s
Statewide Planning Goals & Guidelines.” 1988. Goal 18: Beaches and Dunes. OAR 660-
015-0010(3). Accessed April 25, 2021.
https://www.oregon.gov/lcd/OP/Documents/goal18.pdf
Farris, A. S., Weber, K. M., Doran, K. S., List, J. H. “Comparing Methods Used by the U.S.
Geological Survey Coastal and Marine Geology Program for Deriving Shoreline Position
from Lidar Data.” U.S. Geological Survey Open-file report 2018-1121, 13p.,
60
Feyisa, Gudina L., Henrik Meilby, Rasmus Fensholt, and Simon R. Proud. 2014. “Automated
Water Extraction Index: A New Technique for Surface Water Mapping Using Landsat
Imagery.” Remote Sensing of Environment 140 (January): 23–35.
Fisher, Adrian, Neil Flood, and Tim Danaher. 2016. “Comparing Landsat Water Index Methods
for Automated Water Classification in Eastern Australia.” Remote Sensing of
Environment 175 (March): 167–82.
Gardner, M. 2015. “Analysis of Shoreline Armoring & Erosion Policies,” 64.
Genz, A. S., Fletcher, C. H., Dunn, R. A., Frazer, L. N., Rooney, J. J. 2007. “The Predictive
Accuracy of Shoreline Change Rate Methods and Alongshore Beach Variation on Maui,
Hawaii.” Journal of Coastal Research 23 (1): 87-105.
Gill, S. K., Schultz, J. R. 2001. “Tidal Datums and Their Applications.” NOAA National Ocean
Service Center for Operational Oceanographic Products and Services. 58.
Good, J. W. 1992. “Ocean Shore Protection Policy and Practices in Oregon: An Evaluation of
Implementation Success.” Doctoral Thesis. Oregon State University.
Harris, M., Brock, J., Nayegandhi, A., Duffy, M. 2006. “Extracting Shorelines from NASA
Airborne Topographic Lidar-Derived Digital Elevation Models.” U.S. Geological Survey
Open-file report 2005-1427.
Hiller, Priska H., Leif Lia, and Jochen Aberle. 2019. “Field and Model Tests of Riprap on Steep
Slopes Exposed to Overtopping.” Journal of Applied Water Engineering and Research 7
(2): 103–17.
Himmelstoss, E.A., Henderson, R.E., Kratzmann, M.G., and Farris, A.S., 2018, Digital Shoreline
Analysis System (DSAS) version 5.0 user guide: U.S. Geological Survey Open-File
Report 2018–1179, 110 p.,
IPCC. 2013. “Sea Level Change.” Climate Change 2013: The Physical Science Basics.
Contribution of Working Group I to the Fifth Assessment Report of the Intergovernmental
Panel on Climate Change 1137-1216.
Kauffman, D. 2018 “Additional Riprap Pictures.” Dan Kauffman Excavating, Inc. Accessed May
26, 2021. dan.kauffmanexcavating.com/riprap-more-pics.html.
Kelly, J. T., Gontz, A. M. 2018. “Using GPS-Surveyed Intertidal Zones to Determine the
Validity of Shorelines Automatically Mapped by Landsat Water Indices.” International
Journal of Applied Earth Observation and Geoinformation 65 (March): 92–104.
61
Komar, P. D., Allan, J. C., Ruggiero, P. 2011. “Sea Level Variations along the U.S. Pacific
Northwest Coast: Tectonic and Climate Controls.” Journal of Coastal Research 27 (5):
808-823.
Komar, P. D., Shih, S-M. 1993. “Cliff Erosion along the Oregon Coast: A Tectonic-Sea Level
Imprint Plus Local Controls by Beach Processes.” Journal of Coastal Research 9 (3):
747–65.
Kraus, N. C., McDougal W. G. 1996. “The Effects of Seawalls on the Beach: Part I, An Updated
Literature Review.” Journal of Coastal Research 12 (3): 691–701.
McFeeters, S. K. 1996. “The Use of the Normalized Difference Water Index (NDWI) in the
Delineation of Open Water Features.” International Journal of Remote Sensing 17 (7):
1425–32.
McFeeters, S. K. 2013. “Using the Normalized Difference Water Index (NDWI) Within a
Geographic Information System to Detect Swimming Pools for Mosquito Abatement: A
Practical Approach.” Remote Sensing 5 (2013): 3544-59.
Moore, L. J., Ruggiero, P., List, J. H., “Comparing Mean High Water and High Water Line
Shorelines: Should Proxy-Datum Offsets Be Incorporated into Shoreline Change
Analysis?” Journal of Coastal Research 22 (4): 894-905.
Newman, M., Alexander, M. A., Ault, T. R., Cobb, K. M., Deser, C., Di Lorenzo, E., Mantua, N.
J. 2016. “The Pacific Decadal Oscillation, Revisited.” Journal of Climate 29: 4399-4424.
NOAA. 2021a. “Cold & Warm Episodes by Season.” National Weather Service Climate
Prediction Center. Accessed April 15, 2021.
https://origin.cpc.ncep.noaa.gov/products/analysis_monitoring/ensostuff/ONI_v5.php
NOAA. 2021b. “Data Access Viewer.” 2021. NOAA. Accessed January 21, 2021.
https://coast.noaa.gov/dataviewer/#/LiDAR/search/
NOAA. 2021c. “Economics and Demographics.” 2020. Accessed November 8, 2020.
https://coast.noaa.gov/states/fast-facts/economics-and-demographics.html.
NOAA. 2021d. “Pacific Decadal Oscillation (PDO).” NOAA Physical Sciences Laboratory.
Accessed April 20, 2021. https://psl.noaa.gov/pdo/
NOAA. 2021e. “Station 46050.” NOAA National Data Buoy Center. Accessed March 21, 2021.
https://www.ndbc.noaa.gov/station_history.php?station=46050
NOAA. 2021f. “Tidal Datums.” NOAA Tides & Currents. Accessed April 8, 2021.
https://tidesandcurrents.noaa.gov/datum_options.html#NTDE
62
Oregon Parks and Recreation Department. 2021. “ArcGIS REST Services Directory.” Accessed
March 19, 2021.
https://maps.prd.state.or.us/arcgis/rest/services/Infrastructure/IF_Oregon_Shoreline_Prote
ction_Structures/FeatureServer
Oregon Revised Statutes (ORS). 2019. State and Local Parks; Recreation Programs; Scenic
Waterways; Recreation Trails, ch 390, no. 605 – 770 (2019). Accessed April 25, 2021.
https://www.oregonlegislature.gov/bills_laws/ors/ors390.html
Pilkey, O. H., Wright, H. L. 1988. “Seawalls Versus Beaches.” Journal of Coastal Research
Special Issue (4): 41–64.
Pranzini, E., Wetzel, L., Williams, A. T. 2015. “Aspects of Coastal Erosion and Protection in
Europe.” Journal of Coastal Conservation 19: 445-459.
Pranzini, E., Anfuso, G., Cinelli, I., Piccardi, M., Vitale, G. 2018. “Shore Protection Structures
Increase and Evolution on the Northern Tuscany Coast (Italy): Influence of Tourism
Industry.” Water 10 (11): 1647.
Priest, G. R. 1999. “Coastal Shoreline Change Study Northern and Central Lincoln County,
Oregon.” Journal of Coastal Research Special Issue (28): 140–57.
Reguero, B. G., Beck, M. W., Bresch, D. N., Calil, J., Meliane, I. 2018. “Comparing the Cost of
Nature-based and Coastal Adaptation: A Case Study from the Gulf Coast of the United
States.” PLOS One 13 (4): 1-24.
Ruggiero, P., List, J. H., 2009. “Improving Accuracy and Statistical Reliability of Shoreline
Position and Change Rate Estimates.” Journal of Coastal Research 25 (5): 1069-1081.
Ruggiero, P., Kratzmann, M.G., Himmelstoss, E.A., Reid, D., Allan, J., Kaminsky, G. 2013.
“National assessment of shoreline change—Historical shoreline change along the Pacific
Northwest coast” U.S. Geological Survey Open-File Report 2012–1007, 62 p.,
Shih, S.-M., Komar, P. D. 1994. “Sediments, Beach Morphology and Sea Cliff Erosion within an
Oregon Coast Littoral Cell.” Journal of Coastal Research 10 (1): 144–57.
Smith, M. J., Cromley, R. G. 2012. “Measuring Historical Coastal Change using GIS and the
Change Polygon Approach.” Transactions in GIS 16 (1): 3-15.
Stive, M. J. F., Aarninkhof, S. G. J., Hamm, L., Hanson, H., Larson, M., Wijnberg, M., Nicholls,
R. J., Capobianco, M. 2002. “Variability of Shore and Shoreline Evolution.” Coastal
Engineering 47: 211-235.
63
Stockdon, H. F., Sallenger, A. H., List, J. H., Holman, R. A. 2002. “Estimation of Shoreline
Position and Change using Airborne Topographic Lidar Data.” Journal of Coastal
Research 18 (3): 502-513.
Sunder, S., Ramsankaran, R, Ramakrishnan, B. 2017. “Inter-comparison of Remote Sensing
Sensing-based Shoreline Mapping Techniques at Different Coastal Stretches of India.”
Environmental Monitoring and Assessment 189: 290.
Trenberth, K. 2020. “The Climate Data Guide: Nino SST Indices (Nino 1+2, 3, 3.4, 4; ONI and
TNI).” Accessed May 25, 2021. https://climatedataguide.ucar.edu/climate-data/nino-sst-
indices-nino-12-3-34-4-oni-and-tni.
United States Department of Agriculture. 2017. National Agriculture Imagery Program (NAIP)
Information Sheet. Accessed April 2, 2021. https://www.fsa.usda.gov/Assets/USDA-
FSA-Public/usdafiles/APFO/support-documents/pdfs/naip_infosheet_2017.pdf
USGS. 2021. “Earth Explorer.” USGS Accessed January 21, 2021. https://earthexplorer.usgs.gov
van Rijn, L. C. 2011. “Coastal Erosion and Control.” Ocean & Coastal Management, Concepts
and Science for Coastal Erosion Management (Conscience) 54 (12): 867–87.
Weggel, J. R. 1988. “Seawalls: The Need for Research, Dimensional Considerations and a
Suggested Classification.” Journal of Coastal Research, 29–39.
Xu, H. 2006. “Modification of Normalised Difference Water Index (NDWI) to Enhance Open
Water Features in Remotely Sensed Imagery.” International Journal of Remote Sensing
27 (14): 3025–33.

64
Appendix A DSAS Summary Report for 1997 - 2016
File name: DSAS_Summary_Transects_two_20210415_213807.txt
Timestamp of rate calculation: 04/15/2021 21:40:13
DSAS version: 5.0.20200527.0200
ArcGIS version: 10.8
Rate types run: SCE, NSM, EPR, LRR, WLR
Baseline layer: BaselineBuffer_Line
Shoreline layer: shorelines
Shoreline dates used: 10/17/1997, 4/27/1998, 9/20/2002, 6/23/2009, 6/15/2012, 7/12/2014,
8/20/2014, 4/28/2016
Shoreline threshold: 0
Confidence Interval (CI) selected: 90
Default Uncertainty: 10
Transect spacing length: 50
Smoothing distance: 100
Coordinate system: NAD_1983_NSRS2007_StatePlane_Oregon_North_FIPS_3601
Is bias applied: YES

All rates reported are in meters/year, distance values are in meters.

DISTANCE: SCE (Shoreline Change Envelope, m)

SCE OVERALL AVERAGES:
total number of transects: 432
average distance: 78.88
maximum distance: 374.79
maximum distance transect ID: 385
minimum distance: 9.49
minimum distance transect ID: 16

DISTANCE: NSM (Net Shoreline Movement, m)

NSM OVERALL AVERAGES:
total number of transects: 432
average distance: 5.18
number of transects with negative distance: 185
percent of all transects that have a negative distance: 42.82%
maximum negative distance: -57.67
maximum negative distance transect ID: 72
average of all negative distances: -17.46
number of transects with positive distance: 247
percent of all transects that have a positive distance: 57.18%
maximum positive distance: 168.47
65
maximum positive distance transect ID: 226
average of all positive distances: 22.13

RATE: EPR (End Point Rate, m/yr)

EPR OVERALL AVERAGES:
total number of transects: 432
average rate: 0.32
average of the confidence intervals associated with rates: 1.97
reduced n (number of independent transects): 11
uncertainty of the average rate using reduced n: 0.58
average rate with reduced n uncertainty: 0.32 +/- 0.58

number of erosional transects: 185
percent of all transects that are erosional: 42.82%
percent of all transects that have statistically significant erosion: 3.01%
maximum value erosion: -3.55
maximum value erosion transect ID: 134
average of all erosional rates: -0.96

number of accretional transects: 247
percent of all transects that are accretional: 57.18%
percent of all transects that have statistically significant accretion: 9.49%
maximum value accretion: 12.38
maximum value accretion transect ID: 226
average of all accretional rates: 1.29

RATE: LRR (Linear Regression Rate, m/yr)

LRR OVERALL AVERAGES:
total number of transects: 432
average rate: 1.21
average of the confidence intervals associated with rates: 2.86
reduced n (number of independent transects): 24
uncertainty of the average rate using reduced n: 0.58
average rate with reduced n uncertainty: 1.21 +/- 0.58

number of erosional transects: 126
percent of all transects that are erosional: 29.17%
percent of all transects that have statistically significant erosion: 3.24%
maximum value erosion: -3.15
maximum value erosion transect ID: 135
average of all erosional rates: -0.8
66

number of accretional transects: 306
percent of all transects that are accretional: 70.83%
percent of all transects that have statistically significant accretion: 8.8%
maximum value accretion: 10.57
maximum value accretion transect ID: 385
average of all accretional rates: 2.04

RATE: WLR (Weighted Linear Regression, m/yr)

WLR OVERALL AVERAGES:
total number of transects: 432
average rate: 1.06
average of the confidence intervals associated with rates: 2.98
reduced n (number of independent transects): 25
uncertainty of the average rate using reduced n: 0.6
average rate with reduced n uncertainty: 1.06 +/- 0.6

number of erosional transects: 125
percent of all transects that are erosional: 28.94%
percent of all transects that have statistically significant erosion: 3.47%
maximum value erosion: -3.07
maximum value erosion transect ID: 134
average of all erosional rates: -0.86

number of accretional transects: 307
percent of all transects that are accretional: 71.06%
percent of all transects that have statistically significant accretion: 6.02%
maximum value accretion: 10.21
maximum value accretion transect ID: 385
average of all accretional rates: 1.84

67
Appendix B DSAS Summary Report for 1997 – 2014
File name: DSAS_Summary_Transects_two_20210421_132716.txt
Timestamp of rate calculation: 04/21/2021 13:28:56
DSAS version: 5.0.20200527.0200
ArcGIS version: 10.8
Rate types run: SCE, NSM, EPR, LRR, WLR
Baseline layer: BaselineBuffer_Line
Shoreline layer: shorelines
Shoreline dates used: 10/17/1997, 4/27/1998, 9/20/2002, 6/23/2009, 6/15/2012, 7/12/2014,
8/20/2014
Shoreline threshold: 0
Confidence Interval (CI) selected: 90
Default Uncertainty: 10
Transect spacing length: 50
Smoothing distance: 100
Coordinate system: NAD_1983_NSRS2007_StatePlane_Oregon_North_FIPS_3601
Is bias applied: YES

All rates reported are in meters/year, distance values are in meters.

DISTANCE: SCE (Shoreline Change Envelope, m)

SCE OVERALL AVERAGES:
total number of transects: 413
average distance: 76.74
maximum distance: 374.79
maximum distance transect ID: 385
minimum distance: 12.76
minimum distance transect ID: 107
68

DISTANCE: NSM (Net Shoreline Movement, m)

NSM OVERALL AVERAGES:
total number of transects: 413
average distance: 19.71
number of transects with negative distance: 103
percent of all transects that have a negative distance: 24.94%
maximum negative distance: -43.49
maximum negative distance transect ID: 78
average of all negative distances: -13.24
number of transects with positive distance: 310
percent of all transects that have a positive distance: 75.06%
maximum positive distance: 202.59
maximum positive distance transect ID: 244
average of all positive distances: 30.67

RATE: EPR (End Point Rate, m/yr)

EPR OVERALL AVERAGES:
total number of transects: 413
average rate: 1.2
average of the confidence intervals associated with rates: 1.82
reduced n (number of independent transects): 9
uncertainty of the average rate using reduced n: 0.6
average rate with reduced n uncertainty: 1.2 +/- 0.6

number of erosional transects: 103
69
percent of all transects that are erosional: 24.94%
percent of all transects that have statistically significant erosion: 1.45%
maximum value erosion: -2.7
maximum value erosion transect ID: 134
average of all erosional rates: -0.82

number of accretional transects: 310
percent of all transects that are accretional: 75.06%
percent of all transects that have statistically significant accretion: 32.2%
maximum value accretion: 12.03
maximum value accretion transect ID: 244
average of all accretional rates: 1.88

RATE: LRR (Linear Regression Rate, m/yr)

LRR OVERALL AVERAGES:
total number of transects: 413
average rate: 1.86
average of the confidence intervals associated with rates: 3.08
reduced n (number of independent transects): 19
uncertainty of the average rate using reduced n: 0.7
average rate with reduced n uncertainty: 1.86 +/- 0.7

number of erosional transects: 94
percent of all transects that are erosional: 22.76%
percent of all transects that have statistically significant erosion: 3.63%
maximum value erosion: -2.94
maximum value erosion transect ID: 135
70
average of all erosional rates: -0.94

number of accretional transects: 319
percent of all transects that are accretional: 77.24%
percent of all transects that have statistically significant accretion: 30.99%
maximum value accretion: 16.91
maximum value accretion transect ID: 385
average of all accretional rates: 2.69

RATE: WLR (Weighted Linear Regression, m/yr)

WLR OVERALL AVERAGES:
total number of transects: 413
average rate: 1.42
average of the confidence intervals associated with rates: 3.29
reduced n (number of independent transects): 20
uncertainty of the average rate using reduced n: 0.73
average rate with reduced n uncertainty: 1.42 +/- 0.73

number of erosional transects: 107
percent of all transects that are erosional: 25.91%
percent of all transects that have statistically significant erosion: 3.63%
maximum value erosion: -2.84
maximum value erosion transect ID: 134
average of all erosional rates: -0.91

number of accretional transects: 306
percent of all transects that are accretional: 74.09%
71
percent of all transects that have statistically significant accretion: 14.77%
maximum value accretion: 14.13
maximum value accretion transect ID: 385
average of all accretional rates: 2.24

Abstract (if available)

Abstract The Lincoln Littoral Cell (LLC) contains a 24 km stretch of coastline along Oregon’s central coast. Nearly 50% of the LLC’s coastline has been armored with shoreline protection structures (SPSs), mainly riprap and seawalls. SPSs are constructed to reduce damage to coastal developments caused by breaking waves, flooding, and sediment erosion. Although the SPSs are meant to protect the coast from erosion, they can ultimately cause erosion adjacent to the structure or further down shore. Future projections and models show an increase in frequency of large storm systems that generate larger than average waves and water levels, resulting in increased erosion and flooding. This project utilizes the USGS’s ArcGIS add-on, Digital Shoreline Analysis System (DSAS), to analyze digitized shoreline positions from 1997 to 2016. Visual analysis shows that while on average the shoreline is accreting at a rate of 0.32 m/yr, there is localized erosion adjacent to 53% of the SPSs. Future policies regarding the placement and build of SPSs should take into consideration the long-term negative effects of these structures.

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

A spatial investigation of New York City's historical shoreline

PDF

The use of site suitability analysis to model changes in beach geomorphology due to coastal structures

PDF

Utilizing advanced spatial collection and monitoring technologies: surveying topographical datasets with unmanned aerial systems

PDF

Integrating GIS into farm operations at the Homer C. Thompson Research Farm in Freeville, New York

PDF

Creating a Web GIS to support field operations and enhance data collection for the Animal and Plant Health Inspection Service (APHIS)

PDF

A comparison of two earthquake events in the City of Downey: the Puente Hills and Whittier faults at 7.0 and 6.8 magnitudes

PDF

Radar horizon estimation from monoscopic shadow photogrammetry of radar structures: a case study in the South China Sea

PDF

Modeling historic structure preservation candidacy on Fort Ord

PDF

An application of aerial drones in high definition mapping for autonomous vehicles

PDF

Operational optimization model for Hungry Marketplace using geographic information systems

PDF

Wetland mapping and restoration decision making using remote sensing and spatial analysis: a case study at the Kawainui Marsh

PDF

Using GIS to predict human movement patterns in complex humanitarian emergencies: a test case of the Syrian Conflict

PDF

Roadway hazard analysis: a safe ride for motorcycles

PDF

Designing an early warning system web mapping application for the Atlanta Metropolitan Area before a flooding event

PDF

GIS-aided production of equipment locator maps for Metro rail maintenance and support

PDF

An analysis of the North Rainier Elk Herd area, Washington: change detection and habitat modeling with remote sensing and GIS

PDF

Trojan Food Finder: a web-based GIS campus food sharing application

PDF

GIS analysis of helicopter rescue in San Bernardino County, California

PDF

Using GIS to perform a risk assessment for air-transmitted bioterrorism within San Diego County

PDF

Implementing spatial thinking with Web GIS in the non-profit sector: a case study of ArcGIS Online in the Pacific Symphony

Asset Metadata

Creator Bennett, Melina (author)

Core Title Using the Digital Shoreline Analysis System (DSAS) to analyze changes in shoreline position caused by seawalls along a section of Oregon's coast

School College of Letters, Arts and Sciences

Degree Master of Science

Degree Program Geographic Information Science and Technology

Degree Conferral Date 2021-08

Publication Date 07/26/2021

Defense Date 05/10/2021

Publisher University of Southern California (original), University of Southern California. Libraries (digital)

Tag ArcGIS,Coast,digital shoreline analysis system,GIS,Lincoln City,OAI-PMH Harvest,Oregon,riprap,seawall,shoreline,shoreline protection structures

Format application/pdf (imt)

Language English

Contributor Electronically uploaded by the author (provenance)

Advisor Fleming, Steven (committee chair), Marx, Andrew (committee member), Wilson, John (committee member)

Creator Email melina.rose@me.com,mrbennet@usc.edu

Permanent Link (DOI) https://doi.org/10.25549/usctheses-oUC15622963

Unique identifier UC15622963

Legacy Identifier etd-BennettMel-9885

Document Type Thesis

Format application/pdf (imt)

Rights Bennett, Melina

Type texts

Source University of Southern California (contributing entity), University of Southern California Dissertations and Theses (collection)

Access Conditions The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the author, as the original true and official version of the work, but does not grant the reader permission to use the work if the desired use is covered by copyright. It is the author, as rights holder, who must provide use permission if such use is covered by copyright. The original signature page accompanying the original submission of the work to the USC Libraries is retained by the USC Libraries and a copy of it may be obtained by authorized requesters contacting the repository e-mail address given.

Repository Name University of Southern California Digital Library

Repository Location USC Digital Library, University of Southern California, University Park Campus MC 2810, 3434 South Grand Avenue, 2nd Floor, Los Angeles, California 90089-2810, USA

Repository Email cisadmin@lib.usc.edu