Shoreline Armoring: Impacts on Nearshore Habitat in the Maury Island Aquatic Reserve

Item

Title

Eng Shoreline Armoring: Impacts on Nearshore Habitat in the Maury Island Aquatic Reserve

Date

2018

Creator

Eng Miller, Kirsten

Subject

Eng Environmental Studies

extracted text

SHORELINE ARMORING:
IMPACTS ON NEARSHORE HABITAT
IN THE MAURY ISLAND AQUATIC RESERVE

by
Kirsten Miller

A Thesis
Submitted in partial fulfillment
of the requirements for the degree
Master of Environmental Studies
The Evergreen State College
June 2018

2018 by Kirsten Miller. All rights reserved.

This Thesis for the Master of Environmental Studies Degree
by
Kirsten Miller

has been approved for
The Evergreen State College
by

Erin Martin
Member of the Faculty

Date

ABSTRACT
Shoreline Armoring in the Puget Sound: Impacts on
nearshore habitat in the Maury Island Aquatic Reserve
Kirsten Miller
Shoreline armoring is widespread in the Puget Sound, Washington, but the impacts on the
biological features of nearshore ecosystems have only recently begun to be documented.
Shoreline armoring disrupts the connection between marine and terrestrial ecosystems
along the shoreline and can decrease the availability of prey resources for juvenile
salmon. Most previous work has been conducted in highly urban areas, and this study
aims to strengthen our understanding of residentially-developed, high-bank shorelines
characteristic of the central Puget Sound. Here we determine differences in shoreline
vegetation, terrestrial insect assemblages, wrack coverage and composition, and fish
assemblages between armored and unarmored beaches. Citizen scientists with Vashon
Nature Center’s BeachNET program collected data in the summer of 2017 at three
beaches following protocols from the Washington Sea Grant’s Shoreline Monitoring
Toolbox. Results from this study determine that natural beaches have more overstory
vegetation, trees, and native plant species. Terrestrial insect abundance and taxa richness
was similar at armored and natural beaches, but natural shorelines host a greater percent
composition of Diptera, an important prey species for juvenile salmon. Forage fish
spawning occurred at armored and natural shorelines, however, natural shorelines hosted
a far greater number of sand lance eggs. Natural shorelines had higher abundance and
taxa richness of fish. This study suggests shoreline armoring alters shoreline conditions
and decreases the availability of key habitat and prey resources for key juvenile salmon
species in residentially-developed shorelines of the Puget Sound.

Table of Contents
Page
Table of Contents ............................................................................................................... iv
List of Figures and Tables................................................................................................. vii
Figures......................................................................................................................... vii
Tables ......................................................................................................................... viii
Acknowledgements ............................................................................................................ ix
Chapter 1: Introduction ........................................................................................................1
Background of this study ...............................................................................................3
Roadmap of thesis ..........................................................................................................5
Chapter 2: Literature Review ...............................................................................................5
Shoreline armoring alters physical beach processes ......................................................8
Shoreline armoring alters biological function of the nearshore ...................................10
Shoreline armoring management in Washington State ................................................11
Effects of shoreline armoring on key nearshore habitat features .................................13
Marine riparian vegetation .....................................................................................13
Beach wrack ...........................................................................................................17
Aquatic and terrestrial invertebrates ......................................................................20
Forage fish .............................................................................................................24
Fish assemblages ....................................................................................................26
Research needs .............................................................................................................32
Chapter 3: Methods ............................................................................................................33
Introduction ..................................................................................................................33
Site description.............................................................................................................34
Big Beach ...............................................................................................................38
Lost Lake ...............................................................................................................38
Piner Point..............................................................................................................39
Experimental design.....................................................................................................40
Sample timing and frequency ................................................................................41
Surveys ...................................................................................................................42
Analysis........................................................................................................................48

iv

Chapter 4: Results ..............................................................................................................50
Terrestrial vegetation ...................................................................................................50
Treatment effects on overstory percent cover ........................................................50
Variations in overstory percent cover based on site ..............................................51
Variation in understory percent cover by treatment ..............................................52
Variation in understory percent cover by site ........................................................53
Overhanging trees per treatment ............................................................................54
Overhanging trees per site .....................................................................................55
Native vs. non-native species counts per treatment ...............................................56
Abundance of native vs. non-native species per site .............................................57
Wrack ...........................................................................................................................58
Variation in wrack total percent cover based on treatment....................................58
Site as a basis for variation in wrack total percent cover .......................................59
Variation in marine algae in wrack based on treatment .........................................60
Site as a basis for variation in marine wrack cover ...............................................61
Variation in terrestrial wrack percent cover based on treatment ...........................62
Site as a basis for variation in terrestrial wrack cover ...........................................63
Variations in eelgrass percent cover in wrack based on treatment ........................64
Site as a basis for variation in eelgrass wrack cover ..............................................65
Terrestrial insects .........................................................................................................66
Insect density by treatment ....................................................................................67
Insect density per site .............................................................................................68
Taxa richness per treatment ...................................................................................69
Taxa richness per site .............................................................................................70
Forage Fish...................................................................................................................71
Forage fish spawning at Big Beach .......................................................................74
Forage fish spawning at Lost Lake ........................................................................75
Forage fish spawning at Piner Point ......................................................................76
Fish assemblages ..........................................................................................................76
Fish Observations...................................................................................................79
Chapter 5: Discussion ........................................................................................................81
Marine riparian vegetation ...........................................................................................81
Wrack cover .................................................................................................................82
v

Terrestrial insects .........................................................................................................85
Forage fish ...................................................................................................................87
Fish assemblages ..........................................................................................................88
Suggestions for future fish surveys ........................................................................89
Study design and suggestions for future research ........................................................91
Beaches in the MIAR are unique ...........................................................................91
Localized geomorphology in the MIAR ................................................................93
Citizen science: a challenge and a resource ...........................................................94
Working with small sample sizes ..........................................................................95
Suggestions for planning, management, and restoration .............................................96
Chapter 6: Conclusion........................................................................................................97
References ............................................................................................................................1
Appendices ...........................................................................................................................1
Appendix A. Map of Maury Island Aquatic Reserve ....................................................1
Appendix B. Shoreline Monitoring Toolbox vegetation sampling protocol .................2
Appendix C. Shoreline Monitoring Toolbox wrack sampling protocol ........................3
Appendix D. Shoreline Monitoring Toolbox insect sampling protocol ........................4
Appendix E. Shoreline Monitoring Toolbox fish protocol ............................................5

vi

List of Figures and Tables
Page

Figures
Figure 1: Armoring removal sites on Vashon and Maury Islands .....................................35
Figure 2: Slope stability in the Maury Island Aquatic Reserve .........................................37
Figure 3: Drift cells in the Maury Island Aquatic Reserve ................................................39
Figure 4: Percent cover of overstory vegetation per treatment ..........................................51
Figure 5: Percent cover of overstory vegetation per site ...................................................52
Figure 6: Percent cover of understory vegetation per treatment ........................................53
Figure 7: Percent cover of understory vegetation per site .................................................54
Figure 8: The number of overhanging trees per treatment. ...............................................55
Figure 9: Number of overhanging trees per site ................................................................56
Figure 10: Number of native and non-native species per treatment ..................................57
Figure 11: Number of native and non-native species at each site......................................58
Figure 12: Percent cover of total wrack per treatment. ......................................................59
Figure 13: Percent cover of total wrack per site ................................................................60
Figure 14: Percent cover of marine wrack per treatment ...................................................61
Figure 15: Percent cover of marine wrack per site ............................................................62
Figure 16: Percent cover of terrestrial wrack per treatment ..............................................63
Figure 17: Percent cover terrestrial wrack per site ............................................................64
Figure 18: Percent cover of eelgrass per treatment ............................................................65
Figure 19: Percent cover of eelgrass wrack per site. .........................................................66
Figure 20: Average insect density (individuals/m²) per treatment. ...................................68
Figure 21: Average insect density (individuals/m²) per site. .............................................69
Figure 22: Average taxa richness per treatment ................................................................70
Figure 23: Average taxa richness per site ..........................................................................71
Figure 24: Average number of surf smelt and sand lance eggs per treatment ...................72
Figure 25: Number of surf smelt and sand lance spawning events per treatment .............73
Figure 26: Number of surf smelt and sand lance spawning events per site .......................74

vii

Tables
Page
Table 1: Percent composition of terrestrial invertebrates ..................................................67
Table 2: Fish observations .................................................................................................80

viii

Acknowledgements
I would like to thank Bianca Perla and Maria Metler at the Vashon Nature Center for
working with me for the duration of this thesis. Thank you for your time,
encouragement, support, and resources to make this project happen. Thanks to
Vashon Island BeachNET volunteers for data collection and your interest in healthy
shorelines!
I would like to thank my advisor, Erin Martin, who helped in every part of this thesis
– from initial project planning to providing final edits – spending time to encourage
and provide feedback along the way.
Thanks to Ben Leonard for the assisting with snorkel surveys and statistics.
I would also like to thank my parents, Burt and Debbie Miller, for always lending me
an ear and providing hot showers after long days of snorkeling.

ix

SHORELINE ARMORING IMPACTS IN PUGET SOUND

Chapter 1: Introduction
About a third of the Puget Sound’s shorelines are altered with some form of
shoreline armoring (Puget Sound Partnership, 2012). Shoreline armoring is put into place
to prevent erosion and stabilize shorelines to allow for commercial and residential
development. Armoring can include seawalls, bulkheads, and revetments constructed of
large rock, concrete, wood, or steel. Although these structures are important for
development along the shorelines, there is an increasing understanding that shoreline
armoring may cause adverse ecological impacts along shorelines.
Armoring has been found to alter shorelines, disrupting the connectivity between
marine and terrestrial ecosystems. Armoring is known to reduce shoreline vegetation,
decrease terrestrial insect abundance and diversity, decrease wrack composition, and
reduce egg survival rates for forage fish. Armoring can also alter diet and feeding
behavior of juvenile salmon in the nearshore, as they rely on shallow, productive
nearshore habitats for foraging and refuge from predators during their outmigration from
natal streams to the sea (Heerhartz & Toft, 2015).
Armor removal and beach restoration is a priority in the Puget Sound region in
Washington State, driven by the need to protect Pacific salmon species such as
endangered populations of Chinook salmon (Oncorhynchus tshawytscha), an important
cultural, ecological, and economic resource (Toft et al., 2014). In addition, the Puget
Sound Partnership, a state agency leading the region’s collective effort in Puget Sound

1

SHORELINE ARMORING IMPACTS IN PUGET SOUND
recovery, has set targets for a net decrease of armored shorelines, or more armoring
removed or restored than developed, by 2020.
Although the pace of shoreline armoring development has slowed, armoring in the
Puget Sound is still increasing, as coastal habitats in the Puget Sound face unprecedented
urban growth (Gittman et al., 2015). As shoreline development infringes on Puget Sound,
potentially increasing the need for armoring along beaches, understanding the impacts of
shoreline armoring on terrestrial and aquatic environments along the shoreline is vital for
effective management.
There has been a recent momentum in the Puget Sound region to restore armored
shorelines through removal of armoring structures, addition of sediments, re-planting
native riparian vegetation, and addition of logs and woody debris (Toft et al., 2013, Lee
et al., 2018). However, there is still little scientific information available to assess
impacts of shoreline armoring and beach restoration benefits (Sobosinski, 2003). There is
especially a need for highly localized studies to characterize coastal biota response to
armoring across the highly diverse Puget Sound region (Lee et al., 2018).
This thesis provides a highly localized study of the effects of shoreline armoring
on Vashon and Maury Islands located in the Maury Island Aquatic Reserve (MIAR),
establishing a basic understanding of the physical and biological differences between
armored and natural shorelines. This thesis is the first part in a longer study conducted by
the Vashon Nature Center (VNC) that will assess shoreline armoring removal in the
MIAR. The results presented in this thesis establish a baseline of shoreline conditions
before restoration, which will occur in the summer of 2018. Post-monitoring will occur

2

SHORELINE ARMORING IMPACTS IN PUGET SOUND
after armoring removal in 2018. The results from this study will be used as a comparison
to post-restoration conditions.
This thesis specifically seeks to answer: to what extent are shoreline vegetation
coverage, terrestrial insect assemblages, and wrack accumulation different between
armored and natural sites in nearshore habitats in the MIAR, Puget Sound? Do these
differences affect fish use of these nearshore habitats? Understanding the impacts of
shoreline armoring and whether restoration has the intended benefits is essential for
understanding biological recovery of shorelines and encouraging proper shoreline
management (Lee et al., 2018). Understanding benefits of shoreline restoration may
encourage shoreline restoration, favor alternative stabilization techniques, and reduce
future shoreline armoring development in the Puget Sound region.

Background of this study
The goal of this study is to gain a solid understanding of shoreline conditions in
the MIAR and how shoreline armoring impacts the local nearshore ecosystem. King
County purchased three properties on Maury Island (Big Beach, Lost Lake, and Piner
Point) to remove shoreline armoring and restore natural nearshore processes. An
important restoration goal and project funding for King County is to improve habitat for
out-migrating juvenile salmon. Shoreline armoring alters the key nearshore habitat
features that juvenile salmon depend on in their early life histories. All structures and
bulkheads will be removed from each restoration site and natural shoreline and hillslope
processes will be restored to the maximum extent practical (Booth & Legg, 2017).

3

SHORELINE ARMORING IMPACTS IN PUGET SOUND
Beaches at the study sites are unique as they are steep, highly erosive shorelines.
Highly localized studies are necessary to determine the impacts of armoring and how
restoration benefits the nearshore along beaches in the MIAR. Bulkhead removal may
allow for the return of natural erosional processes which increases habitat benefits,
including increased sediment delivered to the nearshore which helps create shallow water
habitat important for juvenile salmon survival (Booth & Legg, 2017).
Researchers from the VNC lead a group of citizen scientist volunteers to monitor
Big Beach, Lost Lake, and Piner Point in the MIAR. Three shoreline types were
monitored at each beach including a natural shoreline, an armored shoreline, and an
armored shoreline where armoring will be removed in the summer of 2018. For this
thesis, armoring was not removed at the “pre-restoration” site, but data was collected
separately to characterize the habitat prior to the 2018 armoring removal process. This
study implemented standardized monitoring protocols from the Puget Sound
Partnership’s Shoreline Monitoring Toolbox (Shoreline Monitoring Toolbox, 2017). This
study can be used as a model for groups to coordinate systematic studies along beaches
with citizen science volunteers.
The data collected focuses on biotic parameters that serve as a metric for healthy
shoreline habitat, including: shoreline vegetation, terrestrial insect assemblages, wrack
coverage and composition, forage fish spawning, and fish observations from snorkel
surveys. Results will be presented that demonstrate armoring reduces marine riparian
vegetation, alters the composition of terrestrial invertebrates, wrack composition, and
forage fish spawning, and decreases abundance and taxa richness of fish. The results also

4

SHORELINE ARMORING IMPACTS IN PUGET SOUND
establish a pre-restoration baseline of habitat conditions at beaches targeted for
restoration in the summer of 2018.

Roadmap of thesis
I will first summarize what is currently known about the effects of shoreline
armoring on physical and biological shoreline conditions through a literature review
including: marine riparian vegetation, terrestrial invertebrates, wrack accumulation,
forage fish spawning, and fish use. The studies presented in my review focus on research
assessing differences between developed and natural sites along Puget Sound shorelines.
Next, methodology and statistical analysis are presented, highlighting the standardized
Shoreline Monitoring Toolbox protocols. Results will be presented that show armoring
reduces marine riparian vegetation, alters the composition of terrestrial invertebrates,
wrack composition, and forage fish spawning, and decreases abundance and taxa richness
of fish. These results will be placed into context with the current literature in the
Discussion section, in addition to discussing methodology and recommendations for
future VNC beach monitoring surveys. I conclude by summarizing the results and
highlighting the importance of continuing long-term monitoring studies to assess the
biological response to armoring and restoration and potentially encourage armoring
reduction and shoreline restoration in the future.

Chapter 2: Literature Review
Worldwide, shorelines adjacent to bodies of fresh and salt waters face faster
urbanization and population growth than other geographic regions (Neumann et al.,
5

SHORELINE ARMORING IMPACTS IN PUGET SOUND
2015). Coastal regions are known to experience high immigration rates because of their
ease of access to domestic and international shipping, military and defense uses, tourism,
access to recreational activities, and employment opportunities (Bulleri et al., 2005;
Gittman et al., 2015; Neumann et al., 2015). Coastal infrastructure and urban centers are
exposed to various coastal hazards in these areas, such as storms, large waves, flooding,
sea level rise, and erosion. In response, many coastal communities have established
hardened structures including bulkheads, jetties, riprap revetments and seawalls, a
practice commonly called “shoreline armoring” (Chapman & Underwood, 2001;
Heerhartz et al., 2014; Gittman et al., 2015). In some large urban centers, such as San
Diego Bay, Chesapeake Bay, Sydney Harbor, and Hong Kong’s Victoria Harbor, over
50% of the shorelines are armored (Gittman et al., 2015). In the United States alone,
about 14% of the lower 48 states’ shorelines are armored, and 64% of these armored
shorelines are adjacent estuaries and coastal rivers (Gittman et al., 2015). As coastal
immigration and urban centers experience increased growth and development, the rate of
shoreline armoring is expected to rise (Davis et al., 2002; Dugan et al., 2008; Lam et al.,
2009).
Armored shorelines are associated with lower biodiversity, vegetation cover, and
abundances of invertebrates and fish (Moreira et al., 2006; Dugan et al., 2008; Morley et
al., 2012). Armored shorelines can increase beach erosion, as waves reflect off of
armored shorelines (Heatherington & Bishop, 2012). Armoring can reduce the overall
ecological health of coastal ecosystems by degrading shallow intertidal habitats that are
vital to the survival of juvenile fish and aquatic invertebrates (Bilkovic & Roggero, 2008;

6

SHORELINE ARMORING IMPACTS IN PUGET SOUND
Seitz et al., 2006; Gittman et al., 2016). Armored shorelines disrupt the transition
between aquatic and terrestrial habitats and decreases deposition of woody debris and
“wrack”, or organic matter deposited on shorelines (Heerhartz et al., 2014; Lee et al.,
2018). This loss of organic debris affects the aquatic-terrestrial food web including fishes
and macroinvertebrates associated with wrack and vegetated habitats (Bozek & Burdick,
2005; Dugan et al., 2008; Heerhartz et al., 2014; Heerhartz & Toft, 2015; Dethier et al.,
2016).
In Puget Sound, Washington, the shoreline is highly valued, as it serves as a
platform for recreational boating and shipping, commercial growth, and urban and
suburban development (Sobosinski, 2003). Currently, about a third of Puget Sound
shorelines are altered by some form of shoreline armoring (Puget Sound Partnership,
2012). Shoreline armoring in the Puget Sound is increasing. More permitted armor was
gained than lost cumulatively since 2011, resulting in a net cumulative length of 0.8 miles
of new armor between the years of 2011 and 2016 (Puget Sound Partnership, 2012).
Recently, there has been momentum to restore armored shorelines through
removal or armoring structures, nourishment of sediments, replanting native riparian
vegetation, and addition of logs and woody debris (Toft et al., 2013; Toft et al., 2014).
Restoration efforts are driven by the need to protect Pacific salmon (Oncorhynchus
tshawytscha) that are of cultural, ecological, and economic importance to the region
(Rhodes et al., 2006; Munsch et al., 2016). Shallow intertidal areas are known to serve as
nursery habitats for juvenile salmon, providing food and refuge from predators (Toft et
al., 2016). As shoreline development infringes on Puget Sound beaches, understanding

7

SHORELINE ARMORING IMPACTS IN PUGET SOUND
the localized impacts of armoring on terrestrial and marine environments across the Puget
Sound in the is necessary for effective management and to encourage shoreline
restoration and reduce the overall armoring in the Puget Sound region.
The purpose of this literature review is to summarize what is currently known
about the impacts of shoreline armoring on biological shoreline conditions resulting from
shoreline armoring and associated habitat alteration. Guided by regional Puget Sound
recovery goals, this review focuses on the effects of shoreline armoring on shoreline
habitat health, with applications for juvenile salmon. This review is organized in sections
corresponding to each biotic measure included in this study: vegetation, terrestrial
invertebrates, wrack accumulation, forage fish spawning, and fish use along the
shoreline. These biotic parameters serve as a metric for healthy shoreline habitat,
specifically focusing on features vital for juvenile salmon habitat. Throughout the paper, I
will specifically highlight a case study of shoreline restoration project at the Olympic
Sculpture Park, in Seattle, Washington, that looked at similar shoreline parameters before
and after the removal of shoreline armoring. The restoration improved the biological
function of the nearshore in a highly urbanized shoreline, mimicking natural beaches. The
Olympic Sculpture Park case study demonstrates how effective management and
restoration can increase natural shoreline function and increase vital habitat for juvenile
salmon in Puget Sound (Toft et al., 2013).

Shoreline armoring alters physical beach processes
Shoreline armoring, including seawalls, bulkheads, and revetments, is put in place
to protect shorelines from naturally eroding beaches and stabilize areas for upland
8

SHORELINE ARMORING IMPACTS IN PUGET SOUND
commercial and residential uses (Shipman, 2010). Shoreline armoring is known to alter
physical nearshore processes. The nearshore, for the purposes of this document, is
defined as the physical area that extends from the far edge of the photic zone to the
adjacent uplands, including the top of any associated bluffs (Guttman, 2009). Shoreline
armoring replaces natural beaches with hard, vertical surfaces, acting as a physical barrier
between terrestrial and aquatic ecosystems that were once connected (Sobosinski, 2003;
Ecology, 2016). Physical structures cut off, or “lock up”, the natural delivery of sand and
gravel to the shoreline from both marine and terrestrial sources (Ecology, 2016; Dethier
et al., 2016). When waves reflect off these structures, they scour away sediments which
are not replaced (Shipman, 2010). This causes beaches in front of armored sites to erode
slowly, leading to gradual lowering or even the disappearance of the beach (Ecology,
2016). This process is referred to as the “truncation” of the beach (Johannessen &
MacLennan, 2007), where armored beaches tend to be narrower, steeper, and coarsergrained (Nordstrom, 2014).
The “coastal squeeze” is a term used to describe coastal habitat loss due rising sea
levels along armored shorelines. Shoreline armoring creates a static, artificial margin
between land and sea. As sea levels rise and increased storms push the coastal habitats
landward, shoreline armoring prevents the upper beach from migrating inland. Beach
habitats become “squeezed” into a narrowed zone (Doody, 2013; Dethier et al., 2016).
The narrowing of the beach can eliminate shallow water habitat directly adjacent to
shore, which is vital habitat for fish (Munsch et al., 2016). Armoring can alter physical
beach conditions locally and have broader, cumulative impacts across the Puget Sound on

9

SHORELINE ARMORING IMPACTS IN PUGET SOUND
time scales of immediate to years or decades, depending on the location in Puget Sound
(Dethier et al., 2016).

Shoreline armoring alters biological function of the nearshore
Shoreline armoring is known to change the overall biological function of the
nearshore ecosystem. The nearshore is highly productive. It serves as habitat for a
diversity of organisms as well as a refuge and rearing ground for numerous fish species
(Sobosinski, 2003). Physical and chemical processes (e.g., wind and wave energy,
sediment grain size, salinity, tide height) drive biological structure and function in the
intertidal zone. The physical disruption of nearshore ecosystems due to shoreline
armoring can lead to altered biological response.
Shoreline armoring disrupts marine-terrestrial connectivity, alters habitat for midlevel consumers, and ultimately affects prey availability for juvenile salmon (Heerhartz et
al., 2015; Toft et al., 2013). The ecology of the intertidal zone is driven by a connection
between terrestrial and marine processes. Terrestrial ecosystems provide terrestrial leaf
litter input, deposition of large wood, and export of organisms to the beach (Sobosinski,
2010). “Reciprocal subsidization” occurs between the terrestrial-aquatic ecosystems,
where terrestrial plant matter and insects fall into the sea, and marine wrack and
invertebrates are deposited onto the land (Heerhartz et al., 2015). Wrack, or the amount
of seaweeds, seagrasses, and terrestrial plant debris that washes up on shore, provides
nutrients and habitat for terrestrial invertebrates (Heerhartz et al., 2015). The marineterrestrial connection plays an essential role for mid-level consumers which are important
for the diets and early growth rates of juvenile salmon (Rice, 2007).
10

SHORELINE ARMORING IMPACTS IN PUGET SOUND

Shoreline armoring management in Washington State
Understanding the ecological effects of shoreline armoring is important for
guiding policy toward reducing overall armoring and promoting the use of alternative
stabilization techniques that function similarly to natural shorelines. The Puget Sound
region currently has many marine species listed as threatened or endangered, caused in
part by heavy development in the Puget Sound region. Widespread management efforts
around the Puget Sound are focusing on restoring Puget Sound’s health, with a focus on
threatened and endangered species (Guttman, 2009). The listing of Puget Sound
salmonids under the Endangered Species Act has prompted increased attention by
managers and policy makers to focus on the impacts of shoreline armoring on natural
processes that shape Puget Sound. Particular interest has been paid to the functions of
beaches, especially the role of beaches in supporting organisms that occupy important
niches in the food web, such as invertebrates and forage fish (Guttman, 2009).
Several planning and policy documents are designed to protect and restore Puget
Sound, such as Shoreline Master Plan updates and Critical Areas Ordinances. These
documents cite protecting and restoring nearshore habitat functions as an important goal
in overall Puget Sound restoration efforts. Shoreline Master Programs (SMPs) guide
shoreline development in local municipalities, including guidelines for shoreline
armoring. SMPs are guided by state laws but tailored to the specific geographic,
economic, and environmental needs of each community (Ecology, 2016). Local SMPs are
currently being updated to include in-depth guidance on how to implement alternatives to
bulkheads. Alternatives to bulkheading can include “soft” armoring techniques, including

11

SHORELINE ARMORING IMPACTS IN PUGET SOUND
a variety of stabilization methods that mimic site-specific shoreline processes (Van
Zwalenburg, 2016).
In addition, to build, modify, or remove armoring structures, residential and
commercial contractors must obtain a Hydraulic Permit Approval through the
Department of Fish and Wildlife (WDFW, 2016). This permit requires “no net loss” of
ecological functions, and requires the permittee to address mitigative measures to reduce
adverse impacts of the project (WDFW, 2016). Potential mitigation projects can include
enhancing backshore vegetation, addition of large woody debris, and beach nourishment
(Johannessen et al., 2014). Management guidelines aim to protect nearshore function, but
the impacts of armor and mitigation techniques are site-specific. Providing local
restoration of degraded processes, habitat, and ecological function helps maintain health
of nearshore ecosystems. Thoroughly understanding local nearshore ecosystem processes
and impacts of armoring can help maximize mitigation opportunities to provide the
greatest benefits to nearshore systems (Johannessen et al., 2014).
The elevation at which armoring is placed on the beach can influence the scale of
physical impact on nearshore ecosystems (Dethier et al., 2016). Lower elevations of
shoreline armoring, or relative encroachment on the beach, have greater impacts to
biological conditions on local and larger spatial scales (Dethier et al., 2016). Armoring at
low elevations (threshold is approximately 1-2 vertical feet below Mean Higher High
Water (MHHW)) is no longer authorized for shoreline development. However, armoring
at lower elevations than 1-2 feet below MHHW is still present in the Puget Sound. Future

12

SHORELINE ARMORING IMPACTS IN PUGET SOUND
shoreline restoration projects may have greater benefits if projects target shoreline
armoring below this elevation threshold.
Although there has been an increased emphasis to address the impacts of
shoreline armoring, more localized studies are needed to protect and understand critical
habitat in Puget Sound. Continuing research on both localized and Puget Sound-wide
scales will elucidate how shoreline armoring influences both physical and biological
effects of nearshore ecosystems.

Effects of shoreline armoring on key nearshore habitat features
Shoreline armoring is known to change the overall biological function of the
nearshore ecosystem. The following sections will highlight literature documenting the
effects of shoreline armoring on key habitat features including: marine riparian
vegetation, beach wrack, beach wrack associated species, and invertebrate abundances
and composition. In addition, I will look at the effects of shoreline armoring on forage
fish beach spawning. I will then demonstrate how changes in the nearshore due to
armoring physically affects salmon, including their diets and feeding behavior.

Marine riparian vegetation
Shoreline armoring decreases backshore marine riparian vegetation
Riparian vegetation along marine shorelines serves a variety of critical ecological
functions (Brennan, 2007). Coastal trees and other vegetation on backshore areas, banks,
and bluffs help stabilize the soil, control pollution entering marine waters, provide fish
and wildlife habitat. Riparian areas are transitional, providing connections between and

13

SHORELINE ARMORING IMPACTS IN PUGET SOUND
affecting both adjacent aquatic and terrestrial systems. Marine riparian vegetation
communities influence the health and integrity of marine habitats and species and are an
integral part of nearshore ecosystems (Brennan, 2007).
Extensive shoreline development in the Puget Sound region has caused marine
riparian habitat loss. Vegetation characterization is highly variable across the Puget
Sound terrestrial shoreline, however, trees, shrubs, and other ground cover is more
common at natural sites absent of shoreline armoring. The removal of vegetation is
characteristic of armored shorelines in the Puget Sound. In the south-central Puget Sound,
for example, trees make up 80% of the percent cover of marine riparian vegetation at
natural beaches but only 46% percent cover of armored areas, where grass is more
common (Heerhartz et al., 2014). In the central Puget Sound, natural beaches have over
ten times more overhanging vegetation compared to armored beaches (Heerhartz et al.,
2014). Gardens and lawn are more characteristic of armored sites.
Vegetation along shorelines can serve as a metric for habitat quality and a
determinant of available prey resources for juvenile salmon. Over and understory
vegetation is found to host a variety of insect species (Romanuk and Levings, 2003).
Vegetation along the shoreline is vital habitat for terrestrial invertebrates commonly
found in juvenile salmon diets. The lack of shoreline vegetation, including overhanging
trees and shrubs, can affect the abundance and species of invertebrates found along the
shoreline.
Vegetation along the shoreline provides habitat for terrestrial insects, such as
Dipterans (flies), an important dietary component of juvenile Chinook salmon (Munsch et

14

SHORELINE ARMORING IMPACTS IN PUGET SOUND
al., 2016). Terrestrial insects can be carried by wind onto the water surface and provide
food for juvenile salmon in shallow nearshore waters. In addition, leaves, insects, and
other material from overhanging terrestrial plants fall onto the backshore, forming the
basis for multiple terrestrial and aquatic food webs (Guttman, 2009). Armoring can
disrupt this pathway by the associated removal of backshore vegetation and causing a
physical barrier between terrestrial and marine ecosystems (Toft et al., 2007). The
reduction in marine riparian vegetation caused by shoreline armoring decreases available
habitat for invertebrates, and in turn, may have cascading effects on juvenile salmon diets
(Duffy et al., 2010).
Recognition of vegetation as a key function of the nearshore ecosystem is
essential for effective shoreline management. Protecting and restoring backshore
vegetation should be considered an important goal in overall Puget Sound restoration
efforts. Localized studies characterizing naturally occurring vegetation compared to
armored shorelines may increase the understanding of shoreline conditions and assist in
decision making to preserve natural function in these areas.
Olympic Sculpture Park case study and marine riparian restoration
In 2007, the City of Seattle funded the restoration of an armored beach located at
the shoreline of the Olympic Sculpture Park on Elliott Bay in the Puget Sound. This
project is of great interest in the Puget Sound region as an example of habitat
enhancement along urban shorelines (Toft et al., 2013). This project involves an
extensive monitoring plan that is meant to inform future restoration projects in the Puget
Sound.
15

SHORELINE ARMORING IMPACTS IN PUGET SOUND
The overall goal of this project is to support higher numbers of salmon
populations and increase diversity of invertebrate assemblages as prey species for
juvenile salmon. To emulate a natural shallow water habitat, a portion of a seawall was
removed and replaced by a “pocket beach,” and a habitat bench was placed in front of an
existing seawall (Toft et al., 2013). The pocket beach replaced riprap armoring, and the
habitat bench was added as shallow, low-gradient habitat at the base of an adjacent
seawall. Estuarine vegetation, comprised of native plants, was planted above the pocket
beach in the uplands. Biological monitoring was conducted before, during, and after this
enhancement project (pre-enhancement, year 1, and year 3, respectively).
Riparian vegetation was planted in the adjacent supratidal uplands, with a focus
on native species that are common in the Puget Sound coastal zone such as shore pine
(Pinus contorta), alder (Alnus rubra), willows (Salix spp.), beach strawberry (Fragaria
chiloensis), and dune grass (Leymus mollis). Restoring vegetation along the nearshore can
increase habitat complexity and marine-terrestrial connectivity. In response to the
increased plantings along the shoreline, some types of terrestrial insects increased in
abundance and tax richness, which is further discussed in a subsequent section on
terrestrial invertebrates. Other studies have shown insects to be significantly reduced on
armored shorelines where vegetation was removed as well (Romanuk & Levings, 2003;
Sobocinski et al., 2010). Continued development of vegetation communities along
shorelines may increase the input of insects and feeding opportunities for juvenile salmon
(Toft et al., 2013).

16

SHORELINE ARMORING IMPACTS IN PUGET SOUND

Beach wrack
Shoreline armoring and marine-terrestrial connectedness
Wrack, or seaweeds, seagrasses, and terrestrial plant debris deposited on the
beach by an ebbing tide, is habitat for much of the supratidal fauna, and it serves as a
basis for the nearshore detritus-based food web (Heerhartz et al., 2015). Beach wrack can
be comprised of marine (e.g., Ulva spp. and Zostera spp.) or terrestrial (e.g., leaf litter
and wood) sources (Sobosinski, 2003). Wrack functions as a microhabitat by providing
shelter, food, and moisture necessary for many intertidal invertebrates, especially
amphipods, isopods, and insects (Jedrzejczak, 2002a). These organisms are important in
the biogeochemical cycling of marine material (Jedrzejczak, 2002b), and are prominent
consumers in the detritus-based food web (Sobosinski, 2003). The presence of a wrack is
especially important for habitat for invertebrates in areas of low primary productivity,
such as sandy or gravel beaches in Puget Sound (Heerhartz et al., 2016).
The physical disturbance caused by shoreline armoring can reduce the abundance
and composition of wrack that accumulates on Puget Sound shorelines (Heerhartz et al.,
2016). Changes in physical processes due to armoring causes the loss of high shore space,
and therefore the amount of wrack that can accumulate on a beach (Heerhartz et al.,
2015). Overhanging vegetation deposits terrestrial material to the beach, including leaf
litter, sticks, and logs. Local, backshore vegetation is the primary source of terrestrial
detritus in wrack (Heerhartz et al., 2014). Terrestrial detritus in wrack, along with marine
algae, provide food and shelter to diverse communities of invertebrates. With reduced
terrestrial organic debris, armored shorelines lack the resource base to support a leaf-litter

17

SHORELINE ARMORING IMPACTS IN PUGET SOUND
invertebrate community (Heerhartz et al., 2014). Changes in composition and lack of
wrack accumulation lowers taxa richness of insects and benthic macroinvertebrates
associated with wrack, altering prey availability for foraging juvenile salmon (Sobocinski
et al., 2010).
Shoreline armoring reduces beach wrack subsidies
Shoreline armoring can reduce the amount of high shore space on beaches, and in
turn, the amount of wrack and logs that can accumulate on a beach. Reduced wrack
results in significantly different, less taxa-rich and less abundant invertebrate
communities (Heerhartz et al., 2015). Heerhartz et al. (2014) investigated the amount and
composition of wrack and log accumulation across paired armored-unarmored beaches
throughout central and south Puget Sound. They looked at the physical factors that
accounted for these differences, such as beach width, elevation, slope, armoring type, and
uplands vegetation. The width of the armored beaches was significantly reduced by an
average of 8.9 meters, and the elevation of the beach toe was lowered by an average of
0.9 meters (Heerhartz et al., 2014).
They found there was a significant difference between natural and armored
beaches, where there was 66% more total wrack cover in the spring and 76% more in the
fall at natural beaches when compared to armored shorelines. The seasonal variations can
be due to an increase in terrestrial inputs in the fall, such as leaves and sticks, from
upland vegetation. Variations can also be due to the increase of marine algae in the
summer that is susceptible to dislodgement during fall storms (Heerhartz et al., 2014).

18

SHORELINE ARMORING IMPACTS IN PUGET SOUND
Accumulated logs, which provide habitat for many organisms, were almost
completely absent from armored beaches (Heerhartz et al., 2014). By covering the upper
shore with an armoring structure, the space where logs and wrack would normally
accumulate is eliminated (Heerhartz et al., 2014). This has consequences for not only for
primary consumers that depend on the wrack and logs for shelter and food, but also
secondary consumers that are subsidized by resources from these adjacent ecosystems
(Heerhartz et al., 2015).
Shoreline armoring alters wrack composition
Shoreline armoring alters wrack composition, or type of debris deposited on the
beach. Heerhartz et al. (2014), when examining the distribution of wrack on shorelines,
demonstrated that there was a larger proportion of terrestrial material as compared to
seagrass and algal material at natural beaches. The terrestrial component was three to
seven times as abundant on unarmored beaches compared to armored beaches, depending
on the season and beach location (Figure 1). In comparison, the algal proportions were
much higher at armored beaches, where there was on average 74 percent algae at armored
beaches versus 56 percent on natural beaches, demonstrating that there were less
terrestrial inputs on armored beaches (Heerhartz et al., 2014). The clear reduction in the
proportion of terrestrial material in the wrack at armored beaches demonstrates how
shoreline armoring decreases marine-terrestrial connectivity (Heerhartz et al., 2014).
Different types of prey species that associate with either marine or terrestrial wrack
inputs can be affected by the altered composition and abundance of wrack accumulation
at armored sites.
19

SHORELINE ARMORING IMPACTS IN PUGET SOUND

Aquatic and terrestrial invertebrates
Shoreline armoring results in less diverse communities of invertebrates available
for juvenile salmon. Juvenile salmon (Juvenile Chinook and chum salmon) have eclectic
diets, and may benefit from prey from diverse habitats (e.g. terrestrial vegetation, algae,
soft-sediment substrates) (Brennan et al., 2004, Toft et al., 2007; Duffy et al., 2010).
Chinook salmon diet analyses from Puget Sound marine beaches showed a high
proportion of amphipods and insects, specifically Diptera, Homoptera, and Psocoptera,
demonstrating the importance of prey from both marine and terrestrial habitats (Munsch
et al., 2016). Shoreline armoring disrupts the connection between terrestrial and aquatic
ecosystems, a vital function for abundant and diverse invertebrate assemblages.
Shoreline armoring constrains wrack-associated invertebrate communities
Shoreline alterations result in less diverse communities of invertebrates available
for juvenile salmon. Wrack provides food and shelter for diverse communities of
invertebrates, such as talitrid amphipods, isopods, and insects. Wrack invertebrates can
be affected by changes in the physiological requirements of the organisms due to
armoring. For example, armoring changes the sediment moisture and temperature due to
the alterations in wrack cover and composition (Heerhartz et al., 2015). The effects of
altered wrack cover may thus cascade, via altered food webs, to organisms such as fish
(Heerhartz et al., 2015).
Heerhartz et al. (2016) measured the abundance and composition of
macroinvertebrates associated with beach wrack. On average, there were twice as many

20

SHORELINE ARMORING IMPACTS IN PUGET SOUND
terrestrial insects per sample at natural beaches than armored beaches. Insects were
captured at upper shore areas at natural beaches. They also took core samples of wrack to
determine what invertebrates were present. The invertebrate assemblages were found to
be related to the amount and type of wrack found on beaches. Natural beaches had
significantly more talitrid amphipods (sandhoppers), more insects, and fewer aquatic
invertebrates in wrack samples. The talitrid genus, positively correlated with the
proportion of terrestrial wrack, was on average 8.5 times more abundant on natural
beaches than armored (Heerhartz et al., 2016). This result adds strong evidence for the
significant reduction of terrestrial inputs due to shoreline armoring, and how this
reduction can dramatically change invertebrate abundances.
Dethier et al. (2016) looked at broader scale cumulative impacts on invertebrates
in wrack across the entire Puget Sound including 65 pairs or armored and unarmored
beaches across north, central, and south Puget Sound regions. It is difficult to
demonstrate differences attributed to armoring at this scale due to high natural variability
across the Puget Sound. Beach width, riparian vegetation, numbers of accumulated logs,
and amounts and type of beach wrack and associated invertebrates were consistently
lower at armored beaches (Dethier et al., 2016). However, some of results were not
consistent with the localized findings of Heerhartz et al. (2016). Armored beaches
reduced numbers of amphipods and insects only in the central and south regions of the
Puget Sound. When north beaches were included in the analysis, there were no
significant differences in amphipods and insects between armored and natural sites. The
exception was the talitrid amphipod genus, Megalorchestia, that showed a consistent

21

SHORELINE ARMORING IMPACTS IN PUGET SOUND
sensitivity to armoring and where abundances were significantly reduced due to armoring
encroachment on the beach.
The lack of significant difference between wrack-associated invertebrates when
all beaches were compared across Puget Sound is most likely due to regional differences
of shorelines. Impacts to invertebrate abundances in wrack were lower in the northern
regions on Puget Sound where less armoring is present (Dethier et al., 2016). Northern
shorelines also tend to have more habitat space for invertebrates. Northern shorelines
contain overall more wrack due to higher algal populations and mass. In addition, there is
more space for wrack accumulation, as there is lesser encroachment of armoring on the
beach (Dethier et al., 2016).
Shoreline armoring alters epibenthic invertebrate communities - Olympic Sculpture Park
case study
The loss of fine sediment due to shoreline armoring reduces the abundance of
epibenthic invertebrates in shallow water ecosystems (Sobosinski et al., 2010; Toft et al.,
2013), which can reduce epibenthic prey consumption by fish (Morley et al., 2012). Toft
et al. (2013) compared epibenthic invertebrates living at the water-sediment interface
between the pocket beach, habitat bench, and an adjacent armored site. The assemblages
of epibenthic invertebrates became more diverse at the restored sites. Before the site
enhancements, over 93% of amphipod composition consisted of one species
(Paracalliopiella pratti), however, after the enhancement, P. pratti was less dominant
(Toft et al. 2013). Due to the more complex habitat structure created by these

22

SHORELINE ARMORING IMPACTS IN PUGET SOUND
enhancement projects, epibenthic assemblages became more diverse which increases the
diversity of prey available for juvenile salmon (Toft et al. 2013).
Shoreline armoring alters terrestrial invertebrate communities - Olympic Sculpture Park
case study
Toft et al. (2013) looked at differences of terrestrial insects in shoreline vegetation
after shoreline restoration. Taxa richness and abundances of terrestrial insects generally
increased post-enhancement as a result of shoreline plantings and showed higher numbers
of Acari (mites), Collembola (springtails), and aphids which are important prey species
for juvenile salmonids (Toft et al., 2013). Abundances of certain terrestrial insects
associated with marine riparian vegetation can increase with restoration, or planting of
native species, along the shoreline.
Terrestrial invertebrates can be used as a metric for habitat quality and as a
determinant of available prey resources for salmon (Toft et al., 2013). Invertebrate taxa
richness has been found to be greater at sites with intact shoreline vegetation than at
armored sites without (Sobosinski, 2003). Invertebrates may respond to more complex
habitats, as habitat complexity is known to enhance diversity (Chapman, 2003; Morley et
al., 2012).
The enhanced shorelines did not always show definitive improvements over
armored shorelines. For example, some salmon prey items, such as chironomids, a type
of small fly, were abundant at both armored and enhanced sites (Toft et al., 2013). This
could be due to the highly urban and industrialized location of this restoration project as
well as lack of replication on a broader scale (Toft et al., 2013). However, the restorative
23

SHORELINE ARMORING IMPACTS IN PUGET SOUND
plantings along the shoreline did create more complex habitats and increased the marineterrestrial connectivity, both of which can increase inputs of terrestrial invertebrates.

Forage fish
Shoreline armoring and forage fish: an important prey species for salmon
The effects of shoreline armoring may have adverse consequences for forage fish
that spawn in the intertidal zone. Forage fish, such as surf smelt, sand lance, are an
important food source for juvenile and adult Pacific salmon (Rice, 2006). Surf smelt
(Hypomesus pretiosus) and sand lance (Ammodytes hexapterus), utilize areas in the high
intertidal zone for spawning where they deposit their eggs in gravel-sand beaches in the
upper intertidal zone in the Puget Sound. Shoreline armoring may have adverse effects on
egg survival rates due to the reduction of gravel-sand beach habitat and the associated
vegetation loss, causing more exposure to environmental conditions such as increase sun
exposure. Terrestrial vegetation provides shade and increases debris (ex. wrack and logs)
in the upper intertidal zone that protects incubating embryos by providing increased
shade, moisture and protection from sunlight. Removal of terrestrial vegetation that is
associated with shoreline armoring can expose eggs to brighter and hotter conditions,
which are less suitable environments for embryo survival (Rice, 2006).
To look at the influences that shoreline armoring has on forage fish spawning
habitat, Rice (2006) compared the proportion of Surf smelt eggs containing live embryos
at modified and unaltered beaches in Puget Sound, monitoring the light intensity,
substrate and air temperature, and humidity at each shoreline type. The most noteworthy

24

SHORELINE ARMORING IMPACTS IN PUGET SOUND
temperature differences were the substrate temperatures between sites. Natural beaches,
associated with terrestrial vegetation and debris inputs, had a mean temperature of 14.1
degrees Celsius, whereas modified beaches had a mean daily temperature of 18.8 degrees
Celsius. There was a striking difference between the proportion of smelt eggs containing
live embryos between armored and natural beaches. On altered beaches, approximately
half were live as compared to natural beaches (Rice, 2006). Removal of terrestrial
vegetation that is associated with shoreline armoring can expose eggs to brighter and
hotter conditions, which are less suitable environments for embryo survival. Although
this study does document significant differences in environmental conditions between
modified and natural beaches and suggests these differences affect surf smelt embryos,
more detailed information on the specific environmental tolerances of smelt embryos are
needed to fully understand the effects of shoreline armoring on surf smelt embryo
survival (Rice, 2006).
Forage Fish - Olympic Sculpture Park Case Study
The Olympic Sculpture Park enhancement project increased the abundance of
forage fish that utilized the shoreline at the habitat bench and shallow pocket beach (Toft
et al., 2013). The small, pelagic schooling fish may have sought refuge from deeper
waters to avoid predation that is more common in deeper waters or the use of beach
sediments for spawning. Shoreline engineering, such as beach nourishment and
alternative stabilization techniques, may be important for the creation of spawning habitat
and egg survival (Rice, 2006). The enhancement of gravel-sand beaches in this case study

25

SHORELINE ARMORING IMPACTS IN PUGET SOUND
proved important in preserving forage fish spawning habitat and can be used as an
example for future restoration projects.

Fish assemblages
Juvenile salmon in Puget Sound
Puget Sound provides critical rearing habitat for juvenile salmon on their outmigration toward sea. In Puget Sound, several species of Pacific salmon (Oncorhynchus
spp.) rear in nearshore marine areas, including Chinook salmon (Oncorhynchus
tshawytscha) which are of particular concern as they are listed on the federal endangered
species list. Pacific salmon are anadromous species that enter the estuarine or marine
environments as juveniles and have a strong tendency to stay in shallow waters, which
they use for feeding, refuge from predators, and salinity acclimation (Simenstad et al.,
1982). The Puget Sound nearshore is known to be critical for early growth rates of
juvenile salmon as it provides diverse sets of pelagic, benthic, and terrestrial prey
resources (Simenstad et al., 1982). Abundance and quality of prey affect early marine
growth which is critical survival later in their marine life (Duffy et al., 2003).
Shoreline armoring can change the structure of nearshore habitats, reducing the
amount of shallow water habitat available for juvenile salmon (Munsch et al., 2016). The
loss of fine sediment reduces the abundance of epibenthic invertebrates in shallow water
ecosystems (Sobosinski et al., 2010; Toft et al., 2013), which can reduce epibenthic prey
consumption by fish (Morley et al., 2012). Armoring that displaces backshore vegetation
can reduce environmental diversity (Sobosinski et al., 2010) and fish consumption of

26

SHORELINE ARMORING IMPACTS IN PUGET SOUND
terrestrial invertebrates (Toft et al., 2007). This reduction in availability and diversity of
prey resources available for salmon is likely to be the most detrimental effect of armoring
for foraging salmon species (Heerhartz & Toft, 2015).
Shorelines armoring changes juvenile salmon diets in the Puget Sound
Brennen et al. (2004) found that much of the marine mortality for Chinook
salmon is determined by local conditions in the Puget Sound during their first spring and
early summer. Declines in Chinook salmon marine survival since the 1980s may have
been caused by reductions in the quality of feeding and growing conditions during their
early life in the Puget Sound (Duffy et al., 2003). Terrestrial, shallow benthic, and pelagic
habitats are the most important prey production and foraging areas for juvenile Chinook
salmon in shallow marine areas of the Puget Sound. Insects specifically characteristic of
terrestrial vegetated habitats, especially Hymenoptera, Homoptera, and Psocoptera,
dominated the numerical composition of juvenile Chinook diets (Brennen et al., 2004).
Most of the insects in the diets were fully developed winged adult forms, suggesting that
they were likely wind-blown or fell from overhanging vegetation (Brennen et al., 2004).
Benthic and planktonic invertebrates are also important in juvenile Chinook diets. Weight
composition in Chinook salmon diets was similar between benthic, planktonic, and
terrestrial prey categories. Dietary studies on Chinook, coho, and chum salmon from
Hood Canal, and Commencement Bay, Duwamish Head, Skagit Bay, and Shilshole Bay,
in Puget Sound, indicates that terrestrial insects and intertidal amphipods are the largest
components of fish diets throughout Puget Sound. A fish diet analysis from Chinook

27

SHORELINE ARMORING IMPACTS IN PUGET SOUND
salmon demonstrated a high proportion of terrestrial insects in this study as well,
specifically Diptera, Homoptera, and Psocoptera (Rice, 2003).
There are been few studies that focused on salmon diets specifically between
armored and unarmored beaches. Munsch et al. (2015) looked at diets from three species
of juvenile salmon (Chinook, chum, and pink) to look at the differences in prey
availability and feeding patterns among juvenile salmon between armored and unarmored
beaches near the Duwamish River and restored pocket beaches along the urban shoreline
along Elliott Bay. Shoreline armoring affected the composition of prey available in the
environment for specifically Chum salmon along shorelines. Prey selectivity and diet
composition of Chum salmon were different between armored and unarmored sites,
although there was no difference in the diets or stomach fullness of other salmon groups.
Armored sites influenced the diet composition of juvenile chum salmon that select
for epibenthic prey. As mentioned previously, epibenthic invertebrates living at the
water-sediment interface are less diverse at armored sites where habitat structure is less
complex (Toft et al., 2013). Other types of salmon which feed on plankton and
invertebrates along the surface of the water were not affected at either site (Toft et al.,
2013). At beaches, juvenile chum selected for epibenthic copepods, invertebrates living at
the water-sediment interface. However, at seawall sites they selected for planktonic
copepods (invertebrates that drift in deeper waters) (Munsch et al. 2015). This may be
due to the loss of shallow, fine sediment beaches that reduces the availability of
epibenthic prey (Munsch et al. 2015).

28

SHORELINE ARMORING IMPACTS IN PUGET SOUND
It is difficult to assess the effects of shoreline armoring on the diets of juvenile
salmon due to fish mobility which could have an influence results of diet studies.
However, shoreline armoring clearly affects prey resources and can change the feeding
ecology of fish along developed waterfronts. Armoring can change the type of prey
available, such as terrestrial insects and insects that utilize specific shallow water
substrates that are altered due to shoreline armoring.
Shoreline armoring changes the distribution of juvenile salmon related to prey
availability
Smaller fish less abundant along deep shorelines created by intertidal armoring
Shoreline armoring may influence juvenile salmon distribution and feeding
behavior along shorelines. Juvenile salmon prefer unarmored sites that provide estuarine
ecological functions, including shallow water protection and an increased diversity and
abundance of prey species (Heerhartz & Toft, 2015). Heerhartz and Toft (2015)
documented individual-level movement patterns and feeding behavior of juvenile salmon
in shallow water along armored and unarmored shorelines. Snorkel surveys were
conducted at an armored beach, a natural reference beach, and a “restored” beach with
enhanced natural habitat features (at the Olympic Sculpture Park) located in the heavily
armored shoreline in Elliott Bay during peak salmon outmigration periods, April-August
(Heerhartz & Toft, 2015). Juvenile salmon had relatively high feeding rates along both
armored and unarmored sites. However, the most important distinction between armored
and unarmored shorelines for juvenile salmon may be the amount and type of prey
available (Heerhartz & Toft, 2015).
29

SHORELINE ARMORING IMPACTS IN PUGET SOUND
Movement patterns and straightness index values were more diverse at natural
beaches compared to armored shorelines (Heerhartz & Toft, 2015). Juvenile salmon
move in complex paths when feeding, as they change swimming directions and dart to
the surface of the water when attempting to capture prey, causing the fish to diverge from
linear paths (Heerhartz & Toft, 2015). This finding suggests that fish at natural sites have
increased feeding opportunities, as fish exhibited more diversity of swimming speeds and
encompass a broader range of path straightness than fish at armored shorelines (Heerhartz
& Toft, 2015).
Salmon demonstrated more diverse feeding behavior at natural, vegetated
locations than armored sites where food sources are more limited (Toft et al. 2007).
Juvenile salmon were observed darting to the surface to capture insects at more natural
sites (Toft et al. 2007). Unarmored beaches also allow for more complex habitats and
wider shallow intertidal zones which may enable fish to swim with greater path tortuosity
while foraging while remaining in shallow water away from predators (Heerhartz & Toft,
2015).
Due to the highly mobile nature of fish, and their use of large stretches of
shoreline, distinguishing population response to armoring is difficult (Dethier et al.,
2016). Few studies have observed differences in fish response to armored shorelines
using snorkel survey methods (Toft et al., 2013; Heerhartz & Toft, 2015). More studies
using snorkel methods along Puget Sound shorelines may be beneficial to capture fish
behavior along the shorelines.

30

SHORELINE ARMORING IMPACTS IN PUGET SOUND
Distribution of juvenile salmon - Olympic Sculpture Park case study
Toft et al. (2013) measured the effects of shoreline armoring on distributions of
juvenile salmon in the light of available prey resources for salmon. Snorkel surveys were
conducted in peak outmigration, April to July, before and after the site enhancement at
the pocket beach, habitat bench, and adjacent riprap and seawall site for comparison.
Overall, the feeding frequencies of juvenile Chinook salmon increased at the habitat
bench and pocket beach dependent on the year after the site-enhancements (Table 1)
(Toft et al., 2013). After the shoreline enhancement, the feeding frequencies of Chinook
and chum salmon significantly increased after shoreline enhancement compared to the
pre-enhanced period (Toft et al., 2013). Feeding frequencies were characterized by rapid
forays to the surface to feed on neustonic prey, or terrestrial or marine organisms that
float or drift near the surface of the water, and some feeding in the middle of the water
column (Toft et al., 2013). The distribution of juvenile salmon changed in response to the
habitat enhancement locations partly due to increased feeding opportunities (Toft et al.,
2013). The more natural, enhanced beach is important for providing habitat that fosters
increased diversity and abundance of prey species for juvenile salmon.
Although the percentages of salmon feeding generally increased at the enhanced
sites over time, the number of salmon feeding between armored and unarmored sites did
not always increase. For example, in year three, the number of chum salmon feeding at
the seawall site were significantly higher (1525) at the seawall site than at the habitat
bench (504) and pocket beach (163) (Toft et al., 2013). Overall, the enhancements
showed improvements in salmon distribution and prey abundance as compared to heavily

31

SHORELINE ARMORING IMPACTS IN PUGET SOUND
armored shorelines. Most importantly, shoreline restoration, especially in urban areas,
may restore biological functions of the shoreline for fish.

Research needs
The effects of shoreline armoring on salmon should be fully understood when
making management decision regarding the use of armoring structures on Puget Sound
shorelines. This study provides more evidence of the effects of shoreline armoring on
Vashon and Maury Islands, adding to the diversity of studies of armoring across the
Puget Sound, focusing on less developed, residential properties on Vashon and Maury
Islands.
There is widespread recognition by policy and management of potential adverse
biological effects of shoreline armoring in the Puget Sound, however, there is still little
empirical evidence documenting the effects, especially along the diverse regions of the
Puget Sound. Many studies have been focused in highly urban areas, such as the Olympic
Sculpture Park case study in Elliott Bay (near Seattle, WA). Less studies have focus in
residential areas, where shoreline armoring currently makes up most of new armoring
construction projects (Shipman, 2016). Puget Sound’s shorelines are highly diverse, and
it may be important to understand armoring impacts at specific regions across the Puget
Sound. This study adds to the diversity of studies of armoring across the Puget Sound,
focusing on on less developed, residential properties on Vashon and Maury Islands.
Literature regarding the biological effects of armoring is emerging, where
armoring has been shown to negatively impact healthy nearshore ecosystems, including
the estuary functions for juvenile salmon. Puget Sound ecosystems are regionally diverse,
32

SHORELINE ARMORING IMPACTS IN PUGET SOUND
and subject to large differences in physical and biological conditions across space and
time. Localized studies examining the biological response to armoring could benefit
shoreline management policies and local decision making. This thesis is aimed to address
the local impacts of shoreline armoring on key habitat features along shorelines
specifically on Vashon and Maury Islands.
The results from this study are intended to increase our understanding of the
physical and biological differences specifically between armored and natural shorelines
in the MIAR. An important restoration goal is to improve habitat for juvenile salmon that
utilize shorelines in the MIAR. This study establishes site conditions before shoreline
armoring removal, scheduled to take place during the summer of 2018. The suite of
environmental data analyzed in this study can be used as a metric for healthy shoreline
conditions, marine-terrestrial connectivity, and juvenile salmon prey availability at
beaches in the MIAR.

Chapter 3: Methods
Introduction
This thesis is a pre-restoration monitoring study that provides a baseline of
shoreline conditions and how shoreline armoring affects the nearshore in the Maury
Island Aquatic Reserve (MIAR). King County purchased three properties for purposes of
bulkhead removal and environmental restoration at Big Beach, Lost Lake, and Piner
Point. Existing structures and shoreline armoring on each property will be removed
during August-September of 2018, which is after this thesis research was conducted. The
33

SHORELINE ARMORING IMPACTS IN PUGET SOUND
natural shoreline and hillslope processes will be restored to the maximum extent practical
(Booth & Legg, 2017). An important restoration goal and project funding for King
County is to improve habitat for out-migrating juvenile salmon, as shoreline armoring is
known to impact key functions of shoreline habitat for juvenile salmon.
Monitoring for terrestrial invertebrates, shoreline vegetation, beach wrack, forage
fish spawning, and fish use occurred during the summer of 2017. At each site, monitoring
occurred at three treatments: “pre-restoration” (targeted for bulkhead removal),
“armored” (existing bulkhead that will not be removed), and “natural” (no bulkhead)
treatment. The site selection for this project was determined by the Vashon Nature
Center, King County, and the Washington State Department of Natural Resources. This
project used standard field protocols adapted from the Washington Sea Grant’s Nearshore
Monitoring Toolbox, a collection of simple, standardized monitoring protocols that can
be used to evaluate the impacts of shoreline armoring across the Puget Sound (Shoreline
Monitoring Toolbox, 2017).
Permission to access private property was obtained by the Vashon Nature Center.
The sampling was overseen by Vashon Nature Center staff and conducted by trained
citizen-science volunteers from the MIAR stewardship committee.

Site description
The MIAR is located on the eastern shores of Maury Island in central Puget
Sound in the southwestern portion of King County. The reserve is approximately 5,530
acres of state-owned aquatic bedlands and tidelands located in Quartermaster Harbor
(Perla & Metler, 2016). There are three sites in this study including Big Beach, Lost
34

SHORELINE ARMORING IMPACTS IN PUGET SOUND
Lake, and Piner Point. Big Beach and Lost Lake are in Quartermaster Harbor, between
Vashon and Maury Islands (Appendix A). Piner Point is more exposed, as it is located
just outside of the harbor on the southern tip of Maury Island.

Figure 1: Vicinity map of Vashon and Maury Island showing shoreline armoring removal
sites (Booth & Legg, 2017).

Puget Sound is a deep, well-mixed basin with moderately high energy, making for
a unique estuary (Sobocinski, 2003). The estuarine beaches lack severe exposure, as seen
on Washington’s outer coast, but are still subject to physical processes such as wind,
waves, current, longshore current, and swell, not typical of more enclosed estuarine
systems (Nordstrom, 1992).

35

SHORELINE ARMORING IMPACTS IN PUGET SOUND
The climate within the MIAR is influenced by the maritime pacific climate, which
dictates the entire Puget Sound region. Annual temperatures remain mild, while
precipitation levels vary greatly by location and season. Within the reserve, average
rainfall measures differ by 10” from the western to the eastern extent of the reserve.
According to King County records, the western region of the reserve receives an average
of 46” annually, while the eastern edge at Pt. Robinson receives only 36” annually (Perla
& Metler, 2016).
Tides within MIAR are large, with ranges between 3 and 4 meters (Perla &
Metler, 2016). The tides are forced by the tidal variation of sea level at the mouth of the
Salish Sea–the seaward end of the Strait of Juan de Fuca. The tides bring in about 8 km3
of water each high tide, removing that water roughly 12.4 hours later (Perla & Metler,
2016).
The sites included in this study are mainly comprised of the beach type described
as open estuarine intertidal habitats (Dethier, 1990). The primary sediment composition
on these sites was a mix of sand and gravel derived from glacial and interglacial deposits,
delivered to beaches via episodic bluff erosions, and distributed by longshore transport
(Shipman, 2010). Wave energy regime and local geology are the primary drivers of beach
sediment character and gradient in the Salish Sea (Dethier, 2016). Paired beach sites were
within the same drift cell, or independent zone of littoral sediment transport from source
to deposition area, and within the same component of that drift cell (erosional or
depositional) (Dethier, 2016). This study is unique because the study sites are backed by
tremendously steep and actively eroding bluffs. Slopes are very unstable at these sites and

36

SHORELINE ARMORING IMPACTS IN PUGET SOUND
there have been recent landslides in the area (Figure 2). Changes are expected to happen
quite fast after bulkhead removals are completed (Perla & Metler, 2016).

Figure 2: Slope stability in the Maury Island Aquatic Reserve. Slopes are unstable at all
three sampling locations. The red indicates unstable slopes where there have been recent
slides, demonstrating that each sampling location has been subjected to recent slides. The
surrounding areas are also unstable, as the brown indicates unstable slopes with old
slides, and the orange indicates unstable slopes in general (Washington State Coastal
Atlas, 2018).

37

SHORELINE ARMORING IMPACTS IN PUGET SOUND

Big Beach
Big Beach (BB) is an eastward-facing, low-grade beach. This site is south of the Lost
Lake site, toward the mouth of Quartermaster Harbor. This is a residential beach,
however, most of the housing is set back from the beach due to the high-bank nature of
this site. The pre-restoration site (-122.49166, 47.34558) is over 50 meters, and is
comprised of a hodge-podge bulkhead made of wood and large boulders, or rip-rap that
will be removed during the restoration. The armored site is located directly adjacent to
the south of the pre-restoration site, and is characterized by a tall, concrete bulkhead. The
natural site, located north of the pre-restoration and armored sites, is undeveloped and
comprised of dense overhanging vegetation. The beaches in this site are approximately in
the southern end of the same drift cell that drifts from south to north (Figure 3).

Lost Lake
Lost Lake (LL) is an eastward-facing, low-grade beach. This is the innermost site in
Quartermaster Harbor. There is a small housing development comprised of a few houses
along the beach, placed directly above the shoreline armoring. The pre-restoration site (122.48857, 47.36060) is characterized by a 30 m wooden bulkhead with a house and a
few shrubs placed directly above the armoring. The bulkhead and house will be removed
during the restoration. The natural site is directly south of the pre-restoration site and has
visible logs, salt grass, and overhanging trees and shrubs. The armored site is one of the
most northern properties in this small housing development along the beach. The armored
site has a short, wooden bulkhead with adjacent trees, shrubs, and no visible house placed

38

SHORELINE ARMORING IMPACTS IN PUGET SOUND
in the uplands. The beaches at this site are in approximately the center of the same drift
cell that drifts from approximately the south to north (Figure 3).

Figure 3: Drift cells in the Maury Island Aquatic Reserve. Arrows point in the direction
of the drift movement. Lost Lake and Big Beach sampling locations are in the same drift
cell that drifts from south toward the north (green line). Piner Point is in a divergence
zone (black line), which is generally subject to more rapid erosion and significant
sediment sources within littoral cells (Washington State Coastal Atlas, 2018).

Piner Point
Piner Point is a southern-facing, low-grade beach. Piner Point is the outermost site in
Quartermaster Harbor, located at the south tip of Maury Island. This is a highly erosive,
high-bank beach. The pre-restoration site (-122.45894, 47.34329) is a 30 meter, failing
39

SHORELINE ARMORING IMPACTS IN PUGET SOUND
wooden bulkhead. The uplands consist of trees and shrubs that have been subjected to
landsliding in the area. The armored site is located toward the west of the pre-restoration
site, and is a hodge-podge of wooden structures, with a house placed almost directly
above the shoreline armoring, with few overhanging trees and shrubs. The natural site is
directly east of the pre-restoration site, with visibly eroding high-bank, accumulated logs,
and some overhanging trees and shrubs that have be subjected to landslides. The beaches
at this location are in a divergence zone, which is subject to more rapid erosion and
significant sediment sources within littoral cells (Shipman, 2008) (Figure 3).

Experimental design
This study addresses how shoreline armoring can impact natural nearshore
ecosystem functions by comparing paired armored and unarmored beaches at three sites.
Each study site has three treatments: a pre-restoration (where the bulkhead will be
removed), a natural (where no bulkhead exists), and an armored (bulkhead). The different
treatment locations were chosen based on county plans to remove the bulkheads so those
sites anchored subsequent location decisions. The transect lengths were consistent within
each study site. Piner Point and Lost Lake transects are short (30 m) and Big Beach are
long (50 m), which were determined by the length of bulkhead being removed. Transects
were placed parallel to the shore on the fresh wrack line, where the most recent debris
was left behind at the previous high tide. When no wrack line was present (i.e. on many
bulkheaded beaches), transects were placed at the previous high tide line or toe of
armoring.

40

SHORELINE ARMORING IMPACTS IN PUGET SOUND
The three sites at each beach location were selected in close proximity. This site
selection minimizes the effects of physical properties including deposition and movement
of organic debris and sediments that are largely driven by local wind, waves, and currents
(Nordstrom, 1992). Treatments were close, and in some cases directly adjacent to each
other, so they had similar aspects (with respect to sun, waves, and weather), bank slope,
and type of sediment. When possible, armored sites with similar bulkhead structures and
materials to the pre-restoration site were chosen (Perla & Metler, 2016). The three sites
are contained within the same drift cell (Figure 3), reducing the variation in physical
characteristics between sites (Sobocinski, 2003). This sampling regime, while beneficial
in terms of minimizing spatial differences, is inherently biased due to variations between
sites and treatments (Sobosinski, 2003).

Sample timing and frequency
Sampling occurred during summer low tides for maximum accessibility and to
ensure all parameters were measured. Big Beach sampling occurred in June, Lost Lake in
July, Piner Point in August 2017. Surveys occurred during different months due to
scheduling and accessibility restrictions. Variables tested do not differ significantly
between these months (Perla, 2018). Sampling was performed over two days at each site
for all beach surveys. For each beach site survey, data was collected on the same day
during low tide.
Snorkel surveys were conducted over a single day at each site during the high
tide. Data collection occurred at Lost Lake and Big Beach in July and Piner Point in
August 2017.
41

SHORELINE ARMORING IMPACTS IN PUGET SOUND
Forage fish spawning data is continually collected by citizen scientists every other
month at these sites, except during June, July, August, when one site is sampled each
month at the same time as the suite of shoreline surveys.

Surveys
Terrestrial Vegetation
Riparian vegetation provides habitat for terrestrial insects that are important prey
resources for juvenile salmon. Characterizing shoreline vegetation can give valuable
information about the habitat of the upper beach, marine-terrestrial connectivity, as well
as habitat availability for insects (Shoreline Monitoring Toolbox, 2017).
This protocol was taken largely from the Shoreline Monitoring Toolbox
(Shoreline Monitoring Toolbox, 2017) with a few modifications (Appendix B).
Specifically, the overhanging riparian cover measurements were modified to better suit
the conditions at these sites.
For each study site, a list was created of plants in the tree, shrub, and groundcover
layers that occur in the length of the 30 or 50 m transects. Every tree that overhangs the
beach and its species were counted and recorded, using the established transect as a
length to sample. The width of the tree canopy that overhangs the beach was estimated.
All the overhanging widths along each transect were totaled and divided by the entire
transect length to get an estimate of percent cover of overhanging vegetation along the
study transect.

42

SHORELINE ARMORING IMPACTS IN PUGET SOUND
The cover of overstory (trees) and understory (shrubs and groundcover)
vegetation was estimated for three different 5x5 m plots placed along the length of the
established transect at 0-5 m, 10-15 m, and 25-30 m (or if a 50 m transect at 0-5 m; 20-25
m, 45-50 m). The established transect was used as a length to sample, however, plots
were placed at the bluff toe or bulkhead and extended back five meters into the terrestrial
landscape. The general health of the plants was estimated for understory and overstory on
a scale of 1-5 (1=barely alive; 2=major damage over more than 50 percent; 3=some
damage to 30 percent; 4=minor damage less than 20 percent; 5=thriving). Reasons for
low health scores were recorded in notes (i.e. drought stress, slide stress, disease). All
vegetation species were categorized by native and non-native species.
Beach wrack
Beach wrack may be an important source of nutrient exchange between marine
and terrestrial systems and provides shelter, food, and moisture for invertebrates.
Dependent variables include the percent cover of rack and composition of marine and
terrestrial organic debris. Examining composition can give information on the source
material (terrestrial vs. marine sources) and the associated amounts that deposit on the
beach at each site type.
Sampling was based on the Shoreline Monitoring Toolbox methods for wrack
sampling (Appendix C). Briefly, two transects were established: One at the most recent
high tide line with fresh wrack deposition, and a second just above MHHW where older
wrack accumulates. These locations were established by visually observing wrack lines
on the beach. The most recent high tide line targets mobile wrack, whereas the higher
43

SHORELINE ARMORING IMPACTS IN PUGET SOUND
elevation sample targets the more stable wrack layer and extent that wrack is mobilized
during storms. Armored beaches tended to lack wrack and log accumulation, and
therefore, only the lower wrack lines were analyzed in this study. Presence of upper
wrack and logs are noted at each site.
A 0.1 m2 quadrat was placed on the beach surface at ten randomly selected points
along the 30 or 50 m transects placed parallel to shore. A visual estimate of the total
percent cover was taken. The percent composition of marine algae, terrestrial plant
material, and eelgrass was recorded at each quadrat, and therefore, analyzed as
independent of each other.
The visual assessment was based on a percentage of the quadrat, divided into 25
6x6 cm small squares, where each square equals 4 percent. Algae type (e.g., red, green,
brown, or other species) was recorded. When there was less than 0.01 percent cover, we
used a standardized low number (.01) to differentiate between small amounts of wrack
cover to nothing at all.
This study only compares the lower wrack line. All sites had a lower wrack line,
however, not all sites had an upper wrack line, creating a difficulty in comparing upper
wrack lines across sites. Therefore, total percent cover number reported in this study may
not be representative of the total percent cover on the actual beach.
Terrestrial insect fallout traps
Terrestrial insects are an indicator of shoreline conditions and are an important
prey for juvenile salmon. Examining changes in insect assemblages due to armoring can

44

SHORELINE ARMORING IMPACTS IN PUGET SOUND
be important, as the population stability may have cascading impacts on higher trophic
levels, including fish (Shoreline Monitoring Toolbox, 2017).
Insect sampling occurred in the summer months (June-August) as juvenile
Chinook salmon are feeding along the shoreline, and vegetation and insect communities
are developed. The dependent variables measured include taxa richness, the number of
different taxa in the sample, and composition, focusing on key salmon prey species.
Sampling was based on the Shoreline Monitoring Toolbox methods for insect
sampling (Appendix D). Briefly, fallout traps were created using plastic storage bins
(34.6 cm x 21 cm) filled with a weak soap-water solution. Five replicate bins were placed
randomly along a 30 or 50m long study transect parallel to shore and left in place for 24
hours. After 24 hours, the contents were passed through a 0.106 mm sieve, and the
material retained was preserved in 70% isopropyl alcohol. Insect samples were processed
in the laboratory for numerical composition with taxonomic resolution to family.
Density was calculated by summing the number of insects found in each bin and
dividing by the surface area of the bin (0.07266 m2), for each of the five bins at each
treatment. The average was then taken across the fives bins for each treatment at each
site. Taxa richness was calculated by counting the number of species that occurred in
each of the five bins.
Forage fish
Surf smelt (Hypomesus pretiosus) and Pacific sand lance (Ammodytes hexapterus)
spawn on the beach, depositing their eggs in the sediments on the upper beach.
Successful forage fish spawning can be an indicator of a healthy beach. Spawning can be
45

SHORELINE ARMORING IMPACTS IN PUGET SOUND
impacted by changes to the nearshore due to shoreline armoring, since specific sediment
sizes and tidal elevations are targeted by these fish. These fish are a vital part of the food
web, being preyed upon by larger fish (e.g., salmon), marine mammals, and birds
(Shoreline Monitoring Toolbox, 2017).
Bulk beach substrate samples were collected by VNC citizen scientists at each
treatment at each site. Standardized forage fish beach spawning data collection methods
established by Washington Department of Fish and Wildlife were used (see Appendix E).
In short, a 30.48 (100 foot) transect tape was placed parallel to the shore at sandy-gravel
substrates. Tidal elevation of the transect is determined by measuring the distance from
the transect to an identified landmark, such as upland toe of the beach, the last high tide
mark, or the water’s edge. Along the established transect tape, bulk substrate samples
were collected by scooping the top 5-10 cm of sediment (about two foot long scoops) at
10 evenly spaced locations.
Substrate samples were wet-screened through a through set of 4 mm, 2 mm, and
0.5 mm sieves using buckets of shore-side water. The material from the 0.5 mm sieve
was placed into a rectangular dishpan with an inch of water, and winnowed into
subsamples of forage fish egg-sized material. Winnowing consists of rotating or tilting
the dishpan of material to cause lighter material to rise to the surface, and in short,
suspend any forage fish eggs to the top of the sediment sample. Egg subsamples were
collected by scooping the top layer of lighter sediment material (and any eggs) into a 16
oz jar. Sub-samples were sent to the Washington Department of Natural Resources

46

SHORELINE ARMORING IMPACTS IN PUGET SOUND
Aquatic Reserve Program’s laboratory to be analyzed for spawning presence/absence and
number of eggs.
Data was collected every other month at each site starting December 2016.
During June, July, August, data was collected during the full beach surveys, and therefore
only one site was sampled each month during this time.
Fish observations
The natural history of shallow-water fish communities can help identify and
account for critical habitat function. This method was based on the Shoreline Monitoring
Toolbox’s methods for fish observation (Appendix F). During the highest tide of the day,
two 50 meter transects were established parallel to the shore. One transect was
established at 1.5 m depth, about 20 meters from the shore. The second transect was
established at approximately 2.0 meters depth and approximately 30 meters from the
shore.
Observers started by measuring underwater visibility. Ideally, surveys should only
occur when visibility exceeds 2.5 m to maximize the accuracy of observations and
minimize effects of observed on fish behavior (Toft et al., 2007). The shallow water
depth was measured at 1.5 m using a weighted line. The second water depth was
measured 10 m away (away from the beach) at the beginning of the second transect using
a weighted line. The second depths varied, but were consistently around 2 m of depth.
Transects ran parallel to the shore.
Observers recorded the following variables for each fish species encountered:
species, a visual estimate of length to the nearest centimeter, number of individual fish,
47

SHORELINE ARMORING IMPACTS IN PUGET SOUND
and water column position of the fish. When fish were not identifiable to the species
level, names of lower taxonomic resolution were used to describe their identity (e.g.
unknown forage fish). Water column positions were described in thirds: top, middle, and
bottom. Feeding behavior (i.e. darting to the surface) was recorded when applicable.
Number of fish and observations were averaged by treatment type (armored, natural, and
pre-restoration). Taxa richness was calculated by averaging the number of species by
treatment type.
Due to proximity, Big Beach and Lost Lake sites were sampled during the same
day in July. Piner Point was sampled a month later in August.

Analysis
For all analyses, the independent variables were treatment type (armored, restoration,
and natural sites). The following variables were compared at armored and unarmored
beaches where data was collapsed across all sites and examined at each site individually
as well.


Vegetation (percent overstory, percent understory, native vs. non-native species)



Wrack (total percent cover, percent marine, percent terrestrial, percent eelgrass)



Insects (taxa richness, density)



Forage fish spawning (spawning events, number of eggs)



Fish (number of fish, taxa richness, number of observations)

All data were organized in Microsoft Excel XP®. Preliminary data exploration was
performed in Excel. JMP Pro 12 was used for subsequent data analysis when sample

48

SHORELINE ARMORING IMPACTS IN PUGET SOUND
sizes were appropriate. Statistical analysis was not performed when sample sizes or
sampling frequency was low and data was non-normally distributed. All data was
checked for violations of normality and skewness, and was transformed if violations were
detected. Numerical data was log transformed. Arcsine transformations were applied
(p’=ASIN(SQRT(p)) to proportional datasets. Transformed data was then reassessed to
ensure no violations persisted. Statistical analysis was conducted on the transformed
dataset. If the dataset still violated assumptions of normality and skewness upon
transformation, non-parametric statistical tests were used.
One-way analysis of variance (ANOVA) was used to analyze differences between
treatments (armored, natural, pre-restoration). Post-hoc tests were used to further
examine differences using the Tukey test. When significant differences were found
between treatments, a 2-way ANOVA was used to test for differences between groups
while accounting for site and treatment (the two independent variables) as independent of
each other. The critical p-value in assigning statistical significance was α=0.05.
Kruskal-Wallace tests were used as the nonparametric equivalent to the ANOVA test.
A Dunn test was used as a nonparametric post-hoc test. P-values were based the corrected
alpha using the Bonferroni correction of (p=0.0167). The Friedman’s test, or randomizedblock design test, was used as a non-parametric equivalent to the mixed-design ANOVA
to account for site and treatment. Post-hoc tests were not calculated for the Friedman’s
test, as methods for this remain relatively uncommon (Wobbrock et al., 2011). The
critical p-value in assigning statistical significance was α=0.05 for these tests.

49

SHORELINE ARMORING IMPACTS IN PUGET SOUND

Chapter 4: Results
The results section is broken up into sections based on the five beach parameters
assessed in this study: terrestrial vegetation, wrack, insects, forage fish, and fish. Each
section presents the findings from the analysis of each parameter, and assesses whether
there are treatment or site effects for each parameter, when applicable. If there were no
significant differences in parameters across treatments, then site interactions were not
included.

Terrestrial vegetation
Treatment effects on overstory percent cover
The percent cover of overstory vegetation was averaged across all sites based on
treatment (armored, natural, pre-restoration). In general, natural sites had greater average
overstory vegetation coverage (83.89 ± 7.94%), compared to the armored (48.33 ±
12.25%) and pre-restoration (40.56 ± 13.86%) sites (Figure 4), an observation that was
significant based on Tukey’s HSD test (ANOVA F(2,24) = 3.86, P = 0.04).

50

SHORELINE ARMORING IMPACTS IN PUGET SOUND

Average percent (%) cover per
treatment

120
100
80
60
40
20
0
Armored

Natural

Armored (2018
Restoration)

Figure 4: Average overstory vegetation percent cover per treatment.

Variations in overstory percent cover based on site
There is an interaction in that site affects overstory percent cover estimates
(ANOVA F(8,18) = 1.49, P = 0.047). However, the natural sites consistently have higher
percent cover. Percent cover of overstory vegetation was highest at natural treatments at
every study site (Figure 5). For example, the natural site at Piner Point had highest
overstory percent cover (96.67 ± 3.33%), as compared to the armored (60 ± 5.77%) and
pre-restoration (60 ± 15.28%). This was also true for Lost Lake and Big Beach.

51

SHORELINE ARMORING IMPACTS IN PUGET SOUND

100
80
60
40
20

Big Beach

Lost Lake

Armored (2018
Restoration)

Natural

Armored

Armored (2018
Restoration)

Natural

Armored

Armored (2018
Restoration)

Natural

0

Armored

Average percent (%) cover per transect

120

Piner Point

Figure 5: Percent cover of overstory vegetation per site.

Variation in understory percent cover by treatment
Understory vegetation was also averaged across all sites based on treatment
(armored, natural, pre-restoration). The average understory percent cover was
consistently high across treatments at all sites (averages ranged between of 73-88%
across natural, armored, and pre-restoration sites) (Figure 6). When averaging across
sites, the pre-restoration site had the highest average amount of understory (88.33 ±
9.98%), although there were no significant differences between treatments (KruskalWallis 𝝌2(2) = 2.30, P = 0.32).

52

SHORELINE ARMORING IMPACTS IN PUGET SOUND
100

Average percent (%) cover per
treatment

90
80
70
60
50
40
30
20
10
0
Armored

Natural

Armored (2018
Restoration)

Figure 6: Percent cover of understory vegetation per treatment.

Variation in understory percent cover by site
Understory vegetation was high at all sites, ranging from approximately 40% to
100% cover across treatments at all the sites (Figure 7). Vegetation completely covered
the understory at Big Beach and Piner Point pre-restoration sites, raising the average
compared to the armored and natural sites. Pre-restoration sites have not been maintained
for years, allowing for vegetation to grow even though there is development at the site.
At Big Beach, understory almost completely covered all sites (armored (90 ±
5.77%), natural (78.33 ± 14.24%), and pre-restoration (100 ± 0.0%)). The overall plant
health at Big Beach was high (4-5). At Lost Lake, the armored and natural sites were
completely covered with understory plants, and the pre-restoration site was about twothirds covered on average (65 ± 23.61%). The overall health rating for plants at Lost
Lake was high (4-5, with some lower health ratings at armored and pre-restoration sites
53

SHORELINE ARMORING IMPACTS IN PUGET SOUND
due to development). At Piner Point, percent cover of understory was highest at the prerestoration site (100 ± 0.0%), as compared to the armored (53.33 ± 24.04%) and natural
(40 ± 5.77%). The Piner Point natural and pre-restoration beaches had lower average
health ratings (3) than other sites, as invasive species and drought and salt stressed plants
were present.

100
80
60
40
20

Big Beach

Lost Lake

Armored (2018
Restoration)

Natural

Armored

Armored (2018
Restoration)

Natural

Armored

Armored (2018
Restoration)

Natural

0

Armored

Average percent (%) cover per
transect

120

Piner Point

Figure 7: Percent cover of understory vegetation per site.

Overhanging trees per treatment
Overhanging trees were averaged across all sites based on treatment (armored,
natural, pre-restoration). Statistical analysis was not performed due to small sample sizes.
On average, natural sites had more overhanging tree cover (23.33 trees), compared to the
armored (3 trees) and pre-restoration sites (3.67 trees) (Figure 8). All tree species were
native at all sites in this study.

54

SHORELINE ARMORING IMPACTS IN PUGET SOUND

Average number (#) overhanging trees
per treatment

35
30
25
20
15
10
5
0
Armored

Natural

Armored (2018
Restoration)

Figure 8: The number of overhanging trees per treatment.

Overhanging trees per site
Overhanging trees were more abundant at natural sites. Statistical analysis was
not performed due to small sample sizes. Piner Point had the highest number of
overhanging trees overall at the natural site (55 trees), compared to armored (4 trees) and
pre-restoration (8 trees) (Figure 9). Tree species at Piner Point were overall more diverse.
The natural site consisted of clumps of alder trees, madrone, and maple. The armored site
consisted of maple and alder trees. Maple, cedar, salix, shorepine, laurel, and a few dead
trees were present at the pre-restoration site.
At Lost Lake, the highest number of overhanging trees were found at the natural
site (3 trees), compared to armored (1 tree) and pre-restoration (1 tree). All trees along
this transect were alder.

55

SHORELINE ARMORING IMPACTS IN PUGET SOUND
The natural site at Big Beach had the highest number of trees (12 trees), which
were all maple trees, compared to armored (4 trees) and pre-restoration (2 trees). Trees at

70
60
50
40
30
20
10

Big Beach

Lost Lake

Armored (2018
Restoration)

Natural

Armored

Armored (2018
Restoration)

Natural

Armored

Armored (2018
Restoration)

-10

Natural

0

Armored

Average number (#) overhanging trees
per transect

the pre-restoration and armored sites were exclusively alder.

Piner Point

Figure 9: Number of overhanging trees per site.

Native vs. non-native species counts per treatment
Native and non-native species counts were averaged across all sites based on
treatment (armored, natural, pre-restoration). Overall, natural species were more
abundant at natural sites. Differences between the number of native and non-native
species was the highest at the natural sites (37 native versus 7 non-native species). The
number of native compared to non-native species at armored and pre-restoration
shoreline types was similar, where there were 31 native compared to 30 non-native
species at the armored shorelines and 20 native and 20 non-native species at the prerestoration shorelines. Overall, the count of non-native species was highest at the pre56

SHORELINE ARMORING IMPACTS IN PUGET SOUND
restoration site, potentially due to the lack of yard maintenance at pre-restoration sites
(Figure 10).

Number (#) of species per treatment

40
35
30
25
20

Native

15

Non-native

10
5
0
Armored

Natural

Armored (2018
Restoration)

Figure 10: Number of native and non-native species per treatment.

Abundance of native vs. non-native species per site
There were no substantial trends in the number of native species compared to
non-native species per transect when comparing treatments each individual site (Figure
13). However, there were a few findings of interest. There was consistency between the
number of native and non-native species across all treatments at Piner Point. There were
15 native species and no non-native species at the Big Beach natural site. The highest
number of non-native species were found at the Lost Lake pre-restoration site (14 nonnative species).

57

SHORELINE ARMORING IMPACTS IN PUGET SOUND
16
12
10
8
6
4

Native

2

Non-native

Big Beach

Lost Lake

Armored (2018
Restoration)

Natural

Armored

Armored (2018
Restoration)

Natural

Armored

Armored (2018
Restoration)

Natural

0

Armored

Number (#) of species per transect

14

Piner Point

Figure 11: Number of native and non-native species at each site.

Wrack
Variation in wrack total percent cover based on treatment
Total wrack cover varied between treatments when averaged across sites
(Kruskal-Wallis 𝝌 = 12.36, P = 0.002) (Figure 12), with more wrack found at the pre2

(2)

restoration sites than the armored sites (Dunn test Z = 3.64, P = 0.0003). The natural
(21.4 ± 5.2%) and pre-restoration sites (25.5 ± 5.25%) had similar average total percent
cover, whereas average total percent cover was lower at the armored sites (7.8 ± 2.28%).
A Friedman’s test demonstrated that there was no interaction, meaning that treatment
(armoring) has an effect on wrack total percent cover regardless of site (𝝌2(2) = 4.67, p =
0.097).

58

SHORELINE ARMORING IMPACTS IN PUGET SOUND

Figure 12: Percent cover of total wrack per treatment.

Site as a basis for variation in wrack total percent cover
There were no consistent trends in total wrack cover when comparing across
treatments at each individual site (Figure 13). At Big Beach, there was a higher average
percent of total wrack cover (40.3 ± 11.1%) at the pre-restoration site, although this
difference was not statistically significant compared to the natural (19.6 ± 9.69%) and
armored (16.6 ± 5.63%) sites (ANOVA F(2, 27) = 1.87, p=0.17).
At Lost Lake, total percent cover varied across treatments Kruskal-Wallis 𝝌 =
2

(2)

13.91, P = 0.001). Total wrack was highest at the natural site (39.4 ± 9.29%), as
compared to the pre-restoration (28.5 ± 8.80%) and the armored (3.8 ± 1.67%) sites. The
armored site differed significantly than the natural (Dunn test Z = 3.16, P = 0.002) and
pre-restoration sites (Dunn test Z = 3.08, P = 0.002).

59

SHORELINE ARMORING IMPACTS IN PUGET SOUND
At Piner Point, total percent cover did not significantly vary across treatments
(ANOVA F

(2, 27)

= 0.83, p = 0.45). Percent cover was highest at the pre-restoration site (7.7

± 2.18%) compared the armored (3.2 ± 1.71%) and natural sites (5.13 ± 3.27%), although
these results are not significant (Figure 13).

40
30
20
10

Big Beach

Lost Lake

Armored (2018
Restoration)

Natural

Armored

Armored (2018
Restoration)

Natural

Armored

Armored (2018
Restoration)

-10

Natural

0

Armored

Average percent (%) cover per transect

50

Piner Point

Figure 13: Percent cover of total wrack per site.

Variation in marine algae in wrack based on treatment
Percent cover of marine wrack varied across sites (Kruskal-Wallis 𝝌2(2) = 17.003,
p=0.0002) (Figure 14). The natural (24.20 ± 4.89%) and pre-restoration sites (25.10 ±
4.80%) had similar average percent cover, which was higher than the amount of marine
wrack at the armored site (7.87 ± 2.28%) (Dunn test Z = 3.19, P = 0.004; Z = 3.88, P =
0.0003, respectively). A Friedman’s test showed that there was no interaction between

60

SHORELINE ARMORING IMPACTS IN PUGET SOUND
site and treatment (𝝌2(2) = 4.67, p = 0.097). Ulvoid algae dominated the percent cover of
wrack samples.

Figure 14: Percent cover of marine wrack per treatment.

Site as a basis for variation in marine wrack cover
There were no consistent trends in marine wrack cover when assessing the effects
of site (Figure 15). The percent cover of marine wrack mirrors results of total wrack at
each site and treatment, as total percent cover is dominated by marine components. In
contrast to the total cover results, marine wrack cover was highest at the natural site (5.1
± 3.26%) at Piner Point, although this result was not statistically significant (ANOVA F(2,
27)

= 1.03, p = 0.37)

61

SHORELINE ARMORING IMPACTS IN PUGET SOUND
45
35
30
25
20
15
10
5

Big Beach

Lost Lake

Armored (2018
Restoration)

Natural

Armored

Armored (2018
Restoration)

Natural

Armored

Armored (2018
Restoration)

Natural

0
-5

Armored

Average percent (%) cover per transect

40

Piner Point

Figure 15: Percent cover of marine wrack per site.

Variation in terrestrial wrack percent cover based on treatment
When averaging across sites, the averaged terrestrial wrack percent cover varied
across treatments (Kruskal-Wallis 𝝌2(2) = 17.04, p= 0.0002) (Figure 16). The natural
(5.60 ± 1.82%) had a higher average percent cover of terrestrial wrack compared to the
armored (2.1 ± 1.99%) and pre-restoration (1.72 ± 0.62%) sites. The armored had
significantly different medians than the pre-restoration site (Dunn test Z = 2.98, P =
0.0081) and the natural (Dunn test Z = 3.86, P = 0.0003). A Friedman’s test showed no
interaction between site and treatment (𝝌2(2) = 0.63, p = 0.73).

62

SHORELINE ARMORING IMPACTS IN PUGET SOUND

Figure 16: Percent cover of terrestrial wrack per treatment.

Site as a basis for variation in terrestrial wrack cover
Terrestrial wrack was higher at both Lost Lake and Piner Point sites (Figure 17).
At Lost Lake, the terrestrial wrack was significantly different between natural (10.1 ±
4.5%) and armored (0.1 ± 0.1%) sites (Dunn test Z = 3.19, P = 0.004). At Piner Point,
natural (6.2 ± 2.58%) treatments were significantly different than armored (0.2 ± 0.2%)
(Dunn test Z = 3.08, P = 0.006). At Big Beach, there was a higher average of terrestrial
wrack found at the armored site (6.00 ± 6.00%) as compared to the natural (0.5 ± 0.34%)
and pre-restoration (0.9 ± 0.41%) sites, although there were no significant differences
between treatments (Kruskal- Wallis 𝝌2(2) = 2.24, P = 0.33).

63

SHORELINE ARMORING IMPACTS IN PUGET SOUND

10
8
6
4
2

Big Beach

Lost Lake

Armored (2018
Restoration)

Natural

Armored

Armored (2018
Restoration)

Natural

Armored

Armored (2018
Restoration)

-2

Natural

0

Armored

Average percent (%) cover per transect

12

Piner Point

Figure 17: Percent cover terrestrial wrack per site.

Variations in eelgrass percent cover in wrack based on treatment
Eelgrass percent cover differed significantly across the three sites (Kruskal-Wallis
𝝌2 (2) = 35.09, p < 0.0001). Average eelgrass percent cover was highest at natural sites
(3.81 ± 0.85%), compared to the pre-restoration (1.65 ± 0.45%) and armored sites (0.01 ±
0.00%). Eelgrass was highest at Lost Lake natural sites (5.81 ± 1.53%) (Figure 18). The
percent cover of eelgrass differed significantly between the natural and armored
treatments (Dunn test Z = 5.18, P < 0.0001). There was also a significant difference
between eelgrass percent cover between the pre-restoration and armored (Dunn test Z =
5.05, P < 0.0001). A Friedman’s test showed that there was no interaction between site
and treatment (Friedman’s test 𝝌2(2) = 4.67, p = 0.097).

64

SHORELINE ARMORING IMPACTS IN PUGET SOUND

Figure 18: Percent cover of eelgrass per treatment.

Site as a basis for variation in eelgrass wrack cover
Overall, percent cover of eelgrass was higher at natural sites at both Lost Lake
and Piner Point. At Lost Lake, eelgrass cover differed significantly at every treatment
(Kruskal-Wallis 𝝌2(2) = 17.89, P = 0.0001). The natural site (5.8 ± 1.53%) had the highest
average eelgrass cover, whereas the armored (0.02 ± 0.01%) pre-restoration (0.62 ±
0.26%) and sites had trace amounts (Figure 19).
At Piner Point, eelgrass cover differed significant differences between treatments
(Kruskal- Wallis 𝝌2(2) = 15.82, P = 0.0004). Percent cover of eelgrass was significantly
higher at the natural (4.8 ± 1.58%) than the armored (0.01 ± 0.01%) site (Dunn test Z =
3.83, P = 0.0004). Eelgrass at the pre-restoration site was also lower than at the natural
site (2.2 ± 1.08), although results were not significant (Figure 19).

65

SHORELINE ARMORING IMPACTS IN PUGET SOUND
At Big Beach, eelgrass cover differed significantly between treatments (KruskalWallis 𝝌2 (2) = 15.84, p=0.0004). The pre-restoration site had a significantly higher
percent cover of eelgrass (2.1 ± 0.75%), compared to the natural (0.8 ± 0.8; Dunn test Z =
2.72, P = 0.018) and armored (0 ± 0.01; Dunn test Z = 3.63, P = 0.0008) sites (Figure
19).

Figure 19: Percent cover of eelgrass wrack per site.

Terrestrial insects
Overall, there were no significant differences in density or taxa richness across
treatments when averaged across sites. Diptera, or flies, dominated the percent
composition of the samples. Natural treatments had the highest proportion of Diptera
species (Table 1).

66

SHORELINE ARMORING IMPACTS IN PUGET SOUND
Table 1: Percent composition of terrestrial invertebrates per treatment.
%

Armored (2018
Restoration)

%

Diptera

68

Diptera

43

18

Hemiptera

9

Amphipoda

19

Hemiptera

13

Psocoptera

5

Collembola

8

Thysanoptera

10

Hymenoptera

4

Acari

6

Acari

7

Collembola

4

Psocoptera

5

Coleoptera

5

Thysanoptera

2

Hemiptera

5

Hymenoptera

4

Coleoptera

2

Coleoptera

4

Psocoptera

3

Acari

2

Hymenoptera

4

Aranae

1

Aranae

1

Thysanoptera

3

Trichoptera

1

Blattodea

1

Neuroptera

1

Lepidoptera

1

Neuroptera

1

Opiliones

1

Armored

%

Natural

Diptera

37

Collembola

Insect density by treatment
There were no statistically significant differences between insect density based on
treatment (ANOVA F(2, 42) = 0.11, P = 0.89). Average mean densities were similar across
all treatment types, where natural (345.90 ± 108.97/m²) and pre-restoration (336.73 ±
77.76/m²) shorelines had similar insect densities, and densities at armored shorelines
were slightly lower (292.69 ± 58.12/m²).

67

SHORELINE ARMORING IMPACTS IN PUGET SOUND

Figure 20: Average insect density (individuals/m²) per treatment.

Insect density per site
Insect density had a wide range, from 138-726 individuals/m² across all sites and
treatments. The natural beach at Piner Point had the highest insect density (726.67/m²).
There were no substantial trends in insect density when comparing individual sites
(Figure 21).

68

SHORELINE ARMORING IMPACTS IN PUGET SOUND

Figure 21: Average insect density (individuals/m²) per site.

Taxa richness per treatment
Taxa richness was averaged across all sites based on treatment (armored, natural,
pre-restoration). Average taxa richness was similar at each treatment type (armored =
10.93 ± 1.07; natural = 9.8 ± 0.98; pre-restoration = 9.87 ± 1.02) (ANOVA F(2) = 0.38, P
= 0.68).

69

SHORELINE ARMORING IMPACTS IN PUGET SOUND

Average taxa richness (# species) per
treatment

12
10
8
6
4
2
0
Armored

Natural

Armored (2018 restoration)

Figure 22: Average taxa richness per treatment.

Taxa richness per site
Taxa richness at each site and treatment was consistently high, ranging from
approximately 30 to 70 insect species per treatment (Figure 23). There were no
statistically significant patterns in taxa richness between treatments across the sites
(ANOVA F(2) = 0.38, P = 0.68).

70

SHORELINE ARMORING IMPACTS IN PUGET SOUND

Figure 23: Average taxa richness per site.

Forage Fish
When all sites were averaged by treatment, surf smelt spawning occurred at each
treatment type during the sampling window of December 2016-May 2017. On average,
there were more surf smelt eggs at the pre-restoration treatment (110 eggs) than the
armored (66 eggs) and natural (76 eggs) treatments (Figure 24). Sand lance spawning
occurred at natural (350 eggs) and pre-restoration treatments (5 eggs), and was not
present at armored treatments during this time frame (Figure 24).

71

SHORELINE ARMORING IMPACTS IN PUGET SOUND

Figure 24: Average number of surf smelt and sand lance eggs per treatment.

Overall, there were more surf smelt spawning events than sand lance at all
treatments, with the highest number of events at the natural sites (8 events). An event is
defined as any egg found at the site (range from 1 to 350). There were more sand lance
spawning events at the pre-restoration sites (3 events), although there were more sand
lance eggs overall at the natural site.

72

SHORELINE ARMORING IMPACTS IN PUGET SOUND

Figure 25: Number of surf smelt and sand lance spawning events per treatment during
sampling timeframe (December 2016-May 2017).

When comparing each site and treatment individually, surf smelt spawning was
more common than sand lance spawning across sites and treatments. Lost Lake had an
overall higher number of surf smelt spawning events across all sites during this time.
There was one large sand lance spawning event at the Piner Point natural site in
December where around 350 eggs were found (Figure 26).

73

SHORELINE ARMORING IMPACTS IN PUGET SOUND

Figure 26: Number of surf smelt and sand lance spawning events per site during sampling
timeframe (December 2016-May 2017).

Forage fish spawning at Big Beach
There were very few eggs found overall at treatments at the Big Beach site. In
December, there were two surf smelt eggs found at the natural site. No eggs were found
at other sites. There were slightly more surf smelt eggs found at the armored site (4 eggs)
than the natural (1 egg) and pre-restoration (2 eggs) sites in February 2017. There were
no eggs found at any of the treatments in April 2017.

74

SHORELINE ARMORING IMPACTS IN PUGET SOUND
Overall, there were few sand lance eggs found at Big Beach treatments. There
were three sand lance egg found at the pre-restoration site in December 2016. There were
no sand lance eggs found in February or April 2017 at any of the treatments (Figure 28).

Forage fish spawning at Lost Lake
Surf smelt spawned at all three treatments at Lost Lake in both January and May
2017. The only treatment without surf smelt spawning during these sampling events was
the pre-restoration site in March 2017. The highest number of surf smelt eggs were found
at the pre-restoration site in both January (64 eggs) and May (35 eggs) 2017 (Figure 29).

Figure 27: Number of surf smelt eggs per treatment at Lost Lake between January-May
2017.

75

SHORELINE ARMORING IMPACTS IN PUGET SOUND
Overall, there were few sand lance eggs found at Lost Lake treatments. There was
one sand lance egg found at the pre-restoration site in January 2017. There were no eggs
found in March or May 2017 at any of the treatments.

Forage fish spawning at Piner Point
At Piner Point, surf smelt spawning was only present in the winter (December
2016). In December, there were more surf smelt eggs found at the pre-restoration site (8
eggs) than at the natural site (4 eggs) and the armored site (none) in December 2016.
There were no eggs found at any of the treatments in February or April 2017.
There were few sand lance spawning events across treatments at Piner Point.
However, there were approximately 350 sand lance egg found at the natural site in
December 2016. There were no eggs found at any of the treatments in February or April
2017.

Fish assemblages
The number of observations (how many individual times fish were spotted) were
averaged across treatments. In general, there were more observations of fish on average
at natural sites (3 fish) compared to the armored (1.33 fish) and pre-restoration (0.33 fish)
sites (Figure 33).

76

SHORELINE ARMORING IMPACTS IN PUGET SOUND

Figure 28: Number of fish observations per treatment.

Taxa richness of fish was averaged across treatments for all sites. Taxa richness
was higher at natural sites (2.67 species) compared to armored (1.33 species) and prerestoration sites (0.33 species) (Figure 34).

77

SHORELINE ARMORING IMPACTS IN PUGET SOUND

Figure 29: Taxa richness of fish per treatment.

The total number of fish were averaged across treatments for all sites.
There was a higher average number of fish at natural sites (92.33 fish) compared to the
armored (68.33 fish) and pre-restoration (0.33 fish) sites (Figure 35).

78

SHORELINE ARMORING IMPACTS IN PUGET SOUND

Figure 30: Total number of fish observations per treatment.

Fish Observations
At Big Beach, there were two species of fish observed at the natural site:
unknown forage fish and anchovy (approximately 200 fish). There were no observations
at the armored or pre-restoration sites (Table 2). At Lost Lake, there were more total
observations, total fish, and number of species at the natural site (Table 3). At Piner
Point, there were more total observations and number of fish species at the natural site.
There were more overall fish (200 unknown species of forage fish) observed at the
armored site (Table 4).

79

SHORELINE ARMORING IMPACTS IN PUGET SOUND
Table 2: Fish observations.
Site

Big Beach

Treatment

Total
observations

Total fish

Number of
species

Species type

Armored

0

0

0

N/A

Natural

2

101

2

Forage fish
(unknown),
anchovy

Armored (2018
Restoration)

0

0

0

Armored

2

4

2

Sculpin, rock
sol

Natural

4

6

3

Shiner perch,
sculpin, trout
(unknown)

Armored (2018
Restoration)

1

1

1

Saddleback
gunnel

Armored

2

201

2

Sculpin, forage
fish (unknown)
Surf smelt,
salmon
(unknown),
forage fish
(unknown)

Big Beach

Big Beach

Lost Lake

Lost Lake

Lost Lake

Piner Point

Piner Point

Piner Point

80

Natural

3

170

3

Armored (2018
Restoration)

0

0

0

N/A

N/A

SHORELINE ARMORING IMPACTS IN PUGET SOUND

Chapter 5: Discussion
Shoreline armoring reduced complexity of the nearshore. Beaches with
development had altered nearshore habitats compared to natural beaches. Changes in the
nearshore can lead to altered biological response for fish by reducing vital habitat and
prey availability for fish. The following discussion summarizes and provides context to
the results found in this study.

Marine riparian vegetation
Overstory vegetation was higher at natural treatments compared to armored
treatments. In addition, natural treatments had a higher average number of trees. Most of
the trees at all sites were native. The percent of understory cover was similarly high at
each shoreline type, around 70-90% cover across the treatment types.
The study sites in the MIAR had a relatively high amount of over and understory
vegetation cover compared to studies in the literature. Vegetation conditions found in this
study are not always typical of developed shorelines in the Puget Sound. Overall, the
shorelines at these sites are relatively healthy compared to highly urban shorelines. The
average plant health ratings across sites was four, and above three at all sites, where one
is dead and five shows vigorous growth. Shoreline armoring is usually associated with
the removal of over and understory vegetation and the replacement with maintained yards
where grass lawns are more common (Heerhartz et al., 2014). This was not always the
case at armored treatments, and especially the restoration sites where vegetation was
overgrown.
81

SHORELINE ARMORING IMPACTS IN PUGET SOUND
The vegetation present at these sites, and in general, supports the vital connection
between terrestrial and aquatic ecosystems. Terrestrial vegetation fosters habitat for
insects and provides natural beach function, such as shading and moisture retention. More
riparian vegetation contributes to the input of terrestrial insects in the nearshore (Toft et
al., 2013). Terrestrial insects, such as dipterans (flies), can be carried by wind from
terrestrial ecosystems onto the water surface and provide food for juvenile Chinook
salmon (Munsch et al., 2016).
Maintaining shoreline vegetation and an intact upper-beach is necessary for full
function of the supratidal zone. Previous studies have shown that vegetation removal,
which is common at armored treatments, results in significant differences between
backshore invertebrate and insect assemblages (Toft et al., 2014; Heerhartz et al., 2014).
Introducing native riparian vegetation at armored shorelines or after armoring removal
can improve the marine-terrestrial connectivity and may facilitate a rapid response from
terrestrial macroinvertebrate assemblages, a vital part of Chinook diets and coastal food
webs (Toft et al., 2014; Lee et al., 2018).

Wrack cover
Shoreline armoring is known to reduce the accumulation of wrack and logs on
Puget Sound shorelines and reduces the relative proportion of terrestrial wrack (Heerhartz
et al., 2014). A significantly lower amount of wrack was found at armored treatments
compared to the pre-restoration treatments. This study demonstrated significant
differences in overall wrack cover when comparing across treatments, although small

82

SHORELINE ARMORING IMPACTS IN PUGET SOUND
sample sizes made for weaker statistical inferences. Similar amounts of wrack were
found at the natural and pre-restoration treatments in this study.
Wrack was also analyzed by its composition (i.e. marine vs. terrestrial). Natural
treatments had a higher proportion of terrestrial wrack (5.60 ± 1.82), when compared to
armored (2.1 ± 1.99) and pre-restoration (1.72 ± 0.62) treatments, although this finding is
not statistically significantly. Terrestrial inputs are important for nearshore ecosystems in
providing nutrients and habitat for invertebrates in nearshore ecosystems (Heerhartz et
al., 2014). The decrease of vegetation inputs from the uplands due to shoreline
development can decrease the amount of terrestrial organic material that accumulates on
the upper shore (Heerhartz et al., 2014). Terrestrial inputs are known to influence the
abundance and composition of invertebrates in nearshore ecosystems (Heerhartz et al.,
2015). Higher percent cover of terrestrial vegetation cover may be due to the increased
terrestrial vegetation present at natural treatments, although further analysis is needed to
assess this correlation.
Natural sites had a higher proportion of eelgrass wrack (3.81 ± 0.85). Eelgrass
wrack differed significantly across all treatments. A Friedman’s test showed that there
was no interaction between site and treatment, meaning the significant difference
between natural and armored treatments can be accepted regardless of site effects.
Eelgrass was highest at Lost Lake natural treatments (5.8 ± 1.53), potentially reflecting
the local abundance of seagrass meadows in the region (Perla, 2018). Further analysis
could look at site as a factor, as local eelgrass beds and drift cell location may have an
influence on eelgrass cover across sites.

83

SHORELINE ARMORING IMPACTS IN PUGET SOUND
The placement in the drift cell can affect the direction of drift (e.g. longshore
currents, estuarine outflow) and can affect biological response variables. For example,
drift cell placement of sites included in this study (Big Beach, Lost Lake, and Piner
Point) may have caused some of the variation in accumulation of wrack. Total percent
cover of wrack was higher at both Big Beach and Lost Lake, located in the same drift
cell, compared to Piner Point, which is located within a diverging drift cell where net
shore drift goes in either direction and less accumulation occurs (Washington State
Coastal Atlas, 2018).
The overall accumulation and wrack composition measured in this study can be
used as an indicator for healthy habitat for invertebrate populations. Wrack total percent
cover was similar at natural sites (approximately 21% cover) and pre-restoration sites
(approximately 25% cover) and lower at the armored sites (approximately 8% cover).
This is comparable to findings from Heerhartz et al. (2014), where wrack composition
was more diverse at natural sites and amounted to approximately 20% cover at natural
sites compared to 10% percent cover at armored beaches during the summer months.
Diverse organisms take advantage of the shelter and moisture provided by wrack and logs
that accumulate on the beach (Heerhartz et al., 2014). Wrack provides habitat (moisture
and shelter) for talitrid species, which are mobile scavengers that feed on wrack—
especially algae—and other detritus (Heerhartz et al., 2015).
This study compared lower wrack lines, although upper wrack lines and
accumulated logs were present at most natural treatments. As a result, the total percent
cover of wrack reported in this study is likely not representative of the total percent cover

84

SHORELINE ARMORING IMPACTS IN PUGET SOUND
on the actual beach. Future studies should measure total surface area of wrack in the field
(combining upper and lower wrack lines) to get a more accurate representation of wrack
cover on the beach and a better understanding of differences in wrack cover between
armored and unarmored shorelines.
The results from this study represents a snapshot in time, as samples were
collected during one tide at the site scale. Wrack is delivered to beaches after almost
every high tide, and therefore increased sampling frequency and replication may illustrate
temporal effects more thoroughly. In addition, sampling across seasons may show trends
not apparent in this study. For example, increased wrack is common in the autumn as
deciduous trees lose their leaves (Sobocinski, 2003).

Terrestrial insects
This study specifically assessed the differences in abundance and taxa richness of
terrestrial insects, as terrestrial insects may be essential dietary components of fish
throughout the Puget Sound (Lee et al., 2016). Terrestrial invertebrates can be used as a
metric for habitat quality and as an indicator of available prey resources for salmon (Toft
et al., 2013). Natural sites had a slightly higher insect density (345.90 ± 108.97/m²),
although the results were not statistically significant compared to restoration (336.73 ±
77.76/m²) and armored (292.69 ± 58.12/m²) sites. However, when assessing each
individual site, the natural treatment at Piner Point had a much higher insect density
(726.67 m²) than all treatments at all sites. The abundance of insects seemed to show no
correlation with the amount of overhanging vegetation. For example, overstory

85

SHORELINE ARMORING IMPACTS IN PUGET SOUND
vegetation was highest at natural treatments, but insect abundance was similar across
sites.
Taxa richness was slightly higher at armored treatments, although this result was
not statistically significant. Natural sites were the least diverse in taxa richness. Diptera
dominated samples at all shoreline types (armored, natural, and restoration). Diptera,
Psocoptera, and Homoptera terrestrial invertebrate species have been found to dominate
proportions of Chinook salmon diets (Munsch et al., 2016).
Higher insect abundance and taxa richness between natural versus armored
treatments have been documented in previous studies (Sobosinski, 2003, Romanuk &
Levings, 2003, Sobocinski et al., 2010, Toft et al., 2013, Lee at al., 2018). Insect
abundance and taxa richness is known to be greater at sites with intact shoreline
vegetation than at sites lacking vegetation (Sobosinski, 2003). The Olympic Sculpture
Park case study demonstrated clearer results where both density and taxa richness were
higher in areas where shoreline vegetation had been planted than areas that had none
(Toft et al., 2013). The high amount of vegetation at all treatment types in the MIAR
could be a factor in the similarities of taxa richness and density across shoreline types.
Each treatment was 70-90 percent covered with understory vegetation. Overstory was
highest at the natural treatments (about 80 percent), but was still relatively high
(approximately 40 percent cover) at both the armored and restoration sites, so perhaps the
differences in over and understory vegetation were not substantial enough to affect insect
taxa richness and density.

86

SHORELINE ARMORING IMPACTS IN PUGET SOUND
This data represents a snapshot in time of shoreline conditions in the MIAR
during the summer months. Further research with more data collection over a longer time
may elucidate insect community response to treatment type and vegetation differences
more clearly (Sobocinski, 2003).

Forage fish
Forage fish (surf smelt and sand lance) spawning occurred at each beach location
across the sampling window of December 2016 to May 2017. Surf smelt spawning events
were consistently high across treatment types, where the number of spawning events
ranges from 6 to 8 across treatments (armored, natural, pre-restoration), where the most
spawning events were found at natural sites (8 events). The highest amount of sand lance
spawning events occurred at the pre-restoration treatments (3 events), compared to the
natural (1 event) and the armored (no spawning events). When comparing the number of
eggs across treatments, natural treatments had a higher average number of sand lance
eggs, and pre-restoration treatments had the highest number of surf smelt eggs.
The effects of shoreline armoring on forage fish spawning is unclear from this
study. Further analysis should continue to monitor forage fish spawning over longer
periods of time. This may indicate long trends in preferential spawning locations.
Comparing over long time scales would eliminate biases due to spawning seasonality, as
preliminary results in this study suggest that sand lance tend to spawn in the winter,
whereas surf smelt spawn all year round.
The effects of shoreline armoring may be clearer if the health of egg embryos, or
the number of dead versus healthy eggs, is examined between armored and natural
87

SHORELINE ARMORING IMPACTS IN PUGET SOUND
treatments. Shoreline armoring is known to decrease the overall health and survival of
egg embryos (Rice, 2006). Forage fish egg survival depends on sediments, shade, wrack
cover, and other debris characteristic of natural shorelines (Rice, 2006). Accounting for
these site characteristics in future analyses may elucidate the effects of shoreline
armoring on forage fish spawning and embryo survival as well.

Fish assemblages
Fish data collected from snorkel surveys was limited, as only a single survey was
successful at each site. However, the study did indicate preliminary trends in fish use of
the nearshore, where more observations of fish occurred at natural treatments. Taxa
richness of fish was higher at natural treatments compared to the armored and prerestoration treatments as well. Due to limited data, statistically significant trends of fish
use or feeding behaviors were not established. Previous findings from snorkel surveys in
the Puget Sound demonstrate that shoreline armoring alters fish distribution and prey
availability in shallow water habitats (Toft et al., 2007). Development along the shoreline
is likely to change the character of overhanging vegetation and insects, as demonstrated
in this study, which can have cascading effects on fish habitat and prey resource
availability.
Understanding of fish behavior and habitat use in the nearshore is limited
(Munsch et al., 2016). Nearshore fish communities are typically studied using physical
capture (e.g. netting) rather than observing behavior directly. As a consequence, it is
difficult to connect their fine-scale habitat use and behavior to basic ecological theory
(Munsch et al., 2016). Snorkel surveys allow an observational where fish can be observed
88

SHORELINE ARMORING IMPACTS IN PUGET SOUND
behaving more naturally compared to beach sein and netting techniques (Munsch et al.,
2016).
Previous studies using these snorkel survey methods have demonstrated clearer
results between armoring and natural shorelines, where natural shorelines demonstrated
an increase in fish abundance and feeding behavior (Toft et al., 2007; Toft et al., 2013,
Heerhartz et al., 2015; Munsch et al., 2016). These studies were conducted over longer
periods of time and were located in highly urban environments where an intermittent
natural beach may have a greater impact.
Overall, this study provides insight into the localized effects of armoring on fish
communities in the MIAR. This study fine-tuned snorkel survey methods that can be used
by the VNC to continue to monitor fish across time in the MIAR.

Suggestions for future fish surveys
Fish observations occurred at different sites at different months (Big Beach and
Lost Lake in July, Piner Point in August) and captures only a snapshot in time. Fish data
is not directly comparable between sites due to differences in fish migration patterns
across time. The peak out-migration of juvenile salmon is around June and July, although
juvenile Chinook are found along Puget Sound shorelines from late January through
September (Shoreline Monitoring Toolbox, 2012). Only one school of salmon was
observed over the three surveys, where a school of about 20 juvenile pink salmon were
spotted at Piner Point in August 2017. Forage fish (either surf smelt or unknown species)
dominated the observations (both individually and in large schools) at Big Beach in July
2017 and Piner Point in August 2017. A few observations of large schools of forage fish
89

SHORELINE ARMORING IMPACTS IN PUGET SOUND
resulted in the large number of fish observations at these sites in particular. Surf smelt are
known to spawn at various times of the year, whereas Pacific sand lance spawn during
winter months (Rice, 2006).
The differences in physical space between sites could account for some of the
differences in fish use. For example, Big Beach and Lost Lake are located within
Quartermaster Harbor, whereas Piner Point is more exposed to deeper waters and faster
currents found in Dalco and East Passages (Washington State Coastal Atlas, 2018). Many
factors can affect the distribution of salmon and other fishes including, but not limited to,
proximity to freshwater and out-migration corridors, predation risk, and water depth, all
of which may differ between sites included in this study (Toft et al., 2007).
Due to differences in fish use of nearshore habitats across space and time, future
studies should survey all three sites in the same month, and preferably in the same day.
Again, the peak out-migration of juvenile salmon is around June and July, although
juvenile Chinook are found along Puget Sound shorelines from late January through
September (Shoreline Monitoring Toolbox, 2012). Future studies may focus on sampling
between June and July to better document the effects of armoring on juvenile salmon use
in the nearshore. Lastly, surveys were conducted during daylight hours only, which may
not account for all fish use and behavior, although night surveys are not recommended
with volunteers due to safety concerns.
The most favorable conditions for snorkel surveys occurred during the highest,
high tide of the day, during the evening, where high tides reached the bulkhead and
calmer conditions were more common. Fish are known to be more active during the

90

SHORELINE ARMORING IMPACTS IN PUGET SOUND
evening time, according to local fishers in the area. The highest, high tide occurred in the
evening occasionally during the summer, which made for more ideal sampling
conditions. Conditions were not favorable for sampling in the middle of the day (between
10am and 4pm) during the lower-high tide. Wind, increased turbidity, and sunlight
decreased visibility during the mid-day. In addition, the lower-high tide did not always
reach the bulkhead, which could potentially lessen the effect of armoring on fish because
the tide does not reach the bulkhead during this time.
Snorkelers experienced very limited visibility in the water column during all
surveys, where visibility ranged from 1.5 to 2 meters. The Shoreline Monitoring
Toolbox’s fish survey methods advise a 2.5 meter visibility, which was never achieved
during these surveys. Surveys scheduled during earlier months (May-June) were not
conducted due to low visibility (less than one meter). Future studies should ensure at least
a 1.5 to 2 meter visibility before proceeding to sample.
Lastly, future analysis should compare fish data to nearshore habitat parameters
measured in this study. As more fish data is collected at these sites, comparing fish data
to habitat parameters (i.e. canopy cover and terrestrial insect data) may increase the
understanding of fish response to nearshore habitat conditions.

Study design and suggestions for future research
Beaches in the MIAR are unique
Sampling occurred over a small geographic region in Quartermaster Harbor
between Vashon Island and Maury Island in the Puget Sound, and therefore, the results

91

SHORELINE ARMORING IMPACTS IN PUGET SOUND
may not be directly comparable to shorelines across the Puget Sound. Shorelines in the
Maury Island Aquatic Reserve tend have intermittent development, but have some natural
beaches along with armored beaches that may still have beneficial qualities of a natural
shoreline (i.e. over and understory vegetation, wrack accumulation). Studies have focused
on shorelines that are significantly more modified than elsewhere in Puget Sound, such as
in Elliott Bay in Seattle, where about 90% of the shoreline is modified by retaining
structures (Weitkamp et al. 2000) compared to one-third for the rest of the Puget Sound
(Puget Sound Partnership, 2012). The results of this study may not show as drastic results
as shorelines in highly developed areas. That said, the results of this study demonstrate
that shoreline development disrupts marine-terrestrial connectivity and alters important
habitat for fish, which can be widely applied to shorelines across the Puget Sound.
The paired design in this study helps control for variability in environmental
parameters and provides the ability to test for armoring-related differences. However,
differences between armored and unarmored shorelines were not always clear. Armored
(both the armored and pre-restoration) treatments have different levels of development
and yard maintenance (i.e. grass vs. overgrown understory vegetation) that could cause
differences in response variables. For example, the armored treatments are owned and
maintained by property owners, whereas restoration sites that have been purchased by
King County have not been maintained for years (Perla, 2018). This may suggest that the
condition of the habitat (presence of over and understory vegetation) along shorelines
may provide habitat benefits, even when shoreline armoring is present.

92

SHORELINE ARMORING IMPACTS IN PUGET SOUND

Localized geomorphology in the MIAR
Unlike many shoreline restoration removal projects in the literature, the bulkheads
being removed in MIAR are along actively eroding, high-bank shorelines. The
geomorphology at these sites is distinct from other studies and bulkhead removals that
have been done on low-bank waterfront. More immediate changes are expected after
bulkhead removals are completed (Perla & Metler, 2016). The placement of each
treatment in the drift cell should be considered when examining response variables,
especially after restoration. The placement in the drift cell can affect the direction of drift
(e.g. longshore currents, estuarine outflow) and can affect biological response variables.
For example, drift cell placement of sites included in this study (Big Beach, Lost Lake,
and Piner Point) may have caused some of the variation in accumulation of wrack. Total
percent cover of wrack was higher at both Big Beach and Lost Lake, located in the same
drift cell, compared to Piner Point, which is located within a diverging drift cell where
net shore drift goes in either direction and less accumulation occurs (Washington State
Coastal Atlas, 2018).
In contrast to many shoreline armoring removal studies in highly urban areas with
severe loss of beach, sites in the MIAR are characteristic of shallow, low-grade beaches
that provide shallow water habitat for fish. Fish are known to prefer shallow water areas
as a nursey habitat and refuge from prey during juvenile life stages (Ruiz et al., 1993).
Larval forage fish are known to use shallow water and beaches as nursey grounds as well
(Pentilla, 2007).

93

SHORELINE ARMORING IMPACTS IN PUGET SOUND
Analysis did not include beach profile data that was collected in 2016 and 2017.
Future studies can look at the changes in beach profile over the different summers to
analyze the effects of the shoreline armoring and shoreline restoration. In addition,
shoreline armoring elevation on the beach was not examined in this study. Lower
elevations of shoreline armoring negatively affect beach parameters on local and larger
spatial scales (Dethier et al., 2016). Armoring placed below MHHW has substantially
more effect on parameters than armoring higher on the beach (Dethier et al., 2016).
Future analysis in the MIAR should account for shoreline armoring elevation, as
restoring sites with armoring lower on the beach may be more beneficial to restoring
natural physical and biological conditions.

Citizen science: a challenge and a resource
This study was conducted by the VNC with the help of citizen-science volunteers.
This study highlights the importance of citizen science monitoring and the standardized
protocols using Puget Sound Partnership’s Shoreline Monitoring Toolbox. Citizen
science projects can enhance scientific research and assist government agencies with
restoration monitoring for minimized costs and effort (Perla, 2018). Results from this
project established a baseline dataset of pre-restoration shoreline conditions that will
assist King County in understanding the benefits of shoreline restoration. Citizen science
research can also help to meet community involvement and outreach goals established for
many county and state governments. Participation from citizen science volunteers is
beneficial for many reasons, including the understanding of shoreline issues by local
citizen scientists, fostering community ambassadors for shoreline health, and increasing
94

SHORELINE ARMORING IMPACTS IN PUGET SOUND
community buy-in (Perla, 2018). Monitoring in the MIAR is an effective way of
involving and educating the local community in restoration projects along the Vashon
and Maury Island shorelines.
Citizen science projects are often organized by volunteer groups or nonprofits that
often lack funding and resources for long term monitoring. The design of this study is
limited due to the nature of citizen science involvement. The study lacks replication
across time and sample sizes are low. Each beach was sampled one time during the
summer, as it is difficult to train and organize large groups of volunteers with limited
staff capacity at the VNC. Measurements may be inherently biased between individuals
for each study. Different volunteers were involved across the duration of this project, and
therefore some inaccuracy and lack of objectivity may be an issue.
That said, this project demonstrates a strong example of citizen science
involvement in shoreline restoration monitoring projects across the Puget Sound. The
data collected in this study would not have existed without the help and support of citizen
science volunteers. The standardized methods from the Shoreline Monitoring Toolbox
are intended to be used by nonprofits and citizen science groups across the Puget Sound.

Working with small sample sizes
One of the challenges in this analysis is small sample sizes for vegetation, wrack,
and fish parameters. Small sample sizes can increase error rate and potentially distort the
response interpretations, so the results cannot necessarily be generalized to other
shorelines across the Puget Sound (Lee et al., 2018). Ideally, future studies would sample
more often over the summer months and eventually across time to look at long term
95

SHORELINE ARMORING IMPACTS IN PUGET SOUND
trends with larger sample sizes. Increased sample sizes over time is essential for fully
understanding the effects of armoring (Lee et al., 2018). Coordinating monitoring
projects with citizen scientists can be difficult due to lack of time and resources from
small nonprofits and local volunteers. If the VNC wants more conclusive answers about a
certain beach parameter, the non-profit may choose to focus on one beach parameter and
sample more often. As this study evolves, the VNC may consider coordinating with a
small group of volunteers to take responsibility for sampling a certain beach parameter
throughout the year. For example, forage fish beach spawning surveys are currently
monitored throughout the year by trained citizen scientists with little coordination needs
from the VNC.

Suggestions for planning, management, and restoration
Existing and new shoreline management policies should encourage homeowners
and stakeholders to protect natural shorelines and embrace shoreline restoration when it
can simultaneously protect properties and biodiversity. It is critical for policymakers to
consider the benefits of shoreline armoring removal before undertaking new shoreline
development (Lee et al., 2018). When artificial barriers are removed and aquatic habitats
merged with terrestrial habitats, biological and physical processes may be reconnected
and allowed to function more naturally (Toft et al., 2013).
The impacts of shoreline armoring may be reversible, as seen in previous studies
of shoreline restoration (Toft et al., 2014; Dethier et al., 2016). Although few studies
have assessed the effectiveness of armoring removal on restoring coastal ecosystems,
studies generally demonstrate that shorelines without armoring can host higher
96

SHORELINE ARMORING IMPACTS IN PUGET SOUND
abundance and diversity (Toft et al., 2014; Heerhartz et al., 2014; Dethier et al., 2016;
Lee et al., 2018). Maintaining marine-terrestrial linkages should be a top priority for
shoreline restoration, current homeowners, and future development where armoring is
necessary (Heerhartz et al., 2015).
Removal of armoring is not always feasible. Recently, alternatives to shoreline
armoring, including armoring removal, have emerged that can both protect coastal
infrastructure and restore ecological health (Gittman et al., 2016). Shorelines without
armoring can provide the same function as natural erosional barriers (Lee et al., 2018). If
engineered correctly, employing techniques that can stabilize the shoreline by mimicking
site-specific shoreline processes, restoration may still provide the benefits of coastal
protection (Guerry et al., 2012; Toft et al., 2013). For example, large woody debris
protects from beach erosion, but also enhances wrack accumulation and improves
aquatic-terrestrial connectivity (Heerhartz et al., 2014).
Continuing to monitor post-restoration in the MIAR is essential to assess the
recovery of restored coastal ecosystems. Few studies have assessed the effectiveness of
armoring removal on restoring coastal ecosystems (Lee et al., 2018). Increasing the
geographical scope and number of studies of these coastal biota types can increase
knowledge of restoration effectiveness to help inform management policies.

Chapter 6: Conclusion
The ecological impact of large-scale shoreline armoring on nearshore ecosystems
is still largely unknown, though this study demonstrated some proximal effects of
97

SHORELINE ARMORING IMPACTS IN PUGET SOUND
shoreline armoring in the MIAR. Overall, armoring can affect biological conditions of
shorelines on a local scale. The results of this research demonstrate a baseline of
environmental conditions in the MIAR and the localized effects of shoreline armoring.
Understanding the current condition of these beaches is necessary to determine how
coastal biota will respond to shoreline armoring removal and restoration occurring in the
summer of 2018.
This study showed results from studies from five different shoreline parameters
including vegetation cover, wrack cover, insect density and taxa richness, forage fish
beach spawning, and fish observational studies. This study indicated noticeable
differences between shoreline types (armored, natural, and pre-restoration treatments).
The following is a summary of results:

1. Vegetation: The presence of armoring decreased the percent cover of overstory
vegetation and trees compared to natural beaches. Similar percent cover of
understory vegetation was found across treatments (armored, natural, prerestoration). Natural beaches had more native species and a higher plant health
index.
2. Wrack: Pre-restoration sites had a higher abundance of total wrack cover. Natural
beaches had a higher abundance of terrestrial and eelgrass wrack cover. Upper
wrack lines and logs were found at natural beaches.

98

SHORELINE ARMORING IMPACTS IN PUGET SOUND
3. Insects: Insect diversity and taxa richness were similar across treatments. Percent
composition of samples were dominated by Diptera, or flies, at all sites, but was
greatest at natural beaches.
4. Forage fish: Surf smelt and sand lance spawning occurred at all treatment types.
Natural beaches had a higher number of sand lance eggs. Sand lance tended to
spawn in the winter, whereas surf smelt spawning was present throughout the
sampling time frame (December 2016 to May 2017).
5. Fish: Natural sites had a higher taxa richness, number of individual fish sightings,
and number of individual fish.

Shoreline armoring clearly disrupts the connection between terrestrial-aquatic
ecosystems, as demonstrated by the reduction in backshore vegetation and terrestrialassociated wrack cover. Terrestrial-aquatic connection along the shoreline provides vital
habitat functions for invertebrates and fish that utilize the shorelines. Overall, this study
demonstrates that shoreline development changes the biological response to nearshore
habitats by reducing the marine-terrestrial connectivity which provides vital habitat for
invertebrates and fish. The effects of armoring may be minimized if shorelines mimic
natural conditions, as seen in the Olympic Sculpture Park example (Toft et al., 2013), by
ensuring vital nearshore parameters are maintained and marine-terrestrial ecosystems
remain connected. Shoreline restoration and shoreline armoring removal is effective in
improving the health and productivity of coastal ecosystems, and will continue to be
monitored at the three sites in the MIAR.

99

References
Bilkovic, D.M, Roggero, M.M. (2008). Effects of coastal development on nearshore estuarine
nekton communities. Marine Ecology Progress Series, 358:27-39.

Booth, B., Legg, N. (2017). Technical Memorandum: Technical Review Memorandum, Maury
Island Aquatic Reserve Armoring Removal Project (E00376-20). Prepared by Cardno,
Seattle, WA for the Watershed Stewardship Program, King County, WA, USA.

Bozek C.M., Burdick D.M. (2005). Impacts of seawalls on saltmarsh plant communities in the
Great Bay Estuary, New Hampshire USA. Wetlands Ecology and Management,
13(5):553-568

Brennan, J.S., and H. Culverwell. (2004). Marine Riparian: An Assessment of Riparian
Functions in Marine Ecosystems. Published by Washington Sea Grant Program,
UW Board of Regents, Seattle, WA.

Brennan, J.S. (2007). Marine Riparian Vegetation Communities of Puget Sound. Puget Sound
Nearshore Partnership Report No. 2007-02. Published by Seattle District, U.S. Army
Corps of Engineers, Seattle, Washington.

Brennan, J., Culverwell, H., Gregg, R., Granger, P. (2009). Protection of Marine Riparian
Functions in Puget Sound, Washington. Published by Washington Sea Grant Program,
UW Board of Regents Seattle. 139 pp.

1

Brennan, J.S., Higgins, K.F., Cordell, J.R., Stamatiou, V.A. (2004). Juvenile Salmon
Composition, Timing, Distribution, and Diet in Marine Nearshore Waters of
Central Puget Sound in 2001-2002. King County Department of Natural
Resources and Parks, Seattle, WA. 164 pp.

Bulleri F., Chapman M.G., Underwood A.J. (2005). Intertidal assemblages on seawalls
and vertical rocky shores in Sydney Harbour, Australia. Austral Ecology,
30(6):655-667.

Chapman, M.G. (2003). Paucity of mobile species on constructed seawalls: effects of
urbanization on biodiversity. Marine Ecology Progress Series, 264, 21-29.

Chapman, M.G., Underwood, A.J. (2011). Evaluation of ecological engineering of
‘armoured’ shorelines to improve their value as habitat. Journal of Experimental
Marine Biology and Ecology, 400(1–2):302-313.

Conrad, C.C., Hilchey, K.G., (2011). A review of citizen science and community-based
environmental monitoring: Issues and opportunities. Environmental Monitoring
and Assessment, 176(1-4): 273-91.

Coyle, J.M., M.N. Dethier (2010). Review of shoreline armoring literature, in Shipman,
H, Dethier, M.N., Gelfenbaum, G., Fresh, K.L., Dinicola, R.S. (2010). Puget

2

Sound Shorelines and the Impacts of Armoring - Proceedings of a State of the
Science Workshop, May 2009: U.S. Geological Survey Scientific Investigations
Report 2010-5254, pp. 245-265.

Davis J.L.D., Levin L.A., Walther S.M. (2002). Artificial armored shorelines: sites for opencoast species in a southern California Bay. Marine Biology, 140(6):1249-1262

Desbonnet, A., P. Pogue, V. Lee, and N. Wolff. 1994. Vegetated buffers in the coastal Zone
– A summary review and bibliography. Coastal Resources Center Technical Report
No. 2064. University of Rhode Island Graduate school of Oceanography. Narragansett, R
I 02822. 72 pp.

Dethier, M.N., Raymond, W.W., McBride, A.N., Toft, J.D., Cordell, J.R., Ogston, A.S., Berry,
H.D. (2016). Multiscale impacts of armoring on Salish Sea shorelines: Evidence for
cumulative and threshold effects. Estuarine, Coastal and Shelf Science, 175, 106–117.

Department of Ecology (Ecology, 2016). Shoreline Management: Introduction to the Shoreline
Management Act. Accessed February 24, 2016 from
http://www.ecy.wa.gov/programs/sea/sma/st_guide/intro.html

Doody, J.P., 2013. Coastal squeeze and managed realignment in southeast England, does it tell
us anything about the future? Ocean and Coastal Management, 79, 34e41.

3

Duffy, E.J. (2003). Early marine distribution and trophic interactions of juvenile salmon
in Puget Sound (Doctoral dissertation, University of Washington).

Duffy, E.J., Beauchamp, D.A., Sweeting R.M., Beamish, R.J., Brennan, J.S. (2010).
Ontogenetic diet shifts of juvenile Chinook salmon in nearshore and offshore
habitats of Puget Sound. Trans. Am. Fish Soc., 139: 803-823.

Duffy E. J., Beauchamp D.A. (2011). Rapid growth in the early marine period improves
the marine survival of Chinook salmon (Oncorhynchus tshawytscha) in Puget
Sound, WA. Can. J. Fish Aquat. Sci., 68:232-240.

Dugan, J.E., Hubbard, D.M., Rodil, I.F., Revell, D.L., Schroeter, S. (2008). Ecological
effects of coastal armoring on sandy beaches. Marine Ecology, 29(s1):160-170.

Dunagan, C. (2017, October 22). With sea-level rise, waterfront owners confront their
options. Retrieved March 21, 2018, from
https://www.eopugetsound.org/magazine/is/slr.

Fresh, K.L. (2006). Juvenile Pacific Salmon in Puget Sound. Puget Sound Nearshore
Partnership Report No. 2006-06. Published by Seattle District, U.S. Army Corps
of Engineers, Seattle, Washington.

4

Guerry, A.D., Ruckelshaus, M.H., Arkema, K.K., Bernhardt, J.R., Guannel, G., Kim, C.K., Marsik, M., Papenfus, M., Toft, J.E., Verutes, G., et al. (2012). Modeling benefits
from nature: using ecosystem services to inform coastal and marine spatial planning. Int.
J. Bio. Sc., Eco. Serv. Man., 1–15 (19 more authors).

Gittman, R.K., Fodrie, F.J., Popowich, A.M., Keller, D.A., Bruno, J.F., Currin, C.A., Peterson,
C.H., Piehler, M.F. (2015). Engineering away our natural defenses: an analysis of
shoreline hardening in the US. Frontiers in Ecology and Environment, 13(6):301-307.

Gittman, R.K., Peterson, C.H., Currin, C.A., Fodrie, F.J., Piehler, M.F., Bruno, J.F. (2016).
Living shorelines can enhance the nursery role of threatened estuarine habitats.
Ecological Applications, 26(1):249-263.

Guttman, E.S. (2009). Characterizing the Backshore Vegetation of Puget Sound (Unpublished
master's thesis). Prescott College.

Hart, C. (2010). Marine Shoreline Armoring and Puget Sound (10-06-003). Washington State
Department of Ecology, Shorelands and Environmental Assistance Program.

Heatherington, C., Bishop, M.J. (2012). Spatial variation in the structure of mangrove forests
with respect to seawalls. Marine and Freshwater Research, 63(10):926-933.

5

Heerhartz, S.M., Toft, J.D., Cordell, J.R., Dethier, M.N., & Ogston, A.S. (2016).
Shoreline Armoring in an Estuary Constrains Wrack-Associated Invertebrate
Communities. Estuaries and Coasts, 39(1), 171–188.

Heerhartz, S.M., & Toft, J.D. (2015). Movement patterns and feeding behavior of
juvenile salmon (Oncorhynchus spp.) along armored and unarmored estuarine
shorelines. Environmental Biology of Fishes, 98(6), 1501–1511.

Heerhartz, S.M., Dethier, M.N., Toft, J.D., Cordell, J.R., & Ogston, A.S. (2014). Effects
of Shoreline Armoring on Beach Wrack Subsidies to the Nearshore Ecotone in an
Estuarine Fjord. Estuaries and Coasts, 37(5), 1256–1268.

Higgins, K., Schlenger P., Small, J., Hennessy, D., and Hall, J. (2005). Spatial
relationships between beneficial and detrimental nearshore habitat parameters in
WRIA 9 and the City of Seattle. Proceedings of the 2005 Puget Sound Georgia
Basin Research Conference.

Jedrzejczak, M.F. (2002a). Spatio-temporal decay 'hot spots' of stranded wrack in a Baltic
sandy coastal system. Part I. Comparative study of the pattern: 1 type of wrack vs
beach sites. Oceanologia 44:491-512.

Jedrzejczak, M.F. (2002b). Stranded Zostera marina L. vs wrack fauna community
interactions on a Baltic sandy beach (Hel Peninsula, Poland): a short-term pilot

6

study. Part II. Driftline effects of succession changes and colonisation of beach
fauna. Oceanologia 44:367-387.

Lam N.W.Y., Huang R., Chan B.K.K. (2009). Variations in intertidal assemblages and zonation
patterns between vertical artificial seawalls and natural rocky shores: a case study from
Victoria Harbour, Hong Kong. Zoological Studies 48(2):184-195.

Lee T.S., Toft J.D., Cordell J.R., Dethier M.N., Adams J.W., Kelly R.P. (2018) Quantifying the
effectiveness of shoreline armoring removal on coastal biota of Puget Sound. PeerJ
6:e4275.

Moreira, J., Chapman, M.G., Underwood, A.J. (2006). Seawalls do not sustain viable
populations of limpets. Marine Ecology Progress Series, 322:179-188.

Morley, S.A., Toft, J.D., Hanson, K.M. (2012). Ecological effects of shoreline armoring on
intertidal habitats of a Puget Sound urban estuary. Estuaries and Coasts 35(3):774-784.

Munsch, S., Cordell, J., & Toft, J. (2015). Effects of seawall armoring on juvenile Pacific salmon
diets in an urban estuarine embayment. Marine Ecology Progress Series, 535, 213–229.

Munsch, S., Cordell, J., & Toft, J. (2016). Fine-scale habitat use and behavior of a nearshore fish
community: nursery functions, predation avoidance, and spatiotemporal habitat
partitioning. Marine Ecology Progress Series, 557, 1–15.

7

Munsch, S.H., Cordell, J.R., & Toft, J.D. (2015). Effects of shoreline engineering on
shallow subtidal fish and crab communities in an urban estuary: A comparison of
armored shorelines and nourished beaches. Ecological Engineering, 81, 312–320.

Munsch, S. H., Cordell, J.R., Toft, J.D., & Morgan, E.E. (2014). Effects of Seawalls and
Piers on Fish Assemblages and Juvenile Salmon Feeding Behavior. North
American Journal of Fisheries Management, 34(4), 814–827.

National Research Council (NRC) 2000. Clean Coastal Waters: Understanding and Redu
cing the Effects of Nutrient Pollution. National Academy Press, Washington, D.C.

Neumann, B., Vafeidis, A.T., Zimmermann, J., Nicholls, R.J. (2015). Future coastal
population growth and exposure to sea-level rise and coastal flooding—a global
assessment. PLOS ONE, 10(3): e0118571.

Nordstrom, K.F. (1992). Estuarine Beaches. Elsevier Applied Science, New York.

Nordstrom, K.F. (2014). Living with shore protection structures: A review. Estuarine,
Coastal and Shelf Science, 150, 11–23.

Penttila, D.E. (2007). Marine Forage Fishes in Puget Sound. Puget Sound Nearshore
Partnership Report U.S. Army Corps of Engineers, Seattle, Washington, pp. 23
Report No. 2007-03.

8

Pontee, N. (2013). Defining coastal squeeze: A discussion. Ocean & Coastal Management, 84,
204–207.

Perla, B. (2018). Personal communication. Vashon Nature Center, Vashon, Washington.

Perla, B.S., Metler, M.A. (2016). Bulkhead removal monitoring study in Maury Island Aquatic
Reserve. Quality Assurance Project Plan. Vashon Nature Center, Vashon, WA. 52 pp.

Puget Sound Partnership. (2012). 2012 State of the Sound: A Biennial Report on the Recovery of
Puget Sound. Puget Sound Partnership, Tacoma, Washington.

Rhodes, L.D., Durkin, C., Nance, S.L., Rice, C.A. (2006). Prevalence and analysis of
Renibacterium salmoninarum infection among juvenile Chinook salmon Oncorhynchus
tshawytscha in North Puget Sound. Diseases of Aquatic Organisms, 71:179-190.

Rice, C.A. (2006). Effects of shoreline modification on a northern Puget Sound beach:
microclimate and embryo mortality in surf smelt (Hypomesus pretiosus). Estuaries and
Coasts 29: 63-71.

Rice, C.A. (2003, April). Unpublished data from the NOAA/NMFS Northwest Fisheries Science
Center and Skagit System Cooperative. Presented at the Georgia Basin/Puget Sound
Research Conference, Vancouver, BC, Canada.

9

Romanuk, T.N., Levings, C.D. (2003). Associates between arthropods and the
supralittoral ecotone: dependence of aquatic and terrestrial taxa on riparian
vegetation. Environmental Entomology, 32: 1343-1353.

Ruiz, G.M., Hines, A.H., Posey, M.H. (1993). Shallow water as a refuge habitat for fish
and crustaceans in non-vegetated estuaries: an example from Chesapeake Bay.
Mar. Ecol. Prog. Ser. 99, 1-16.

Seitz, R.D., Lipcius, R.N., Olmstead, N.H., Seebo, M.S., Lambert, D.M. (2006).
Influence of shallow-water habitats and shoreline development on abundance,
biomass, and diversity of benthic prey and predators in Chesapeake Bay. Marine
Ecology Progress Series, 326:11-27.

Shipman, H. (2008). A Geomorphic Classification of Puget Sound Nearshore Landforms.
Puget Sound Nearshore Partnership Report No. 2008-01. Published by Seattle
District, U.S. Army Corps of Engineers, Seattle, Washington.

Shipman, H. (2010). The geomorphic setting of Puget Sound: implications for shoreline
erosion and the impacts of erosion control structures, in Shipman, H., Dethier,
M.N., Gelfenbaum, G., Fresh, K.L., and Dinicola, R.S., eds., 2010, Puget Sound
Shorelines and the Impacts of Armoring— Proceedings of a State of the Science

10

Workshop, May 2009: U.S. Geological Survey Scientific Investigations Report
2010-5254, p. 19-34.

Shoreline Monitoring Toolbox, Washington Sea Grant (2017).
https://sites.google.com/a/uw.edu/toolbox/

Simenstad, C.A., Fresh, K.L., Salo, E.O. (1982). The role of Puget Sound and Washington
coastal estuaries in the life history of Pacific salmon: an unappreciated function. Pages
343-364 in V.S. Kennedy, editors. Estuarine comparisons. Academic Press, New York.

Sobocinski, K.L. (2003). The Impact of Shoreline Armoring on Supratidal Beach Fauna of
Central Puget Sound. University of Washington School of Aquatic and Fishery Sciences.

Sobocinski, K.L., Cordell, J.R., & Simenstad, C.A. (2010). Effects of Shoreline Modifications on
Supratidal Macroinvertebrate Fauna on Puget Sound, Washington Beaches. Estuaries and
Coasts, 33(3), 699–711. http://doi.org/10.1007/s12237-009-9262-9.

Toft, J.D., Cordell, J.R., Heerhartz, S.M., Armbrust, E.A., Simenstad, C.A. (2010). Fish and
invertebrate response to shoreline armoring and restoration in Puget Sound, in Shipman,
H., Dethier, M.N., Gelfenbaum, G., Fresh, K.L., and Dinicola, R.S., eds. (2010). Puget
Sound Shorelines and the Impacts of Armoring—Proceedings of a State of the Science
Workshop, May 2009: U.S. Geological Survey Scientific Investigations Report 20105254, p. 161-170.

11

Toft, J.D., Cordell, J.R., Armbrust, E.A. (2014). Shoreline Armoring Impacts and Beach
Restoration Effectiveness Vary with Elevation. Northwest Science, 88(4), 367–375.

Toft, J.D., Cordell, J.R., Simenstad, C.A., Stamatiou, L.A. (2007). Fish Distribution,
Abundance, and Behavior along City Shoreline Types in Puget Sound. North
American Journal of Fisheries Management, 27(2), 465–480.

Toft, J.D., Ogston, A.S., Heerhartz, S.M., Cordell, J.R., Flemer, E.E. (2013). Ecological
response and physical stability of habitat enhancements along an urban armored
shoreline. Ecological Engineering, 57, 97–108.

Van Zwalenburg, Kim (2016, February 1). Personal interview on shoreline armoring.
Department of Ecology, Olympia, Washington.

Washington State Department of Fish and Wildlife (WDFW, 2016). Hydraulic Project
Approval (HPA): Frequently Asked Questions. Accessed February 20, 2016 at
http://wdfw.wa.gov/licensing/hpa/faq.html.

Weitkamp, D.E., G.R. Ruggerone, L. Sacha, J. Howell, B. Bachen. (2000). Factors
affecting Chinook populations. Parametrix, Natural Resource Consultants, and
Cedar River Associates, Report to City of Seattle, Washington.

12

Wobbrock, J.O., Findlater, L., Gergle, D., Higgins, J.J. (2011). The Aligned Rank
Transform for nonparametric factorial analyses using only ANOVA procedures.
Proceedings of the ACM Conference on Human Factors in Computing Systems
(CHI '11). Vancouver, British Columbia (May 7-12, 2011). New York: ACM
Press, pp. 143-146. Honorable Mention Paper.

13

Appendices
Appendix A. Map of Maury Island Aquatic Reserve

1

Appendix B. Shoreline Monitoring Toolbox vegetation sampling
protocol

2

Appendix C. Shoreline Monitoring Toolbox wrack sampling protocol

3

Appendix D. Shoreline Monitoring Toolbox insect sampling protocol

4

Appendix E. Shoreline Monitoring Toolbox fish protocol

5

Media

Miller_Thesis_2018.pdf