-
Title
-
Zostera marina and sea level rise: Estimating future habitat availability on armored shorelines in the Puget Sound.
-
Creator
-
Kingen, Greyson
-
Identifier
-
Thesis_MES_2023_KingenG
-
extracted text
-
ZOSTERA MARINA AND SEA LEVEL RISE:
ESTIMATING FUTURE HABITAT AVAILABILITY ON ARMORED AND UNARMORED
SHORELINES IN THE PUGET SOUND.
by
Greyson Kingen
A Thesis
Submitted in partial fulfillment
Of the requirements for the degree
Master of Environmental Studies
The Evergreen State College
June 2023
©2023 by Greyson Kingen. All rights reserved.
This Thesis for the Master of Environmental Studies Degree
by
Greyson Kingen
has been approved for
The Evergreen State College
by
_______________________________
John Kirkpatrick, Ph.D.
Member of Faculty
_______________________________
Date
ABSTRACT
Zostera marina and sea level rise:
Estimating future habitat availability on armored and unarmored shorelines in the Puget Sound
Greyson Kingen
Global sea level rise is a growing concern for coastal environments and poses significant risks to
many habitats and species. Zostera marina habitat in the Puget Sound, however, has been
predicted to expand in the event of sea level rise (SLR) due to the elevation range of its subtidal
habitat. Shoreline armoring is extensive along Puget Sound shorelines and occurs at many
locations where eelgrass is present. This study used geographic information system (GIS)
analysis of publicly available survey data and digital elevation models to summarize the spatial
extent of eelgrass and armoring in the Puget Sound. It then used site analysis and inundation
models of armored and unarmored site pairs in the Central Puget Sound. Armoring was found to
occur on 30% of shorelines where eelgrass has been observed. Individual study sites showed
considerable variability in characteristics and SLR projections. Habitat expansion was projected
by the year 2100 for 3 sites and losses were modeled at 1 site. Differences in habitat changes
between armored and unarmored site pairs were variable and did not suggest any patterns of
armoring impacts. Elevation profiles from each site indicate that slope and profile shape may be
a predictor of habitat loss or gain. Overall, the analysis showed that while total area of habitat
change at small scale locations is variable, the ideal depth range of eelgrass habitat will shift
closer to armored surfaces and their potential influences.
Table of Contents
List of Figures ................................................................................................................................ vi
List of Tables ................................................................................................................................ vii
Acknowledgements ...................................................................................................................... viii
Introduction ..................................................................................................................................... 1
Literature Review............................................................................................................................ 4
Ecology of Zostera marina ......................................................................................................... 4
Sea Level Rise ............................................................................................................................. 6
Causes and effects of sea level rise ......................................................................................... 6
Shoreline Armoring ..................................................................................................................... 9
What is shoreline Armoring? ................................................................................................... 9
Extent of Armoring in Puget Sound ...................................................................................... 10
Physical effects of Armoring ................................................................................................. 10
Z. marina habitat change and response to sea level rise ........................................................... 11
Projected future habitat area. ................................................................................................. 11
Accretion ............................................................................................................................... 13
Substrate ................................................................................................................................ 14
Slope ...................................................................................................................................... 15
Conclusion................................................................................................................................. 16
Methods......................................................................................................................................... 18
Data sources .............................................................................................................................. 18
Study area .................................................................................................................................. 20
Summary Statistics and Elevation Profiles ............................................................................... 21
Site Selection ............................................................................................................................. 22
ArcGIS Pro modeling ................................................................................................................ 25
SLAMM modeling .................................................................................................................... 27
Results ........................................................................................................................................... 29
ArcGIS Pro projections ............................................................................................................. 29
Profiles ...................................................................................................................................... 37
SLAMM .................................................................................................................................... 45
iv
Discussion ..................................................................................................................................... 48
Project limitations and Assumptions ......................................................................................... 51
Conclusion .................................................................................................................................... 53
Bibliography ................................................................................................................................. 55
Appendix A. .................................................................................................................................. 61
ArcGIS Pro projection maps ..................................................................................................... 61
Appendix B. ................................................................................................................................ 101
SLAMM Analysis Results ...................................................................................................... 101
v
List of Figures
Figure 1 Puget Sound study area. ................................................................................................. 21
Figure 2 Central Puget Sound Site pair selections. ....................................................................... 24
Figure 3 Central Puget sound study site pairs. .............................................................................. 25
Figure 4 CPS site 1 SLR Projections. ........................................................................................... 31
Figure 5 Percentage of habitat change CPS site 1. ....................................................................... 32
Figure 6 CPS site 2 SLR Projections ............................................................................................ 33
Figure 7 Percentage habitat change CPS site 2. ............................................................................ 33
Figure 8 CPS site 3 SLR Projections. ........................................................................................... 34
Figure 9 Percentage habitat change CPS site 3. ............................................................................ 35
Figure 10 CPS site 4 SLR Projections .......................................................................................... 36
Figure 11 Percentage habitat change CPS site 4. .......................................................................... 36
Figure 12 CPS Site 1 Armored site profiles.................................................................................. 40
Figure 13 CPS Site 1 Non-armored site profiles. ......................................................................... 40
Figure 14 CPS Site 2 Armored site profiles.................................................................................. 41
Figure 15 CPS Site 3 Non-armored site profiles. ......................................................................... 43
Figure 16 CPS Site 3 Armored site profiles.................................................................................. 43
Figure 17 CPS Site 4 Non-armored site profiles. ......................................................................... 44
Figure 18 CPS Site 4 Armored site profiles.................................................................................. 45
Figure 19 Map results from SLAMM analysis of CPS site 1 ....................................................... 47
vi
List of Tables
Table 1 IPCC scenario SLR values ............................................................................................... 27
Table 2 Tidal datums and corrections. .......................................................................................... 27
Table 3 Summary Statistics Results. ............................................................................................. 29
Table 4 Results from ArcGIS Pro SLR raster analysis. ................................................................ 30
Table 5 Profile characteristics of site transects. ............................................................................ 38
vii
Acknowledgements
I would like to thank Mike Ruth for introducing me to the world of maps and spatial analysis and
teaching me everything I know about GIS. Thank you to my thesis reader, John Kirkpatrick, for
your encouragement and advice through this process; after every meeting with you I felt less
stressed and more confident in my work. Thank you, Janis, for spending so many hours caring
for my son while I worked, I could not have done this without your help. Thank you to Roger for
the enduring words “Isn’t there something you should be doing?”. And finally, I would like to
thank my wife Rhonda, who always gave me the encouragement, support, and the motivation to
continue.
viii
Introduction
The most recent Intergovernmental Panel on Climate Change (IPCC) report projections
show that there will likely be a between 0.43 m and 0.84 m rise in global mean sea level (GMSL)
by the year 2100 (Oppenheimer et al., 2019). Some projections are over 1m rise in GMSL
(Oppenheimer et al., 2019). Many studies in the Puget Sound have attempted to model or
quantify affects from sea level rise (SLR) on nearshore and intertidal habitats and have projected
massive losses of some habitat types (Glick et al., 2007). In contrast to these losses of habitat,
potential eelgrass habitat has been modeled to expand with sea level rise in some cases (Kairis &
Rybczyk, 2009; Smith & Liedtke, 2022).
The eelgrass Zostera marina is a subtidal seagrass species that provides essential habitat
in the Puget Sound for many species including salmonids. Specifically, juvenile salmon as well
as feeder fish use this habitat to forage for invertebrates during vulnerable transitional stages
(Kennedy et al., 2018). Eelgrass is also considered by the Washington State Department of
Natural Resources to be a vital sign of the health of Puget Sound waters because of its sensitivity
to environmental pressures like temperature, light availability and physical disturbance
(Christiaen et al., 2022). Despite its sensitivity, healthy eelgrass can also provide significant
ecosystem services such as carbon storage, reduction of sediment resuspension, and mitigation of
ocean acidification (Christiaen et al., 2022).
Due to its significance in the ecosystem, Z. marina is well studied from a variety of
angles. However, many studies of SLR focus on intertidal species rather than subtidal nearshore
species as they are less at risk from inundation. The models that are widely used to predict these
changes rely on the assumption of conversion of nearshore habitat on a basis of elevation. In
1
some cases, though, these models generalize or omit erosion, accretion data and the effects of
shoreline armoring on these processes (Kairis & Rybczyk, 2009; Smith & Liedtke, 2022).
Shoreline armoring of various types exist on about 30% of Puget sound shorelines
(Morley et al., 2012). These structures can form a barrier between sediment sources to shorelines
and reflect wave energy, which can lead to increasing beach sediment size and lowering beach
elevation (Smith & Liedtke, 2022; Thom & Williams, 2001). Studies of SLR that discuss
shoreline armoring recognize these structures as a physical barrier to landward migration of
species. While armoring does present a barrier, it also is recognized to alter shoreline substrate,
elevation, and slope. Nonetheless, the existing studies of eelgrass habitat response to SLR do not
appear to address shoreline armoring or its affects to habitat quality (Kairis & Rybczyk, 2009).
Shoreline armoring is widespread in the Puget Sound and efforts are taking place to
remove barriers and restore natural shorelines. However, with the increasing threat of rising
water, landowners may decide to place more barriers to protect their assets (Smith & Liedtke,
2022). Understanding the extent of the environmental risk from shoreline armoring is important
for conservationists and decision makers considering removal or construction of barriers as well
as locations for eelgrass preservation and restoration efforts.
The purpose of this study was to find to what extent shoreward migration of eelgrass in
response to sea level rise will be affected by shoreline armoring in the Puget Sound. The study
used GIS spatial analysis and the open source Sea Level Affecting Marshes Model (SLAMM) to
examine armored shorelines near eelgrass habitat (Clough et al., 2016). The extent of eelgrass
habitat in Puget Sound that may be affected by shoreline armoring was quantified, and several
site pairs were selected for more focused analysis. Site pairs of armored and unarmored beaches
1-5 km in length were chosen with the methods of Dethier et al. (2016) by attempting to match
2
geomorphic and bathymetric characteristics, location in the same drift cell and close proximity of
the pair member (Dethier et al., 2016). At each site elevation profiles measure were created to
measure beach slope, width, and armor elevation before raster analysis was used to estimate
habitat change under three SLR scenarios. These spatial measurements and elevation analyses
provide a preliminary exploration of the spatial relationship between shoreline armoring and
Zostera marina communities and may serve as a basis for more robust studies.
3
Literature Review
The topics of sea level rise, shoreline armoring, and the seagrass Zostera marina are each
well studied with a wealth of literature spanning decades. An overview of the current knowledge
about them is necessary to support any research into their interactions. This literature review will
provide a basic background of each subject before detailing the intersecting literature and its
implications. First the review will introduce Zostera marina and its significance in the Puget
Sound region. Next the problem of sea level rise, its effects, and current predictions will be
outlined. Thirdly, shoreline armoring and its effects will be described, as well as its extent in
Puget Sound. The fourth section will bring the three previous topics together in a review of what
is known specifically about how eelgrass may be affected by Sea Level Rise in the presence of
an armored shoreline. The last section will be a brief conclusion that will describe gaps in the
literature and where my study could add knowledge for the benefit of conservationists, shoreline
habitat managers, and future researchers.
Ecology of Zostera marina
Zostera marina, or eelgrass, is one of the most widespread native seagrasses in the Puget
Sound. It is a monecious flowering plant belonging to the Family Zosteraceae (Hitchcock &
Cronquist, 2018). They have perennial rhizomes, flowering stems, and vegetative leaves that are
around 80 cm in length, though morphology can vary greatly depending on physical and
environmental conditions (Moore & Short, 2006). Z. marina grows in lower intertidal to subtidal
nearshore habitats and in Puget Sound it can be found everywhere but in the southernmost inlets
(Christiaen et al., 2022). Generally, Z. marina grows in beds that are unoccupied by other
seagrass species, but in some cases the non-native Nanozostera japonica (commonly known as
4
Zostera japonica) overlaps its habitat in and mixed stands or a mosaic of bed patches (Hannam,
2013; Hitchcock & Cronquist, 2018).
Eelgrass can be found on fine, soft substrates like sand and muddy sediments of beaches
and tidal flats. The primary limiting factors of its distribution are substrate, light availability, and
desiccation (Hannam et al., 2015; Moore & Short, 2006). At the lower end of its depth range it is
limited by light penetration and at the upper end by exposure and desiccation during low tides
(Christiaen et al., 2022; Hannam, 2013; Hannam et al., 2015).
Eelgrass habitat depth distribution can vary between locations and shoreline types. On
flat shorelines eelgrass depth range can be shallower than at steeper fringing beds, but eelgrass at
fringe beds can also grow at deeper depths (Hannam et al., 2015). Washington State Department
of Natural Resources Submerged Vegetation Monitoring Program (SVMP) has found that in
Puget Sound, Z. marina occurs between +1.4 m and -12.0 m MLLW with an preferred depth
range of 0 to -2 m MLLW (Hannam et al., 2015). They calculate that over 60% of eelgrass is
subtidal, or 1 meter below MLLW (Hannam et al., 2015).
Z. marina has several ecosystem services that it provides to the Puget Sound region. It is
an essential habitat for fish and provides both cover and foraging opportunities. Several species
of salmon and beach spawning forage fish use eelgrass extensively in their juvenile stages as
they transition from freshwater to marine habitats (Dumbauld et al., 2015). Due to its importance
to juveniles of many species, eelgrass habitat is sometimes referred to as a nursery habitat
(Moore & Short, 2006). Eelgrass beds increase beach complexity by providing habitat and
substrate for benthic invertebrates and epiphytes, which in turn provide food for fish and
waterfowl (Duffy et al., 2005; Moore & Short, 2006).
5
Eelgrass is highly productive, and its biomass contributes to carbon storage through
deposition in marine sediment. They also provide beach stabilization with their root systems and
increased water clarity by decreasing resuspension of sediment (Christiaen et al., 2022; Moore &
Short, 2006). Eelgrass has even been recognized to benefit water quality by limiting harmful
algae and bacteria as well as provide some mitigation for ocean acidification (Christiaen et al.,
2022; Pacella et al., 2018). In contrast to these abilities to mitigate climate change, eelgrass can
be very sensitive to certain environmental influences like increasing water temperature and
elevated nutrient input. Eutrophic conditions can affect the seagrass by decreasing light
availability through phytoplankton blooms and algal overgrowth (Christiaen et al., 2022; Moore
& Short, 2006). Because of its widespread distribution and significance in Puget Sound, Zostera
marina is used as an indicator of ecosystem health. It is monitored closely by the state of
Washington and is the subject many studies and restoration efforts. (Cereghino et al., 2012;
Christiaen et al., 2022).
Sea Level Rise
Causes and effects of sea level rise
Sea level rise is a climate change phenomenon that is a cause of concern for coastal
regions around the world. The Global Mean Sea Level (GMSL) is increasing and has accelerated
from 1.4 mm per year during 1901-1990 to 3.6 mm per year between 2006-2015 (Oppenheimer
et al., 2019). Rising temperatures from anthropogenic climate change have shifted the balance of
multiple interconnected hydrologic processes that control the flow and storage of water around
the globe. The warming atmosphere causes both thermal expansion of the ocean water as well as
the loss of water stored in the form of ice. Thermal expansion occurs as rising temperature
decreases the density of the water, resulting in a greater volume without any increase in mass
6
(National Research Council et al., 2012; Oppenheimer et al., 2019). Yet, ocean mass is also
increasing due to the addition of stored water. Most of the fresh water on the earth is stored in the
Antarctic and Greenland ice sheets. These ice sheets can increase sea level through sub surface
melting, loss of ice at marine edges, and loss of surface mass from ablation (National Research
Council et al., 2012; Oppenheimer et al., 2019). Like the ice sheets, glaciers also contribute to
sea level through melting of their stored water and have, to date, added more mass to the ocean
than both the Antarctic and Greenland ice sheets (Oppenheimer et al., 2019). However, the total
mass of glacial water storage is only a fraction of the ice sheets capacity to add to sea level rise
(National Research Council et al., 2012; Oppenheimer et al., 2019).
Climate induced SLR will expose coastal and nearshore habitats to a variety of effects
including inundation, salinization, erosion, and more frequent extreme sea level events (Glick et
al., 2007; Miller et al., 2019; Oppenheimer et al., 2019). As water levels rise, habitats can
transition to new types, are forced to migrate landward, or are converted to open water. Inland
habitat that is not inundated may also be influenced as groundwater can become salinated and the
water table can rise. Subtidal habitat may be affected by reduced light availability and erosion. In
some cases sedimentation and accretion may accumulate quickly enough for nearshore habitats
to match rising water levels, but as rates of SLR are increasing, this process may be outpaced
(Fagherazzi et al., 2020; Moritsch et al., 2022; Poppe & Rybczyk, 2022). There is still a great
deal of uncertainty in how nearshore vegetation will respond to lateral and horizontal migration
(Fagherazzi et al., 2020). Migration of coastal habitats is a natural process in response to changes
in water level and shoreline conditions, but the rapid pace of SLR and anthropogenic
modifications of the coastlines may restrict this ability (Glick et al., 2007; Oppenheimer et al.,
2019; Thom & Williams, 2001).
7
The most recent IPCC report projections show that there will likely be between a 0.43 m
(low emission scenario RCP2.6) and 0.84 m (high emission scenario RCP8.5) rise in Global
Mean Sea Level by the year 2100 (Oppenheimer et al., 2019). However, Sea level will not occur
proportionately around the globe due to geodynamic processes. The shifting distribution of water
to the ocean from storage in ice and on land, will influence the gravity, shape, and rotation of the
earth. These variations will cause disproportionate sea level changes at the regional scale
(Oppenheimer et al., 2019). Geologic processes like uplift and subsidence can also affect sea
level relative to the land at the regional level. Human activity can also influence regional sea
level by causing Anthropogenic subsidence through extraction of groundwater and hydrocarbon
(Candela & Koster, 2022; Miller et al., 2019; Oppenheimer et al., 2019). While GMSL is a good
measure for explaining sea level rise in general, localized studies are necessary for predicting
effects of SLR at the regional and local levels (Miller et al., 2019). The IPCC estimates that
regional variation can be ±30% around the global mean, with greater than 30% departure
possible in regions with rapid vertical movement of the land (Oppenheimer et al., 2019).
Regional calculations of sea level rise have both Absolute Sea Level (ASL) and Relative Sea
Level (RSL) to consider. ASL is the average height of the ocean as compared to a fixed baseline
like the center of the earth, while RSL is the average height of the ocean compared to a fixed
point on the land (Miller et al., 2019). Washington State may have local variations in absolute
sea level rise of around 10cm by 2100, but there is greater potential for variation due to vertical
uplift and subsidence (Miller et al., 2019). Regional rates of vertical land movement for the
entire state have been estimated at about +0.10 cm per year, but more localized studies have
shown that there are areas in the Central Puget Sound that may experience subsidence by the end
of the century (Miller et al., 2019; National Research Council et al., 2012). The Washington
8
Coast Resilience Project collected and modeled data from 171 locations around the state,
allowing them to show these more local projections (Miller et al., 2019). For example, they
project RSL at an inner Puget Sound location (Tacoma) at between +2.1 ft (low emission
scenario RCP4.5 central estimate) and +2.5 ft (high emission scenario RCP8.5 central estimate),
and a Coastal location (Neah bay) at between +0.5 ft (RCP4.5 central estimate) and 1.0 ft
(RCP8.5 central estimate) (Miller et al., 2019).
Shoreline Armoring
What is shoreline armoring?
As coastlines are developed, modifications are made to the shoreline for several purposes
including controlling wave energy and stabilizing the shore. Armoring is a method of shoreline
stabilization used to protect land from natural physical processes like erosion, inundation, and
storm surge. Typically armoring is placed to protect structures and development in upland areas
that may be affected by erosion or flooding.
There are many different forms of armoring that are used depending on the location and
the level of protection intended. Some forms of armoring like sea walls create barriers against the
water with concrete that are impermeable and very reflective of wave energy (Shipman et al.,
2010; Thom & Williams, 2001). Similarly, bulkheads are wall like structures made of concrete
or wood that both hold back erosion from the land and protect from waves. Another common
form of armoring is called revetment. This style is made of thick layers of permeable stones that
are commonly deposited from the upper shore at the high water line all the way down to the low
water line (Thom & Williams, 2001). Revetments are generally more simple to construct and can
follow the natural contour of the shoreline. Riprap is the most common type of revetment and is
9
composed of random stone rubble, but there are other types formed from interlocking concrete or
metal cages filled with stone. (Shipman et al., 2010; Thom & Williams, 2001).
Extent of Armoring in Puget Sound
Puget Sound is one of the largest estuaries in the country and is extensively armored
along its shorelines. Around a third of the approximate 4000 km of estuarine coast has some
form of armoring due to the urbanization and industrialization of the region (Shipman et al.,
2010). A review in 2001 reported that 1.7 miles of shoreline are armored per year in the Puget
Sound (Thom & Williams, 2001). The expanding urbanization of Puget Sound in combination
with rising sea levels has caused concern that armoring will continue despite the potential
ecological harm that it may cause (Dethier et al., 2016, 2017; Smith & Liedtke, 2022).
Restoration and removal of armored shore is becoming more widespread but there is also
evidence that new armoring is often being placed without permitting or proper adherence to
regulatory standards (Dethier et al., 2017; Kinney et al., 2015).
Physical effects of Armoring
The intended purpose of armoring is to interrupt the natural progression of physical
processes like erosion, but there are consequences of this interruption. Impoundment blocks the
supply of sediment to the fronting shore and possibly to other shorelines connected by currents
and drift cells (Simenstad et al., 2011; Thom & Williams, 2001). Armoring can lead to erosion of
the beach surface if its sediment supply is blocked, and longshore transport continues to move
material from the shore (Shipman et al., 2010). The hard surfaces of armoring structures reflect
wave energy away from the shore or back onto the beach surface. Reflected wave energy can
resuspend sediments and allow currents to draw them away from the beach in a process called
scour, resulting in a coarser substrate and lowered beach elevation. As the wave energy is
10
redirected, it may also cause scouring effects on adjacent shores (Shipman et al., 2010). Scour
and impoundment at armored locations results in vertical but not horizontal erosion, which can
lead to narrowing of the beach and steepening of its slope (Shipman et al., 2010)
The severity and speed of sediment supply changes may vary greatly by location and
extent of armoring. A study by Dethier et al. in 2016 found that differences in sediment size
between armored and unarmored beaches were difficult to distinguish at the local scale, but were
significant at the regional scale (Dethier et al., 2016). They were able to conclude that drift cells
that were extensively armored had larger sediment grain size. This may confirm the theory of
Thom and Williams that there are thresholds of drift cell armoring extent beyond which sediment
supply loss becomes significant (Thom & Williams, 2001). Both Thom and Williams and
Dethier et al. suggest that there are likely cumulative effects of extensive armoring at the larger
regional and temporal scales (Dethier et al., 2016; Thom & Williams, 2001).
Z. marina habitat change and response to sea level rise.
Projected future habitat area.
Unlike many habitats on Puget Sound shorelines, eelgrass habitat has been projected to
expand as sea level rise occurs. This is to be expected if we consider that loss of terrestrial
habitat from inundation results in a corresponding expansion of marine habitat. Some of the most
concerning effects of SLR (including inundation, salinization, and storm surge) are not a
significant issue for Z. marina due to its subtidal habitat range. As a result of its depth range,
eelgrass habitat change is not often the focus of modeling. For example, the National Wildlife
Federation used the SLAMM model to estimate habitat changes over 10 large areas of the Pacific
Northwest, but the only habitat type in the model that covers eelgrass is “estuarine open water”.
The study showed an overall expansion of estuarine open water in all scenarios, but this is an
11
imprecise measure when considering that only a fraction of the open water depth profile fits
eelgrass’s narrow elevation range of 1 to -14 m MLLW (Glick et al., 2007).
Smith and Liedke also used the SLAMM model in their study and overlaid eelgrass
habitat data to make up for the lack of corresponding habitat type in the model. At their site they
projected potential eelgrass habitat gains of over 10% for a 0.4 m SLR scenario and over 22% for
a 1 m scenario by the year 2100 (Smith & Liedtke, 2022). They noted that this habitat expansion
was contrary to their initial hypothesis, but also acknowledged the many unaccounted variables
of erosion, accretion, and changes in wave energy (Smith & Liedtke, 2022).
In 2009, Kairis & Rybczyk used a relative elevation model to project eelgrass
productivity and coverage in Padilla Bay. Their results also predicted overall increases in habitat
by 2100 in all but the highest of eight SLR scenarios (Kairis & Rybczyk, 2009). They do note
that the large buffer of flats surrounding existing eelgrass habitat is a major factor in its ability to
expand unchecked in their model (Kairis & Rybczyk, 2009). This may suggest that bays and
locations with gently sloping beaches will have more resilient eelgrass habitat. Another
consideration that the authors note is that the ability of eelgrass to migrate laterally is slower than
the predicted rate of expansion (Kairis & Rybczyk, 2009). However, eelgrass can spread over
large distances through seed dispersal and spread of reproductive shoots, so is possible that
lateral spreading rate may not present an issue (Kairis & Rybczyk, 2009; Moore & Short, 2006).
Each of these projections of sea level rise account for a variety of variables to predict
water depth and even changes in shore elevations, but some of these variables can be influenced
by shoreline armoring in ways that might affect predictions of eelgrass habitat change. Some of
these variables include accretion, substrate, and slope. Accretion is the process of material
deposition that builds and expands shorelines and as a result changes depth profiles and habitat
12
distribution. Substrate refers to the material composition and size of the shore’s surface, which
can determine its suitability as habitat. Slope simply describes the angle of the beach surface,
which can affect exposure to wave energy and determines the width of the available habitat
along the shoreline. Changes in each of these variables are difficult to accurately predict, but
they are known to be vulnerable to shoreline armoring as well as important for eelgrass habitat
suitability.
Accretion
The geomorphic processes that form beaches in the Puget Sound play an important role in
determining how eelgrass habitat is distributed along shorelines. Finlayson (2006) characterized
the majority of Puget Sound beaches as low energy with a composite profile consisting of a steep
foreshore and a low-tide terrace. They are typically mixed sediment beaches with a distinct
transition of sediment coarseness between the foreshore and the terrace (Finlayson, 2006).
However, variations in the profiles and sediment characteristics between locations are variable
with some uncertainty in which oceanographic and geomorphic factors are most important in
their formation (Finlayson, 2006).
The Puget Sound has a unique tidal climate with a diurnal pattern and wide tidal range.
This pattern focuses tidal and wave energy on the upper foreshore. The concentration of energy
results in the majority of sediment transport and elevation changes occurring in the upper shore
with very little on the terrace and the subtidal zone (Finlayson, 2006)
Fringing eelgrass beds may be more susceptible to SLR because they may lack the direct
sediment input that delta beds receive to vertically shift their habitat (Dethier et al., 2017;
Finlayson, 2006). These areas rely on erosion and longshore transport of material to create
13
suitable fine substrate for eelgrass habitat. As many of the shorelines in Puget Sound have some
sort of armoring, erosion and the natural geomorphic process of beach creation will be impeded.
It is possible that near deltas of freshwater input some effects of SLR could be
counteracted or slowed to a degree by sediment deposition (Fagherazzi et al., 2020; Poppe &
Rybczyk, 2022). If the sediment supply remains high enough, the beach surface might increase
in height along with the increasing depth. In 2009 a study of Padilla Bay in Puget Sound by
Kairis and Rybczyk found that accretion rates in that location would not keep pace with SLR
(Kairis & Rybczyk, 2009). Also, a more recent study that modeled sediment deposition on
eelgrass beds in Padilla bay suggested that there would likely be negative affects to eelgrass if
sediment concentrations were increased to the level required to match SLR (Poppe & Rybczyk,
2022). They found that almost four times the current suspended sediment concentration would be
needed to keep pace with SLR. That level of suspended sediment could reduce light penetration
and inhibit eelgrass growth potential (Poppe & Rybczyk, 2022). However, eelgrass was modeled
to significantly expand its habitat shoreward in Padilla bay despite the lack of accretion (Kairis
& Rybczyk, 2009).
Substrate
Accretion rates and sediment concentrations not only affect beach elevation, but the
composition of the beach surface material. This material, or substrate, is very important when
determining habitat suitability for seagrasses. Due to shoreline armoring and the accelerated
speed of climate induced sea level rise, the substrate at the ideal tidal depth for eelgrass may not
remain suitable as the range shifts landward. Shoreline armoring can reduce the deposition of
fine materials on beaches by blocking material input from the land which may result in more
coarse and rocky substrates (Dethier et al., 2017; Finlayson, 2006; Shipman et al., 2010). The
14
severity of armoring impacts like coarsening of beach sediments is thought to be dependent on
the location of the structure along the beach profile (Shipman et al., 2010). Rising sea level will
bring the preferred habitat depth range closer to barriers and increase the risk of coarsening
substrate. However, substrate grain size is variable between locations and is dependent on a
variety of factors that influence accretion that are difficult to predict.
In a study on Bainbridge Island, Smith and Liedtke found that upper beach mean
sediment size was smaller on an armored site as compared to an adjacent unarmored site, though
the low wide terrace was very similar between the two (Smith & Liedtke, 2022). They concluded
that the difference in mean grain size did likely indicate effects from shoreline armoring like
scour and wave reflection. In that study they modeled increases of eelgrass habitat in all three
SLR scenarios that they used, but they recognized that physical processes like accretion and
scour were unaccounted for in the model, and that the unarmored site demonstrated greater
plasticity (Smith & Liedtke, 2022).
Slope
Changes in variables like accretion and substrate can often be very difficult to measure or
predict depending on the study scale or timeline, but slope can be related to both and can be
easier to use as a predictor of accretion, erosion, and substrate. On armored shorelines there is
also the possibility of erosion resulting from increasing storm surge and wave energy caused by
SLR, as well as armoring induced wave scour and increased wave energy (Finlayson, 2006;
Shipman et al., 2010). As is the intention of armoring, this erosion does not widen the beach
landwards, but occurs on the beach surface. This process results in a steeper beach slope.
Increasing depth can also result in greater wave energy and sediment resuspension that could
15
decrease light penetration and reduce photosynthetic potential (Kairis & Rybczyk, 2009; Moore
& Short, 2006).
The studies that model eelgrass habitat expansion tend to do so at very flat sites like
Padilla Bay because they contain a large percentage of the eelgrass beds in the Puget sound.
Sound wide surveys have found that around 50% of eelgrass is found at locations with “Flats”
type profiles like Padilla bay in the Northern Puget sound, and the rest is located on narrower
“Fringe” sites (Christiaen et al., 2019). At Padilla bay and similar delta sites there is very little
change in slope between current and projected habitat (Kairis & Rybczyk, 2009; Poppe &
Rybczyk, 2022). However, at fringing sites where eelgrass beds are smaller, there has been less
modeling. At these locations there may be a more drastic transition from the low wide terrace to
the steeper slope of the foreshore.
Steeper slopes may not necessarily be a limiting factor for eelgrass habitat, but they are
associated with more coarse substrate and higher wave energy (Finlayson, 2006; Moore & Short,
2006). A greater slope will also decrease the overall surface area of available habitat in Z.
marina’s preferred depth range, though this effect may be negligible depending on the degree of
slope change and the scale of study.
Conclusion
Eelgrass is an essential species for Puget Sound habitats and an important indicator of
ecosystem health. Current literature and modeling suggest that SLR may increase available
habitat area, yet complex variables like accretion, erosion, and sediment supply add uncertainty
to these predictions. Shoreline Armoring exacerbates this uncertainty by its influence on these
geomorphic variables. Few studies attempt to address the effects of shoreline armoring on
16
eelgrass habitat. My study will attempt to better understand future eelgrass habitat availability by
linking the science of SLR modeling to the developing research of armored shores.
17
Methods
The goal of this project was to estimate the amount of Z. marina habitat that may be
influenced by shoreline armoring, model potential future habitat availability, and compare SLR
models at armored and unarmored habitat locations. The following methods section covers the
project data sources, study region, and analysis techniques. The methods described show the
steps that were taken to obtain eelgrass and armoring coverage statistics, select study sites,
collect site characteristics, and estimate habitat change with SLR inundation models.
Data sources
The primary spatial data sources for this research project were obtained from several
sources including the Puget Sound Nearshore Ecosystem Restoration Project (PSNERP), the
WSDNR Submerged Vegetation Monitoring Program (SVMP), ShoreZone, and the United
States Geological Survey (USGS).
Data from the PSNERP was downloaded in a geodatabase (GDB) file structure which
contained all of the vector data layers and tables used in the PSNERP comprehensive change
analysis published in 2011. (Simenstad et al.). The data type and quality that the PSNERP used
to create this geodatabase ranges widely, but only two layers were used in this study. Drift cell
line features were used in the selection of study sites. Armoring line features from this dataset
were adapted from the ShoreZone study as well as other sources with a date range between 1994
and 2008 (Anchor QEA, 2009).
SVMP data was also obtained in a GDB that contains all of the programs survey data
from the years 2000 to 2020 (Christiaen et al., 2022). These layers include study site polygons,
18
generalized eelgrass polygons, and survey transects. These layers were used for the estimation of
eelgrass presence and site selection.
ShoreZone inventory data was also used to fill in gaps of unsurveyed shoreline from the
other sources when necessary. As the survey was completed between 1994-2000, the Shorezone
data is relatively old and may not be reliable for many uses due to the collection methods. This
data was collected via aerial videography and manually interpreted to create a comprehensive
survey of shoreline characteristics. The high potential for human error in this method makes it
necessary to treat it with a degree of uncertainty (Berry et al., 2001). Despite these uncertainties,
it is in some cases the most comprehensive and complete survey of Puget Sound shoreline
characteristics that is available at this time. The SVMP has not yet collected data on all the Puget
Sound shoreline and so this dataset became useful in the estimation of eelgrass coverage and in
study site selection.
Elevation data was obtained from the US Geological Survey and NOAA. The USGS has
created an integrated topobathymetric digital elevation model (TBDEM) with 1 meter resolution
for their Coastal National Elevation Database (CoNED). This elevation model combines data
from 186 sources into a single model that prioritized the most recent and accurate data available
(OCM Partners, 2023).
All data was downloaded to ArcGIS Pro v. 3.0 for exploration, analysis, and data
management. All layers used or were transformed to the projected coordinate system UTM 1983
Zone 10 with a vertical coordinate system of NAVD88. Horizontal and vertical units were set to
meters.
19
Study area
The study area was adapted from the study area of the Submerged Vegetation Monitoring
Program (Christiaen et al., 2022). The study area of the entire Puget Sound region extends from
the United States-Canada border in the North, to the full southern extent of the Puget Sound. It
also encompasses the San Juan Islands and the US shoreline along the strait of Juan De Fuca to
the end of Cape Flattery.
The Puget Sound region was divided into subregions for the selection of small scale site
pairs. The South Puget Sound (SPS) includes the basin and all the inlets south of the Tacoma
Narrows (Figure 1). The Central Puget Sound (CPS) extends from the Tacoma Narrows to the
opening of Admiralty Inlet into the Strait of Juan de Fuca. North Puget Sound (NPS) includes
Whidbey Basin, the US coastline of the Salish sea North of Admiralty Inlet. The regions
including the San Juan Islands and the stretch of the Strait of Juan de Fuca from the Pacific
Ocean to Admiral were excluded from use in the small scale site analysis due to their difference
in shoreline characteristics, presence of eelgrass and armoring, as well as the limitations of time.
20
Figure 1: Puget Sound study area. Summary statistics include all areas within all regions.
Summary Statistics and Elevation Profiles
Summary statistics of the extent of eelgrass and armoring in the PS region were collected
using vector data from the SVMP and the PSNERP. Line features were used to calculate length
of shoreline segments that were reported to have eelgrass presence. The line features were
manually edited with the trimming tool to only extend to the boundary of the study area
polygons. The SVMP dataset provided shoreline segments with vegetation codes corresponding
to Zostera marina habitat, but some locations have not been surveyed and have a nodata value
21
that cannot be assumed to have no Z. marina presence. To supplement this data set on shore
segments with no data, the PSNERP eelgrass data which is derived from the Shorezone Project
was used to estimate eelgrass extent.
Armoring presence was obtained from PSNERP as a line feature attribute. The armoring
attribute is only a Yes/No option assigned to shoreline segments. The PSNERP used Shorezone
data to calculate this attribute and the “Yes” value was only assigned to line segments with 50%
or more armoring coverage (Anchor QEA, 2009; Simenstad et al., 2011).
Shoreline profiles were created with ArcGIS Pro using the CoNED DEM as the elevation
source. Line features of 4 transects per site pair were drawn from 6 m to -2 m MLLW at sites 1
and 3 and from 6 m to 0.5 m MLLW at site 2. Contours line features were generated at each
required elevation to ensure accurate snapping of transect endpoints. The lines were manually
drawn at an approximate 150 to 200 m interval and an approximate right angle from the
shoreline. The line vertices were densified to 0.5 m intervals and elevation values were
interpolated from the MLLW corrected DEM. Profile charts were then created from each set of
line features. Additional statistics like slope and distances were measured along the transect lines
and elevation contours with exploratory analysis tools and recorded in Excel.
Site Selection
Smaller scale sites were chosen out of these regions in order to characterize the eelgrass
habitat in the Puget Sound. Due to the restriction of time and the narrow selection parameters,
only 4 sites in the Central Puget Sound region were able to be used in the study. Site pairs of
armored and unarmored beaches 2 km or less in length that have evidence of Z. marina presence
were selected with the criteria used by Dethier et al. (2016) by attempting to match geomorphic
22
and bathymetric characteristics, location in the same drift cell and close proximity of the pair
member.
Sites selection used the eelgrass and armoring line features to select study site polygons
from the SVMP database. In each region sites were filtered by the presence of eelgrass and
armoring, then filtered by their proximity to a site with eelgrass but no armor. These remaining
site pairs were selected by their occurrence in the same drift cell and drift direction, then they
were manually filtered for similar direction of shore face. Site pairs that occurred in different
drift cells or were in differing zones of the same drift cell, for example convergence zones or
divergence zones, were discarded. Pairs that had shorelines oriented in significantly differing
directions were also discarded from the selections. Site 1 is located on Maury Island at the
coordinates 122.400°W 47.400°N. Site 2 is located at Magnolia Bluff, Seattle at the coordinates
122.427°W 47.654°N. Site 3 is located near Brownsville at the coordinates 122.613°W
47.667°N. Site 4 is on Ledgewood Beach, Whidbey Island at the coordinates 122.608°W
48.146°N (Figure 2).
23
Figure 2: Central Puget Sound Site pair selections.
24
Figure 3: Central Puget sound study site pair locations as indicated by the blue stars.
ArcGIS Pro modeling
Estimating eelgrass habitat change was completed with elevation analysis using raster
datasets. ArcGIS Pro was used to perform raster analysis of digital elevation models and create
inundation models of preferred Z. marina habitat depth range. The CoNED elevation model was
25
used as the source data for these models. Local NOAA tidal datums, IPCC projected SLR data,
and habitat ranges defined by the SVMP were used to calculate potential Z. marina habitat
coverage layers (Hannam et al., 2015; Oppenheimer et al., 2019).
The raster calculator geoprocessing tool was used to correct the DEM from NAVD88 to
MLLW so that the elevation model would correspond to the Z. marina habitat range values.
Datum correction values were obtained from the Tacoma and Seattle NOAA tidal gage stations
(National Oceanic and Atmospheric Administration, 2020). The DEM raster was then processed
with the raster calculator by separately adding each scenario’s sea level rise value to simulate the
change in MLLW by 2100. The DEM collection date was 2015 and the IPCC scenario base year
was in 2005, so a correction was also applied to each scenario SLR value to account for sea level
rise that had already occurred between 2005 and 2015. The correction calculation used 2.07 mm
per year from the mean annual sea level rise over the past century reported by local NOAA tidal
gauge data (Table 1).
The cells within the depth ranges of eelgrass habitat were selected from each corrected
DEM with the raster calculator to estimate habitat coverage in each scenario. Calculations were
run with the RCP 2.6, 4.5, and 8.5 scenarios as well as the base MLLW datum corrected DEM.
Elevation range for Z. marina habitat was set at the preferred depth of between 0 m and -2 m
MLLW as described by the SVMP (Hannam et al., 2015). To separately calculate area for the
armored and unarmored site pairs, the Extract by Mask geoprocessing tool was used to select the
raster cells within their respective site polygons. The source DEM cell size was 1 meter so area
could be calculated with cell count values obtained from attribute tables. Resulting area values
were input into Excel for data management and calculation of percent change.
26
Table 1: IPCC scenario projected SLR values and date corrected values
IPCC Scenario
RCP 2.6
RCP 4.5
RCP 8.5
SLR (m)
SLRcorr (m)
0.43
0.4093
0.55
0.5293
0.84
0.8193
SLAMM modeling
The Sea Level Affecting Marshes Model (SLAMM) was used to model SLR and habitat
change with an erosion model component. The program required an elevation model, slope layer,
and land cover class layer in the numeric SLAMM format. Each of these layers was required to
be in the ASCII raster format with the identical location, dimension, and projected coordinate
system (Clough et al., 2016). The DEM at each site pair was clipped to a standardized extent that
encompassed both the armored and unarmored site and a slope layer was generated. The Raster
Calculator was used to apply a vertical datum correction and define 0 elevation as the local Mean
Tide Level as defined by the local NOAA tidal gauge (National Oceanic and Atmospheric
Administration, 2020). The Tacoma and Seattle tidal gauges were used for sites in the CPS
region (Table 2).
Table 2: Tidal datums and corrections using nearest NOAA tidal gauge.
Location
MTL
NAVD88
NAVD88corr
Sites
Tacoma
2.094
0.729
1.365 CPS 1
Seattle
2.032
0.715
1.317 CPS 2, 3 and 4
To fulfil the requirement for land cover class layers a Washington State National
Wetlands Index (NWI) was obtained from the U.S. Fish & Wildlife Service (USFWS, 2023).
The NWI layer was reclassed to SLAMM Classic categories and converted to raster format. The
NWI dataset only included wetland polygons and required manual classification of dry land
27
categories into the converted raster. The site DEM, slope layer and NWI categories were then
converted to the ASCII file format and input into SLAMM.
The SLAMM model was run with custom sea level rise scenarios as the programs preset
options use data and SLR scenarios originating from the 2001 IPCC report. To match the 2019
IPCC report, the custom SLR scenarios were set with a base year of 2005 and used the median
values of global mean sea level rise (GMSL) in the 2046-2065 and 2100 projections for three
Representative Concentration Pathways (RCP) (Oppenheimer et al., 2019). Custom scenarios of
RCP 2.6, 4.5, and 8.5 were input as well as a preset 1 m scenario. The site parameters were set
using a combination of default settings, data from NOAA, and from settings described in Glick
et al. (2007) and Smith and Liedtke Outputs were saved as both tabular and GIS data formats.
28
Results
Summary Shoreline Statistics
Examination of shoreline surveys found that the study region contained over 3900
kilometers of shoreline with 1071 km of armored shore and 1701 km of shoreline with eelgrass.
Within the entire Puget Sound study region about 27% of the shorelines are armored on 50% or
more of the shoreline segment, and 43% of the total shoreline has been surveyed to have eelgrass
presence. Of these shores with reported eelgrass presence, 30% have armoring on 50% or more
of the shoreline segment.
Table 3: Summary statistics calculated in ArcGIS Pro from spatial survey data. Armored shore refers to
shoreline segments that are armored on 50% or more of their length.
Summary Shoreline Statistics
Total
Shoreline
(km)
3962.07
Total Armored Total Shoreline
Percentage Percentage Percentage Eelgrass
Total Armored Shore
Shoreline
with Eelgrass
of Shore of Shore with
shoreline with
with Eelgrass (km)
(km)
(km)
Armored
eelgrass
armor
1071.14
1701.82
512.32
27%
43%
30%
ArcGIS Pro projections
The projections of habitat change from ArcGIS analysis are focused on the narrow depth
range of preferred habitat and generally follow the expected patterns of change and habitat
expansion from each SLR scenario. In 3 out of 4 sites, the SLR projections resulted in expansion
of potential habitat. As expected, the RCP 2.6 scenario showed the least percent change in
habitat and the RCP 8.5 showed the greatest changes at all sites and scenarios.
29
Table 4: Results from ArcGIS Pro SLR raster analysis modeling. All scenarios use the DEM year of 2015
as the base year and 2100 as the projected year. "Depth" refers to the elevation range used as an
estimation of preferred Z. marina habitat. “ChangeDiffArmor/non” is the difference between the
percentage habitat change of the armored and non-armored site pair in the same scenario. Negative
values are in bold italic.
Central Puget Sound ArcGIS Projections of Eelgrass Habitat Change by 2100
Region
CPS
CPS
CPS
CPS
CPS
CPS
CPS
CPS
Region
CPS
CPS
CPS
CPS
CPS
CPS
CPS
CPS
Region
CPS
CPS
CPS
CPS
CPS
CPS
CPS
CPS
Region
CPS
CPS
CPS
CPS
CPS
CPS
CPS
CPS
Site Armor
1 Armor
1 Non
1 Armor
1 Non
1 Armor
1 Non
1 Armor
1 Non
Site Armor
2 Armor
2 Non
2 Armor
2 Non
2 Armor
2 Non
2 Armor
2 Non
Site Armor
3 Armor
3 Non
3 Armor
3 Non
3 Armor
3 Non
3 Armor
3 Non
Site Armor
4 Armor
4 Non
4 Armor
4 Non
4 Armor
4 Non
4 Armor
4 Non
Depth
0m to-2m
0m to-2m
0m to-2m
0m to-2m
0m to-2m
0m to-2m
0m to-2m
0m to-2m
Depth
0m to-2m
0m to-2m
0m to-2m
0m to-2m
0m to-2m
0m to-2m
0m to-2m
0m to-2m
Depth
0m to-2m
0m to-2m
0m to-2m
0m to-2m
0m to-2m
0m to-2m
0m to-2m
0m to-2m
Depth
0m to-2m
0m to-2m
0m to-2m
0m to-2m
0m to-2m
0m to-2m
0m to-2m
0m to-2m
Scenario
Base
Base
RCP2.6
RCP2.6
RCP4.5
RCP4.5
RCP8.5
RCP8.5
Scenario
Base
Base
RCP2.6
RCP2.6
RCP4.5
RCP4.5
RCP8.5
RCP8.5
Scenario
Base
Base
RCP2.6
RCP2.6
RCP4.5
RCP4.5
RCP8.5
RCP8.5
Scenario
Base
Base
RCP2.6
RCP2.6
RCP4.5
RCP4.5
RCP8.5
RCP8.5
Year Area (sq. m)
2015
65,700
2015
65,500
2100
80,858
2100
71,338
2100
85,566
2100
74,685
2100
109,957
2100
101,159
Year Area (sq. m)
2015
147,449
2015
80,308
2100
212,558
2100
121,745
2100
240,722
2100
148,369
2100
413,362
2100
284,796
Year Area (sq. m)
2015
40,140
2015
81,861
2100
47,896
2100
99,949
2100
51,171
2100
107,903
2100
55,148
2100
118,376
Year Area (sq. m)
2015
39,516
2015
40,853
2100
34,571
2100
33,501
2100
32,343
2100
31,054
2100
28,657
2100
27,456
GainLoss PCT change ChangeDiffArmor/non
15,158
0.231
0.142
5,838
0.089
-0.142
19,866
0.302
0.162
9,185
0.140
-0.162
44,257
0.674
0.129
35,659
0.544
-0.129
GainLoss PCT change ChangeDiffArmor/non
65,109
0.442
-0.074
41,437
0.516
0.074
93,273
0.633
-0.215
68,061
0.847
0.215
265,913
1.803
-0.743
204,488
2.546
0.743
GainLoss PCT change ChangeDiffArmor/non
7,756
0.193
-0.028
18,088
0.221
0.028
11,031
0.275
-0.043
26,042
0.318
0.043
15,008
0.374
-0.072
36,515
0.446
0.072
GainLoss PCT change ChangeDiffArmor/non
(4,945)
(7,352)
(7,173)
(9,799)
(10,859)
(13,397)
-0.125
-0.180
-0.182
-0.240
-0.275
-0.328
0.055
-0.055
0.058
-0.058
0.053
-0.053
At site 1 the armored beach showed greater habitat expansion than the unarmored beach
in each scenario. In each scenario the armored site showed 13-16% more change than the
30
unarmored site (Table 4). In the RCP 8.5 scenario, differences in percent change between the
armored and unarmored beaches were less than in RCP 2.6 and 4.5, indicating that the
differences in change may be reduced with greater increases in water height.
Figure 4: CPS site 1 unarmored and armored beaches projected habitat area for Initial (DEM date
2015), RCP 2.6, 4.5 and 8.5. Projected habitat area layers are stacked so that only the Initial habitat
area and areas of shoreward expansion are visible. Individual projections availible in Appendix A.
31
Figure 5: Percentage of habitat change between non-armored and armored beaches at CPS site 1.
Site 2 results aligned with expectations and showed greater habitat expansion at the
unarmored site than the armored site in every scenario. The difference between the armored and
unarmored sites percent change was relatively larger in the higher severity scenarios. In RCP 2.6
the difference between the two was only 7%, but in RCP 4.5 and 8.5 the differences were 22%
and 74%. The RCP 8.5 scenario at site 2 showed the greatest habitat expansion out of all sites
with 180% gain at the armored site and 255% gain at the unarmored site (Table 4).
32
Figure 6: CPS site 2 unarmored and armored beaches projected habitat area for Initial (DEM date
2015), RCP 2.6, 4.5 and 8.5. Projected habitat area layers are stacked so that only the Initial habitat
area and areas of shoreward expansion are visible. Individual projections availible in Appendix A.
Figure 7: Percentage habitat change between non-armored and armored beaches at CPS site 2.
33
Site 3 also showed expected habitat expansion with more gain at the unarmored site than
the armored site. The difference between percent change at armored and unarmored sites
followed a similar pattern to site 2, but the difference between scenarios was less severe. RCP
2.6 showed a difference of 3% followed by 4% in RCP 4.5 and 7% in RCP 8.5 (Table 4). At site
3 there was also the lowest percent change in RCP 8.5 compared to sites 1 and 2 (Table 4).
Figure 8: CPS site 3 unarmored and armored beaches projected habitat area for Initial (DEM date
2015), RCP 2.6, 4.5 and 8.5. Projected habitat area layers are stacked so that only the Initial habitat
area and areas of shoreward expansion are visible. Individual projections availible in Appendix A.
34
Figure 9: Percentage habitat change between non-armored and armored beaches at CPS site 3.
At site 4 the results were nearly opposite from the other 3 sites. Every scenario at this
location projected losses of habitat by 2100. The RCP 2.6 scenario showed the least amount of
loss with a 13% decrease in area at the armored site and 18% at the unarmored. RCP 8.5 showed
the greatest losses with 28% decrease at the armored site and 33% at the unarmored site. In each
scenario the non-armored site had over 5% more losses than the armored site (Table 4).
35
Figure 10: CPS site 4 unarmored and armored beaches projected habitat area for Initial (DEM date
2015), RCP 2.6, 4.5 and 8.5. Projected habitat area layers are stacked so that only the Initial habitat
area and areas of shoreward expansion are visible. Individual projections availible in Appendix A
Figure 11: Percentage habitat change between non-armored and armored beaches at CPS site 4.
36
During the analysis of these results, it became apparent that a distinct pattern of steep
elevation change was occurring near mean low water. This pattern forms a small band or seam
along the shoreline at each site that is likely where the topographic and bathymetric elevation
models were stitched together to form the original CoNED dataset. This seam appears to be
about 2-4 m wide, spanning from around 0.6 to 1-meter MLLW elevation, but the depth and
width vary between locations. The seam runs just above the edge of the RCP 8.5 coverage in
every scenario. The CoNED metadata explains that the edges between these datasets were
blended with the best available methods using several generalization and interpolation tools that
maintained a 1 m resolution and a smooth transition at the seamlines (OCM Partners, 2023). At
the regional level, a seamline with one meter or less elevation difference would not be a major
issue for most applications, but a SLR inundation model focused at low tidal and subtidal
elevations could be hindered by this boundary.
Profiles
Each site that was examined in the Central Puget Sound had a significantly different
shoreline profile. To visualize both the transitions from upland to foreshore as well as from
foreshore to the low wide terrace, the profiles at 1, 3, and 4 were drawn from 6m elevation to -2
m MLLW. To better represent the transitions at site 2 the profile was drawn from 6 m to 0.5m
because the greater width of the terrace at this location.
Foreshore and terrace width can be estimated from most of these profiles as well as slope.
However, the seamline in the DEM runs along this transition point and obscures the accurate
measurement of any of these parameters. At sites 2 and 3 there is enough unobscured foreshore
to estimate slope and widths and get a general picture of the beach profile. In the case of site 1,
the seam is over 2 m high and covers the majority of the foreshore, so it is difficult to estimate a
37
slope or width. To measure slope and widths at sites 2, 3, and 4, the average foreshore transition
elevation of 1.22 m MLLW from Finlayson (2006) was used as an estimation to differentiate
between foreshore and terrace.
Table 5: Profile characteristics of site transects. Slope is measured in percent slope. Foreshore is
estimated between MHW and transition elevation. Terrace is estimated between transition elevation and 2 m. "Avg Slope" refers to average slope between MHW and -2 m "Armorheight" refers to the
approximate MLLW elevation in meters of the base of shore armoring. "NA" = Not Applicable.
Profile Characteristics
Site
Foreshore
Armor Transect
#
(m)
1
N
1
NA
1
N
2
NA
1
N
3
NA
1
N
4
NA
1
Y
1
NA
1
Y
2
NA
1
Y
3
NA
1
Y
4
NA
FS
Terrace
T
Transition Avg.
Armorheight
slope
(m)
slope Elevation Slope
NA
NA
3.1
NA
4.0
NA
NA
NA
3.3
NA
4.0
NA
NA
NA
2.7
NA
4.1
NA
NA
NA
4.0
NA
5.5
NA
NA
NA
2.3
NA
3.8
NA
NA
NA
2.8
NA
4.0
NA
NA
NA
2.9
NA
4.2
2.5-4.5
NA
NA
5.0
NA
3.7
2.5-5
Site
Foreshore
Armor Transect
#
(m)
2
N
1
20.6
2
N
2
18.35
2
N
3
12.39
2
N
4
23.49
2
Y
1
25.24
2
Y
2
22.9
2
Y
3
7.93
2
Y
4
22.49
FS
Terrace
T
Transition Avg.
Armorheight
slope
(m)
slope Elevation Slope
9.64
294
1.2
1.22
1.7
NA
10.77
292
1.4
1.22
2.0
NA
16.26
373
1.2
1.22
2.0
NA
9.28
324
1.5
1.22
2.2
NA
7.93
309
1.3
1.22
1.9
NA
10.04
359
2.1
1.22
3.3
2.5
24.89
509
0.8
1.22
1.3
2.3
9.06
606.9
0.8
1.22
1.2
2.6
Site
Foreshore
Armor Transect
#
(m)
3
N
1
23.98
3
N
2
28.16
3
N
3
61.14
3
N
4
26.65
3
Y
1
31.96
3
Y
2
13.9
3
Y
3
14.16
FS
Terrace
T
Transition Avg.
Armorheight
slope
(m)
slope Elevation Slope
8.26
200.4
1.6
1.22
2.3
NA
7.03
194.1
1.7
1.22
2.3
NA
3.37
149.2
2.2
1.22
2.5
NA
7.37
103.7
3.1
1.22
4.0
NA
6.69
100.9
3.2
1.22
4.0
3.2
13.98
80.4
4.0
1.22
5.5
3.5
13.53
59.37
5.5
1.22
7.1
3.6
38
3
Y
4
13.48
Site
Foreshore
Armor Transect
#
(m)
4
N
1
13.88
4
N
2
14.43
4
N
3
12.3
4
N
4
12.96
4
Y
1
7.33
4
Y
2
10.34
4
Y
3
10.12
4
Y
4
8.831
14.55
48.15
6.7
1.22
8.4
4
FS
Terrace
T
Transition Avg.
Armorheight
slope
(m)
slope Elevation Slope
14.32
43.3
8.9
1.22 10.4
NA
16.06
50.39
9.2
1.22 10.7
NA
15.87
56.86
8.5
1.22 10.0
NA
14.74
41.72
8.2
1.22
9.9
NA
25.67
41.91
9.6
1.22 12.4
2.5
18.63
50.71
7.4
1.22
9.5
2.5
19.48
51.53
7.7
1.22
9.7
2.4
22.46
45.07
8.1
1.22 10.6
NA
Site 1 is a narrow fringe beach with a steep foreshore and a 150-200 m wide terrace. The
foreshore slope could not be accurately measured, but the terrace had a 3-5% slope (Table 5).
The terrace has a relatively convex shape with the slope increasing with depth (Figure 12, 13).
On the armored shore, the base of the armoring was obscured, but would fall somewhere
between 2.5 and 5m elevation (Table 5).
39
Figure 12: Central Puget Sound Site 1 armored site profiles measured from 6 m to -2 m MLLW. Mean
High Water for reference indicated by dashed red line.
Figure 13: Central Puget Sound Site 1 Non-armored site profiles measured from 6 m to -2 m MLLW.
Mean High Water for reference indicated by dashed red line.
40
Site 2 is a flat fringe beach with an 8-25 m wide foreshore and a 300-600 m wide terrace.
The foreshore slope was around 10-16% on unarmored and 8-24% on the armored side. The
terrace slope was around 1.5% at both unarmored and armored sites. The terrace shape was very
flat to slightly convex (Figure 14, 15). Armoring elevation was around 2.5 m on the armored site
(Table 5).
Figure 14: Central Puget Sound Site 2 armored site profiles measured from 6 m to 0.5 m MLLW. Mean
High Water for reference indicated by dashed red line.
41
Figure 15: Central Puget Sound Site 2 Non-armored site profiles measured from 6 m to 0.5 m MLLW.
Mean High Water for reference indicated by dashed red line
Site 3 is a very narrow fringe beach with a 14-60 m wide foreshore and a 50-200 m
terrace. The foreshore was wider at the unarmored side with around 24-60 m width as compared
to the 14-30 m wide armored beach. Foreshore slope was around 3-8% on the unarmored beach
and 7-14% on the armored beach. The terrace slope was approximately 2-3% on the unarmored
beach and 3-7% on the armored beach. The terrace had a convex shape with slope increasing
with depth (Figure 16, 17). Armoring elevation was estimated at around 3-4 m (Table 5).
42
Figure 16: Central Puget Sound Site 3 Non-armored site profiles measured from 6 m to -2 m MLLW.
Mean High Water for reference indicated by dashed red line.
Figure 17: Central Puget Sound Site 3 Armored site profiles measured from 6 m to -2 m MLLW. Mean
High Water for reference indicated by dashed red line.
43
Site 4 is a very narrow fringe beach with a steep foreshore and steep terrace. The
foreshore at the unarmored site was around 12-14 m wide with a 14-16% slope. At the armored
beach the foreshore was 7-10 m wide with a 19-25% slope. The terrace at both site pairs was 4055 m wide with a 7-9% slope. Armoring elevation was estimated to be around 2.5 m (Table 5).
The terrace has an even to concave shape with little to no increase in slope with depth (Figure
18, 19). This site is located at a very shallow section of the Puget Sound and it is difficult to
visualize any deep end of the terrace due to the lack of a characteristic steep transition from
terrace to deeper water.
Figure 18: Central Puget Sound Site 4 Non-armored site profiles measured from 6 m to -2 m MLLW.
Mean High Water for reference indicated by dashed red line.
44
Figure 19: Central Puget Sound Site 4 Armored site profiles measured from 6 m to -2 m MLLW. Mean
High Water for reference indicated by dashed red line.
SLAMM
The results from the SLAMM analysis were less useful than anticipated for investigating
eelgrass habitat change. While it was understood prior to analysis that there was no equivalent to
an eelgrass habitat category within the model, the outputs revealed less applicable information
than expected. Most of the land categories in SLAMM either were not present at the locations or
do not apply to the analysis and many had null or unchanged values. The categories that are most
applicable are “Tidal Flat”, “Estuarine Beach”, and “Estuarine Open Water”. The degree of
conversion of tidal flats and beaches to open water can to some degree be assumed to indicate
the amount of conversion to available habitat for eelgrass. However, the model unfortunately
does not provide an elevation output and so the lower limits and total area of the eelgrass habitat
are indiscernible from within the estuarine open water category output (Figure 20). Model
45
outputs also appeared to show very little difference between the scenarios. For site 2 and 3 the
results for RCP 2.6, 4.5 and 8.5 are nearly identical and show little to no change between the
emission scenarios. (Appendix B). These results may indicate errors in the selection of model
parameters and scenario preparation, or they may illustrate some of the limitations of using the
program. The restriction of the model to small shoreline areas with low wetland category
complexity is likely not what the model creators intended. Due to these issues the use of this
program was abandoned to focus on other study methods and only sites 1-3 were analyzed.
The most useful results can be summarized by comparison of the habitat change
differences between armored and unarmored sites while the detailed results are available in the
appendices. Armored sites 1 and 3 showed greater losses of tidal flat and estuarine beach and
corresponding increases in estuarine open water when compared to their unarmored site pairs
(Appendix B). At site 2 both the armored and unarmored sites lost 100% of their tidal flat and
estuarine beach, but the unarmored site had a more than 16% greater increase in estuarine open
water (Appendix B).
46
Figure 20: Example of map results from SLAMM analysis of CPS site 1. Panels show entire modeled area
of site 1 including both armored and unarmored site pairs. Individual armored and unarmored site
results tables located in Appendix B.
47
Discussion
Results of this study indicate that there is likely a significant percentage of Z. marina
habitat in the Puget Sound that is adjacent to armored shorelines and will be subject to their
influence on beach morphology. Nearly a third of estimated eelgrass habitat is on shoreline that
is at least 50% armored, and in a highly developed region like the Puget Sound there is high
potential that armoring will continue as SLR progresses (Dethier et al., 2016; Smith & Liedtke,
2022). Like other literature has indicated, the simulations of SLR have shown positive changes in
the total area of available habitat in the majority of sites and scenarios (Glick et al., 2007; Kairis
& Rybczyk, 2009; Smith & Liedtke, 2022).
The projections of habitat change from the ArcGIS analysis showed substantial
variability between sites in both overall change and comparisons of site pairs. While habitat
expansion occurred as expected in 3 out of 4 sites, percent change ranged between 9-52% for
RCP 2.6 and 38-255% for RCP 8.5 (Table 4). Projections displayed less habitat expansion at the
armored locations in two out of the three sites that showed overall habitat increases. At sites 2
and 3 the armored beaches showed a lower positive percent change in each scenario. At sites 1
and 4 the unarmored pairs do not appear to offer greater protection to habitat as they resulted in
less expansion or greater loss in the case of site 4. The steep and narrow Site 3 had the lowest
positive percent change in RCP 8.5 for both armored and unarmored sites and the flatter wider
Site 2 had the largest percentages of expansion.
The results support the idea that steep slopes near to the shoreline will naturally limit the
total area of habitat expansion as the ideal depth is confined to a narrower band of land surface.
However, any conclusions regarding armoring as a determinant of beach morphology should not
be made with these limited results. Shoreline armoring could be a cause of steeper beach slope in
48
these locations; however, it is also possible that locations with steeper slopes are more likely to
need stabilization and the relationship between armor and higher slopes may not be causal.
Despite this uncertainty regarding the role of armoring in determining the slope represented in
the elevation model, literature supports the idea that armoring causes increased slope (Dethier et
al., 2016; Shipman et al., 2010).
Beach profiles appear to have a large effect on the amount of habitat expansion in any
given SLR scenario, even in less severe scenarios. In the Central Puget Sound, the majority of
eelgrass is found on locations with fringe type profiles, and all the sites used in this study were
categorized as fringe type sites (Christiaen et al., 2022). Previous studies of flats type sites have
found significant habitat expansion in SLR scenarios, but fringe sites are understudied (Kairis &
Rybczyk, 2009; Poppe & Rybczyk, 2022). The results from this study show high variability
between each fringe site. Very steep sites may show habitat loss, but flatter fringe beach
locations with a large “low wide terrace”, like Site 2, appear to provide large areas for new
habitat in the preferred habitat depth for Z. marina. In the Puget Sound, even flat beaches are
often characterized by steep foreshores which appear to begin around 1 m MLLW (Finlayson,
2006). This steep shore will likely slow the rate of habitat expansion, especially if it is resistant
to erosion as in the case of an armored shore. Steep and narrow fringe locations that have less
terrace, like site 4, may experience little expansion or even habitat loss as the preferred habitat
range is compressed against the shoreline. If SLR progresses near or past 1 meter, then
reductions in expanded habitat may begin to occur from the lower limit of habitat as water
deepens on the terrace. Accretion and erosion processes will occur over time as SLR progresses,
and may alter the beach profile to elevate the terrace and foreshore, but this plasticity is likely
reduced on armored beaches (Smith & Liedtke, 2022).
49
These results may also show that profile shape on the low wide terrace may be a predictor
of projected habitat gain or loss. At sites 1-3, where there was projected habitat gain, the terrace
profiles had a relatively convex shape that exhibited increasing slope with increasing depth. At
the only site with projected habitat losses, site 4, the terrace profile had a relatively straight to
concave shape that did not gradually increase slope with depth. SLR on concave profiles will
push habitat onto steeper slopes and reduce available habitat area in the preferred depth range.
Alternatively, SLR on convex profiles will result in habitat expansion over the decreasing slope
until impeded by the foreshore transition.
The preferred habitat area for Z. marina appeared to remain below the foreshore and on
the low wide terrace in all scenarios. Unfortunately, this result may have been forced by the
boundary of the seam in the elevation data. However, the foreshore to terrace transition typically
occurs near 1 meter MLLW in Puget Sound, which is above where any of the scenarios used
would have shown projected habitat expansion (Finlayson, 2006). If SLR remains below 1 meter,
then eelgrass preferred habitat will stay on what is currently the low wide terrace. This part of the
shoreline profile is typically characterized by finer sand and sediment and should provide
suitable substrate. If SLR surpasses 1 meter rise, then habitat may encroach upon the foreshore
where grain size is larger. On unarmored beaches erosion and sedimentation could decrease grain
size over time, but this would be more difficult at armored locations.
Another factor to consider, especially at flat sites, is the rate of habitat expansion and the
ability of Z. marina to migrate into potential habitat. Eelgrass has been estimated to have a lateral
spreading migration rate of around 12-15cm per year, which would result less than 10-12 m
expansion by 2100 (Kairis & Rybczyk, 2009; Neckles et al., 2005). In contrast the lateral habitat
expansion modeled would be about 10-100 m at sites 1 and 3 and up to around 350 m at site 2.
50
This means that the primary method of new habitat colonization would have to be through seed
dispersal. Spreading eelgrass through seed has been known to result in rapid expansion or
recolonization of beds, but the rates of this type of spread are variable and rely heavily on factors
such as flowering intensity, environmental conditions and new shoot mortality (Neckles et al.,
2005; Olesen & Sand-Jensen, 1994)
As expected, projections did not show Z. marina habitat directly overlapping with
armoring because of the low tidal and subtidal range of eelgrass, but the distance between the
two closed significantly. Most armoring was located near the Mean High Water line (around 3 m
in Puget Sound) with some areas extending down to nearly 2 m MLLW. Projections did show
that the preferred habitat range would move from 50-250m distance from the armoring to within
10-15m in the RCP 8.5 scenario at every location. The severity of negative armoring impacts are
thought to be greater when armoring is located lower along the beach profile (Shipman et al.,
2010). This increasing proximity of Z. marina to armoring will expand the risk of exposure to the
negative effects of wave energetics, increased substrate grain size, and turbidity (Poppe &
Rybczyk, 2022; Shipman et al., 2010). While the future conditions of beach slope, profile shape,
and habitat area are uncertain and dependent on individual location and fluctuating sediment, the
closing distance between habitat and armoring is a certainty in the face of rising sea levels.
Project limitations and Assumptions
While considering the results and implications of this study it is important to reiterate
potential inaccuracies resulting from the data and methods used in this study. Firstly, some older
and lower resolution data like the Shorezone survey were used as a basis for estimating eelgrass
presence. Secondly the digital elevation model source used as the basis for SLR projections was
formed from separate topographic and bathymetric datasets which may have resulted in a narrow
51
band of steep slope with generalized values that passed through each study site. Thirdly, erosion
and accretion processes were not modeled in this study and its results are based upon the
assumption of unchanged bathymetry from 2015 to 2100. The analysis conducted in this uses
extant digital elevation models that do not change in the varying SLR scenarios; modeling
erosion or sedimentation processes in various scenarios could be beneficial in future work.
During the initial stages of this research project the intention was to use the Sea Level
Affecting Marshes Model (SLAMM) as a method of modeling beach erosion. As erosion and
accretion processes are one of the main differences between armored and unarmored shores,
modeling these changes would have been greatly beneficial for the results and conclusions of the
study. It was understood initially that the program could not directly model eelgrass habitat but,
considering its use in Smith and Leidtke (2022) and Glick et al. (2006), it appeared promising for
providing at least a basis for comparison to other survey and analysis results. However, using the
SLAMM program proved to be more time and labor intensive than initially anticipated, and its
results indicated that user inexperience and model limitations made the data output ineffective
for this study.
It is important to note that these projections assume that eelgrass will retain the same
preferred habitat range while under the influence of other climate change impacts. Estimates of
climate change induced sea level rise include calculations of thermal expansion which implies
that any significant increase of sea level will have a corresponding increase of water temperature
(Oppenheimer et al., 2019). Thermal changes also can cause alterations to water chemistry and
water quality. Increased temperature and low water quality are known to have negative effects on
eelgrass health. While eelgrass may have available habitat according to geomorphological and
52
bathymetric characteristics, that does not guarantee overall environmental suitability for the
species.
Future studies of eelgrass habitat expansion would benefit greatly from continued survey
and data collection on as many shorelines as possible. At the Puget Sound regional scale there
are large datasets like Shorezone and CoNED that can be incredibly useful but can also contain
data that is over 20 years old. New surveys should include accurate elevation models of the
shoreline at the land/ water interface between MHHW and MLLW. This elevation band is very
difficult to digitally map because water reflectance restricts aerial remote sensing and the
shallow depth restricts boat borne remote sensing, but new technologies like green laser LIDAR
are making advances in making these elevation models possible (OCM Partners, 2023). It may
also be possible to use the separate elevation models used in the CoNED DEM and use different
methods of interpolation to generalize the land/ water interface more accurately. The CoNED
project used a convex hull method which may have misrepresented the curvature of the beach
surface at the foreshore/ terrace transition, and future studies could determine a more suitable
tool to create a smoother transition between the datasets.
Conclusion
This study explored the effects of global sea level rise on the habitat of Z. marina and the
degree to which shoreline armoring will impact the species response to this rising threat. The
results of this study indicate that nearly one third of known eelgrass habitat is potentially
exposed to the effects of armoring as shown by the overall extent of armoring present in Puget
Sound. Habitat change projections show habitat expansion at a majority of sites as sea levels
increase. However, results are inconclusive about the differences in Z. marina habitat response to
SLR between armored and unarmored beaches. Fringing habitat locations with steep elevation
53
profiles were shown to have high variability in the level of habitat change, but profile shape was
shown to be an indicator of potential gain or loss. Of potentially more concern than overall
habitat area changes is the increasing proximity of eelgrass habitat to armored surfaces, as effects
are unknown regarding eelgrass’s ability to effectively colonize habitat near these anthropogenic
structures.
Some of the insights provided by this study into potential eelgrass habitat changes may be
useful for land management and ecosystem restoration. Most studies have found that eelgrass
habitat will expand with sea level rise at flat locations, but this study suggests that outcomes on
fringe locations may be more difficult to predict. While shoreline armoring may not individually
present significant concerns to Z. marina at this time, it is important to be aware that rising sea
levels could increase the risks of impact. When land managers and conservationists evaluate
plans for SLR mitigation or ecosystem restoration at locations with armoring and eelgrass
presence, they should be sure to consider the increasing proximity of armoring to habitat in the
future.
54
Bibliography
Anchor QEA. (2009). Geospatial Methodology Used in The PSNERP Comprehensive Change
Analysis of Puget Sound. (pp. 1–373). Geospatial Methodology Used in The Puget Sound
Nearshore Ecosystem Restoration Project.
Berry, H. D., Harper, J. R., Mumford, T. F., Bookheim, B. E., Sewell, A. T., & Tamayo, L. J.
(2001). The Washington State ShoreZone Inventory User’s Manual. Nearshore Habitat
Program. Washington State Department of Natural Resources.
Candela, T., & Koster, K. (2022). The many faces of anthropogenic subsidence. Science,
376(6600), 1381–1382. https://doi.org/10.1126/science.abn3676
Cereghino, P., Toft, J., Simenstad, C., Iverson, E., Campbell, S., Behrens, C., & Burke, J. (2012).
Strategies for Nearshore Protection and Restoration in Puget Sound (Technical Report
No. 2012–01; p. 184). Puget Sound Nearshore Ecosystem Restoration Project.
Christiaen, B., Ferrier, L., Dowty, P., Gaeckle, J., & Berry, H. (2019). Puget Sound Seagrass
Monitoring Report, Monitoring Year 2016—2017. Nearshore Habitat Program.
Washington State Department of Natural Resources.
Christiaen, B., Ferrier, L., Dowty, P., Gaeckle, J., & Berry, H. (2022). Puget Sound Seagrass
Monitoring Report, Monitoring year 2018-2020 (p. 71). Nearshore Habitat Program.
Washington State Department of Natural Resources.
Clough, J. S., Park, R., Propato, M., & Polaczyk, A. (2016). SLAMM 6.7 Technical
Documentation. Warren Pinnacle Consulting.
Dethier, M. N., Raymond, W. W., McBride, A. N., Toft, J. D., Cordell, J. R., Ogston, A. S.,
Heerhartz, S. M., & Berry, H. D. (2016). Multiscale impacts of armoring on Salish Sea
55
shorelines: Evidence for cumulative and threshold effects. Estuarine, Coastal and Shelf
Science, 175, 106–117. https://doi.org/10.1016/j.ecss.2016.03.033
Dethier, M. N., Toft, J. D., & Shipman, H. (2017). Shoreline Armoring in an Inland Sea:
Science-Based Recommendations for Policy Implementation. Conservation Letters,
10(5), 626–633. https://doi.org/10.1111/conl.12323
Duffy, E. J., Beauchamp, D. A., & Buckley, R. M. (2005). Early marine life history of juvenile
Pacific salmon in two regions of Puget Sound. Estuarine, Coastal and Shelf Science,
64(1), 94–107. https://doi.org/10.1016/j.ecss.2005.02.009
Dumbauld, B., Hosack, G., & Bosley, K. (2015). Association of Juvenile Salmon and Estuarine
Fish with Intertidal Seagrass and Oyster Aquaculture Habitats in a Northeast Pacific
Estuary. Transactions of the American Fisheries Society, 144, 1091–1110.
https://doi.org/10.1080/00028487.2015.1054518
Fagherazzi, S., Mariotti, G., Leonardi, N., Canestrelli, A., Nardin, W., & Kearney, W. S. (2020).
Salt Marsh Dynamics in a Period of Accelerated Sea Level Rise. Journal of Geophysical
Research: Earth Surface, 125(8), e2019JF005200. https://doi.org/10.1029/2019JF005200
Finlayson, D. (2006). The Geomorphology of Puget Sound Beaches: (Technical Report No.
2006–02). Puget Sound Nearshore Partnership. https://doi.org/10.21236/ADA477548
Glick, P., Clough, J., & Nunley, B. (2007). Sea Level Rise and Coastal Habitats in the Pacific
Northwest (pp. 1–106). National Wildlife Federation.
Hannam, M. (2013). The Influence of Multiple Scales of Environmental Context on the
Distribution and Interaction of an Invasive Seagrass and its Native Congener [Thesis,
University of Washington].
https://digital.lib.washington.edu:443/researchworks/handle/1773/23711
56
Hannam, M., Dowty, P., Christiaen, B., Berry, H., Ferrier, L., Gaeckle, J., Stowe, J., & Sutton, E.
(2015). Depth Distribution of Eelgrass in Greater Puget Sound (p. 68). Nearshore
Habitat Program, Aquatic Resources Division, WSDNR.
Hitchcock, C. L., & Cronquist, A. (2018). Flora of the Pacific Northwest: An illustrated manual
(Second). University of Washington Press.
Kairis, P. A., & Rybczyk, J. M. (2009). Sea level rise and eelgrass (Zostera marina) production:
A spatially explicit relative elevation model for Padilla Bay, WA. Ecological Modelling,
221(7), 1005–1016. https://doi.org/10.1016/j.ecolmodel.2009.01.025
Kennedy, L. A., Juanes, F., & El-Sabaawi, R. (2018). Eelgrass as Valuable Nearshore Foraging
Habitat for Juvenile Pacific Salmon in the Early Marine Period. Marine and Coastal
Fisheries, 10(2), 190–203. https://doi.org/10.1002/mcf2.10018
Kinney, A., Francis, T., & Rice, J. (2015). Analysis of Effective Regulation and Stewardship
Findings (p. 57). University of Washington Puget Sound Institute.
Miller, I., Morgan, H., Mauger, G., Newton, T., Weldon, R., Schmidt, D., Welch, M., &
Grossman, E. (2019). Projected Sea Level Rise for Washington State – A 2018
Assessment. A collaboration of Washington Sea Grant, University of Washington Climate
Impacts Group, Oregon State University, University of Washington, and US Geological
Survey. https://cig.uw.edu/publications/projected-sea-level-rise-in-washington-state-a2018-assessment/
Moore, K. A., & Short, F. T. (2006). Zostera: Biology, Ecology, and Management. In
Seagrasses: Biology, Ecology and Conservation (pp. 361–386). Springer Netherlands.
https://doi.org/10.1007/978-1-4020-2983-7_16
57
Moritsch, M. M., Byrd, K. B., Davis, M., Good, A., Drexler, J. Z., Morris, J. T., Woo, I.,
Windham-Myers, L., Grossman, E., Nakai, G., Poppe, K. L., & Rybczyk, J. M. (2022).
Can Coastal Habitats Rise to the Challenge? Resilience of Estuarine Habitats, Carbon
Accumulation, and Economic Value to Sea-Level Rise in a Puget Sound Estuary.
Estuaries and Coasts. https://doi.org/10.1007/s12237-022-01087-5
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. https://doi.org/10.1007/s12237-012-9481-3
National Oceanic and Atmospheric Administration. (2020). Station Selection—NOAA Tides &
Currents. Tides and Currents.
https://tidesandcurrents.noaa.gov/stations.html?type=Datums
National Research Council, Division on Earth and Life Studies, Ocean Studies Board, Board on
Earth Sciences and Resources, & Committee on Sea Level Rise in California, O., and
Washington. (2012). Sea-Level Rise for the Coasts of California, Oregon, and
Washington: Past, Present, and Future. National Academies Press.
http://ebookcentral.proquest.com/lib/esu/detail.action?docID=3379304
Neckles, H., Short, F., Barker, S., & Kopp, B. (2005). Disturbance of eelgrass Zostera marina by
commercial mussel Mytilus edulis harvesting in Maine: Dragging impacts and habitat
recovery. Marine Ecology Progress Series, 285, 57–73.
https://doi.org/10.3354/meps285057
OCM Partners. (2023). 1887—2017 USGS CoNED Topobathy DEM (Compiled 2020): Puget
Sound, WA from 2010-06-15 to 2010-08-15. NOAA National Centers for Environmental
Information. https://www.fisheries.noaa.gov/inport/item/59971.
58
Olesen, B., & Sand-Jensen, K. (1994). Patch dynamics of Eelgrass Zostera marina. Marine
Ecology-Progress Series, 106, 147–156. https://doi.org/10.3354/meps106147
Oppenheimer, M., Glavovic, B., Hinkel, J., van de Wal, R., Magnan, A. K., Abd-Elgawad, A.,
Cai, R., Cifuentes-Jara, M., Deconto, R. M., Ghosh, T., Hay, J., Isla, F., Marzeion, B.,
Meyssignac, B., & Sebesvari, Z. (2019). Sea Level Rise and Implications for Low-Lying
Islands, Coasts and Communities. In: IPCC Special Report on the Ocean and Cryosphere
in a Changing Climate (pp. 321–445). Cambridge University Press.
https://doi.org/10.1017/9781009157964.006.
Pacella, S. R., Brown, C. A., Waldbusser, G. G., Labiosa, R. G., & Hales, B. (2018). Seagrass
habitat metabolism increases short-term extremes and long-term offset of CO2 under
future ocean acidification. Proceedings of the National Academy of Sciences, 115(15),
3870–3875. https://doi.org/10.1073/pnas.1703445115
Poppe, K. L., & Rybczyk, J. M. (2022). Assessing the future of an intertidal seagrass meadow in
response to sea level rise with a hybrid ecogeomorphic model of elevation change.
Ecological Modelling, 469, 109975. https://doi.org/10.1016/j.ecolmodel.2022.109975
Shipman, H., Dethier, M. N., Gelfenbaum, G., Fresh, K. L., & Dinicola, R. S. (2010). Puget
Sound Shorelines and the Impacts of Armoring—Proceedings of a State of the Science
Workshop, May 2009 (USGS Scientific Investigations Report No. 2010–5254; p. 262 p.).
https://pubs.usgs.gov/sir/2010/5254/
Simenstad, C., Ramirez, M., Burke, J., Logsdon, M., Shipman, H., Tanner, C., Toft, J., Craig, B.,
Davis, C., Fung, J., Bloch, P., Fresh, K. L., Campbell, S., Meyers, D., Iverson, E., Bailey,
A., Schlenger, P., Kiblinger, C., Myre, P., … MacLennan, A. (2011). Historical Change
and Impairment of Puget Sound Shorelines: Puget Sound Nearshore Ecosystem Project
59
Change Analysis (Puget Sound Nearshore Report No. 2011-01; p. 313). Washington
Department of Fish and Wildlife, Olympia, Washington, and U.S. Army Corps of
Engineers, Seattle, Washington.
Smith, C. D., & Liedtke, T. L. (2022). Potential effects of sea level rise on nearshore habitat
availability for surf smelt (Hypomesus pretiosus) and eelgrass (Zostera marina), Puget
Sound, Washington. In Potential effects of sea level rise on nearshore habitat availability
for surf smelt (Hypomesus pretiosus) and eelgrass (Zostera marina), Puget Sound,
Washington (USGS Numbered Series No. 2022–1054; Open-File Report, Vols. 2022–
1054). U.S. Geological Survey. https://doi.org/10.3133/ofr20221054
Thom, R. M., & Williams, G. D. (2001). Executive Summary: Marine and Estuarine Shoreline
Modification Issues (p. 140) [White Paper]. Battelle Marine Sciences Laboratory and
Pacific Northwest National Laboratory.
USFWS. (2023). Wetlands Mapper | U.S. Fish & Wildlife Service. FWS.Gov.
https://www.fws.gov/program/national-wetlands-inventory/wetlands-mapper
60
Appendix A.
ArcGIS Pro projection maps
Appendix A 1. CPS unarmored site 1 projected habitat area for Initial (DEM date 2015), RCP 2.6, 4.5
and 8.5. Projected habitat area layers are stacked so that only the Initial habitat area and areas of
shoreward expansion are visible.
61
Appendix A 2. CPS armored site 1 projected habitat area for Initial (DEM date 2015), RCP 2.6, 4.5 and
8.5. Projected habitat area layers are stacked so that only the Initial habitat area and areas of shoreward
expansion are visible
62
Appendix A 3. CPS unarmored site 2 projected habitat area for Initial (DEM date 2015), RCP 2.6, 4.5
and 8.5. Projected habitat area layers are stacked so that only the Initial habitat area and areas of
shoreward expansion are visible
63
Appendix A 4. CPS armored site 2 projected habitat area for Initial (DEM date 2015), RCP 2.6, 4.5 and
8.5. Projected habitat area layers are stacked so that only the Initial habitat area and areas of shoreward
expansion are visible
64
Appendix A 5. CPS unarmored site 3 projected habitat area for Initial (DEM date 2015), RCP 2.6, 4.5
and 8.5. Projected habitat area layers are stacked so that only the Initial habitat area and areas of
shoreward expansion are visible
65
Appendix A 6. CPS armored site 3 projected habitat area for Initial (DEM date 2015), RCP 2.6, 4.5 and
8.5. Projected habitat area layers are stacked so that only the Initial habitat area and areas of shoreward
expansion are visible
66
Appendix A 7. CPS unarmored site 4 projected habitat area for Initial (DEM date 2015), RCP 2.6, 4.5
and 8.5. Projected habitat area layers are stacked so that only the Initial habitat area and areas of
shoreward expansion are visible
67
Appendix A 8. CPS armored site 4 projected habitat area for Initial (DEM date 2015), RCP 2.6, 4.5 and
8.5. Projected habitat area layers are stacked so that only the Initial habitat area and areas of shoreward
expansion are visible
68
Appendix A 9. CPS unarmored site 1 projected habitat area for Initial (DEM date 2015).
69
Appendix A 10. CPS unarmored site 1 projected habitat area for RCP 2.6.
70
Appendix A 11. CPS unarmored site 1 projected habitat area for RCP 4.5
71
Appendix A 12. CPS unarmored site 1 projected habitat area for RCP 8.5
72
Appendix A 13. CPS armored site 1 projected habitat area for Initial (DEM date 2015).
73
Appendix A 14. CPS armored site 1 projected habitat area for RCP 2.6.
74
Appendix A 15. CPS armored site 1 projected habitat area for RCP 4.5
75
Appendix A 16. CPS armored site 1 projected habitat area for RCP 8.5
76
Appendix A 17. CPS unarmored site 2 projected habitat area for Initial (DEM date 2015).
77
Appendix A 18. CPS unarmored site 2 projected habitat area for RCP 2.6.
78
Appendix A 19. CPS unarmored site 2 projected habitat area for RCP 4.5
79
Appendix A 20. CPS unarmored site 2 projected habitat area for RCP 8.5
80
Appendix A 21. CPS unarmored site 2 projected habitat area for Initial (DEM date 2015).
81
Appendix A 22. CPS armored site 2 projected habitat area for RCP 2.6.
82
Appendix A 23. CPS armored site 2 projected habitat area for RCP 4.5
83
Appendix A 24. CPS armored site 2 projected habitat area for RCP 8.5
84
Appendix A 25. CPS unarmored site 3 projected habitat area for Initial (DEM date 2015).
85
Appendix A 26. CPS unarmored site 3 projected habitat area for RCP 2.6.
86
Appendix A 27. CPS unarmored site 3 projected habitat area for RCP 4.5.
87
Appendix A 28. CPS unarmored site 3 projected habitat area for RCP 8.5.
88
Appendix A 29. CPS armored site 3 projected habitat area for Initial (DEM date 2015).
89
Appendix A 30. CPS armored site 3 projected habitat area for RCP 2.6.
90
Appendix A 31. CPS armored site 3 projected habitat area for RCP 4.5
91
Appendix A 32. CPS armored site 3 projected habitat area for RCP 8.5
92
Appendix A 33. CPS unarmored site 4 projected habitat area for Initial (DEM date 2015).
93
Appendix A 34. CPS unarmored site 4 projected habitat area for RCP 2.6.
94
Appendix A 35. CPS unarmored site 4 projected habitat area for RCP 4.5
95
Appendix A 36. CPS unarmored site 4 projected habitat area for RCP 8.5
96
Appendix A 37. CPS armored site 4 projected habitat area for Initial (DEM date 2015).
97
Appendix A 38. CPS armored site 4 projected habitat area for RCP 2.6.
98
Appendix A 39. CPS armored site 4 projected habitat area for RCP 4.5
99
Appendix A 40. CPS armored site 4 projected habitat area for RCP 8.5
100
Appendix B.
SLAMM Analysis Results
Appendix B 1. CPS Unarmored Site 1 SLAMM Projections of Wetland Habitat Change by 2100. Percent
Change represented in decimal format. “PCT change” = Percent Habitat Change. “NA” = Not
Applicable.
101
Appendix B 2. CPS Armored Site 1 SLAMM Projections of Wetland Habitat Change by 2100. Percent
Change represented in decimal format. “PCT change” = Percent Habitat Change. “NA” = Not
Applicable.
102
Appendix B 3. CPS Unarmored Site 2 SLAMM Projections of Wetland Habitat Change by 2100. Percent
Change represented in decimal format. “PCT change” = Percent Habitat Change. “NA” = Not
Applicable.
103
Appendix B 4. CPS Armored Site 2 SLAMM Projections of Wetland Habitat Change by 2100. Percent
Change represented in decimal format. “PCT change” = Percent Habitat Change. “NA” = Not
Applicable.
104
Appendix B 5. CPS Unarmored Site 3 SLAMM Projections of Wetland Habitat Change by 2100. Percent
Change represented in decimal format. “PCT change” = Percent Habitat Change. “NA” = Not
Applicable.
105
Appendix B 6. CPS Armored Site 3 SLAMM Projections of Wetland Habitat Change by 2100. Percent
Change represented in decimal format. “PCT change” = Percent Habitat Change. “NA” = Not
Applicable.
106
