-
Title
-
Eng
An Evaluation of Eelgrass (Zostera marina) Epifaunal Communities Following Large-Scale Restoration of the Nisqually River Delta, Washington
-
Date
-
2015
-
Creator
-
Eng
Blakely, Sierra Sellers
-
Subject
-
Eng
Environmental Studies
-
extracted text
-
AN EVALUATION OF EELGRASS (Zostera marina) EPIFAUNAL COMMUNITIES
FOLLOWING LARGE-SCALE RESTORATION of the
NISQUALLY RIVER DELTA, WASHINGTON
by
Sierra Sellers Blakely
A Thesis
Submitted in partial fulfillment
of the requirements for the degree
Master of Environmental Studies
The Evergreen State College
June 2015
© 2015 by Sierra Sellers Blakely. All rights reserved.
This Thesis for the Master of Environmental Studies Degree
by
Sierra Sellers Blakely
has been approved for
The Evergreen State College
by
________________________
Dina Roberts, Ph.D.
Member of the Faculty
________________________
Date
Abstract
AN EVALUATION OF EELGRASS (Zostera marina) EPIFAUNAL COMMUNITIES
FOLLOWING LARGE-SCALE RESTORATION of the
NISQUALLY RIVER DELTA, WASHINGTON
Sierra Sellers Blakely
Zostera marina, a species of eelgrass native to the Puget Sound, serves as an ecosystem
engineer in marine environments, supports an abundance of commercially and culturally
important species, and is widely considered an indicator of ecosystem health. Eelgrass
beds located along estuary boundaries benefit outmigrating juvenile salmon by providing
refugia and hosting a diverse community of invertebrate prey, thus facilitating the
transition from freshwater to saltwater. In particular, the federally-threatened Nisqually
fall Chinook (Oncorhynchus tshawytscha) stock stands to benefit from eelgrass habitat.
Here I evaluate the epifaunal invertebrate prey composition of four distinct eelgrass beds
located along the Nisqually River Delta in Puget Sound, Washington for abundance and
diversity during the spring outmigration season (March - September 2014). Epifaunal
abundance increased through time at all sites but one, and community structure
experienced a shift in dominance from amphipods to increasing proportions of
polychaetes and annelids. Invertebrate abundance was most significantly influenced by
site (p<0.0001) and month (p<0.001), and eelgrass shoot complexity was positively
correlated with increased abundance for many key species. These data help to validate
and quantify the contribution of Nisqually eelgrass beds as a valuable source of prey for
juvenile Chinook salmon, and reinforce the ecological value of these habitats for the
management of threatened salmon stocks in Puget Sound.
Table of Contents
LIST OF FIGURES……………………………………............................... v
LIST OF TABLES ……………………………………………...…………vi
ACKNOWLEDGEMENTS..………………………………...…….............vii
INTRODUCTION…………………………………………..…………....... 1
LITERATURE REVIEW….......................................................................... 4
Overview of Seagrasses………………………………..………….. 4
Life History of Eelgrass…………………………………………… 5
Salmon Utilization of Eelgrass Habitat …………………….………8
Patterns of Eelgrass Decline …………………………………......... 10
Eelgrass Trends in Puget Sound …………………………………... 12
Eelgrass Restoration………………………………………………...14
Impact of Dikes on Coastal Habitats ……………………………… 16
METHODS ……………………...………………………………………… 22
Study Area ………………………………………………………… 19
Experimental Design and Data Analysis ………………………...... 23
Plant Collection…………………………………………………….. 24
Invertebrate Collection …………………………………………......24
RESULTS ………………………………………………………………..... 26
Community Structure ……………………………………………… 26
Biodiversity ………………………………………………………. 30
Key Species Analysis …………………………………………….. 31
Biophysical Variables by Site……………………………………… 37
DISCUSSION ……………………………………………………………. 40
Site Variation ……………………………………………………… 40
Patterns of Epifaunal Diversity…………………………………….. 41
Suggestions for Further Research …………………………………. 42
TABLES…………………………………………………………………… 44
LITERATURE CITED…………………………………………………….. 48
APPENDIX A……………………………………………………………… 53
APPENDIX B……………………………………………………………… 54
APPENDIX C……………………………………………………………… 55
iv
List of Figures
Figure 1 – Zostera marina technical illustration …………………................6
Figure 2 – the eelgrass meadow: a world of microhabitats. …………….......8
Figure 3 – Salmon habitat use in a restored estuarine wetland. …………… 10
Figure 4 – Site-wide decadal trends in eelgrass abundance………………... 15
Figure 5 – The Nisqually River Delta, Washington USA…………………. 22
Figure 6 – Nisqually National Wildlife Refuge Restoration………………. 23
Figure 7 – Nisqually Eelgrass Sites………………………………………... 24
Figure 9a – Percent composition of epifaunal phylum abundance……….... 28
Figure 9b – Percent composition of epifaunal species abundance………… 29
Figure 10 – Mean epifaunal abundance by site and month………………... 30
Figure 11 – Phylum abundance by month and site……………………….... 30
Figure 12 – nMDS ordination of taxa by site. …………………………….. 31
Figure 13 – Mean sample diversity by site and month. …………………… 32
Figure 14 – Arthropoda abundance by site & month……………………… 33
Figure 15 – Percent composition of Arthropoda abundance………………. 34
Figure 16 – Amphipoda abundance by site and month…………………..... 35
Figure 17 – Copepoda abundance by site & month………………………... 36
Figure 18 – Polychaeta abundance by site & month…………….………… 37
Figure 19 – Ostracoda abundance by site & month……………...………… 38
Figure 20 – Tanaidacea abundance by site & month…………...………….. 39
Figure 21 – Temperature by site and month…………………...…………... 40
Figure 22 – Salinity by site and month………………………...…………... 41
v
List of Tables
Table 1 – Multivariate Analysis (MANOVA) of the effects of site,
month, nodes and mean blade length on epifaunal abundance…………….. 48
Table 2 – One-way ANOVA output for Shannon’s biodiversity
index and site characteristics………………………………………………. 48
Table 3 – AIC model selection of one-way ANOVA fit of Shannon’s
Biodiversity Index………………………………………………………….. 48
Table 4 – AIC model selection of influences on Arthropoda abundance….. 49
Table 5 – AIC model selection of influences on Amphipoda abundance…..49
Table 6 – AIC model selection of influences on Copepoda abundance…… 49
Table 7 – AIC model selection of influences on Polychaeta abundance……49
Table 8 – One-way ANOVA of the effect of month, site and
month x site on temperature……………………………………………….. 49
vi
Acknowledgements
This study would not have been possible without the help from a great many people:
My thesis reader Dr. Dina Roberts at the Evergreen State College, whose dedication and
commitment to my success provided encouragement and inspiration throughout this
process: I could not have done this without your support.
Isa Woo and Susan De La Cruz of the USGS Western Ecological Research Center
provided access to the epifaunal data for this project. Steve Rubin of the USGS Western
Fisheries Research Center provided the 2014 data of eelgrass bed density. Melanie Davis
of the USGS Western Ecological Research Center was instrumental in providing
guidance, statistical help and general advice, and was an invaluable resource and
inspiration. Lennah Shakeri and Anna Hissem (USGS) provided fieldwork assistance
throughout the 2014 sampling season. Charles Norton (USGS) identified the epifaunal
invertebrates used in this project. Glynnis Nakai, Doug Roster, Brian Root and Marian
Bailey with the Nisqually National Wildlife Refuge provided site access, field equipment
and shore contacts that allowed us to safely and successfully complete our sampling
regime.
Acknowledgments go towards my fellow graduate student peers for their support and
advice, especially Chelsea Waddell and Sean Greene who provided valuable assistance
with statistical analyses.
Lastly, I must thank my family and friends for being a source of endless support,
especially my fiancé Ian Shopland for his unwavering dedication throughout this process.
vii
I. INTRODUCTION
The 2009 restoration of the Nisqually River Delta in South Puget Sound,
Washington, USA, completed a multi-stage restoration effort that reconnected over 300
hectares of previously-diked land to tidal inundation, allowing for the reclamation of
approximately 75% of historic tidal marsh habitat (Barham, 2010). The removal of the
historic Brown Farm Dike at Nisqually National Wildlife Refuge represents the largest
estuary restoration in the Pacific Northwest, and has increased salt marsh habitat in
Southern Puget Sound by 50% (Woo et al., 2011). Restoration was stimulated by the
need to protect habitat for migratory birds and enhance forage opportunity for salmon,
including the federally listed Nisqually fall Chinook salmon (U.S. Fish and Wildlife
Service, 2005). Four eelgrass beds are located near the mouth of the restored Nisqually
River Delta, each experiencing varying, indirect effects of restoration from alterations to
sediment transport and water flow rates (Davenport, 2012).
Zostera marina, or common eelgrass provide a wide range of ecological services
that benefit both humans and wildlife along coastal boundaries (Plummer et al. 2013).
Eelgrass plants produce and sequester high volumes of organic carbon, aid in nutrient
cycling and help stabilize fine sediments to reduce erosion (Boström et al. 2006; Duarte
& Chiscano 1999; Duarte et al. 2005). Eelgrass beds form a structurally complex habitat
that support a higher diversity and abundance of ecologically and economically important
species as compared to unvegetated habitats (Bell & Westoby 1986; Orth et al. 1984).
Eelgrass blades provide spawning substrate for marine forage fishes (Penttila 2007),
create a rich nursery habitat for juvenile organisms, and support outmigrating
1
anadromous fish during their transition from freshwater to saltwater (Beck et al. 2001;
Costa et al. 1994; Carr et al. 2011).
Eelgrass beds are highly sensitive to environmental stressors that alter light
attenuation, and are often the first to respond to habitat disturbances that alter light
availability in the nearshore environment (Short & Wyllie-Echeverria 1996). Increasing
development along coastal areas and increased pollution inputs have contributed to an
unprecedented decline of eelgrass species (Waycott et al. 2009). Moreover, this decline is
accelerating, and may have negative impacts on organisms that rely on these habitats
(Hyndes et al., 2003; Irlandi & Crawford, 1997). Eelgrass beds in Puget Sound have been
monitored since 2000 by the Washington Department of Natural Resources (WA DNR).
No long-term (multi-decadal) data exists for these stocks, but preliminary population data
(2000 – 2013) indicate that eelgrass beds in Puget Sound are following this global
negative trend, with some variability between years (Christiaen et al., 2015; Gaeckle, et
al., 2011; Orth et al., 2006).
Eelgrass beds that occur along estuary boundaries provide a valuable link between
freshwater and marine habitats for outmigrating juvenile Pacific salmon by offering
refuge from predators and providing a rich food base before fish enter the open ocean
(Takekawa et al., 2013). The largest proportion of prey consumed in eelgrass beds is
thought to come from epifaunal species that attach to eelgrass blades, though the full
extent of how salmon utilize the invertebrate and habitat resources of eelgrass beds is still
poorly understood. This study represents a preliminary effort to evaluate the biological
contribution of eelgrass beds to outmigrating salmon in the restored Nisqually River
Delta. To quantify this contribution, we sought to measure the availability and
2
composition of invertebrate communities found in the eelgrass beds on the Nisqually
Delta by assessing patterns of biodiversity and abundance of epifaunal invertebrates,
conducting key species analysis for several known invertebrate prey species, and
examining how site characteristics such as temperature, salinity and plant complexity and
density impact these patterns of diversity both spatially and temporally.
Assessing changes in the distribution and community characteristics of associated
invertebrate populations will provide a necessary baseline to quantify the forage
opportunity of resident eelgrass beds to salmon populations. Eelgrass epifaunal
communities have not yet been studied at Nisqually, despite growing evidence that these
nearshore habitats provide a vital transitional habitat for salmon growth. Understanding
these patterns will allow managers to better identify benchmarks for eelgrass recovery at
Nisqually. A comprehensive evaluation of epifaunal invertebrates on the Nisqually river
delta will inform adaptive management strategies to protect and enhance eelgrass habitat
that supports outmigrating salmon populations. Lastly, this project will contribute to a
more comprehensive understanding of food web interactions in delta habitats, allowing
for a targeted allocation of resources to species critical for salmon forage.
This project was conducted in collaboration with the U.S. Geological Survey
(USGS), U.S. Fish and Wildlife Service (USFWS) and the Nisqually Indian Tribe at the
Nisqually National Wildlife Refuge. This research served as a pilot project to evaluate the
ability of nearshore eelgrass habitats to provide enhanced opportunity to juvenile salmon
populations in Puget Sound. The study falls within the larger research aims of the
Nisqually National Wildlife Refuge’s plan to document post-restoration recovery rates
and habitat characteristics of the Nisqually River Delta.
3
II. LITERATURE REVIEW
Overview of Seagrasses
Seagrasses are a globally distributed class of marine flowering plants
characterized by long, narrow leaf blades that form underwater meadows along shallow,
coastal margins and within estuaries (Figure 1). Seagrass species have relatively low
species diversity with only 58 extant species worldwide (Larkum, 2007). These plants
evolved from a terrestrial plant ancestor 100 million years ago, and represent the only
plant that has successfully transitioned from land back into the marine environment
(Larkum, et al. 2007). Seagrass plants increase habitat complexity by forming dense,
meadow-like mats of submersed vegetation. These meadows support higher density and
abundance of invertebrates when compared to adjacent unvegetated habitat, and provide
critical habitat for many commercially and recreationally important species (Beck et al.,
2001). Seagrasses have some of the most substantial light requirements of any submersed
aquatic vegetation, requiring up to 29% of incident radiation to reach blades for growth
(Dennison et al., 1993; Short & Wyllie-Echeverria, 1996). As a result, they are acutely
sensitive to environmental changes that reduce light availability. Furthermore, seagrasses
may act as a ‘coastal canary’ to highlight preliminary impacts of environmental
degradation (Orth et al., 2006).
Seagrass beds are a critical driver of nearshore carbon sequestration. Despite
occupying 0.2% of the seafloor, seagrass species cumulatively store 50% of total organic
carbon (Corg) in ocean sediments and serve as one of the most effective carbon sinks on
Earth (Duarte et al., 2005; Waycott et al., 2009). Seagrasses are estimated to bury 27.4
4
tetragrams of Corg per year, or approximately 10% of yearly Corg sequestration in the
oceans (Fourqurean et al., 2012). It is estimated that seagrass ecosystems store between
4.2 – 8.4 petagrams of organic carbon overall, and possibly as much as 19.9 petagrams
(Fourqurean et al., 2012). For reference, one petagram represents over two trillion
pounds. In an environment where anthropogenic climate change threatens our ocean’s
ability to uptake CO2 through increased temperatures and ocean acidification, the
contribution of seagrass to the global carbon sequestration budget may become
increasingly important in the years to come (Doney, 2009).
Despite growing recognition of the importance of these habitats, seagrasses are
experiencing an unprecedented decline. Since the late 19th century, seagrass meadows
have declined in all areas of the globe where data exists (Waycott et al. 2009). Aerial
cover of seagrass beds has declined 29% since 1879, at a rate of 1.5% per year through
1980 (Fourqurean et al., 2012). The rate of this decline has accelerated since 1980 to 5%
loss of seagrass extent per year, and the causes of this acceleration are still poorly
understood (Hughes et al., 2008; Short et al., 2011; Short & Wyllie-Echeverria, 1996;
Waycott et al., 2009). Nearly 15% of all seagrass species are considered threatened by the
International Union for Conservation of Nature (IUCN) (Hughes et al., 2008; Short et al.,
2011), including eelgrass.
Life History of Eelgrass
Common eelgrass (Zostera marina) is the most prolific of six species of
seagrasses along the Pacific Coast of North America. Eelgrass plants grow in soft
substrates along shallow embayments and coastal margins ranging from the Bearing Sea
5
in Alaska to the Gulf of California (Larkum et al., 2007; Phillips, 1983). Eelgrass plants
propagate through the lateral spread of perennial root-rhizomes and annual seed dispersal.
Its leaves are long thin blades ranging from 2 – 20 millimeters wide and up to 53
centimeters long that radiate upward from the root rhizome (Phillips, 1983).
Figure 1. Technical illustration of Zostera marina blades, seeds and root rhizomes.
Reprinted from: The Families of Flowering Plants (website) (Watson & Dallwitz, 2015).
6
Eelgrass is a powerful ecosystem engineer that greatly enhances community
diversity and biomass. Much like terrestrial plants, rhizome root structures stabilize soft
sediment and reduce rates of erosion (Larkum, 2007). Eelgrass blades form a thick
understory of vegetation that provides refugia for clinging epifauna and macroalgae and
greatly enhances the diversity and biomass of fish and invertebrate species (Attrill et al.,
2000; Blackmon et al., 2006.; Edgar & Robertson, 1992; Orth et al., 1984; Curras et al.,
1993). For example, Carr et al. (2011) compared the abundance and diversity of
associated mesograzers with eelgrass shoot density over time in San Francisco Bay. They
found that eelgrass plants with flowering shoots supported higher densities of epifauna as
compared to non-flowering blades or unvegetated sediment.
Eelgrass shoots alter the physical characteristics of water parcels that interact with
eelgrass beds, further enhancing nursery habitat. Eelgrass blades reduce water flow rate,
wave action, and sediment re-suspension, trapping suspended sediment particles and
organic matter. This results in the largest sediment accretion rate of any aquatic
vegetation (Duarte et al., 2005; Orth et al., 2006). While some organic matter is
transported to adjacent environments, the majority is buried within these soft sediments,
and is a significant source of carbon sequestration among types of submerged aquatic
vegetation (Larkum, 2007; Duarte et al., 2005).
7
Figure 2. The eelgrass meadow: a world of microhabitats. Reprinted from Kelp and Eelgrass in
Puget Sound (Mumford, 2007).
Salmon Utilization of Eelgrass Habitat
Nearshore habitats such as eelgrass beds provide an important link between
freshwater and marine habitats for multiple species of juvenile Pacific salmon. These
habitats are especially important for ocean-bound species of Chinook salmon,
Oncorhynchus tshawytscha, which have a longer residency time in nearshore and
8
estuarine habitats than any other salmon species (Magnusson & Hilborn, 2003). Chinook
salmon are one of 27 stocks of salmon in Puget Sound, and are listed as threatened under
the Endangered Species Act (National Marine Fisheries Service Northwest Region,
2005). Eelgrass beds that occur along estuary boundaries are used by outmigrating
juvenile salmon to help enhance their early growth and development (Figure 3). During
salmon life cycles, the largest proportion of natural mortality for salmon occurs in the
first few months spent in the marine environment. Since predation dynamics are largely
size-selective, juvenile salmon that are able to rapidly increase their size are afforded a
competitive advantage during their early life cycle (Brodeur et al., 2007). Juvenile
salmon utilize the nearshore environment for most of the year, with eelgrass beds used
consistently from May through September (Thom, et al., 1989). Salmon are not bound by
their natal stream, and migrate extensively around Puget Sound to access different
nearshore areas (Shaffer, 2004).
Figure 3. Conceptual model of salmon habitat use in a restored estuarine wetland. The largest
proportion of prey consumed in eelgrass beds is thought to come from epifaunal species that
attach to eelgrass blades. Model created by USGS Western Ecological Research Center (2013).
9
Thom et al. (1989) evaluated the abundance of fish and prey species in four
habitat types in Puget Sound to evaluate the relative importance of different habitat types
to fisheries resources. For eelgrass beds, they found that salmon abundance varied over
time and showed a quick response to changes in preferred prey abundance. Chinook
salmon preferentially selected copepods, and overall prey biomass was positively
correlated with eelgrass biomass such that larger eelgrass plants supported greater
abundances of prey species. Chinook prey utilization on the Nisqually Delta nearshore
before estuary restoration varied with salmon body size, and was largely composed of
amphipods and copepods (Pearce, et al., 1982). Elsewhere in Puget Sound, euphausiid
shrimp were also an important prey species (Brodeur, 1990).
Patterns of Eelgrass Decline
In all coastal regions of the United States, eelgrass populations have experienced
large fluctuations in range and extent in response to environmental stressors over the last
100 years (Orth et al., 2006). North Atlantic eelgrass populations experienced a
debilitating epidemic in the 1930s that killed 90% of all eelgrass within two years (Short
& Wyllie-Echeverria, 1987). Many populations were completely extirpated along the east
coast of North America, ranging from Maine to Florida (Dexter, 1985). This wasting
disease was caused by a pathogenic form of Labyrinthula zosterae, a marine slime mold
that had significant economic and ecological impacts on dependent eelgrass species
(Dexter, 1985; Short et al., 1987).
Labyrinthula zosterae spreads through direct contact between infected blades. It
presents as black or brown spots that spread to cover eelgrass blades at a rate of 0.8mm
10
per hour. The pathogen acts as a secondary decomposer on living and senesced blades,
dramatically reducing photosynthetic capacity of the infected blades and decreasing plant
fitness (Larkum, 2007; Ralph & Short, 2002). Both pathogenic and non-pathogenic
strains of L. zosterae have been found on eelgrass beds in the Atlantic and Pacific oceans
(Short et al., 1987). Wasting disease continues to be observed in eelgrass meadows
throughout North America and Europe including the northeastern Pacific ocean, although
no infestation since has led to the precipitous declines observed in the 1930s (Ralph &
Short, 2002). Risk of infection is believed to be greater in plants that are already subject
to other stressors such as increased temperatures, disturbance or pollution (Larkum,
2007). As coastal areas are increasingly subject to environmental stressors, beds infected
with L. zosterae are at a heightened risk of widespread decline.
Loss of eelgrass beds from the 1930s collapse catastrophically altered nearshore
ecosystem processes and negatively impacted many dependent species (Short et al.,
1987). Areas with widespread eelgrass loss observed a decline of migratory waterfowl,
including the Brant goose Branta bernicla, which rely on eelgrass as a preferred forage
material (>80%) in the winter (Addy & Aylward, 1944), as well as the collapse of a
commercial scallop fishery in Chesapeake Bay. This decline also resulted in the first
recorded extinction of an eelgrass dependent limpet species (Lottia aleveus) (Larkum et
al., 2007). Eelgrass populations of the Atlantic largely recovered after 30 years, but
remained stagnant or nonexistent in areas where erosion was strengthened in absence of
eelgrass beds (Larkum et al., 2007). From the 1960s to present day populations have
declined from historic extent due to anthropogenic stressors caused by shoreline
development and nearshore pollution (Short et al., 2011). This decline is not evenly
11
distributed and can cause severe negative impacts on urbanized coastal ecosystems (Orth
et al., 2006).
Eelgrass Trends in Puget Sound
In Puget Sound, Washington, the greatest threat to eelgrass beds comes from
aggressive coastal development due to the rapid and accelerating urbanization of
metropolitan areas (Gaeckle, et al., 2011). Development-associated stressors to eelgrass
include construction of overwater structures such as docks and marinas that shade benthic
habitats, and shoreline armoring. Armored shorelines alter the energy budgets of
nearshore systems by disrupting habitat connectivity and flow of water and nutrients from
terrestrial to nearshore environments (Rehr et al., 2014). Armored shorelines are
frequently at a sediment deficit as compared to natural shorelines (Rehr et al., 2014).
Twenty-seven percent of shorelines in in Puget Sound are armored, thereby changing the
sedimentation pattern and characteristics of adjacent nearshore habitats (Shipman, et al.,
2009).
Eelgrass beds occupy approximately 43% of shorelines and 9% of the sea floor in
Puget Sound, ranging from 1 – 3,000 hectares in size (Gaeckle, et al., 2011). The Puget
Sound Partnership (PSP) has tracked eelgrass abundance and distribution since 2000 in a
comprehensive monitoring effort through the Submerged Vegetation Monitoring Project
(SVMP; Figure 4). In 2010, the PSP identified eelgrass as one of 20 dashboard
ecosystem indicators of ecosystem health using trend data compiled from the SVMP
(Gaeckle et al., 2011). The following year, PSP identified a target goal to increase
eelgrass extent by 20% in 2020, using a combination of passive and active restoration
12
methods (Gaeckle et al., 2011). In addition, the Washington Department of Fish and
Wildlife has designated eelgrass beds as a habitat of special concern under their authority
over hydraulic projects (WAC 220-110-250), and the Washington Department of
Ecology designated eelgrass areas as critical habitat through the Washington Shoreline
Management Act (WAC 172-26-221) (Gaeckle et al., 2011).
It is difficult to predict long-term trends without multi-decadal data, but
preliminary trends of eelgrass populations have been evaluated throughout Puget Sound.
Since 2000, the total area occupied by eelgrass in Puget Sound has remained stable. Sitewide trends are more difficult to interpret. In 2009, overall populations were stable, yet
site data showed a greater proportion of sites with declining eelgrass populations, which
suggests that a net loss of eelgrass area is being obscured by a few stable populations
(Gaeckle et al., 2011). From 2009 – 2013 site populations experienced an overall
increase, especially in South Puget Sound and at the site of two large-scale delta
restoration projects. Both the Skokomish River Delta and the Nisqually River Delta
showed significant increases in eelgrass bed extent, suggesting that changing
sedimentation patterns from dike removal are benefitting the nearshore environment
(Christiaen et al., 2015).
13
Figure 4. Site-wide decadal trends in eelgrass abundance, 2003 – 2013. Adapted from the Puget
Sound Submerged Vegetation Monitoring Project (Christiaen et al., 2015).
Eelgrass Restoration
As eelgrass populations continue to decline, research has increasingly shifted
towards methods to mitigate stressors and restore existing eelgrass beds (Orth et al.,
2006). The mechanics of eelgrass restoration bear many similarities to terrestrial
restoration projects, where seed dispersal and erosion are influenced by water currents
14
instead of air currents (Seddon, 2004). Seagrass restoration presents a unique challenge in
that restoration is often strongly limited by access, since these habitats are submerged
most of the time. Restoration projects involve costly and specialized equipment, often
requiring certified SCUBA divers to complete basic restoration techniques (Seddon,
2004).
Active restoration methods include physical interventions that are used to boost or
enhance eelgrass recruitment in degraded habitats by seeding or replanting diminished
eelgrass beds (Seddon, 2004). These methods have mixed success rates, and can produce
especially low survival rates when used as the only restoration intervention (Short &
Wyllie-Echeverria, 1996). Passive restoration include natural enhancements of eelgrass
habitat, through reductions to factors that stress or inhibit eelgrass growth and
recruitment by enhancing preferred habitat conditions (Seddon, 2004).
Moore & Short (2006) found that “improvement of water clarity is the single
greatest factor that will aid in the restoration of Zostera species.” Interventions to
improve water clarity include steps to reduce nutrient loading, eutrophication, and
stormwater runoff, or to limit sediment discharge into coastal waters (Short & WyllieEcheverria, 1996). Improving water clarity encourages eelgrass to expand laterally and
re-establish historic depth limits (Larkum, 2007). Research into restoration interventions
grew in interest during the 1930s in response to decimation of eelgrass populations by
wasting disease (Addy, 1947; Addy & Aylward, 1944), however the lack of plant
response to reseeding and planting efforts was acknowledged even then and has not
improved since (Fonseca, 2011).
15
Despite this stagnation in restoration success, scientists have broadened their
understanding of the key factors that influence success and failure rates in restoration
projects. McGlathery et al. (2012) mapped functional and structural recovery trajectories
in eelgrass beds by evaluating metrics of primary productivity, sediment deposition,
shoot density and plant biomass of large experimental beds in successive years.
Unvegetated beds were seeded in 0.4 hectare plots with a total of 4.4 million seeds.
Habitat characteristics were compared to adjacent unvegetated sites for 9 years. The
authors observed an initial 4-year delay in shoot density across all sites, with a linear
increase from 4 – 9 years. Compared to eelgrass beds at 3-4 years after planting, 9-year
old beds had 20 times more productivity, 2 times more organic matter, 3 times more
carbon and 4 times more nitrogen sequestered in progressively finer sediment. Such
results are encouraging for eelgrass restoration efforts.
Impact of Dikes on Coastal Habitats
Removal of shoreline armoring is one method used to enhance preferred eelgrass
habitat and restore habitat connectivity between terrestrial and marine environments
(Heerhartz & Toft, 2015; Shipman, et al., 2009). Diking along river deltas prevents the
formation of distributary channels that help to slow rates of coastal erosion by depositing
sediment along the delta mouth. Without these distributary channels, water enters the
delta at a faster rate, increasing turbulence and decreasing light penetration and water
clarity (Giesen, et al., 1990). Armored deltas limit the distribution of sediment both
behind and in front of the dike, and a higher percentage of sediment is lost offshore
(Stevens & Lacy, 2012).
16
Hood (2004) evaluated changes to adjacent diked habitat on the Skagit River,
Washington, and found that dikes are directly and indirectly responsible for estuarine and
nearshore habitat loss. Dike construction reduced the tidal prism, resulting in an increase
of channel filling and deposition along seaward dike boundaries, and a decrease in
channel sinuosity with enhanced erosion landward. Dikes are both directly and indirectly
responsible for estuarine and nearshore habitat loss. Coastal development has increased
over the past 200 years, primarily due human interventions to preserve and extend arable
land. These actions reduce tidal flow to a thin ring along manipulated boundaries, and
have significant negative impacts on delta extent and function of critical habitats
(Boumans, et al., 2002).
Removal of shoreline armoring has been shown to alter sedimentation dynamics,
water hydrology and geomorphology for adjacent nearshore environments, while
increasing estuary tidal prisms (Hood, 2004). Restoration of armored deltas modifies
patterns of sedimentation at river mouths, leading to accretions of soft sediment
characteristics in the nearshore environment (Davenport, 2012; McGlathery et al., 2012).
Restoration can also have a direct positive benefit to juvenile salmon species, especially
Pacific salmon (Oncorhynchus spp.) that utilize nearshore seagrass habitats as a staging
ground before heading to the open ocean (David et al., 2014; Gray et al., 2002; Lind-Null
& Larsen, 2010; Shreffler et al., 1992).
Initial trends from the 2009 Nisqually Delta restoration indicate that the tidal flat
geomorphology has changed since restoration. Sediment grain size within Nisqually has
decreased with time, likely as a result of increased sediment transport through
reconnected tidal channels (Davenport, 2012). The addition of sediment to the nearshore
17
has the opportunity to support expanded eelgrass habitat. Preliminary trends of eelgrass
abundance indicate that the eelgrass beds at Nisqually are increasing in size following
dike removal, further reinforcing the positive benefit of dike removal in this estuary
environment (Christiaen et al., 2015).
18
III. METHODS
Study Area
The Puget Sound is a polyhaline fjord encompassing 1.6 million acres in western
Washington, USA. Freshwater input from multiple large river systems combines to create
a mixed-salinity system. Shorelines are moderate to steeply-sloped, and substrates range
from rocky shoreline common in northern Puget Sound to softer sand and clay sediments
in southern Puget Sound. Puget Sound coastlines are primarily nourished by inputs from
nearby feeder bluffs or deposition of sediment from river deltas, which contribute to
support a diverse assemblage of flora and fauna, including five species of commercially
and culturally important Pacific salmon (Oncorhynchus spp.) and top predator species
such as orca whales, California sea lions, and seals.
The Nisqually River Delta is a river valley estuary located at the southern end of
Puget Sound. The Nisqually River originates from the Nisqually Glacier on Mount
Rainier, and flows northwest for 125 kilometers before entering into Puget Sound
(47.08°N, 122.70°W; Figure 5). As the largest river flowing into southern Puget Sound,
and the largest restored estuary in this region, the Nisqually River Watershed covers
approximately 1,900 square kilometers and includes one national park, a United States
military base, two counties, and a rapidly-growing urban population (Karlstrom, 1971).
The Nisqually River Delta lies within the boundaries of the Nisqually National Wildlife
Refuge, which was designated as a national natural landmark in 1971 due to its
ecological significance. It is operated by the U.S. Fish and Wildlife Service (USFWS)
19
and managed by both the USFWS and the Nisqually Indian Tribe (U.S. Fish and Wildlife
Service, 2005).
In 1905, construction of the 5-mile Brown Farm Dike converted emergent marsh
and tidal slough habitat into agricultural and pasture land, isolating 600 hectares of
wetland habitat from tidal influence. Land behind the dike subsided over time due to a
lack of sediment inputs and decomposition of previously submerged organic matter such
as peat (Barham, 2010).
Figure 5. The Nisqually River Delta, Washington with study area and refuge boundary identified. Map
created by Sierra Blakely, May 2015, ESRI Basemap.
Removal of five-miles of earthen dike in October 2009 restored 308 hectares of
previously diked habitat to tidal inundation, allowing reclamation of 75% of historic tidal
habitat. Restoration of the Nisqually River Delta reconnected over 35 kilometers of
20
historic tidal channels, improved upstream riparian habitats and enhanced habitat for
wildlife, migratory birds and federally threatened salmon including the Nisqually fall
Chinook stock. The Nisqually Delta restoration increased salt marsh habitat in Southern
Puget Sound by 50%, and ranks as the largest estuary restoration in the Pacific Northwest
(Davenport, 2012, Figure 6).
Figure 6.Nisqually National Wildlife Refuge restoration map, created by J. Cutler, Nisqually Indian Tribe.
We conducted eelgrass site surveys at four Zostera marina beds located along the
Nisqually River Delta: Hogum Bay (HGB), McAllister Creek (MCA), Red Salmon
Slough (RSS) and Cormorant Passage (CMP) (Figure 7). These eelgrass beds were
identified by the Washington Department of Natural Resources as part of the Submerged
Vegetation Monitoring Project (Gaeckle et al., 2011) and are positioned along the West
21
and East sides of the Nisqually Delta. McAllister Creek is located on the West side of the
Nisqually Delta at the mouth of McAllister Creek, a spring-fed creek (47.06°N,
122.43°W). Red Salmon Slough is located on the East Side of the Nisqually Delta at the
mouth of Red Salmon Slough on the Nisqually River (47.06°N, 122.41°W). Hogum Bay
is located northwest of the Delta and just north of an aquaculture farm operated by Taylor
Shellfish Farm (47.07°N, 122.44°W). Cormorant Passage is the farthest site from the
Delta, and occurs along a steeper shoreline closer to the main channel opposite Anderson
Island (47.09°N, 122.37°W).
Figure 7. Location and approximate size of Nisqually River Delta eelgrass beds. Blue imagery
represents 2014 aerial imagery of delta landforms, ESRI basemap used. Map created by Sierra
Blakely, February 2015.
All eelgrass samples were collected by the U.S. Geological Survey Western
Ecological Research Center (USGS WERC) as part of the Estuary and Salmon
22
Restoration Program (ESRP), an ongoing monitoring project that aims to assess the
capacity of the Nisqually delta to support invertebrate prey following restoration. We
chose sites to align with fish capture sites used by the Nisqually Indian Tribe for stock
assessment purposes. McAllister Creek was sampled from March through September
2014, and the additional three sites (Hogum Bay, Red Salmon Slough & Cormorant
Passage) were sampled from May through July 2014.
Experimental Design and Data Analysis
We measured temperature and salinity six times per site using a YSI Pro 2030
(Yellow Springs, Ohio USA), and averaged these values across each sample. Shoot
density was collected once in June 2014 by Steve Rubin (USGS Western Fisheries
Research Center), from the center of each eelgrass bed during low tide when the beds
were exposed. To determine changes in biophysical variables, we used a general
hierarchical cluster analysis to evaluate each site’s compositional response to specific
predictor variables including temperature, salinity, number of nodes and mean blade
length. Month was used as a covariate to account for sampling variation and seasonal
differences. We used a multivariate analysis of variance (MANOVA) to identify which
predictor variables explained the most variation between sites. Prior to running each
analysis, a histogram of the data distribution was evaluated, and a Shapiro-Wilk
normality test was run to evaluate normality. Data were log-transformed when necessary.
Several candidate models were compared using the Akaike Information Criterion (AIC)
to determine the best fit model. Model selection was performed using R version 3.1.3 (R
Core Team, 2015) and Excel.
23
Plant Collection
We collected ten eelgrass plants per site during the lowest tide of the sampling
cycle (0.0 to -3.42 feet, Mean lower low water). Eelgrass shoots collected in March and
April were gathered opportunistically from plants caught on the boat anchor, as the
combined water and air temperature was too cold to safely collect samples by hand. From
May and continuing through September, plants were collected by hand by separating the
rhizome from the shoot at the sediment surface, with care taken to avoid jostling the
plants to reduce loss of epifauna during collection. No distinction was made between
vegetative and flowering eelgrass shoots. Collected shoots were transferred to an 8-ounce
jar filled with ambient seawater, then placed into a cooler until processing. All samples
were processed within 24 hours of collection. Once rinsed for invertebrates, each eelgrass
plant was measured for blade length (mm) and number of nodes per plant.
Invertebrate Collection
We processed epifaunal samples following the protocol established by Carr et al.
(2011) and Holmlund et al. (1990). Each jar was emptied onto a 500 µm sieve, before the
eelgrass shoot was placed in three 1-minute freshwater baths to remove clinging
epifaunal. This technique has been shown to remove 92 – 100% of epifauna from algae
and 90% from eelgrass plants. We poured each freshwater bath through a 500 µm sieve
to capture invertebrates, which were preserved in 95% ethanol and identified to the
lowest possible taxa by technicians at the USGS San Francisco Bay Estuary field station
in Vallejo, California.. We used a subset (75) of the total samples that had been processed
24
by February 2015. That subset contained 29,504 invertebrates, identified to various
taxonomic levels.
We used a multiple analysis of variance (MANOVA) to identify patterns of
invertebrate community characteristics between sites. Amphipods, copepods, nematodes,
ostracods, polychaetes, and tanaids were selected as key species groups, due to greater
abundances of these groups across all months, as well as their established function as
prey for juvenile salmon (Brennan, et al., 2004). To understand which predictor variables
(temperature, salinity, mean blade length, number of nodes and mean shoot density) had
the strongest effect on community structure, we used generalized linear models to
compare both the additive and interactive effects of our predictor variables. Both
temperature and salinity showed a unimodal distribution and were therefore compared
using a polynomial model. All candidate models were compared using AIC to determine
which site variables were the best predictors of species abundance.
Invertebrate community similarity was calculated using multi-response
permutation procedures (MRPP) for each site using PC ORD (McCune & Mefford,
2011), and a Sorenson distance measure. The primary matrix of species abundances was
relativized by species maximum to control for the effect of overabundant species in our
results. Due to the non-normality of our data, we performed a non-metric
multidimensional scaling (nMDS) using PC ORD to linearize the relationships of species
abundance and our predictor variables of site, temperature, salinity, mean blade length,
number of nodes per plant and mean shoot density.
25
IV. RESULTS
Community Structure
Six unique phyla of epifaunal invertebrates were identified from our samples
collected March – July 2014. Twenty five unique taxa were observed at Cormorant
Passage, 29 unique taxa observed at McAllister Creek, 28 unique taxa observed at
Hogum Bay and 19 unique taxa observed at Red Salmon Slough (Appendix A). We
observed a shift in abundance across all sites, from a community dominated by
arthropods in March through May, to a community dominated by nematodes, annelids
and molluscs in June and July (p<0.001, Figures 9a-b).
Figure 9a. Percent community composition of total phylum abundance observed.
26
Figure 9b. Percent community composition of total species abundance observed to lowest
taxonomic identification level.
Abundance varied significantly by both site and time (Figure 10). Hogum Bay
had significantly higher abundances than any other site, and supported abundances at
least one magnitude higher than those observed at other sites (p<0.05). This finding was
visualized using a hierarchical cluster analysis that aggregated the predictor values of site
and month to identify compositional responses. Hogum Bay in June was the most
divergent site for community composition, and McAllister Creek was the site most
similar across months (Figure 11). Mean epifaunal abundance increased between March
and July (p<0.05, Figure 10). Among months, March and April were showed the greatest
difference of phylum abundance from June and July. McAllister creek was the only site
sampled during March and April, therefore abundance for these months represent a
smaller sample size with less representation across the four sampling sites.
27
Figure 10. Mean epifaunal abundance by site and month, +/- s.e. Data was log-transformed for ease of
visualization. No July samples were counted for RSS.
Figure 11. Hierarchical clustering analysis of differences in phylum abundance by month and site.
Branches indicate degrees of separation using Euclidian distance measure.
28
We confirmed this significance using a multiple analysis of variance of site
characteristics of site + month + nodes + mean blade length (Table 1). While all site
characteristics exhibited statistical significance, site was the strongest predictor of
phylum abundance (p<0.0001).
A multi-response permutation procedure with Sorenson’s distance measure
showed a significant difference between sites (A=10462810, p<0.0001). We visualized
this using non-metric multidimensional scaling (nMDS) to linearize the relationship of
species abundance and site, temperature, salinity, mean blade length, nodes and mean
shoot density. nMDS showed a strong distinction between groups in the species space
(Figure 12).
Figure 12. nMDS ordination analysis of taxa abundance by site characteristics using
Sorenson’s (Bray-Curtis) distance measure. MCA = Green triangles, HGB = Blue triangles,
CMP = Red triangles, RSS = Pink triangles.
29
Biodiversity
Biodiversity varied significantly across site and time (Figure 13). An ANOVA of
site, month, number of nodes and mean blade length showed that month was the most
significant predictor variable of biodiversity (p<0.005, Table 2). Mean blade length and
the interactive effect between month x site were also significant (p<0.01). Mean
biodiversity at McAllister Creek was on average 40% higher than any other site, and this
relationship differed depending on month. McAllister Creek experienced the lowest
biodiversity during May and June, while Cormorant Passage, Hogum Bay and Red
Salmon Slough had the greatest diversity during those months. Analysis using AIC shows
that a combination of predictor variables of month, site, nodes and mean blade length
explains more of the pattern of biodiversity than month alone (Table 3).
Figure 13. Mean sample diversity (as measured by Shannon’s Diversity Index) for epifaunal
abundance by site.
30
Key Species Analysis
Arthropoda
Arthropods exhibited both highest overall abundances and contained the largest
number of individuals identified to a lower taxonomic level, allowing for additional
analysis of community structure for this phylum. Arthropod abundance increased in May
and remained high through July, with the exception of McAllister Creek, where arthropod
abundance dipped in June. Hogum Bay populations were two magnitudes higher than any
other site (Figure 14). Arthropod abundance varied seasonally by site, and was positively
correlated with number of nodes per plant (p<0.05, Table 4). The interactive effect of
month, site and nodes was also found to be significant, but was rejected due to the small
sample size and high volume of null values in the generalized linear model.
Figure 14. Mean Arthropoda abundance by site & month, +/- s.e.
31
Arthropod community structure varied significantly across months (p<0.05).
March and April were dominated by high densities of ostracods and tanaids, which
shifted shifted to a population dominated by high densities of copepods, caprellids and
amphipods from May through July (p<0.05, Figure 15).
Figure 15. Percent composition of Arthropoda taxa by month across all sites and sampling
months.
Amphipoda
Amphipod abundance varied seasonally, and the degree of variation was
dependent upon site. Populations were low in March and April, and then increased to a
maximum mean abundance of 687 individuals in July. Abundance varied dramatically by
site, with lowest overall abundance observed at Cormorant Passage. Hogum Bay had the
highest overall abundances, and was two orders of magnitude higher (p<0.005, Figure
16). Analysis of candidate generalized linear models showed that site was the most
32
significant predictor variable in amphipod abundance (p<0.05, Table 5). We observed a
positive relationship between blade length and amphipod abundance. Any difference in
amphipod abundance due to number of nodes was explained by site differences, since
node number varied widely across sites from a minimum mean of 6.5 nodes at Cormorant
Passage to a maximum mean of 7.6 nodes at Red Salmon Slough (p<0.05).
Figure 16. Mean Amphipoda abundance by site and month, +/- s.e.
Copepoda
Copepod abundance varied significantly over time, and the degree of this
variation was dependent on site (p<0.05). Populations at Cormorant Passage and
McAllister Creek steadily increased across the sampling period, while Hogum Bay
populations declined in July. Red Salmon Slough abundance was only recorded for May,
although this is likely due to a lack of samples processed for July at this time (Figure 17).
Copepod abundance was positively correlated with blade length and number of nodes. A
33
full interactive model of month x site x nodes was indicated as a better fit for abundance
patterns, but was rejected due to a small sample size that limited statistical certainty
(Table 6).
Figure 17. Mean Copepoda abundance by site & month, +/- s.e.
Polychaeta
Polychaete abundance varied significantly over time, and the degree of this
variation was dependent on site (p<0.05). Abundance increased through time for
Cormorant Passage, McAllister Creek and Red Salmon Slough, while abundance peaked
in June at Hogum Bay (Figure 18). The interactive model of month, site and nodes was
the best predictor of polychaete abundance (Table 7).
34
Figure 18. Mean Polychaeta abundance by site & month, +/- s.e.
Ostracoda
McAllister Creek was sampled over a longer period of time, and exhibited two
peaks in abundance both early in the season (March/April) and late in the season (July)
with low abundances of ostracods between those months. Conversely, Cormorant
Passage, Hogum Bay and Red Salmon Slough only had ostracods present in June, when
abundance was lowest at McAllister (Figure 19). Ostracod abundance varied over time,
and the degree of this variation was again dependent on site differences. A full interactive
model incorporating month x site + mean blade length + salinity2 was a better fit based
on the AIC index, but this model was rejected due to a high proportion of blanks in the
output model summary, which suggests that the sample size for this analysis was too
small.
35
Figure 19. Mean Ostracoda abundance by site & month, +/- s.e.
Tanaidacea
Tanaid abundance followed a similar pattern to that of ostracods. McAllister
Creek was the only site to exhibit two peaks of abundance early (March/April) and later
in the season (July) with low abundances in the intervening months. Cormorant Passage,
Hogum Bay and Red Salmon Slough all peaked in abundance in June or July, while
McAllister Creek abundances were 2 magnitudes greater than the other three sites (Figure
20). This variation was dependent on an interaction effect of month and site, and was
positively correlated to mean number of nodes (p<0.05).
36
Figure 20. Mean Tanaidacea abundance by site & month, +/- standard error.
Biophysical Variables by Site
Temperature increased steadily from March through July across all sites (Figure
21). We observed a significant change in water temperature through time that was
strongly dependent on site (p<0.01). A one-way analysis of variance showed that these
seasonal differences varied by site (Table 9). McAllister Creek showed the greatest
variation over time of any other site (p<0.05). Temperature through time was dependent
on site, and month was the greatest predictor of temperature.
37
Figure 21. Mean temperature by site and month. Error bars signify +/- s.e.
Salinity varied significantly over time by month and site (p<0.01, Figure 22). An
overall trend towards decreasing salinity measurements supports our assumption that sites
located closer to the Nisqually Delta (McAllister Creek & Red Salmon Slough)
experienced greater freshwater input than sites located farther from the Nisqually River
and McAllister Creek (Cormorant Passage & Hogum Bay). A one-way ANOVA found
that both month (p<0.05) and site (p<0.01) were significant predictors of salinity (Table
11). Of the candidate linear models used, the full interactive model using month x site
was the best predictor of salinity at each site (p<0.05; p<0.01). We conducted the same
analysis for node number and mean blade length, and found no statistically significant
results, indicating that month and site were not significant predictors of eelgrass node
number or mean blade length.
38
Figure 22. Mean Salinity by site and month. Error bars signify +/- standard error.
39
V. DISCUSSION
Site Variations
Epifaunal communities play an important role in providing foraging opportunity
for outmigrating juvenile salmon in Puget Sound In this study, we observed an abundance
of epifaunal invertebrates across all four sampling sites, including known salmon prey
species such as amphipods, copepods, ostracods, tanaids and polychaetes. These
abundances varied widely for each species through time, and were not evenly distributed
across sites. Among the four eelgrass beds sampled, Hogum Bay exhibited the greatest
difference in monthly invertebrate community structure with peak densities in June, while
the remaining sites of McAllister Creek, Cormorant Passage and Red Salmon Slough saw
increases in available prey biomass throughout the sampling period.
Hogum Bay site characteristics of mean shoot density, nodes, mean blade length,
temperature and salinity were all similar to the biological ranges observed at the other
three eelgrass beds. This suggests that there are other drivers beyond site characteristics
that support higher epifaunal abundances. Hogum Bay is adjacent to 300 acres of
commercial shellfish aquaculture property operated by National Fish and Oyster that
produces oysters, manila clams and geoduck. Studies have shown that filter feeders such
as bivalves exert a top-down control on aquatic vegetation by removing phytoplankton
and particulate organic matter from the water column, thereby increasing light
penetration and resulting in a more amenable environment for aquatic plant growth
(Newell, 2004). This enhancement could be responsible for the greater surface area and
epifaunal invertebrate abundances at Hogum Bay.
40
Our site at McAllister Creek is also located adjacent to the aquaculture farm;
however, this site did not support epifaunal abundances as great as Hogum Bay.
Invertebrate densities at McAllister Creek were at least one magnitude higher than at
Cormorant Passage or Red Salmon Slough. This site is also impacted by its proximity to
the mouth of the spring-fed McAllister Creek, which runs north into Puget Sound. Unlike
rivers, spring-fed creeks supply a seasonally constant source of freshwater, which may
limit any impact of the shellfish aquaculture from interacting with this site. This was
reinforced by the ordination results of our site epifaunal communities, which confirmed a
clear separation of McAllister Creek site from the other delta sites. Anecdotal
observations indicated that water clarity was lowest at McAllister Creek, likely due to the
constant current from the creek mouth and associated turbidity. Further studies that assess
light penetration depth in Nisqually eelgrass beds are recommended for a clearer
understanding of these dynamics.
Patterns of Epifaunal Diversity
Eelgrass beds on the Nisqually River Delta varied strongly in epifaunal
community composition and total abundance, and this abundance varied widely over time
and among all sites. Overall community structure was characterized by a greater
abundance of relatively few dominant taxa per site, and followed patterns observed at
eelgrass sites in the San Francisco Bay Estuary (Carr & Boyer, 2011). Abundance was
significantly impacted by the interactive effect between time of year and site for all
invertebrates. Taxa dominance can be attributed to a number of factors of biological
factors that can facilitate the availability of select species during different months.
Biological communities are largely influenced by the accessibility of prey resources and
41
nutrient sources. Many ostracod species have greater abundances during March and April
with a dip in populations during May and June when species reproduce (Hull, 1997) We
observed this pattern with ostracod populations across our four eelgrass sites, although
ostracod populations did not regain dominance later in the season. Rates of development
for epifaunal species have also been linked to temperature fluctuations, where increased
temperatures can negatively affect hatch and development rates for juveniles due to
increasing water temperature and salinity fluctuations over the spring to summer period
(Hull, 1997).
Suggestions for Further Research
This project serves as a starting point for examining post-restoration epifaunal
invertebrate abundance within and among Nisqually eelgrass beds. These data have a
variety of applications to help strengthen the link between invertebrate prey communities
and patterns of juvenile salmon abundance in nearshore vegetated habitats, but additional
studies are needed to identify the direct contribution of eelgrass prey sources to salmon
during periods of outmigration. Lampara netting is a technique used to sample fish in
nearshore areas inaccessible to beach seining, and was conducted at all eelgrass sites
during 2014. It is our hope that catch data can be compared to our findings of epifaunal
species composition to evaluate whether juvenile salmon occur in eelgrass beds during
periods when preferred prey species are present in high abundances. Fish gut contents
were also collected in 2014, and have the potential to be used in gut content analysis to
calculate a percent similarity index of observed prey species within eelgrass beds and
salmon diets. Since salmon have been shown to move rapidly into habitats with greater
preferred prey abundances, we would expect to see an increase in salmonid abundance
42
during periods dominated by high abundances of amphipods and copepods (Brennan et
al., 2004).
Eelgrass bed size was mapped at Nisqually during 2014, and could be used to
explore the relationship between eelgrass patch size and epifaunal abundance, which has
the potential to inform which eelgrass beds may provide the greatest benefit to resident
salmon for prioritizing ongoing management projects. The UGSG is sampling the same
parameters of epifaunal invertebrates and eelgrass site characteristics for 2015, which
will allow for a multi-year comparison of abundance and diversity that may help predict
how Nisqually eelgrass invertebrate communities change over time.
The abundance of amphipods, copepods and polychaetes observed in May
through July overlaps with periods of delta utilization during salmon outmigration (May
through September), and supports our conclusion that eelgrass beds of the Nisqually delta
have the potential to provide a valuable source of prey for juvenile salmon that utilize
these habitats as a transitional habitat during outmigration. We can expect this forage
opportunity to increase through time, as eelgrass bed extent continues to increase
following the 2009 delta restoration (Christiaen et al., 2015). These data reinforce the
ecological value of restoration of these habitats for the management and support of
threatened salmon stocks in Puget Sound.
43
Tables
Table 1. Multivariate Analysis (MANOVA) of the effects of site, month, nodes and mean
blade length on epifaunal invertebrate abundance.
Month
Site
MSD
Nodes
Df
4
3
1
1
Pillai
1.4263
1.4686
0.32094
0.49584
Approx F
1.9234
3.2717
1.6125
3.3555
Num DF
68
51
17
17
Den DF
236
174
58
58
Pr(>F)
0.0003129*
3.943e-09 *
0.09059
0.0002946 *
Table 2. One-way ANOVA output for Shannon’s biodiversity index and site
characteristics of month, site, nodes, mean blade length and Month x Site. Asterisk
denotes significant values.
Month
Site
Nodes
Mean Blade Length (MBL)
Month x Site
Month x MBL
Site x MBL
Month x Site x MBL
Df
4
3
1
1
5
4
3
3
Sum Sq
1.918
0.14
0.138
0.937
1.546
0.528
0.505
0.17
Mean Sq
0.4794
0.0468
0.1383
0.9372
0.3093
0.132
0.1683
0.0567
F Value
5.171
0.505
1.492
10.108
3.335
1.281
1.634
0.551
Pr(>F)
0.00118*
0.68055
0.22659
0.00232*
0.00996*
0.28954
0.19283
0.64997
Table 3. AIC model selection of one-way ANOVA fit of Shannon’s Biodiversity Index.
Asterisk denotes best fit model.
Df
Month x Site
14
Month x Site + Nodes
15
Month x Site x Nodes
25
Month x Site + MBL
15
Month x Site x MBL
25
Month x Site + Nodes + MBL 16
AIC
58.02555
54.4779
62.82086
59.51478
64.12012
50.22716 *
44
Table 4. AIC test of linear fit models of influences on Arthropoda abundance. Asterisk
denotes best fit model.
Site
Month + Site
Month x Site
Nodes + Site
Nodes x Site
MBL + Site
MBL x Site
Df
2
8
13
5
8
5
8
AIC
15269.89
15133.92
12683.14
14101.55
12361.37 *
18213.26
17577.63
Table 5. AIC model selection of linear fit models of influences on Amphipoda
abundance. Asterisk denotes best fit model.
Df
AIC
Site
4 5931.25
Month + Site
8 4218.57
Month x Site
13 2565.841*
Nodes + Site
5 5294.481
Nodes x Site
8 5262.67
MBL+ Site
5 5932.53
MBL x Site
8 5906.374
Table 6. AIC model selection of linear fit models of influences on Copepod abundance.
Asterisk denotes best fit model.
Df
AIC
Site
4 6648.059
Month + Site
8 5454.787
Month x Site
2 3837.772 *
Nodes + Site
13 4898.091
Nodes x Site
5 6368.736
MBL + Site
8 8725.354
MBL x Site
5 6633.271
45
Table 7. AIC test of linear fit models of influences on Polychaeta abundance. Asterisk
denotes best fit model.
Df
AIC
Nodes
2 18231.517
Month x Site
12 3818.25
Month x Site + Nodes
8 8735.551
Month x Site + MBL
14 3737.537
Month x Site x Nodes
24 3054.403 *
Month x Site + MBL
14 3820.183
Month x Site + Temp^2
15 3301.505
Table 8. One-way ANOVA of the effect of month, site and month x site on temperature
Month
Site
Month x Site
Df
F
4 91.248
3 31.914
5 6.646
P
< 2e-16 *
1.14e-12 *
5.00e-05 *
46
Literature Cited
Addy, C. E. (1947). Eel grass planting guide. Maryland Conservationist, (24), 16–17.
Addy, C. E., & Aylward, D. A. (1944). Status of Eelgrass in Massachusetts during 1943. The
Journal of Wildlife Management, 8(4), 269–275. http://doi.org/10.2307/3796019
Barham, J. (2010, October). Nisqually National Wildlife Refuge: Estuary Restoration. Oral
Presentation presented at the OSU Spotlight on Science, Portland, OR. Retrieved from
https://media.oregonstate.edu/media/t/0_wzwaqh9o
Beck, M. W., Heck, K. L., Able, K. W., Childers, D. L., Eggleston, D. B., Gillanders, B. M., …
Weinstein, M. P. (2001). The Identification, Conservation, and Management of Estuarine
and Marine Nurseries for Fish and Invertebrates. BioScience, 51(8), 633–641.
http://doi.org/10.1641/0006-3568(2001)051[0633:TICAMO]2.0.CO;2
Bell, J. D., & Westoby, M. (1986). Abundance of Macrofauna in Dense Seagrass Is Due to
Habitat Preference, Not Predation. Oecologia, 68(2), 205–209.
Boström, C., Jackson, E. L., & Simenstad, C. A. (2006). Seagrass landscapes and their effects on
associated fauna: A review. Estuarine, Coastal and Shelf Science, 68(3–4), 383–403.
http://doi.org/10.1016/j.ecss.2006.01.026
Boumans, R. M. J., Burdick, D. M., & Dionne, M. (2002). Modeling Habitat Change in Salt
Marshes After Tidal Restoration. Restoration Ecology, 10(3), 543–555.
http://doi.org/10.1046/j.1526-100X.2002.02032.x
Brennan, J. S., Higgins, K. F., Cordell, J. R., & Stamatiou, V. A. (2004). Juvenile Salmon
Composition, Timing, Distribution, and Diet in Marine Nearshore Waters of Central
Puget Sound in 2001-2002 (p. 164). Seattle, WA: King County Department of Natural
Resources and Parks.
Brodeur, R. D. (1990). A Synthesis of the Food Habits and Feeding Ecology of Salmonids in
Marine Waters of the North Pacific (Technical Report No. FRI-UW-9016) (p. 43).
FIsheries Research Institute: University of Washington.
Brodeur, R. D., Daly, E. A., Sturdevant, M. V., Miller, T. W., Moss, J. H., Thiess, M. E., …
Norton, E. C. (2007). Regional Comparisons of Juvenile Salmon Feeding in Coastal
Marine Waters off the West Coast of North America. American Fisheries Society, 57,
183–203.
Carr, L. A., & Boyer, K. A. (2011). Spatial patterns of epifaunal communities in San Francisco
Bay eelgrass (Zostera marina) beds. Marine Ecology, 32(1), 88 – 103.
http://doi.org/10.1111/j.1439-0485.2010.00411.x
47
Christiaen, B., Dowty, P., Ferrier, L., Berry, H., Hannam, M., & Gaeckle, J. (2015). Puget Sound
Submerged Vegetation Monitoring Program 2010 - 2013 Report (Technical Report) (p.
61). Olympia, WA: Washington State Department of Natural Resources Aquatic
Resources Division. Retrieved from
http://www.eopugetsound.org/sites/default/files/DNR_SVMP_2013_03_13.pdf
Costa, M. J., Costa, J., de Almeida, P. R., & Assis, C. A. (1994). Do eel grass beds and salt marsh
borders act as preferential nurseries and spawning grounds for fish? An example of the
Mira estuary in Portugal. Ecological Engineering, 3(2), 187–195.
http://doi.org/10.1016/0925-8574(94)90045-0
Davenport, A. (2012, May). Modeling Geomorphic Effects on Eelgrass Before and After
Restoration, Nisqually Delta, Washington. San Francisco State University, San Francisco,
Calif.
David, A. T., Ellings, C. S., Woo, I., Simenstad, C. A., Takekawa, J. Y., Turner, K. L., …
Takekawa, J. E. (2014). Foraging and Growth Potential of Juvenile Chinook Salmon after
Tidal Restoration of a Large River Delta. Transactions of the American Fisheries Society,
143(6), 1515–1529. http://doi.org/10.1080/00028487.2014.945663
Dennison, W. C., Orth, R. J., Moore, K. A., Stevenson, J. C., Carter, V., Kollar, S., … Batiuk, R.
A. (1993). Assessing water quality with submersed aquatic vegetation. BioScience, 43(2),
86–94. http://doi.org/10.2307/1311969
Dexter, R. W. (1985). Changes in the standing crop of eelgrass, Zostera marina L. at Cape Ann,
Massachusetts, since the epidemic of 1932. Rhodora, 87(851), 357–366.
http://doi.org/10.2307/23314553
Duarte, C. M., & Chiscano, C. L. (1999). Seagrass biomass and production: a reassessment.
Aquatic Botany, 65(1–4), 159–174. http://doi.org/10.1016/S0304-3770(99)00038-8
Duarte, C. M., Middelburg, J. J., & Caraco, N. (2005). Major role of marine vegetation on the
oceanic carbon cycle. Biogeosciences, 2(1), 1–8. http://doi.org/10.5194/bg-2-1-2005
Fonseca, M. S. (2011). Addy Revisited: What Has Changed with Seagrass Restoration in 64
Years? Ecological Restoration, 29(1-2), 73–81. http://doi.org/10.3368/er.29.1-2.73
Fourqurean, J. W., Duarte, C. M., Kennedy, H., Marbà, N., Holmer, M., Mateo, M. A., …
Serrano, O. (2012). Seagrass ecosystems as a globally significant carbon stock. Nature
Geoscience, 5(7), 505–509. http://doi.org/10.1038/ngeo1477
Gaeckle, J., Dowty, P., Berry, H., & Ferrier, L. (2011). Puget Sound Submerged Vegetation
Monitoring Project 2009 Report. Washington: Washington State Department of Natural
48
Resources Puget Sound Assessment and Monitoring Program. Retrieved from
http://www.dnr.wa.gov/Publications/aqr_eelgrass_svmp_report.pdf
Giesen, W. B. J. T., van Katwijk, M. M., & den Hartog, C. (1990). Eelgrass condition and
turbidity in the Dutch Wadden Sea. Aquatic, 37(1), 71–85.
Gray, A., Simenstad, C. A., Bottom, D. L., & Cornwell, T. J. (2002). Contrasting Functional
Performance of Juvenile Salmon Habitat in Recovering Wetlands of the Salmon River
Estuary, Oregon, U.S.A. Restoration Ecology, 10(3), 514–526.
http://doi.org/10.1046/j.1526-100X.2002.01039.x
Heerhartz, S. M., & Toft, J. D. (2015). Movement patterns and feeding behavior of juvenile
salmon (Oncorhynchus spp.) along armored and unarmored estuarine shorelines.
Environmental Biology of Fishes, 1–11. http://doi.org/10.1007/s10641-015-0377-5
Hood, W. G. (2004). Indirect environmental effects of dikes on estuarine tidal channels: Thinking
outside of the dike for habitat restoration and monitoring. Estuaries, 27(2), 273–282.
http://doi.org/10.1007/BF02803384
Hughes, A. R., Williams, S. L., Duarte, C. M., Heck, K. L., & Waycott, M. (2008). Associations
of concern: declining seagrasses and threatened dependent species. Frontiers in Ecology
and the Environment, 7(5), 242–246. http://doi.org/10.1890/080041
Hull, S. L. (1997). Seasonal changes in diversity and abundance of ostracods on four species of
interidal algae with differing structural complexity. Marine Ecology Progress Series,
161, 71–82.
Hyndes, G. A., Kendrick, A. J., MacArthur, L. D., & Stewart, E. (2003). Differences in the
species- and size-composition of fish assemblages in three distinct seagrass habitats with
differing plant and meadow structure. Marine Biology, 142(6), 1195–1206.
http://doi.org/10.1007/s00227-003-1010-2
Irlandi, E. A., & Crawford, M. K. (1997). Habitat Linkages: The Effect of Intertidal Saltmarshes
and Adjacent Subtidal Habitats on Abundance, Movement, and Growth of an Estuarine
Fish. Oecologia, 110(2), 222–230.
Karlstrom, E. L. (1971). Notes on the Marine Biology of the Nisqually, the Outer Flats, Delta
Front, and Reach.
Larkum, A. W. D. (2007). Seagrasses Biology, Ecology and Conservation. Springer.
Lind-Null, A., & Larsen, K. (2010). Otolith Analysis of Pre-Restoration Habitat Use by Chinook
Salmon in the Delta-Flats and Nearshore Regions of the Nisqually River Estuary (U.S.
Geological Survey Open-File Report No. 2010-1238) (p. 28). Olympia, Washington.:
U.S. Geological Survey.
49
Magnusson, A., & Hilborn, R. (2003). Estuarine Influence on Survival Rates of Coho
(Oncorhynchus kisutch) and Chinook Salmon (Oncorhynchus tshawytscha) Released
from Hatcheries on the U. S. Pacific Coast. Estuaries, 26(4), 1094–1103.
McCune, B., & Mefford, M. J. (2011). PC-ORD. Multivariate Analysis of Ecological Data.
(Version 6). Glendon Beach, Oregon, U.S.A.: MjM Software.
McGlathery, K. J., Reynolds, L. K., Cole, L. W., Orth, R. J., Marion, S. R., & Schwarzschild, A.
(2012). Recovery trajectories during state change from bare sediment to eelgrass
dominance. Marine Ecology Progress Series, 448, 209–221.
http://doi.org/10.3354/meps09574
National Marine Fisheries Service Northwest Region. (2005). 5-Year Review: Summary and
Evaluation of Puget Sound Chinook, Hood Canal Summer Chum, Puget Sound Steelhead
(Technical Report) (p. 51). Portland, OR.
Newell, R. I. E. (2004). Ecosystem influences of natural and cultivated populations of
suspension-feeding bivalve molluscs: a review, 23(1), 51–61.
Orth, R. J., Carruthers, T. J. B., Dennison, W. C., Duarte, C. M., Fourqurean, J. W., Heck, K. L.,
… Williams, S. L. (2006). A Global Crisis for Seagrass Ecosystems. BioScience, 56(12),
987–996. http://doi.org/10.1641/0006-3568(2006)56[987:AGCFSE]2.0.CO;2
Orth, R. J., Heck, K. L., Jr., & Montfrans, J. van. (1984). Faunal Communities in Seagrass Beds:
A Review of the Influence of Plant Structure and Prey Characteristics on Predator: Prey
Relationships. Estuaries, 7(4), 339–350. http://doi.org/10.2307/1351618
Pearce, T. A., Meyer, J. H., & Boomer, R. S. (1982). Distribution and Food Habits of Juvenile
Salmon in the Nisqually Estuary, Washington, 1979-1980 (Technical Report). Olympia,
WA: U.S. Fish and Wildlife Service Fisheries Assistance Office.
Penttila, D. (2007). Marine Forage Fishes in Puget Sound (Technical Report No. 2007-03).
Washington Department of Fish and Wildlife, Puget Sound Nearshore Partnership.
Retrieved from http://www.pugetsoundnearshore.org/technical_papers/marine_fish.pdf
Plummer, M. L., Harvey, C. J., Anderson, L. E., Guerry, A. D., & Ruckelshaus, M. H. (2013).
The Role of Eelgrass in Marine Community Interactions and Ecosystem Services: Results
from Ecosystem-Scale Food Web Models. Ecosystems, 16(2), 237–251.
http://doi.org/10.1007/s10021-012-9609-0
Ralph, P. J., & Short, F. T. (2002). Impact of the wasting disease pathogen, Labyrinthula
zosterae, on the photobiology of eelgrass Zostera marina. Marine Ecology Progress
Series, 226, 265–271. http://doi.org/10.3354/meps226265
50
R Core Team. (2015). R: A language and environment for statistical computing. Vienna, Austria:
R Foundation for Statistical Computing. Retrieved from http://www.R-project.org/
Rehr, A. P., Williams, G. D., Tolimieri, N., & Levin, P. S. (2014). Impacts of Terrestrial and
Shoreline Stressors on Eelgrass in Puget Sound: An Expert Elicitation. Coastal
Management, 42(3), 246–262. http://doi.org/10.1080/08920753.2014.904195
Seddon, S. (2004). Going with the flow: Facilitating seagrass rehabilitation. Ecological
Management & Restoration, 5(3), 167–176. http://doi.org/10.1111/j.14428903.2004.00205.x
Shaffer, A. (2004). Salmon in the Nearshore: What do we know and where do we go? (Synthesis
of Conference Discussion) (p. 10). Port Townsend, WA: Pacific Estuarine Research
Society.
Shipman, H., Dethier, M. N., Gelfenbaum, G., Fresh, K. L., & Dinicola, R. S. (2009). Puget
Sound Shorelines and the Impacts of Armoring (No. U.S. Geological Survey Scientific
Investigations Report 2010-5254) (pp. 35–42). U.S. Geological Survey.
Short, F. T., Muehlstein, L. K., & Porter, D. (1987). Eelgrass Wasting Disease: Cause and
Recurrence of a Marine Epidemic. Biological Bulletin, 173(3), 557–562.
Short, F. T., Polidoro, B., Livingstone, S. R., Carpenter, K. E., Bandeira, S., Bujang, J. P., …
Zieman, J. C. (2011). Extinction risk assessment of the world’s seagrass species.
Biological Conservation, 144(7), 1961–1971.
http://doi.org/10.1016/j.biocon.2011.04.010
Short, F. T., & Wyllie-Echeverria, S. (1996). Natural and human-induced disturbance of
seagrasses. Environmental Conservation, 23(01), 17–27.
Shreffler, D. K., Simenstad, C. A., & Thom, R. M. (1990). Temporary Residence by Juvenile
Salmon in a Restored Estuarine Wetland. Canadian Journal of Fisheries and Aquatic
Sciences, 47(11), 2079–2084. http://doi.org/10.1139/f90-232
Shreffler, D. K., Simenstad, C. A., & Thom, R. M. (1992). Foraging by Juvenile Salmon in a
Restored Estuarine Wetland. Estuaries, 15(2), 204–213. http://doi.org/10.2307/1352693
Stevens, A., & Lacy, J. (2012). The Influence of Wave Energy and Sediment Transport on
Seagrass Distribution. Estuaries & Coasts, 35(1), 92–108. http://doi.org/10.1007/s12237011-9435-1
Takekawa, J. Y., Smith, A., & Woo, I. (2013). Assessing effects of restoration on the Nisqually
River Delta: enhancing invertebrate prey to increase capacity for salmon (Statement of
Work: 1 July 2014 - 30 June 2016 No. 7-1683). Olympia, WA: U.S. Geological Survey.
Retrieved from
51
https://salishsearestoration.org/images/e/ed/Takekawa_%26_Woo_2013_niqually_delta_
monitoring_proposal.pdf
Thom, R. M., Simenstad, C. A., Cordell, J. R., & Salo, E. O. (1989). Fish and their epibenthic
prey in a marina and adjacent mudflats and eelgrass meadow in a small estuarine bay
(Technical Report No. FRI-UW-8901) (p. 31). Seattle, Washington: FIsheries Research
Institute University of Washington. Retrieved from
https://digital.lib.washington.edu/researchworks/handle/1773/4104
U.S. Fish and Wildlife Service. (2005). Nisqually National Wildlife Refuge Final Comprehensive
Conservation Plan (p. 304). Nisqually National Wildlife Refuge Complex: U.S. Fish and
Wildlife Service.
Waycott, M., Duarte, C. M., Carruthers, T. J. B., Orth, R. J., Dennison, W. C., Olyarnik, S., …
Williams, S. L. (2009). Accelerating loss of seagrasses across the globe threatens coastal
ecosystems. Proceedings of the National Academy of Sciences, 106(30), 12377–12381.
http://doi.org/10.1073/pnas.0905620106
Woo, I., Turnker, K., Smtih, A., Markos, P., & Takekawa, J. Y. (2011). Assessing habitat
development in response to large scale restoration at the Nisqually River Delta.
(Unpublished Data Summary Report) (p. 21). Vallejo, CA: USGS Western Ecological
Research Center, San Francisco Bay Estuary Field Station.
52
Appendices
Appendix A. Epifaunal invertebrate species observed, broad taxonomic group and
classification by site, March – July 2015.
Phylum
Species
Annelida Eteone
Neanthes
Taxonomic Group Classification Level CMP MCA HGB RSS
Polychaeta
Polychaeta
Genus
Genus
X
X
Oligochaeta
Piscolidae
Annelida
Annelida
SubClass
Family
X
X
X
X
X
Polychaeta
Polychaeta
SubClass
X
X
X
X
Arthropoda Americorophium
X
Amphipoda
Genus
Ampeliscidae
Amphipoda
Family
X
Amphipoda
Ampithoe lacertosa
Amphipoda
Ampithoidae
Order
Species
X
X
Ampithoe valida
Amphipoda
Species
X
Ampithoidae
Amphipoda
Family
X
Aoridae
Aoroides columbiae
Caprellidae
Amphipoda
Amphipoda
Caprellidae
Family
Species
Family
X
Chironomidae
Diptera
Family
X
Cirripedia
Arthropoda
InfraClass
Copepoda
Arthropoda
SubClass
Corophiidae
Amphipoda
Cumacea
Eobrolgus
Eogammarus
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
Family
X
X
X
Cumacea
Amphipoda
Order
Genus
X
X
X
X
Amphipoda
X
Genus
X
X
X
Grandidierella japonica Amphipoda
Species
X
X
X
X
Harpacticoida
Copepoda
Order
X
X
X
X
Idotea
Isaeidae
Isopoda
Amphipoda
Genus
Family
X
X
X
X
X
Isopoda
Leptochelia
Isopoda
Arthropoda
Order
Genus
X
Lysianassoidea
Amphipoda
Superfamily
Monocorophium
Amphipoda
Genus
Ophelina
Amphipoda
Genus
Ostracoda
Ostracoda
Order
Pagarus
Arthropoda
Genus
Tanaidaceae
Tanaidaceae
Order
Chordata Ascidiacea
Tunicata
Class
Mollusca Bivalvia
Mollusca
Class
Gastropoda
Mollusca
Gastropoda
Mytilus
Nudibranchia
Nematoda Nematoda
Platyhelminthes Platyhelminthes
Trematoda
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
Class
Genus
X
X
X
X
X
Nudibranchia
Nematoda
Order
Phylum
X
X
X
X
X
X
Platyhelminthes
Phylum
X
X
X
Platyhelminthes
Class
X
53
Appendix B. Eelgrass field collection techniques: a-b) Collecting eelgrass in the field. c)
Freshwater station & sieve. d). Measuring eelgrass blade length and node number. d)
Processed eelgrass prepared for drying.
A
B
C
E
D
54
Appendix C. Eelgrass beds at a) Hogum Bay and b) Cormorant Passage.
A
B
55
