The Evergreen State College Institutional Repository

Nutrient Cycling Between Cultivated Manila Clams (V. Philippinarium) and Green Macroalgae (Ulva Spp.) on the Northern Hood Canal

Item

Title (dcterms:title)

Eng Nutrient Cycling Between Cultivated Manila Clams (V. Philippinarium) and Green Macroalgae (Ulva Spp.) on the Northern Hood Canal

Date (dcterms:date)

2016

Creator (dcterms:creator)

Eng Sahli, Natalie C

Subject (dcterms:subject)

Eng Environmental Studies

extracted text (extracttext:extracted_text)

NUTRIENT CYCLING BETWEEN CULTIVATED MANILA CLAMS (V.
PHILIPPINARIUM) AND GREEN MACROALGAE (ULVA SPP.) ON THE
NORTHERN HOOD CANAL

by
Natalie C. Sahli

A Thesis
Submitted in partial fulfillment
of the requirements for the degree
Master of Environmental Studies
The Evergreen State College
June, 2016

©2016 by Natalie C. Sahli. All rights reserved.

This Thesis for the Master of Environmental Studies Degree
by
Natalie C. Sahli

has been approved for
The Evergreen State College
by

________________________
Erin Martin, Ph. D.
Member of the Faculty

________________________
Date

ABSTRACT
Nutrient Cycling between Cultivated Manila clams (V. philippinarium) and Green
Macroalgae (Ulva spp.) on the Northern Hood Canal
Natalie C. Sahli
Large blooms of green marcoalgae (Ulva spp.) occur seasonally throughout the
Puget Sound basin; densely covering cultivated manila clams (Venerupis philippinarum)
and possibly acting as a source of POM to the organisms. Shellfish growers observe a
biomass increase of Ulva spp. in the presence of V. philippinarum monocultures. This
observation suggests a fertilization effect between species. Though V. philippinarium
aquaculture provides millions of dollars of annual revenue to the State of Washington,
nutrient cycling dynamics between Ulva spp. and V. philippinarum remain unquantified
for the Northern Hood Canal. To further examine this relationship, fifteen standard
commercial V. philippinarum mesh bags served as part of an experiment to assess the
potential for symbiotic nutrient cycling between the two species on Thorndyke Bay, in
Hood Canal, WA from June to September 2015. The randomized bags consisted of
treatments containing exclusively V. philippinarum, exclusively Ulva spp., and the
combination of Ulva spp. and V. philippinarum. C:N ratios of Ulva directly exposed to
clams were significantly lower (8.6 ±0.41) relative to the C:N ratios of Ulva spp. (12.5
±0.67) raised without clams, suggesting that ammonium secreted as a metabolic
byproduct by the clams provides essential nitrogen to macroalgae tissues. Additionally,
unanticipated drought conditions during the experiment yielded δ 13C evidence that V.
philippinarium (-20.82±0.26‰) feed on sources more isotopically depleted than Ulva
spp. (12.87±0.67‰) and phyto-POM (15.09± 0.63‰). Evidence for a commensal
relationship between these species justifies to shellfish growers that a well-managed
integrated system provides means to remove excess nutrients from aquaculture systems
via the extraction of nitrogen-rich macroalgae.

Table of Contents
INTRODUCTION……………………………………………………………………….1
CHAPTER 1: LITERATURE REVIEW……………………………………………....3
Introduction……………………..…………………………………………………………3
V. philippinarum Aquaculture in Washington State……………………………..………..6
Economic Importance and Growing Region Characteristics……………………..………7
V. philippinarum monocultures: dynamics within regional ecosystems………..………..11
Primary Producers in the Puget Sound in Hood Canal…………………………………..14
Macroalgae Blooms: dynamics of a nutrient source and sink……….…………………..15
Seasonal ecology of phytoplankton and response to nutrient variability…..……………17
Inter-trophic Nutrient Cycling………………...…………………………………………18
Organic and Inorganic Nutrient Sources and Fates in Puget Sound Estuaries…………19

Stable isotope methodology: tracing source elements through trophic levels...…22
Nutrient exchange and quantification in integrated aquaculture systems……….………25
Examining Case Studies and Identifying Existing Questions……………………………27
The impact of shellfish monoculture on macroalgae proliferation……..……………….27
Potential V. philippinarium dietary shift during seasonal macroalgae blooms…………30
Potential symbiosis between Ulva spp. and V. philippinarum…..……………………….33
CHAPTER 2: MANUSCRIPT……………..…………………………………………..36
Introduction………………………………………………………………………………36
Materials and Methods…………………………………………………………………...40
Study Location and Background…………………………………………………………40
Duration of Study……...…………………………………………………………………42
Field Design……………………………………………………………………………...42


 

iv
 

Sample Collection and Processing………………………………………………………45
Statistical Analysis…………………….…………………………………………………50
Results……………………………………………………………………………………51
V. philippinarium growth…………………………………………………...……………51
Ulva spp. Abundance…………………………………………………………………….57
Carbon to Nitrogen Ratios……………………………………………………………….59
Dietary Isotope Analysis…………………………………………………………………61
Mixing Model………………………….…………………………………………………64
Discussion………………………………..………………………………………………65
V. philippinarium Growth…………………………..……………………………………65
Carbon to Nitrogen Ratios……………………………………………………………….66
Isotope Analysis………………………………………………………….………………69
CONCLUSION…...………………………….………………………........……………77
Evidence for a Mutualistic Interaction…………………………………….…………….77
Ecosystem Implications…………………………………………………………………..79
Confounding Factors…………………...………………………………………………..80
Recommendations for Future Research……………………...…………………………..83
Integrating Findings into industry Practices…………………………….………………85
LITERATURE CITED……………...…………………………………………………87
APPENDICES………………………………………………………………………….96
Appendix A: Carbon and Nitrogen raw data and atomic ratios…………………………94
Appendix B: Stable isotope raw data……………………………………………………96
Appendix C: Detailed Materials and Methods……………………………….………….97


 

v
 

List of Figures
Figure 1. Change in WA state shellfish production from 1986 to 2013……….…………8
Figure 2. Standard mesh growing bags……………………………………….…………10
Figure 3. Location of the study site, Thorndyke Bay, situated in the Northern Hood
Canal……………………………………………………………..………………………41
Figure 4. The standard layout of in-ground aquaculture bags along the intertidal zone &
cross-section of an in-ground aquaculture bag…………..………………………………43
Figure 5. The experimental row in the field & diagram of the three treatments randomly
distributed throughout the experimental row………………………………………….…44
Figure 6. Examples of clam bags receiving a low, medium, and high Ulvoid density
designation…………………………….…………………………………………………48
Figure 7. Length in centimeters of V. philippinarum control group over time……….…53
Figure 8. Length in centimeters of V. philippinarum treatment group over time…...…..53
Figure 9. The mean % growth values for height and length between V. philippinarium
treatment and control groups……………….……………………………………………57
Figure 10. Frequency distribution of high, medium, and low Ulvoid density designations
between control and treatment groups…………………………………………………..58
Figure 11. A graphical depiction of C/N ratios in Ulva spp. tissues for treatment and
control groups by date……………………………………………………………………60
Figure 12. Mean isotopic signature with ranges for Ulva spp. control and treatment
groups, V.philippinarium control and treatment stomach gland and phyto-POM………62
Figure 13. The differences in mean δ13C values for and control Ulva spp. groups, and
phytoplankton-based POM……………………………………………………..………..63
Figure 14. Linear regression using temperature as a predictive variable of C/N ratio.…67
Figure 15. Map of Thorndyke Bay, Dosewallips, and Hamma Hamma site location along
the Hood Canal……………..……………………………………………………………74
Figure 16. A mixing model comparing δ 13C and δ 15N ratio values of Ulva spp.,
phytoplankton-based POM, combined V. philippinarium signatures, and the combined
signatures of G. maritima and S. virginica…………………...………………………….76
Figure 17. V. philippinarium cultivation site on July 18th, 2015……………..…………81

 

vi
 

Figure 18. V. philippinarium cultivation site on August 31st, 2015……………….....….82


 

vii
 

List of Tables
Table 1. Average length measurements in centimeters for each of the five V.
philippinarum control and treatment replicates…………………………………………52
Table 2. Mean length growth rates for each of the treatment and control replicates
between June 27th and August 15th….…………..……………………………………….54
Table 3. Average height measurements in centimeters for each of the five V.
philippinarum control and treatment replicates..……………………….………………..55
Table 4. Mean height growth rates for each of the treatment and control replicates
between June 27th and August 15th…………………...………………………………….56
Table 5. The mean range of estimated source contributions in percentages to V.
philippinarium diet using δ 13C values for Ulva spp, Phyto-POM, and G. maritima and S.
virginica……………………………….…………………………………………………75


 

viii
 

Acknowledgements
This thesis would not have been possible without the support of faculty and staff
at the Evergreen State College. Dr. Kevin Francis was an incredibly adept and responsive
program director throughout my time in the program, whose excellent leadership and
communication allowed for students and faculty to feel comfortable and supported in
their professional learning environment. Jenna Nelson was a great help to me in terms of
methods development and laboratory training. Her high standards for quality lab work
ensured excellent outcomes for my field and lab data. Kalie Adney at the science support
center was incredibly helpful in acquiring materials and lab space in a timely manner.
Her willingness to go beyond her required duties to accommodate lab students allowed
me to complete my project on time and under budget. Dr. Erin Martin was the largest
positive influence on my thesis project. Erin took the time to read over multiple drafts of
my thesis, methods, and grant proposals. She advised me throughout the entire process on
writing, lab techniques, and field design. Erin even took the time to visit my site and to
help with data collection. I am truly grateful for Erin’s high standards and the
commitment she shows to her students.
I am also incredibly thankful to the external support I received. Dr. Joth Davis of
the Puget Sound restoration Fund was incredibly helpful in allowing me to set up an
experiment on his shellfish farm. Additionally, Joth provided me with indispensible
industry and local ecological knowledge. I am grateful for the opportunity to have
worked with Joth towards a common goal of IMTA in the Puget Sound region. Audrey
Lamb of Taylor Shellfish was additionally helpful in guiding my field research design
and providing me with materials and advice. I learned an immense amount from Audrey
pioneering Ulva spp.-shellfish interactions on the Thorndyke Bay site. Additionally, my
undergraduate research assistant, Lauren Johnson helped me greatly in terms of data
collection and sample processing. Lauren’s positive attitude and volunteered commitment
made a huge difference when undertaking data collection of this magnitude.
I would also like to thank my family and friends, most notably, Kieran Lavelle,
for his continued commitment as my partner. Kieran made my research efforts and thesis
process a priority, dedicating long hours in the field and in support at home. Without
Kieran’s love and support, this project would not have been the enjoyable experience that
it was. Additionally, thank you to all my new friends in the MES cohort, and all my
persisting friends, who are now dispersed far and wide. This project was enriched via the
support and comic relief you all provided. Thank you also to my family, who continue to
love me throughout all the stages of my life.
This research project was funded in full by the MESA Professional Development
Fund and the Evergreen Student Foundation Activity Grant.


 

ix
 

INTRODUCTION
Shellfish monocultures exist within complex functioning shoreline ecosystems.
Currently, Washington State shorelines host the most productive shellfish industry in the
nation, comprising 31% of the total national market (Washington Sea Grant [WSG],
2015). Washington State’s extensive shellfisheries warrant further scientific investigation
into potential ecosystem impacts, especially as projections predict substantial industry
growth (WSG, 2015; Booth, 2014).
One of the most relevant concerns with shellfish monocultures is their contribution
to local nutrient cycling. The Washington coast experiences seasonal upwelling events of
nutrient-rich water (Newton et al, 2007). Though nutrients are essential to the functioning
of coastal systems, excess nutrients act as a detriment to ecosystems by over stimulating
primary production (Newton & Voorhis, 2002). The eventual degradation of primary
producers, utilizing these nutrients, depletes local oxygen concentrations and can result in
eutrophic conditions. Though upwelling is the main contributor of nutrients to
Washington’s coastal ecosystems, locally-sourced nutrients can additionally effect water
quality conditions (Newton, et al 2007).
The contribution of shellfish monocultures to the marine inorganic nitrogen pool
has been clearly documented. Shellfish produce ammonium (NH4+) as a metabolic
byproduct. This inorganic form of nitrogen is readily taken up by primary producers,
such as Ulva spp., a seasonal native green macroalgae. The high ammonium
concentrations result in increased biomass of Ulva spp. near monoculture sites (Saurel et
al, 2014; Zertuche-Gonzalez et al, 2008). Though this relationship between shellfish and
macroalgae biomass has been demonstrated for certain growing regions, the contribution


 

1
 

of shellfish-produced nitrogen to primary producer tissue in all highly profitable growing
regions remains unquantified. If Ulvoid tissue acts as a temporary sink for monoculturederived nitrogen, the species can be managed to mitigate seasonal negative effects of
shellfish monocultures.
Ulvoids may also transcend their negative designation as an aquaculture pest
species to act as a food source to monocultures. Bivalves have the capacity to incorporate
detrital material into their diets when available (Suh & Shin, 2014; Dunton & Schell,
1999). In this sense, shellfish may act to remove labile detritus from the water column,
interfering with the harmful decomposition pathway. Evidence for a symbiotic
relationship between the two species would support a transition within the industry away
from monocultures, towards an integrated macroalgae-shellfish aquaculture system.
Investigating the relationship between seasonal primary producer blooms and
shellfish monocultures will lead to a more clear understanding of how the industry can
manage aquaculture plots in the face of growth. This thesis attempts to quantify the
relationship between seasonal Ulvoid blooms and a commercially significant
monoculture, as they exist on a productive cultivation site in one of Washington’s
estuarial bays. The goal of this research is to quantify the contribution of Ulva spp. to the
diet of a commercially relevant species and to address the contribution of inorganic
nitrogen from monocultured species to seasonally occurring macroalgae tissue. These
findings will contribute to a larger investigation of how to manage Ulvoid blooms in
growing regions. Additionally, this research may provide evidence into the sensibility of
incorporating primary producers into cultivation sites, as a means of sequestering and
removing the nitrogen produced by shellfish during the summer months.


 

2
 

To illustrate the relationship between monocultures and primary producers, the
following document will elaborate on the economic and ecological importance of the
species, and locale, of interest, ultimately leading to a novel research question.
Additionally, a review of relevant research undertaken to this point will be included.
Following this will be an elaboration on field and laboratory methods used to quantify the
relationship between selected species. The quantitative results of this research and
subsequent discussion will follow. Finally, a conclusion, which addresses the economic
and ecological implications of these findings, will end this document.
CHAPTER 1: LITERATURE REVIEW
Introduction
Washington State shellfish are globally renowned for their fine quality. As early as
the 1860’s, the region’s abundant native shellfish resources entered into external markets
(Pacific Shellfish Institute [PSI]). The high demand for these delicacies brought about a
rapid decline in wild native populations, most specifically the native Olympia Oyster
(Ostrea lurida) (PSI). The solution came with the introduction of the larger Japanese
Pacific Oyster (Crassostrea gigas) to the region to supplement declining wild stocks
(Humphreys et al, 2015). Inevitably, additional non-native species were accidentally
introduced through shipments of these Pacific Oysters, including the now commercially
significant Manila clam (Venerupis philippinarum)(Quayle, 1949 in Humphreys et al,
2015). Both Crassostrea gigas and V. philippinarum adapted seamlessly to Washington’s
tidal flats (Humphreys et al, 2015). Today, both populations exist in wild and commercial
settings, bringing millions of dollars worth of revenue to the state annually (IEc, 2014).
To maintain a high level of production for global markets, shellfisheries rely on


 

3
 

intensive cultivation of Washington’s intertidal flats (PSI). Shellfish are grown often in
bags or racks, segregated by species (Toba, Dewey, & King, 2005). These expansive
bivalve monocultures require little or no external feed inputs, relying almost exclusively
on pre-existing ambient nutrients from the local environment. Shellfish growers and
citizens alike herald the nutrient extractive properties of filter-feeding monocultures as an
environmental solution to nitrified waters (Shumway et al, 2003). However, an in-depth
analysis of the complex nutrient cycling interactions between bivalves and their larger
ecosystem reveals a much more complicated story.
Locally concentrated nutrients negatively impact the water quality in certain
regions of the Puget Sound (Newton et al, 2007). Nutrients concentrate in certain areas,
due to reduced exchange with outside water bodies (USGS). The eutrophic conditions
resulting from excess nutrients create harmful conditions for local wild, and cultivated
marine, life (Newton, et al 2007). Shellfish accumulate organic and inorganic forms of
nutrients into their tissues over their life span, ultimately allowing for the removal of
these harmful elements from the system upon harvest (Shumway et al, 2003). However,
shellfish also release dissolved inorganic nutrients in the form of ammonium as a
metabolic byproduct (Peterson & Heck, 1999). This ammonium fuels the growth of green
macroalgae. Although macroalgae temporally sequester nutrients, these nutrients are
reintroduced back into the local system upon degradation (Saurel et al, 2014). Evidence
exists that macroalgae blooms negatively impact growth of certain shellfish species
(though the mechanism by which they do this is unclear ).(Lamb, 2015). Before
attempting to quantify the potentially complex inter-trophic dynamic between shellfish
and macroalgae blooms, understanding the relevant local, and comparable global,


 

4
 

nutrient cycling scenarios is imperative.
One way by which nutrient cycling can be assessed is through using carbon and
nitrogen isotopes, in combined with elemental ratios. Stable isotopes of carbon are often
used to delineate the dietary composition of a given individual or population. There are
two stable isotope forms of carbon: 13C and 12C. Each primary producer has its own
unique 13C signature, which carries over into consumers. Hence, in knowing the
signatures of all potential food sources and the signature of the primary consumer, the
relative contribution of each food source to the total diet can be deduced via a system of
equations. Elemental ratios are similar in that primary producer tissues mimic the exact
proportions of nitrogen and carbon in the environment (Zertuche-Gonzales et al, 2008;
Peterson & Heck, 1999). Hence, lower carbon to nitrogen (C/N) ratios indicate higher
ambient nitrogen concentrations. Each of these techniques has been used, respectively, in
the study of bivalve diets, and in studying the effect of shellfish on primary producer
nutrient assimilation (Suh & Hin, 2014; Zertuche-Gonzales et al, 2008; Dunton & Schell,
1999; Peterson & Heck 1999). Up to this point, a dietary stable isotope analysis for V.
philippinarum and an elemental ratio analysis of Ulva spp. growing near shellfish
cultivation plots in the Puget Sound are absent from the current literature.
The following literature review will serve to illuminate the essential pieces of this
complex ecological web, using Venerupis philippinarum in the productive Northern
Hood Canal as the commercially relevant centerpiece. To start, an exploration of this
commercially significant bivalve species’ dependency on regional baseline conditions,
including primary productivity, will be examined. Following, will be an examination of
seasonally and spatially relevant primary producers and their specific nutrient exchanges


 

5
 

with aquacultured bivalves. Furthermore, a detailed discussion of regional organic and
inorganic nutrient cycling will give specific context to trophic interactions in
Washington’s estuarial bays. Nutrient cycling at the ecosystem level will provide a
transition into the discussion of specific inter-trophic nutrient cycling. Concluding will be
a synthesis of all represented scales, presenting an integrated multi trophic systems
solution to the environmental stressors emergent from V. philippinarum monocultures.
This review will demonstrate the need for monoculture diversification in the Washington
State shellfish industry, and propose a method of study providing quantitative evidence in
favor of this paradigm shift to shellfish growers and marine ecologists, alike.
V. philippinarum Aquaculture in Washington State: ecology of a profitable
monoculture
The following section demonstrates the value of V. philippinarum aquaculture to
the State of Washington and its relevance for intensive study. Research by the industry,
into physical applications to monoculture growing sites, has the potential to increase
production in the state. Already, certain growing regions yield greater abundances of wild
and commercial V. philippinarum; notably the Hood Canal and, to a lesser extent, South
Puget Sound. These regions are indispensable to V. philippinarum aquaculture in
Washington. Thus, the commercial and ecological components of these highly productive
locales provide context for further research into potential industry innovations. In these
regions, the monocultured clams are dependent on larger ecosystem forces and thus
cannot be separated from the ecology as a whole. This section will illuminate the
connection between the aquaculture industry and the biogeochemical components at
growing sites. It will demonstrate a needed shift in the paradigm from monoculture-


 

6
 

intensive research to intensive investigations surrounding the integration of natural
ecosystem dynamics into the structure of aquaculture itself.

Economic Importance and Growing Region Characteristics
Washington State boasts the largest grossing shellfish aquaculture industry in the
nation (Industrial Economics Incorporated [IEc], 2014), providing 96.9 million dollars of
revenue to the state annually (Booth, 2014). V. philippinarum aquaculture comprises
eleven to sixteen percent of total commercial cultivation, second only to the Pacific
Oyster (IEc, 2014). Currently, the industry is exploring new markets for V.
philippinarum, intending to increase production of the clams without drastically
expanding growing sites (IEc, 2014). Innovations in cultivation techniques, which reduce
predation by dispensing durable netting, and increase juvenile survivorship, are fueling
the growth of the industry (Thompson, 1995). Altering substrates in order to increase the
recruitment of spat has been a focus in industry and conservation research. Providing
gravely substrates, often enhanced with crushed bivalve shells increases recruitment for
hardshell clams (The Nature Conservancy [TNC]). In hatchery settings, individuals are
selectively bred to create more robust brood stocks (Johnson, 2008). Historically,
industry research in Washington focuses primarily on increasing the production of V.
philippinarum by selectively breeding populations and/or altering the surrounding
environment (TNC; Johnson, 2008; Thompson, 1995). Diversification from the current
monoculture paradigm, to more closely mimic natural ecosystem cycling may further
improve yields on commercial sites. The following figure exemplifies the increase of
manila clam aquaculture in the Puget Sound and Washington Coast.


 

7
 

Figure 1. Change in WA state shellfish production from 1986 to 2013 (Washington Sea
Grant [WSG] 2015).
Industry projections indicate an increase in V. philippinarum production for the all
growing regions throughout Washington (Booth, 2014). There are five major growing
regions for shellfish in Washington State (Willapa Bay, Grays Harbor, North Puget
Sound, South Puget Sound and the Hood Canal) all which support the cultivation of V.
philippinarum. The regional variability amongst the various growing sites significantly
influences the predominance of V. philippinarum as compared to other molluscan
species. (Booth, 2014). As of 2014, the Hood Canal supports the majority of V.
philippinarum aquaculture in the state (Booth, 2014). In 2013, manila clam aquaculture
accounted for 47% of total aquaculture production in the Hood Canal (WSG, 2015).
The Hood Canal is a primary sub-basin of the greater Puget Sound (Warner &
Kawase, 2001). This deep, estuarial fjord is unique in that its mouth is located between
two high glacial sills, separating it from the main influx of water from the north. These
sills, termed the Admiralty sills, allow water to flow over into the Hood Canal via the
Strait of Juan de Fuca. Though the sills do not prevent the influx of water, they greatly


 

8
 

inhibit the outflow of water from the Canal, creating highly stratified conditions in the
water column (Warner & Kawase, 2001). These stratified conditions become problematic
in the late Summer and Fall, when dissolved oxygen concentrations are low as a result of
seasonal upwelling and primary producer degradation (Newton et al, 2007). Riverine
inputs into the Hood Canal, additionally remain stratified on the upper surface, further
contributing to primary production and eutrophic conditions (Warner & Kawase, 2001).
The unique physical characteristics of the Hood Canal make it well-suited to
support V. philippinarum cultivation. V. philippinarum readily settle in gravel, sand, and
mud substrates (Department of Fisheries and Oceans [DFO], 1999). However, the species
display higher survivorship in finer-sediment substrates (Thompson, 1995). The upper
layers of Hood Canal are comprised predominantly of glacial till and glacial outwash
sediments (Washington State Department of Transportation [WSDOT], 2008), providing
an ideal habitat type for the clams. Several small streams flow into the Canal, providing a
mixture of fine and course sediment to beaches (WSDOT, 2008). The Hood Canal and
Puget Sound have experienced significant sediment delivery (Puget Sound Partnership
[PSP], 2006). However, the Hood Canal is less populated and has experienced less land
change use, which impacts sedimentation in the Puget Sound, reinforcing it as an ideal
growing site. Tides in the Hood Canal are less severe than in other growing regions,
reducing exposure of intertidal clams to desiccation (Toba, Dewey, & King, 2005). Less
drastic tides also reduce the intensity of wave action and beach erosion (WSDOT, 2008).
Strong wave action inhibits clam survivorship by washing away particulate substrate.
Thus calmer bays, such as those found in the Canal, are preferred for V. philippinarum
cultivation (Toba, Dewey, & King, 2005).


 

9
 

Inherent physical factors create obstacles to Hood Canal aquaculture. The late
summer and early fall is defined by poor water quality in the Canal (Newton et al, 2007).
Low dissolved oxygen levels, resulting primarily from the seasonal upwelling of oxygen
depleted waters along the Eastern Pacific, present an issue to Hood Canal marine life
(Newton et al, 2007). Though shellfish posses a higher tolerance to hypoxic conditions
than marine vertebrates, severe oxygen reduction can cause significant stress within
organisms (Diaz & Rosenburg, 1995). Harmful algal blooms (HABs) of large green
macroalgae similarly stress marine invertebrates. Additionally, certain species of
dinoflagellate phytoplankton have been shown to create a starvation response in
invertebrates during the late summer and early fall months throughout thr Puget Sound
Basin (Jerry Borchert, personal communication, 2015).
The shellfish industry has developed some measures of defense against biotic
stressors. Large nets help contain wild juvenile clams and decrease predation risk
throughout the two to five year growing period (Booth, 2014). On large scale aquaculture
operations, hard plastic 1/2-inch mesh bags protect clams from predators and reduce
migration throughout their life cycle (Toba, Dewey, & King, 2005). Aquaculture bags are
almost completely submerged in the sediment, allowing clams to burrow to their
preferred depth of two to three inches below the surface (Washington Department of Fish
an Wildlife [WDFW]).


 

10
 


 

Figure 2. Standard mesh growing bags (Toba, Dewey, & King, 2005).
Often, harvesters remove macro algae from the outsides of the bags during the summer
months. Macroalgae reduces the availability of dissolved oxygen within the bags and
prevents the circulation of seawater (Toba, Dewey, & King, 2005). Circulating seawater
carries phytoplankton-based particulate organic matter, the bivalve’s primary food
source.

V. philippinarum monocultures: dynamics within regional ecosystems
Cultured and wild V. philippinarum derive nutrients from several different sources.
V. philippinarum primarily consume marine particulate organic matter (POM) (Poulain et
al, 2010). The composition of POM in marine waters is composed of several components
including “ phytoplankton, microphytobenthos, resuspended sediment, terrestrial carbon,
[and] marine macro algae detritus” (Poulain et al, 2010). Depending on the availability of
POM, filter feeders will consume less preferable organic matter, such as that from
sewage runoff (Rensel, Bright & Siegrist, 2011). This is testament to their highly variable
and adaptive diets. Cultured bivalves depend most heavily on phytoplankton from the
ambient environment, evident from decreased growth rates when phytoplankton are
scarce in comparison to alternate nutrient sources (Spillman et al, 2008). In the midst of


 

11
 

seasonally-induced large microalgal blooms in the Puget Sound, bivalve species
experience increased growth rates and will selectively feed on phytoplankton over other
organic nutrient sources (Rensel, Bright & Siegrist, 2011).
Dissolved inorganic phosphorous increases microalgal productivity (Newton,
2011), which corresponds to higher growth rates in V. philippinarum. Phosphorous
represents a critical nutrient input affecting the availability of POM. However, in the
Hood Canal, nitrogen is considered to be a more significant limiting nutrient than
phosphorous (Newton, 2011). Hence, a detailed discussion of phosphorous cycling will
be largely omitted from this document in favor of a greater emphasis on nitrogen cycling.
Marine macrophytic detritus can contribute significantly to bivalve diets when
abundant (Dunton and Schell 1999). However, detrital material is found less preferable to
phytoplankton (Dang et al, 2009 & Rensel, Bright & Siegrist, 2011). In Arachon Bay,
France, detritus was consumed only when phytoplankton was less abundant (Dang et al,
2009). Bivalves sort particles by size prior to digestion excreting larger masses of
material as pseudofeces (Tucker & Hargreaves, 2009). Hence, the consumption of
phytoplankton may be more energetically favorable than the consumption of detritus.
In estuarial food webs, primary consumers rely on local sources of primary
producers (Dang et al, 2009). Local sources of organic and inorganic nutrients, and thus
the diets of the bivalves utilizing those nutrients, vary regionally. (Rensel, Bright &
Siegrist, 2011). Location on the tidal flat (Rensel, Bright & Siegrist, 2011). proximity to
freshwater inputs (Kasai, Horie, & Sakamoto, 2004), relative abundance of macroalgae
and macrophytes, as well as seasonal productivity of phytoplankton control the type and
abundance of nutrients. Assessing these environmental parameters could result in a more


 

12
 

intensive understanding of bivalve-environment interaction to determine the relative local
contributors to V. philippinarum diets. This understanding could illuminate ways in
which aquaculture systems could be manipulated to potentially maximize the availability
of preferred food sources. If the preferred food sources of cultivated shellfish have an
associated economic value, shellfish growers may be inclined to incorporate a more
diverse set of species into cultivation plots.
Shellfish aquaculture exists within the context of a larger ecosystem. The
ecosystem services necessary for shellfish survival represent only one aspect of a larger
interaction. The byproducts of these monocultures similarly feed back into the
surrounding environment. Environmental impacts from shellfish monocultures can be
both positive and negative. Shellfish aquaculture increases biodeposition of undigested or
excreted POM into the sediment, fueling denitrifying bacteria, theoretically removing
nitrogen from the system (Shumway et al, 2003). This sedimentary organic matter (SOM)
fuels benthic primary consumers at the base of the food chain (Shumway et al, 2003).
Additionally, shellfish filter both particulate organic and inorganic forms of nitrogen,
resulting in N assimilation in their tissue, helping mitigate potentially harmful eutrophic
conditions (Shumway et al, 2003). The structure of aquaculture additionally provides
refuge for native species, and supports wild shellfish recruitment (Saurel et al, 2014).
Unfortunately, extensive shellfish monocultures have the potential to negatively
impact the environment. Dense monocultures adversely affect benthic biodiversity
(Sequeira et al, 2008). In addition, cultivated V. philippinarum compete with native clams
and oysters for desirable POM, reducing growth rates in wild species (Sequeira et al,
2008) and changing phytoplankton species composition (Saurel et al, 2014). Evidence in


 

13
 

the literature, which conflicts with industry publications, suggests shellfish farms
contribute further to eutrophication through biodeposition (De Casbaianca, Laugier, &
Marinho-Soriano, 1997). Ammonium excretion by V. philippinarum increases with
summer temperatures (Mann & Glomb, 1978), fueling the growth of primary producers
(Saurel et al, 2014). Shellfish monocultures can act as significant nutrient sources to local
primary producers. Zertuche-Gonzalez, 2008, concluded that ammonium concentrations
were high enough to significantly increase the proliferation of the green seaweed, Ulva
spp., sharing the same bay as oyster monocultures in Baja California. Local evidence
from the Puget Sound suggests that manila clam monocultures similarly increase the
abundance of seaweed species in a harvest area (Saurel et al, 2014). Dense Ulva spp.
blooms outcompete and shade other macrophyte communities, reducing marine plant
diversity. Overall, alternative aquaculture systems warrant investigation as a means to
reduce any potential negative impacts of monocultures on local ecosystems.
Primary Producers in the Puget Sound in Hood Canal: spatial and seasonal
dynamics of Ulva spp. and Phytoplankton
The previous section demonstrates the importance of regional ecology to productive
V. philippinarum aquaculture in the state of Washington. This section will further expand
on the dynamics of primary producers, specifically phytoplankton and Ulva spp. While
phytoplankon is the established preferred primary food source of cultivated bivalves,
detrital macroalgae can contribute to bivalve diets when present. Distribution of both
primary producers is variable amongst growing sites in the Puget Sound. Similar to V.
philippinarum, primary producer populations respond to the biogeochemical forcing in a
given region. The abundance and composition of primary producers in an area can have
significant impacts on growing shellfish. Additionally, shellfish can influence the


 

14
 

abundance and physiology of certain primary producers. Understanding the ecology of
these primary producers will illuminate the key components of seasonal and spatial algal
nutrient inputs into productive aquaculture settings. Additionally, this information will
contribute further insight into the feasibility and possible benefits of incorporating native
macroalgaes into integrated mariculture systems.

Macroalgae Blooms: dynamics of a nutrient source and sink
Ulva spp. is a genus of green sea lettuce native to the Puget Sound and Hood Canal.
During the growing months of June-September, Ulva spp. is the dominant macroalgal
species in the intertidal zone (Western Washington University [WWU]). Ulva spp.
attaches to substrates, yet the majority exists as floating in large masses in shallow, sandy
protected bays (WWU). Ulva spp. generally prefers areas of softer substrates over rocky
intertidal zones (Nelson, 2008). Reduced nitrogen input into marine ecosystems in the
late summer months, due to decreased river export, limits Ulva spp. growth (Nelson,
2008). Decay typically happens during the late summer/early fall (Nelson, 2008). Low
salinity environments and low light levels have a negative effect on the macro algae
blooms (Nelson, 2008). This combination of factors make sandy, well-exposed V.
philippinarum growing sites excellent habitats for Ulva spp..
Often Ulva survives better in these near shore environments than other
macrophytes. Evidence in the last decade suggests that Ulva spp. outcompetes marine
macrophytes in their traditional habitat, as it more readily adapts to high intensity light
levels, partial desiccation and eutrophic conditions (Nelson, 2008). Economically feasible
macroalgae show declining population levels throughout the certain regions of the Sound


 

15
 

due to increasing sedimentation from anthropogenic sources (Mumford, 2007). Ulva spp.
thrive in highly sedimented zones, and they can serve as an experimental proxy by which
to assess detrital contributions, and nutrient uptake of macroalgae as a whole to altered
near shore ecosystems.
Seaweeds are generally perceived to be the least represented primary producer
contributors to global marine food webs (Nelson &Tjoelker, 2003). However, Ulva spp.
is a significant contributor of nutrients in marine near shore ecosystems. In the case of
eutrophic near shore coastal ecosystems, seaweeds represent a main component of
primary production (Zertuche-Gonzalez et al, 2008). Ulva spp. decays rapidly, with a half
life of 8 days (Zertuche-Gonzalez et al, 2008). Some nutrients are remineralized into the
environment; however, labile organic components are reintroduced into the food web
during decomposition (Zertuche-Gonzalez et al, 2008).
Other macroalgaes, such as kelp, contribute significantly to benthic food webs
(Dunton & Schell, 2003). Though detrital incorporation is variable among benthic
invertebrates, significant evidence exists supporting the importance of detrital macro
algae throughout near shore trophic levels (Dunton &Schell, 2003). Juvenile V.
philippinarium readily incorporate marine detritus, which can compose over 50% of their
diet (Suh & Shin, 2013). In comparable environments, adult clams incorporate 20-30%
floating or settled detritus (Suh and Shin, 2013). Detrital incorporation in benthic
communities varies based on locale (Suh and Shin, 2013). Incorporation rates of
macroalgae detritus remain unknown for adult V. philippinarium in the Hood Canal
growing region.
The relative abundance of Ulva spp. in the intertidal zone allows it to be a


 

16
 

significant sink for nutrients (Hanisak 1993 in Zertuche-Gonzalez et al, 2008). Nitrogen,
phosphorous and carbon comprise three ambient nutrients able to be stored in Ulva tissue
(Zertuche-Gonzalez et al, 2008). Ulva spp. abundance increases significantly near oyster
and manila clam beds due to higher ammonium concentrations (Saurel et al, 2014;
Zertuche-Gonzalez et al 2008). The tissue of Ulva spp. near oyster beds in Bahia San
Quintin lagoon in Mexico reach maximum values for Ulva nitrogen saturation (around
~2.3%) for most of the growing season (Zertuche-Gonzalez et al, 2008). In certain high
light intensity, low-nutrient environments, maximum nitrogen absorption reflects
maximum growth rate (Zertuche-Gonzalez et al, 2008).
If the Ulvoids are left to naturally decay, which is likely, unless physically
removed, the absorbed nutrients will return to the environment during decomposition in
the Fall. If the area experiences reduced water circulation over the winter, these nutrients
may possibly linger to exacerbate the following Spring’s bloom, with the onset of warmer
temperatures and increased light. This phenomenon as can be seen in the Bahia San
Quintin lagoon (Zertuche-Gonzalez et al, 2008). Removal of Ulva spp. tissue prior to
degradation can reduce eutrophic conditions (Zertuche-Gonzalez et al, 2008). A wellmanaged integrated aquaculture system could result in the removal of this short-term
nutrient sink from the ecosystem. Thus, the resulting ecosystem would be less susceptible
to eutrophic conditions resulting from shellfish monocultures and natural seaweed
degradation.

Seasonal ecology of phytoplankton and response to nutrient variability
Phytoplankton blooms occur in the Puget Sound and Hood Canal primarily between


 

17
 

the months of April to September (Nakata & Newton, 2001). Generally blooms will
occur at temperatures greater than thirteen degrees Celsius (Greengrove et al, 2014), and
thus are not dependent exclusively on season. Blooms are uncommon during the winter
months due to limitations in temperature and light (Greengrove et al, 2014). A variety of
physical environmental factors influence bloom severity “…including vertical advection
and turbulence, modulation of underwater light intensity by self-shading and inorganic
particles, sinking of algal cells, and occasional rapid horizontal advection of population
from [a given region] by sustained winds.” (Winter et al, 1975).
Microalgae blooms comprise the largest component of total marine primary
production. Marine phytoplankton blooms are fueled primarily by available nitrogen in
marine ecosystems (Winter, 1975). Phytoplankton assimilate dissolved organic nitrogen
(DON), ammonium, nitrate, nitrite, and, to a lesser degree, atmospheric nitrogen gas
(Voss et al, 2011). Ammonium is the most readily absorbed form of N by phytoplankton,
especially in low light intensity and nutrient-limited conditions (Dortch, 1990). Though
the rate of ammonium uptake in phytoplankton exceeds that of nitrate, conditions of high
nitrate saturation produces more productive blooms for a majority of species (Dortch,
1990). Recent increases in nutrient introduction by anthropogenic sources has resulted in
increasing localized bloom size throughout the Puget Sound region (Kangaonkar et al,
2012).
In the Puget Sound, seasonal nutrient upwelling during the summer months and
anthropogenic nutrient inputs control the proliferation of phytoplankton (Kangaonkar et
al, 2011). These large seasonal blooms act to effectively feed cultured bivalves
throughout the region (Rensel, Bright & Siegrist, 2011). However, the byproduct of


 

18
 

microalgae degradation can significantly contribute to eutrophic conditions (Voss et al,
2011). The decomposition of phytoplankton reduces dissolved oxygen concentrations
below the surface, especially in the summer months when water layers are highly
stratified by temperature (Newton et al, 2007, 2005; Voss et al, 2011). Thus, the
degradation of these large blooms negatively impacts water quality and habitat conditions
for native species.
Inter-trophic Nutrient Cycling: tracing fundamental elements through complex
ecosystems
The previous two sections demonstrate 1) the dependence of a commercially
important species on local physical and ecological dynamics 2) the seasonal and spatial
ecology of primary producers, the food sources of V. philippinarium, in relation to
aquaculture sites. This section will attempt to further synthesize the fundamental
relationship amongst environmental nutrient availability, primary consumers (i.e.,
shellfish), and primary producers in aquaculture settings. Elemental carbon and nitrogen
comprise the core transferable units upon which to quantify the story of dynamic intertrophic exchange. The physical forcings of the local environment come into play as a
baseline for available nutrients. Additionally, the differential characteristics of available
carbon and nitrogen species can be used to assess the relative movement of the elements
from their ecological baseline through trophic levels. Similarly, the relative abundance of
specific elemental components in primary producer tissue serves to illuminate the effect
of external mechanisms on the ready available forms of these elements. Synthesizing
components of environmental nutrient determination and elemental transfer between
trophic levels can act to quantify a symbiotic relationship within a given locale.
Symbiotic multi-trophic relationships in aquaculture settings represent solutions to effect


 

19
 

nutrient determinations in eutrophically stressed coastal environments.

Organic and Inorganic Nutrient Sources and Fates in Puget Sound Estuaries
Organic nutrients enter the Puget Sound through primarily allochthonous pathways.
The majority of Dissolved Organic Nitrogen (DON) enters the Puget Sound by riverine
transport (Mackas & Harrison, 1997). Evidence suggests that DON and particulate
organic N, exported by wetland-dominated Hood Canal tributaries, can exceed inorganic
N export in other tributaries (Steinburg et al, 2010). Particulate and dissolved organic
nutrients contribute significantly to eutrophic conditions in estuaries by fueling primary
productivity and increasing local sedimentation (Mackas & Harrison, 1997). The local
degradation of marine primary producer tissue acts as an autochthonous source of
particulate organic nitrogen (PON) and dissolved organic nitrogen (DON) (Mackas &
Harrison, 1997).
Average inputs of particulate organic nitrogen (PON) and dissolved organic
nitrogen (DON) into Puget Sound estuaries were assessed in 1997 to be 1400-1500
tonnes/day (Mackas & Harrison, 1997); a value that likely has increased with land use
change and increasing population in the Puget Sound basin. Of this organic input, ~75%
terminates in primary producer tissue (Mackas & Harrison, 1997). Organic nitrogen
leaves estuarial systems through advective export of incorporated, particulate, and
dissolved organic nitrogen, marine-life harvest, predation, and to a lesser extent,
denitrification (Harrison et al, 1997). In poorly flushed basins, DON and PON compound
localized eutrophic conditions (Mackas & Harrison, 1997).
Nutrient cycling in marine near shore ecosystems represents a complex process


 

20
 

inclusive of terrestrial and open ocean influences (Khangaonkar et al, 2012; Voss et al,
2011). Terrestrial nutrient inputs dominated by “rivers, non-point source runoff and
nearly 100 wastewater discharges” threaten the health of the Puget Sound, primarily
poorly flushed basins (Khangaonkar et al, 2012). In basins such as the South Puget Sound
and Hood Canal, these stagnant inorganic and organic terrestrial nutrients exacerbate
eutrophic and harmful hypoxic conditions (Khangaonkar et al, 2012; Newton et al, 2007).
Seasonal nutrient upwelling events from the open oceans (which contribute the
majority of nutrients into the Puget Sound basin) occur from November to Februrary
(Khangaonkar et al, 2012). Upwelled nutrients from the Pacific Ocean enter the Puget
Sound basin by way of the Admiralty Inlet (Khangaonkar et al, 2012). In the Hood Canal,
inorganic N inputs from upwelling comprise about 98% of the total N delivered to the
surface (Steinnburg et al, 2010). Certain sub-basins receive significant N inputs via
watershed export (Steinnburg et al, 2010). However, the majority of the Hood Canal
cycles according to seasonal upwelling influence. Despite the dominance of upwelling as
a source of nutrient availability in the Canal, seasonal autochthonous and allochthonous
sources represent important factors in localized nutrient cycling.
Inorganic nutrient loading significantly affects local marine chemistry. Dissolved
inorganic nutrients (DIN) exceed organic forms of N in the Puget Sound (Steinnburg et
al, 2010; Mackas & Harrison, 1997). Streams and rivers provide additional contributions
of DIN to marine systems (Newton et al, 2011). Freshwater inorganic nutrient inputs
account for 421 ± 162 metric tons of inorganic nitrogen per year (Paulson et al, 2004 in
Newton et al, 2011). Bacterially mediated atmospheric fixation of N contributes
additional inorganic nutrients into the Hood Canal at 30 ± 11 metric tons per year

 

21
 

(Paulson et al, 2004 in Newton et al, 2011). In contrast, the primary mechanisms by
which nitrogen is naturally lost from the Washington coastal shelf are through
bacterially-mediated denitrification and burial in sediment (Christiansen, Smethie, &
Devol, 1987). Inorganic nutrient cycling represents an incredibly complex process
involving both abiotic and biotic components. Due to the limited scope of this project, the
discussion of inorganic nutrient cycling will be constrained primarily to the inorganic N
exchange between macrobiota.
The balance of nitrogen in a system is dependent on a variety of environmental
factors. These variable concentrations of nutrients represent the fundamental
environmental parameters that shape nutrient assimilation and movement in an
ecosystem. Understanding nutrient exchange on a local level can illuminate forcings
which may have a profound impact on species within a regional ecosystem.

Stable isotope methodology: tracing source elements through trophic levels
Stable isotopes behave in a predictable manner from individual to ecosystem.
Stable nitrogen and stable carbon isotopes are most commonly utilized to study intertrophic dynamics and nutrient cycling in complex systems (Ryabenko, 2013). Stable
nitrogen isotopes are found in two forms: 15N and 14N (Ryabenko, 2013), whereas stable
carbon isotopes take the form of 12C and 13C (Farquhar, Ehleringer, & Hubick,1989). The
forms differ in their neutron counts; the “heavier isotope” having the higher amount of
neutrons of the two forms (Ryabenko, 2013). The heavier form of each isotope is
significantly less abundant than the lighter form (Ryabenko, 2013). The exact
environmental abundance varies with local biogeochemical cycling and can experience


 

22
 

wide temporal variation. However, species will assimilate both forms present in the
environment in a specific and predictable manner. Isotope relative abundances allow for
the delineation and comparison of trophic levels. Measure of relative abundance is
expressed in comparison to a standard and are defined by the following equations for
carbon and nitrogen (Ryabenko, 2013)

1a.

! !"

! !"

 δ𝐶 !" vs. V − PDB = [(! !" )!"#$%& ÷ (! !" )!!!"# − 1]×1000
*where V-PDB represents the Vienna PeeDee Belemnite International Standard

b.

! !"

! !"

 δ𝑁 !" 𝑣𝑠. 𝑎𝑖𝑟 = [(!!" )!"#$%& ÷ (!!" )!"# − 1]×1000
Primary producers differ in their rates of assimilation of heavy and light N isotopes

(Ryabenko, 2013). Nitrogen isotope abundance in primary producers is a reflection of
both the relative occurrence of species-specific discriminatory molecular reactions and
environmental 15N/14N ratio (i.e. the 15N/14N of nutrients the primary producers
consume). The tissue of local primary producers will proportionally reflect environmental
abundance of 15N /14N (Cloern, 2002). The ratio of heavy to light isotope in tissues is
referred to as the “relative abundance” (Ryabenko, 2013). Relative abundance can differ
amongst individuals within the same trophic level (Cloern, 2002). However, the change
in relative abundances of N isotopes from one trophic level to the next is large enough to
delineate food webs in a given ecosystem (Ryabenko, 2013).
Fractionation of available inorganic N species within a given ecosystem depends
on a variety of environmental parameters. Some of the determining factors in available N
species isotope relative abundance baseline values include microbial species composition
(and rate of metabolization), dissolved oxygen concentration, the availability of N in a
system, the presence of N-assimilating biota, and the biogeochemical source of the Nspecies (Hoefs, 2009; Hein et al, 1995). Macroalgae more readily assimilate the heavier

 

23
 

form of nitrogen into their tissues, especially if their N source is ammonium (Altabet,
1988). High ambient ammonium concentrations correspond to a higher environmental
relative abundance of 15N (Cloern, 2002). Ammonium’s enrichment, relative to other
species of inorganic N, is attributed to significant fractionation during the nitrification
process (NH4+ to NO2) (Brandes & Devol, 1997). However, this enrichment does not
occur in anoxic environments, which may be observed in the Hood Canal during late
summer (Brandes & Devol, 1997). In N limited systems, fractionation is relatively zero
between forms of DIN, as N-consumers do not exhibit an isotope preference during
limitation (Hoefs, 2009). In anoxic systems, the δ15N value for nitrate is triple the value
of nitrate in oxic systems due to the dominance of denitrification (Hoefs, 2009). Field
observations aiming to average isotope fractionation between anoxic and oxic marine
environments yield an average value of 4.5‰ for NH4+, and a value -10‰ for NO3(Hoefs, 2009). These values exemplify the tendency of ammonium to have a higher
enrichment value compared to more oxidized nitrate species.
Micro and macroalgae differ in their capacity to assimilate ammonium. Ammonium
assimilation occurs at a much faster rate than nitrate assimilation for both macro and
microalgae (Hein et al, 1995). However, microalgae assimilate ammonium at a more
rapid rate than do macroalgae (Hein et al, 1995). Additionally, microalgae simultaneously
display a preference for less enriched ammonium species (Altabet,1988). Microscopic
primary producers have a much higher turnover rate and therefore disproportionately
utilize the lighter, more energetically favorable, isotope forms (Altabet,1988).
Incorporation of heavy isotopes is positively correlated with the residence time of an
organism in a given environment (referred to as incubation period) (Ryabenko, 2013).
Incubation period is inversely related to turnover rate. Hence, this discrimination against
isotopically enriched ammonium results in higher concentrations of ambient enriched


 

24
 

NH4+.When ammonium, in any form, is present in high concentrations, the enzyme which
facilitates nitrate uptake is severely limited in microalage, resulting in disproportionate
ammonium uptake (Hein et al, 1995). Hence, the fractionation values for ammonium in
nutrient dense, productive systems far exceeds that of nitrate.
Carbon isotope signatures are the primary metric used to quantify the dietary
contributions to consumers (whereas nitrogen isotopes, because of their high
fractionation rates through trophic positions, delineate trophic level). Moving up trophic
positions in an ecosystem will result in an isotopic enrichment proportional to trophic
level for C and N isotopes. Both macro and microalgae synthesize carbohydrate
structures from carbon dioxide through the photosynthetic process (Farquhar, Ehleringer,
& Hubick,1989). The photosynthetic process selectively discriminates between the heavy
and light isotope, preferring the light isotope (Farquhar, Ehleringer, & Hubick,1989).
Green macroaglae is more enriched relative to microalgae (Dunton & Schell, 2003). In
the Puget Sound Ulva spp. has a relative abundance of δ13C -13.0 to -6.7‰ and
phytoplankton is more depleted with δ13C values ranging from -20.0 to -18.1 ‰ (Howe,
Simenstad,& Ogsto, 2012). V. philippinarum incorporate both detrital macroalgae and
phytoplankton into their diet. The relative abundance of C isotope contributed from each
component will be proportionately reflected in the tissue of V. philippinarum, as
represented by the following equation (barring trophic fractionation values for V.
philippinarum) (Duedero et al, 2009):
2.

δ 13C phytoplankton x F phytoplankton + δ 13C ulva x F ulva = δ 13C clam

*where δ 13C represents the relative abundance of 13C; and F represents the fraction of
each dietary component
As elements move through a trophic system, each higher trophic level will
metabolically incorporate the heavier isotope at a species-specific rate (Zanden &
Rasmussen, 2001). This processes is referred to as fractionation and leaves the body

 

25
 

tissues relatively enriched with the heavier isotope (Suh & Shin, 2013). V. philippinarum
fractionate nitrogen by an average amount of 2.9‰ (Duedero et al, 2009) and carbon by
0.6 ‰ (Suh & Shin, 2013). Often a standard trophic fractionation rate of 3.4 ‰ and 0.8
‰ for nitrogen and carbon, respectively, can be applied to most trophic studies (Suh &
Shin, 2013). Not all tissues in a consumer will assimilate isotopes equally (Kidd et al,
1995). Muscle tissues retain the average signatures of diet over longer periods of time
(Kidd et al, 1995). In bivalves, the stomach gland tissue best reflects short term diet
(Raikow & Hamilton, 2001). Hence, the stomach gland of V. philippinarum will most
accurately reflect diet based on environmental composition of primary producers at a
given point in time after applying the 0.6 ‰ fractionation rate (Suh & Shin 2014).

Nutrient exchange and quantification in integrated aquaculture systems
Integrated multitrophic aquaculture (IMTA) systems are a sustainable alternative to
traditional monocultures. These systems can operate in temperate marine waters and
incorporate commonly cultured regional species (Barrington et al., 2009). The most
commonly utilized model comprises of fed finfish, filter feeding shellfish, and extractive
macroalgae (Barrington et al., 2009). However, systems can be simplified from this
model to a primary producer/primary consumer system based on fundamental principles
of nutrient cycling. In this context, macroalgae acts as an extractive component for
inorganic nutrients produced by shellfish (Barrington et al., 2009). Filter feeding bivalves
incorporate detrital macroalgae into their diets in traditional monoculture settings
(Ruesnick et al, 2014). Thus it is reasonable to assume that in an integrated aquaculture
system, shellfish will similarly ingest a portion of ambient detrital cultivated macroalgae.
IMTA systems help mediate negative aquaculture contributions to eutrophication by


 

26
 

confining nutrient cycling within the system. Despite extensive literature supporting the
benefits of IMTA, only a 16% increase in the practice has been observed in temperate
marine environments for 1999-2009 (Barrington et al., 2009).
Shellfish monocultures produce high concentrations of ammonium, which
compound with environmentally sourced nutrients to exacerbate harmful local conditions.
C:N ratios indicate the relative nitrogen assimilation of a primary producer in relation to a
nitrogen source (Peterson & Heck, 1999). The carbon nitrogen ratio within the tissue of a
primary producer is a direct reflection of available environmental nitrogen (Peterson &
Heck, 1999). Thus, increased amounts of available ammonium accessible to Ulva spp.
tissue will presumably result in a lower C:N ratio within that tissue. Assessing the C:N
ratio of Ulva spp. near V. philippinarum would quantify the relative assimilation of
bivalve-produced DIN for applications to a bioremediative IMTA setting.
Quantifying detrital contribution from primary producer to primary consumer is
possible using stable isotope methodologies (Dunton & Schell, 2003). A case for the cocultivation of seaweed and bivalves could result from the quantification of detrital
material into bivalve diets. Evidence from Pacific Oysters (Crassostrea gigas) suggest
that in the presence of abundant detrital material, dietary incorporation reaches a
maximum of ~9% detrital material (Ruesnick et al, 2014). However, inter and intra
species variability as well as regional environmental conditions may allow V.
philippinarum in the Hood Canal to incorporate higher levels of detrital macroalge into
their diet. Additionally, smaller V. philippinarum have an increased capacity for detritus
uptake (Suh & Shin, 2013). If stable carbon isotope analysis illuminates V. philippinarum
as a significant sink for detrital material, it would represent in piece of an


 

27
 

environmentally beneficial symbiosis in an aquaculture context.
Examining Case Studies and Identifying Existing Questions: a novel project
proposal
Marine ecological literature warrants further quantitative investigation into the
interaction between V. philippinarum and Ulva spp. in the Northern Hood Canal. The
following section extensively examines case studies relevant to nutrient exchange
between these species in attempts to diversify the literature applied to a commercially
relevant scenario in the Northern Hood Canal. Additionally, this section will pose a novel
research question surrounding this distinct set of species, and unique locale. This question
will ultimately serve to address the quantitative component of a larger purpose: the
possibility of supporting innovation within the shellfish industry.

The impact of shellfish monoculture on macroalgae proliferation
The ability for dense bivalve monocultures to positively impact macrophyte growth
is well-documented in the literature. Peterson and Heck use a simple experiment to
quantify the effect of ammonium, introduced by suspension feeding bivalves, specifically
Modiolis americanus, on the growth of the seagrass Thalassia testudinum. The study
controls for confounding factors by conducting the experiment within laboratory settings.
This controlled environment allows for the researchers to determine exact species
densities and proximity to one another. The results demonstrate two fundamental
concepts. 1) Ambient ammonium concentration increases with increasing bivalve density;
2) C:N ratios in macrophyte leaf tissue are negatively correlated to shellfish density. In
treatments where shellfish density is at 0, Thalassia testudinum tissue displays a C:N
ratio of 16.31 ±0.38. Whereas treatments containing 500 and 1500 mussels display C:N


 

28
 

ratios of 14.85±0.67 and 13.37±0.38 respectively (Peterson & Heck, 1999). As C:N ratio
in Thalassia testudinum tissue is inversely correlated with shellfish density, it suggests
that the presence of bivalves increases nitrogen assimilation in primary producers,
lowering C:N ratios.
More recent studies attempt to quantify the effect of pre-existing monocultures on
macroalgae growth in the field. Similar to the Perterson and Heck study, a study
undertaken by Zertuche-González et al in 2008, focuses on species of Ulva found in a
sub-tropical bay near intensive shellfish monocultures. The study relies primarily on Ulva
dry weights obtained from plots, and quantitative results from a C,H,N analysis.
Researchers additionally monitored total dissolved inorganic nitrogen (TDIN) over the
two year study. An increase in TDIN, specifically ammonium, resulted in greater %N in
Ulva tissue. Increases in %N also corresponded to greater biomass. The study concluded
that Ulva responds to environmental fluxes in TDIN and can act as a temporary nutrient
sink.
A Puget Sound study, undertaken by Saurel et al, 2014 utilized a combination Farm
Aquaculture Resource Management (FARM) model, combined with mid-year Ulva
biomass measurements. Results from the simulations and field support the prediction that
ammonium contributes significantly to macroalgal production. Each V. philippinarum
individual produces 0.3 g of ammonium over a growing cycle of three years. In the first
year of the study, the average difference between Ulvoid biomass growing on predator
exclusion nets was 1 gdw (gram dry weight) m-3 higher for nets exposed toV.
Philippinarum relative to controls that contained no V. philippinarum. The subsequent
two years showed an increasing difference between nets associated with V. philippinarum


 

29
 

and nets unassociated with V. philippinarum (differences were ~2.5 and ~4.0 gdw m-3
respectively).This indicates accumulation of nutrients over the study period, despite the
seasonality of Ulvoid blooms. The Saurel et al study, proposes the crucial consideration
of utilizing the harvestable Ulva biomass to prevent nitrogen from reentering the system
via tissue degradation.
The above studies serve to support further quantitative assessment of nutrient
cycling on a V. philippinarum cultivation plot in the northern Hood Canal. Both the
Peterson and Heck study and the Zertuche-González et al 2008 study quantify the amount
of N assimilated into macrophyte tissue in the presence of shellfish. However, neither
study uses V. philippinarum as a species of interest. Additionally, both of these studies
focus on subtropic shellfish cultivation scenarios, leaving space to quantify interactions in
temperate regions. The Saurel et al. (2014) study examines the temperate Puget Sound,
and utilizes V. philippinarum as a species of interest. However, this study investigates
biomass gains apart from nutrient composition within Ulva tissue. Though these studies
reflect and support the conception that shellfish cultivation leads to increased nitrogen
assimilation/biomass increase in primary producers, further specific investigation is
warranted.
Extrapolating research from V. philippinarum cultivation sites in the North Puget
Sound can help illuminate the biomass proliferation scenario taking place on aquaculture
sites in the northern Hood Canal. The extent of N assimilation into Ulva tissue as a result
of V. philippinarum cultivation remains undefined for this region. The following question
poses to address this unknown:
Do V. philippinarium monocultures result in increased nitrogen assimilation, as
measured by C:N ratios, in Ulva spp. tissue?


 

30
 

If successfully answered, the results from this investigation will better inform the
management of seasonal Ulvoid blooms on aquaculture sites in the region, and have
positive environmental implications.

Potential V. philippinarium dietary shift during seasonal macroalgae blooms
Shellfish are well-known to have variable diets, largely dependent on the nutrient
sources available in their local environments. Research abroad of V. philippinarum, and
local studies focusing primarily on the Pacific Oyster C. gigas, give insight to the
environmental and physiological factors that dictate dietary preferences in cultivated
bivalves. However, the literature alludes to circumstances which remain largely uninvestigated. The capacity of V. philippinarum to alter their diet to consume detrital
material during summer algal blooms in the northern Hood Canal is a circumstance
which calls for further scientific exploration.
Case studies from outside the temperate Pacific Northwest provide evidence of V.
philippinarum’s ability to incorporate detrital material as a dietary supplement. Suh and
Shin 2013 quantify the intra-specific differences in dietary preferences of manila clam
age classes in the Korean Yellow Sea. The study investigates the ability of size class to
determine the dietary incorporation of detrital vs. microalgal material. Results indicate
that larger clams more readily incorporate microalgae into their diets as opposed to
detrital material. However, the incorporation of detrital material is variable based on
seasonal availability, and in some cases comprises 20-30% of the larger clam diet.
Smaller juvenile clams incorporate much higher proportions of detritus into their diets


 

31
 

(47.1 to 51.2%) (Suh & Shin 2013).
A separate study from the W. Pacific, specifically the Miya Estuary of Japan,
investigates the potential of V. philippinarum to incorporate terrestrial material into their
diets. The paper by Kasai, Horie, & Sakamoto 2004, concludes that marine POM
(MPOM) was highly preferred over terrestrial POM (TPOM) because of the high lignin
content of TPOM, and the high concentration of nitrogen in MPOM. However, during
heavy rain events, when high concentrations of TPOM were present, the isotopic
signatures of the clams shifted to more closely reflect the TPOM signature due to the
availability of this material.
Many of the case studies involving delineating diet via isotopic signature in the
Pacific Northwest focus primarily on C. gigas. In one study, conducted by Ruesink et al
in 2013, researchers assessed growth rates in C. gigas as correlated with environmental
parameters such as temperature, salinity, sediment type, dissolved oxygen concentration
and resource availability, throughout the Puget Sound. This study takes into account the
variability in primary producer composition, such as the abundance of microalgae,
microphytobenthos, and macroalgae, over the study area. Organism growth was strongly
positively correlated with temperature. However, no significant conclusions could be
drawn in terms of broad dietary preference comparing within seasons. A relative isotopic
depletion occurred in clam tissues in the winter when terrestrial inputs were presumably
higher. Otherwise, the isotopic signature of the organisms similarly reflected that of the
primary producers in the local environment, which was not a significant factor in
determining growth. Hence, this comparative study over wide geographic range yielded
results indicating a certain dietary specificity for each test region.


 

32
 

A more recent study conducted by Conway-Cranos et al in 2015 focuses on a wider
range of the Puget Sound basin, including the Hood Canal. The study examines the
primary dietary inputs to C. gigas for each specific locale. Isotope data indicate that
macrophytes, including Ulvoid species, upland vegetation, sea marsh, and marine grasses,
comprise a large portion (over 40%) of C. gigas diets in the Hood Canal. A stable
isotope study, conducted by NOAA in 2011, reports the capacity of C. gigas to consume
particulate effluent when present nearby, despite an indicated preference for microalgae
(Rensel, Bright, and Seigrist, 2011). Overall, the literature from within the Puget Sound
region suggests the tendency of C. gigas to uptake the POM characteristic to a given
region.
The aforementioned studies indicate that bivalves have the capacity to shift their
diets as a reflection of available food sources. Literature focusing on V. philippinarum
diets in the W. Pacific gives insight into the environmental and physiological parameters
affecting food preference. However, these studies do not take into account the high
Ulvoid bloom density, which occurs in N. Hood Canal growing areas. Studies from the
Puget Sound region often focus on quantifying the readiness of commercially significant
C. gigas to incorporate available organic matter. Additionally, this research does not
focus specifically on seasons, or subregions, most affected by dense Ulvoid blooms.
Hence, there remains a need to assess the assimilation of available environmental
components into V. philippinarum diets during the prolific summer macroalgae blooms.
The following question poses to address this unknown:
Does V. philippinarium’s microalgae-based diet shift to include proportions of detrital
Ulva spp. tissue during seasonal macroalgae blooms?
Answering this question will inform the aquaculture industry of the seasonal nutrient


 

33
 

cycling dynamics which take place on V. philippinarum plots during the summer Ulvoid
blooms.

Potential symbiosis between Ulva spp. and V. philippinarum
Based on the above conclusions, the possibility of symbiotic nutrient cycling
between V. philippinarum and Ulva spp. seems highly likely. A vast body of literature
exists, which addresses nutrient exchange in controlled integrated multitrophic
aquaculture (IMTA) contexts. Integrated Multitrophic Aquaculture is the co-cultivation
of two or more species in order to mimic the nutrient cycling scenarios present in natural
ecosystems. One of the more comprehensive studies of IMTA, undertaken by
Barrigngton et al in 2009, remarks on the effectiveness of co-cultivating shellfish and
macroalgae. The role of bivalves and macroalgae in IMTA systems remain consistent
with the species’ ability to cycle nutrients in their native ecosystems.
Macroalgae acts as an inorganic nutrient extractive agent for the inorganic effluent
produced by other species in the system (Barrington et al, 2009). In 2007, Buschmann
conducted a study focusing on the Pacific Coast of S. America. In Chile, macroalgae is
commercially farmed in conjunction with salmon, abalone, and mussels (Buschmann,
2007). The study attempts to optimize the depth and structure of macroalgae cultures to
maximize inorganic N extraction to prevent the proliferation of eutrophying algal blooms.
Studies of IMTA off the coast of South Africa report that using macroalgae with abalone
aquaculture reduces inputs of nitrogen into surrounding ecosystems by an average of 4.4
tons per year, once harvested (Nobre, et al, 2010). The international success of
macroalgae in reducing large quantities of inorganic off puts from shellfish aquaculture


 

34
 

farms is promising for areas of intensive shellfish monocultures, such as the Puget Sound.
Bivalves act to incorporate particulate organic matter as a result of effluent or tissue
degradation within IMTA systems. A study by Bolton et al in 2008 assesses the strengths
of co-cultivating Ulva spp. and abalone. As abalone readily consume particulate Ulva
spp., co-cultivating the species would reduce the required feed input into the abalone. The
inorganic nitrogen produced by the abalone would, in turn, fertilize the Ulva spp.
Additionally, the harvested Ulva spp. would be processed for meal on shore to be fed
back to the abalone, supplying the majority of their dietary needs. The ability of
particulate Ulvoids to contribute to abalone diets may closely mimic the dietary intake of
V. philippinarum during summer Ulvoid blooms on the Hood Canal.
Integrated Multitrophic aquaculture systems are not standard practice in
Washington. However, strong local interest exists in researching IMTA systems. NOAA
conducted a study in 2011 which placed Pacific Oysters (C. gigas) and Blue Mussels (M
edulitus) near working finfish aquaculture pens in an open-exchange system in the Puget
Sound. This system allowed for both effluent and phytoplankton-based POM to flow
through to the shellfish. The study uses stable isotope signature from the pen effluent to
quantify the uptake of the effluent in adjacent shellfish plots. The results indicate that
growth near effluent increases, though shellfish species selectively feed on
phytoplankton. These findings support the benefits of IMTA systems (Rensel, Bright,
Seigrist, 2011). Currently there is strong interest in understating the benefits of shellfishmacroalgae open exchange systems. Research utilizing pre-existing species on potential
sites in the Hood Canal and Puget Sound may give insight into the interaction of
cultivated macroalgae and pre-existig bivalve infrastructure, barring the finfish


 

35
 

component often common to IMTA systems.
As a vast body of literature hails the positive environmental impacts of IMTA
systems, Washington’s willingness to participate in IMTA-related research could greatly
benefit local aquaculture and ecosystems. IMTA has a strong foothold in S. Africa and
the Pacific coast of South America. However, the ecosystem dynamics and commercially
relevant species of these locales do not directly translate to Washington aquaculture
systems. The above studies provide strong support for the integration of macroalgae and
shellfish aquaculture based on quantitative principles of nutrient cycling. Studies in
temperate regions, such as the Bay of Fundy (Barrington et al, 2009) and Puget Sound,
demonstrate the ability for shellfish to act as an organic nutrient extractor from farm
effluent. The missing piece is quantifying nutrient exchange in the field using
commercially relevant V. philippinarum and native Ulva spp.. Fortunately, the dense
summer Ulvoid blooms offer the perfect naturally occurring opportunity to study nutrient
exchange between these species prior to investing in additional infrastructure.

CHAPTER 2: MANUSCRIPT
Introduction
The Hood Canal is a deep fjord comprising part of the main Puget Sound basin in
Washington State. This canal provides vital habitat to a wealth of native and established
marine species. Several molluscan species endemic to the Canal serve to support
Washington State’s flourishing seafood economy. The most profitable of these
commercial species, the Pacific Oyster (Crassostrea gigas) and the Manila clam
(Venerupis philippinarum), were introduced into the region from the Western Pacific in

 

36
 

the early 20th century (Humphreys et al, 2015). Over time, these species established
themselves into Washington’s tidal flats, providing the foundation for the modern multimillion dollar shellfish industry. Today, Hood Canal growing regions lead the State in
manila clam production (Booth, 2014). As the Canal is vital to the health and continued
success of the shellfish industry, understanding the interactions of the local
environmental fluxes and clam monocultures is of utmost importance.
Though the Hood Canal provides ideal physical habitat for the introduced V.
philippinarum, biogeochemical cycling in the Canal can result in undesirable seasonal
growing conditions within the region. Evidence suggests that the seasonal blooms of
macroalgae, including Ulva spp., negatively impact marine life, including certain species
of cultivated shellfish (Newton et al, 2007). Ulvoid bloom intensity has increased in
recent years for aquaculture sites on the Northern Hood Canal, presumably as a result of
increased anthropogenic nutrient loading. When large blooms of primary producer
degrade, dissolved oxygen is rapidly depleted. For the Hood Canal, eutrophication is
especially severe, as the exchange of water from the Canal basin into the Strait of Juan de
Fuca is limited by geographic barriers (United States Geological Survey [USGS]).
Negative impacts from the deterioration of large macroalgae blooms on V.
philippinarum growth remain unquantified for the northern Hood Canal growing regions.
Existing evidence suggests, that when present in high densities, Ulvoids decrease the
overall growth of Pacific Oysters (Lamb, 2015). However, there is some question as to
whether the Ulvoid densities of 0, 1.5, and 3.0 kg in the C. gigas experiment accurately
reflected true densities present on the bags (Lamb, 2015). Additionally, the mechanism
inhibiting successful growth in this species remains unknown. Discussions among


 

37
 

growers hypothesize that Ulva spp. draws down vital DO within the bags, or that the
Ulvoids reduce the flow of more labile food sources through the bags. Uncertainty exists
as to whether the dense blooms have a similar impact on commercially significant
cultivated species, such as V. philippinarum. Field and industry observations suggest that
mortality in V. philippinarum populations does not increase during the summer harmful
bloom events (Joth Davis, personal communication, 2015). Ulva may provide a food
source to shellfish grown for aquaculture purposes, negating some of the negative
impacts described previously. Hence, a more complex, and possibly synergistic,
mechanism may underlie the seasonal coexistence of these two species.
It is unknown as to what degree the Ulva spp. blooms, which cover the V.
philippinarum cultivation site, provide an alternate food source to the local clam
population. Growers speculate that dense blooms inhibit to the flow of phytoplanktonbased seston to the clams. V. philippinarum depend heavily on phytoplankton-based
seston as a primary food source (Conway-Cranos et al, 2015; Suh & Shin, 2014; Resnsel,
Bright, Seigrist, 2011; Kasai, Horie, & Sakamoto, 2004). However, Ulva spp. degrades
rapidly (Zertuche-González et al, 2008). Hence, the detrital material could be
supplementing V. philippinarum diets. The negative impacts of Ulva spp. on C. gigas
growth may not directly translate to V. philippinarum due to the intraspecies differences
in feeding behavior and/or cultivation method. Hence, the effect of Ulva spp. on V
philippinarum may represent a relationship that exceeds the initial negative stigma.
In addition, recent evidence suggests that macroalgae Ulva spp. benefits from
cultivated shellfish (Saurel et al, 2014; Zertuche-Gonzalaz et al, 2008). Monocultures
contribute to the annual increase in prolific blooms in monoculture-supporting bays


 

38
 

(Saurel et al, 2008; Zertuche-Gonzalaz et al, 2008). The capacity for shellfish cultivation
to significantly increase the nutrient assimilation and proliferation of primary producers
is true for the Puget Sound (Saurel et al, 2014). However, within the Canal, specifically
the Northern Hood Canal, the transfer of inorganic nutrients from V. philippinarum
monocultures to Ulva spp. remains unexamined. Question remains as to whether shellfish
monocultures exacerbate local blooms, further contributing to regional poor water quality
conditions post degradation. If Ulva spp. proves to act as a significant temporary sink for
inorganic nutrients released by shellfish cultivation, industry could effectively manage
the blooms so as to reduce negative environmental impacts. Management (ie. removal)
could additionally decrease the proliferation of Ulva spp. in subsequent years by
removing excess nutrients from the poorly flushed system.
Ecological and economic and value exists in investigating the relationship between
seasonal macroalgae blooms and V. philippinarum monocultures. The question which
most adeptly addresses the concerns of ecologists and industry follows:
Are commercially farmed manila clams (Venerupis philippinarium) and seasonally
occurring green macroalgae (Ulva spp.) involved in symbiotic nutrient cycling in the
context of bag aquaculture on the Northern Hood Canal?
Do V. philippinarium monocultures result in increased nitrogen assimilation, as
measured by decreased C:N ratios, in Ulva spp. tissue?
Does V. philippinarium’s microalgae-based diet shift to include higher proportions of
detrital Ulva spp. tissue during seasonal macroalgae blooms?
To answer this question, an experiment will take place allowing for the
quantification of nitrogen assimilation in primary producer tissue via elemental ratios.
Stable isotope analysis will allow for the quantification of Ulvoid contribution to V.
philippinarum diets. Tissue will be examined from a treatment, which mimics the status


 

39
 

quo summer conditions, as well as control treatments, which isolate V. philippinarum and
Ulva spp. tissue from one another.
Quantifying the extent of nutrient cycling between the two species will illuminate
the role of V. philippinarium cultivation in enhancing seasonal harmful macroalgae
blooms. Additionally, the contribution of Ulva spp. to summer V. philippinarium diets,
and the subsequent effect on growth rate will be illuminated. Findings could help
promote sustainable management practices, and encourage the diversification of industry
towards the co-cultivation of shellfish and macroalgae. Ideally, a more thorough
understanding of this multitrophic interaction will result in a net positive ecosystem
benefit surrounding V. philippinarium cultivation.

Materials and Methods
Study Location and Background
This study took place on the Northern Hood Canal. As a whole, the Hood Canal is a
deep estuarial fjord comprising part of the main Puget Sound basin (Warner & Kawase,
2001). It is separated from external water bodies via the Admiralty Sill, which restrict
water from circulating out of the Canal in the north (Warner & Kawase, 2001).
Thorndyke Bay, which comprises the location of this study, is an estuarial bay located in
the Northern Hood Canal, at the Admiralty Sill. The Thorndyke Bay ecosystem
represents a rare, pristine mid-sized creek estuary, identified as a “priority conservation
area.” (Harrington, 2005). The beach provides excellent habitat for wintering waterfowl
(WDFW, 2004 in Harrington, 2005) and shellfish. Due to the lack of development or
alteration in and around the Thorndyke Creek estuary, the area functions as high value


 

40
 

native species habitat (Harrington, 2005).

Figure 3. Location of the study site, Thorndyke Bay, situated in the Northern Hood
Canal. All field measurements were collected at this site.
The Thorndyke Bay estuary is composed primarily of sandy glacial sediment,
deposited via the erosion of exposed bluffs in the north (Harrington, 2005). The absence
of shoreline development in the area allows for continued natural deposition of
sediments to the beaches. Vegetation along the shoreline is comprised primarily of salt
marsh commonly associated with small stream mouths. The stream running into the area,


 

41
 

Thorndyke Creek, supports a 32 acre marsh above the high tide line (Harrington, 2005).
This marsh is classified as a low, silty marsh, and forms at the mouth of a mid-sized
creek. Thordyke Bay itself is a shallow body of water with relatively weak tides (Cannon,
2005). As the bay is shallow, the low tide line is relatively far from shore.
The Baywater Shellfish Company operates in the intertidal zone just north of
Thorndyke Creek. Baywater Inc harvests four types of shellfish for local commercial
distribution, including the manila clam, Venerupis philippinarum. Every year, from May
through August, green macroalgae blooms of Ulva spp. completely cover V.
philippinarum plots (Joth Davis, personal communication, 2015). Hence, this site
provides the essential components for studying the extent of nutrient cycling between
Ulva spp. and V. philippinarum. To quantify nutrient cycling at peak Ulva spp. density
and maximum V. philippinarum metabolic rate, data collection took place from June
2015 through September, 2015. The site’s rare position amidst and unfettered estuarial
ecosystem, and unique susceptibility to intensive Ulvoid blooms, designated it as the
ideal site for this research project.
Duration of study
This study began on June 27th, 2015 and ended on September 5th 2015. This time
frame was chosen in attempts to capture the majority of the Ulvoid growing season. All
data was collected on an outgoing tide of < 0.0 ft.

Field Design
All samples were collected from a working V. philippinarum aquaculture plot. In
this plot, V. philippinarum were segregated into in-ground aquaculture bags containing


 

42
 

approximately 100-300 live individuals each. The bags were laid in several dozen rows
perpendicular to the shoreline, each row containing individuals of a similar age. All bags
were industry standard of 1/2’’ hard plastic mesh, with the dimensions of 18” x 32” x 4.”

Figure 4. The image on the left portrays the standard layout of in-ground aquaculture
bags along the intertidal zone. Bags are distributed in a uniform grid pattern. Rows of
similarly aged clams run perpendicular to the waterline. The Right-hand image is a cross
section of an in-ground aquaculture bag. Clams remain submerged in the sediment, while
the top of the bag remains exposed. Toba, Dewey, & King, 2005.
One row containing 15 total bags, running perpendicular to the waterline (~20 m
from the high tide line) was used for this experiment. The 15 bags were segregated into
three separate treatments, 5 replicates of each, to assess the extent of nutrient cycling
between the bag-cultivated clams and seasonally occurring green macroalgae. Replicates
were randomized. The first treatment acted as a control and contained exclusively V.
philippinarum, in order to assess the dietary intake of V. philippinarum in the absence of
a dense Ulva spp. covering. The second treatment, also a control, contained no clams
within the bag, but maintained a dense Ulva spp. covering on top of the bag. The purpose
of this treatment was to assess the extent of nitrogen uptake in Ulva spp. in the direct
absence of V. philippinarum. The third treatment mimicked the status quo during the


 

43
 

summer growing season, containing both V. philippinarum and Ulva spp.. The goal of
this treatment was to assess dietary composition in V. philippinarum individuals directly
contacting Ulva spp.. Additionally, this treatment addressed nitrogen uptake in Ulva spp.
directly in contact with V. philippinarum. The three treatments contained within the
experimental row are summarized below:
Control 1: Exclusively V. philippinarum (no Ulva spp.)
Control 2: Exclusively Ulva spp. (no V. philippinarum)
Treatment: V. philippinarum and Ulva spp

Figure 5. The lefthand picture is the experimental row in the field. The righthand image is
a digram of the three treatments as they were randomly distributed throughout the
experimental row.
Treatments containing V. philippinarum, were standardized to include 150 clams,
two years of age, for each bag prior to the experiment. The quantity 150 was chosen to
reflect the average number of live clams per bag as observed in the surrounding
aquaculture plot. It was essential to ensure that all clams be of similar size and age, as
ammonium excretion (Mann & Glom, 1978) and preference for detritus (Suh and Shin,
2014) are size/age-dependent factors. In choosing clams all of the same age and
broodstock, it was assumed that size be consistent amongst individuals.

 

44
 

Initially, each bag within the experimental row was covered in thick attachments of
Ulva spp.. These attachments had formed in May, and as such, were well established by
the time the experiment began. The Ulva spp. only control bags, as well as the
combination treatment bags, were left with the pre-existing attachments. These
attachments were left in tact to most accurately mimic the status-quo conditions on the
bags during summer bloom events. No additional Ulva spp. was introduced into the
treatments. This experiment relied exclusively on the regeneration of Ulva spp.
attachments on the in-ground bags to capture the fluctuations in natural density
throughout the growing season. Ulva spp. also existed in a free-floating form. However,
this unattached Ulva spp. is randomly deposited and removed via daily tidal action.
Hence, free-floating Ulva spp. was not considered as a variable in this experiment. Preexisting attachments on V. philippinarum only treatments were severed at the onset of the
experiment, and scraped weekly to prevent further growth and potential reduction in
treatment effect.

Sample Collection and Processing
Data Collection in the Field
V. philippinarum:
To detect any potential treatment effect unaccounted for by quantitative nutrient
analysis, clam growth was measured in each of the 10 total clam bags every other week
from June 27th 2015 to August 15th, 2015. Bags were dumped into large plastic sorting
trays and the height and width of every tenth clam measured with calipers. Any
mortalities in the bags were recorded, and the dead individuals were removed from the


 

45
 

total population. Weekly, 3 randomly chosen individuals were collected from each bag
and put on ice for transportation to the laboratory for isotopic content and elemental ratio
analyses preparation.
Ulvoid abundance fluctuated in response to ambient conditions, which was
unanticipated. Hence, qualitative designations were utilized to capture these changes in
Ulvoid abundance for July 31, 2015 through September 5, 2015. In order to estimate the
amount of Ulva spp. attached to the Ulva spp. control and combination treatment bags,
qualitative proxy measurements were taken twice during the experiment. To not disturb
the tissue on the experimental row, proxy high, med, and low samples were collected
from working V. philippinarum bags in the surrounding cultivation plot to obtain dry
weight measurements. All ten Ulva spp.-containing bags in the experimental row were
visually assessed and designated high, med, and low in relation to their density of Ulva
spp. attachments. A general guideline designated an Ulvoid covering from 0-33% of the
surface of the exposed bag as low, 33-66% as medium, and 66-100% as high. Depending
on daily conditions, bags ranged from three-quarters of the bags covered in “low” density
Ulva spp. to four fifths of the bags covered in “high” densities of Ulva spp. These
samples were transported in ziplock bags on ice to the laboratory to obtain dry weight.


 

46
 

Figure 6. Examples of bags receiving a low, medium, and high designation. This photo
was taken during the mid-July Ulvoid die-off. Hence, bags in this picture represent the
lower end of Ulva spp. coverage.
Weekly, approximately 5 g of Ulva spp tissue was collected from each of the ten bags in
the experimental row containing Ulva spp., and transported on ice in small ziplock bags
to the laboratory to be prepared for isotope and C/N ratio analysis.
Phytoplnkton-based Seston
In order to assess the relative contribution of phytoplankton-based seston to V.
philippinarium diets in the various treatments, phytoplankton was collected weekly
starting July 13, 2015 and running through the end of the experiment. Samples were


 

47
 

collected on an incoming tide using a phytoplankton net into an acid washed (HCl) amber
bottle. All excess seston on phytoplankton net was rinsed into bottle with DI water in the
field. Bottles were then transported indirectly on ice to the laboratory for isotope analysis
preparation.

Laboratory Sample Preparation
V. philippinarum:
In the laboratory, samples from each of the ten bags were prepared for isotopic analyses
of δ13C and 15N following the procedures of Levin and Currin (2012). To do so, clams
were separated into 24 oz plastic containers and covered with ~20 oz of seawater
collected on site. The clams were then allowed to purge their stomach contents for 24
hours to prevent any potential cross-contamination into the stomach gland during
dissection. Dissection tools and surface were sterilized with 70% ethanol prior to the
dissection of each individual. Abductor mussels on the clam were cut using dissecting
scissors and the clam was internally rinsed with DI water. The stomach gland was
removed, rinsed with DI water, and stored in a small plastic ziplock. The stomach gland
was chosen as it most accurately reflects short-term dietary preferences in marine
invertebrates (Levin and Currin, 2012). Clam stomach glands from individuals in the
same bag were combined to form a composite sample to reflect the average diets of the
clams within each individual aquaculture bag. Stomach glands were then frozen at -20 °C
for 5-8 months prior to analysis preparation.
Three composite samples, belonging to each of the treatment categories, from each
week of data collection were removed from the freezer in February 2016 and allowed to


 

48
 

thaw. Samples were then placed in combusted petri dishes and dried at 60 °C for 24 hrs
(Levin and Currin, 2012). Samples were then placed into combusted ceramic mortar and
pestle and pulverized to a fine powder prior to preparation for isotopic analysis, described
below.
Ulva spp.:
Once in the laboratory, the 5 g samples were squeezed to remove DI water
and frozen at -20 °C (Levin and Currin, 2012) for 5-8 months prior to analysis
preparation.
Samples were removed from the freezer in February 2016 and allowed to thaw.
Samples were then placed in combusted 60 x15 mm petri dishes via sterilized forceps and
dried at 60 °C for 24 hrs (Levin and Currin, 2012). Samples were then placed into
combusted ceramic mortar and pestles and pulverized to a fine powder prior to
encapsulation.
The three qualitative Ulva samples were rinsed 3x with DI water. Qualitative
samples were spun in separate batches according to classification in a plastic salad
spinner for 10 seconds to remove excess water. The samples were then transferred to foil
to be dried in the oven at 60°C for 24 hrs. Dry weights were and recorded.
Phytoplankton-based Seston:
Once in the laboratory, samples were filtered through a combusted glass
vacuum filtration apparatus using a GF/F filter. Filters were dried in the oven at 60 °C for
24 hrs (Levin and Currin, 2012). Dried filters were wrapped in combusted foil and stored
in a cool, dry place for 5-8 months prior to encapsulation.


 

49
 

Stable Isotope Analysis
Pulverized composite samples of both Ulva spp. and V. philippinarium were later
transferred into pre-combusted 5x8 mm tin capsules for isotope analyses. Approximately
2.0 mg of Ulva spp. sample, and 1.25 mg of V. philippinarium sample were weighed into
separate capsules using the Perkin Elmer AD 6000 Ultra Microbalance, folded and
shipped to the UC Davis Stable Isotope Facility for C/N ratio determination and δ13C and
15

N analysis via a PDZ Europa ANCA-GSL elemental analyzer interfaced to a PDZ

Europa 20-20 isotope ratio mass spectrometer (stableisotopefacility.ucdavis.edu,
2/22/2016).
In February 2016, filters were removed from foil and prepared on a surface
sterilized with 70% ethanol. A sterilized 5 mm diameter cork borer was used to remove 8
small sections of the filter. Measurements were made of the diameter of seston on the full
filter in order to calculate the proportion of material removed and sent for analysis. The 8
sections were placed in a pre-combusted 9x10 mm tin capsule. The capsule was then
folded into a cube <8 mm and placed in a 48-well tray. The tray was sent for C/N ratio
determination and 13C and 15N analysis at UC Davis as described previously.

Statistical Analysis
Statistical analysis took place in JMP pro version 12. To visualize growth trends, V.
philippinarium measurement values were averaged from every replicate, for each
collection date. Growth rate was calculated for each replicate by taking the difference
between the final and initial measurements and dividing by the time elapsed. For growth
rates, variance homogeneity was tested using Levene’s test, normality was assessed using


 

50
 

Shapiro-Wilke’s test. Data was normally distributed and displayed the similar variance.
Treatment effect on V. philippinarium growth was analyzed using one-way ANOVA,
followed by a Tukey’s HSD test. Carbon and nitrogen ratio data was transformed from
micrograms to moles, which were used to find the atomic ratio. Data was normally
distributed and displayed equal variance. Hence, means between treatments were
compared using one way ANOVA. Analysis via Tukey’s HSD test verified significant
differences between treatment means.
Isotope relative abundance values for phytoplankton and Ulva spp. were compared
using a t-test and assessed for normality via a Shapiro-Wilke’s test. The two groups were
shown to be normally distributed. Carbon isotope relative. abundance values for the V.
philippinarium stomach glands were adjusted to reflect a fractionation of +0.6‰. This
fractionation value reflected that used in Suh & Shin 2013 dietary isotope analysis of V.
philippinarium. Nitrogen relative abundance values were adjusted to reflect a +2.9‰
fractionation rate. Values for both isotopes were normally distributed via ShapiroWilke’s test.
Results

V. philippinarum growth
Mean length measurements of V. philippinarum, taken at intervals throughout the
summer, are displayed in Tables 1 and 2. Initial average measurements ranged from 3.764.12 cm for individuals in V. philippinarum only (control) bags. For individuals in V.
philippinarum and Ulva spp. (treatment) bags, initial average values ranged from 3.854.10 cm. Final measurements for individuals in control bags ranged from 4.12-4.27 cm,
and for individuals in treatment bags from 4.18-4.29.


 

51
 

Replicate/Date
Control 1
Control 2
Control 3
Control 4
Control 5
Mean
Treatment 1
Treatment 2
Treatment 3
Treatment 4
Treatment 5
Mean

6/27/15
4.12 ± 0.13
4.12±0.08
4.11±0.06
4.11±0.06
3.76±0.06
4.04±0.07
3.92±0.07
3.92±0.07
3.85±0.09
3.87±0.06
4.10±0.07
3.91±0.04

7/18/15
4.01±0.09
4.15±0.06
4.11±0.06
4.11±0.09
3.82±0.09
4.04±0.06
4.00±0.03
3.96±0.03
3.91±0.03
3.96±0.02
4.02±0.02
4.00±0.02

8/1/15
3.92±0.09
4.11±0.05
4.14+/0.09
4.30±0.08
3.97±0.11
4.09±0.07
4.00±0.09
3.90±0.08
4.05±0.08
3.96±0.07
4.00±0.07
3.98±0.02

8/15/15
4.24±0.08
4.27±0.05
4.27±0.06
4.27±0.02
4.12±0.06
4.26±0.05
4.18±0.06
4.24±0.06
4.24±0.09
4.27±0.05
4.29±0.05
4.24±0.01

Table 1. Average length measurements in cm coupled with standard error for each of the
five V. philippinarum only control and treatment replicates. Measurements were taken
from late June- Mid August.
Control values show a universal increase in average length over the measurement
period. Wide variation exists between the means of the initial measurements (0.36 cm), as
well as the means of the final measurements (0.14 cm). Additionally, negative growth
trends between initial and final measurements were recorded. This may be due to small
sample size and natural wide variability in shell lengths as the same individuals were not
repeatedly sampled.
All length values show an increase over the measurement period. Lesser variation
(0.18 cm) exists in the initial measurements of the treatment bags, as compared to that of
the control bags. Means of the final measurements also exhibit a lesser difference from
initial values than those of the control replicates (0.11 cm). Additionally, negative growth
trends between initial and final measurements were recorded. This may be due to small
sample size, as clams were only measured four times throughout the experiment, and
natural wide variability in shell lengths. To account for measurement inconsistencies
throughout the measurement period, only final and initial values were taken into


 

52
 

consideration for the growth rate (Fig. 7)
4.5

Length in cm

4.4
4.3

Control 1

4.2

Control 2
Control 3

4.1

Control 4

4

Control 5

3.9
3.8
3.7

Date

Figure 7. Length in centimeters of V. philippinarum control group over time. Each line
represents a different replicate within the control group.
4.4
4.3

Length in cm

4.2
Treatment 1

4.1

Treatment 2

4

Treatment 3

3.9

Treatment 4

3.8

Treatment 5

3.7

Date

Figure 8. Length in centimeters of V. philippinarum grown with ulva spp. (treatment
group) over time. Each line represents a different replicate within the treatment group.


 

53
 

Treatment
 1

%
 Growth
 
length
6.8

Control
 1

%
 Growth
 
length
3.4

Treatment
 2

8.3

Control
 2

7.3

Treatment
 3

10.0

Control
 3

3.9

Treatment
 4

10.6

Control
 4

6.3

Treatment
 5

5.8

Control
 5

9.5

Mean

8.3±1.0
 

Mean

6.1±2.7
 

Table 2. Mean length growth rates for each of the treatment and control replicates
between June 27th and August 15th.
In this case, the experimental treatment has a is 2.2% higher growth rate than the V.
philippinarum control group. This difference, however, is not significant (t=1.53, df=4,
p=0.16).
Mean height measurements of V. philippinarum, taken at intervals throughout the
summer, are displayed in Table 3. Initial average measurements ranged from 1.83-2.11
cm for individuals in V. philippinarum only (control) bags. For individuals in V.
philippinarum and Ulva spp. (treatment) bags, initial average values ranged from 1.90 to
2.02 cm. Final measurements for individuals in control bags ranged from 1.98 to 2.02 cm,
and for individuals in treatment bags from 1.92 to 2.09 cm.


 

54
 

Replicate/Date
Control 1
Control 2
Control 3
Control 4
Control 5
Mean
Treatment 1
Treatment 2
Treatment 3
Treatment 4
Treatment 5
Mean

6/27/15
1.96±0.05
2.03±0.05
2.11±0.03
2.05±0.01
1.83±0.03
2.00±0.05
1.98±0.04
1.97±0.05
1.96±0.05
1.90±0.03
2.02±0.03
1.96±0.02

7/18/15
1.96±0.04
2.14±0.05
2.05±0.04
2.02±0.01
1.97±0.05
2.01±0.04
1.98±0.04
1.88±0.03
1.92±0.04
1.91±0.03
2.02±0.04
1.94±0.03

7/31/15
1.99±0.04
2.02±0.03
1.93±0.03
1.96±0.01
1.97±0.05
1.98±0.02
2.04±0.04
1.96±0.04
1.85±0.06
1.94±0.03
1.93±0.06
1.95±0.03

8/15/15
2.10±0.03
1.98±0.04
2.01±0.01
2.02±0.01
1.76±0.04
1.98±0.06
1.92±0.05
2.09±0.04
1.95±0.05
1.86±0.03
1.88±0.03
1.94±0.04

Table 3. Average height measurements in cm coupled with standard error for each of the
five V. philippinarum only control and treatment replicates. Measurements were taken
from late June- Mid August.
The measurements for the control group indicate that only one of the five replicates
showed positive growth. Of the remainder of the control group, one showed no growth
and three showed negative growth. Of the replicates displaying no, or negative, growth
between initial and final values, intermediate values indicate evidence of positive growth
rates in two. Additionally, initial value measurements vary widely (0.38 cm). Final
measurement values range by 0.24 cm. The lowest observed mean of 1.76 cm was taken
on the final measurement day. Whereas the highest mean value was recorded on the
second day of measuring.
The treatment group mirrored the control in that one bag had an increase in growth,
one did not change, and three decreased from initial to final measurement. Again, two of
the four replicates, which either did not grow, or remained the same between initial and
final measurements, showed an increased mean size in intermediate measurements.
Additionally, the mean initial measurement valued varied by 0.12 cm. The final values
varied more widely (0.21 cm). Both the lowest and highest means were recorded on the


 

55
 

final measurement date. Measured shell heights between treatments and over time
displayed stochastic tendencies.
Using the means displayed in the above tables, % growth over time was calculated
for each replicate between the treatment and control groups. The results are displayed in
the Table 6:

Treatment

%
 Growth
 
height
-­‐2.7

Control

%
 Growth
 
height
7.0

Treatment

6.0

Control

-­‐2.4

Treatment

0.0

Control

-­‐4.6

Treatment

-­‐1.6

Control

-­‐1.3

Treatment

-­‐7.3

Control

-­‐3.9

Mean

-­‐1.1+/1.9

Mean

-­‐1.1±2.1

Table 4. Mean growth rates for each of the treatment and control replicates between June
27th and August 15th. The means of all replicates in each treatment and control were
taken for both length and height.
The values demonstrate the difference between the mean of % growth between the
treatment and control groups using the parameter of height. The mean % growth for
height between both groups is equal at -1.1%. Taking into account the standard error of
±2.0 %, it is unlikely the shell heights actually experienced negative growth.
Figure 9 demonstrates the difference between treatments on using the parameters of
length and height.


 

56
 

% Growth from Initial Measurement

10
 
8
 
6
 
Length

4
 

Height
2
 
0
 
-­‐2
 
-­‐4
 
Presence of Ulva spp.

Absence of Ulva spp.

Figure 9. The mean % growth values for height and length between V. philippinarium
treatment and control groups. Means do not appear to be significantly different in either
case.
The above figure demonstrates the relative equality of the % growth means between
the control and treatment groups for each parameter.
Statistical analysis of the mean percent growth for shell length between control and
treatment groups does not demonstrate a significant difference between groups (t=1.53,
df=4, p=0.16). Additionally, analysis shows there is no significant difference between
mean% height growth between control and treatment groups (t=0.33, df=4, p=0.98). This
analysis concludes that the presence of Ulva spp. does not have a significant effect on the
growth of V. philippinarum for this experiment.

Ulva spp. Abundance
Ulva spp. abundance varied between sampling dates. As a large storm took place
8/20/2015 which deposited deeper water Ulvoids onto the entire cultivation plot,

 

57
 

abundances were only recorded for 7/31/15 and 9/5/15.
Qualitative estimates, as achieved by drying and weighing proxy Ulvoid masses,
indicate 7/31/15 had an overall lower Ulvoid density covering (2.4-8.1 grams dry weight
[gdw]). The second collection date, 9/5/15 had slightly more dense Ulvoid attachments
(3.4-10.2 gdw). For both dates, all high designations were concentrated in the treatment
groups, whereas low and medium designations were present throughout. Figure 10 shows
the differences in frequency distribution for high, medium, and low values for the both
collection dates combined
6
5

Count

4
3

Treatment

2

Control

1
0
High

Medium
Ulvoid Density

Low

Figure 10. Frequency distribution of high, medium, and low Ulvoid density designations
between control and treatment groups. N=10.
From figure 10, it is clear that the treatment group displays a greater amount of high
Ulvoid density counts, and relatively reduced medium and low Ulvoid density counts.
Substituting the proxy dry weight values for the high, medium, and low
designations, yields no significant difference for the means of the control (ulva only) (5.0
±0.8 gdw) and treatment (5.4 ± 0.8 gdw) groups for the 7/31/15 date (t=0.17, df=9,


 

58
 

p=0.56). The control (3.4 ±1.0 gdw) and treatment (7.3 ±0.9 gdw) means are additionally
not significantly different for the 9/5/15 date (t=1.47, df=9, p=0.18).
Despite the lack of significance, it is clear that a pattern exists between control and
treatment bags. Bags containing V. philippinaurium have an observably higher Ulvoid
density than bags not containing V. philippinaurium. A larger sampling size, and longer
sampling period is necessary to determine if there is a true statistical difference between
the treatments.

Carbon to Nitrogen Ratios
Carbon to nitrogen ratios for Ulva clustered around lower values for the treatment
group containing V. philippinarum and Ulva spp.(X=8.6±0.41), and higher values for the
control group containing Ulva spp. exclusively (X=12.5±0.67). Values for the treatment
group ranged from 6.6 to 11.6, whereas values for the control group ranged from 9.6 to
16.3. The highest value for the control group was collected on September 5th (16.3), the
lowest on July 19th (9.6). For the treatment group, the highest value was collected August
15th (11.6), the lowest on July 13th (6.6).
A majority of treatment and control C/N ratios showed significant differences when
partitioned by collection date. Only two of the three dates, 7/19/15 (p=0.84) and 7/31/15
(p=0.07) did not display a significant difference between the treatment groups. The
following figure displays the differences between mean values by date.


 

59
 

18
16
14

C/N Ratio

12

Ulva only

10
8
6
4
2
0
July 13th

July 19th

July 31st

August 15th August 29th September 5th

Collection Date

Figure 11. A graphical depiction of C/N ratios for treatment (C/U) and control (U) groups
by date. Treatment groups represent an average of 3 measurements, whereas 3 control
group averages were taken July 13th. All other control values were based off of one
measurement per date.
This visualization represents the clear tendency of C/N ratios for replicates in the
treatment group to be lower than those of the control group.
The mean values are significantly different from one another. The mean C:Nulva for
all replicates in the treatment group was 8.6 ±0.41. The mean value for the Ulva spp.
treatment was 12.5 ±0.67. The results of the t-test indicate that the two values are
significantly different from each other p<0.001. These results indicate that the V.
philippinarium in the treatment group had a significant effect on the nitrogen assimilation
into the Ulva spp. tissue.
The C/N ratios for the phyto-POM ranged from 4.11 to 5.10, and did not overlap
with the C/N ratio range of Ulvoid tissues on the control or treatment bags. There was a
significant difference between the C/N ratio of Ulvoid tissue and phyto-POM (p<0.0001).


 

60
 

The C/N ratios for V. philippinarium ranged from 4.07 to 5.86 for the group grown
without Ulva and from 3.82 to 5.19 for the treatment group grown with the Ulva. Neither
control nor treatment V. philippinarium ratios overlapped with ratios of control or
treatment Ulva tissue. There was a significant difference between control and treatment
groups for the V. philippinarium ratios (p=0.038). There was a significant difference
between V. phillippinarium control group and the phyto-POM (p=0.27) and V.
phillippinarium control group and the phyto-POM (p=0.71). The difference between the
control and treatment groups does not reflect the expectation that V. phillippinarium
exposed to Ulva would have a higher C/N ratio (see Appendix 1).

Dietary Isotope Analysis
δ 13C analysis
For V. philippinarium control and treatment stomach glands, δ 13C values were
similar in range (control= -19.1 to -22.6 ‰; treatment=-18.7 to -22.0‰). Though the
range for the treatment values was slightly lower than that of the control, mean δ 13C
values between treatment and control V. philippinarium stomach glands did not show a
significant difference between groups (p=0.16).

For δ 13C values Ulva spp. overall exhibited the widest range of all test groups (-9.0
to -17.1‰). Ulva spp. tissue from the Ulva only control (-13.0 to-17.1‰, X=-14.30 ±
0.67) and the Ulva tissue from the Ulva and V. philippinarium treatment (-9.0 to -16.5‰,
X=-12.15± 0.58) showed a significant difference between control and treatment groups in
terms of mean δ 13C signature (t=-2.4, df=25, p=0.03). This finding suggests that the
presence of V. philippinarium had an unexpected enrichment effect on the δ 13C signature


 

61
 

of the associated Ulva spp.
δ 13C values for the stomach glands in the V. philippinarium only group (-19.1 to 22.6 ‰, X= -20.82±0.26) and for the stomach glands in the combination treatment group
(-18.7 to -22.0 ‰, X=-20.26±0.27) did not intersect with the range of values from any of
the primary producer groups, including phytoplankton-based particulate organic matter
(POM). δ 13C values for phytoplankton ranged from -14.9 to -17.4 ‰
Figure 12 summarizes the ranges for control and treatment Ulva spp., control and
treatment V. philippinarium groups, and the range of phytoplankton-based POM.

-5
-7
-9

δ 13C

-11

Ulva
 spp-­‐T
 

-13

Ulva
 spp-­‐C
 

-15

phyto-­‐POM
 
V.
 philippinarium
 

-17
-19
-21
-23
Species

Figure 12. This figure exhibits the mean isotopic signature with ranges for Ulva spp.
control and treatment groups and the V.philippinarium control and treatment stomach
gland (after taking into account 0.6‰ fractionation factor for the stomach gland). The
mean δ13C value for phyto-POM is also displayed. Ulva spp. groups have significantly
different δ13C signatures from one another. V.philippinarium control and treatment do not
show a significant difference. However, V.philippinarium combined control and
treatment group lies outside of the range for all primary producers.


 

62
 

δ13C mean values for V.philippinarium were statistically indistinguishable (t=-9.66,
df=37, p=0.16). Hence, Ulva spp. application had no detectable effect on
V.philippinarium dietary preference as indicated from the stable isotope analysis.
However, V. philippinarium presence had an unexpected enrichment effect on δ13C
signature of Ulva spp.. As such, Ulva spp. tissue must be divided into two classes
(treatment and control) for accurate comparison with other primary producers.
The average mean δ 13C for all Ulva spp. in the experiment (12.87±0.67‰) and
phytoplankton-based seston (15.09± 0.63‰) were found to be significantly different from
one another (t=-9.41, df=28,p=0.03). There was a significant difference between the
Ulva spp. treatment group and phytoplankton (t=2.52, df= 21, p<0.0001). However, there
was no significant difference between Ulva spp. control (X=-14.30 ±0.67‰) and
phytoplankton (t=0.62, df= 11, p=0.75). Figure 13 represents the δ13C values for
treatment and control Ulva spp. groups, as well as the δ13C value for phytoplankton.
-­‐18
 
-­‐16
 
-­‐14
 

δ13C

-­‐12
 
-­‐10
 

phyto-­‐POM
 

-­‐8
 

Ulva
 spp.-­‐C
 
Ulva
 spp.-­‐T
 

-­‐6
 
-­‐4
 
-­‐2
 
0
 

Primary Producers

Figure 13. The differences in mean δ13C values for treatment (XT= -12.15± 0.58‰) and
control (XC=-14.30 ±0.67‰) Ulva spp. groups, and phytoplankton-based POM (XPhyto=12.87±0.67‰). There is no significant difference between XC and XPhyto values. All other
groups are significantly different.

 

63
 

δ 15N analysis
V.philippinarium δ15N were enriched comparatively to all primary producer values
when adjusted with a 2.9 ‰trophic fractionation factor. Control δ15N values for
V.philippinarium ranged from 10.0 to 10.8‰. Treatment δ15N values ranged from 10.1 to
10.8. Control and treatment V.philippinarium groups did not display a significant
difference.
Primary producer values were relatively similar in terms of δ15N values. The range
of values for the Ulva spp. control group was 7.2 to 9.2‰. The treatment group ranged
from 7.6 to 9.6‰. Though the treatment group had a slightly higher range of values, Ulva
spp. δ15N values did not display any significant difference between control and treatment
groups (t=1.65, df=25, p=0.51). The same was true for means δ 15N values between
treatment (X=10.4 ±0.12‰.) and control (10.4±0.05‰.) groups of stomach glands
(p=0.78). Phytoplankton δ15N values ranged from 7.9 to 8.8‰. Mean δ15N phytoplankton
values (8.3±0.88‰) were not significantly different from mean Ulva spp. δ15N values
(t=,0.29, df=35, df=28, p=0.66).
Mixing Model
As the mean 13C signatures of both primary producer groups were significantly
different from one another, the following system of linear equations was used to deduce
the proportion of dietary source contributions to the control and treatment V.
philippinarium diets.
3a.

δ 13Cphyto-POM x Fphyto-POM + δ13CUlva x FUlva = δ 13CV. philippinarium

b.

F phyto-POM + F Ulva= 1

*where δ 13CV. philippinarium only ranged from -19.1 to -22.6 ‰; δ 13CUlva only ranged from -13.0 to17.1‰,; δ 13Cphyto-POMranged from 14.9 to -17.4 ‰ and δ 13CV. philippinarium combination= -18.7 to -22.0


 

64
 

‰; δ 13CUlva combination=--9.0 to -16.5‰; δ 13Cphyto-POM=-14.9 to -17.4 ‰ and Fphyto-POM=the fraction
of phyto-POM in the diet; FUlva=the fraction of Ulva spp. in the diet.

In attempting the mixing model for the V. philippinarium only control group, I used
the range of V. philippinarium control signatures, the range of Ulva spp. control
signatures and the range of POM signatures. For the V. philippinarium combination
treatment, I used the range of V. philippinarium combination signatures, the range of
Ulva spp. combination signatures, and the range of POM signatures. However, the
signatures of both source contributors were enriched compared to the signature of the
stomach gland. Hence, any combination of both test sources will not account for the
signature found in the stomach glands of the control and treatment groups.

Discussion

V. philippinarum growth
The results from the growth data do not reveal any significant effect of the presence
of Ulva spp. on V. philippinarium shell growth. Shell length exhibits an average increase
that is slightly higher in the presence of Ulva relative to the absence of Ulva
spp.(control). For shell height, both the treatment and control groups did not differ in
growth rate. The growth rate increased over the summer most likely due to increased
metabolic activity in the warmer months. This data was collected over a much shorter
time period than studies in the literature focusing primarily on growth rates, which
typically last from 6 months to 2 to two years (Lamb unpublished data, 2015; Suh &
Shin, 2012). Results suggest that a longer period of data collection is needed to yield a
more meaningful result.
The effect of Ulva spp on the growth rates of V. philippinarium does not compare


 

65
 

to the effect of Ulva spp. on C. gigas at the same site. When C. gigas was exposed to
dense concentrations of Ulva spp. at the Thorndyke Bay site during the summer of 2015,
growth rate was negatively impacted (Lamb, 2015). However, C. gigas growth rate did
not become negatively impacted until mid-July, and the trend was more pronounced later
in the study (August-October) (Lamb, 2015). This reinforces the necessity of extending
the measurements of V. philippinarium further into the season for a more accurate crossspecies comparison.
Despite the discrepancy between the response of C. gigas and V. philippinarium in
the field, it is possible that the summer Ulvoid blooms do not have a significant impact on
V. philippinarium growth. The study by conducted by Lamb in 2015 examines the effect
of 0, 1.5, and 3.0 kg of Ulva spp. biomass on C. gigas growth. In contrast, this study,
using V. philippinarium, depends exclusively on natural density of Ulva spp. on site to
mimic the natural conditions. This approach did not allow for the control of Ulva spp.
densities for the duration of the project. The summer of 2015 was especially warm and
dry; resulting in drastically decreased natural Ulvoid density (Joth Davis, personal
communication). Hence, the discrepancies in Ulvoid density may account for the
differences in treatment effect observed between the Lamb 2015 results and these results.

Carbon to Nitrogen Ratios
Carbon/nitrogen (C:N) ratios show a marked decrease for Ulva spp. tissue with V.
philippinarium; Values ranged from 8.6±0.41 in the presence of V. philippinarium, and
12.5±0.67 in the absence, being significantly lower in the presence of clams. One reason
for the lower C:N ratios in the presence of V. philippinarium could be that there is


 

66
 

increased nitrogen assimilation in the tissue of Ulva spp. when grown in proximity to the
clam. V. philippinarium actively produces ammonium as a metabolic byproduct (Saurel
et al, 2014), which is the most readily utilized form of nitrogen by Ulva spp. (Saurel et al,
2014; Zertuche-Gonzalez et al, 2008). Hence, it is assumed that the ammonium produced
by V. philippinarium has a fertilization effect for Ulva spp. tissue. Additionally, the
production of ammonium is positively correlated to higher temperatures due to
heightened metabolic activity (Mann & Glomb, 1978). Hence, the seasonality of the
Ulvoid blooms coincides with an assumed increase in local ammonium concentrations
near V. philippinarium beds in the warm summer months.
To test if daily mean temperature had an effect on nitrogen assimilation in Ulva
spp. tissue (hypothetically by increasing V. philippinarium metabolic activity) a
regression was run using weekly average temperature data from the National Weather
Service (NWS) for Seattle. Figure 14 demonstrates the relationship between daily
average temperature and C/N ratio for Ulvoid tissue exposed to V. philippinarium.
10

C/N Ratio

9.5

R2
 = -0.23

9
8.5
8
7.5
7
65

70

75
80
Temperature (F)

85

90

Figure 14. Linear regression using temperature as a predictive variable of C/N ratio. The
relationship between daily temperature and C/N is not significant (R2=-0.23, p=0.18).


 

67
 

This analysis indicates that weekly temperature and C/N ratio are not significantly
correlated (R2=0.-0.23, p=0.18). However, the C:N ratio decreased with increasing
temperature, as expected. Despite the weekly temperature not acting as a strong force on
nitrogen assimilation, the continued presence of V. philippinarium did have a significant
effect on overall nitrogen assimilation.
As the presence of V. philippinarium appears to have a fertilization effect for Ulva
spp., it can be assumed that V. philippinarium monocultures exacerbate local bloom
severity. In Samish Bay, V. philippinarium monocultures have been shown to
significantly increase the biomass of local Ulva spp.(Saurel et al, 2014). Though no
equivalent biomass measurements exist for Thorndyke Bay, elevated nitrogen levels in
Ulvoid tissue imply increased growth biomass production due to heightened nutrient
acquisition. This biomass is of concern for the aquaculture industry, as it makes shellfish
harvest more difficult, and has been shown to negatively impact the growth of
commercially significant C. gigas when present in high densities (Lamb, 2015).
The capacity for Ulvoid blooms to act a significant sink for nitrogen in Thorndyke
Bay has significant implications for the surrounding ecosystem. The tissue effectively
stores excess nitrogen released by V. philippinarium monocultures. However, the
degradation of this tissue releases the stored nitrogen back into the local ecosystem.
Additionally, the process of primary producer degradation consumes dissolved oxygen
(DO) in the water column, increasing harmful eutrophic conditions. The Hood Canal
experiences seasonal decrease in DO due to oxygen-poor upwelling and limited flushing
of the Hood Canal Basin (Newton et al, 2007). In the Southern reaches of the Canal, this
poor water quality can result in lethal conditions for local sea life (Newton et al, 2007).


 

68
 

Though Thorndyke Bay is not highly susceptible to low DO of this capacity, V.
philippinarium monocultures and Ulvoid blooms co-occur in the South Hood Canal.
Hence, the nitrogen from degrading Ulvoid blooms may be locally exacerbating preexisting DO issues on aquaculture farms throughout the Hood Canal. Ultimately, research
into biomass proliferation and nitrogen assimilation in areas more strongly impacted by
poor water quality is necessary to determine the extent of the problem in these areas.

Isotope Analysis
δ 13C stable isotope values for phytoplankton reflected the typical range of values
for the Hood Canal (Conway-Cranos et al, 2015), and for Ulva in the N. Puget Sound
(Howe, Simenstad,& Ogsto, 2012). As expected, the overall signature of the Ulvoid
tissue was slightly enriched compared to that of the phytoplankton-based POM. This is
due to the increased utilization of the heavier isotope of CO2 in macroscopic primary
producers (Altabet,1988).
Interestingly, Ulvoid tissue collected from bags containing V. phillipinarium were
significantly enriched when compared to Ulvoid tissue grown independently of V.
phillipinarium. Analyzed separately, tissue not exposed to V. phillipinarium was
statistically indistinguishable from pytoplankton-based seston. Though phytoplankton
and both control and treatment Ulva spp. isotopic ranges fall within values supplied by
the literature, phytoplankton-based seston values fall toward the bottom end of the range.
Conway–Cranos et al provides a range of -14.9 to -25.3 ‰ for phytoplankton-POM in
nearby Dosewallips. Phytoplankton-POM values, collected from July 13th through
September 5th, lie from -14.9 to -17.4 ‰ . It is possible that phytoplankton is naturally
relatively enriched at the Thorndyke Bay site from July to September. Alternatively, the


 

69
 

decomposition of high abundances of Ulvoid species may have resulted in enriched
POM. As Ulvoid species degrade more rapidly than other macrophytes (ZertucheGonzalez et al, 2008), it is possible that larger proportions of the enriched tissue were
present in the POM samples, further enriching the signature.
The significant difference between control and treatment Ulvoid tissue indicates
that the presence of V. phillipinarium has an enrichment effect on Ulva spp.. An
investigation into this mechanism reveals that Ulva spp. growing on V. phillipinarium
bags may be disproportionately integrating bicarbonate (HCO3 -) relative to CO2. Of the
species of dissolved inorganic carbon (DIC), HCO3 – is relatively enriched (Boutton,
1991). The rate of HCO3 – uptake in Ulvoids positively correlates with temperature and
desiccation (Axelsson, Larsson, & Ryburg, 1999). Assimilation of HCO3 – also occurs
with lower levels of dissolved oxygen (DO) (Axelsson, Larsson, & Ryburg, 1999). This
is due to an adaptive mechanism within the Ulvoid tissue, which adjusts for the reduced
concentrations of all DIC species in the absence of oxygen (Axelsson, Larsson, &
Ryburg, 1999). HCO3 – forms faster than CO2 in marine environments. As the treatment
and control bags were relatively similar in terms of temperature and desiccation status,
reduced DO, caused by V. phillipinarium respiration, may partially account for the tissue
enrichment difference between treatments.
To determine if V. philipinnarium respiration has an effect on bicarbonate uptake,
it is important to demonstrate that Ulvoid net photosynthesis is greater than the V.
philipinnarium respiration rate. In the North Adriatic clams respire at a rate of 0.014 ±
0.009% grams dry weight per day (gdw d-1) at 20 degrees C (Solidoro et al, 2000). At 18
degrees C, Ulva fenestrata experiences a max gross primary production rate of 22.63 ±

 

70
 

2.7 gdw h-1, when completely submerged in seawater and a reparation rate of 7.54±1.60
gdw h-1. In air, the gross photosynthetic rate is 15.46 ± 2.33, with a respiration rate of
8.06±0.53 gdw h-1 (Quadir, Harrison, & DeWreede, 1979). Normalizing units, it becomes
clear that Ulva fenestrata photosynthesizes at a rate 8 to 25 times the respiration rate of
V. philipinnarium. Though there is a difference of 2 degrees C between these
measurements, V. philipinnarium respires slightly slower at 18 degrees C (Solidoro et al,
2000), implying that the photosynthetic rate of Ulva fenestrata would continue to outpace
V. philipinnarium respiration at this temperature.
It is hypothesized that Ulva spp. in the Northern Hood Canal uptakes nitrogen
produced by V. philipinnarium respiration, perhaps resulting in an initial increase in
growth rate. However, as biomass continues to increase, V. philipinnarium CO2 may not
be able to meet the DIC demands of the rapidly photosynthesizing Ulvoids. Where
fertilized Ulva spp. has rapidly consumed CO2, the DIC composition would shift in favor
of HCO3 –. In these circumstances, the Ulvoids could employ their HCO3 – uptake
mechanism. Relying primarily on HCO3 – would account for a tissue enrichment in the
treatment setting. The control Ulva spp. bags may also experience a shift from CO2 to
HCO3- due to their own photosynthetic needs. However, it can be assumed that their
growth rate, and therefore, photosynthetic capacity is not as great, due to lack of initial
fertilization effect from V. philipinnarium nitrogen.
Results from this study do not show a significant difference between Ulvoid
density between bags cointaining V. philipinnarium and empty bags. However, density
data available in this study is minimal, and the trend implies that biomasses are higher in
Ulvoids exposed to V. philippinarium as compared to Ulvoids not directly exposed.

 

71
 

Ultimately, a more thorough and long-term study into relative biomasses of Ulvoids on V.
philipinnarium bags should be done in conjunction with an Ulvoid tissue relative
abundance study.
For V. phillipinarium, δ 13C values were indistinguishable between the presence
or absence of Ulva spp.. This indicates that the presence of Ulva spp. in this particular
experiment did not have a significant effect on the dietary preference of V.
phillipinarium. These results, however, may not directly translate to years displaying
more average bloom densities. This experiment was designed to mimic the status-quo
conditions of Ulvoid attachment to V. phillipinarium bags by not controlling for Ulvoid
densities on treatment bags. However, the summer drought shifted the bloom season
earlier and visibly reduced the amount of Ulva spp. attached to all V. phillipinarium
growing bags on the site. It is unclear if non-drought bloom conditions would result in a
dietary shift, due to their usual capacity to completely smother the bags, hypothetically
restricting the inflow of POM to V. phillipinarium. Preferably, this experiment would be
repeated during a non-drought year to test the effect of status quo Ulva spp. covering on
V. phillipinarium diet.
Though the effect of Ulvoid presence on V. philippinarium could not be
determined, the results illuminated interesting information about V. philippinarium diets
during seasonal drought. There was no significant difference between δ 13C signatures of
V. philippinarium between treatments. Hence, it can be assumed that the stomach gland δ
13

C signatures were an accurate representation of diet in the absence of obstruction.

Analyzing the data under this assumption revealed that the combination of detrital Ulva
spp. and phytoplankton-POM did not account for the entirety of V. philippinarium diet.

 

72
 

The δ 13C isotopic signature for V. philippinarium fell far outside of the ranges of both
primary producers, being significantly depleted.
To account for the depleted isotopic values of the stomach glands, values of salt
marsh grass, specifically Glaux maritima and Salicornia virginica, from nearby
Dosewallips and Hamma Hamma were substituted into a source proportion estimator.
Marsh grasses and upland vegetation are depleted in regards to Ulva and phytoplankton
signatures. The signatures found in the 2015 Conway-Cranos et al study showed averages
signatures of δ 13C= -28.0 ±0.6‰ and δ 13C= -27.6±0.9‰ for each of the two Hood Canal
sites. Sea marsh detritus was found to comprise 35-45% of C. gigas diet at these sites.
Upland vegetation was not found to be a significant dietary contributor to C. gigas (~2%)
(Conway-Cranos et al, 2015). As Thorndyke Bay is surrounded on both sides by sea
marsh, it can be assumed that there is a supply of marsh detritus to bivalve diets in this
region (Harrington, 2005).


 

73
 

Figure 15. Map of Thorndyke Bay, Dosewallips, and Hamma Hamma site location along
the Hood Canal. The Hamma Hamma site showed an average Salt Marsh signature of δ
13
C -28.0 ±0.6‰ and the Dosewallips site, an average signature of δ 13C -27.6±0.9‰
(Conway-Cranos et al, 2015).
As there was no significant difference between treatment and control groups for
V. philippinarium stomach gland signatures, the δ 13C signature ranges were taken from
the combined group (-22.1 to -19.1
 ‰) to assess the possibility of a third dietary
contributor. For Ulva spp., the experimentally measured total δ 13C values of -15.7 to -9.9
 
‰ were input into the mixing model. Phytoplankton δ 13C measured range values of -16.44

to -14.07
 ‰ were used. For sea grasses, the entirety of the reported values were used (28.0 to -27.0
 ‰) (Conway-Cranos et al, 2015). Table 5 shows the results from the
analysis performed by ISOSOURCE addressing the mean value of probable %
contribution of primary producer to overall dietary composition. ISOSOURCE computes
a frequency histogram of all possible solutions to the three-end member mixing model


 

74
 

using δ 13C data.
Primary Producer

Ulva spp.

Phyto-POM

Mean Estimated
Contribution

23-24%

26-32%

G. maritima and S.
virginica
44-50%

Table 5. The mean range of estimated source contributions in percentages to V.
philippinarium diet using δ 13C values for Ulva spp, Phyto-POM, and G. maritima and S.
virginica. Estimates indicate a high contribution from G. maritima and S. virginica,
followed by, Phyto-POM and Ulva spp. respectively.
The lower end of the salt marsh and phyto-POM values found in the ISOSOURCE
model are consistent with the values presented in the Conway-Cranos 2015 C. gigas
study (35-45%). The values in table 5 may over represent actual contribution of the three
primary producer categories to V. phillippinarium diet, as the full range of potential
source contributions was not included in this post-hoc analysis. Additionally, values for
Ulva spp. could not be directly compared to this study, as Ulvoids were included amongst
nine species as intertidal macrophytes (Conway-Cranos et al, 2015). Figure 16 displays
the value ranges for all separately considered primary producers and the combined
primary consumer values.


 

75
 

10
9.5
9

Control

d15N

8.5

Treatment
8

Ulva (T)

7.5

Ulva (C)
phyto-POM

7

Sea Marsh

6.5
6
-29

-24

-19

-14

-9

d13C

Figure 16. A mixing model comparing δ 13C and δ 15N ratio values of primary producers
and V. philipinarium (y-range extends to 5). Raw values for treatment and control groups
for V. philipinarium were included, though the groups did not show any significant
difference for δ 13C between treatments (p=0.16). Ulva control (Ulva C) and treatment
(Ulva T) group means, along with the range of values were included. Ulva treatment and
control groups showed significant differences between groups and hence were treated as
separate dietary contributors. The phytoplankton-based POM mean and range was also
included. The combined isotopic means and ranges ofG. maritima and S. virginica were
also included.
Figure 16 demonstrates that with the added combined signatures of G. maritima
and S. virginica, V. philippinarium values from this study fall between the ranges of food
sources. Hence, incorporating depleted sources into a model or future study is crucial for
this site.
Overall, this study was able to illuminate that Ulva spp. has no observable impact
on V. philippinarium dietary preference during years of severe drought. Additionally, this
study uncovered the need to include more dietary source components, especially local sea
marsh grasses, in subsequent analysis of bivalve diets at this site. Thus, more isotope-


 

76
 

based studies are needed to 1) determine the effect of normal density Ulvoid blooms on
V. philippinarium diets, and 2) illuminate the baseline source contributors to V.
philippinarium diet in the Thorndyke Bay region of the Northern Hood Canal.
Conclusion

Evidence for a Mutualistic Interaction
The original intent of this study was to explore whether V. philippinarium and Ulva
spp. are involved in a symbiotic relationship during the seasonal proliferation of Ulvoid
blooms. This study was part of an effort to explore the relationship between cultivated
bivalves and macroalgae on a future macroalgae cultivation site in the Northern Hood
Canal. Though it remains undetermined as to whether a symbiotic relationship exists, it is
evident that at the very least, a commensal relationship exists between V. philippinarium
and Ulva spp..
This study confirmed the capacity for V. philippinarium to increase nitrogen
assimilation in Ulva spp. tissue. This contribution of nitrogen remained previously
unquantified for these species in the Northern Hood Canal. However, increased inorganic
nutrient assimilation into macroalgae tissue in the presence of bivalves is one of the
fundamental principles behind integrated (IMTA) multitrophic aquaculture systems
(Barrington, 2009). It may not be desirable, from an ecosystem perspective, to increase
nutrient flow to Ulvoid tissue (for reasons discussed later). However, if the capacity for
V. philippinarium to contribute nitrogen to macroalgae tissue translates to a cultivatable
species, IMTA could be successful in the Northern Hood Canal. Whole bay studies in the
Northern Puget Sound show that Ulvoid tissue abundance increases near whole plots of
V. philippinarium. Hence, it is reasonable to assume that the fertilization effect is not

 

77
 

limited strictly to tissue attached to growing bags.
Various IMTA systems count detrital algae from the system as a significant
contributor to bivalve diets. These systems are often closed systems with reduced variety
of feed input (Barrington, 2009). Hence, bivalve diets in these circumstances would
mirror the availability of food sources. However, in open ecosystem settings, such as in
Thorndyke Bay, bivalves can be more selective of their food sources. V. philippinarium
does not always feed proportionally based on food availability (Suh & Shin, 2014). As
seen in this study, V. philippinarium, uninhibited by barriers, incorporates isotopically
depleted food sources. It is uncertain whether dense Ulvoid barriers would cause a
significant shift in this behavior. Macroalage cultivation systems may not cause a large
disruption of POM. Hence, their application to the V. philippinarium growing site may
not force an unfavorable dietary shift. The response of V. philippinarium to these new
source components would be a more relevant study in terms of the site-specific IMTA
configuration.
Overall, the commensal relationship illuminated by this study is evidence for the
potential success of macroalgae cultivation in Thorndyke Bay. Further investigation is
necessary to determine how various macroalgae species and densities would impact V.
philippinarium diet, and most importantly, growth. However, findings from this study
indicate that low-interference macroalgae growth does not have a significant negative
impact on V. philippinarium diet or growth. Meanwhile, cultivated V. philippinarium
significantly increase nitrogen assimilation in Ulvoid tissues. These findings support the
inclusion of macroalgae cultivation on the Northern Hood Canal.


 

78
 

Ecosystems Implications
During a time of increasing anthropogenic impacts on Washington’s coastal
ecosystems, understanding the sources and fates of nutrients is essential. Shellfish
monocultures serve to sequester particulate organic nitrogen (PON), integrating it into
their tissues, and allowing it to be removed from the local ecosystem upon harvest
(Shumway et al, 2003). However, during the summer months, when shellfish are most
metabolically active, this organic PON is more rapidly transformed into ammonium
(Mann & Glomb, 1978). Ammonium is the most readily uptaken form of nitrogen by
Ulvoids and phytoplankton (Saurel et al, 2014; Zertuche-Gonzalez et al, 2008; Dortch,
1990). This study provides evidence of increased nitrogen assimilation in Ulvoid tissue
associated with V. philippinarium. Hence, the transformative capacity of V.
philippinarium provides an easily accessable fertilizer to macroalgae blooms.
Though the fertilization effect of V. philippinarium on primary producer tissue can
be harnessed to promote macroalgae cultivation, its current relationship with Ulva spp.
has presumably negative ecosystem impacts. As demonstrated by this study, Ulvoid
tissue assimilates higher nitrogen concentrations near V. philippinarium monocultures.
This increases overall Ulvoid biomass (Saurel et al, 2014; Zertuche-Gonzalez et al, 2008)
and locally concentrates nutrients. Upon degradation of Ulva spp. tissue, these nutreints
are released back into the ecosystem as particulate organic matter. The aerobic digestion
of this POM consumes dissolved oxygen and leaves areas more susceptible to seasonal
eutrophic conditions.
Shoreline systems, which experience dense Ulvoid blooms, high volumes of
riverine nitrogen inputs, and low rates of flushing, may be most affected by nitrogen


 

79
 

outputs from V. philippinarium monocultures. These more eutrophically susceptible areas
occur widely in the Southern Hood Canal (Newton et al, 2007). Though the primary
driver of eutrophication in the Hood Canal are geochemical forcing from the open ocean,
anthropogenic effects such as industry and nutrient runoff have compounding, localized
effects. In order to help negate any negative contributions of the shellfish industry to
seasonally concentrating nutrients, Ulvoid blooms should be removed from cultivation
sites. In doing so, shellfish will have a higher positive net benefit on surrounding
shoreline ecosystems.

Confounding Factors
This research project took place in the midst of a historic drought in Washington
State. The warmer and drier than average conditions abnormally impacted the growing
season of Ulva spp.. Industry observations and Puget Sound literature indicate the normal
Ulvoid growing season lasts from June through September (Joth Davis, personal
communication, 2015; Nelson et al, 2000). However large quantities, Ulvoid species were
observed in Thordyke Bay as early as May during the study year (Joth Davis, personal
communication, May 2015). Hence, the proposed study timeline was unable to capture
the beginning of the true Ulvoid growing season. Additionally, the extreme warm, dry
temperatures, in combination with the large tide runs of July, caused severe desiccation in
Ulva spp. covering the V. philippinarum cultivation plots. The original experimental
design was based on the capacity for Ulva spp. to form strong attachments to the mesh
growing bags. However, these attachments were weakened and severed with the mass
degradation of Ulva spp. tissue from July 18th- July 31st, 2015. Though the nearshore


 

80
 

Ulva spp. tissue was significantly affected, deeper free-floating Ulva spp. continued to
proliferate observably less affected.

Figure 18. This photo was taken of the entire V. philippinarium cultivation site on July
18th, 2015. The severity of Ulvoid die off is evident by the amount of mesh growing bag
exposed. In normal conditions, bag surfaces would be entirely covered in Ulvoid
attachments.
Free-floating Ulvoid and terrestrial and marine macrophyte species posed an
obstacle to the integrity of the experimental design throughout the field data collection
process. Established Ulva spp. attachments persisted on their respective bags throughout
the experiment (barring the July 18th- July 31st, 2015 window). However, tidal deposits
of free-floating Ulvoids and macrophytes species served to contaminate the non-Ulva
spp. control bags. In attempts to prevent the reduction of treatment affect by free-floating
tidal vegetation, the control bags were thoroughly scraped each week throughout the


 

81
 

experiment. However, abundance and residence time of free-floating masses remains
unknown for the periods between visits.
Additionally, an uncharacteristically severe storm hit the Washington coast on the
data collection day of August 31, 2015. This storm redistributed large quantities of free
floating Ulvoids into the upper intertidal zone, completely covering the experimental row.
Additionally, sediment and particulate marcrophytes were redistributed into the water
column. This temporary shift in seston composition did not mirror the week proceeding
or following the storm.

Figure 19. This photo was taken August 31st following the large late-summer storm. The
photo is of the high shoreline, usually free of Ulvoid species. Clearly, large masses of
Ulva spp. were redistributed much higher onto the intertidal zone and shoreline than
normal. Photo credit: Joth Davis
To better buffer the experimental row from confounding environmental factors, and

 

82
 

increase treatment effect, a shift in the experimental structure is recommended. This
buffer could be achieved by implementing a secondary containing structure on the
outside of the V. philippinarum mesh bags. This structure would act as a barrier to
prevent the influx of free-floating Ulvoid masses on incoming tides. Additionally, the
barrier would help contain the degrading Ulva spp. within the surface area of the bag.
This method would also allow for the direct quantification of mass algal mass, and allow
the researcher to maintain Ulva spp. biomass at consistent levels throughout the
experiment. This barrier, however, would not allow for the true quantification of
particulate Ulva spp. in clam diets given environmental fluxes present in the natural
environment.

Recommendations for Future Research
This study warrants further investigation into the seasonal dynamics taking place in
Thorndyke Bay. The results from this study confirmed increased nitrogen assimilation in
Ulvoid tissue associated with V. philippinarum. Hence, it would be interesting to quantify
the associated response in Ulva spp. biomass at this site. Quantifying the abundance of
Ulvoid biomass over several seasons would give insight into how locally concentrated
nutrients, from shellfish cultivation and from degrading Ulvoids, are effecting subsequent
year’s blooms. Additionally, expanding this study to include the eutrophically-stressed
Southern Hood Canal could illuminate the seasonal contribution of shellfish cultivation to
increasing primary production, and subsequent decomposition.
After a preliminary IMTA system is introduced in the Northern Hood Canal, it
would be beneficial to quantify the effect of bivalve species on nitrogen assimilation and


 

83
 

biomass accumulation in cultivated macroalgal species. Findings from this type of study
would help the industry make a case for the expansion of IMTA, especially if macroalgae
proves an economically viable product. Additionally, this type of study would allow for
ecologists to quantify the removal of nitrogen from these ecosystems through macroalgae
harvest. Increased removal of nutrients from these systems may also aid in bolstering the
argument for IMTA systems throughout the state. Additionally, it would be interesting to
investigate the effect of macroalgae cultivation on seasonal Ulvoid blooms. It is possible
that through the repeated harvest of seasonal nutrients from the system, Ulvoid blooms
may become less severe. This would have positive implications for growers, and for the
surrounding ecosystem.
Many questions about V. philippinarum diet were posed by this study. This research
took place under the assumption that V. philippinarum were eating primarily Ulva spp.
and phyto-POM. However, it was illuminated that their diet is likely to be far more
expansive, consisting of much more depleted sources. An investigation into source
contributors to V. philippinarum diets is imperative to understanding how these
preferences respond to seasonal influences. Additionally, it would be interesting to
investigate the dietary preferences of V. philippinarum in non-drought years.
Traditioanlly, dense Ulvoid mats completely cover V. philippinarum plots. It is possible
they are pressured to switch their diets to incorporate more Ulvoid detritus during normal
years. It would also be interesting to see how these normal summer conditions impact the
growth of V. philippinarum in the covered bags.
After the introduction of the IMTA system, it would be interesting to note any
change in V. philippinarum feeding preferences. It is possible that a different species of


 

84
 

macroalgae may be more preferable to V. philippinarum. If this is the case, IMTA
systems have a higher chance at industry acceptance. Investigating the interaction of
cultivated macroalgae with C. gigas is another important study. As C. gigas is the most
profitable bivalve in Washington state (Booth, 2014), it is important to monitor its
success in regards to new industry innovations.

Integrating Findings into industry Practices
The results from this study have both short-term and long-term implications for the
shellfish industry. As Ulva spp. acts as a temporary nutrient sink, and significantly
reduces C. gigas growth when present in dense quantities (Lamb, unpublished data,
2015), there is rationale for better management practices. For these reasons, the removal
of Ulvoid blooms from shellfish cultivation sites would have positive impacts
ecologically and economically. Currently, the removal of Ulvoid species is seen by the
industry to be expensive and time consuming (Joth Davis, personal communication,
2015). A cost-benefit analysis would be necessary to determine the amount of resources
optimal to devote to the removal of Ulvoid species. It is reasonable to assume that over
time, removing Ulva spp. from growing sites will result in reduced need for removal in
subsequent years. Hence, a larger upfront investment in removing the blooms could have
lasting benefits to a given harvest sites.
The integration of macroalgae aquaculture to shellfish growing beds would, too,
have positive benefits for growers. Cultivated macroalgae would sequester nutrients in a
similar manner to Ulvoid species, without experiencing rapid decomposition rates
(Zertuche-Gonzalez, 2008). This would allow for the longer-term sequestration of
nutrients prior to harvest. Additionally, cultivating macroalgae for developing local


 

85
 

markets would yield a direct economic benefit to growers.
Overall, removing inorganic nitrogen from shellfish growing regions during the
summer months is of utmost importance to Hood Canal growers. Macroalage has the
capacity to sequester nitrogen during the summer growing season. Removing the
currently dense Ulvoid blooms from growing areas is a practice growers can adopt to
reduce the negative indirect effects that shellfish cultivation has on surrounding estuarial
ecosystems. For a longer-term solution, macroalgae cultivation can be introduced to
growing areas to more reliably sequester nitrogen throughout the entire growing season.
The harvest of these species would result in increased profits to growers, and allow for
the active removal of shellfish-based inorganic nitrogen. Regardless of tactic, growers
must look to utilizing the benefits of integrated ecosystem functions for an
environmentally conscious approach shellfish cultivation in the Hood Canal.


 

86
 

Bibliography
Anderson, D. M., Glibert, P. M., & Burkholder, J. M. (2002). Harmful algal blooms and
eutrophication: nutrient sources, composition, and consequences. Estuaries, 25(4)704726.
Altabet, M. A. (1988). Variations in nitrogen isotopic composition between sinking and
suspended particles: Implications for nitrogen cycling and particle transformation in the
open ocean. Deep Sea Research Part A. Oceanographic Research Papers, 35(4),535-554.
Barrington, K., Chopin, T., & Robinson, S. (2009). Integrated multi-trophic aquaculture
(IMTA) in marine temperate waters. Integrated mariculture: a global review. FAO
Fisheries and Aquaculture Technical Paper, 529, 7-46.
Bode, A., Alvarez-Ossorio, M. T., & Varela, M. (2006). Phytoplankton and macrophyte
contributions to littoral food webs in the Galician upwelling estimated from stable
isotopes. Marine Ecology Progress Series, 318(8).
Bolton, J. J., Robertson-Andersson, D. V., Shuuluka, D., & Kandjengo, L. (2009).
Growing Ulva (Chlorophyta) in integrated systems as a commercial crop for abalone feed
in South Africa: a SWOT analysis. Journal of Applied Phycology, 21(5), 575-583.
Booth, S.R. (2014).Crop Profile for Bivalve (Oysters, Clams, Geoduck Clams and
Mussels) Aquaculture in Washington. Retrieved from
http://www.ipmcenters.org/cropprofiles/docs/WAbivalve.pdf
Boutton, T. W. (1991). Stable carbon isotope ratios of natural materials: II. Atmospheric,
terrestrial, marine, and freshwater environments. Carbon isotope techniques, 1, 173.
Buschmann, A. H., Varela, D. A., Hernández-González, M. C., & Huovinen, P. (2008).
Opportunities and challenges for the development of an integrated seaweed-based
aquaculture activity in Chile: determining the physiological capabilities of Macrocystis
and Gracilaria as biofilters. Journal of Applied Phycology, 20(5), 571-577.
Cloern, J. E., Canuel, E. A., & Harris, D. (2002). Stable carbon and nitrogen isotope
composition of aquatic and terrestrial plants of the San Francisco Bay estuarine system.
Limnology and oceanography, 47(3), 713-729.
Christensen, J. P., Smethie, W. M., & Devol, A. H. (1987). Benthic nutrient regeneration
and denitrification on the Washington continental shelf. Deep Sea Research Part A.
Oceanographic Research Papers, 34(5), 1027-1047.
Conway-Cranos, L., Kiffney, P., Banas, N., Plummer, M., Naman, S., MacCready, P., &
Ruckelshaus, M. (2015). Stable isotopes and oceanographic modeling reveal spatial and


 

87
 

trophic connectivity among terrestrial, estuarine, and marine environments. Marine
Ecology Progress Series, 533, 15.

Dang, C., Sauriau, P. G., Savoye, N., Caill-Milly, N., Martinez, P., Millaret, C., & De
Montaudouin, X. (2009). Determination of diet in Manila clams by spatial analysis of
table isotopes. Marine Ecology Progress Series, 387, 167-177.

De Casabianca, M. L., Laugier, T., & Marinho-Soriano, E. (1997). Seasonal changes of
nutrients in water and sediment in a Mediterranean lagoon with shellfish farming activity
(Thau Lagoon, France). ICES Journal of Marine Science: Journal du Conseil, 54(5), 905916.
Department of fisheries and Oceans (1999). Stock Assessment Report. Retrieved
from:http://www.dfo-mpo.gc.ca/CSAS/CSAS/status/1999/c6-03e.pdf
Deudero, S., Cabanellas, M., Blanco, A., & Tejada, S. (2009). Stable isotope
fractionation in the digestive gland, muscle and gills tissues of the marine mussel Mytilus
galloprovincialis. Journal of Experimental Marine Biology and Ecology, 368(2), 181188.
Diaz, R. J., & Rosenberg, R. (1995). Marine benthic hypoxia: a review of its ecological
effects and the behavioural responses of benthic macrofauna. Oceanography and marine
biology. An annual review, 33, 245-03.
Dortch, Q. (1990). The interaction between ammonium and nitrate uptake in
phytoplankton. Marine ecology progress series. Oldendorf, 61(1), 183-201.
Dunton, K. H., & Schell, D. M. (1987). Dependence of consumers on macroalgal
(Laminaria solidungula) carbon in an arctic kelp community: δ13C evidencee. Marine
Biology, 93(4), 615-625.
Farquhar, G. D., Ehleringer, J. R., & Hubick, K. T. (1989). Carbon isotope discrimination
and photosynthesis. Annual review of plant biology, 40(1), 503-537.
Feely RA, Scott, DC, Cooley SA. (2009). Ocean acidification: present conditions and
future changes in a high CO2 world. Oceanography. 22(4), 36-47.
Feely RA, Sabin CL, Hernandez-Ayon JM, Ianson D, Hales B. (2008). Evidence for
upwelling of corrosive “acidified” water onto the continental shelf. Science. 320, 14901492.


 

88
 

Greengrove, C., Masura, J., Moore, S., Bill, B., Hay, L., Eldred, K., ... & Stein, J.
Alexandrium bloom ecology in Puget Sound: cyst distribution, growth and viability.
Hallegraeff, G. M. (1993). A review of harmful algal blooms and their apparent global
increase. Phycologia, 32(2), 79-99.
Hein, M., Pedersen, M. F., & Sand-Jensen, K. (1995). Size-dependent nitrogen uptake in
micro and macroalgae. Marine ecology progress series. Oldendorf, 118(1), 247-253.
Hoefs, J. (1997). Stable isotope geochemistry (Vol. 201). Berlin: Springer.
Howe, E. R., & Simenstad, C. A. (2015). Using stable isotopes to discern mechanisms of
connectivity in estuarine detritus-based food webs. Marine Ecology Progress Series, 518,
13-29.
Humphreys, J., Harris, M. R., Herbert, R. J., Farrell, P., Jensen, A., & Cragg, S. M.
(2015). Introduction, dispersal and naturalization of the clam Ruditapes philippinarum in
British estuaries, 1980–2010. Journal of the Marine Biological Association of the
United Kingdom, 1-10.
Industrial Economics, Incorporated (2014). Marine Sector analysis Report. Retrieved
from:http://msp.wa.gov/wp-content/uploads/2014/03/AquacultureSectorAnalysis.pdf
Kasai, A., Horie, H., & Sakamoto, W. (2004). Selection of food sources by Ruditapes
philippinarum and Mactra veneriformis (Bivalva: Mollusca) determined from stable
isotope analysis. Fisheries science, 70(1), 11-20.
Kidd, K. A., Schindler, D. W., Hesslein, R. H., & Muir, D. C. G. (1995). Correlation
between stable nitrogen isotope ratios and concentrations of organochlorines in biota
from a freshwater food web. Science of the Total Environment, 160, 381-390.
Khangaonkar, T., Sackmann, B., Long, W., Mohamedali, T., & Roberts, M. (2012).
Simulation of annual biogeochemical cycles of nutrient balance, phytoplankton bloom
(s), and DO in Puget Sound using an unstructured grid model. Ocean Dynamics, 62(9),
1353-1379.
Mackas, D. L., & Harrison, P. J. (1997). Nitrogenous nutrient sources and sinks in the
Juan de Fuca Strait/Strait of Georgia/Puget Sound estuarine system: assessing the
potential for eutrophication. Estuarine, Coastal and Shelf Science, 44(1), 1-21.
Miller WA, Reynolds AC, Sobrino C, Reidel GF (2009). Shellfish face uncertain future
in high CO2 world: influence of acidification on oyster larvae calcification and growth in
estuaries. PLOS1.4(5), 5661.


 

89
 

Mumford, T.F. 2007. Kelp and Eelgrass in Puget Sound. Puget Sound Nearshore
Partnership Report No. 2007-05. Published by Seattle District, U.S. Army Corps of
Engineers, Seattle, Washington.Available at www.pugetsoundnearshore.org
Mann, R., & Glomb, S. J. (1978). The effect of temperature on growth and ammonia
excretion of the Manila clam Tapes japonica. Estuarine and Coastal Marine Science,
6(3), 335-339.
Nakata, K., & Newton, J. A. (2000). Seasonal patterns and controlling factors of primary
production in Puget Sound’s Central Basin and Possession Sound. In Puget Sound Water
Quality Action Team, 2002. Proceedings of the 2001 Puget Sound Research
Conference. T. Droscher, editor. Puget Sound Water Quality Action Team. Olympia,
Washington.
Needoba, J. A., Sigman, D. M., & Harrison, P. J. (2004). The Mechanism of Isotope
Fractionation During Algal Nitrate Assimilation as Illuminated by the 15N/14N of
Intracellular Nitrate. Journal of Phycology, 40(3), 517-522.
Nelson, T. A., Haberlin, K., Nelson, A. V., Ribarich, H., Hotchkiss, R., Alstyne, K. L.
V., & Fredrickson, K. (2008). Ecological and physiological controls of species
composition in green macroalgal blooms. Ecology, 89(5), 1287-1298.
Nelson, T. A., Nelson, A. V., & Tjoelker, M. (2003). Seasonal and spatial patterns of"
green tides"(ulvoid algal blooms) and related water quality parameters in the coastal
waters of Washington State, USA. Botanica Marina, 46(3), 263-275.
Newton, J., Bassin, C., Devol, A., Kawase, M., Ruef, W., Warner, M., & Rose, R. (2007,
March). Hypoxia in Hood Canal: An overview of status and contributing factors. In
Proceedings of the 2007 Georgia Basin Puget Sound Research Conference. Puget Sound
Action Team, Olympia, Washington.
Newton, J. A., Bassin, C., Devol, A., Richey, J., Kawase, M., & Warner, M. (2011).
Hood Canal dissolved oxygen program integrated assessment and modeling report: I.
Overview and results synthesis.
Newton, J. and Van Voorhis, K., 2002, Seasonal Patterns and Controlling Factors of
Primary Production in Puget Sound’s Central Basin and Possession Sound, Washington
State Department of Ecology, Environmental Assessment Program, Publication #02-03059, Olympia, WA.
Nobre, A. M., Robertson-Andersson, D., Neori, A., & Sankar, K. (2010). Ecological–
economic assessment of aquaculture options: comparison between abalone monoculture


 

90
 

and integrated multi-trophic aquaculture of abalone and seaweeds. Aquaculture, 306(1),
116-126.
Peterson, B. J., & Heck, K. L. (1999). The potential for suspension feeding bivalves to
increase seagrass productivity. Journal of Experimental Marine Biology and Ecology,
240(1), 37-52.
Poulain, C., Lorrain, A., Mas, R., Gillikin, D. P., Dehairs, F., Robert, R., & Paulet, Y. M.
(2010). Experimental shift of diet and DIC stable carbon isotopes: Influence on shell δ 13
C values in the Manila clam Ruditapes philippinarum. Chemical Geology, 272(1), 75-82.
Quadir, A., P. J. Harrison, and R. E. DeWreede. "The effects of emergence and
submergence on the photosynthesis and respiration of marine macrophytes." Phycologia
18, no. 1 (1979): 83-88.
Raikow, D. F., & Hamilton, S. K. (2001). Bivalve diets in a midwestern US stream: a
stable isotope enrichment study. Limnology and Oceanography, 46(3), 514-522.
Resnsel,J., Bright,K., and Segreist,Z. (2011). Intergrated Shellfish Mariculture in the
Puget Sound. Retrieved from
http://www.aquamodel.net/Downloads/Puget%20Sound%20IMTA%20final%20report%
2031May2011.pdf
Robinson, S. M. C., Martin, J. D., Cooper, J. A., Lander, T. R., Reid, G. K., Powell, F., &
Griffin, R. (2011). The role of three dimensional habitats in the establishment of
integrated multi-trophic aquaculture (IMTA) systems. Bulletin of the Aquaculture
Association of Canada, 109(2), 23-29.
Ruesink, J. L., Trimble, A. C., Berry, H., Sprenger, A. G., & Dethier, M. N. (2014).
Environmental Correlates of Growth and Stable Isotopes in Intertidal Species Along an
Estuarine Fjord. Estuaries and coasts, 37(1), 149-159.
Ryabenko, E. (2013). Stable isotope methods for the study of the nitrogen cycle. Topics
in Oceanogr, 1-40.
Saurel, C., Ferreira, J. G., Cheney, D., Suhrbier, A., Dewey, B., Davis, J., & Cordell, J.
(2014). Ecosystem goods and services from Manila clam culture in Puget Sound: a
modelling analysis. Aquaculture Environment Interactions, 5, 255-270.
Sequeira, A., Ferreira, J. G., Hawkins, A. J. S., Nobre, A., Lourenço, P., Zhang, X. L., &
Nickell, T. (2008). Trade-offs between shellfish aquaculture and benthic biodiversity: a
modelling approach for sustainable management. Aquaculture, 274(2), 313-328.


 

91
 

Shumway, S. E., Davis, C., Downey, R., Karney, R., Kraeuter, J., Parsons, J., & Wikfors,
G. (2003). Shellfish aquaculture–in praise of sustainable economies and environments.
World Aquaculture, 34(4), 8-10.
Solidoro, C., Pastres, R., Melaku Canu, D., Pellizzato, M., & Rossi, R. (2000). Modelling
the growth of Tapes philippinarum in Northern Adriatic lagoons. Marine Ecology
Progress Series, 199(137-148).
Spillman, C. M., Hamilton, D. P., Hipsey, M. R., & Imberger, J. (2008). A spatially
resolved model of seasonal variations in phytoplankton and clam (Tapes philippinarum)
biomass in Barbamarco Lagoon, Italy. Estuarine, Coastal and Shelf Science, 79(2), 187203.
Suh, Y. J., & Shin, K. H. (2013). Size-related and seasonal diet of the clam (Ruditapes
pinarum), as determined using dual stable isotopes. Estuarine, Coastal and Shelf Science,
135, 94-105.
The Pacific Shellfish Institute (2013). Where we Work Washington. Retrieved
from:http://www.pacshell.org/washington.asp
Thompson, D. S.. (1995). Substrate Additive Studies for the Development of Hardshell
Clam Habitat in Waters of Puget Sound in Washington State: An Analysis of Effects on
Recruitment, Growth, and Survival of the Manila Clam, Tapes philippinarum, and on the
Species Diversity and Abundance of Existing Benthic Organisms. Estuaries, 18(1), 91–
107. Retrieved from http://www.jstor.org/stable/1352285
Toba, D., Dewey,B., & King, T. (2005). Small Scale Clam Farming for Pleasure and
Profit in Washington State. Retrieved from:
http://articles.extension.org/sites/default/files/Smallscale%20clam%20farming%20for%20pleasure%20and%20profit%20in%20Washington.
pdf
Tucker, C. S., & Hargreaves, J. A. (Eds.). (2009). Environmental best management
practices for aquaculture. John Wiley & Sons.
United States Geological Survey. An Introduction to Hood Canal. Retrieved from
http://wa.water.usgs.gov/projects/hoodcanal/data/HC.pdf on 2/23/16.
Voss, M., Baker, A., Hermann, W. B., Conley, D. J., Deutsch, B., Engel, A., ... &
Grizzetti, B. (2011). Nitrogen processes in coastal and marine ecosystems. The European
Nitrogen Assessment: Sources, Effects and Policy Perspectives, 1, 147-176.


 

92
 

Warner, M. J., Kawase, M., & Newton, J. A. (2001, February). Recent studies of the
overturning circulation in Hood Canal. In Proceedings of the 2001 Puget Sound Research
Conference, Puget Sound Action Team, Olympia, WA.
Washington Department of Fish and Wildlife (2015). Clams: manila clams. Retrieved
from:http://wdfw.wa.gov/fishing/shellfish/clams/manila_clams.html
Western Washington University. Ecology of Harmful Macroalgal Blooms. Retrieved
from: http://www.wwu.edu/algae/eblooms.htm
Winter, D. F., Banse, K., & Anderson, G. C. (1975). The dynamics of phytoplankton
blooms in puget sound a fjord in the northwestern united states. Marine Biology, 29(2),
139-176.
Zanden, M., & Rasmussen, J. B. (2001). Variation in δ15N and δ13C trophic
fractionation: implications for aquatic food web studies. Limnology and oceanography,
46(8), 2061-2066.
Zertuche-González, J. A., Camacho-Ibar, V. F., Pacheco-Ruíz, I., Cabello-Pasini, A.,
Galindo-Bect, L. A., Guzmán-Calderón, J. M., ... & Espinoza-Avalos, J. (2009). The role
of Ulva spp. as a temporary nutrient sink in a coastal lagoon with oyster cultivation and
upwelling influence. Journal of Applied Phycology, 21(6), 729-736.


 

93
 

Appendices
APPENDIX A: Carbon and Nitrogen raw data and atomic ratios
C Amount (ug)
Ulva spp.
443.77
461.22
468.47
610.92
555.17
257.63
324.52
319.53
489.41
365.03
550.33
546.62
340.22
397.43
421.13
297.28
812.12
528.27
507.28
422.16
592.63
716.63
226.03
374.07
550.5
655.97
V. philippinarium
808.27
794.50
689.42
805.05
562.21
719.17
697.96
885.58
537.46
290.70


 

N Amount (ug)

C/N mass ratio

C/N atomic ratio

ID number

57.9
54.71
38.39
53.65
48.24
24.8
24.14
39.45
49.17
37.64
51.84
40.49
36.51
44.19
34.83
36.82
103.09
67.77
36.8
30.5
42.13
63.927
15.88
23.45
39.45
34.4

7.66
8.44
12.2
11.39
11.51
10.39
13.44
8.1
9.95
9.7
10.61
13.5
9.32
8.99
12.09
8.07
7.88
7.8
13.78
13.84
14.07
11.21
14.23
15.95
13.95
19.07

6.57
7.23
10.46
9.76
9.86
8.9
11.52
8.53
8.31
9.1
11.58
7.98
7.71
10.36
6.92
6.75
6.69
11.82
11.85
12.05
12.2
9.61
12.2
13.67
11.96
16.35

1-CU-270
1-CU-05
1-CU-32
2-CU-32
2-CU-270
2-CU-02
3-CU-32
3-CU-270
3-CU-60
4-CU-05
4-CU-02
4-CU-32
5-CU-60
5-CU-270
5-CU-02
6-CU-60
6-CU-05
6-CU-32
1-U-36
1-U-28
1-U-04
2-U-04
3-U-28
4-U-36
5-U-88
6-U-227

124.65
131.96
106.11
133.64
100.41
118.94
115.13
129.47
98.64
56.57

6.48
6.02
6.49
6.02
5.6
6.05
6.06
6.84
5.45
5.14

5.56
5.16
5.57
5.16
4.8
5.19
5.2
5.86
4.67
4.41

1-C-228
1-C-213
1-C-402
1-CCU-60
1-CCU-05
1-CCU-32
2-C-228
2-C-92
2-C-402
2-CCU-32
94
 


 

568.95
702.19
573.02
607.93
720.66

104.33
120.34
105.04
106.56
118.71

5.45
5.84
5.46
5.7
6.07

4.67
5
4.68
4.89
5.2

703.23
490.04
569.25
565.37
614.81
524.94
593.35
566.94
550.99
596.60
666.73
698.21
630.16
613.15
469.47
751.06
549.36
692.86
626.12
700.07
546.77
Phyto-POM
668.87
1916.68
569.88
2568.18
375.23

131.40
93.77
103.85
104.10
121.18
107.53
119.72
102.16
123.37
99.61
130.00
124.54
127.55
120.67
93.35
117.64
115.67
144.11
115.14
136.27
109.88

5.35
5.23
5.48
5.43
5.07
4.88
4.95
5.55
4.47
5.99
5.13
5.61
4.94
5.08
5.03
6.38
4.75
4.81
5.44
5.14
4.98

4.59
4.48
4.7
4.66
4.35
4.18
4.25
4.76
3.82
5.13
4.4
4.81
4.24
4.36
4.31
5.47
4.07
4.12
4.66
4.4
4.27

2-CCU-60
2-CCU-02
3-C-99
3-C-213
3-C-92
3-CCU270
3-CCU-60
3-CCU-02
4-C-228
4-C-270
4-C-99
4-CCU-02
4-CCU-32
4-CCU-05
5-C-228
5-C-99
5-C-92
5-CCU-05
5-CCU-32
5-CCU-60
6-C-402
6-C-213
6-C-99
6-CCU-02
6-CCU-60
6-CCU-05

122.25
373.25
118.75
452.16
63.05

5.47
5.14
4.80
5.68
5.95

4.69
4.4
4.11
4.87
5.1

P-1
P-2
P-3
P-4
P-5

95
 

APPENDIX B. Stable isotope raw data
ID number
Ulva spp.
1-CU-270
1-CU-05
1-CU-32
2-CU-32
2-CU-270
2-CU-02
3-CU-32
3-CU-270
3-CU-60
4-CU-05
4-CU-02
4-CU-32
5-CU-60
5-CU-270
5-CU-02
6-CU-60
6-CU-05
6-CU-32
1-U-36
1-U-28
1-U-04
2-U-04
3-U-28
4-U-36
5-U-88
6-U-227
V. philippinarium
1-C-228
1-C-213
1-C-402
1-CCU-60
1-CCU-05
1-CCU-32
2-C-228
2-C-92
2-C-402
2-CCU-32
2-CCU-60
2-CCU-02

 

δ13C

δ15N

-16.42
-16.52
-16.62
-14.60
-10.94
-14.33
-10.90
-9.90
-10.85
-9.79
-9.22
-9.03
-12.70
-12.19
-11.58
-10.69
-10.85
-11.90
-17.08
-16.28
-12.29
-16.32
-14.57
-12.40
-13.95
-12.95

7.63
8.16
7.88
7.95
8.28
8.05
8.86
8.67
8.23
8.72
8.51
8.57
9.23
9.64
9.34
8.10
8.23
8.41
8.15
7.18
8.90
7.65
8.21
8.90
9.18
8.26

-19.50
-19.15
-19.67
-19.10
-18.79
-19.21
-19.49
-20.41
-19.32
-18.91
-19.00
-19.43

7.67
7.85
7.56
7.59
7.91
7.77
7.48
7.47
7.53
7.61
7.60
7.61
96
 

3-C-99
3-C-213
3-C-92
3-CCU-270
3-CCU-60
3-CCU-02
4-C-228
4-C-270
4-C-99
4-CCU-02
4-CCU-32
4-CCU-05
5-C-228
5-C-99
5-C-92
5-CCU-05
5-CCU-32
5-CCU-60
6-C-402
6-C-213
6-C-99
6-CCU-02
6-CCU-60
6-CCU-05
phyto-POM
P-1
P-2
P-3
P-4
P-5

-20.54
-20.62
-21.06
-19.43
-18.96
-20.56
-21.01
-20.40
-21.78
-20.60
-21.66
-20.03
-22.69
-21.64
-21.97
-20.90
-22.23
-21.28
-23.01
-21.27
-21.49
-21.97
-21.23
-21.67

7.50
7.67
7.42
7.56
7.85
7.65
7.57
7.67
7.10
7.60
7.36
7.58
7.18
7.36
7.37
7.45
7.15
7.28
7.57
7.47
7.46
7.40
7.42
7.48

-15.52
-14.22
-13.92
-14.44
-17.36

8.36
8.84
8.20
7.88
8.22

APPENDIX C: Detailed materials and methods
Field Data Collection
*All data to be collected on tide run of <0.0 ft
Materials:
• 15x Manila clam bags covered in Ulva: bags must be of 1/2 in mesh (industry
standard) and partially submerged in intertidal sediment
• 1,500 Manila clams: clams of intermediate age/size (clams=1.5 yr)
• 2x 200 mL dark plastic bottles: Bottles must be rinse thoroughly 3x with DI water.
They must then be submerged in a 1.2M acid bath for 24 hours prior to sample
collection. Bottles must be re-rinsed 3x with DI water after acid bath.
• 2x 2L bottles: bottles must be rinsed 3x with DI water and 3x with seawater

 

97
 

•

•
•
•
•

before sample collection
Phytoplankton net: rinse net with 3x with DI water before sample collection to
clean and after sample collection to wash remnant phytoplankton into collection
bottle
2x Hand Calipers: 1x electronic, 1x manual
3x gallon ziplock bags: bags must be thoroughly rinsed clean in standard water
prior to sample collection
1 large plastic tub: able to hold ~150 manila clams
20 small ziplock bags

Site Preparation:
1) Designate one row of 15 clam bags containing 1.5 year old clams
2) Remove bags from sediment and standardize 10 bags to hold 150 manila clams
3) Remove all clams from 5 bags, replace with several heavy rocks for weight (designate
bags “Ulva only” and mark with blue zip-tie)
4) Randomize bags and return to original location in the sediment
5) Remove Ulva from 5 randomized clam bags (designate bags “clam only” and mark
with pink zip-tie)
6) Remaining 5 bags will be designated “both” and marked with a yellow zip-tie
7) Attach a numerical marker to each of the 15 bags
8) Leave experiment for one week before collecting the first round of data
Data collection:
Clams:
1) Every other week measure clam growth in each of the 10 clam bags
2) Dump each bag into plastic tray to count
3) Count off 10 clams, measuring the height and width of every tenth clam with calipers
until all clams are counted
4) Record measurements and total number of clams in bag. Also record mortalities.
Remove dead clams from the bag.
5) Collect three random individuals from from each bag to put in ziplock container
6) Store individuals on ice, avoiding direct contact with ice
Ulva:
1) Collect Ulva samples (>5g) from the 10 bags weekly
2) Samples will be placed in small ziplock bags and transported on ice, avoiding direct
contact with ice
3) Weekly, note Ulva abundance on bags (Low, med, high)
4) Qualitatively match “low, med, high” coverings on 10 treatments to “low, med, high”
coverings on rows outside of experiment
5) Scrape a representative sample of each of the three categories from external bags
6) Put each sample in large ziplock bag
7) Transport on ice
ISOTOPE METHODOLOGY
UC Davis Stable Isotope Facility Protocol
Materials:


 

98
 

•

•

•
•

•
•
•
•
•
•
•
•
•
•
•

2 x plastic leader bottles: Bottles must be rinse thoroughly 3x with DI water. They
must then be submerged in a 1.2 M acid bath for 24 hours prior to sample
collection. Bottles must be re-rinsed 3x with DI water after acid bath.
20 x GF/F 47 micrometer filters: Filters must be stored in an aluminum foil pouch
and combusted at 450 degrees C in a muffler furnace for 4.5 hours. After
combustion, pouch must be stored in a dry, isolated place.
20 x (10x10) squares of aluminum foil: foil must be stored in a larger foiled
pouch and combusted with the GF/F filters
Vacuum stopper & vacuum column: The top of the stopper, as well as both
openings on the column must be wrapped in tin foil. The wrapped pieces must be
put in a 500 degree C oven for 4 hours.
Dissection tools: All dissection tools and tray must be sterilized (rinsed with 70%
ethanol) before touching the organisms.
54 x small glass petri dishes: Cover glass the openings of petri dishes in foil.
Combust in muffle furnace at 500 degrees C for 4 hours.
Metal scupula: Wash with lab soap between samples. Follow with DI water rinse.
Rinse with 70% ethanol. Repeat between samples
Small cork border: Wash with lab soap between samples. Follow with DI water
rinse. Rinse with 70% ethanol. Repeat between samples
2x forceps: Wash with lab soap between samples. Follow with DI water rinse.
Rinse with 70% ethanol
Mortar and pestle: Wash with lab soap between samples. Follow with DI water
rinse. Rinse with 70% ethanol Repeat between samples.
150x 5x8 mm Tin Capsules: Combust at 500 degrees C for 4 hours inside a 200
mL glass beaker covered in foil.
2x 96-well tray: Assign unique name to trays. Group samples of similar material
together.
1x 48 well tray
crushing rod
cup holder

V. Phillippinarium
Freezer Prep:
1) In the lab, put samples from each treatment into separate labeled 20 oz tupperware
2) Samples must be completely submerged in filtered seawater for at least 24 hours to
allow for the cleansing of gut contents
3) The 10 treatments will require 4 L of seawater which will be filtered through a course
mesh
4) Using nitrile gloves, organisms must be moved to a sterilized dissecting tray
5) A sterilized dissecting tool must be inserted into the partially open clam under water
and the abductor muscles carefully cut with sterilized dissecting scalpel
6) The inside of the organism must then be rinsed with DI water
7) After the gills are moved aside by a sterilized dissecting pin, the visceral mass must be
located
8) Using sterilized dissection scissors, the membrane of the visceral mass must be
carefully cut to expose the stomach gland

 

99
 

9) The stomach gland then must be properly removed using sterilized tweezers
10) The 3 stomach glands associated with a given treatment must then be transferred in to
a properly labeled plastic bag to be frozen in a -20 degree C freezer until further
processing
11) The dissecting tray must be rinsed with DI water and 70% ethanol between each
separate dissection
Shipping Prep:
1) Remove samples from freezer and allow to de-thaw until sample separates from bag
2) Dry each sample in separate, labeled glass petri dish at 60 degrees C for 24 hrs
3) Pulverize each sample separately in mortar and pestle
4) Use scupula to weigh 1.25 mg of dry sample from mortar into Sn capsule
5) Secure material inside of capsule
6) Organize capsules into 96-well tray leaving no empty wells between samples
8) Secure small samples by placing an index card (cut to size) over wells before securing
lid
Ulva spp.
Freezer Prep:
Approximately >5 grams of ulva should be collected randomly from the tops of the ten
treatments and put in fresh, labeled plastic bags
Additionally, high, medium, and low coverage clam bags must be identified to
qualitatively reflect the amount of ulva on treatment bags in the same growing area
All the ulva must be removed and stored in separately labeled plastic bags from these
clam bags once every two weeks
All ulva samples contained in the plastic bags must be transported back to the lab on ice
Samples for isotopic analysis must be removed using nitrile gloves and rinsed with DI
water
Isotope samples must then be placed into fresh, properly labeled miniature plastic bags to
be frozen in a -20 degree C freezer until further processing
Qualitative Ulva abundance samples must be removed from bags and rinsed 3x with
water
After the final rinse, all excess water must be removed from samples by manually
compressing tissue until no runoff is observed
These 3 samples must then be placed on foil and dried at 60 degrees C for 24 hours
before weighing.
Shipping Prep:
1) Remove samples from freezer and allow to de-thaw until sample separates from bag
2) Dry each sample in separate, labeled glass petri dish at 60 degrees C for 24 hrs
3) ulverize each sample separately in mortar and pestle
4) Use scupula to weigh 2.0 mg of dry sample from mortar into Sn capsule
5) Secure material inside of capsule
6) Organize capsules into 96-well tray leaving no empty wells between samples
8) Secure small samples by placing an index card over wells before securing lid


 

100
 

Phytoplankton
Freezer Prep:
1) Sample must be collected using a 20 micrometer plankton net to capture the water.
2) The water must be drained from the bottom of the net to fill 2 dark plastic liter bottles
3) The bottle must be stored on ice during transport to the lab
4) To prep for vacuum filtration of the water sample, all pre-combusted glassware must
be thoroughly cleaned with ethanol
5)The apparatus will include a cylinder clamped on top of GF/F filter which rests on a
stopper platform.
6) The stopper will seal a 1000 mL flask
7) The first flask will be connected to a second waste trap 1000 mL flask by a rubber
hose. The rubber hose will be attached atop a second dual valve stopper. This stopper will
also be connected to a vacuum by a hose
8) The vacuum must be turned on slowly before pouring the contents of the liter bottle
through the filer
9) After the liquid has passed through the filter, the column must be unclamped, and the
G/F filter removed by sterilized tweezers and placed into a pre-combusted 10x10 cm
sheet of aluminum foil.
10) Fold the aluminum foil into a pouch around the filter using sterilized forceps
11) Place pouch in a 60 degree C drying oven for 24 hours
12) After contents have dried, fold filter into quarters using sterilized forceps
13) Re-wrap pouch and place in medium-sized desiccator until analysis
Shipping Prep:
1) Using sterilized forceps remove filter from foil pouch
2) Use sterilized hole punch to remove a piece of filter
3) Place circular piece into tin capsule
4) Secure material inside of capsule
5) Organize capsules into 96-well tray leaving no empty wells between samples
6) Secure small samples by placing an index card over wells before securing lid


 

101
 

Item sets

The Evergreen State College Masters Theses: Environmental Studies

Media

Sahli_NMESthesis2016.pdf