The Influence of Eastern Oyster (Crassostrea Virginica) Reef Restoration on Nitrogen Cycling in a Eutrophic Estuary

Loyola University Chicago Loyola eCommons

Master's Theses Theses and Dissertations

2015

The Influence of Eastern Oyster (Crassostrea virginica) Reef Restoration on Nitrogen Cycling in a Eutrophic Estuary.

Michael Hassett Loyola University Chicago

Follow this and additional works at: https://ecommons.luc.edu/luc_theses

Part of the Water Resource Management Commons

Recommended Citation Hassett, Michael, "The Influence of Eastern Oyster (Crassostrea virginica) Reef Restoration on Nitrogen Cycling in a Eutrophic Estuary." (2015). Master's Theses. 2785. https://ecommons.luc.edu/luc_theses/2785

This Thesis is brought to you for free and open access by the Theses and Dissertations at Loyola eCommons. It has been accepted for inclusion in Master's Theses by an authorized administrator of Loyola eCommons. For more information, please contact [email protected].

LOYOLA UNIVERSITY CHICAGO

THE INFLUENCE OF EASTERN OYSTER (CRASSOSTREA VIRGINICA)

REEF RESTORATION ON NITROGEN CYCLING IN A EUTROPHIC ESTUARY

A THESIS SUBIMTTED TO

THE FACULTY OF THE GRADUATE SCHOOL

IN CANDIDACY FOR THE DEGREE OF

MASTER OF SCIENCE

PROGRAM IN BIOLOGY

MICHAEL CHRISTOPHER HASSETT

CHICAGO, ILLINOIS

MAY 2015

ACKNOWLEDGMENTS

This research was funded by National Science Foundation grant DEB 0918152 to

Timothy Hoellein and Chester Zarnoch, and by a Hudson River Foundation grant to

Timothy Hoellein, Chester Zarnoch, Denise Bruesewitz, Brett Branco, and Wayne

Gardner. Thank you to my thesis adviser Timothy Hoellein and committee members

Martin Berg, John Kelly, and Christopher Peterson for their comments on this thesis.

Thanks also to Steven Polaskey, Matthew Girard, Melaney Dunne, the Hoellein Lab at

Loyola University Chicago, and Christopher Calderaro for their assistance in the laboratory. Thank you to Joseph Schluep, Allison Mass Fitzgerald, Doris Law, and

Samantha Lindemann for their assistance in this research. Finally, thank you to my family and friends for encouraging me throughout the thesis writing process.

iii

To grandpa, for reminding me to believe in myself.

TABLE OF CONTENTS

ACKNOWLEDGMENTS ...... iii

LIST OF TABLES ...... vii

LIST OF FIGURES ...... ix

ABSTRACT ...... xi

CHAPTER I: INTRODUCTION ...... 1 Worldwide oyster reef loss ...... 1 Oyster ecosystem services ...... 3 Nitrogen cycle and denitrification ...... 4 Oysters and sediment N cycling ...... 7 Oyster reef restoration in NYC ...... 9 Thesis research ...... 10

CHAPTER II: OYSTER REEF RESTORATION AND DENITRIFICATION PATHWAYS IN AN URBANIZED EUTROPHIC ESTUARY ...... 13 Introduction ...... 13 Changes in oyster reefs and their ecosystem services ...... 13 Oysters and sediment N cycling ...... 14 Oyster reef restoration in NYC ...... 17 Objectives ...... 17 Methods...... 18 Study site and reef construction ...... 18 Collection of sediment cores ...... 19 Flow-through measurements of nutrient and gas fluxes ...... 20 Sediment Organic Matter ...... 22 Dissolved solute and gas chemistry ...... 22 Flux calculations ...... 24 Statistical analysis ...... 25 Results ...... 26 Nutrient and gas fluxes in reef-adjacent and reef-distal cores ...... 26 Sediment organic matter in reef-adjacent and reef-distal cores ...... 28 Nutrient and gas fluxes in spat-on-shell and shell-only cores ...... 28 Discussion ...... 39 Oyster reef proximity and denitrification ...... 39 Spat-on-shell and N cycling ...... 44 Comparison of SOD and N-fixation rates to literature values ...... 46 Expectations for reef restoration on sediment biogeochemistry in urban habitats ... 47

CHAPTER III: THE INTERACTION OF OYSTER PRESENCE AND EUTROPHICATION IN REEF AQUARIA MICROCOSMS ...... 49 Introduction ...... 49 Changes in oyster reefs and their ecosystem services ...... 49 Oysters and sediment N cycling ...... 50 Oyster reef restoration in NYC ...... 52 Thesis research ...... 53 Methods...... 55 Aquarium experiment with oyster and eutrophication treatments ...... 55 Collection of sediment caps ...... 57 Flow-through measurements of nutrient and gas fluxes ...... 58 Sediment and oyster organic matter ...... 59 Dissolved solute and gas chemistry ...... 60 Flux calculations ...... 61 Statistical analysis ...... 62 Results ...... 62 Physicochemical measurements ...... 62 Nutrient and gas fluxes ...... 63 Sediment organic matter and oyster tissue biomass ...... 64 Discussion ...... 73 Eutrophic conditions increased denitrification ...... 73 Oysters increased direct denitrification in eutrophic and oligotrophic conditions ... 74 Oysters had little effect on total denitrification, SOD, and nutrient fluxes ...... 76 Comparison of results to field study ...... 78 Expectations for reef restoration ...... 79

LITERATURE CITED ...... 80

VITA ...... 91

LIST OF TABLES

Table 1. Water column physicochemical properties on each sampling date at the Soundview Park oyster reef ...... 31

Table 2. Mean (±SE) for nutrient fluxes in sediment cores taken adjacent to (near) and 10-15 m away (far) from the Soundview Park oyster reef ...... 31

Table 4. Mean (±SE) of nitrogen regeneration in July 2011 in cores taken at the Soundview Park oyster reef (near = adjacent to oyster reef, far = 10-15 m away from reef) ...... 33

Table 5. Mean (±SE) for nutrient fluxes in sediment cores with spat-on-shell and shell- only substrates from the Soundview Park oyster reef ...... 35

Table 6. Mean (±SE) for sediment oxygen demand (SOD) and nitrogen gas (N2) fluxes in sediment cores with spat-on-shell and shell-only substrates ...... 36

Table 7. Mean (±SE) of nitrogen regeneration in July 2011 in cores taken at the Soundview Park oyster reef with spat-on-shell and shell-only substrates...... 38

Table 8. Mean (±SE) water column physicochemical values for oligotrophic and eutrophic aquaria in the laboratory experiment. Field measurements are the average water column nutrient concentrations in seasonal samples from the East River, New York City ...... 65

Table 9. Mean (±SE) nutrient fluxes in sediment cores taken from oligotrophic and eutrophic aquaria, with and without oysters ...... 66

Table 10. Mean (±SE) sediment oxygen demand (SOD) and nitrogen gas (N2) fluxes in sediment cores taken from oligotrophic and eutrophic aquaria with and without oysters ...... 67

vii

Table 11. Mean (±SE) oyster soft tissue ash-free dry mass (AFDM), aquaria sediment % organic matter (OM), aquaria sediment % carbon, and % nitrogen by trophic state and oyster presence ...... 70

viii

LIST OF FIGURES

Figure 1. Simplified diagram of the nitrogen (N) cycle modified from Herbert (1999) .... 5

Figure 2. Conceptual diagram of predictions in oyster reef sediment core experiment ... 10

Figure 3. Mean (±SE) sediment oxygen demand (SOD) by season and reef proximity (near = adjacent to reef, far = 10-15 m away from reef) ...... 33

Figure 4. Mean (±SE) denitrification (A) and potential denitrification (B) by season and reef proximity (near = adjacent to reef, far = 10-15 m away from reef) ...... 34

Figure 5. Mean (±SE) coupled denitrification (A) and direct denitrification (B) by season and reef proximity (near = adjacent to reef, far = 10-15 m away from reef) ...... 34

Figure 6. Mean (±SE) organic matter by season and reef proximity (near = adjacent to reef, far = 10-15 m away from reef) ...... 35

Figure 7. Mean (±SE) sediment oxygen demand (SOD) for spat-on-shell (spat) and shell- only (shell) reef substrate ...... 37

Figure 8. Mean (±SE) denitrification (A) and potential denitrification (B) in spat-on-shell (spat) and shell-only (shell) cores ...... 37

Figure 9. Mean (±SE) coupled denitrification (A) and direct denitrification (B) in spat- on-shell (spat) and shell-only (shell) cores ...... 38

Figure 10. Mean (±SE) sediment oxygen demand (SOD) by trophic state and oyster presence. 68Figure 11. Mean (±SE) denitrification (A) and potential denitrification (B) by trophic state and oyster presence ...... 68

Figure 11. Mean (±SE) denitrification (A) and potential denitrification (B) by trophic state and oyster presence ...... 68

Figure 12. Mean (±SE) coupled denitrification (A) and direct denitrification (B) by trophic state and oyster presence ...... 69

Figure 13. Mean (±SE) organic matter from oligotrophic and eutrophic experimental aquaria ...... 71

Figure 14. Mean (±SE) sediment % carbon by mass (A) and sediment % nitrogen by mass (B) of artificial reef sediment by trophic state and oyster presence ...... 71

Figure 15. Mean (±SE) organic matter of oyster visceral mass from oligotrophic and eutrophic aquaria ...... 72

ABSTRACT

Oysters (Crassostrea virginica) are functionally extinct in the urbanized Hudson-

Raritan estuary (HRE) in New York City, however, oyster reef restoration is promoted to mitigate nitrogen (N) pollution via oysters’ filtration and excretion. In order to determine the effect of a recently restored oyster reef on sediment denitrification, I seasonally took

12 sediment cores (45 cm2) adjacent to and 10 m away from a recently constructed reef in the HRE. Cores were incubated in flow-through chambers with site water containing (1) no amendments, (2) 15N-ammonium, or (3) 15N-nitrate, from which I calculated coupled nitrification-denitrification and direct denitrification as isotope-enriched N2. Coupled denitrification was minimal at all sites, whereas direct denitrification was elevated near the reef, suggesting organic matter in oyster waste stimulated direct denitrification of water column nitrate, however, high variability in field samples precluded statistical significance. To further investigate these effects, I designed a laboratory study to test how oysters influence direct and coupled denitrification in oligotrophic and eutrophic sediments. I created a 2x2 aquaria testing matrix with oyster presence and trophic state as factors and repeated the isotope treatments and sediment core methods used in the field study. Oyster presence significantly increased direct denitrification, but caused only small changes in coupled denitrification, suggesting that oyster reef restoration may be useful for removing nitrate from coastal ecosystems.

CHAPTER I

INTRODUCTION

Worldwide oyster reef loss

Oyster reef loss is a global phenomenon. An estimated 85% of oyster reefs have been lost worldwide, with some areas becoming ‘functionally extinct’ and unable to sustain an oyster fishery or sustain reef-related ecosystem services (Beck et al. 2011).

Reef losses are driven by many factors, including overharvesting and pollution

(Rothschild 1994, MacKenzie et al. 1997, Halpern et al. 2008). In urban areas, oyster reef loss is exacerbated by eutrophication, siltation of oyster reefs, and disease (Jackson et al.

2001, MacKenzie 1996).

Oyster reefs on the Atlantic coast of North America provided a valuable fishery resource in the past (NOAA 1997). Oyster harvests in the US peaked between 1880 and

1910, representing a removal of 7.26 x 107 kg of oyster meat annually (Coen and

Luckenbach 2000, MacKenzie 1996). Over-harvesting was one factor contributing to the collapse of the wild Atlantic oyster fishery in the 1930’s (Jackson et al. 2001). In 2012, oyster harvest from the middle Atlantic states totaled 8.62 x 105 kg, or less than 6% of the national oyster harvest for that year (Lowther 2013).

A major influence of urban land-use on organisms inhabiting coastal ecosystems, such as oysters, is a change in hydrology. Impervious surfaces and combined sewers alter the timing and delivery of storm water to the shallow marine environment. The ‘flashier’

1 2 hydrology typical of urban environments means increased magnitude and frequency of floods, which increase erosion and sediment transport (Walsh et al. 2005). In addition, when flooding in urban environments overwhelms capacity of combined sewer systems, untreated sewage and street runoff enter adjacent aquatic habitats (Newbold 2006).

Overflow events and normal operation of wastewater treatment plants enrich the benthic sediments and water column with organic carbon and nutrients, thereby generating conditions typical of eutrophic systems (Kennish 2002, Nixon 1995).

Organisms inhabiting urban coastal ecosystems are affected by periodic dredging of shipping channels required to sustain port commerce. Channel maintenance can damage oyster reefs’ health and reduce recruitment of juveniles. In addition to removing sediment from shipping channels, dredging is used to provide sand and sediment to fill in coastal landscapes (Gornitz et al. 2001, Jackson et al. 2001). Oyster larvae require solid substrates to attach their byssal threads, where they settle to metamorphose into adults

(Carriker 1990). Reefs are sustained by positive feedback, where shells of previous generations provide an ideal habitat for settling (Gutierrez et al. 2003). Live C. virginica release chemicals that cause free-swimming larvae to exhibit settling behavior (Coon et al. 1985, Zimmer-Faust and Tamburri 1994). Siltation and dredging disrupt the physical and biological requirements for natural recruitment. Oysters may survive temporary burial by siltation, but not long-term burial (Baker and Mann 1992).

Parasites also contribute to oyster reef loss. The protist parasites ‘MSX’

(Multinucleated Sphere X, Haplosporidium nelsoni, Minchinia nelsoni) and ‘dermo’

(Perkinsus marinus) were discovered in the 1950’s (MacKenzie 1996, Rothschild 1994).

3 MSX was introduced from Asia and was first documented in Delaware Bay, where it spread along the eastern United States (Carnegie and Burreson 2011). Dermo was first discovered in the Gulf of Mexico and spread along the eastern United States by the

1950’s (Ewart and Ford 1993). MSX thrives in systems with low salinity (i.e., 15 ppt)

(Carnegie and Burreson 2011, Ewart and Ford 1993) and dermo is common in warm water (i.e., >25°C) at salinities ≥8-12 ppt (Ewart and Ford 1993). Methods to eradicate the parasites without negatively affecting oyster hosts have yet to be developed, although resistance to MSX may be inherited in oysters (Carnegie and Burreson 2011, Barber

1999).

Oyster ecosystem services

Intact oyster reefs provide many ecosystem services, including habitat creation and filtration. Oyster reefs stabilize benthic surfaces of intertidal habitats, which is beneficial for other animals. Oyster settling and reef-building creates complex three- dimensional benthic structures (Coen et al. 2007, Gutiérrez et al. 2003). These porous structures provide habitat for mobile invertebrates and vertebrates (e.g., benthic fishes), and increase surface area for epibiotic organism attachment (e.g., benthic algae, bacteria)

(Nestlerode et al. 2007, Gutiérrez et al. 2003). Reefs may provide refugia or nursery habitats for other organisms, which can increase overall species diversity (Coen et al.

2007). The process of reef creation stabilizes benthic and shoreline sediments, which could help reduce the effects of rising sea level and shoreline erosion (Coen et al. 2007).

Finally, filter-feeding organisms, such as oysters, can increase water clarity, which can

4 then increase property and recreational value of local environments (Grabowski and

Peterson 2007, Coen et al. 2007).

Similar to other bivalve suspension feeders (Bruesewitz et al. 2006), oysters increase water clarity by filtering food from the water column and delivering carbon (C) and nitrogen (N) in biodeposits (i.e., feces and pseudofeces) to benthic sediments, decreasing suspended solids, turbidity, and water column phytoplankton, which can enhance removal of N from the aquatic ecosystem via denitrification (see below) (Newell

2004, Hoellein et al. 2015). Filtered items are sorted and selectively digested (Haven and

Morales-Alamo 1966, Newell and Jordan 1983). Feces, which are materials that have been swallowed, digested, and defecated, are deposited as mucus-bound pellets of digested food items. Pseudofeces are loosely bound mucus strings containing rejected or excess items filtered by the oyster (Newell and Langdon 1996). Biodeposits are up to 8 times larger by volume than naturally settling sediments and contain C and N, which can stimulate biogeochemical processes (Lund 1957, Hoellein et al. 2015).

Nitrogen cycle and denitrification

Nitrogen is usually a limiting nutrient in coastal ecosystems and high N concentrations in urban environments (Vitousek and Howarth 1991) can lead to harmful algal blooms, hypoxic “dead zones”, and reduced species richness (NRC 2000, Howarth et al. 2006, Diaz and Rosenberg 2008). While some N enters aquatic ecosystems via natural processes, anthropogenic sources of N pollution have greatly increased N concentrations over the past 200 years and surpassed natural fixation in the late 20th century (Vitousek et al. 1997, Galloway and Cowling 2002). Anthropogenic N-loading

5 includes point and non-point sources from agriculture, urbanization, and industry

(Carpenter et al. 1998, Carmago and Alonso 2006). In addition to anthropogenic sources, naturally existing dinitrogen gas (N2) in the atmosphere can enter ecosystems via nitrogen-fixing bacteria (Vitousek et al. 2002, Gardner et al. 2006).

Nitrogen Sources and Transformations

Figure 1. Simplified diagram of the nitrogen (N) cycle modified from Herbert (1999). Solid lines indicate N transformations in anaerobic environments. Dotted lines indicate N transformations in aerobic environments.

Through assimilation of inorganic N or N-fixation, organisms incorporate N into their tissues, where it can be incorporated into food webs and recycled following death

+ and excretion (Figure 1). Organic N is typically converted first into ammonium (NH4 ).

+ + For example, live organisms excrete NH4 or organic N (which is mineralized to NH4 )

+ and organic matter from dead tissue is mineralized to NH4 by decomposers. Ammonium is a highly reduced form of N, and is transformed via a number of microbially mediated

6 + - processes. Nitrification by aerobic bacteria converts NH4 into nitrate (NO3 ). Anaerobic

+ - ammonia oxidation (anammox) converts NH4 and nitrite (NO2 ) into N2 (Strous et al.

+ 1999, Thamdrup and Dalsgaard 2002). Finally, NH4 can be re-assimilated directly by microbes and plants to be used in construction of organic N compounds.

+ - Like NH4 , NO3 is a common inorganic N species in aquatic habitats, can be assimilated directly by plants and microbes, and has several fates through dissimilatory

- + transformations. For example, dissimilatory NO3 reduction to NH4 (DNRA) by

- + anaerobic bacteria reverts NO3 into NH4 , creating a feedback loop that retains N in an environment and enhances eutrophication (An and Gardner 2002, Giblin et al. 2013). In

- contrast, denitrification is the aerobic reduction of NO3 to N2 gas through a series of microbially-mediated dissimilatory transformations (with creation and reduction of nitrous oxide as an intermediate step). Denitrification results in the removal of biologically reactive N from aquatic habitats, so factors that control denitrification are a focus of research in eutrophic ecosystems. I note that anammox also represents loss of biologically available N, but does not produce as much N2 as denitrification in coastal ecosystems (Thamdrup and Dalsgaard 2002, Devol 2003, Francis et al. 2007).

- Denitrification is controlled by three primary factors: (1) NO3 availability, (2) favorable redox conditions for denitrifying microbes, and (3) organic carbon availability

(Herbert 1999, Philippot 2002). Most microbes do not have all enzymes required for full

- denitrification, and so complete reduction of NO3 to N2 may require a consortium of closely associated microbes with complementary enzymes (Wallenstein et al. 2006). In addition, denitrification that occurs through the coupling of nitrification and

7 denitrification, described as ‘coupled denitrification’, can be the dominant pathway for

N2 production in continental shelf sediments (Heiss et al. 2012 and sources therein).

Coupled denitrification requires adjacent anoxic-oxic microsites, which support close associations of organisms and chemical substrates for each process. Denitrification that

- occurs independently of nitrification (i.e., using NO3 in the water column or sediment) is

‘direct denitrification’ and is typical in well-mixed coastal habitats in urban environments

(Hoellein and Zarnoch 2014). Enhancing N delivery to sediments and the removal of N from ecosystems via denitrification may mitigate effects of cultural eutrophication.

Confirming these effects requires careful measurement of separate denitrification pathways (i.e., coupled vs. direct) and environmental controls on N cycling rates.

Oysters and sediment N cycling

Restoration of native oyster reefs in coastal ecosystems has been proposed as a method to promote denitrification. As previously mentioned, oysters are filter feeders and contribute to improved water quality by removing phytoplankton from the water column, and redistribute nutrients and carbon from their food to the benthos as biodeposits, to be processed by sediment microbes (Newell 1988, Pomeroy et al. 2006). Oyster feces are tightly packed with strings of mucus up to several millimeters long while pseudofeces are less tightly bound and may break apart during excretion (Newell et al. 2005). Both types of biodeposits sink up to 40% faster than unaggregated organic matter (Newell et al.

2005). Oyster filtration rate is highly variable and dependent on oyster size, seston quality, and temperature (Riisgard 1988, Newell and Langdon 1996, Cranford et al.

2011). The nutritional content of oyster biodeposits varies with seston quality (Newell

8 2004) and with biodeposit type. Hoellein et al. (2015) found C. virginica reefs with higher quality seston (5-28 um diameter) correlated with significantly lower C:N ratios

(i.e. higher quality organic matter) in biodeposits compared to reefs with lower quality seston.

Understanding the conditions that control oyster-mediated denitrification is important for planning reef conservation and restoration (Hoellein et al. 2015). Previous research has produced equivocal results, where oysters increase denitrification rates in some cases, but have little effect in others. Several studies have measured denitrification in flow-through cores containing oyster reef sediments compared to sediment further from reefs. For example, Hoellein et al. (2015) showed denitrification in oyster reef- adjacent sediments was higher than denitrification in reef-distal sediments at one reef in

New Hampshire, but found no effect at another reef. Comparisons of restored oyster reefs to oyster-free control sites at a reef in Chesapeake Bay showed significantly higher denitrification in the reefs (Kellogg et al. 2013). Finally, experimental enclosures of oysters at increasing densities increased sediment organic matter, but had no effect on denitrification potential (Hoellein and Zarnoch 2014) or denitrification gene abundance

(S. Lindeman, in review) in Jamaica Bay, NYC. In summary, oyster reefs may enhance denitrification, but rates vary among studies and sites (Kellogg et al. 2014). Research has begun to account for the biotic and abiotic factors that could explain oysters’ effect on denitrification, but the effect of oyster reefs on denitrification pathways in urban ecosystems has not previously been measured.

9 Oyster reef restoration in NYC

Prior to 1900, oysters were common throughout the Atlantic coast of the United

States, including New York Harbor (Levinton et al. 2013, Zu Ermengassen 2012,

Mackenzie et al. 1997). Many oyster reefs in the region were unsustainably harvested and the New York and New Jersey fishery was depleted in the 1920’s (Franz 1982). During peak harvest years, for example, the state of New York harvested more than 5.51 x 107 kg of oysters in 1910 alone (Lyles 1969). Other contributions to oyster reef loss in the area include the creation of shipping passages and dredging (Mackenzie et al. 1997). Sewage and wastewater in NYC waterways decreased water quality as the population grew.

Although wastewater treatment infrastructure has improved over the last 50 years, NYC waterways receive more N that any other estuary in the world, an average of 290 g N m-2 y-1 (Howarth et al. 2006).

Eastern oyster (Crassostrea virginica) reef restoration in NYC is well supported by local and state governmental agencies. Reef restoration is ongoing at several locations including the Hudson River, Jamaica Bay, and East River. One primary objective for restoration is to improve water quality via oyster suspension feeding and enhanced denitrification (Cornwell et al. 1999, Newell et al. 2005, Coen et al. 2007, Grizzle et al.

2013). Although recent evidence suggests that oysters can influence sediment carbon and may affect denitrification potential in the region (Hoellein and Zarnoch 2014), but more studies on the environmental controls and pathways for denitrification in oyster reefs in urban ecosystems are needed.

10 Thesis research

In this thesis, I examined the effects of oyster reefs on N biogeochemistry in a highly urbanized coastal ecosystem. In Chapter 2, I evaluate the effect of a recently restored oyster reef in New York City on denitrification pathways. This field experiment compared reef-adjacent and reef-distal sediment cores and fluxes of N solutes and gasses at the sediment-water interface. In the lab, I installed cores in a continuous-flow system where inflow water was amended with 15N isotopes to create three treatments: control,

15 + 15 - NH4 , and NO3 . The isotopes allowed me to follow the chemical pathways of N2 production (i.e., coupled vs. direct denitrification). I predicted that sediment in reef- adjacent cores would have higher concentrations of organic matter and nitrogenous waste from oyster biodeposits, with higher rates of total denitrification (Figure 2). I expected

Figure 2. Conceptual diagram of predictions in oyster reef sediment core experiment. N transformations are indicated in italics. Dots in water column represent phytoplankton, ovals represent oysters, and dots in sediment represent organic matter. Line thickness indicates relative magnitude of N transformations.

11 reef-adjacent cores would show higher coupled nitrification-denitrification than direct

+ denitrification, stimulated by mineralization of NH4 from oyster biodeposits and the ability of the 3-dimensional structure of the oyster reef to maintain oxic-anoxic microsites

(Kellogg et al. 2013). In contrast, I predicted the pathway for N2 production in reef-distal cores would be more evenly divided between coupled and direct denitrification.

In Chapter 3, I compare oyster reef effects on sediment N biogeochemistry in aquaria by manipulating sediment C and water column nutrient concentrations, mimicking oligotrophic and eutrophic conditions. Using artificial seawater and homogenized sediment from the field site in NYC, I built artificial ‘reefs’ in aquaria. I designed a fully crossed experiment to compare the effect of oysters on sediment N cycling under conditions mimicking eutrophication (i.e., high water column nutrients and high sediment organic matter) and under oligotrophic conditions (i.e., no added water column nutrients and low sediment organic matter). I collected sediment cores from aquaria and analyzed N cycling using the same analytical approaches as the field study. I predicted that aquaria with oysters would have higher rates of denitrification than aquaria without oysters. Additionally, I predicted that eutrophic aquaria would have higher rates of denitrification than oligotrophic aquaria. I expected coupled nitrification- denitrification would be the dominant N pathway for denitrification in both oligotrophic and eutrophic aquaria.

Results from both chapters will inform management practices as well as assist oyster reef restoration efforts of the Hudson River Foundation and other stakeholders. In

12 addition, these results provide insight to N biogeochemistry in urbanized ecosystems, which has not been extensively studied.

CHAPTER II

OYSTER REEF RESTORATION AND DENITRIFICATION PATHWAYS

IN AN URBANIZED EUTROPHIC ESTUARY

Introduction

Changes in oyster reefs and their ecosystem services

Oyster reef loss is a global phenomenon. Approximately 85% of oyster reefs have been lost worldwide, with some areas deemed ‘functionally extinct’ or unable to sustain an oyster fishery or reef-related ecosystem services (Beck et al. 2011). Reef losses are driven by multiple stressors such as overharvesting (Rothschild 1994, MacKenzie et al.

1997, Halpern et al. 2008), eutrophication, disease, and altered hydrology, which cause siltation of reefs and inhibit larval recruitment (Jackson et al. 2001, MacKenzie 1996).

Intact oyster reefs provide many ecosystem services. Oyster reefs create complex three-dimensional structures (Coen et al. 2007, Gutiérrez et al. 2003), which provide habitat to invertebrates, fishes, and organisms that attach to hard surfaces (e.g., mussels, algae, and other microbes) (Nestlerode et al. 2007, Gutiérrez et al. 2003). Reefs provide refugia and nursery sites, which increase abundance and diversity of fish with nearshore juvenile stages (Coen et al. 2007). Reefs also stabilize benthic and shoreline sediments, which may help mitigate rising sea level and shoreline erosion (Coen et al. 2007).

Finally, filter-feeding organisms can increase water clarity, with beneficial effects on the

13 14 recreational and property values of nearshore habitats (Grabowski and Peterson 2007,

Coen et al. 2007). Similar to other bivalve suspension feeders (Bruesewitz et al. 2006), oysters filter particulates from the water column and deliver carbon (C) and nitrogen (N) in biodeposits (i.e., feces and pseudofeces) to the sediment. Oyster filtrate is sorted and selectively ingested (Haven and Morales-Alamo 1966, Newell and Jordan 1983).

Pseudofeces contain filtered items that are rejected (e.g., sand), and are released loosely bound in mucus strings (Newell and Langdon 1996). Feces are produced following digestion, and are released from the oyster as mucus-bound pellets. Oyster filtration is highly variable and dependent on oyster size, seston quality, and temperature (Riisgard

1988, Newell and Langdon 1996, Cranford et al. 2011). However, biodeposits are larger, contain more C and N, and sink up to 40% faster than un-aggregated, naturally settling organic matter (Newell et al. 2005), and thereby likely stimulate biogeochemical processes in underlying sediment (Lund 1957, Hoellein et al. 2015).

Oysters and sediment N cycling

Nitrogen is considered a primary limiting nutrient in marine ecosystems, and may undergo multiple potential transformations prior to removal from the aquatic environment. Through assimilation of inorganic N or N-fixation, organisms incorporate

N into their tissues, where it can be incorporated into food webs and recycled following death and excretion. Organic N in dead tissues and wastes are converted first into

+ ammonium (NH4 , a highly reduced form of N), which can be directly assimilated by microbes or plants, and/or transformed via several microbially mediated processes.

+ - Nitrification by aerobic bacteria converts NH4 into nitrate (NO3 ). Anaerobic ammonia

15 + - oxidation (anammox) converts NH4 and nitrite (NO2 ) into N2 (Strous et al. 1999,

+ - Thamdrup and Dalsgaard 2002). Like NH4 , NO3 can be assimilated directly by plants and microbes, and has several fates through dissimilatory transformations. For example,

- + - dissimilatory NO3 reduction to NH4 (DNRA) by anaerobic bacteria reverts NO3 into

+ NH4 , creating a feedback loop that retains N in an environment and enhances eutrophication (An and Gardner 2002, Giblin et al. 2013). In contrast, denitrification is

- the aerobic reduction of NO3 to N2 gas through linked dissimilatory transformations

(with creation and reduction of nitrous oxide as an intermediary step). Denitrification results in the removal of biologically reactive N from aquatic habitats, so factors that control denitrification are a focus of research in eutrophic ecosystems.

Oyster biodeposits may enhance removal of N from the aquatic ecosystem by stimulating sediment denitrification (Newell 2004, Kellogg et al. 2014) by altering one or

- more of the three primary drivers of denitrification: organic C, NO3 , and redox conditions (Herbert 1999, Philippot 2002). Oyster biodeposits may enhance the quality or

- quantity of sediment C, thereby driving denitrification of water column NO3 (i.e., direct

- denitrification). This is typical in well-mixed coastal habitats high in water column NO3

(Hoellein and Zarnoch 2014). In addition, organic N within biodeposits can be

+ + mineralized to ammonium (NH4 ), and oysters excrete NH4 directly. If oyster-derived

+ - NH4 is nitrified, the NO3 produced may be subsequently denitrified (i.e., coupled

- + nitrification-denitrification). In areas with low water column NO3 relative to NH4 , this can be the dominant pathway for N2 production (Herbert 1999, Heiss et al. 2012, Smyth

2013, Kellogg et al. 2014). Coupled nitrification-denitrification requires adjacent anoxic-

16 oxic microsites, which support close association of organisms and chemical substrates for each process. Enhancing N delivery to sediments and the removal of N from ecosystems via denitrification may help mitigate effects of cultural eutrophication, but quantifying these effects requires careful measurement of separate denitrification pathways (i.e., coupled vs. direct) and environmental controls on N cycling rates to understand the processes involved.

Restoration of native oyster reefs in coastal ecosystems has been proposed as a method to promote denitrification. Understanding the environmental conditions that control oyster-mediated denitrification is important for planning reef conservation and restoration (Hoellein et al. 2015). Previous research has produced equivocal results, however, as oysters increase denitrification rates in some cases, but have little effect in others. For example, Hoellein et al. (2015) showed denitrification in oyster reef-adjacent sediments was higher than denitrification in reef-distal sediments at one reef in New

Hampshire, but found no effect at another reef in the same estuary. Comparisons of restored oyster reefs to oyster-free control sites at a reef in Chesapeake Bay showed significantly higher denitrification in the reefs (Kellogg et al. 2013). Finally, experimental enclosures of oysters at varying densities increased sediment organic matter, but had no effect on denitrification potential (Hoellein and Zarnoch 2014) or denitrification gene abundance (S. Lindeman, in review) in Jamaica Bay, NYC. In summary, oyster reefs may enhance denitrification, but their effect on the underlying controls for various denitrification pathways require further study, especially for reefs in waters influenced by urbanization.

17 Oyster reef restoration in NYC

Prior to 1900, oysters were common throughout the Atlantic coast of the United

States, including New York Harbor (Levinton et al. 2013, Zu Ermengassen 2012,

Mackenzie et al. 1997). Unsustainable harvesting depleted the New York and New Jersey fishery by the 1920’s (Franz 1982). Other contributions to oyster reef loss in the area include the creation of shipping passages and dredging (Mackenzie et al. 1997) as well as cultural eutrophication and disease (Levinton 2013).

Eastern oyster (Crassostrea virginica) reef restoration in NYC is supported by local and state governmental agencies. Reef restoration is ongoing at several locations including the Hudson River, Jamaica Bay, and East River. One primary objective for restoration is to improve water quality via oyster filtration and enhanced denitrification

(Cornwell et al. 1999, Newell et al. 2005, Coen et al. 2007, Grizzle et al. 2013). Recent evidence using experimental enclosures suggests that oysters can increase sediment C and may affect denitrification potential (Hoellein and Zarnoch 2014). However, no previous studies have examined environmental controls and pathways for denitrification in restored oyster reefs in urban ecosystems.

Objectives

In this study, I measured the effects of a recently restored oyster reef in New York

City on sediment denitrification pathways on a seasonal basis. I compared fluxes of N solutes and gasses at the sediment-water interface from cores collected immediately adjacent to a restored reef and 10-15 m away. In the lab, I installed cores in a continuous- flow system where inflow water was amended with 15N isotopes to create three

18 15 + 15 - treatments: control, NH4 , and NO3 . The isotopes allowed me to follow the chemical pathways of N2 production (i.e., coupled vs. direct denitrification). I predicted that sediment in reef-adjacent cores would have more C and N from oyster biodeposits, with higher rates of total denitrification. I expected reef-adjacent cores would show higher coupled nitrification-denitrification than direct denitrification, stimulated by

+ mineralization of NH4 from oyster biodeposits and the ability of the 3-dimensional structure of the oyster reef to maintain oxic-anoxic microsites (Kellogg et al. 2013). In contrast, I predicted the pathway for N2 production in reef-distal cores would be more evenly divided between coupled and direct denitrification, with lower rates overall.

Methods

Study site and reef construction

The study site was located near Soundview Park in the East River, Bronx, New

York City. The East River is a 26 km tidal strait that connects Upper New York Bay to

Long Island Sound. Daily tides occur at both ends of the strait due to tidal head differences and can reach speeds of up to 5 knots, leading the waters to be well-mixed and not stratified (O’Shea and Bronson 2000). Waters at Soundview Park receive freshwater input from the Bronx River, creating brackish conditions of 25 ppt at the constructed oyster reef (Table 1). The Bronx River watershed covers 263 km2 with 24% impermeable surface cover (Hoellein et al. 2011), contributing to N loading in the estuary. Additional N loading occurred via combined sewer overflow pipes located in the

Bronx River estuary and East River.

19 Oyster reef construction at Soundview Park began in autumn 2010, directed by the Army Corps of Engineers. First, a 50 m2 (5 x 10 m) rock base of 20-30 cm diameter rubble was laid on the sediment, followed by a layer of surf clam shell. Juvenile oysters as spat-on-shell (SOS) were raised at the NY Harbor School, transported to the site, and evenly distributed by hand throughout the reef in October 2010. A total of 58,500 SOS were initially placed at the reef. Oyster reef monitoring began in November 2010 with wading observers using 9, 0.1 m2 quadrats. Observers noted that substantial erosion and transport of SOS had occurred during autumn. Therefore, in June 2011, an additional

55,700 SOS were added to 18 m2 on the northern half of the reef (Grizzle et al. 2013).

Collection of sediment cores

We collected sediment cores to measure sediment biogeochemistry on July 31,

2011, October 1, 2011, and May 7, 2012. We accessed the reef via wading during low tide. Six sediment cores (7.6 cm in diameter and 15-20 cm depth) were taken directly adjacent to the oyster reef, and six cores were taken 10-15 m away from the reef. The core sampler was fitted with a one-way rubber flow valve, which preserved the sediment- water interface (SWI) with minimal disturbance during collection (Gardner et al. 2006).

The 6 reef-distal locations were chosen by a concurrent hydrologic study that showed oyster biodeposits were least likely to be transported to that location (Brett Branco, unpublished data). In July 2011 only, we collected 12 additional cores to measure the effects of live oyster spat on nutrient transformations. Of those additional twelve, we collected 6 cores from within the reef that contained spat-on-shell, and collected 6 cores in areas containing shell only. After sampling, all cores were closed with tightly fitting

20 rubber caps, sealed with electrical tape, stored in a dark cooler, and taken back to laboratory. We also collected six, 20 L carboys of unfiltered site water to use in flow- through measurements. There was no evidence for DO stratification on any date, attributed to shallow depth during collection (~1 m) and tidal flow.

+ - On each sampling date, we measured concentrations of SRP, NH4 , NO2 , and

- NO3 (see below). In addition, triplicate water samples (100-200 ml) were used to quantify total chlorophyll a concentration. We filtered samples through a 0.45 µm nitrocellulose membrane. Membranes were stored in the dark at -20°C until analyzed.

Samples were extracted overnight at 4°C in 90% acetone. Chlorophyll a was measured fluorometrically (Turner Designs, Sunnyvale, CA) (Parsons et al. 1984). Hudson River

Foundation researchers also collected water quality data throughout my sampling period during regular monitoring, including temperature, pH, salinity, and dissolved oxygen using handheld YSI meters (YSI, Inc., Yellow Springs, OH).

Flow-through measurements of nutrient and gas fluxes

The sediment cores were set up in continuous-flow incubations within 2-3 hours of sampling, using the protocols described in Gardner et al. (2001). First, we removed water from each core, leaving 5 cm of water above the SWI (approximately 227 mL). If present, we carefully removed benthic fauna such as eastern mud snails (Ilyanassa obsoleta). A plunger with a rubber O-ring was fitted into each core to create a tight seal.

The plunger was plumbed with a Teflon inlet and outlet tubing (Lavrentyev et al. 2000;

An et al. 2001). Aerated site water flowed into each core via a 16 channel peristaltic pump (model 205U, Watson Marlow Pumps Group, Falmouth, Cornwall, UK) at a rate of

21 1.1 mL minute-1, and core outflows were collected in beakers. We carefully removed any air bubbles from the core headspace at the beginning of the analysis, and cores were monitored for bubbles throughout. We wrapped cores in aluminum foil to prevent bubble formation via photosynthesis, although we observed no benthic algae and little bubble formation inside cores. We conducted all measurements at room temperature.

We established three treatments in the site water that flowed into the cores: two

15 15 + cores received no N (control), two cores had added NH4 (final concentration 10 μM

15 15 - 15 NH4-N), and two cores had added NO3 (final concentration 10 μM K NO3-N)

(Nielsen 1992, Gardner and McCarthy 2009). These isotope treatments enriched mean

14,15 + 14,15 - (±SE) inflowing water column nutrient concentrations of NH4 and NO3 by a factor of 1.45 (±0.11) and 1.55 (±0.05), respectively. The experiment was repeated for cores collected adjacent to the reef (N=6) and cores collected 10-15m away from the reef

(N=6). In summer only, this experiment was repeated for cores containing spat-on-shell

(N=6) and cores with shells only (N=6). Water was passed through the cores for 24 hours to establish steady-state environment prior to sample collection (Gardner and McCarthy

2009). Inflow and outflow water was collected at 24, 48, and 72 hours after the start of the incubation (Bruesewitz et al. 2013). We did not measure steady-state conditions directly, but core outflow water was stable and well oxygenated on each collection period. For example, reef-adjacent mean (±SE) final O2 concentration was 32.1 (±26.4)

µM O2 in July, 160.9 (±3.1) µM O2 in October, and 153.8 (±26.4) µM O2 in May.

At each sample period, we collected water for dissolved inorganic nutrients and gasses. Water samples for dissolved nutrients were filtered with 0.2 µm nylon syringe

22 filters (Thermo Scientific, Rockwood, TN, USA) into triplicate 20 mL liquid scintillation vials (Wheaton, Millville, NJ, USA). In addition, we filtered three replicate samples in 8

15 + mL glass sample vials (Wheaton) for later measurement of dissolved NH4 . Water samples were frozen until measurement of dissolved nutrients. Separately, we collected three replicate water samples for dissolved gas analyses. Water from each inflow and outflow were collected in 15 mL glass vials with ground glass stoppers (Chemclass,

Vineland, NJ, USA). For this process, we filled each vial slowly from the bottom and allowed them to overflow for several volumes (Bruesewitz et al. 2013, Hoellein and

Zarnoch 2014). Samples were preserved with 200 µL of 50% zinc chloride, and stored in glass vials with ground glass stoppers underwater below room temperature until analysis of dissolved gasses (Kana et al. 1994, An et al. 2001, An and Gardner 2002, McCarthy and Gardner 2003).

Sediment Organic Matter

After the final sample collection from the continuous-flow measurements were completed, we collected the top 5 cm of sediment from each core to measure sediment ash-free dry mass. Sediment was placed in 500 mL plastic containers and frozen until analysis. We transferred thawed samples into pre-ashed and weighed aluminum pans, dried samples at 60°C for at least 48 hours, and measured dry mass. We combusted samples for 4 hours at 550 °C. Samples cooled for at least 1 hour in a room temperature desiccator before final measurement of ash mass.

Dissolved solute and gas chemistry

Water chemistry was performed on a SEAL AutoAnalyzer 3 (Seal Analytical, Inc.,

23 Mequon, WI). We measured SRP using the ascorbic acid method (Murphy and Riley

+ - - - 1962), NH4 with the indophenol blue method (Solorzano 1969), and NO2 + NO3 (NOx ) using the sulfanilamide method with a cadmium reduction column. Samples were also

- tested with the sulfanilamide method but without the reduction column to measure NO2

- alone, and NO3 was determined by difference (Greenburg 1985). We modified the sulfanilamide method by amending the ammonium chloride (NH4Cl) reagent with 0.5 g

L-1 of ethylenediaminetetraacetic acid (EDTA) buffer, as EDTA is known to bond with

- metals that can inhibit NO3 reduction.

Dissolved gas analysis was performed using a membrane-inlet mass spectrometer

(MIMS, Bay Instruments, Easton, MD). For each sample, a peristaltic pump moved water from the bottom of the sample vial through a vacuum chamber, which pulled dissolved gasses out of solution across a membrane (Kana et al.1994, An et al. 2001, Bruesewitz et al. 2013). MIMS plumbing includes a trap submerged in liquid N2 to prevent water vapor

28 29 30 contamination. Gasses measured on the MIMS include N2 isotopes ( N2, N2, and N2), argon, and oxygen. We ran periodic standards of equal salinity at 21°C and 30°C to correct for instrument drift throughout the run (Kana et al. 1994, Hoellein and Zarnoch

2014). The quadrupole mass spectrometer in the MIMS can produce O+ ions that form

28 30 nitric oxide (NO) in the presence of N2, thereby affecting N2 and N2 measurements

(Eyre et al. 2002; Kana and Weiss 2004). The error magnitude is machine-specific

(McCarthy and Gardner 2003), and for the machine used in this study (Dr. Wayne

Gardner’s laboratory at the University of Texas Marine Science Institute), the effect was well measured and shown to be negligible (McCarthy et al. 2007, 2008).

24 Flux calculations

We calculated fluxes of dissolved nutrients and gasses by subtracting the outflow from the inflow concentrations, where a positive number indicates net production or release from sediment, and a negative number indicates net retention (An et al. 2001). We multiplied change in element (µmol) by the flow rate of the pump, and then divided by the area of the SWI to obtain a final value for flux in units of mol element m-2 h-1

(McCarthy et al. 2008).

Multiple N2 gas transformations were calculated using the relative N2 flux values

15 + 15 - across the three treatments (control, + NH4 , and + NO3 ). Net N2 flux was the balance of denitrification and N-fixation in control cores. N-fixation was calculated by using a

28,29,30 28,29,30 quadratic equation based on gross N2 denitrification and net N2 production in

15 - the NO3 cores (An et al. 2001). Denitrification (DNF) was calculated as the difference between N-fixation and net N2 flux in the control cores. Potential DNF, or the rate of

- denitrification in the presence of elevated NO3 , was measured as total N2 production

15 - 15 15 + (accounting for N-fixation) in the NO3 cores. Finally, I followed N from the NH4

15 - 29,30 15 and NO3 into the isotopically labeled N2. Coupled DNF represented the N

29,30 15 + converted to N2 in cores that were amended with NH4 . Direct DNF represents the

15 29,30 15 - N converted to N2 in cores that were amended with NO3 . Sediment oxygen demand (SOD) was measured simultaneously with N2 gasses and was calculated as the difference in O2 flux in control cores. I note anammox can also contribute to isotope labeled N2, but studies from nearby estuaries suggest it is a relatively minor contributor of total N2 production (i.e., 2-9% of denitrification; Engström et al. 2005; Koop-Jakobsen

25 and Giblin 2009). Finally, I acknowledge that denitrification rates do not account for incomplete denitrification to nitrous oxide.

Dissimilatory nitrate reduction to ammonium (DNRA) is an anaerobic pathway

- that can represent a significant pathway for NO3 under certain conditions (An and

15 - 15 + 15 - Gardner 2002). I measured DNRA by tracing NO3 into the NH4 pool in the + NO3

15 + cores. I measured NH4 concentrations using high-performance liquid chromatography

(HPLC) as described in Lin et al. (2011), which were derived from Blackburn (1979).

15 + HPLC was used to measure NH4 in water samples from cores that were amended with

15 + + 15 - either NH4 (to measure NH4 uptake) or NO3 (to measure DNRA). The difference

15 + + between DNRA and uptake was net NH4 flux. Rates of DNRA, NH4 uptake, and net

15 + 15 + 15 - NH4 flux are all considered potential values because I measured NH4 and NO3 in

+ - addition to background concentrations of NH4 and NO3 . Additionally, I did not measure

+ 15 + pore-water NH4 or loss of NH4 from cation exchange (Bruesewitz et al. 2013, Smyth et al. 2013, Gardner and McCarthy 2009, An and Gardner 2002).

Statistical analysis

For all fluxes, I first calculated the mean flux on each sampling day, and then calculated mean flux from each individual core by averaging across the 3 sampling days

(24, 48, and 72 hours post incubation). Individual cores were replicates for the statistical

15 + 15 - analysis, where N=2 for each treatment (control, + NH4 , and + NO3 ) (Gardner and

McCarthy 2009, Bruesewitz et al. 2013). I acknowledge the low replication for each treatment’s reduced statistical power; however, there was limited capacity for the number of core measurements that could be completed simultaneously. I decided to prioritize the

26 number and types of N transformations measured, rather than increase the number of replicates and thus complete fewer types of measurements. I used a 2-way ANOVA with reef proximity and date as factors to analyze patterns in dissolved solute flux, sediment organic matter, SOD, N2 gas fluxes (i.e., net N2 flux, N-fixation, denitrification, potential denitrification, coupled denitrification, and direct denitrification), as well as DNRA and

15 + NH4 uptake. For the spat and shell cores collected in July 2011, I used a t-test to compare those response variables between spat-on-shell and shell alone cores. I

14 15 15 - 14 - calculated N DNRA rates by comparing N DNRA rates and NO3 and NO3 concentrations. I also calculated DNRA:direct denitrification ratios, using t-tests to compare reef-distal vs. reef-adjacent cores and spat vs. shell cores. Analyses were performed in Systat v.13 (Systat Software, Inc.) with p-values ≤0.05 as a threshold for significance.

Results

Physicochemical conditions at the Soundview Park oyster reef varied by season

(Table 1). Mean daily water temperature was highest in summer and ranged from 13.8 –

22.1 ˚C. Dissolved oxygen (DO) during daytime was lower in summer and spring, and

+ - ranged from 4.7 – 6.2 mg/L. Water column concentrations of NH4 and NO3 were much

- greater in fall than in summer or spring, while SRP and NO2 were less variable.

Chlorophyll a was higher in summer than fall or spring (Table 1). Finally, pH and salinity were relatively uniform across sampling dates (Table 1).

Nutrient and gas fluxes in reef-adjacent and reef-distal cores

Fluxes of nutrients and SOD were not affected by reef-proximity, but differed

27 - among seasons. NO2 flux was highest in fall (ANOVA, p=0.001; Table 2), and SOD was

- + highest in summer (Figure 3). While fluxes of NO3 , NH4 , and SRP were variable and not significantly affected by reef proximity or season, there was a trend that cores taken further away from the reef had greater positive fluxes than cores adjacent to the reef, with

- the exception of NO3 , which had greater negative fluxes further away from the reef.

Fluxes of N2 were also not affected by reef-proximity, but did show significant seasonal variation, following seasonal trends similar to those seen in previous studies

(Kemp 1990). For example, denitrification was highest in fall and lowest in spring

(ANOVA p=0.006), and rates were not affected by reef proximity (ANOVA, p=0.959;

15 - Table 3; Figure 4A). Potential denitrification (i.e., total N2 produced with added NO3 ) was an exception to this pattern, as there was a significant interaction between reef proximity and season (ANOVA p=0.023; Table 3), resulting from higher rates of direct denitrification in reef-adjacent sediments in the summer, but identical rates between reef- adjacent and reef-distal sediments in spring and autumn (Figure 4B). Despite the added

15 - NO3 , rates of potential denitrification were no higher than denitrification, suggesting

- that denitrification was not NO3 limited. Finally, patterns of net N2 flux (data not shown) by season and reef-proximity were the same as those for denitrification and potential denitrification because N-fixation was low or not detected (Table 3).

15 - 15 + I used NO3 and NH4 as tracers to follow movement of dissolved inorganic N into N2 gas. Rates of both direct and coupled denitrification were lowest in spring, and there was no effect of reef proximity on either rate (Figure 5; Table 3). Across all

15 15 - sampling dates, rates of N-N2 production were much higher when NO3 was added to

28 15 + inflow water (i.e., direct denitrification) than when NH4 was added to inflow water

(coupled denitrification; Figure 5). Rates of coupled denitrification for reef-adjacent cores represented 11.3% of direct denitrification in summer, 13.5% in autumn, and 0% in spring. Rates of coupled denitrification in reef-distal cores represented approximately the same proportion of direct denitrification: 7.7% in summer, 19.6% in autumn, and 0% in spring.

15 - 15 + I also used NO3 as a tracer to measure DNRA as the rate of NH4 production in reef-adjacent and reef-distal cores in summer. There was no difference in DNRA between reef-adjacent and reef-distal cores in summer (t-test p=0.141), no difference in

15 + 15 + NH4 uptake (t-test, p=0.192), and no difference in net NH4 flux (t-test, p=0.638;

15 15 - 15 + Table 4). The relative rate of N from NO3 converted into NH4 via DNRA relative to the rate converted to N2 via direct denitrification was 5.0% in the reef-adjacent cores, and 28.4% in the reef-distal cores.

Sediment organic matter in reef-adjacent and reef-distal cores

There was a significant difference in organic matter content in cores due to reef proximity (ANOVA, p=0.008), but not season (ANOVA, p=0.126; Table 3), showing a consistent pattern that reef-distal cores had higher amounts of organic matter than reef- adjacent cores (Figure 6). While I did not measure size-fractioned organic material in cores, reef-adjacent cores had more shell material in sediments, which likely accounts for the trend of reduced organic matter proportion.

Nutrient and gas fluxes in spat-on-shell and shell-only cores

In general, nutrient fluxes were higher in spat-on-shell cores relative to those with

29 + only shells (Table 5). Flux of NH4 out of cores was significantly higher in spat-on-shell

- - treatment (t-test p=0.006), and both NO2 and NO3 fluxes were on the margin of statistical significance for higher uptake in spat-on-shell cores (t-test p=0.057 and p=0.081, respectively). In contrast, there was no significant difference between spat and shell SRP fluxes (t-test p=0.940).

I observed no significant differences in SOD (t-test p=0.607; Figure 7) or N2 fluxes between spat-on-shell and shell-only cores, although several N2 fluxes were on the margins of statistical significance (Table 6). For example, cores with spat-on-shell tended to have higher denitrification (t-test p=0.067) and potential denitrification (t-test p=0.099) than shell-only cores (Figure 8). Spat-on-shell and shell-only cores had the same rates of coupled denitrification (t-test p=0.893) and direct denitrification (t-test, p=0.130; Table 6;

Figure 9). As with reef-adjacent and reef-distal cores, direct denitrification was greater than coupled denitrification, and potential denitrification rates were the same as

15 - denitrification, despite the added NO3 (Figure 8). I observed no N-fixation in spat-on- shell or shell-only cores (Table 6). Finally, there was no significant difference in DNRA

15 + between spat treatments (t-test p=0.254) or NH4 uptake (t-test, p=0.170, Table 7).

15 + 15 + However, the balance between NH4 uptake and NH4 regeneration was negative (i.e., net uptake) in both spat-on-shell and shell-only cores, and there was significantly more uptake in shell-only cores than in spat-on-shell cores (t-test, p=0.016; Table 7).

After completing flow-through core measurements, I disassembled spat and shell cores to count oyster spat density. As expected, mean (±SE) in spat cores was 8.5 (±2.4) individuals core-1, corresponding to a density of 1,889 (±541) individuals m-2. The mean

30 spat density in the cores was slightly lower than the spat density stocked in June 2011

(prior to my sampling) of 3,094 individuals m-2. Among the 6 shell-only cores, I found a single core that contained one spat.

31 Table 1. Water column physicochemical properties on each sampling date at the Soundview Park oyster reef. Abbreviations: DO = dissolved oxygen, SRP = soluble + - - reactive phosphorus, NH4 = ammonium, NO2 = nitrite, and NO3 = nitrate. Date

Measurement Units 7/31/2011 (SE) 10/1/2011 (SE) 5/7/2012 (SE)

Temperature ˚C 23.2 21.4 13.0

pH 7.4 7.7 7.8

Salinity ppt 25 25 25

DO mg L-1 5.8 8 6.2

Chlorophyll a µg L-1 33.73 9.60 9.63

SRP µM 2 (0.01) 6 (0.3) 2 (1.2)

NH + µM 9 (0.1) 37 (0.8) 14 (8.0) 4 NO - µM 3 (0.04) 3 (0.1) 2 (0.5) 2 NO - µM 11 (0.9) 30 (4.9) 15 (6.2) 3

Table 2. Mean (±SE) for nutrient fluxes in sediment cores taken adjacent to (near) and 10-15 m away (far) from the Soundview Park oyster reef. Units for fluxes are µmol of P or N m-2 h-1. P-values are from 2-way ANOVA by reef proximity and season, where interact. Interact = interaction of site and season. P-values ≤0.05 are in bold. + - Abbreviations: SRP = soluble reactive phosphorus, NH4 = ammonium, NO2 = nitrite, - and NO3 = nitrate. Near Far p-value Flux Mean SE Mean SE Season Proximity Interact SRP 9.1 11.2 13.6 7.7 0.097 0.667 0.798 + NH4 34.6 64.5 150.4 62.0 0.155 0.167 0.708 - NO2 -3.8 6.7 0.6 8.1 0.001 0.199 0.735 - NO3 -14.6 36.1 -33.7 8.0 0.572 0.637 0.357

Table 3. Mean (±SE) of sediment oxygen demand (SOD), organic matter, and nitrogen gas (N2) fluxes in sediment cores taken adjacent to (near) and 10-15 m away from (far) from the Soundview Park oyster reef. Denitrification (DNF) is measured in control 15 cores (no N added) and represents total N2 production after accounting for N-fixation. Potential DNF is measured in cores amended 15 15 - 15 29,30 15 15 + with N nitrate ( NO3 ). Coupled DNF represents the N in N2 produced in cores amended with N ammonium ( NH4 ). Direct 15 29,30 15 - DNF represents the N in N2 produced in cores that were amended with NO3 . Seasonal samples were taken July 2011, October 2011, and May 2012. P-values are from 2-way ANOVA, with p-values ≤0.05 in bold. Near Far 2-way ANOVA p-value Rate Units Mean SE Mean SE Season Proximity Interaction SOD µmol O m-2 h-1 1139.1 332.2 1291.8 316.3 0.045 0.570 0.405 2 Organic Matter % 2.8 0.7 5.6 0.8 0.126 0.008 0.540 14 -2 -1 N-Fixation µmol N m h 40.5 40.5 0.0 0.00 0.422 0.356 0.422 Denitrification (DNF) µmol 14N m-2 h-1 311.6 96.7 314.4 99.8 0.006 0.959 0.683 Potential DNF µmol 14,15N m-2 h-1 302.0 99.4 234.9 94.1 <0.001 0.012 0.023 Coupled DNF µmol 15N m-2 h-1 1.2 1.5 0.4 0.9 0.001 0.149 0.273 Direct DNF µmol 15N m-2 h-1 20.9 5.6 13.6 5.8 0.041 0.174 0.945

33 Table 4. Mean (±SE) of nitrogen regeneration in July 2011 in cores taken at the Soundview Park oyster reef (near = adjacent to oyster reef, far = 10-15 m away from 15 - reef). All values were measured from cores amended with NO3 and all units are 15 + -2 -1 15 measured in µmol NH4 -N m h . Nitrogen regeneration ( DNRA) represents 15 + dissimilatory nitrate reduction to ammonium, or the amount of NH4 formed when 15 - 15 NO3 was added to cores. DNRA:DirDNF is the ratio of N transformation via DNRA to direct denitrification, with units µmol 15N m-2 h-1. Higher values favor DNRA and 14 14 + -2 -1 lower values favor direct denitrification. DNRA is measured in µmol NH4 -N m h 15 14 - 15 - and was calculated from DNRA rates, NO3 concentrations, and NO3 concentrations. Total nitrogen regeneration (14,15DNRA) is the sum of 14N and 15N DNRA. P-values are taken from t-tests. Near Far t-test Rate Mean SE Mean SE p-value 15DNRA 2.07 0.49 7.10 2.07 0.141 DNRA:DirDNF 0.08 0.04 0.28 0.06 0.122

14DNRA 1.48 0.35 12.48 3.63 0.094 14,15DNRA 3.55 0.83 19.58 5.69 0.108

2500

2000 a ab

)

-1

-2 1500

mol O mol b 1000



SOD

500

0 Near Far Near Far Near Far Summer Fall Spring Figure 3. Mean (±SE) sediment oxygen demand (SOD) by season and reef proximity (near = adjacent to reef, far = 10-15 m away from reef). Fluxes are shown from control cores (i.e., no 15N addition to inflow water). Seasons with different letters represent significant differences as indicated by Tukey’s multiple comparison test. Error bars represent standard error for 2 replicates.

1000 1000 A B

800 800

)

-1

-2 600 a a 600

mol N m N mol

 400 400 b

DNF

200 200

0 0 Near Far Near Far Near Far Near Far Near Far Near Far Summer Fall Spring Summer Fall Spring

Figure 4. Mean (±SE) denitrification (A) and potential denitrification (B) by season and reef proximity (near = adjacent to reef, far = 10-15 m away from reef). Fluxes in (A) are from control cores (no 15N addition) and fluxes in (B) are from cores amended with 15 - NO3 . All error bars represent standard error for 2 replicates.

50 50 A B

40 40

)

-1 a ab b

-2

m 2 30 30

29,30

mol 20 20



DNF (

15 10 a a b 10

B.D. 0 0 Near Far Near Far Near Far Near Far Near Far Near Far Summer Fall Spring Summer Fall Spring

Figure 5. Mean (±SE) coupled denitrification (A) and direct denitrification (B) by season and reef proximity (near = adjacent to reef, far = 10-15 m away from reef). Fluxes in (A) 15 + are from cores amended with NH4 and fluxes in (B) are from cores amended with 15 - NO3 . Spring measurements of coupled denitrification were below detection (B.D.). Seasons with different letters represent significant differences as indicated by Tukey’s multiple comparison test. Error bars represent standard error for 2 replicates.

% Organic Matter Organic %

0 Near Far Near Far Near Far Summer Fall Spring

Figure 6. Mean (±SE) organic matter by season and reef proximity (near = adjacent to reef, far = 10-15 m away from reef). Error bars represent standard error for 2 replicates.

Table 5. Mean (±SE) for nutrient fluxes in sediment cores with spat-on-shell and shell- only substrates from the Soundview Park oyster reef. Units for fluxes are µmol of P or N m-2 h-1. P-values are from a t-test, with p-values ≤0.05 in bold. Abbreviations: SRP = + - - soluble reactive phosphorus, NH4 = ammonium, NO2 = nitrite, and NO3 = nitrate. Spat-on-shell Shell-only t-test Flux Mean SE Mean SE p-value SRP 8.6 7.3 7.9 3.1 0.940 + NH4 168.3 5.4 79.7 4.0 0.006 - NO2 -31.2 4.1 -14.6 0.5 0.057 - NO3 -166.9 43.2 22.8 38.3 0.081

Table 6. Mean (±SE) for sediment oxygen demand (SOD) and nitrogen gas (N2) fluxes in sediment cores with spat-on-shell and shell- only substrates. Denitrification (DNF) is measured in control cores with no 15N added. Potential DNF is measured in cores that were 15 15 - 29,30 15 amended with N nitrate ( NO3 ). Coupled DNF represents N2 produced in cores that were amended with N ammonium 15 + 29,30 15 - ( NH4 ). Direct DNF represents N2 produced in cores that were amended with NO3 . P-values are from t-tests. Spat-on-shell Shell-only t-test Rate Units Mean SE Mean SE p-value -2 -1 SOD µmol O2 m h 1475.6 379.3 1232.3 136.6 0.607 N-Fixation µmol 14N m-2 h-1 0.0 0.0 0.0 0.0 - Denitrification µmol 14N m-2 h-1 266.1 22.6 181.4 5.4 0.067 Potential DNF µmol 14,15N m-2 h-1 243.3 34.2 142.9 1.2 0.099 Coupled DNF µmol 15N m-2 h-1 4.2 1.4 4.8 3.9 0.893 Direct DNF µmol 15N m-2 h-1 34.6 8.2 9.1 6.1 0.130

2000

1500

)

-1

-2

2 1000

mol O mol



SOD 500

0 Spat Shell

Figure 7. Mean (±SE) sediment oxygen demand (SOD) for spat-on-shell (spat) and shell- only (shell) reef substrate. Fluxes are shown from control cores (i.e., no 15N addition to inflow water). Error bars represent standard error for 2 replicates.

350 350 A B 300 300

) 250 250

-1

-2 200 200

mol N m N mol 150 150 Data Y



DNF DNF 100 100

50 50

0 0 Spat Shell Spat Shell

Figure 8. Mean (±SE) denitrification (A) and potential denitrification (B) in spat-on-shell (spat) and shell-only (shell) cores. Fluxes in (A) are from control cores (no N addition) 15 - and fluxes in (B) are from cores amended with NO3 . Error bars represent standard error for 2 replicates.

50 50 A B

) 40 40 B

-1 h B

-2

30 30

N m

mol 20 20



DNF ( DNF 10 10

0 0 Spat Shell Spat Shell

Figure 9. Mean (±SE) coupled denitrification (A) and direct denitrification (B) in spat- on-shell (spat) and shell-only (shell) cores. Fluxes in (A) are from cores amended with 15 + 15 - NH4 and fluxes in (B) are from cores amended with NO3 . Error bars represent standard error for 2 replicates.

Table 7. Mean (±SE) of nitrogen regeneration in July 2011 in cores taken at the Soundview Park oyster reef with spat-on-shell and shell-only substrates. All values were 15 - 15 + measured from cores amended with NO3 and all units are measured in µmol NH4 -N m-2 h-1. Nitrogen regeneration (15DNRA) represents dissimilatory nitrate reduction to 15 + 15 - ammonium, or the amount of NH4 formed when NO3 was added to cores. DNRA:DirDNF is the ratio of N transformation via DNRA to direct denitrification, with units µmol 15N m-2 h-1. Higher values favor DNRA and lower values favor direct 14 14 + -2 -1 denitrification. DNRA is measured in µmol NH4 -N m h and was calculated from 15 14 - 15 - DNRA rates, NO3 concentrations, and NO3 concentrations. Total nitrogen regeneration (14,15DNRA) is the sum of 14N and 15N DNRA. P-values are taken from t- tests.

Spat-on-shell Shell-only t-test Rate Mean SE Mean SE p-value Nregen 5.1 0.3 7.1 1.3 0.254 DNRA:DirDNF 0.2 0 1.2 0.7 0.253

14DNRA 11.0 0.6 9.5 1.7 0.504 14,15DNRA 16.1 0.9 16.7 3.0 0.862

39 Discussion

My research goals were to measure the effect of 1) oyster reef restoration and 2) oyster spat-on-shell on sediment denitrification pathways in a eutrophic coastal ecosystem. Reef-adjacent cores had significantly higher rates of potential denitrification in summer relative to the reef-distal cores, and cores with spat-on-shell had significantly

+ higher denitrification and NH4 efflux than shell alone. These data, combined with high variation in biogeochemistry among seasons, substrate types, and N2 production pathways, illustrate the underlying environmental controls on sediment N dynamics at this site and allow us to project the conditions under which oyster reef restoration is most likely to affect N2 production in urbanized coastal environments elsewhere.

Oyster reef proximity and denitrification

- Major drivers of denitrification rates include NO3 availability, sediment organic matter, and favorable redox conditions (Groffman et al. 1999, Herbert 1999). Oysters should have the strongest effect on N cycling if biodeposition enhanced N or C pools at time periods when either substrate is limiting denitrification. My results showed potential

28,29,30 15 - denitrification (i.e., N2 in cores amended with NO3 ) was higher in reef-adjacent cores than reef-distal cores in summer, with no differences during the other two seasons.

- This pattern may be attributable to organic matter or NO3 limitation of denitrification.

Nitrate concentrations were lowest in summer (Table 1), and organic matter was lower in summer and fall relative to spring (Table 3, Figure 6), both of which suggest N and/or C limitation of denitrification would be most likely in summer. In addition, I observed significantly higher potential denitrification in reef-adjacent relative to more distant

40 cores, but no effects on denitrification (i.e., unamended control cores). This further

- suggests that reef-adjacent denitrification was limited by NO3 concentrations, while denitrifiers in reef-distal cores were not.

A second explanation for why oyster reef proximity effects on potential denitrification were restricted to summer could be the evolving physical structure of the restored reef. My first sampling date was in July 2011, after the reef was seeded with spat for a second and final time in June 2011. Anecdotally, I observed the dispersal of oysters and shell substrate during the following autumn and spring. This occurred because the spat-on-shell were not permanently affixed to immobile structures, and shifted away from the reef with wave action. The agency responsible for reef construction and monitoring acknowledged the movement of oysters after the reef’s initial construction in October

2010, and added 55,700 additional spat-on-shell to the reef in June 2011. It is likely that oyster export continued throughout my sampling period (July 2011 to May 2012), which may have lead to oyster burial and death. Soft, organic-rich sediments typical of urban coastal environments may face this challenge elsewhere, which could inhibit reef growth, reduce recruitment, and minimize the chances that oysters could affect sediment N dynamics. Allowing oysters to attach to large, stable substrates (e.g., concrete blocks) before reef construction may benefit future restoration in this ecosystem.

A third reason that oyster reef proximity effects on potential denitrification were limited to summer was the relative abundance of oysters and food availability.

Approximately 58,500 SOS were initially seeded to form the reef in Fall 2010 and an additional 55,700 were seeded in June 2011 (Grizzle et al. 2013). A month later, the

41 oyster population on the reef in July 2011 was estimated at 23,656 individuals (Alison

Mass Fitzgerald, unpublished data). Although I do not know the oyster population size for October 2011 and May 2012, the population most likely continued to decline. In addition to waves and sedimentation decreasing oyster populations on the reef, recruitment after my summer sampling period was likely minimal because oyster reproduction in this area ceases by September (Levinton 2013). Finally, chlorophyll α concentration was also highest in summer compared to the other two seasons (Table 1).

When combined, the higher concentration of phytoplankton, larger oyster populations,

- more consolidated reefs, and lower NO3 concentrations in the summer offer strong explanations for the seasonally specific oyster-mediated effect on increased denitrification potential. However, it’s possible that as seasons were only sampled once each throughout one year, that seasonality may be less important than the time passed since reef construction. My results could also be affected by the time of sample collection, as I always took cores and water samples from the oyster reef at periods of low tide.

Contrary to my predictions, the primary inorganic N substrate for denitrification

- + was NO3 rather than NH4 , regardless of oyster reef proximity. Rates of direct

15 - 29,30 denitrification (i.e., NO3 to N2) were consistently much higher than rates of coupled

15 + 29,30 denitrification (i.e., NH4 to N2), indicating that coupled nitrification-denitrification and/or anammox was not a major pathway for N2 production across seasons or oyster reef proximity. This is in agreement with recent research in Great Bay, New Hampshire

(Hoellein et al. 2015), but is inconsistent with patterns observed elsewhere (Newell et al.

42 2002, Laursen and Seitzinger 2002, Smyth et al. 2013). However, it is possible the pattern documented here may change with oyster reef age and water column conditions. I expect mature reefs may promote greater coupled nitrification-denitrification as the 3- dimensional complexity of reef structure may facilitate the adjacent oxic and anoxic microsites required. Coupled nitrification-denitrification may be more prevalent in

- systems with lower NO3 and higher O2 concentrations (Seitzinger 2006, Newell 2002).

- An abundance of available NO3 in the water column decreases the importance of

- nitrifying bacteria to provide NO3 and can decouple nitrification and denitrification

- (Hoellein and Zarnoch 2014, Risgaard and Peterson 2003). High NO3 concentrations are typical of well-mixed coastal waters in urban environments such as my study site and nearby ecosystems (Hoellein and Zarnoch 2014).

Seasonal variation in the effect of oysters on sediment N cycling has been shown in previous research. For example, in Bogue Sound, North Carolina, sediment adjacent to oyster reefs had higher denitrification rates compared to intertidal and subtidal flats, with highest rates in summer and the lowest rates in winter (Piehler and Smyth 2011). In addition, Kellogg et al. (2013) showed that intact oyster reefs had significantly higher denitrification rates than nearby sediments with no oysters, and the difference between the two types of sediment was greatest in summer and lowest in winter. Those results are consistent with my data by indicating summer as a period of stronger influence of oysters on denitrification than spring or fall.

Water column nutrient concentrations may affect the role of oysters on sediment

N cycling, and concentrations can be driven by land-use patterns and/or from sediment

43 effluxes of N and P. For example, Hoellein et al. (2015) found that eutrophic conditions (i.e., high water column chlorophyll a and sediment organic matter) at an oyster reef in New Hampshire were correlated with increased sediment denitrification relative to an oligotrophic reef. In Jamaica Bay, NYC, Hoellein and Zarnoch (2014) showed oysters increased sediment C but did not affect potential denitrification, because potential denitrification was not primarily C limited. Finally, sediments underlying

+ bivalve aquaculture may be sinks for particulate and dissolved N, and sources of NH4

(Higgins et al. 2013, Dame 2012, Hatcher et al. 1994, Dame 1989), thereby affecting the balance between denitrification and DNRA (Nizzoli et al. 2006).

- The relative rates of denitrification and potential DNRA suggest that NO3 was

+ more likely to be denitrified than converted to NH4 via DNRA, similar to results documented in temperate coastal environments elsewhere. Ratios of DNRA to DNF were

8% in reef-adjacent cores and 28% in reef-distal cores, with no significant effect of reef- proximity. Converting data from Smyth et al. (2013) into identical units as this study, I

−2 −1 estimate that mean N2 production near oyster reefs in summer was 214 μmol N m h while mean potential DNRA was 86.0 μmol N m−2 h−1, giving a DNRA to denitrification ratio of 40%, slightly higher than documented here. Similarly, in nearby Jamaica Bay, the abundance of genes for nitrite reductase used in denitrification (i.e., nirS and nirK) outnumbered the abundance of genes for nitrite reductase used in DNRA (i.e., nrfA)

(Lindemann, in review). The relative abundance of the nrfA gene to nir genes was 0.47% in sediment exposed to oysters, and 0.50% in sediment with no oysters (Lindemann, in

44 review). While the assembled ratios cover a large range, rates and genes for denitrification were consistently higher than those for DNRA.

- Like denitrification, DNRA requires anoxia, organic matter, and NO3 . Previous studies have suggested that DNRA may exceed denitrification in conditions of high sediment organic matter, sulfide, and sustained anoxia (Koop-Jakobsen and Giblin 2010,

McGlathery 2007, Gardner 2006). Although I did not see a significant difference in

DNRA in sediment cores due to oyster reef proximity and did not measure sulfide, higher sediment organic matter in reef-distal cores may in part explain the trend of higher

DNRA relative to reef-adjacent cores.

Spat-on-shell and N cycling

+ Spat-on-shell cores showed significantly higher NH4 fluxes and denitrification rates relative to shell-only cores, suggesting that live oysters impact N cycling beyond the effects of sediment that are simply exposed to oyster activity (i.e., reef-adjacent cores).

+ Excretion by spat was the most likely reason for increased NH4 flux. However, I note

+ that increased NH4 efflux did not correspond with increased coupled nitrification-

- denitrification. This suggests the N2 produced from denitrification originated from NO3

+ (i.e., direct denitrification). Therefore, NH4 excreted by spat or mineralized from biodeposits was not a source of N for denitrification when measured using sediment exposed to oysters (i.e., reef-adjacent cores), or sediment cores that contained live oysters. Instead, it is more likely that oysters contribute to direct denitrification via the C

+ in their feces and pseudofeces, and not by supplying NH4 in excretion or mineralized from biodeposits.

45 + Other studies have found that live bivalves increase NH4 flux and denitrification more than sediment alone or sediment that was exposed to the activity of

+ bivalves. Kellogg et al. (2013) measured significantly higher denitrification and NH4 flux in cores with live oysters relative to sediment alone, attributed to oyster excretion

- and coupled-denitrification, as water column NO3 was low in that study. Smyth et al.

(2013) compared nutrient and gas fluxes from cores that contained sediment alone, cores with sediment and oysters, and cores with oysters and no sediment. Fluxes of net N2 and

+ NH4 were highest in the oyster-only cores, followed by oysters with sediment, both of which had significantly higher fluxes than sediment-alone. Using identical continuous-

+ flow cores as this study, Turek and Hoellein (2014) found higher net N2 flux and NH4 flux in sediment cores that contained live clams (Corbicula fluminea), relative to sediment alone or sediment that had been exposed to C. fluminea biodeposits in an urban river.

In this study and those listed above, bivalves can stimulate N gas and solute fluxes via feces and pseudofeces, however, several additional pathways are possible. For example, microbial communities that colonize shell biofilms can carry out nitrification and denitrification (Svenningsen et al. 2012, Heisterkamp et al. 2010). Similarly, microbes in digestive systems of zebra mussels (Dreissena polymorpha) increase N gas fluxes via denitrification (Svenningsen et al. 2012). Finally, infaunal bivalves (e.g., clams such as C. fluminea) increase flux of inorganic nutrients and gasses between the water column and sediment microbes, and thereby enhance rates of nitrification and denitrification (Zhang et al. 2011, Turek and Hoellein 2014). To my knowledge, the

46 relative effects of sediment microbes and oyster shell or gut containing microbes to N gas fluxes have not been measured.

Comparison of SOD and N-fixation rates to literature values

Sediment oxygen demand (SOD) was not affected by proximity to the oyster reef, and although it was variable among seasons, SOD values were similar to previous studies of oyster reef sediments and positively related to sediment organic matter and DNRA.

-2 -1 Overall, mean SOD ranged from 1100-1300 µmol O2 m h with the highest rates from

-2 -1 individual cores measured in summer (1800 µmol O2 m h ). Similar values were found

-2 -1 in oyster reefs in North Carolina, where SOD was ~1500 µmol O2 m h , with a summer

-2 -1 maximum of 2700 µmol O2 m h (Piehler and Smyth 2011). A recent study in the

Chesapeake Bay demonstrated the potential for restored oyster reef SOD to reach very

-2 -1 high rates of 12,000-38,000 µmol O2 m h (Kellogg et al. 2013). I found a positive relationship between SOD and sediment organic matter, which has been documented elsewhere (Smyth et al 2013, Walker and Snodgrass 1986).

Nitrogen fixation was below detection in most cores, was not affected by reef proximity, and is likely not a major component of N cycling at this site. I found N-

15 - fixation in only 1 reef-adjacent sediment core in summer, and in 3 of the NO3 amended cores in the spring (1 reef-adjacent, 2 reef-distal). Because incubations were completed in the dark, N-fixation was likely heterotrophic. These results are similar to Hoellein et al.

(2015), who showed oyster reef proximity had no effect on N-fixation rates, including reefs at both an oligotrophic and eutrophic location. N-fixation was also not found in oyster reefs in North Carolina (Smyth et al. 2013). Nutrient-rich oyster biodeposits may

47 reduce N limitation in sediments near oyster reefs, thereby decreasing N-fixation, although this has not been experimentally evaluated.

Expectations for reef restoration on sediment biogeochemistry in urban habitats

I predicted that oysters would enhance coupled and direct denitrification adjacent to the reef relative to sediment more distant from the reef. However, results suggest this expectation should be tempered, as the effects of oysters were significant in summer alone, and there was little evidence for coupled nitrification-denitrification as a significant pathway for N2 production. Managers in this region likely will continue to support oyster reef restoration as a technique for mitigating eutrophication and providing other ecosystem services. Therefore, I suggest reef construction that maintains oyster density to maximize the potential for oyster reefs to affect biogeochemical processes. If this occurs, organic matter from oyster biodeposits may increase direct denitrification of

- water column NO3 . Expanding reef area may also promote natural recruitment and reef complexity, thereby increasing benthic habitat heterogeneity and associated ecosystem services (Coen et al. 2007).

In addition to increased oyster density, reducing reef displacement may be beneficial. During data collection, I noticed gradual movement of spat-on-shell away from the reef over the span of several months. ‘Mucky’ urban sediments, such as those found in the East River, are less likely to provide strong anchoring positions favored by

C. virginica and may result in oyster burial. Restoration efforts may benefit from a mixture of reef construction techniques including immobile substrate for settling larvae,

48 such as concrete blocks or shells in plastic mesh bags (Grizzle et al. 2013, Yianopoulos and Anderson 2003, Coen et al. 2011).

Oyster reef restoration may be one useful component of larger N mitigation strategies for urbanized coastal habitats such as New York and New Jersey, but I suggest this strategy should not be employed in isolation. Although there is a potential to increase

N2 production, the size and density of oyster reefs required is unlikely to mitigate N inputs entirely (Higgins 2013). Physical improvements to wastewater systems and combined sewer overflows could reduce N input to NY waterways to a greater degree.

Supporting programs that implement N credit trading could provide economic incentives to businesses and increase awareness of N pollution (Jones et al. 2010). These benefits build upon the economic value of oysters as low-maintenance nutrient regulation systems, and will require continued measurements of N fluxes as reef management policies evolve in this region.

CHAPTER III

THE INTERACTION OF OYSTER PRESENCE AND EUTROPHICATION

IN REEF AQUARIA MICROCOSMS

Introduction

Changes in oyster reefs and their ecosystem services

1997, Halpern et al. 2008), eutrophication, disease, and altered hydrology, which cause siltation of reefs and inhibit larval recruitment (Jackson et al. 2001, MacKenzie 1996).

Intact oyster reefs provide many ecosystem services. Oyster reefs create complex three-dimensional structures (Coen et al. 2007, Gutiérrez et al. 2003), which provide habitat to invertebrates, fishes, and organisms that attach to hard surfaces (e.g., mussels, algae, and microbes) (Nestlerode et al. 2007, Gutiérrez et al. 2003). Reefs provide refugia and nursery sites, which increase abundance and diversity of fish with nearshore juvenile stages (Coen et al. 2007). Reefs also stabilize benthic and shoreline sediments, which may help mitigate rising sea level and shoreline erosion (Coen et al. 2007). Finally, filter- feeding organisms can increase water clarity, with beneficial effects on the recreational

49 50 and property values of nearshore habitats (Grabowski and Peterson 2007, Coen et al.

2007). Similar to other bivalve suspension feeders (Bruesewitz et al. 2006), oysters filter particulates from the water column and delivering carbon (C) and nitrogen (N) in biodeposits (i.e., feces and pseudofeces) to the sediment. Oyster filtrate is sorted and selectively ingested (Haven and Morales-Alamo 1966, Newell and Jordan 1983).

Pseudofeces contain filtered items that are rejected (e.g., sand), and are released in loosely bound in mucus strings (Newell and Langdon 1996). Feces are produced following digestion, and are released from the oyster as mucus-bound pellets. Oyster filtration is highly variable, and dependent on oyster size, seston quality, and temperature

(Riisgard 1988, Newell and Langdon 1996, Cranford et al. 2011). However, biodeposits are larger, contain more C and N, and sink up to 40% faster than un-aggregated, naturally settling organic matter (Newell et al. 2005), and thereby likely to stimulate biogeochemical processes in underlying sediment (Lund 1957, Hoellein et al. 2015).

Oysters and sediment N cycling

Nitrogen is considered a primary limiting nutrient in marine ecosystems, and has multiple transformations prior to removal from the aquatic environment. Through assimilation of inorganic N or N-fixation, organisms incorporate N into their tissues, where it can be transferred into food webs and recycled following death and excretion.

+ Organic N in dead tissues and wastes are converted first into ammonium (NH4 , a highly reduced form of N), which can be directly assimilated by microbes or plants, and/or transformed via several microbially mediated processes. Nitrification by aerobic bacteria

+ - converts NH4 into nitrate (NO3 ). Anaerobic ammonia oxidation (anammox) converts

+ - NH4 and nitrite (NO2 ) into N2 (Strous et al. 1999, Thamdrup and Dalsgaard 2002). Like

51 + - NH4 , NO3 can be assimilated directly by plants and microbes, and has several fates

- through dissimilatory transformations. For example, dissimilatory NO3 reduction to

+ - + NH4 (DNRA) by anaerobic bacteria reverts NO3 into NH4 , creating a feedback loop that retains N in an environment and enhances eutrophication (An and Gardner 2002,

- Giblin et al. 2013). In contrast, denitrification is the aerobic reduction of NO3 to N2 gas through linked dissimilatory transformations (with creation and reduction of nitrous oxide as an intermediary step). Denitrification results in the removal of biologically reactive N from aquatic habitats, so factors that control denitrification are a focus of research in eutrophic ecosystems.

Oyster biodeposits may enhance removal of N from the aquatic ecosystem by stimulating sediment denitrification (Newell 2004, Kellogg et al. 2014) by altering one or

- more of the three primary drivers of denitrification: organic C, NO3 , and favorable redox conditions (Herbert 1999, Philippot 2002). Oyster biodeposits may enhance the quality or

- quantity of sediment C, thereby driving denitrification of water column NO3 (i.e., direct

- denitrification). This is typical in well-mixed coastal habitats high in water column NO3

(Hoellein and Zarnoch 2014). In addition, biodeposits contain organic N, which can be

+ + mineralized to ammonium (NH4 ), and oysters excrete NH4 directly. If oyster-derived

+ - NH4 is nitrified, the NO3 produced may be subsequently denitrified (i.e., coupled

- + nitrification-denitrification). In areas with low water column NO3 relative to NH4 , this can be the dominant pathway for N2 production (Herbert 1999, Heiss et al. 2012, Smyth

2013, Kellogg et al. 2014). Coupled nitrification-denitrification requires adjacent anoxic- oxic microsites, which support close association of organisms and chemical substrates for each process. Enhancing N delivery to sediments and the removal of N from ecosystems

52 via denitrification may help mitigate effects of cultural eutrophication, but requires careful measurement of separate denitrification pathways (i.e., coupled vs. direct) and environmental controls on N cycling rates.

Hampshire, but found no effect at another reef. Comparisons of restored oyster reefs to oyster-free control sites at a reef in Chesapeake Bay showed significantly higher denitrification in the reefs (Kellogg et al. 2013). Finally, experimental enclosures of oysters at varying densities increased sediment organic matter, but had no effect on denitrification potential (Hoellein and Zarnoch 2014) or denitrification gene abundance

(S. Lindeman, unpublished data) in Jamaica Bay, NYC. In summary, oyster reefs may enhance denitrification, but the effect of oyster reefs on the underlying controls for various denitrification pathways require further study, especially for reefs in waters influenced by urbanization.

Oyster reef restoration in NYC

Prior to 1900, oysters were common throughout the Atlantic coast of the United

States, including New York Harbor (Levinton et al. 2013, Zu Ermengassen 2012,

Mackenzie et al. 1997). Unsustainable harvesting depleted the New York and New Jersey

53 fishery by the 1920’s (Franz 1982). Other contributions to oyster reef loss in the area include the creation of shipping passages and dredging (Mackenzie et al. 1997) as well as cultural eutrophication and disease (Levinton 2013).

Eastern oyster (Crassostrea virginica) reef restoration in NYC is supported by local and state government agencies. Reef restoration is ongoing at several locations including the Hudson River, Jamaica Bay, and East River. One primary objective for restoration is to improve water quality via oyster filtration and enhanced denitrification

Thesis research

In the previous chapter of this thesis, I examined nitrogen transformation rates at a recently restored oyster reef in New York City. By using a mass spectrometer to measure

N2 species in sediment cores, I found few differences between oyster reef sediments and a nearby control site due to high ambient nutrient concentrations and variation in samples. Although denitrification rates were not significantly different due to the reef, I was able to find a consistent pattern in denitrification pathways using 15N isotopes, where direct denitrification rates were consistently higher than coupled nitrification- denitrification rates. High nutrient concentrations and the similarity between reef and control sediments may have caused a lack of significant differences in field data.

Physicochemical characteristics at this restored oyster reef (e.g., unique hydrology and

54 tides, urbanization) make it difficult to find comparable control sites that account for all relevant variables. Finding an oligotrophic reference site with an oyster reef in New York

City is extremely unlikely, however these conditions could be replicated in a laboratory environment. The utilization of aquaria would allow me to determine potential effects of oyster reefs on denitrification rates in the presence of varying eutrophication levels while also reducing the amount of variation observed in the field.

In this study, I compared oyster reef effects on sediment N biogeochemistry in aquaria by manipulating sediment C and water column nutrient concentrations. Using artificial seawater and homogenized sediment from the field site in NYC, I built artificial

‘reefs’ in aquaria. I designed a fully crossed experiment to compare the effect of oysters on sediment N cycling under conditions mimicking eutrophication (i.e., high water column nutrients and high sediment organic matter) and under oligotrophic conditions

(i.e., no added water column nutrients and low sediment organic matter). I collected sediment cores from aquaria and analyzed N cycling using the same analytical approaches as the field study (Chapter 2).

I predicted that aquaria with oysters would have higher rates of denitrification than aquaria without oysters. Additionally, I predicted that eutrophic aquaria would have higher rates of denitrification than oligotrophic aquaria. I expected coupled nitrification- denitrification would be the dominant N pathway for denitrification in both oligotrophic and eutrophic aquaria. These results provide insight to N biogeochemistry in urbanized ecosystems, which has not been extensively studied. Results will inform management practices as well as assist oyster reef restoration efforts of the Hudson River Foundation and other stakeholders.

55 Methods

Aquarium experiment with oyster and eutrophication treatments

This experiment was carried out in glass aquaria (42 L, 26 cm x 32 cm x 51 cm) with 2 fully crossed treatments: oysters and eutrophication. The oyster treatment consisted of 6 aquaria with oysters (density 200 m-2) and 6 aquaria without oysters.

Aquaria oyster density was designed to match field densities observed in previous studies

(Mann et al. 2009). The eutrophication treatment consisted of 6 aquaria with organic-rich sediment collected from the study site at the East River, NYC, and water column

+ 3- - enrichments of ammonium (NH4 ), phosphate (PO4 ), and nitrate (NO3 ) at concentrations matching the study site. The low nutrient (i.e., oligotrophic) treatment was established in 6 aquaria, which received no organic matter or water column enrichment

(N=12 aquaria total). The four separate treatments in this study design are 1) oligotrophic without oysters, 2) oligotrophic with oysters, 3) eutrophic without oysters, and 4) eutrophic with oysters. The experiment was carried out in 3 separate trials of 4 aquaria each, where one of each treatment type was present in each trial. Each trial lasted 11 days, after which sediment was collected for measurements of N transformations via continuous-flow cores.

Sediment for the eutrophication treatment was collected from Soundview Park in the East River, New York City on October 3, 2012, November 26, 2012, and January 28

2013. Sediment was collected by shovel in a shell-less area several meters away from the site of the previous oyster reef field study in the East River. Sediment was kept on ice and shipped overnight. Immediately upon delivery, sediment was homogenized with a putty knife and mixed with small amount of artificial seawater (30 ppt, Instant Ocean, United

56 Pet Group, Blacksburg, VA) until it became a thick paste. Macroinvertebrates, rocks, shells, and garbage were removed from sediment if present. For each eutrophic aquarium,

I filled 5 rubber caps (8.3 cm diameter, 5 cm depth) with 3 cm of playground sand and 2 cm of East River sediment on top (not mixed together). For each oligotrophic aquarium, caps were filled with playground sand only. The 5 caps for each aquarium were then placed in foil pans (10 cm x 24 cm x 30 cm), which had been filled 8 cm deep with playground sand, so that the cap’s top surface was level with the surface of the sand in the foil pan. Pans covered approximately 60% of the bottom of aquaria, leaving room for water exchange during the 10-day incubation without disturbing sediments or caps.

Aquaria were slowly filled with artificial seawater, minimizing disturbance to sediments and caps. Each aquarium was inoculated with microbes from the study site. To do this, I mixed 100 mL of homogenized sediment with 100 mL of artificial seawater and slowly poured the slurry over the entire water surface of the aquaria. In eutrophic experimental tanks, I added a nutrient amendment based on measurements from the field study (Table 1). Bubblers were added to all aquaria.

For each trial replicate, I obtained fifty, 2-year-old live eastern oysters

(Crassostrea virginica) from Fisher’s Island Oyster Hatchery (Fisher’s Island, NY), following the precedent set by the Hudson River Foundation to use hatchery oysters when restoring NYC oyster reefs. Oysters were shipped overnight and refrigerated upon arrival. I randomly distributed individual oysters to the two experimental oyster aquaria.

Oyster density was set at 200 oysters m-2 (14 oysters aquarium-1), based on previous research (Mann et al. 2009) and for maximizing the number of oysters that could fit in the aquaria. Oysters were held above the sediment on a wire mesh stand (1.27 cm) fitted to

57 the top of each foil pan before placement in aquaria. Remaining oysters were placed in 2 separate reserve aquaria.

All aquaria (including non-oyster treatments), received 25 mL daily inputs of marine microalgae (Instant Algae Shellfish Diet 1800, Reed Mariculture, Campbell, CA) diluted to 1 L with artificial seawater. Algae were added to aquaria via a gravity-fed slow drip system made of 0.64 cm tubing and a plastic stopcock. Tubing released algae at the water surface over the bubblers, which dispersed algae throughout the aquarium. Oyster survivorship was monitored for 10 days. Any dead oysters were replaced with living oysters from the reserve tanks. Mortality was variable, but improved in sequential trials

(48% in trial 1, 10% in trial 2, and 0% in trial 3). Mortality in trial 1 was attributed to weekend air conditioning failure that also affected other concurrent experiments in the facility. However, all dead individuals were immediately replaced so the impact on sediment microbial processes was minimal. Artificial seawater was siphoned and replaced every 2 days. Nutrient amendments were added to eutrophic tanks every time aquaria water was changed.

Collection of sediment caps

After 10 days, all aquaria water was carefully siphoned from aquaria and oysters were removed and frozen. I removed the 5 sediment core caps from each foil pan. Acrylic cores (7.6 cm diameter, 30.48 cm length) were placed snugly inside the caps (N=3 per aquarium), creating a tight seal with minimal disturbance of sediment. Caps were sealed to acrylic cores with electrical tape and placed in a dark cooler. The contents of the remaining 2 caps from each aquarium were used for organic matter and C:N analysis (see below).

58 Flow-through measurements of nutrient and gas fluxes

Acrylic cores with sealed rubber caps filled with sediment were set up in continuous-flow incubations within 30 minutes of sampling (N=3 treatment-1). I gently filled each core with artificial seawater to a depth of 5 cm above the sediment water interface (SWI), a volume of approximately 227 mL, which minimized SWI disturbance.

A plunger with a rubber O-ring was placed into each core to create a seal. The plunger was plumbed with a Teflon inlet and outlet tubing. (Lavrentyev et al. 2000; An et al.

2001). I prepared twelve 20 L carboys of artificial seawater as water sources for the cores and added bubblers to the carboys. Aerated site water flowed into each core via a 16 channel peristaltic pump (model 205U, Watson Marlow Pumps Group, Falmouth,

Cornwall, UK) at a rate of 1.0 mL minute-1, and core outflows were collected in beakers.

I carefully removed any air bubbles from the core headspace at the beginning of the analysis, and cores were monitored for bubbles throughout. I covered cores with aluminum foil to prevent bubble formation via photosynthesis, although I observed no benthic algae or bubble formation inside cores. I conducted the entire analysis at room temperature.

I established three treatments in the artificial seawater that flowed into the cores:

15 15 + four cores received no N (control), four cores had added NH4 (final concentration 10

15 15 - 15 μM NH4-N), and four cores had added NO3 (final concentration 10 μM K NO3-N)

(Nielsen 1992, Gardner and McCarthy 2009). These isotope treatments enriched mean

14,15 + 14,15 - (±SE) inflowing water column nutrient concentrations of NH4 and NO3 by a factor of 2.47 (±0.44) and 1.66 (±0.14), respectively, in oligotrophic cores and by a factor of 1.33 (±0.07) and 1.23 (±0.05), respectively, in eutrophic cores. Water was passed

59 through the cores for 24 hours to establish steady-state environment prior to sample collection (Gardner and McCarthy 2009). Inflow and outflow water was then collected at

24, 48, and 72 hours after the start of the incubation (Bruesewitz 2013).

At each sample collection period, I collected samples for dissolved inorganic nutrients and gasses. Water samples for dissolved nutrients were filtered with 0.2 µm nylon syringe filters (Thermo Scientific, Rockwood, TN, USA) into triplicate 20 mL liquid scintillation vials (Wheaton, Millville, NJ, USA). Water samples were frozen until measurement of dissolved nutrients. Separately, I collected three replicate water samples for dissolved gas analyses. Water from each inflow and outflow were collected in 12 mL flat-bottomed vacutainers with screw caps (Labco Limited, Lampeter, UK). For this process, I filled each vial slowly from the bottom and allowed them to overflow for several volumes (Bruesewitz et al. 2013). Water samples for dissolved gas analysis were preserved with 200 µL of 50% zinc chloride and stored underwater and below room temperature until analysis of dissolved gasses (Kana et al. 1994, An et al. 2001, An and

Gardner 2002, McCarthy and Gardner 2003).

Sediment and oyster organic matter

After starting the continuous-flow measurements, I retrieved the 2 sediment core caps reserved for sediment analysis. From one cap, I transferred 270 ml of sediment to pre-ashed and weighed aluminum pans, dried samples at 60°C for a minimum of 48 hours, and measured dry mass. I combusted the samples at 550 °C for 4 hours, allowed samples to cool for at least 1 hour in a room temperature desiccator, then measured ash mass to calculate ash-free dry mass (AFDM). From the final cap, I placed 0.75 ml of sediment in three 2 mL microcentrifuge tubes (#02-681-266, Fisher Scientific, USA).

60 Samples were dried at 60C, acidified with two treatments of 25% HCl, re-dried at 60C

(Nieuwenhuize et al. 1994), and then measured with a Perkin Elmer 2400 Series II CHN analyzer (PerkinElmer, Waltham, Massachusetts, USA), as described by Hoellein and

Zarnoch (2014), to determine sediment % carbon and % nitrogen. Analyses of C:N in artificial reef sediments started by calculating the moles of C and N in each sample by multiplying the total sample weight by the % of C or N, then dividing by the atomic mass of C or N, respectively. Finally, I measured organic content of oyster visceral mass using the process described for sediment AFDM.

Dissolved solute and gas chemistry

Water chemistry was performed on a SEAL AutoAnalyzer 3 (Seal Analytical, Inc.,

Mequon, WI). I measured SRP using the ascorbic acid method (Murphy and Riley 1962),

+ - - - NH4 with the indophenol blue method (Solorzano 1969), and NO2 + NO3 (NOx ) using the sulfanilamide method with a cadmium reduction column. Samples were also tested

- with the sulfanilamide method but without the reduction column to measure NO2 alone,

- and NO3 was determined by difference (Greenburg 1985). We modified the sulfanilamide method by amending the ammonium chloride (NH4Cl) reagent with 0.5 g

L-1 of ethylenediaminetetraacetic acid (EDTA) buffer.

Dissolved gas analysis was performed using a membrane-inlet mass spectrometer

(MIMS, Bay Instruments, Easton, MD). For each sample, a peristaltic pump moved water from the bottom of the sample vial through a vacuum chamber, which pulled dissolved gasses out of solution across a membrane (Kana et al. 1994, An et al. 2001, Bruesewitz et al. 2013). MIMS plumbing includes a trap submerged in liquid N2 to prevent water vapor

28 29 30 contamination. Gasses measured on the MIMS include N2 isotopes ( N2, N2, and N2),

61 argon, and oxygen. We ran periodic standards of equal salinity at 24.5°C and 37°C to correct for instrument drift throughout the run (Kana et al. 1994, Hoellein and Zarnoch

2014). The quadrupole mass spectrometer in the MIMS can produce O+ ions that form

28 30 nitric oxide (NO) in the presence of N2, thereby affecting N2 and N2 measurements

(Eyre et al. 2002; Kana and Weiss 2004). The error magnitude is machine-specific

(McCarthy and Gardner 2003), but has not been measured for the machine used in this analysis.

Flux calculations

(McCarthy et al. 2008).

Multiple N2 gas transformations were calculated using the relative N2 fluxes

15 + 15 - across the three treatments (control, + NH4 , and + NO3 ). Net N2 flux was the balance of denitrification and N-fixation in control cores. N-fixation was calculated by using a

28,29,30 28,29,30 quadratic equation based on gross N2 denitrification and net N2 production in

15 - the NO3 cores (An et al. 2001). Denitrification (DNF) was calculated as the difference between N-fixation and net N2 flux in the control cores. Potential DNF, or the rate of

- denitrification in the presence of elevated NO3 , was measured as total N2 production

15 - 15 (accounting for N-fixation) in the NO3 cores. Finally, we followed the N from the

15 + 15 - 29,30 15 NH4 and NO3 into the isotopically labeled N2. Coupled DNF represented the N

62 29,30 15 + converted to N2 in cores that were amended with NH4 . Direct DNF represents the

15 29,30 15 - N converted to N2 in cores that were amended with NO3 . We note anammox can also contribute to isotope labeled N2, however, studies from nearby estuaries suggest it is a relatively minor contributor of total N2 production (i.e., 2-9% of denitrification;

Engström et al. 2005; Koop-Jakobsen and Giblin 2009). Finally, we acknowledge that denitrification rates do not account for incomplete denitrification to nitrous oxide.

Statistical analysis

For all fluxes, we first calculated the mean flux on each sampling day, and then calculated mean flux from each individual core by averaging across the 3 sampling days

(24, 48, and 72 hours post incubation). The three separate experimental trials were replicates for the statistical analysis, where N=3 for each treatment (oligotrophic without oysters, oligotrophic with oysters, eutrophic without oysters, eutrophic with oysters). I used a 2-way ANOVA with trophic level and oyster presence as factors to analyze patterns in dissolved solute flux, sediment organic matter, SOD, and N2 gas transformations (i.e., net N2 flux, N-fixation, denitrification, potential denitrification, coupled denitrification, direct denitrification), sediment organic matter, and sediment C and N content. We used t-tests to analyze patterns in oyster soft tissue organic matter.

Analyses were performed in Systat v.13 (Systat Software, Inc.) with p-values ≤0.05 as a threshold for significance.

Results

Physicochemical measurements

Physicochemical measurements including temperature, DO, and salinity were identical across the 3 sequential trials of the laboratory experiment and were similar to

63 mean seasonal values from the field experiment. As expected, water column nutrient concentrations of oligotrophic aquaria were much less than eutrophic aquaria (Table 8).

+ - However, mean concentrations of NH4 and NO3 were higher in eutrophic aquaria than in the East River by approximately 10 uM. In addition, water column SRP concentrations in both aquaria treatments were consistently higher than field concentrations. I did not

- include NO2 in the aquaria nutrient amendments, which was lower in aquaria than in the field study (Table 8).

Nutrient and gas fluxes

Fluxes of dissolved nutrients were generally positive (i.e., net flux out of the sediment) across oyster and trophic state treatments (Table 9). One exception to this pattern was in the eutrophic treatment with oysters, where sediment showed net uptake of

+ - - - NH4 , NO2 , and NO3 . Uptake of NO3 was significantly higher in eutrophic aquaria

(ANOVA, p=0.003), but the effect of oyster presence was only marginally significant

- (ANOVA p= 0.077). NO2 flux was not significantly affected by either trophic state or

+ oyster presence. For SRP and NH4 flux, there were significant effects of trophic state,

+ oysters, and oyster x trophic state interaction (ANOVA, SRP p=0.001, NH4 p<0.001,

Table 9).

All gas fluxes were significantly higher in eutrophic aquaria, and oysters had no effect on any fluxes (except direct DNF; Table 10). SOD was higher in eutrophic aquaria and was not affected by oyster presence (ANOVA, p=0.952, Table 10), and SOD in eutrophic aquaria was similar to mean SOD in the field study (Figure 10). Rates of DNF and potential DNF were also not affected by oyster presence, and potential DNF

(measured in cores amended with 15N) was consistently greater than DNF (Table 10,

64 Figure 11). Direct DNF was the only flux that was significantly higher when oysters were present (ANOVA, p=0.035; Table 10, Figure 12B). Direct DNF was also higher in high nutrient aquaria (ANOVA, p<0.001), and there was no significant interaction between oysters and nutrients (ANOVA p= 0.560). Coupled DNF was also not affected by oyster presence, and was lower than direct DNF across all treatments (Table 10, Figure 12).

This pattern is identical to that observed in the field study.

Sediment organic matter and oyster tissue biomass

As expected, there was more organic matter in eutrophic aquaria (ANOVA, p<0.001, Table 11, Figure 13), which also had a greater proportion of sediment C

(ANOVA, p<0.001) and N (ANOVA, p=0.001) relative to low nutrient aquaria (Table

11, Figure 14). Oyster presence had no effect on total organic matter (ANOVA, p=0.380,

Table 11); however, oysters significantly increased the relative sediment N content

(ANOVA, p = 0.029) and marginally increased the sediment C content (ANOVA, p =

0.071) (Table 11, Figure 14) in both high and low nutrient aquaria. Interaction between oyster presence and trophic level was not significant (ANOVA, % N p=0.081, % C p=0.177) (Table 11). Finally, oyster visceral mass averaged 1 g AFDM oyster-1 in all aquaria, with no difference in low and high nutrient aquaria (t-test, p=0.403, Figure 15).

65 Table 8. Mean (±SE) water column physicochemical values for oligotrophic and eutrophic aquaria in the laboratory experiment. Field measurements are the average water column nutrient concentrations in seasonal samples from the East River, New York City. + Abbreviations: DO = dissolved oxygen, SRP = soluble reactive phosphorus, NH4 = - - ammonium, NO2 = nitrite, and NO3 = nitrate. Experimental aquaria Measurement Units Oligotrophic Eutrophic Field Temperature ˚C 20.7 (0.2) 20.8 (0.2) 19 (3) Salinity ppt 30 (0) 30 (0) 25 (0) DO mg L-1 7.9 (0.1) 8 (0.1) 7 (1) SRP µM 5 (0.5) 16 (1) 2 (1) + NH4 µM 5 (1) 24 (1) 14 (8) - NO2 µM 0.4 (0.01) 0.4 (0.03) 2 (1) - NO3 µM 7 (1) 25 (0.2) 15 (6)

Table 9. Mean (±SE) nutrient fluxes in sediment cores taken from oligotrophic and eutrophic aquaria, with and without oysters. Units for fluxes are µmol of P or N m-2 h-1. P-values are from 2-way ANOVA by trophic state and oyster presence, with values ≤0.05 in bold. Abbreviations: TS = trophic state, OP = oyster presence, Int. = interaction of trophic state and oyster presence, SRP = soluble reactive + - - phosphorus, NH4 = ammonium, NO2 = nitrite, and NO3 = nitrate. Oligotrophic Eutrophic Control +Oysters Control +Oysters p-value Flux Mean SE Mean SE Mean SE Mean SE TS OP Int. SRP 20.6 2.1 111.9 16.7 6.9 6.2 0.4 5.7 <0.001 0.002 0.001 + NH4 88.2 1.5 118.4 17.7 53.4 4.6 -28.3 5.7 <0.001 0.028 <0.001 - NO2 0.2 0.9 -5.4 12.1 -5.2 2.9 -18.8 16.5 0.391 0.381 0.71 - NO3 92.9 29.7 4.1 30.3 27.4 10.2 -44.8 10.3 0.003 0.077 0.268

Table 10. Mean (±SE) sediment oxygen demand (SOD) and nitrogen gas (N2) fluxes in sediment cores taken from oligotrophic and eutrophic aquaria with and without oysters. Denitrification (DNF) was measured in control cores (no 15N added) and represents total N2 production after accounting for N-fixation (N-fixation was zero in all aquaria). Potential denitrification (Pot. DNF) was measured 15 15 - 15 29,30 in cores amended with N nitrate ( NO3 ). Coupled denitrification (Cpl. DNF) represents the N in N2 produced in cores 15 15 + 15 29,30 amended with N ammonium ( NH4 ). Direct denitrification (Dir. DNF) represents the N in N2 produced in cores that were 15 - amended with NO3 . P-values are from 2-way ANOVA by trophic state and oyster presence, with values ≤0.05 in bold. Abbreviations: TS = trophic state, OP = oyster presence, Int. = interaction of trophic state and oyster presence. Oligotrophic Eutrophic Control +Oysters Control +Oysters p-value Rate (Units) Mean SE Mean SE Mean SE Mean SE TS OP Int. -2 -1 SOD (µmol O2 m h ) 691 221 583 47 1079 62 1170 145 0.008 0.952 0.489 DNF (µmol 14N m-2 h-1) 8.3 4.2 6.1 2.8 62.4 1.6 57.1 21.6 0.001 0.891 0.743 Pot. DNF (µmol 14,15N m-2 h-1) 34 11 34 3 108 12 127 31 0.002 0.598 0.590 Cpl. DNF (µmol 15N m-2 h-1) 0.9 0.4 0.5 0.2 6.2 2.3 4.6 2.1 0.018 0.545 0.703 Dir. DNF (µmol 15N m-2 h-1) 4.0 1.7 24.7 3.8 50.5 11.5 63.2 5.1 <0.001 0.035 0.560

68 1400

1200 Avg E. River SOD (1136.3)

) 1000

-1

-2

2 800

mol O mol 600



SOD 400

200

0 Control +Oysters Control +Oysters Oligotrophic Eutrophic Figure 10. Mean (±SE) sediment oxygen demand (SOD) by trophic state and oyster presence. Fluxes are shown from control cores (i.e., no 15N addition to inflow water). Error bars represent standard error for 3 replicates. Reference line represents observed values from the field study.

Figure 11. Mean (±SE) denitrification (A) and potential denitrification (B) by trophic state and oyster presence. Fluxes in (A) are from control cores (no 15N addition) and fluxes in (B) are from cores amended with 15NO3-. Error bars represent standard error for 3 replicates.

80 A B

)

-1 60

-2

29,30 40

mol



DNF ( 20

0 Control +Oysters Control +Oysters Control +Oysters Control +Oysters Oligotrophic Eutrophic Oligotrophic Eutrophic Figure 12. Mean (±SE) coupled denitrification (A) and direct denitrification (B) by 15 + trophic state and oyster presence. Fluxes in (A) are from cores amended with NH4 and 15 - fluxes in (B) are from cores amended with NO3 . Error bars represent standard error for 3 replicates.

Table 11. Mean (±SE) oyster soft tissue ash-free dry mass (AFDM), aquaria sediment % organic matter (OM), aquaria sediment % carbon, and % nitrogen by trophic state and oyster presence. P-value for oyster AFDM is from t-test comparing oysters in each trophic state. P-values for sediment, %C and %N are from 2-way ANOVA by trophic state and oyster presence, with values ≤0.05 in bold. Abbreviations: TS = trophic state, OP = oyster presence, I = interaction of trophic state and oyster presence. Oligotrophic Eutrophic Control +Oysters Control +Oysters p-value Mean SE Mean SE Mean SE Mean SE TS OP I Oyster Tissue - - 0.82 0.35 - - 1.22 0.14 0.403 - - (g AFDM) Sediment OM (%) 0.32 0.12 0.37 0.08 1.68 0.17 1.84 0.06 <0.001 0.380 0.627 Sediment C (%) 0.03 0.01 0.09 0.04 0.52 0.17 0.88 0.10 <0.001 0.071 0.177 Sediment N (%) <0.01 <0.01 0.01 0.01 0.02 0.01 0.04 <0.01 0.001 0.029 0.081

2.0

1.5

1.0

% Organic% Matter

0.5

0.0 Control +Oysters Control +Oysters Oligotrophic Eutrophic Figure 13. Mean (±SE) organic matter from oligotrophic and eutrophic experimental aquaria. Error bars represent standard error for 3 replicates.

1.2 0.05 A B

1.0 0.04

0.8 0.03

0.6

0.02 0.4

Sediment % Carbon by Mass by Carbon % Sediment

0.01 Mass by Nitrogen % Sediment 0.2

0.0 0.00 Control +Oysters Control +Oysters Control +Oysters Control +Oysters Oligotrophic Eutrophic Oligotrophic Eutrophic

Figure 14. Mean (±SE) sediment % carbon by mass (A) and sediment % nitrogen by mass (B) of artificial reef sediment by trophic state and oyster presence. Error bars represent standard error for 3 replicates. Note the y-axis scale for %N is several orders of magnitude smaller than the y-axis for %C.

1.6

1.4

1.2

1.0

0.8

0.6

g AFDM / oyster / AFDM g

0.4

0.2

0.0 Oligotrophic Eutrophic

Figure 15. Mean (±SE) organic matter of oyster visceral mass from oligotrophic and eutrophic aquaria. Error bars represent standard error for 3 replicates. T-test, p=0.403.

73 Discussion

This experiment was designed to test how oyster presence and trophic state affect total denitrification and to determine the pathways used for N2 production in aquaria. As expected, eutrophic conditions increased denitrification, however, oyster presence only

- stimulated direct denitrification (i.e., denitrification of water column NO3 ), likely via enhanced organic matter quality. These data illustrate the underlying controls on sediment N dynamics that occur in oyster reefs, which are needed to predict where oyster restoration can be expected to provide the ecosystem service of enhanced denitrification in eutrophic ecosystems.

Eutrophic conditions increased denitrification

- Three primary factors control denitrification: redox conditions, NO3 availability, and sediment organic matter (Groffman et al. 1999, Herbert 1999). In the eutrophic aquaria, I influenced all three characteristics relative to the oligotrophic conditions via the

+ - addition of NH4 , NO3 , and sediment organic matter from the East River. As expected, denitrification rates were strongly affected by trophic state, regardless of oyster presence.

All measured forms of denitrification in this study were significantly increased by eutrophic conditions, including denitrification, potential denitrification (14,15N in cores

15 - 15 + amended with NO3 ), coupled denitrification (cores amended NH4 ), and direct

15 - denitrification (cores amended with NO3 ).

Contrary to my predictions, there was evidence of nutrient limitation of denitrification rates in both eutrophic and oligotrophic conditions. My prediction was that denitrification in the eutrophic aquaria would not be nutrient limited, given the N and C

15 - enrichments. However, denitrification rates increased with added NO3 in both

74 - oligotrophic and eutrophic aquaria (Table 10, Figure 11), suggesting NO3 limitation in

- both treatments. N-limitation in eutrophic aquaria is also supported by the greater NO3 fluxes into the sediment in eutrophic aquaria relative to oligotrophic aquaria, indicating

- higher sediment NO3 demand (Table 9). Additionally, although water column DO concentrations were the same in oligotrophic and eutrophic aquaria, sediment oxygen demand was significantly greater in eutrophic cores, which is also commonly correlated with higher denitrification (Kellogg et al. 2013, Piehler and Smyth 2011). Overall, the microbial communities in eutrophic and oligotrophic aquaria responded to changes in N availability with high N uptake and thus have the capacity to respond to the effect of oysters on sediment N and C pools.

Oysters increased direct denitrification in eutrophic and oligotrophic conditions

Under both oligotrophic and eutrophic conditions, oysters increased direct

15 - denitrification, or the amount of water column NO3 that entered the N2 pool, but the magnitude of change was different between treatments. In oligotrophic aquaria, oysters increased direct denitrification by a factor of 6.2 (i.e., from 4 to 24 µmol N m-2 h-1), but in eutrophic aquaria, oysters increased direct denitrification by a factor of 1.3 (from 50 to 63

µmol m-2 h-1). The pathways whereby oysters could affect direct denitrification include the organic matter in biodeposits and the altered hydrology around reefs due to oyster filtration.

Results suggested oysters increased sediment organic matter quality, as the relative abundance of C and N was higher in sediment exposed to oysters in both oligotrophic and eutrophic aquaria (Table 11, Figure 12). This is most likely the effect of oyster biodeposits (Gallagher 1988, Baldwin and Newell 1991, Baldwin and Newell

75 1995). Providing sediment denitrifiers with higher quality organic matter could

- increase denitrification of water column NO3 , especially in oligotrophic conditions where C is likely to be limiting. Contrary to expectations, however, oysters did not increase total sediment organic matter (Table 11, Figure 11). One explanation for this could be that measurements of organic matter included the entire volume of each core caps in the aquaria (8.3 cm diameter, 5 cm depth). It is possible that the oyster-mediated effect on organic matter only occurred only at the sediment surface (Hoellein and

Zarnoch 2014), and including the entire contents of the caps may have obscured any surface differences in organic matter between control and oyster aquaria. Additionally, the 10-day incubation period of aquaria trials may have been an insufficient amount of time in order to accumulate significant differences between control and oyster aquaria.

Oyster-enhanced direct denitrification could also be facilitated by their influence on aquarium hydrology. Anecdotally, I observed oyster aquaria had clearer water relative to aquaria without oysters, indicating a higher rate of algae and nutrient delivery from the water column to the sediment due to the filtration. This may have also increased

- movement of dissolved NO3 in the water column to sediment microbes. Larger reefs may alter in situ hydrology at the reef scale, although the effect on denitrification would depend on the oyster density and natural hydrodynamic patterns (e.g., tides, currents, and estuary inputs).

Other studies have found links between oyster filter-feeding and organic matter in sediments near reefs. Similar to results from the laboratory study, Hoellein et al. (2015) showed that sediments near reefs have higher quality and quantity of C than reef-distal sediments, which was correlating with higher oyster feeding and direct denitrification

76 rates. In addition, Kellogg et al. (2013) compared sediments in restored oysters reefs to control (no oyster sites) and found reefs had 15 times more total C and 11 times more organic C than control sites. Finally, Newell (2002) used aquaria experiments to show oysters decreased turbidity, and the organic matter removed from the water column by oysters increased benthic organic matter, N remineralization, and denitrification rates.

+ I expected oysters to increase coupled nitrification-denitrification via NH4 in biodeposits and oyster waste, however, coupled denitrification rates were consistently low and not affected by oysters. This conflicts with previous studies that found coupled denitrification to be the dominant N2 production pathway (Newell 2002, Smyth et al.

2013). In this study, the ratio between coupled nitrification and direct denitrification was highest (~23%) in the oligotrophic, no oyster aquaria (i.e., 0.9 and 4.0 µg N m-2 h-1, respectively; Figure 12) and 12% in the eutrophic, no oyster aquaria. In aquaria with oysters the ratio was ≤7%. Oyster aquaria likely limited nitrification by increasing oxygen demand via organic matter biodeposits at the benthic surface, limiting oxygen availability to nitrifying microbes.

Oysters had little effect on total denitrification, SOD, and nutrient fluxes

I predicted that oysters would have a greater effect on sediment gas and nutrient fluxes in oligotrophic aquaria, where oyster biodeposits would be more likely to alleviate

N and C limitation of microbial activity. However, while oysters affected the pathway of

15 - 29,30 direct denitrification ( NO3 to N2), oysters had little effect on total denitrification rates or sediment oxygen demand in both oligotrophic and eutrophic conditions. Potential explanations for these patterns include considerations of oyster density and reef complexity, incubation time, and the uniform nature of high-quality algal feedstocks.

77 The ‘artificial reef’ conditions in aquaria cannot account for some ecologically significant components of natural reefs, including variation in oyster density, demographics, and structural complexity. I set aquaria density set at 200 oysters m-2 based on studies that calculated reef density for the Mid-Atlantic (Mann et al. 2009,

Kellogg et al. 2014). In addition, the aquarium oysters were all the same age. Aquarium reef complexity is low compared to older, wild reefs with mixed age classes and greater abundance of shells and associated organisms. Physical complexity of structured habitat, such as oyster reefs, enhances denitrification relative to unstructured habitats, such as mudflats (Piehler and Smyth 2011, Kellogg et al. 2013, Hoellein et al. 2015). Future aquaria studies could include these elements by establishing treatments with different densities, population structures, and variation in shell abundance.

The incubation time in this experiment (10 days) was established based on oyster survivorship in a pilot study, and may play a role in the capacity to detect an effect of oysters on microbial N transformations. It’s possible that the effect of oysters on total organic matter abundance may require a greater duration of oyster exposure. In addition, denitrifying microbes may require more time to acclimate to changes in C and N pools.

The lack of difference in total denitrification rates may be due to eutrophic aquaria not having reached microbial carrying capacity. Future studies may avoid this complication by sampling aquaria over multiple time points to assess growth and succession of reef- associated biofilm communities and activities.

The capacity for oysters to affect sediment processes is mediated in part by feeding selection and variability in food stocks may drive oysters’ effect on sediment denitrification. In this study, I provided high quality mixed algae shellfish-feed to all

78 aquaria. Thus, the effects of oysters on sediment microbes via food selection (i.e., production of feces and pseudofeces) by oysters were minimized. Previous laboratory work examining effects of oysters on N transformation rates also used algal monoculture feeds (Newell 2002), however, these conditions do not reflect typical algal diversity in oligotrophic or eutrophic ecosystems. Field studies show higher seston and biodeposit quality correlates with higher direct denitrification rates (Hoellein et al. 2015). Future studies could evaluate the effect of oyster food selection on denitrification by using algal cultures of mixed nutritional quality.

Comparison of results to field study

Eutrophic aquaria were established to match conditions at a restored oyster reef in the East River, NYC. My measurements indicate physicochemical variables including temperature, salinity, and DO were consistent throughout the lab study and approximated field conditions (Table 8). In addition, nutrient and gas fluxes showed similar patterns between eutrophic oyster aquaria and rates observed in the field. For example, fluxes of

+ SRP and NH4 showed sediments were generally sources of nutrients into the water

+ column, except for eutrophic oyster aquaria, which acted as NH4 sinks (Table 9). In the cores collected adjacent to the restored oyster reefs in situ, sediments were also sources

- of SRP. Eutrophic aquaria and the cores from the field were both net sinks of NO3 .

Although the lab study showed marginally significant increases with oyster presence

- (Table 9, p=0.077), the field study showed no increase in NO3 uptake with oyster reef proximity (Table 3, p= 0.572). This difference could be due to experimental design, where field sediment cores were taken adjacent to reefs so as to not disturb the restored oyster reef, but lab aquaria suspended oysters directly above sediments. Finally, although

79 SOD was similar between lab and field data, denitrification rates (Table 10) were generally lower in the eutrophic aquaria than in the field.

Patterns in coupled nitrification-denitrification and direct denitrification were very similar in both the laboratory and field measurements. In both settings, direct denitrification was higher than coupled nitrification-denitrification. In the field, nitrification rates for both oyster reef-adjacent and distal cores were highly variable among replicate cores and seasons. This variability precluded a significant effect of oyster reef proximity on denitrification rates for the field project. In contrast, rates in the controlled aquaria environment were less variable among replicates, and showed oyster presence significantly influenced denitrification pathways by encouraging direct denitrification.

Expectations for reef restoration

Overall, the aquaria experiment suggests some role for oysters in denitrification of eutrophic habitats. In particular, oyster reef restoration may prove useful for promoting

- loss of water column NO3 . This point is critical for in restoration sites like the East

River, where harvest of N from oyster soft tissues and shells is discouraged or illegal, limiting the use of oyster harvest as a method of N removal. My data show the denitrifying ‘power’ of reef-associated microbes is likely restricted by surface area, oyster density, and seasonality. If paired with a method to increase nitrification or anammox rates, oyster reef restoration could be applied more broadly to mitigate N pollution in coastal ecosystems. In addition, restoration would best be utilized in conjunction with other nutrient mitigation strategies, techniques, or policy changes to limit the nutrients entering aquatic ecosystems.

LITERATURE CITED

An, S., Gardner, W. S., & Kana, T. (2001). Simultaneous measurement of denitrification and nitrogen fixation using isotope pairing with membrane inlet mass spectrometry analysis. Applied and environmental microbiology, 67(3): 1171- 1178.

An, S., & Gardner, W. S. (2002). Dissimilatory nitrate reduction to ammonium (DNRA) as a nitrogen link, versus denitrification as a sink in a shallow estuary (Laguna Madre/Baffin Bay, Texas). Marine Ecology Progress Series, 237(4).

Baker, S. M., & Mann, R. (1992). Effects of hypoxia and anoxia on larval settlement, juvenile growth, and juvenile survival of the oyster Crassostrea virginica. The Biological Bulletin, 182(2), 265-269.

Baldwin, B. S., & Newell, R. I. E. (1991). Omnivorous feeding by planktotrophic larvae of the eastern oyster Crassostrea virginica. Mar. Ecol. Prog. Ser, 78, 285-301.

Baldwin, B. S., & Newell, R. I. (1995). Relative importance of different size food particles in the natural diet of oyster larvae (Crassostrea virginica). Oceanographic Literature Review, 42(10).

Barber, B. J. (1999). A Guide to Bivalve Diseases for Aquaculturists in the Northeastern US. Sea Grant Maine/New Hampshire.

Beck, M. W., Brumbaugh, R. D., Airoldi, L., Carranza, A., Coen, L. D., Crawford, C., Defeo, O., Edgar, G. J., Hancock, B., Kay, M. C., Lenihan, H. S., Luckenback, M. W., Toropova, C. L., Zhang, G., & Guo, X. (2011). Oyster reefs at risk and recommendations for conservation, restoration, and management. Bioscience, 61(2), 107-116.

+ Blackburn, H.T. (1979). Method for measuring rates of NH4 turnover in anoxic marine 15 + sediments using a NH4 dilution technique. Applied and Environmental Microbiology 37, 760–765.

Branco, B. (2011). Unpublished data.

80 81 Bruesewitz, D. A., Tank, J. L., Bernot, M. J., Richardson, W. B., & Strauss, E. A. (2006). Seasonal effects of the zebra mussel (Dreissena polymorpha) on sediment denitrification rates in Pool 8 of the Upper Mississippi River. Canadian Journal of Fisheries and Aquatic Sciences, 63(5), 957-969.

Bruesewitz, D. A., Gardner, W. S., Mooney, R. F., Pollard, L., & Buskey, E. J. (2013). Estuarine ecosystem function response to flood and drought in a shallow, semiarid estuary: Nitrogen cycling and ecosystem metabolism. Limnology and Oceanography, 58, 2293-2309.

Camargo, J. A., & Alonso, Á. (2006). Ecological and toxicological effects of inorganic nitrogen pollution in aquatic ecosystems: a global assessment. Environment international, 32(6), 831-849.

Carnegie, R. B., & Burreson, E. M. (2011). Declining impact of an introduced pathogen: Haplosporidium nelsoni in the oyster Crassostrea virginica in Chesapeake Bay. Mar. Ecol. Prog. Ser, 432, 1-15.

Carpenter, S. R., Caraco, N. F., Correll, D. L., Howarth, R. W., Sharpley, A. N., & Smith, V. H. (1998). Nonpoint pollution of surface waters with phosphorus and nitrogen. Ecological applications, 8(3), 559-568.

Carriker, M. R. (1990). Functional significance of the pediveliget in bivalve development. In The Bivalvia: Proceedings of a Memorial Symposium in Honour of Sir Charles Maurice Yonge (Vol. 1, p. 267). Hong Kong University Press.

Coen, L. D., Brumbaugh, R. D., Bushek, D., Grizzle, R., Luckenbach, M. W., Posey, M. H., Powers, S. P., & Tolley, S. (2007). Ecosystem services related to oyster restoration. Marine Ecology Progress Series, 341, 303-307.

Coen, L. D. (2011). Managing oysters in South Carolina: a five year program to enhance/restore shellfish stocks and reef habitats through shell planting and technology improvements. Marine Resources Center, SCDNR, 2011.

Coen, L. D., & Luckenbach, M. W. (2000). Developing success criteria and goals for evaluating oyster reef restoration: ecological function or resource exploitation?. Ecological Engineering, 15(3), 323-343.

Coon, S. L., Walch, M., Fitt, W. K., Weiner, R. M., & Bonar, D. B. (1990). Ammonia induces settlement behavior in oyster larvae. The Biological Bulletin, 179(3), 297- 303.

82 Cornwell, J. C., Kemp, W. M., & Kana, T. M. (1999). Denitrification in coastal ecosystems: methods, environmental controls, and ecosystem level controls, a review. Aquatic Ecology, 33(1), 41-54.

Cranford, P. J., Ward, J. E., & Shumway, S. E. (2011). Bivalve filter feeding: variability and limits of the aquaculture biofilter. Shellfish aquaculture and the environment, 81-124.

Dame, R. F. (2012). Bivalve Filter Feeders: In Estuarine and Coastal Ecosystem Processes. Springer Publishing Company, Incorporated.

Dame, R. F., Spurrier, J. D., & Wolaver, T. G. (1989). Carbon, nitrogen and phosphorus processing by an oyster reef. Marine ecology progress series. Oldendorf, 54(3), 249-256.

Devol, A. H. (2003). Nitrogen cycle: solution to a marine mystery. Nature, 422(6932), 575-576.

Diaz, R. J., & Rosenberg, R. (2008). Spreading dead zones and consequences for marine ecosystems. science, 321(5891), 926-929.

Engström, P., Dalsgaard, T., Hulth, S., & Aller, R. C. (2005). Anaerobic ammonium oxidation by nitrite (anammox): Implications for N< sub> 2 production in coastal marine sediments. Geochimica et Cosmochimica Acta, 69(8), 2057-2065.

Ewart, J. W., & Ford, S. E. (1993). History and impact of MSX and Dermo diseases on oyster stocks in the northeast region. Northeastern Regional Aquaculture Center, University of Massachusetts Dartmouth.

Eyre, B. D., Rysgaard, S., Dalsgaard, T., and Christensen, P. B. (2002). Comparison of isotope pairing and N2:Ar methods for measuring sediment denitrification rates - Assumptions, modifications and implications. Estuaries 25:1077–1087.

Fitzgerald, A.M. (2013). Unpublished data.

Francis, C. A., Beman, J. M., & Kuypers, M. M. (2007). New processes and players in the nitrogen cycle: the microbial ecology of anaerobic and archaeal ammonia oxidation. The ISME journal, 1(1), 19-27.

Franz, D. R. (1982). An historical perspective on mollusks in Lower New York Harbor, with emphasis on oysters. Ecological Stress and the New York Bight: Science and Management. Columbia SC: Estuarine Research Federation, 181-197.

83 Gallagher, S. M. (1988). Visual observations of particle manipulation during feeding in larvae of a bivalve mollusc. Bulletin of Marine Science 43, 344-365.

Galloway, J. N., & Cowling, E. B. (2002). Reactive nitrogen and the world: 200 years of change. AMBIO: A Journal of the Human Environment, 31(2), 64-71.

Gardner, W. S., Yang, L., Cotner, J. B., Johengen, T. H., & Lavrentyev, P. J. (2001). Nitrogen dynamics in sandy freshwater sediments (Saginaw Bay, Lake Huron). Journal of Great Lakes Research, 27(1), 84-97.

Gardner, W. S., McCarthy, M. J., An, S., Sobolev, D., Sell, K. S., & Brock, D. (2006). Nitrogen fixation and dissimilatory nitrate reduction to ammonium (DNRA) support nitrogen dynamics in Texas estuaries. Limnology and Oceanography, 51(1), 558-568.

Gardner, W. S., & McCarthy, M. J. (2009). Nitrogen dynamics at the sediment–water interface in shallow, sub-tropical Florida Bay: why denitrification efficiency may decrease with increased eutrophication. Biogeochemistry, 95(2-3), 185-198.

Giblin, A. E., Tobias, C. R., Song, B., Weston, N., Banta, G. T., & Rivera-Monroy, V. H. (2013). The importance of dissimilatory nitrate reduction to ammonium (DNRA) in the nitrogen cycle of coastal ecosystems.

Gornitz, V., Couch, S., & Hartig, E. K. (2001). Impacts of sea level rise in the New York City metropolitan area. Global and Planetary Change, 32(1), 61-88.

Grabowski, J. H., & Peterson, C. H. (2007). Restoring oyster reefs to recover ecosystem services. Theoretical ecology series, 4, 281-298.

Greenberg, A. (1985). Standard methods for the examination of water and wastewater (16th ed.). Washington, DC: APHA.

Grizzle, R., Ward, K., Lodge, J., Suszkowski, D., Mosher-Smith, K., Kalchmayr, K., & Malinowski, P. (2013). Oyster Restoration Research Project (ORPP) Final Technical Report. ORRP Phase 1: Experimental Oyster Reef Development and Performance Results. US Army Corps of Engineers—New York District, The Port Authority of New York & New Jersey, New York-New Jersey Harbor & Estuary Program.

Groffman, P. M., Holland, E. A., Myrold, D. D., Robertson, G. P., XiaoMing, Z., Coleman, D. C., Bledsoe, C. S., & Sollins, P. (1999). Denitrification. Standard soil methods for long-term ecological research., 272-288.

84 Gutiérrez, J. L., Jones, C. G., Strayer, D. L., & Iribarne, O. O. (2003). Mollusks as ecosystem engineers: the role of shell production in aquatic habitats. Oikos, 101(1), 79-90.

Halpern, B. S., Walbridge, S., Selkoe, K. A., Kappel, C. V., Micheli, F., D'Agrosa, C., Bruno, J. F., Casey, K. S., Elbert, C., Fox, H. E., Fujita, R., Heinemann, D., Lenihan, H. S., Madin, E. M. P., Perry, M. T., Selig, E. R., Spalding, M., Steneck, R., & Watson, R. (2008). A global map of human impact on marine ecosystems. Science, 319(5865), 948-952.

Hatcher, A., Grant, J., & Schofield, B. (1994). Effects of suspended mussel culture (Mytilus spp.) on sedimentation, benthic respiration and sediment nutrient dynamics in a coastal bay. Marine Ecology-Progress Series, 115, 219-219.

Haven, D. S., & Morales-Alamo, R. (1966). Aspects of biodeposition by oysters and other invertebrate filter feeders. Limnol. Oceanogr, 11(4), 487-498.

Heiss, E. M., Fields, L., & Fulweiler, R. W. (2012). Directly measured net denitrification rates in offshore New England sediments. Continental Shelf Research, 45, 78-86.

Heisterkamp IM, Schramm A, de Beer D, Stief P (2010). Nitrous oxide production associated with coastal marine invertebrates. Mar Ecol Prog Ser 415:1–9

Herbert, R. A. (1999). Nitrogen cycling in coastal marine ecosystems. FEMS microbiology reviews, 23(5), 563-590.

Higgins, C. B., Tobias, C., Piehler, M. F., Smyth, A. R., Dame, R. F., Stephenson, K., & Brown, B. L. (2013). Effect of aquacultured oyster biodeposition on sediment N2 production in Chesapeake Bay. Mar Ecol Prog Ser, 473, 7-27.

Hoellein, T. J., Arango, C. P., & Zak, Y. (2011). Spatial variability in nutrient concentration and biofilm nutrient limitation in an urban watershed. Biogeochemistry, 106(2), 265-280.

Hoellein, T. J., Zarnoch, C. B., & Grizzle, R. E. (2015). Eastern oyster (Crassostrea virginica) filtration, biodeposition, and sediment nitrogen cycling at two oyster reefs with contrasting water quality in Great Bay Estuary (New Hampshire, USA). Biogeochemistry, 122(1), 113-129.

Hoellein, T. J., & Zarnoch, C. B. (2014). Effect of eastern oysters (Crassostrea virginica) on sediment carbon and nitrogen dynamics in an urban estuary. Ecological Applications, 24(2), 271-286.

85 Howarth, R. W., and Marino, R. (2006). Nitrogen as the limiting nutrient for eutrophication in coastal marine ecosystems: Evolving views over three decades. Limnology and Oceanography, 51(1) 364-376.

Jackson, J. B. C., Kirby, M. X., Berger, W. H., Bjorndahl, K. A., Botsford, L. W., Bourque, B. J., et al. (2001). Historical overfishing and the recent collapse of coastal ecosystems. Science 293, 629–638.

Jones, C., Branosky, E., Selman, M., & Perez, M. (2010). How nutrient trading could help restore the Chesapeake Bay. World Resources Institute Working Paper.

Kana, T. M., Darkangelo, C., Hunt, M. D., Oldham J. B., Bennett, G. E., and Cornwell, J.C. (1994). Membrane inlet mass spectrometer for rapid high-precision determination of N2, O2, and Ar in environmental water samples. Analytical Chemistry 66: 4166–4170.

Kana, T.M., and D.L. Weiss. (2004). Comment on “Comparison of isotope pairing and N-2: Ar methods for measuring sediment denitrification”. By Eyre, B.D., Rysgaard, S., Daisgaard, T., and P. Bondo Christensen. 2002. Estuaries 25: 1077– 1087. Estuaries 27:173–176.

Kellogg, M. L., Cornwell, J. C., Owens, M. S., & Paynter, K. T. (2013). Denitrification and nutrient assimilation on a restored oyster reef. Mar Ecol Prog Ser, 480, 1-19.

Kellogg, M. L., Smyth, A. R., Luckenbach, M. W., Carmichael, R. H., Brown, B. L., Cornwell, J. C., Piehler, M. F., Owens, M. S., Dalrymple, D. J., & Higgins, C. B. (2014). Reprint of: Use of oysters to mitigate eutrophication in coastal waters. Estuarine, Coastal and Shelf Science.

Kemp, W. M., Sampou, P., Caffrey, J., Mayer, M., Henriksen, K., & Boynton, W. R. (1990). Ammonium recycling versus denitrification in Chesapeake Bay sediments. Limnology and Oceanography, 35(7), 1545-1563.

Kennish, M. J. (2002). Environmental threats and environmental future of estuaries. Environmental conservation, 29(01), 78-107.

Koop-Jakobsen, K., & Giblin, A. E. (2009). Anammox in tidal marsh sediments: the role of salinity, nitrogen loading, and marsh vegetation. Estuaries and Coasts, 32(2), 238-245.

Koop-Jakobsen, K., & Giblin, A. E. (2010). The effect of increased nitrate loading on nitrate reduction via denitrification and DNRA in salt marsh sediments. Limnology and Oceanography, 55(2), 789.

86 Laursen, A. E., & Seitzinger, S. P. (2002). Measurement of denitrification in rivers: an integrated, whole reach approach. Hydrobiologia, 485(1-3), 67-81.

Lavrentyev, P. J., Gardner, W. S., & Yang, L. (2000). Effects of the zebra mussel on nitrogen dynamics and the microbial community at the sediment-water interface. Aquatic Microbial Ecology, 21(2), 187.

Levinton, J., Doall, M., & Allam, B. (2013). Growth and Mortality Patterns of the Eastern Oyster Crassostrea virginica in Impacted Waters in Coastal Waters in New York, USA. Journal of Shellfish Research, 32(2), 417-427.

Lin, X., McCarthy, M. J., Carini, S. A., & Gardner, W. S. (2011). Net, actual, and + potential sediment–water interface NH4 fluxes in the northern Gulf of Mexico + (NGOMEX): Evidence for NH4 limitation of microbial dynamics. Continental Shelf Research, 31(2), 120-128.

Lindemann, S., Castignetti, D., Zarnoch, C. B., and T. J. Hoellein. (2015). Effect of eastern oysters (Crassostrea virginica) and seasonality on nitrite reductase gene abundance (nirS, nirK, nrfA) in an urban estuary. Manuscript submitted for publication. Estuaries and Coasts.

Lowther, A. (2013). Fisheries of the United States 2012. National Oceanic and Atmospheric Administration.

Lund E. J. (1957) Self silting by the oyster and its significance for sedimentation geology. Publications of the Institute of Marine Science 4:320 – 327

Lyles, C. H. (1969). Historical catch statistics (shellfish). U.S. Department of the Interior, Fish and Wildlife Service, Curro Fish. Stat. 5007, 116 p.

MacKenzie Jr., C. L. (1996). History of oystering in the United States and Canada, featuring the eight greatest oyster estuaries. Marine Fisheries Review 58(4), 1-78.

MacKenzie Jr, C. L., Burrell Jr, V. G., Rosenfield, A., & Hobart, W. L. (1997). The History, Present Condition, and Future of the Molluscan Fisheries of North and Central America and Europe: Volume 2, Pacific Coast and Supplemental Topics.

McCarthy, M. J., Gardner, W. S., Lavrentyev, P. J., Moats, K. M., Jochem, F. J., & Klarer, D. M. (2007). Effects of hydrological flow regime on sediment-water interface and water column nitrogen dynamics in a Great Lakes coastal wetland (Old Woman Creek, Lake Erie). Journal of Great Lakes Research, 33(1), 219- 231.

87 McCarthy, M. J., McNeal, K. S., Morse, J. W., & Gardner, W. S. (2008). Bottom-water hypoxia effects on sediment–water interface nitrogen transformations in a seasonally hypoxic, shallow bay (Corpus Christi Bay, TX, USA). Estuaries and Coasts, 31(3), 521-531.

McCarthy, M. J., & Gardner, W. S. (2003). An application of membrane inlet mass spectrometry to measure denitrification in a recirculating mariculture system. Aquaculture, 218(1), 341-355.

McGlathery, K. J., Sundback, K., & Anderson, I. C. (2007). Eutrophication in shallow coastal bays and lagoons: the role of plants in the coastal filter. Marine Ecology Progress Series, 348, 1-18.

Murphy, J., & Riley, J. P. (1962). A modified single solution method for the determination of phosphate in natural waters. Analytica chimica acta, 27, 31-36.

Nestlerode, J. A., Luckenbach, M. W., O’Beirn, F. X. (2007). Settlement and Survival of the Oyster Crassostrea virginica on Created Oyster Reef Habitats in Chesapeake Bay. Restoration Ecology, 15(2), 273–283.

Newbold, J. D., Bott, T. L., Kaplan, L. A., Dow, C. L., Jackson, J. K., Aufdenkampe, A. K., Martin, L. A., Van Horn, D. J., and Long, A. A. (2006). Uptake of nutrients and organic C in streams in New York City drinking-water-supply watersheds. Journal of the North American Benthological Society, 25(4), 998-1017.

Newell, R. I. (1988). Ecological changes in Chesapeake Bay: Are they the result of overharvesting the American oyster, Crassostrea virginica. Understanding the estuary: advances in Chesapeake Bay research, 129, 536-546.

Newell, R. I., Cornwell, J. C., & Owens, M. S. (2002). Influence of simulated bivalve biodeposition and microphytobenthos on sediment nitrogen dynamics: A laboratory study. Limnology and Oceanography, 47(5), 1367-1379.

Newell, R. I. E. (2004) Ecosystem influences of natural and cultivated populations of suspension-feeding bivalve molluscs: a review. J Shellfish Res 23(1):51–61

Newell, Roger I. E., et al. "Influence of eastern oysters on nitrogen and phosphorus regeneration in Chesapeake Bay, USA." The comparative roles of suspension- feeders in ecosystems. Springer Netherlands, 2005. 93-120.

Newell, R. I. E., & Jordan, S. J. (1983). Preferential ingestion of organic material by the American oyster Crassostrea virginica. Marine ecology progress series. Oldendorf, 13(1), 47-53.

88 Newell, R. I. E., & Langdon, C. J. (1996). Mechanisms and physiology of larval and adult feeding. The Eastern Oyster Crassostrea virginica. Maryland Sea Grant, College Park, MD, 185-229.

Nielsen, L. P. (1992). Denitrification in sediment determined from nitrogen isotope pairing. FEMS Microbiology Ecology, 9(4), 357-361.

Nieuwenhuize, J., Maas, Y. E., & Middelburg, J. J. (1994). Rapid analysis of organic carbon and nitrogen in particulate materials. Marine Chemistry, 45(3), 217-224.

Nixon, S. W. (1995). Coastal marine eutrophication: a definition, social causes, and future concerns. Ophelia, 41(1), 199-219.

Nizzoli, D., Welsh, D. T., Fano, E. A., & Viaroli, P. (2006). Impact of clam and mussel farming on benthic metabolism and nitrogen cycling, with emphasis on nitrate reduction pathways. Marine Ecology Progress Series, 315, 151-165.

NRC. (2000). Clean coastal waters: Understanding and reducing the effects of nutrient pollution. National Academies Press.

O’Shea, M. L., & Brosnan, T. M. (2000). Trends in indicators of eutrophication in western Long Island Sound and the Hudson-Raritan estuary. Estuaries, 23(6), 877-901.

Parsons, T. R., Maita, Y., & Lalli, C. M. (1984). A manual of chemical and biological methods for seawater analysis. Pergamon press.

Philippot, L. (2002). Denitrifying genes in bacterial and archaeal genomes. Biochimica et biophysica acta (BBA)-Gene structure and expression, 1577(3), 355-376.

Piehler, M. F., & Smyth, A. R. (2011). Habitat-specific distinctions in estuarine denitrification affect both ecosystem function and services. Ecosphere, 2(1), art12.

Pomeroy, L. R., D'Elia, C. F., & Schaffner, L. C. (2006). Limits to top-down control of phytoplankton by oysters in Chesapeake Bay.

Riisgard, H. U. (1988). Efficiency of particle retention and filtration rate in 6 species of Northeast American bivalves. Marine Ecology Progress Series, 45, 217-223.

Risgaard-Petersen, N. (2003). Coupled nitrification–denitrification in autotrophic and heterotrophic estuarine sediments: on the influence of benthic microalgae. Limnology and Oceanography 48:93–105.

89 Rothschild, B., Ault, J. S., Goulletquer, P., & Heral, M. (1994). Decline of the Chesapeake Bay oyster population: a century of habitat destruction and overfishing. Marine Ecology Progress Series, 111, 29-39.

Seitzinger, S., Harrison, J. A., Böhlke, J. K., Bouwman, A. F., Lowrance, R., Peterson, B., Tobias, C., & Drecht, G. V. (2006). Denitrification across landscapes and waterscapes: a synthesis. Ecological Applications, 16(6), 2064-2090.

Smyth, A. R., Geraldi, N. R., & Piehler, M. F. (2013). Oyster-mediated benthic-pelagic coupling modifies nitrogen pools and processes. Marine ecology. Progress series, 493, 23-30.

Smyth, A. R., Thompson, S. P., Siporin, K. N., Gardner, W. S., McCarthy, M. J., & Piehler, M. F. (2013). Assessing nitrogen dynamics throughout the estuarine landscape. Estuaries and coasts, 36(1), 44-55.

Solorzano, Lucia. (1969). Determination of ammonia in natural waters by the phenolhypochlorite method. Limnol. Oceanogr 14.5, 799-801.

Strous, M., Fuerst, J. A., Kramer, E. H., Logemann, S., Muyzer, G., van de Pas- Schoonen, K. T., Webb, B., Kuenen, J. G. & Jetten, M. S. (1999). Missing lithotroph identified as new planctomycete. Nature, 400(6743), 446-449.

Svenningsen, N. B., Heisterkamp, I. M., Sigby-Clausen, M., Larsen, L. H., Nielsen, L. P., Stief, P., & Schramm, A. (2012). Shell biofilm nitrification and gut denitrification contribute to emission of nitrous oxide by the invasive freshwater mussel Dreissena polymorpha (Zebra Mussel). Applied and environmental microbiology, 78(12), 4505-4509.

Thamdrup, B., & Dalsgaard, T. (2002). Production of N2 through anaerobic ammonium oxidation coupled to nitrate reduction in marine sediments. Applied and environmental microbiology, 68(3), 1312-1318.

Turek, K.A., and T.J. Hoellein. (In press). The invasive Asian clam (Corbicula fluminea) increases sediment denitrification and ammonium flux in 2 streams in the Midwestern USA. Freshwater Science.

Vitousek, P.M., H.A. Mooney, J. Lubchenco and J.M. Melillo. (1997). Human domination of Earth's ecosystems. Science 277: 494-499.

Vitousek, P.M., K. Cassman, C. Cleveland, T. Crews, C.B. Field, N.B. Grimm, R.W. Howarth, R. Marino, L. Martinelli, E.B. Rastetter, and J.I. Sprent. (2002). Towards an ecological understanding of biological N fixation. Biogeochemistry 57: 1–45.

90 Vitousek, P. M., & Howarth, R. W. (1991). Nitrogen limitation on land and in the sea: how can it occur?. Biogeochemistry, 13(2), 87-115.

Walker, R. R., & Snodgrass, W. J. (1986). Model for sediment oxygen demand in lakes. Journal of Environmental Engineering, 112(1), 25-43.

Wallenstein, M. D., Myrold, D. D., Firestone, M., & Voytek, M. (2006). Environmental controls on denitrifying communities and denitrification rates: insights from molecular methods. Ecological Applications, 16(6), 2143-2152.

Walsh, C. J., Roy, A. H., Feminella, J. W., Cottingham, P. D., Groffman, P. M., & Morgan, R. P. (2005). The urban stream syndrome: current knowledge and the search for a cure. Journal of the North American Benthological Society, 24(3), 706-723.

Yianopoulos, G. M., & Anderson, W. D. (2003). Intertidal oyster restoration along an eroding shoreline: An assessment of substrate types for stabilization and propagation. Journal of Shellfish Research, 22(1), 361.

Zhang, L., Shen, Q., Hu, H., Shao, S. and Fan, C. (2011). Impacts of Corbicula fluminea on oxygen uptake and nutrient fluxes across the sediment-water interface. Water, Air, and Soil Pollution 220:399-411.

Zimmer-Faust, R. K., & Tamburri, M. N. (1994). Chemical identity and ecological implications of a waterborne, larval settlement cue. Limnology and Oceanography, 39(5), 1075-1087.

Zu Ermgassen, P. S., Spalding, M. D., Blake, B., Coen, L. D., Dumbauld, B., Geiger, S., Grabowski, J. H., Grizzle, R., Luckenbach, M., McGraw, K., Rodney, W., Ruesink J. L., Powers, S. P., & Brumbaugh, R. (2012). Historical ecology with real numbers: past and present extent and biomass of an imperilled estuarine habitat. Proceedings of the Royal Society B: Biological Sciences, 279 (1742), 3393-3400.

VITA

Michael Hassett was born and raised in southeast Michigan. He attended

Michigan State University in East Lansing, MI and earned a Bachelor of Science degree in Fisheries and Wildlife in 2009 with classwork focusing on fisheries and aquatic ecology. While an undergraduate, he was selected for a study abroad program in

Antarctica and a NSF-REU summer fellowship at Fordham University in Armonk, NY.

After graduating, he worked as an intern at the Arizona Game and Fish Department in

Pinetop, AZ and at the U.S. Fish and Wildlife Service’s Fish Technology Center in Warm

Springs, GA. At Loyola, he taught introductory biology laboratory classes in the

Department of Biology.