Citation
Past and Novel Approaches to Coastal Restoration

Material Information

Title:
Past and Novel Approaches to Coastal Restoration an Overview of Oyster Reef Restoration and a Hybrid Living Shoreline Technique
Creator:
Bersoza Hernandez, Ada Cecilia
Place of Publication:
[Gainesville, Fla.]
Florida
Publisher:
University of Florida
Publication Date:
Language:
english
Physical Description:
1 online resource (93 p.)

Thesis/Dissertation Information

Degree:
Master's ( M.S.)
Degree Grantor:
University of Florida
Degree Disciplines:
Environmental Engineering Sciences
Committee Chair:
ANGELINI,CHRISTINE
Committee Co-Chair:
KAPLAN,DAVID A
Committee Members:
FREDERICK,PETER C
PALMER,TODD
Graduation Date:
5/4/2018

Subjects

Subjects / Keywords:
artificial-reef -- biodiversity -- biofouling -- breakwall -- costs -- crassostrea-virginica -- ecological-engineering -- ecosystem-services -- habitat -- living-shoreline -- oyster-reef -- resilience -- shipworms -- teredo
Environmental Engineering Sciences -- Dissertations, Academic -- UF
Genre:
bibliography ( marcgt )
theses ( marcgt )
government publication (state, provincial, terriorial, dependent) ( marcgt )
born-digital ( sobekcm )
Electronic Thesis or Dissertation
Environmental Engineering Sciences thesis, M.S.

Notes

Abstract:
For centuries, humans have been drawn to coastal ecosystems because of the abundant ecosystem services they provide, such as productive fisheries and sheltered waterways. However, extensive development and intensive exploitation has led to high levels of ecosystem degradation reflected in habitat loss, collapsed fisheries, and loss of shoreline. In response, humans have been implementing a variety of restoration projects for decades in efforts to recover the ecosystem services that have been lost. Among these projects have been efforts to recover oyster reefs and prevent shoreline erosion. Here, I present a review of over 1700 oyster reef restoration projects implemented in the Atlantic and Gulf Coasts of the United States that show efforts have been concentrated in the Gulf and Chesapeake regions, have relied heavily on oyster-based substrates, and that the profitability of these projects has varied by material used and project size. This review is followed by the results of a field experiment looking at shipworm burrowing on four tree species across different distances from the sediment which suggests that infestations are more prevalent on the bottom 20 cm from the sediment and on tree species with lower wood densities. Findings from these first two projects serve to inform a novel living shoreline technique - a hybrid structure consisting of wooden breakwalls and oyster restoration structures designed to slow down shoreline erosion in high energy environments. ( en )
General Note:
In the series University of Florida Digital Collections.
General Note:
Includes vita.
Bibliography:
Includes bibliographical references.
Source of Description:
Description based on online resource; title from PDF title page.
Source of Description:
This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Thesis:
Thesis (M.S.)--University of Florida, 2018.
Local:
Adviser: ANGELINI,CHRISTINE.
Local:
Co-adviser: KAPLAN,DAVID A.
Electronic Access:
RESTRICTED TO UF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE UNTIL 2020-05-31
Statement of Responsibility:
by Ada Cecilia Bersoza Hernandez.

Record Information

Source Institution:
UFRGP
Rights Management:
Applicable rights reserved.
Embargo Date:
5/31/2020
Classification:
LD1780 2018 ( lcc )

Downloads

This item has the following downloads:


Full Text

PAGE 1

PAST AND NOVEL APPROACHES TO COASTAL RESTORATION: AN OVERVIEW OF OYSTER REEF RESTORATION AND A HYBRID LIVING SHORELINE TECHNIQUE By ADA CECILIA BERSOZA HERNNDEZ A THESIS PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE UNIVERSITY OF FLORIDA 2018

PAGE 2

2018 Ada Cecilia Bersoza Hernndez

PAGE 3

To my family

PAGE 4

4 ACKNOWLEDGMENTS I thank my advisor, Christine Angelini for giving me the opportunity to undertake these projects and for her continued guidance, trust, support, and kindness I would also like to thank my committee members David Kaplan, Peter Frederick, and Todd Palmer for their advice and help. A n enormous thank you is due to Sinead Crotty and Kimberly Pr in ce for their unconditional love and friendship In addition, I would like to thank my lab mates Sean Sharp, Audrey Batzer, Gregory Kusel, Daniel Gallagher, and Emma Johnson for their help with field and lab work. I would also like to thank everyone in my family for their constant encouragement throughou t the years Finally, I would like to acknowledge the University of Florida and the National Estuarine Research Reserve Science Collaborative for funding my work.

PAGE 5

5 TABLE OF CONTENTS page ACKNOWLEDGMENTS ................................ ................................ ................................ ............... 4 LIST OF TABLES ................................ ................................ ................................ ........................... 7 LIST OF FIGURES ................................ ................................ ................................ ......................... 8 LIST OF OBJECT S ................................ ................................ ................................ ....................... 10 LIST OF ABBREVIATIONS ................................ ................................ ................................ ........ 11 ABSTRACT ................................ ................................ ................................ ................................ ... 12 CHAPTER 1 INTRODUCTION ................................ ................................ ................................ .................. 13 2 RESTORING THE EASTERN OYSTER: HOW MUCH PROGRESS HAS BEEN MADE IN 53 YEARS OF EFFORT? ................................ ................................ .................... 16 Temporal and Regional Trends in Restoration ................................ ................................ ....... 18 Trends in Restoration Materials ................................ ................................ .............................. 21 Restoration Costs and Benefits ................................ ................................ ............................... 22 Restoring on Historical Scales ................................ ................................ ................................ 27 3 INTERACTIVE EFFECTS OF WOOD TRAITS AND INTERTIDAL EXPOSURE ON SHIPWORM INFESTATION IN SOUTHEASTERN U.S. ESTUARIES ............................ 35 Methods ................................ ................................ ................................ ................................ .. 37 Study Sites ................................ ................................ ................................ ....................... 38 Experiment 1: Tree Species, Branch Diameter, Elevation and Site Effects on Shipworm Infestation ................................ ................................ ................................ ... 38 Experiment 2: Regional Study of Tree Species, Elevation and Site Effects on Shipworm Infestation ................................ ................................ ................................ ... 41 Experiment 3: Anti Fouling Techniques for Wo oden Substrates ................................ ... 42 Statistical Analyses ................................ ................................ ................................ .......... 42 Results ................................ ................................ ................................ ................................ ..... 43 Biofouling ................................ ................................ ................................ ........................ 43 Experiment 1 ................................ ................................ ................................ ............ 43 Experiment 2 ................................ ................................ ................................ ............ 44 Experiment 3 ................................ ................................ ................................ ............ 44 Shipworm Damage ................................ ................................ ................................ .......... 45 Experiment 1 ................................ ................................ ................................ ............ 45 Experiment 2 ................................ ................................ ................................ ............ 46 Experiment 3 ................................ ................................ ................................ ............ 46

PAGE 6

6 Discussion ................................ ................................ ................................ ............................... 47 Implications for Wooden Structures in Coastal Environments ................................ .............. 52 4 A MANUAL FOR RE ENGINEERING LIVING SHORELINES TO HALT EROSION AND RESTORE COASTAL HABITAT IN HIGH ENERGY ENVIRONMENTS ............. 59 Project Planning ................................ ................................ ................................ ...................... 61 Design and I nstallation ................................ ................................ ................................ ........... 61 Breakwalls ................................ ................................ ................................ ....................... 61 Oyster Restoration Structures ................................ ................................ .......................... 62 Maintenance ................................ ................................ ................................ ............................ 63 Breakwalls ................................ ................................ ................................ ....................... 63 Oyster Restoration Structures ................................ ................................ .......................... 64 Monitoring ................................ ................................ ................................ .............................. 65 Shoreline Monitoring ................................ ................................ ................................ ....... 65 Breakwal l Monitoring ................................ ................................ ................................ ..... 66 Oyster Monitoring ................................ ................................ ................................ ........... 66 Observed condition ................................ ................................ ................................ ... 66 Vertical height of oysters ................................ ................................ ......................... 67 Percent cover of oysters and other taxa ................................ ................................ .... 67 Oyster density and size ................................ ................................ ............................. 67 Extensive Monitoring ................................ ................................ ................................ ...... 68 Shoreline ................................ ................................ ................................ ................... 68 Breakwalls ................................ ................................ ................................ ................ 69 Cost Consideration s ................................ ................................ ................................ ................ 69 5 CONCLUSIONS ................................ ................................ ................................ .................... 73 APPENDIX A LIST OF SOURCES FOR OYSTER RESTORATION DATABASE ................................ .. 76 B SIZE OF CONSTRUCTED OYSTER REEF AREA BY REGION ................................ ...... 82 C MONITORING SCHEDULE AND SAMPLE DATA SHEETS FOR LIVING SHORELINES PROJECT ................................ ................................ ................................ ...... 83 Monitoring Schedule ................................ ................................ ................................ .............. 83 Sample Data Sheets ................................ ................................ ................................ ................ 83 LIST OF REFERENCES ................................ ................................ ................................ ............... 86 BIOGRAPHICAL SKETCH ................................ ................................ ................................ ......... 93

PAGE 7

7 LIST OF TABLES Table page 2 1 Number of cost database entries as well as mean, minimum, and maximum costs per hectare of constructed reef for each material category. ................................ ..................... 29 4 1 Itemized list of the materials and their cost needed for the construction of the wooden breakwalls. ................................ ................................ ................................ ............ 70 A 1 Names and affiliations of restoration practitioners contacted by region. .......................... 76 A 2 List of databases accessed and the regions for which they had oyster restoration project data ................................ ................................ ................................ ......................... 79 A 3 List of publications detailing oyster restoration projects ................................ ................... 79 B 1 Average, minimum, and maximum sizes of constructed reefs in each region. .................. 82

PAGE 8

8 LIST OF FIGURES Figure page 2 1 Oyster reef area constructed in each region of the Atlantic and Gulf Coasts of the US .... 30 2 2 Examples of oyster r estoration materials ................................ ................................ ........... 31 2 3 Oyster reef area constr ucted by different substrates ................................ .......................... 32 2 4 Percentage of constructed reef area by each material type across study regio ns ............... 33 2 5 Return on investment by project size for 88 oyster restoration projects ............................ 34 3 1 Map of experimental sites in St. Augustine, FL, United States. ................................ ........ 53 3 2 Experimental design of ladders. Rungs represent tree branches positioned at the shown distances from the sediment, which is represented by the gray shading ................ 54 3 3 Average number of barnacles per branch at each distance from the sediment. ................. 55 3 4 Regression trees for experiment 1 ................................ ................................ ...................... 56 3 5 Average percent wood volume lost for small branches (b lack bars) and large branches (gray bars). Bars show mean and standard error for five replicates at each site, species, and diameter class ................................ ................................ ......................... 57 3 6 Palmer Drought Severity Index (PDSI) from July 2016 to September 2017, spanning the periods of experiments 1 and 2. ................................ ................................ ................... 58 3 7 Salinity data for the two experimental sites at the time periods corresponding to experime nts 1 and 2 ................................ ................................ ................................ ........... 58 4 1 Diagram of breakwall design. Each circle represents one fencepost. ................................ 71 4 2 Diagram of experimental set up relative to the shoreline showing breakwall placement in black and oyster restoration structures in gray. ................................ ............ 71 4 3 Photograph of oyster restoration structures. BESE are shown in the far left and in the middle. Gabions are shown second from the left and on the far right ............................... 72 4 4 Photograph showing full experimental setup. PVC poles on the right mark the initial shoreline position and the spots selected for ecologic al monitoring ................................ 72 C 1 Sample data sheet for monitoring wall height. Eight measurements are recorded for each wall on the seaward (front) and landward (back) sides. ................................ ............ 83 C 2 Sample data sheet for monitoring biofouling on walls. Six measurements are taken on the seaward (front) and landward (back ) sides. ................................ ............................ 83

PAGE 9

9 C 3 Sample data sheet for vegetation and community monitoring. ................................ .......... 84 C 4 Sample data sheet for percent ground cover. ................................ ................................ ..... 84 C 5 Sample data sheet for shoreline monitoring. DP indicates distance to the most seaward patch of remnant peat and DV indicates distance to most seaward vege tation. ................................ ................................ ................................ .......................... 85

PAGE 10

10 LIST OF OBJECTS Object page 2 1 Database of oyster reef restoration projects in the Atlantic and Gulf Coasts of the US (.xlsx file 152KB) ................................ ................................ ................................ .............. 18 2 2 Database of oyster reef restoration projects cost data and analyses (.xlsx file 83KB) ...... 23 2 3 List of all water bodies with historic reef area, degraded/lost reef area, and restored reef area data. ................................ ................................ ................................ ..................... 27

PAGE 11

11 LIST OF ABBREVIATIONS ANOVA Analysis of variance BESE Biodegradable Elements for Starting Ecosystems PVC Polyvinyl chloride ROI Return on investment

PAGE 12

12 Abstract of Thesis Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Master of Science PAST AND NOVEL APPROACHES TO COASTAL RESTORATION: AN OVERVIEW OF OYSTER REEF RESTORATION AND A HYBRID LIVING SHORELINE TECHNIQUE By Ada Cecilia Bersoza Hernndez May 2018 Chair: Christine Angelini Major: Environmental Engineering Sciences For centuries, humans have been drawn to coastal ecosystems because of the abundant ecosystem services they provide, such as productive fisheries and sheltered waterways However, extensive development and intensive exploitation has led to high levels of e cosystem degradation reflected in habitat loss, collapsed fisheries, and loss of shoreline. In response humans have been implementing a variety of restoration projects for decades in efforts to recover the ecosystem services that have been lost. Among the se projects have been efforts to recover oyster reefs and prevent shoreline erosion. Here I present a review of over 1700 oyster reef restoration projects implemented in the Atlantic and Gulf Coasts of the United States that show efforts have been concent rated in the Gulf and Chesapeake regions, have relied heavily on oyster b ased substrates, and that the profitability of these projects has varied by material used and project size This review is followed by the results of a field experiment looking at shipworm burrowing on four tree species across different distances from the sediment which suggests that infestations are more prevalent on the bottom 20 cm from the sediment and on tree species with lower wood densities Findings from these first two projects serve to inform a novel living shoreline technique a hybrid structure consisting of wooden breakwalls and oyster restoration structures designed to slow down shoreline erosion in high energy environments

PAGE 13

13 CHAPTER 1 INTRODUCTION For centuries, productive fisheries, sheltered waterways, and many other goods and services have drawn humans to settle near the coast (Worm et al. 2006). This development has been so intensive that n o pristine coastal marshes, seagrass meadows forests or reefs remain (Halpern et al. 2008). With over one third of the global population residing within 100 kilometers of the coast and 38 million people deriving employment from coastal fisheries and fish products (UNEP 2006, FAO 2014), it is clear that we continue to rely heavily on these ecosystems. However, centuries of human activity have put extreme pressure on these coastal habitats and have led to widespread ecosystem degradation (Lotze et al. 2006). Anthropogenic im pacts such as overexploitation pollution, and eutrophication have driven habitat and biodiversity loss es which, in turn, have compromised the ecosystem services coastal communities receive (Worm et al. 2006). Moving forward, i f coastal e cosystems and human s are to sustainably interact, it is vital to not only protect the natural ecosystems that remain but also to restore habitat that has been lost This widespread loss of coastal habitat has spurred many ecological restoration efforts, which can be defined as (SER 2004). These efforts typically involve an intentional intervention within an ecosystem that eliminates a disturbance and/or re introduces key native species, with the objective of achieving ecosystem health, integrity, and stability. Ultimately the goal of restoration is to obtain an ecosystem whose structure and function is stable, does not require any further interventions, and is resilient enough to withstand stresses and disturbances within normal limits (SER 2004). Restoration efforts have been carried out for decades and across a variety of ecosystems (e.g. seagrass beds, mangrove forests, coral reefs, Bayraktarov et

PAGE 14

14 al. 201 6), all with the goal of recovering enough ecosystem structure and function to support the ecosystem services on which we have grown so dependent. However, despite decades of restoration projects being implemented, there is still a lot that we do not know One important k nowledge gap includes the lack of sy nthesis of restoration efforts. For many ecosystems, there has been no comprehensive review of the projects, the techniques used for restoration, or the costs of different methods or techniques As such, it is difficult for other restoration practitioners to learn from past efforts and build upon what has already been done. To begin to address this knowledge gap, this thesis focuses on oyster reef restoration efforts and presents a synthesis of 1768 oyste r restoration projects implemented in the Atlantic and Gulf Coasts of the United States in the past 53 years. Through this review, the second chapter of this thesis identifies regional and temporal trends in oyster restoration effort and materials used. In addition, cost data were collected for 88 restoration projects and were used to estimate the profitability of different materials using return on investment calculations based on materials costs ecosystem service benefits gained. With these analyses, rest oration practitioners can make decisions for future projects (e.g. what materials to use or how big of an area to restore) based on what has historically been done and what has been most profitable. This effort to compile oyster restoration projects in a c omprehensive database should also act to encourage restoration practitioners to more widely and openly d isseminate the details (e.g. materials, restoration footprint, and costs) of the projects they are involved in. Another issue in restoration ecology is if one particular restoration method is successful in one area then it will also succeed anywhere else it is implemented (Hilderbrand et al. 2005). In efforts to prevent this from happening, the third chapter pre sents the findings of a field experiment designed to inform the design of wooden

PAGE 15

15 breakwalls built to aid in shoreline restoration. Specifically, chapter three looks at shipworm infestation patterns across different sites, tree species, intertidal heights, and branch diameters in Southeastern US estuaries Finally, results from the second and third chapters come together and are put into practice in the form of a manual for a brand new restoration technique presented in chapter four. This restoration method is a type of living shoreline a set of techniques used to prevent shoreline erosi on and recover coastal habitat where oyster restoration structures are used in conjunction with wooden breakwalls in order to slow down and perhaps even halt shoreline er osion in a Florida estuary. The oyster restoration component of this project draws from recommendations in chapter two that oyster restoration substrates be locally sourced, readily available, and easily deployed. In addition, breakwall maintenance regarding biofouling relies on findings from the shipworm experiment detailed in chapter three showing that the bottom 20 cm are the most vulnerable to infestations Together, these three chapters show an example of how current restoration can be informed by past efforts and by experimental ecology.

PAGE 16

16 CHAPTER 2 RESTORING THE EASTERN OYSTER: HOW MUCH PROGRESS HAS BEEN MADE IN 53 YEARS OF EFFORT? Of the many threatened coastal habitats oyster reefs are one of the most imperiled (Beck et al. 2011). Reef building oysters have b een harvested for centuries for example b y 1 st century R omans (Kuijper and Turner 1992), Native Americans 3200 years ago ( Rick et al. 2014), and European settlers in colonial America (Berry 2008). To meet growing demand, commercial harvesting began in the Middle Ages in Europe (Lotze et al. 2005) and in the early 1800s in the US (MacKenzie 1996), and intensif ied dramatically with the advent of dredging Due to prolonged and intensive harvesting o nce dominant species Crassostrea virginica, Ostrea lurida, and Ostreola conchaphila in North America (Kirby 2004, White et al. 2009), Ostrea edulis in Europe (Smyth et al. 2009), and Saccostrea glomerata in Australia (Kirby 2004) are considered overexploited and approximately 85% of oyster reefs have been lost worldwide (Beck et al 2011). Other stressors, such as disease, habitat destruction, and eutrophication, further limit persistence and recovery (Jackson 2008). This is of direct concern to humans becau se of the economic value of oysters; indeed, oyster landings generat e $1 9 0 million in annual revenue in the US alone ( NMFS 201 5 ) Oysters also play an important, sustaining e cological role as foundation species that protect shoreline s sequester carbon, enhance water quality and support fisheries by creating reef habitat ( Grabowski et al. 2005 Coen et al. 2007). Motivated by th ese economic and ecological value s there has been growing interest in restoring oyster s (Coen et al. 2007). N umerous studies have evaluated the f easibility of restoring the European oyster, Ostrea edulis in the United Kingdom ( e g Laing et al. 2006 Alleway and Connell 2015 ) and while initial attem pts to restock mudflats with native oyster s were unsuccessful and led to the introduction and subsequent invasion of the Pacific oyster, Crassostrea gigas in the Wadden Sea (Died e rich et al. 2005) there are renewed efforts to restore

PAGE 17

17 native oysters in tho se waters Similarly, significant e ffort ha s been made to rebuild Olympia oyster, Ostrea lurida reefs in Washington Oregon and California by planting oyster shell, as well as hatchery produced spat on shell cultch (Dinnel et al. 2009 McGraw 2009 ). Howe ver, no species has been the focus of more effort than the Eastern oyster, Crassostrea virginica P ractitioners have been restoring this species across the US Atlantic and Gulf C oasts for decades (Powers et al. 2009, Brown et al. 2014, La Peyre et al. 2004 a ). Because of the significant cost and scope of effort s dedicated to promoting this species recovery as well as its diminished reef habitat there is a need for synthesis of the scope and outcomes of restoration conducted across its range. Insights deriv ed from such large scale review can inform the design of future efforts and identify where additional investments in restoration may yield the greatest economic and ecological benefits Here we present a synthesis of Eastern o yster (hereafter, oyster) restoration and how the scale of this effort compares to the scale of historic oyster loss extends to Brazil, we focus on restoration implemented along the US Atlantic and Gulf Coasts because of the large number of projects initiated on these shorelines and a general lack of restoration outside the US ( Laing et al. 2006). W e assembled an oyster restoration project dataset by: searching Web of Science for peer reviewed studies using keywords accessing databases (e g restoration database), and contacting practitioners (project sources listed in Appendix A ). We then reviewed each project that had oysters as a focal species or identified oyster restoration as an added benefit (e g seawalls, breakwaters), defined a geographic location, and were completed or underway. We included only projects that met our i e those that deployed settlement substrate or planted live oysters

PAGE 18

18 specifically to enhance or establish reefs and support an expansion of local oyster populations. We excluded fisheries enhancement projects that deployed oyster shell or seed with the expli cit intent that this material and the oysters established on it would be subsequently harvested, and also excluded efforts to protect oysters through sanctuary or marine protected area delineation without additional substrate enhancement. In some states in cluding Maryland, Virginia and North Carolina, these actions are common strategies for managing oyster populations and supporting restoration; however, they are not the focus of this review. For the 1768 projects in the database, we recorded the project na me, location / water body, year the project started, d ata source, areal footprint (reported for 1178 projects), and substrate type (reported for 1437 projects, see project details in Object 2 1). Object 2 1. Database of oyster reef restoration projects in the Atlantic and Gulf Coasts of the US (.xlsx file 152KB) Here, we summarize: 1) how restoration effort and approaches have varied over time ; 2) how cost and estimated return on investment of projects differ with substrate used and project size; and 3) how much reef area has been restored relative to historic baselines. We explored each question within the five regions defined by NOAA fisheries classification system: the Northeast (ME, NH, MA, RI, CT) Mid Atlantic (NY, NJ, DE), Chesapeake (MD, VA), Southeast (NC, SC, GA, east coast of FL) and Gulf (west coast of F L AL, MS, LA, TX). Based on these analyses, we identify key areas in need of improvement in order for restoration to make greater strides i n rehabilitating oyster reefs and other coastal habitats in the 21 st century. Temporal and Regional Trends in R estoration Oyster restoration efforts meeting our definition began in 1964, 1992, and 1993 in the Gulf, Southeast, and Chesapeake, respectivel y, and since then a total of 5199 hectares of settlement substrate have been deployed according to the projects that report their size in our

PAGE 19

19 database (Fig. 2 1 A) Despite a slow start from the 1960s to 1990s effort ha s escalated with an average of 190 he ctares of reef built each year across the US since 2000 The majority of this effort ha s been focused in the Gulf and Chesapeake where 3186 and 1828 hectares of reef (61% and 35% of total area reported) have been constructed, respectively (Fig. 2 1B). All projects initiated before 1987 in the Gulf utilized repurposed materials, from recycled tires and scrap metal, as settlement substrate although the scale of these efforts is unknown because their sizes were not reported ( Furlong 2012 La Peyre et al. 2014a ). Between 1987 and 2005, restoration practitioners i n the Gulf primarily focused on enhancing reef generated shoreline stabilization, habitat provisioning, and water quality improvement services (Brown et al. 2014, La Peyre et al. 2014 b ) restored 14 hectares per year, on average, with the exception of 1995 when four Alabama projects collectively constructed 1214 hectares of reef. Since the mid 2000s, oyster restoration has further accelerated across the Gulf following the passing of The American Recov ery and Reinvestment Act (2009) and The Restore Act (2012), policies that have made significantly more funding available for coastal restoration. Similar to the Gulf, restoration in the Chesapeake began relatively slowly before accelerating in the mid 200 0s (Fig. 2 1A). These restoration activities that left substrate in place to rejuvenate and sustain overfished reefs marked a new strategy in the region, where coastal resource managers since the 1920s had been implementing oyster repletion and broodstock enhancement programs whose purpose was to support commercial oyster fisheries by deploying substrate that would ultimately be harvested (Southworth and Mann 1998, Schulte 2017). Through many, generally large scale projects (average project size between 199 9 2016: 2.85 hectares see Appendix B for project size data by region), practitioners have steadily increased constructed reef area in the Chesapeake over the last quarter century (Fig. 2 1).

PAGE 20

20 In the Southeast, the total reef area restored or established is less than in the Gulf and Chesapeake, and the focus on restoration has been slower to accelerate (Lenihan and Peterson 1998, Powers et al. 2009, Kingsley Smith et al. 2012). This may reflect a smaller historic area of reef and commensurately smaller emp hasis on oyster rehabilitation relative to other regions. However, practitioners have been constructing an average of 7 hectares of reef annually across the Southeast since 1999, much of which has been distributed across many smaller scale shoreline stabil ization projects. Recently, effort in this region has surged following the passing of the American Recovery and Reinvestment Act (Powers et al. 2009, Appendix B ). In contrast, our database indicates only 22 and 19 hectares of non harvest reef have been co nstructed to date north of Chesapeake Bay in the Mid Atlantic and Northeast, respectively, (Fig. 2 1B). This reduced activity can be partially attributed to concern that re establishing reefs in these heavily populated regions poses a risk to those who may harvest oysters from polluted waters (Martin 2010). Because oyster recruitment is generally low and/or episodic, simply deploying substrate is not expected to be sufficient to recover oyster populations or their ecological functions in these regions (Griz zle and Ward 2016). Instead, positioning substrates with seed oysters on historic reef locations is projected to be a critical, though not well tested (Geraldi et al. 2013), strategy for re establishing oysters in their northern range (Yozzo et al. 2004). However, uncertainty about historic reef locations impedes the identification of such restoration sites in many Northeast estuaries (Larsen et al. 2013). Despite these challenges, investment in feasibility studies to identify where restoration may succeed is increasing (e g Project and the Mass Oyster Project to re establish reefs across the Northeast and Mid Atlantic.

PAGE 21

21 Trends in Restoration M aterials A variety of materials, which we categorize into six substrate types (Fig. 2 2 ), have been used in restoration projects where settlement substrate is thought to be limiting oyster recovery Of these, the most commonly used has been oyster shell (recycled and fossilized ), which accounts for 1173 of the 3390 (34%) hectares of constructed reef area for which we have substrate information and is typically deployed as loose shell, shell bags, shells attached to plastic mats, spat on shell, and through gardening programs in w hich homeowners grow oysters in cages and then transp lant them onto restored reefs (Fig. 2 3, 2 4). Although oyster shell has been moderately utilized across all regions, the Chesapeake has been particularly reliant on it (81% of constructed reef area, Fig 2 4D) Rising costs, insufficient quantities, and limited sources of recycled and fossilized shell as well as the vulnerability of this material to dislodgement or burial constrain its use in many locations (La Peyre et al. 2014b). In addition, shell susceptibility to boring sponge infestation and other degradation processes has prompted some practitioners to incorporate other substrates in restoration designs (Powell et al. 2006). In particular, mixed oyster substrates that combine oyster shell with other types of more readily available shell (e g surf clam, whelk) or more durable materials, such as limestone or granite, have become increasingly popular in the last decade and currently account for 35% of total constructed reef area ( Fig. 2 3A 2 4A ). In contrast, concrete and mixed concrete substrates those combining concrete with other materials such as limestone or crab traps have been the least employed substrates, accounting for only 1.5% and 2.4% of the total constructed reef area, respective ly ( Fig. 2 3, 2 4A). C oncrete based projects have been implemented most commonly in the Southeast (26% of project area: 11% concrete, 15% mixed concrete), largely due to widespread deployment of nservancy South Carolina

PAGE 22

22 chapter (Fig. 2 4E ) Despite its durability and ability to be cast in shapes designed to f acilitate oyster establishment (e g ReefBall s, Fig. 2 2B ) or recycled from demolished infrastructure, high manufacturing and tran sportation costs often limit its use Another approach practitioners have taken is to combine the durability of concrete with the habitat complexity technique accoun t for 2.4% of total constructed reef area and never account for more than 5% of reef area in any region (Fig. 2 3B, Fig 2 4). Finally, we classified reefs that did not employ oyster or concrete substrates alone or in granite (Powers et al. 2009, Brown et al. 2014). These projects comprise 9.8% of total reef area and are used primarily in the Southeast and Gulf where they account for 32% and 13% of constructed reef area, respectively. Probably due to their low cost and high availability relative to oyster and concrete, the use of other substrates has more than quadrupled from 4.8 to 24 hectares per year prior to and post 2013 Restoration C osts and B enefits To establish baseline information about the cost of oyster restoration, we asked practitioners to share data on projects for which they had detailed records of incurred construction, planning, permitting, labor, and monitoring expenses. Reflecting the vast va riability in approaches to establishing reefs, all of these cost categories tend to differ among projects based on substrate accessibility, practitioner experience, volunteer participation, project siting (e g subtidal versus intertidal), project goals (e g research and restoration versus restoration only), and other factors. Since construction, which we define as the cost of substrate and its placement on site, was the only cost consistently reported across all 88 projects in our dataset (e g some proje cts utilized volunteers while others paid staff, contractors and/or

PAGE 23

23 consultants for labor), we report this value, standardized to be the cost of constructing one hectare of reef in 2016 USD, as a representative measure of restoration costs (see Object 2 2 for project cost details) Object 2 2. Database of oyster reef restoration projects cost data and analyses (.xlsx file 83KB) It must be acknowledged that this database is by no means comprehensive in coverage due to the difficulty in obtaining records of project costs and that project designs influence the areal costs reported in our database for instance, higher relief subtidal reefs often require larger volumes of substrate compared to intertidal r eefs and therefore tend to cost more to restore per hectare Nonetheless, these areal construction cost estimates enable us to provide a first order assessment of the cost effectiveness of different restoration approaches and should act to encourage practi tioners and funding agencies to compile and share information on how project funds are allocated to identify cost saving opportunities ( e g Westby et al. 2016) With these considerations in mind, oyster restoration construction costs ranged widely from $ 3,826 $2,180,361 hectare 1 The average cost of oyster restoration, $299,999 hectare 1 is 4 times higher than mangrove restoration ($69,387 hectare 1 ) and approximately 20 times lower than coral reef restoration ($5,990,208 hectare 1 values from Bayrakta rov et al. 2016, all reported in 2016 USD). On average, concrete has been the most expensive ($1,225,579 hectare 1 ) and mixed oyster substrates the least expensive ($120,166 hectare 1 ) material type (Table 2 1). According to our database, over $23 million were invested to build these 88 projects. To evaluate the potential yield of these investments, it is necessary to not only quantify restoration costs but also the health of resultant reefs and their production of ecosystem services. Grabowski et al. (201 2) estimated ecosystem service benefits derived from oyster reefs to vary between $5500 and $99,000 hectare 1 year 1 depending on reef location and which services, and at what level, are achieved. Assuming $10,325 hectare 1 year 1 as the average annual value of ecosystem

PAGE 24

24 service benefits derived from restored reefs starting one year post construction, an estimate that includes a 3% annual discount as described in Grabowski et al. (2012), we estimated the 14 year return on investment (ROI =(Benefits Costs ) /Costs*100) for the 88 projects in our database. We use the 14 costs should be recover ed in 2 14 years However, if restored reefs are left undisturbed it is possible for t hem to last up to 20 years (Powers et al. 2009), meaning they could potentially have ecosystem service benefits exceeded their initial cost after 14 years, and thos e with negative ROIs, meaning their costs were not regained in service benefits. Estimated ROI ranged from 93% to 368%, with exactly half the projects falling into positive and half into negative categories (Fig. 2 5) We found ROI to vary substantially among substrate types, with only 1 of 12 concrete based reefs yielding a positive ROI compared to 33 of 50 oyster shell reefs. ROI also varied with project size, with 29% of projects < 0. 4 hectares compared to 75% of projects > 0. 4 hectares yielding a posi tive ROI (Fig. 5B) Feedback from practitioners suggested that significant differences in substrate costs and in the circumstances under which projects were implemented were key factors driving this variation in ROI. For instance, the construction cost of an oyster shell restoration project in Massachusetts was quite low because the site was located close to the shell source, while the construction cost of a limestone boulder reef in Matagorda Bay, TX was relatively high because heavy equipment and experien ced operators were needed to place this substrate on site. Practitioners also noted that the cost of some substrate are rapidly changing: for example, the price of shell rose from $1 $2 to $2.50 $5 per bushel between 2010 and 2017 in the Chesapeake, respec tively, and bagged shell increased from $12.50 to $15 per bag from 2015 to

PAGE 25

25 2017, significantly altering restoration costs. Together, these findings highlight that gross differences in the cost efficiency of oyster restoration can arise depending on a proje ct's substrate, size, and setting, and that restoration costs are evolving. Thus, minimizing the cost of new restoration projects requires careful evaluation of all of these factors. Further consideration and quantification of how ecosystem service benef its vary among projects is also important for assessing the cost efficiency of oyster restoration. If, for instance, we calculated ROI using the highest annual value of ecosystem service benefits derived from restored oyster reefs of $99,000 hectare 1 year 1 estimated by Grabowski et al (2012), 83 of the 88 projects would have had a positive 14 year ROI. This simple exercise suggests that the implementation of more costly projects such as concrete based intertidal projects that can withstand high wave ene rgy can be financially justified if resultant reefs produce very high levels of ecosystem services. Although it must be noted that this higher value of $99,000 hectare 1 year 1 is largely driven by shoreline stabilization services, so projects geared towa rd shoreline protection might be the only ones achieving these high ecosystem service values. Importantly, t here continues to be progress made in measuring and predict ing these ecosystem service outcomes. Rigorous monitoring has revealed little evidence t hat deploying spat on shell is effective in rebuilding vibrant reefs in areas with low natural recruitment in the Southeast (Geraldi et al. 2013) or Gulf (Wallace et al. 2002), suggesting a low oyster related ecosystem service benefits and ROI potential fo r projects utilizing this high cost technique (but see CBT Sustainable Fisheries GIT 2017). In addition, recent meta analyses are provid ing a basis for predicting fish production from new oyster reef habitat ( Z u Ermgassen et al. 2015, building on Peterson et al. 2003) and T he Nature C onservancy is now incorporating these d ata into their Oyster Calculator which is being tested as a tool for scaling restoration projects in Florida

PAGE 26

26 supported through the Restore Act. These and similar future datasets that util ize universal metrics to monitor restored oyster reef health and service provisioning (Baggett et al. 2015, NAS 2017), as well as tools that enable managers to forecast the potential success of planned projects in sustaining high ecosystem service levels a re critical for inform ing where projects should be placed and what substrates they should utilize to improve the ROI of future efforts Finally, there are likely other substantial benefits to oyster restoration not captured in our ROI analyses that should be factored into decisions about when, where and how new projects should be implemented. In particular, our estimates do not explicitly account for avoidance costs, meaning those that would be accrued if the restoration activity had not taken place, such a s damage to property or loss of shoreline habitats protected by restored reefs. They also do not account for potential multiplicative benefits whereby restored reefs, through their improvement of water quality, stabilization of sediment, facilitation of vi brant food webs, and production of oyster larvae improve the health and resilience of surrounding habitats (e g mudflats, beaches, salt marshes, and other oyster reefs). They also do not capture hard to quantify societal benefits, such as those that volun teers derive from connecting with nature while constructing living shorelines or gardening oysters, or educational benefits students accrue from studying restored reefs. Furthermore, our ROI analyses do not capture how constructed reefs interact with other natural and anthropogenic factors, such as the evolution of disease resistant oysters and larval supply from sanctuary reefs, to mediate oyster population dynamics. Because these interactions are complex, determining what percentage of oyster recovery and ecosystem service benefits may be due to each factor in cases where populations appear to be on the rise like the Chesapeake Bay and designing management strategies to further augment recovery remains a challenge. Further research to measure these add itional benefits and to resolve these outstanding questions about the

PAGE 27

27 relative importance of oyster restoration versus fisheries management, sanctuaries, and natural process in affecting oyster population dynamics are critical to informing where further in vestment in restoration may be most profitable. Restoring on Historical Scales This synthesis documents a significant and accelerating effort to restore the Eastern oyster. However, when we compared the restored area in our database to the historic reef area lost (detailed in Zu Ermgassen et al. 2012), we discovered that 0.07 0.9, 3.6, 17.1 and 4.5% of degraded reef area has been restored via substrate deployment in the Mid Atlantic, Chesapeake, Southeast, Gulf and the combined total in the U.S. Atlantic and Gulf Coasts, respectively (see Object 2 3 for a list of all water bodies with historic reef area, degraded/ lost reef area, and restored reef area data). Object 2 3. List of all water bodies with hi storic reef area, degraded/ lost reef area, and restored reef area data. These percentages would no doubt be higher were we to have included in our calculations the natural areas without any restoration activity that have been set aside as sanctuaries, oys ter reefs that are now under more sustainable harvest management, and additional restoration efforts not captured in our dataset. It is important to note that we could not calculate these values for the Northeast region and our summary values for the other regions do not include data from specific water bodies where our database indicates there has been considerable oyster restoration, including Lynnhaven River, VA, and Tampa Bay, FL due to the lack of historic reef area data. Regardless, these analyses sug gest that practitioners have made a more significant dent in rebuilding historic reefs in certain water bodies, such as Mobile Bay, AL and Pamlico Sound, NC, than others and highlight that a tremendous area may still need to be restored to rehabilitate oys ter reefs at large scales. Given that recovery of historically reflective oyster reef area seems

PAGE 28

28 unachievable under our definition of oyster restoration unless dramatically more funding is made available for reef construction, these findings suggest that m arine protected areas and well managed harvest reefs must be incorporated into oyster recovery and management efforts. Furthermore, because a lack of suitable substrate is by no means the only reason for oyster decline, oyster habitat management must also consider the role of other environmental factors (e g disease, salinity, water quality) in mediating oyster decline and recovery. Whether the societal goal of restoration is to fully recover historic habitat that has been lost, or simply to regain enough habitat to generate desired ecosystem services, our analyses indicate we have a long way to go to achieving either endpoint for oys ter reefs. Our finding that the ROI for restoration tends to increase with project size (Fig. 2 5) suggests that there is critical need for new, cost effective strategies that allow restoration of much larger areas. In particular, innovations in substrate types and/or how they are sourced (e g recycled concrete rubble, mass manufactured textured substrates) and placed on site are essential for driving down construction costs. Similarly, novel approaches for minimizing labor costs, including homeowner gard ening and other volunteer programs, are needed to trim project budgets. In reviewing the literature, we also found widespread evidence that site characteristics, including elevation and access to adequate larval supply (Colden et al. 2017) are key to resto ration success. Consequently, first identifying sites where restoration activities are likely to stimulate the recovery of self sustaining, productive reefs is essential for ensuring positive ROIs. We therefore support prior recommendations that longer ter m, larger scale and standardized monitoring and research be devoted to understanding how and where projects are successful in order to improve project site selection and make more effective use of funds for restoration of this degraded coastal habitat.

PAGE 29

29 T able 2 1. Number of cost database entries as well as mean, minimum, and maximum costs per hectare of constructed reef for each material category. Material Number of entries Average cost/Ha Min. cost/Ha Max. cost/Ha Oyster 50 $137,148 $3,826 $411,339 Mixed Oyster 17 $120,166 $4,580 $245,028 Concrete 12 $1,225,579 $22,408 $2,180,361 Mixed Concrete 2 $136,527 $59,046 $214,008 Mixed 5 $428,483 $67,336 $827,690 Other 2 $188,622 $184,900 $192,345 Total 88 $299,999 $3,826 $2,180,361

PAGE 30

30 Figure 2 1. Oyster reef area constructed in each region of the Atlantic and Gulf Coasts of the US: A) reef area constructed over time, B) total number of reef hectares constructed by regions

PAGE 31

31 A B Figure 2 2. Examples of oyster restoration materials: A) oyster shell bags a type of oyster substrate August 21, 2016. Photograph courtesy of the author ; B) reef domes a type of concrete substrate August 21, 2016. Photograph courtesy of the author.

PAGE 32

32 Figure 2 3. Oyster reef area constructed by different substrates: A) reef hectares constructed over time; B) total reef hectares constructed by each material type

PAGE 33

33 Figure 2 4. Percentage of constructed reef area by each material type across study regions : A) data for all sites, B) data for the Northeast, C) data for the Mid Atlantic, D) data for Chesapeake Bay, E) data for the Southeast, F) data for the Gulf Coast

PAGE 34

34 Figure 2 5. Return on investment by project size for 88 oyster restoration projects: A) projects with a positive ROI B) projects with a negative ROI

PAGE 35

35 CHAPTER 3 INTERACTIVE EFFECTS OF WOOD TRAITS AND INTERTIDAL EXPOSURE ON SHIPWORM INFESTATION IN SOUTHEASTERN U.S. ESTUAR IES An important challenge for manm ade marine infrastructure is biofouling, the growth of organisms such as barnacles, algae, and sponges on submerged structures which can severely affect the functioning and limit the lifespan of structures found in coastal environments. For instance, hull fouling on ships increases drag, resulting in increased fuel consumption and higher costs (Schultz et al. 2011) and in metal structures, biofouling organisms can create a corrosive environment that degrades the materials (Yang et al. 2014). One material that is particularly vulnerable to biofouling is wood due to the risk of being infested by shipworms These are a group of wood boring bivalve molluscs of the genera Teredo, Bankia and Lyrodus that settle on wooden structures and aided by symbiotic bacte ria in their guts digest cellulose and excavate burrows for shelter ( Rice et al 1990, Lopez Anido et al. 2004, Nelson 2015). Because their larvae are abundant in the water and can settle and bore into wooden substrates, shipworms have been a problem for humans for centuries In the 1730s, shipworms caused extensive damage to Netherlands vulnerable to storm surge and damage (Sundberg 2015). I n the United States, the int roduction of the Teredo navalis into San Francisco Bay in the 1910s led to a loss of half a billion US dollars due to the destruction of wharves and piers that resulted in high maintenance costs and put a halt to shipping activities (Nelson 2015). Recent e stimates indicate that annual costs associated with losses and damages to infrastructure by shipworms is around $205 million US dollars per year (Pimentel et al. 2000), indicating these burrowing molluscs function as a pervasive and persistent drain on mar ine and coastal economies.

PAGE 36

36 Although shipworms are prevalent in marine systems, they are especially problematic in coastal and estuarine environments where wooden structures (e.g. wharves, piers, docks, mangroves) and debris (e.g. fallen trees) are abundan t. Shipworms have been shown to be tolerant of a wide range of salinities (between 5 and 35 ppt), but thrive at salinities of 10 ppt and above (Paalvast & van der Velde 2011a). Shipworms also tolerate temperatures ranging from 15 C to 25 C ( Steinmayer & Turfa 1996 ), but can spawn as soon as temperatures rise above 11 12 C (Graves 1942 ). In addition, shipworms can release between one and two million veliger larvae at a time, which occur in a free floating stage for three or fewer weeks before settl ing on w ooden substrates ( Grave 1928, Grave 1942). In currents, ballast water, and driftwood shipworms are able to disperse hundreds of kilometers (Scheltema 1971). Collectively, these life history traits make shipworms well equipped to survive and persist in bot h tropical and temperate estuarine environments worldwide, and to pose a threat to wooden structures found in these systems. To prolong the life of wooden structures in coastal and estuarine environments humans have sought out different methods. Historic a ccounts show that ancient Egyptians and Chinese protected wooden structures with resin, pitch, and pain t (Steinmeyer and Turfa 1997). Other more modern approaches include placing copper or lead plates on wooden ships, using paraffin, tar, and asphalt (Paal vast & van der Velde 2011 b ). However, none of these methods effectively prevent shipworm burrowing, and the one material that has worked best at preventing larval settlement, creosote which is a category of materials derived from the carbonization of coal, has toxic and carcinogenic properties (Hoppe 2002). Given that there are no treatments that are fully effective at preventing shipworm boring (Borges 2014), it is necessary to understand how environmental (e.g. location, distance from the sediment) and su bstrate characteristics (e.g. type

PAGE 37

37 of wood, branch/plank/piling size) mediate the prevalence of shipworm burrowing to inform the design and installation of ecological ly engineered docks, pilings and other infrastructure. This lack of treatment against ship worms and of understanding of their boring preferences is problematic because wood continues to be a commonly used construction material in estuarine environments (Borges et al. 2003). In particular, wood is often used in the construction of shoreline prot ection features, including bulkheads and breakwalls, in estuaries worldwide. Breakwalls, also known as groynes in Europe, are composed of wooden piles or fence posts that are filled with brush, branches or small trees and, given their porous nature and con struction just off shore, are designed to decrease wave or boat wake energy acting on and facilitate sediment deposition along the shoreline edge. They were first built in the North Sea in Germany in 1815 and are preferable to conventional hardened, non pe rmeable groynes because they are less expensive and result in less erosion in the adjacent unprotected shoreline (Bakker et al. 1984, Orford 1988, Weichbrodt 2008). For these same reasons, shoreline stabilization and restoration methods are starting to ado pt more natural approaches, and in the coming years, the use of wooden shoreline stabilization structures may increase in the face of sea level rise and increased shoreline erosion (Bulleri and Chapman 2010). Despite the forecasted increase in their use, i t remains unclear how their design might be optimized to enhance their longevity in the face of shipworm infestation. To better understand the environmental and substrate characteristics that modulate patterns in shipworm infestation of intertidal breakwal l branches and pilings, we conducted a 6 month field experiment (Experiment 1) to test how proximity to the sediment surface, tree species identity, branch diameter, and site interact to mediate the density of shipworm burrows and percentage of wood volume lost to burrowing in two northeast Florida estuaries. We then

PAGE 38

38 tested how tree species identity and distance from sediment mediate patterns in shipworm infestation across the southeastern US region by replicating this experiment for three months in the sam e two sites and at four additional sites (Experiment 2) Finally, we compared barnacle and oyster colonization and shipworm infestation of two non chemical (tape and silicone wraps) and two chemical techniques (pressure treated wood and copper based antifo uling paint) meant to protect wooden posts against biofouling and enhance their longevity (Experiment 3) For experiments 1 and 2 we hypothesized that: (1) shipworm burrow density and wood volume loss would be highest on branches located close to the sedi ment surface and negligible for branches buried underneath the sediment surface due to anoxic conditions, (2) small branches will lose a higher percentage of wood volume, but large branches will have higher burrow densities, and (3) branches with high wood densities (laurel oak and mangroves) will experience less damage than those with low wood densities (crepe myrtle and sweetgum). For experiment 3 we hypothesized that chemical techniques would result in wooden posts having fewer barnacles and s hipworm burrows than all other treatments, but that non chemical techniques would have less damage than unprotected controls. The findings of these experiments can potentially inform the ecologically engineered design of wooden breakwalls for shoreline pro tection in this region where lateral loss of shorelines is pervasive due to boat traffic and high energy wave environments. Methods Study S ites Experiments 1 and 3 were conducted in two tidal creeks within the Matanzas River Estuary in St. Augustine, Flori da, USA (site 1: 29 45' 47.9592'' N 81 15' 46.242'' W and site 2: 29 51' 57.7584'' N 81 18' 48.5316'' W Fig. 3 1). These sites are exposed to semidiurnal tides ranging from 0.25 m to 1.25 m, experience temperatures between 22 and 35 C in summer and from 4 to 25 C in winter, and receive an annual average of 113 mm of precipitation per month

PAGE 39

39 (NOAA National Centers for Environmental Information 2018). The tidal creeks selected were surrounded by salt marsh dominated by smooth cordgrass ( Spartina altern iflora ). Black ( Avicennia germinans ) mangroves were also present at both sites and occurred as isolated trees. Experiment 1: Tree Species, Branch Diameter, Elevation and Site Effects on S hipwo rm I nfestation hipworm infestation: laurel oak ( Quercus hemisphaerica) sweetgum ( Liquidambar styraciflua ), crepe myrtle ( Lagerstroemia spp. ), and black mangrove ( Avicennia germinans ). The first three species were selected because of their abundance in the region and thu s availability for use in breakwall construction Laurel oak and sweetgum are native to Florida, while crepe myrtle is an ornamental species that was introduced and now is widely established in the region Black mangroves are also common in Florida and, du e to their natural exposure to shipworms as a result of their intertidal estuarine distribution, we anticipated that this species would be more resistant to shipworm infestation However, this species cannot be harvested without a permit and is presented h ere simply as a useful comparison from which to gauge the infestation of the other tree species. For each tree species, we tested two diameter classes relevant for filling breakwalls given their availability and ease of handling a large and a small one. B ecause of the natural distribution of branch sizes, the diameter classes differed among species Laurel oak had a small diameter class ranging from 1 2.5 cm and a large diameter class ranging from 2.5 5 cm Crepe myrtle and sweetgum had a small diameter class ranging from 1 2 cm and a large diameter class ranging from 2 4 cm Mangrove branches had a small diameter class ranging from 1 1.5 cm and a large diameter class ranging from 1.5 3.5 cm. To compare shi pworm infestatio n rates between tree species, diameter classes, and distances from the sediment surface we built ladders using PVC poles as the ladder sides and

PAGE 40

40 tree branches as the rungs (Fig. 3 2). Each ladder from the small diameter class had a total of 9 50 cm long branch rungs that were secured to the PVC pipes with cable ties at positions of 10, 5, 0, 5, 10, 20, 30, 40, and 50 cm from the sediment surface Each l adder in the large diameter classes had a total of 7, 50cm branches that were secure d at positions of 10, 0, 10, 20, 30, 40, and 50 cm from the sediment ; note that the 5 and 5 cm rungs were omitted from large branch ladders due to space limitations Five replicate l adders were built for each study site and diameter class for laurel oak, sweetgum, and crepe myrtle branches However, d ue to a lack of available tree branches and legislation in Florida that restricts trimming of mangroves small and large diameter m angrove ladders had only 5 branches each, positioned at 10, 0, 10, 20, and 3 0 cm from the sediment (N=5 small diameter and N=4 large diameter class ladders) The ladders were driven by hand into the sediment at a spacing of 1m in the intertidal mudflat 2 3m from the salt marsh shoreline edge at each study site at elevations of 0 .572 to 0.732 m and 0.506 to 0.593 m above mean low water at sites 1 and 2, respectively in July 2016. These locations were selected to mimic the location where breakwalls are typically built to provide shoreline protection. The ladders were retrieved 6 months later in January 2017 In the lab, we counted the number of barnacles on each branch to evaluate biofouling (oysters were also observed but were extremely rare i.e. less than 10 oysters were observed across all 562 branches deployed). We then mea sured the length and diameter of each branch in order to calculate the initial wood volume by multiplying the area of the base of the branch times its length. After calculating this, we cut each branch into ten 5cm long segments. For each segment, we count ed the number of shipworm burrows and, using calipers, we measured the diameter and depth of 10 burrows per segment. If less than 10 burrows were observed, all burrows were measured. We used the burrow diameter and depth to calculate the volume of each

PAGE 41

41 of these 10 burrows by multiplying the area of the burrow opening times the burrow depth. WE then averaged these volumes, multiplied this value by the number of burrows per segment, and summed these values across each branch to estimate the total volume of wo od lost to shipworm burrowing Finally, we calculated the percent of branch wood volume lost to shipworms by dividing the total volume of wood lost to shipworm burrowing by the initial wood volume. For a sub set of 109 branches randomly distributed across all tree species, diameters and distances from sediment, we measured the diameter and depth of every observed burrow (between 1 112 burrows per branch) and compared the percent of wood volume lost from this detailed assessment of burrow distribution to that calculated using the average of 10 burrows per segment using a t test. Because the p value was not below 0.05, we consider the two values to be not significantly diffe rent and only measured 10 burrows for all branch segments and report the percent of wood volume lost values based on these measurements. Experiment 2: Regional Study of Tree Species, Elevation and Site Effects on Shipworm I nfestation To evaluate potential spatial variation across the region and assess interannual variability in shipworm infestation rates at the two sites evaluated in experiment 1, we repeated this experiment in June 2017 in our same two sites and deployed ladders at 4 additional sites along the southeastern US coast (Cedar Key, FL, Withlacoochee Bay, FL, Suwannee River, FL, and Sapelo Island, GA). Based on results from experiment 1, we used only small diameter laurel oak and sweetgum ladders with branches positioned at heights from 0 to 30 cm in 10 cm intervals as these tree species varied in their vulnerability to shipworm infestations, and these diameter class and positions were most vulnerable to shipworm infestation and thus best suited for assessing spatial and inte rannual variation in infestation rates. At each site, we deployed 5 replicates of each ladder in June 2017 and retrieved them September 2017. Branches were brought back to the

PAGE 42

42 lab and analyzed in the same way as branches in experiment 1 barnacles were co unted on each branch and branches were cut into 5 cm segments and inspected for shipworm burrows. Experiment 3: Anti Fouling T ech niques for Wooden S ubstrates We tested five treatments of wood protection against biofouling: pressure treated fence posts, com mercial copper anti fouling paint, silicone wraps, tape, and an untreated control on wooden 2x 2 posts at distances of 0.5 to 0.5m from the sediment. Posts were deployed on July 2016 at the same five replicates in sites 1 and 2 used in experiment 1. They w ere retrieved on January 2018 and brought back to the laboratory for processing. Biofouling was assessed for each post by counting the number of barnacles and shipworm damage was quantified by cutting each post at the 10, 5, 0, 5, 10, and 20 cm mark and estimating the percent of the area burrowed by shipworms. This technique for assessing shipworm damage differs from that used in the previous two experiments because previously we had individual branches for each height and could differentiate individual b urrows. In this case, however, we have one continuous wooden block for each treatment, so we assessed the percent area burrowed rather than the wood volume lost because burrows would often span multiple heights. In addition, posts were left in the ground f or a longer time than the branches because when considering their applications as wooden breakwalls, branches can be more easily replaced whereas fence posts need to last longer in the natural environment. Statistical A nalyses To evaluate the significance and relative importance of site, distance from the sediment, species type, and diameter in explaining variation in barnacle density per branch, percent wood volume lost, and shipworm burrow density, we developed regression trees using the analysis of varia nce (ANOVA) method of recursive partitioning. We then pruned over fitted trees using k

PAGE 43

43 fold cross validation. (see Gittman et al. 2015 for details). Regression trees were made using R Initial results from regression tree analyses for barnacle density in experiment 1 revealed site and branch diameter to be of little to no significance in predicting the extent to which tree branches would experience biofouling. Thus, we analyzed the effect size and significance of only distance from the sediment and species type on barnacle density found on tree branches by performing a two way ANOVA with these variables as main factors. Post hoc analyses were performed using Tukey HSD test. For experiment 2, bar nacles were analyzed in the same way as in experiment 1. However, because in this case site was significant, a three way ANOVA was run with site, species, and height as main factors. Due to the results obtained for experiment 2, no further statistical anal yses were run for shipworm data. To evaluate the significance and relative importance of site and treatment on barnacle density and percent area burrowed by shipworms on wooden posts in experiment 3, two way ANOVAs were run with these two variables as main factors. For analyzing shipworm damage, separate ANOVAs were run for each distance from the sediment. Results Biofouling Experiment 1 For experiment 1, a two way ANOVA found height (p< 0.0001 ), species (p< 0.0001 ), and the interaction of height and tree spe cies (p< 0.0001 ) to have significant effects on the number of barnacles observed per branch. We found that distance from sediment strongly mediated barnacle distribution with the average number of barnacles increasing with increasing height and peaking at 4 0 cm above the sediment where approximately 50 barnacles per branch were observed (Fig.3 3 ). Branches positioned 30 cm and above from the sediment to have

PAGE 44

44 significantly more barnacles than those below (F 1,554 =318.0, p<0.0001). The number of barnacles per b ranch on branches at the lower heights ( 10 to 5 cm from the sediment) were lowest and did not differ between branches. We also found the number of barnacles that settled on each type of branch to be highest on laurel oak and sweetgum branches with 36 and 27 barnacles per branch, respectively. Crepe myrtle had 18 barnacles per branch while mangrove branches had 7 barnacles per branch and was significantly lower than all other species. Experiment 2 For experiment 2, a three way ANOVA found height (p<0.0005), site (p<0.0001), and the interaction between height and site (p<0.0001) to have significant effects on the number of barnacles observed per branch. We found branches positioned 0cm from the sediment to have on average 45 barnacles per branch, a significan tly lower number than branches located above this height. Branches at the 10, 20 and 30 cm mark had on average 161, 190, and 196 barnacles per branch, respectively, but these values were not significantly different from one another. We also found the numbe r of barnacles per branch to be significantly higher on branches deployed in Cedar Key with 664 barnacles per branch on average. All other sites did not significantly differ in their number of barnacles. Experiment 3 For experiment 3, a two way ANOVA found treatment (p<0.0001), site (p<0.0 1 ), and the interaction between treatment and site (p<0.05) to have significant effects on barnacle density observed in the wooden posts. Treated fence posts had a significantly higher number of barnacles than all other tr eatments with 0.5 barnacles cm 1 on average. Silicone had the lowest number of barnacles with 0 barnacles observed on any silicone treated posts. However, this number was not significantly different from the remaining treatments, whose averages ranged from 0.03 to 0.1

PAGE 45

45 barnacles cm 1 We also found barnacle density to be significantly higher at Site 1 ( 0.2 barnacles cm 1 on average), than at Site 2 ( 0.07 barnacles cm 1 on average). Shipworm Damage Experiment 1 Despite both sites having similar shipworm boring damage (75 out 278 branches with at least one shipworm burrow at site 1 and 79 out of 284 branches at site 2), damage, measured both in terms of burrow density and the volume of wood lost, differed significantly between the two sites. Regression tre e analysis explained 99.9 and 94.4% of the variation in the per cent of wood volume lost (Fig. 3 4 A, tree root node error = 0.0149) and burrow density (Fig.3 4 B, tree f both metrics of shipworm damage. Specifically, while 2.6% of wood volume loss to burrows was observed at site 2, only 0.4% wood volume loss was observed at site 1. Similarly, while 10 burrows were observed per branch on average at site 2, only 1 burrow / branch was observed at site 1 (F 1,530 =31.07, p<0.0001). Shipworm burrowing also varied significantly with distance from the sediment surface at both si tes and across all four species. W hile we detected shipworm burrows at all distances, the percent of woo d volume lost and burrow density peaked in branches located 0 and 2 5 cm from the sediment layer (Fig. 3 5 ) Specifically, regression trees for both shipworm damage metrics (Fig. 3 4 A and 3 4 B) identified 2.5 cm (i.e. 2.5cm below the sediment surface) to b e the lower limit and 25 cm to be the upper limit of the area at which most shipworm damage occurs. Within this area, between 3.7 and 5% of wood volume was lost to shipworm burrows compared to only 0.4% wood volume loss outside this area. Similarly, shipwo rm burrow densities ranged between 19 and 28 burrows per branch if the branch was positioned within these boundaries versus 1 to 2

PAGE 46

46 burrows per branch if the branch was positioned further into the sediment or higher in the water column. Tree species identi ty was the third most important factor mediating variation in shipworm burrow damage such that sweetgum and crepe myrtle branches experienced more than 2.5 times higher percent wood volume loss and 3 times higher burrow density than laurel oaks and mangrov es (F 3,530 =2.89, Figs. 3 4A and 3 5 ) While we detected no significant effect of branch diameter class on the percentage of wood volume lost, this factor did explain significant variation in burrow density (Fig. 3 4 B) such that large diameter branches had almost three times higher burrow densities than small diameter branches. Experiment 2 Out of 240 branches deployed across six sites in experiment 2, only one sweetgum branch deployed at Site 2 in St. Augustine, FL had shipworm presence. A total of three burrows were found in this branch, accounting for approximately 1% of the wood volume of the branch. Experiment 3 Two way ANOVAs performed for shipworm percent burrowed data at each of the six heights ( 10, 5, 0, 5, 10, 20 cm) found treatment and site to have significant effects on the percent of the area burrowed by shipworm s but only at distances of 10cm (F 4,36 =4.45, p<0.01, F 1,36 =10.09 p<0.005) and 5cm (F 4,36 =4.58 p<0.005, F 1,36 =7.53 p<0.01). At 10 and 5cm, shipworm burrowing was significantly h igher on the control posts (28% and 32% of area burrowed, respectively) and significantly lower on posts treated with paint and tape (0% burrowed at both distances). Posts deployed at site 2 experienced significantly higher burrowing (17% and 19% of area b urrowed, respectively) than those deployed at site 1 (1.9% and 3.7% of area burrowed, respectively). At distances of 10cm only treatment was significant (F 4,36 =3.79

PAGE 47

47 p=0.011), with higher burrowing in control posts (23% of area burrowed) and lower burrowi ng in paint and tape treatments (3.5% and 2.2% respectively). However, paint and tape were not significantly different from silicone and fence posts. A t all other distances there was no significant difference in shipworm burrowing between treatments or sit es although shipworm damage ranged from 0% to 27% area burrowed Discussion Similar biofouling results were found for experiments 1 and 2, with the number of barnacles per branch increasing with increasing height. However, biofouling differed across sites in the southeastern US ( experiment 2 ) but not in St. Augustine sites ( exper iment 1 ) When looking at treatments to ward off barnacles, the most effective treatment in experiment 3 was silicone, which had no barnacles on any post. Viewed in the context of wooden marine infrastructure, this means that barnacles are likely to become more of a problem in the upper results suggest that pressure treated fence posts might be more likely to get encrusted with barnacles faster. Although the averag e percent wood volume lost was less than 7% (Fig. 3 4 ) over six months this can be a substantial amount of wood loss over time. Most importantly, these averages by site, height, and species can give an indication of when and how often wooden structures in coastal environments (e.g. wooden breakwalls) will need maintenance. Furthermore, it is important to note that some species have great variation in their percent wood volume lost, sometimes reachin g close to 25% wood loss (Fig. 3 5 ). This serves as anothe r indicator of what to expect with different tree species when they are exposed to shipworm infested waters. With these considerations in mind, we see there was a strong context dependence in shipworm infestation rates by site, height, and tree species, wi th great variability in the extent of shipworm

PAGE 48

48 damage between sites and between tree species. Branches deployed at site 2 experienced significantly more burrowing than those at site 1, and sweetgum and crepe myrtle branches, the tree species with the lowes t wood densities, experienced more shipworm damage compared to laurel oak and mangrove branches. Another important difference in the patterns of shipworm damage is that despite burrows being present at all heights, the majority of the damage was concentrat ed in the top 20 cm above the sediment. These differences in site, species, and height seen in experiment 1 suggest that shipworm burrowing may be exacerbated under certain conditions. In addition, the difference in the extent of burrowing from experiment 1 to experiment 2 suggests that there can be strong interannual effects within sites. The first initial difference in shipworm burrow density and the amount of wood volume lost was dependent on site. We compared elevation at each site and found Site 2 ladd ers were positioned at an average elevation of 0.593 m above sea level, while Site 1 ladders were placed at an average elevation of 0.6482 m above sea level. We would expect ladders at lower elevations to have a longer inundation time and in turn, to be exposed to shipworm infested waters for a longer time This was not the case, however, in our experiment since Site 1 ladders, positioned at lower elevation s, experience d less shipworm damage than ladders at S ite 2 Another difference may be the distance f rom each site to the mouth of the estuary. Given shipworms preference for higher salinities (Barrows 1917), the closer a site is to a saltwater source the more hospitable the environment is for shipworms. In our case, both sites were located along the Mat anzas River Estuary, but Site 1 was located 8.3 km from where the estuary meets the Atlantic Ocean, while Site 2 was 6.5 km away We expected Site 2 to experience higher salinities which might explain the higher infestation rates seen on branches deployed there. However, salinity data from nearby water quality monitoring stations revealed average salinities during the study

PAGE 49

49 period to be nearly the same (34.8 ppt in Site 1 and 34.1 ppt for Site 2). While salinity and elevation do not explain the results seen in this experiment, t hese differences in shipworm damage across sites could potentially be attributed to differences in larval delivery driven by creek morphology Site 1 consisted of a narrow, meandering creek, while Site 2 was a wider creek that was mor e open to the main river. This could potentially mean that a larger volume of water was present in Site 2 than at Site 1, and with it came a larger larval supply. The second most important indicator of shipworm boring was distance from the sediment surfac e. Results from this experiment show shipworm damage to be concentrated in the top twenty centimeters above the sediment layer, which is consistent with what has been found in other studies. Tuente et al. (2002) found in the harbors of Germany that shipwor m burrow densities on wooden piles increase with decreasing height above the sea floor. This experiment, however, took place in an estuary with a larger tidal range (0.30 m to 4.4 m) than ours and the wooden piles spanned a broader range of elevations (0 t o 5 m). Scheltema and Truitt (1956) found similar results in the coastal waters of Maryland with more shipworms being found on wooden panels closer to the sediment surface over a range of depths from 0 to 2.1 m. Finally, Paalvast and van der Velde (2011a) also found shipworm burrowing to be negatively correlated with distance to the sea floor at depths from 0 to 1 m. This pattern in height and shipworm burrowing likely arises because shipworms cannot access branches deep in the substrate ( 5 and 10 cm) bec ause of anoxic conditions and are inundated for less time at the upper limits (i.e. exposed to shipworm larvae for less time). However, it is important to note that for shipworms in subtidal systems with greater depth ranges, salinity also plays a role. As Barrows (1917) suggests, shipworm boring activity is largely driven by salinity, which increases with increasing depth (Paalvast and van der Velde 2011a), making the bottom most areas of estuaries

PAGE 50

50 particularly habitable to shipworms. Given these environme ntal preferences, it would be expected for shipworm activity to be more prevalent closer to the sediment and would explain the patterns seen in this and previous experiments. Tree species also influenced the extent of shipworm boring and we found species specific differing hardness, with shipworm burrows being more prevalent in softer branches than in harder ones. One measure of tree hardness is wood density, which is an indicator of how much wood substance is present in a given volume of wood and is typically calculated by the ratio of dry weight of wood divided by its green volume (Zobel and Jett 1995). Looking at the four different tree species used in this experiment, we found significantly more shipworm damage on sweetgum and crepe myrtle branches, which have wood densities of 0.42 and 0.55 g/cm 3 respectively (Holbrook and Putz 1989, Reyes et al. 1992) In contrast, black mangrove and oak species have significantly higher wood densities of 0.87 and 0.70 g/cm 3 respectively (Reyes et al. 1992, Saenger 2002), and experienced considerably less shipworm damage. These shipworm preferences for certain tree species is also consistent with Paalvast and van der experiment where they found more shipworm settlement in fir than in oak panels. They attribute shipworms to invest more energy boring into harder woods than softer woods. In addition, oak and mangrove trees produce tannins, a compound known to limit protein availability to organisms consuming their bark and leaves (Hathway 1958, Robbins et al 1987, Kimura and Wada 1989), potentially limiting digest ibility of these types of wood in shipworms. These results suggest that the type of tree species used can be a significant driver in the long term vulnerability of wooden structures built in coastal environments.

PAGE 51

51 From the regional study carried out during the second year of this experiment we can see that there can be interannual variability in shipworm activity. One possibility is that the patterns seen from one year to the next are a result of the time the wood was left in the water. The first set of ladd ers were left in the field for six months from July to January, but the second were left only for three months, from June to September. In their field experiment conducted in Port of Rotterdam, Netherlands, Paalvast and van der Velde (2011a) report that al though shipworm larvae were present in the water between April and November, they did not observe visible infestations in their wooden panels before September. This might mean that shipworm larvae might have been present but the wood was not in the water f or enough time for the larvae to grow and develop and cause actual damage. However, this serves to show that the timing of when wood gets placed in the water matters. Construction of structures such as wooden breakwalls might have a longer life span if str ategically built to maximize the period between larval exposure and shipworm boring. However, a s stated previously, shipworms prefer environments with higher salinity. This may have influenced the results we saw during the regional study given that 2016 ex perienced higher levels of drought and coincident salinity levels in estuaries we worked in than 2017. During the study period in 2016, the Palmer Drought Severity Index ranged from 2.73 to 2.06, while during 2017 it ranged from 0.76 to 2.48 (Figure 3 6 ). Salinities differed between the two years mainly in the minimum values reached. While sites 1 and 2 experienced minimum salinities of 28.6 and 23.2 ppt in 2016, respectively, in 2017 salinity dropped to minimums of 16.2 and 9.7 ppt (Figure 3 7) It is i mportant then to consider how climate change and other anthropogenic drivers interact to affect the severity and duration of drought, thus creating conditions for persistent shipworm activity.

PAGE 52

52 Implications for Wooden Structures in Coastal E nvironments Shi pworm boring preferences seen here can be used to inform the construction of wooden structures in coastal environments, such as wooden breakwalls used as living shorelines techniques. Particularly, understanding how shipworm burrowing varies across differe nt tree species can help identify the optimal building materials that will prolong the life span of these structures. For instance, when selecting filling for wooden wave breaks, choosing tree species with higher wood densities may result in structures tha t are more resistant to shipworm damage and that may not require such immediate maintenance as other lower wood density species In addition, knowing where shipworm burrowing is concentrated along the water column can help predict which will be the most vu lnerable areas of wooden structures. With this knowledge, priority can be given to the bottommost 25 cm of wooden structures when maintenance is required. Furthermore, being aware of the spatial variability in shipworm boring and of characteristics such as larval delivery to a site can help determine how long wooden structures will last in a particular area. Knowing if a site has high or low larval delivery can indicate the extent of shipworm damage that can be expected. Finally, being aware of the environm ental conditions (e.g. drought, decreased river discharge, warmer temperatures) that can create a hospitable environment for shipworms can help coordinate the timing of the deployment of new structures in order to maximize their life span or can dictate ma intenance efforts for wooden structures already installed in coastal areas. Ultimately, shipworms will be a persistent threat to wood in coastal environments, but a better understanding of the conditions under which shipworm boring occurs can promote a sma rter approach to extending the life span of wooden structures. Knowing the areas that are vulnerable to shipworm damage and addressing these vulnerabilities through innovative techniques such as combinations of natural and manmade materials can help build more resistant and longer lasting structures.

PAGE 53

53 Figure 3 1. Map of experimental sites in St. Augustine, FL, United States.

PAGE 54

54 Figure 3 2. Experimental design of ladders. Rungs represent tree branches positioned at the sh own distances from the sediment which is represented by the gray shading : A) d esi gn of small ladders for crepe myrtle, sw eetgum, and laurel oak branches; B) design of large ladders for crepe myrtle, sweetgum, and laurel oak branches; C) design for both small and large mangrove ladders.

PAGE 55

55 Figure 3 3. Average number of barnacles per branch at each distance from the sediment.

PAGE 56

56 Figure 3 4. Regression trees for experiment 1 : A) regression tree for percent wood volume lost mean percent wood volume lost indicated after each split along w ith the number of branches (n) included in the analysis ; B) regression tree for burrow density. Mean number of burrows indicated after each split along with number of branches (n) included in the analysis.

PAGE 57

57 Figure 3 5. Average p ercent wood volume lost for small branches (black bars) and large branches (gray bars). Bars show mean and standard error for five replicates at each site, species, and diameter class : A) data for crepe myrtle branches in site 1 ; B) data for sweetgum branches in site 1; C) data for laurel oak branches in site 1; D) data for mangrove branches in site 1; E) data for crepe myrtle branches in site 2; F) data for sweetgum branches in site 2; G) data for laurel oak branches in site 2; H) data for mangrove branches in site 2

PAGE 58

58 Figur e 3 6. Palmer Drought Severity Index (PDSI) from July 2016 to September 2017, spanning the periods of experiments 1 and 2. Figure 3 7 Salinity data for the two experimental sites at the time periods corresponding to experiments 1 and 2: A) salinity data for site 1 at time period 1 (left plot) and time period 2 (right plot) ; B) salinity data for site 2 at time period 1 (left plot) and time period 2 (right plot).

PAGE 59

59 CHAPTER 4 A MANUAL FOR RE ENGINEERING LIVING SHORELINES TO HALT EROSION AND RESTORE CO ASTAL HABITAT IN HIGH ENERGY ENVIRONMENTS For centuries, humans have settled on coastal systems because of their rich resources and high recreational and aesthetic value. This preference has been so extensive that to date over one third of the human popula tion lives within a hundred kilometers from the coast (MA 2005) which has created a strong dependence on coastal resources. The dependence communities have on the coastal resources has led to extensive development which, in turn, has resulted in high coa stal degradation. One particular effect of coastal development has been erosion damage. Coastal erosion is a physical process in which sediments are rearranged and shoreline is transformed due to both natural and anthropogenic factors. Natural factors, suc h as sediment sources and sinks, geological coastal processes, waves, and tides can contribute to shoreline erosion. However, many anthropogenic factors, including recreational activities such as boating, and dredging, construction, and other development a long the coast act together to transform the shorelines of beaches and estuaries (National Research Council 1990). The erosional changes in the shorelines and desire to protect the developed communities, prompted the construction of coastal defense struct ures such as concrete seawalls, groins, and bulkheads. Ideally, these structures are designed to act as a barrier and keep wave energy from impacting the shoreline, thus preventing erosion. However, these hardened coastal protection approaches solve short term erosional problems and can have several unintended consequences, such as erosion of adjacent shoreline and destruction of estuarine and coastal habitats. Furthermore, hardened shorelines have lower biodiversity and reduced habitat complexity when comp ared to natural shorelines (Gittman et al. 2016).

PAGE 60

60 Given the complications that can be caused by hardened shorelines, living shorelines have gradually become an alternative to hardened structures, and often have components of both based strategies of shoreline protection. The living shoreline approach comprises a variety of techniques, all of which operate under the premise of building with nature. Examples include replanting man grove seedlings and marsh grasses along eroded banks, using oyster shell or other substrates to repopulate degraded reefs, and using coir fiber logs to stabilize shorelines of estuaries. These natural materials not only dissipate wave energy without scouri ng and eroding the adjacent shoreline, but also create natural habitat and increase biodiversity and ecosystem functioning. While there are several advantages to living shorelines, it is important to understand that not all shorelines are the same which me ans that not all living shorelines approaches can be applied to every site. A challenge that several practitioners have encountered is working with living shorelines in high energy environments. Some common approaches are not suitable for habitats with hig h wave impacts, such as areas with high boat traffic that creates high energy boat wakes. For instance, mangrove seedlings and marsh grasses are easily uprooted and bagged oyster shell could become dislodged, leaving the shoreline once again unprotected. T hus, figuring out how to naturally protect high energy shorelines remains an outstanding challenge. To address this issue, the team of researchers and natural resource managers designed the project engineering of living shorelines to halt erosion and r estore coastal habitat in high which present s a novel living shorelines technique designed to address erosional impacts along high energy shorelines. The project team designed a hybrid living shoreline structure that acts as a double b arrier to dissipate boat wake energy along the Atlantic Intracoastal Waterway in St. Augustine, Florida. This technique consists of a set of wooden

PAGE 61

61 breakwalls placed in front of oyster restoration structures located in front of the shoreline. The breakwall s act as a first line of defense against boat wakes, while the oyster restoration structures are meant to serve as substrate to jump start oyster reefs that can further dissipate wave energy. Within this manual are project details including testing the hyb rid living shorelines technique and explanations of planning and site selection, design, installation, monitoring, and maintenance of the structures. Project Planning This living shoreline project was conducted along the shoreline of the Tolomato River in the Atlantic Intracoastal Waterway in Ponte Vedra, Florida, USA. The sites are intertidal, with semidiurnal tides ranging from 0.42 to 1.8 m, and are bordered on their landward edge by salt marsh habitat consisting mainly of smooth cordgrass ( Spartina alt erniflora ) along with some isolated black mangrove ( Avicennia germinans ) trees and patches of marsh succulents including Batis maritima, Borrichia frutescens and Sarcocornia perennis Historically, extensive Eastern oyster ( Crassostrea virginica ) reefs populated the intertidal margins of this estuary, but due to heavy boat traffic creating a high energy environment, only dead shell rakes now occur along the ICW channel edge. Live oyster reefs occur widely in tidal creeks and inlets that experienc e lowe r levels of boating activity in the area, however, and oyster larval supply in the area is fairly high during the reproductive season, which lasts from March through October (Dix et al. unpublished data ). Design and Installation Breakwalls Breakwalls are living shorelines structures built parallel to the shoreline with the intention of dissipating wave energy. The breakwalls consists of fourteen wooden fence posts assembled in a rectangle with a material within the posts (Figure 4 1). The pr oject breakwalls

PAGE 62

62 measure 14 feet long, 1.5 feet wide, and between 15 28 inches high and can be built in sets of three to protect a stretch of shoreline roughly 50 feet long. Each breakwall is built using 6.5 foot long wooden fence posts as the wall frame, tree branches as the fill material, and plastic coated multipurpose wire and fence post nails to hold the tree branches within the fence post wall frame. A breakwall is built by first installing a set of 14 wooden fence posts arranged in two rows of seven posts (Figure 4 1 ). Fence posts are driven 1.5 to 2.5 feet into the ground through the use of augers and large wooden mallets and positioned one in front of the other, leaving a 1 ft space between. With the fence posts set up, the space inside the wall can be filled with tree branches ensuring that branches are spread evenly across the length of the wall. It is recommended that tree branches measure between 7 and 14 feet long to ensure that the branches remain inside the wall as well as minimal loss of mat erial. The tree branches used for this project were crepe myrtle, however other tree species can also serve the same purpose. Crepe myrtle branches were chosen because it was a locally available, abundant, easily transported, and inexpensive resource. Tree branches inside the fence posts should reach roughly twice the desired height (i.e. 30 56 inches high). The tree branches should then be compressed by stepping on top of the branches until the wall is tightly packed. With the branches compressed, a second person can begin securing the branches in place with the plastic coasted multipurpose wire. To do this, multipurpose wire is woven in a zig zag pattern across the length of the wall, pulling on the wire to tighten it and securing it with a nail at every f ence post. Oyster Restoration S tructures Twenty feet behind each breakwall, four oyster restoration structures were placed within sediment (Figure 4 2 ). The oyster restoration structures will further dissipate wave energy and jump start the regrowth and r estoration of oyster reefs. For this project, two types of oyster

PAGE 63

63 restoration structures were tested: oyster gabions and BESE (Biodegradable Elements for Starting Ecosystems). These two types alternated within each breakwall (Figure 4 2 ). Oyster gabions ( Figure 4 3) were wire cages filled with recycled oyster shell. These were built using wire to form rectangular cages measuring 20 x 8 x 6 inches ( l*w*h ), which were then filled with recycled oyster shell obtained from a local s hell recycling program. BESE (Figure 4 3 ) were developed in the Netherlands and are currently being used for restoration of mussel beds in the Wadden Sea and several experim ental projects in Florida. BESE are structures made of potato starch waste that consist of multiple interlocking sheets in a honeycomb pattern. The sheets can be stacked to a desired height. For this project, BESE structures were assembled to have final dimensions of 35 x 17 x 2 inches (l*w*h). Oyster gabions are weighed down by the oyster shell and are able to be k ept in placed. BESE however, need to be secured using an L shaped rebar driven into the ground in between the middle of the structure. Maintenance Natural degradation and biofouling as well as the high energy environment will likely damage the structural integrity of the living shorelines structures. This is why periodic maintenance might be necessary in order to extend the life span of the breakwalls and oyster restoration structures. Breakwalls The most immediate threat to the breakwalls is dislodgment o f tree branches resulting in a loss of height. This can happen due to the high energy conditions the walls are in and due to loosening of the multipurpose wire holding the branches together. This is the reason why wall height must be monitored seasonally a nd if there is a substantial decrease in height from one monitoring period to the next, branches should be added on top of the remaining ones and secured once again with multipurpose wire.

PAGE 64

64 A second threat to breakwalls is biofouling. Barnacles can settle o n top and shipworms can infest the insides of branches and severely compromise their structural stability. Shipworms in particular can be especially damage as they burrow and remove wood volume from branches, making them more likely to break. Unfortunately there is no known treatment that is 100% preferences. Shipworms bore more easily into tree species with lower wood density and they concentrate their burrowi ng in the top 20 cm (Bersoza and Angelini, unpublished data ). Thus, to prolong the life span of the breakwalls, a higher density wood would be optimal. Also, this indicates that the bottom 20 cm of the wall will be the most vulnerable to shipworm boring an d will require the most maintenance. It is also important to consider that shipworms thrive in higher salinities, so periods of drought will result in more shipworm damage and thus increased maintenance. Oyster Restoration S tructures Due to the accumulati on of sediment, both oyster gabions and BESE run the risk of being buried over time. To avoid this, it is necessary to reposition them every monitoring period. The oyster gabions need to be lifted and shaken slightly in order for the sediment that has accu mulated within the structure to be dislodged. The gabions can then be placed back in their original position, making sure they are on top of the newly accumulated sediment. If BESE are buried, the rebar securing them in place should be removed. The process used with the oyster gabions should then be repeated for the BESE Oyster gabions also run the risk of the wire caging disintegrating. It is important to periodically examine the wire for signs of deterioration. If the wire is disintegrating before a matu re oyster reef is established, then it will be necessary to replace the gabion cages using a higher gauge wire.

PAGE 65

65 Because of their lightweight structure, BESE could potentially be dislodged due to the high energy wave environment. If that occurs, it will be necessary to secure them with additional rebar. Monitoring recommended set of shoreline, breakwall, and oyster parameters should be monitored. Monitoring efforts should o ccur seasonally, once every four months. The monitoring schedule and data sheets for this pr oject can be found in Appendix C for reference. Shoreline Monitoring e dge of the marsh was marked and defined as the point where the continuous mass of organic peat ended, along three, 14 meter long segments of shoreline using 15 diameter x 0.9 m long PVC poles that extended 0.5 m above the surface and were spac ed 1m apart. PVC poles were numbered 1 through 15 starting at the northernmost point at each point (Figure 4 4 ). These markers were used throughout the project to monitor shoreline position pre and post construction which allows evaluation of whether the shoreline was retreating, advancing, or staying the same by measuring the distance from each PVC pole to the most seaward cordgrass stem and patch of peat. Ground cover data was collected for each shoreline segment at specific PVC poles (numbered 3, 6, 9, 12, and 15). Ground cover measurements were taken from a 0.5 m x 0.5 m quadrat positioned at the PVC marker indicating the 1) shoreline edge (0 m), 2) 1.5 m behind the shoreline, and 3) 1.5 m in front. At each quadrat, the percent ground cover of sand, pe at, root mat, shell hash, and live oyster on the sediment surface was recorded to evaluate whether surficial sediment composition was shifting over time.

PAGE 66

66 Breakwall Monitoring The primary aspect to monitor in breakwalls is their height, since it is an indicator of p otential loss of material (i.e. tree branches) and when maintenance may be necessary. To measure breakwall height, height was recorded for each wall at eight points along the landward side and eight points along the seaward side. Oyster Monit oring Oyster monitoring methods were designed in order to collect data on three of the four proposed by Baggett et al. (2015) for assessing oyster restoration projects The metrics include oyster de nsity, oyster size, and reef height. A fourth universal metric from Baggett et al. (2015) reef area, is not appropriate for experimental substrates because both oyster gabions and BESE ha ve a fixed bottom areal coverage. Additionally, the general condition of the experimental s ubstrates and percent cover of oysters and other organisms on the substrates was assessed Observed condition The observed condition of the oyster gabions and BESE was characterized by visual inspection, with notes and photographs of degradation of wire o r materials (rust or deterioration), movement from original position, and sediment buildup and/or subsidence These notes were determined by making measurements with a ruler (to nearest 0.5 cm) at each corner of the structure and near the mid point on each side, of the vertical distance between the sediment surface and the bottom of the oyster gabion or BESE. Measurements and photos were made on three randomly selected oyster gabions or BESE behind each of the two types of wave breaks (tall and short) yiel ding a total of six assessed experimental substrate units per site.

PAGE 67

67 Vertical height of oysters The vertical height of live oysters on the oyster gabions and BESE assessed for observed condition (above) was determined by measuring with a ruler to the neares t mm the distance above the oyster gabion or BESE surface that live oysters project Measurements were made at 10 haphazardly selected locations on each of the 12 replicate substrates. Areal coverage of categories including live oysters, dead oysters, and other organisms was determined on three replicate oyster gabions using a point intercept method (modified from Berquist et al. 2006). Intercepts were recorded separately for the top, front (facing the wave breaks), and landward side of each oyster gabion. Percent cover of oysters and other taxa A template was made from the oyster gabion wire material contai ning fifty openings and was placed on the surface of the oyster gabion (or BESE) Percent cover was sampled by sliding a flag pin down from the intersect ion of the wire in one corner and identifying the object on the substrate as either If it was not oyster or sediment, the identification of ot f observations for each category was recorded, and per cent cover calculated using equation 4 1. % cover =(# of observations/50) x 100 (4 1) Oyster density and size To monitor oyster density and size in the oyster gabions, t hree oyster gabions were ra ndomly selected from each of the two types of wave breaks at each site (= 6 total per site). One smaller oyster gabion was removed from each larger oyster gabion selected, where the vertical height of live oysters was measured (see below) and contents were emptied into a fish box to be rinsed in order to remove sediments and debris. All shells in the smaller oyster gabion were counted and 25 of those shells were haphazardly chosen for measurement All live oysters on the

PAGE 68

68 25 selected shells were counted and measured (shell height to nearest mm ) with calipers or ruler. After processing, all shell and live oysters were placed back into the smaller oyster gabion which was returned to the larger gabion from which it was taken. Live oyster density (#oysters/0.1 m 2 ; and # oysters/0.015 m 3 ) was determin ed using equation 4 2. #oysters per 0.1 m 2 and per 0.015 m 3 = (#shells in mini gabion) (total live spat/25) (4 2) Size frequency distributions of live oysters were also determined for the measured individuals and asses sed by site and wave break type. To monitor oyster density and size in the BESE elements, t hree BESE were randomly selected from each of the two types of wave breaks at each site for a total of 6 per site). A 0.015 m 3 portion with the same dimensions as each smaller gabion (6 in [=15 cm] height x 8 in [=20 c m] width x 20 in [50 cm] length = 0.1 m 2 surface area) of each of the selected BESE elements was removed and processed as described above for the oyster gabions. After processing, all oysters and BESE material were re fastened to the entire BESE from which they were taken. Extensive M onitoring For a deeper understanding of how these living shoreline structures are shaping the ecosystem around them, as well as to get a better picture of the condition of the breakwalls and oyster gabions, this project included a second set of parameters to monitor. Shoreline Community data was collected for each shoreline segment at the PVC poles numbered 3, 6, 9, 12, and 15. Vegetation percent cover, cordgrass stem height (8 stems measured), stem density, snail, oyster, and mussel counts, as well as number of crab burrows wer e recorded within a 0.5 m x 0.5 m quadrat positioned at the PVC marker indicating 1) the shoreline edge (0 m), 2) 1.5 m behind the shoreline, and 3) 1.5 m in front.

PAGE 69

69 Breakwalls In addition to the breakwalls losing branches over time, a second threat to the structural integrity is the presence of fouling organisms on the tree branches and fence posts. Biofouling was measured by estimating the percent algae cover and the number of barnacles and oysters within a 0.1 m x 0.1 m quadrat which was posi tioned at 6 locations along the landward side of the wall and 6 locations along the seaward side. The quadrats were positioned halfway up the breakwall between the fence posts. These same biofouling metrics were recorded on 4 fence p osts on the landward si de and 4 fence posts on the seaward side of each wall to assess fence post biofouling. Cost C onsiderations For the breakwall the main expenses in terms of materials were the wooden fence posts. Because we built a total of 33 walls, this required 462 fence posts. Each post was priced at $4.79 resulting in a total of $2,212.98 Other materials included the multi purpose wire, priced at $9.98 and a box of fence post nails priced at $11 .98 For this project, we used approximately one 100 ft roll of wire per w all and three large boxes of nails, bringing the t otal for these materials to $341.32 Finally, equipment costs totaled at approximately $ 489 which were spent on the m anual augers, wooden mallets, and hammers necessary to build the walls. This brings the total costs for the breakwalls to $3,043 (see Table 4 1 for details). However, it is important to note that we obtained all of the tree branches through donations and that if a local source of branches is not available, this could drive up the cost of cons truction In terms of labor costs, each wall required at least four people working together and took approximately four to five hours to build. In addition, given the location of our living shoreline sites, we required a boat and boat captain for approxima tely five to six hours each day we were working.

PAGE 70

70 Table 4 1. Itemized list of the materials and their cost needed for the construction of the wooden breakwalls. Material Amount Cost per unit Total cost Fence posts 462 $4.79 $2,212.98 Multipurpose wire 33 100 ft rolls $9.98 $329.34 Fence post nails 1 box $11.98 $11.98 Augers 2 $113.95 $227.90 Wooden mallets 3 $77.03 $231.09 Hammers 3 $9.98 $29.94 Total $3043.23

PAGE 71

71 Figure 4 1. Diagram of breakwall design. Each circle represents one fencepost. Figure 4 2. Diagram of experimental set up relative to the shoreline showing breakwall placement in black and oyster restoration structures in gray.

PAGE 72

72 Figure 4 3. Photograph of oyster restoration structures. BESE are shown in the far left and in the middle. Gabions are shown second from the left and on the far right June 23, 2017. Photograph courtesy of the author. Figure 4 4. Photograph showing full experimental setup. PVC poles on the right mark the initial shoreline position and the spots selected for ecological monitoring. June 23, 2017. Photograph courtesy of the author.

PAGE 73

73 CHAPTER 5 CONCLUSIONS It is clear from the state of our coastal ecosystems that restoration needs to be an ongoing effort across different habitats. Although it is unlikely that we will restore our seagrass beds, mangrove forests, and other ecosystems to historic abundances and health, we can aim to restore enough structure and function to provide the necessary ecosystem services we depend on. To achieve this, howeve r, it is necessary that we learn from past efforts and make informed decisions based on data that is specific to the region and habitat we are trying to restore. From the review of oyster restoration projects in the United States we can learn that restorat ion effort can vary greatly by region and, in this case, can be heavily dependent on few substrates. This is concerning because certain restoration materials, while relatively readily available, can become scarce in the near future. Thus it is no t only im portant to consider what restoration materials have historically been used, but also to consider how the availability of these materials is likely to change. In addition, the review of oyster restoration projects presented in chapter 2 also shows that mate rial costs vary widely, and this can also limit how big restoration projects can be. For restoration efforts to be most profitable, it is impor tant to use materials that are cost effective locally sourced, durable, and easily deployed. In this way, the ar ea restored and, hopefully, the ecosystem services obtained are maximized. From chapter 3 we learn the importance of knowing the environment in which restoration takes place. It highlights the importance of knowing the specific threats to restoration proje cts, in this case shipworms, and understanding what to expect when restoring in a specific area. From this experiment specifically, we learn that for wooden structures in coastal environments shipworm infestations vary regionally, but that burrowing is con centrated in the bottom 20 cm. In addition, shipworm boring varies by tree species, with denser wood species being more

PAGE 74

74 resistant to burrowing. Thus, with this knowledge, we can modify the design of wooden restoration structures in such a way that maximize s their life span with respect to shipworm biofouling. Finally, lessons learned from chapters 2 and 3 come together in the form of a manual for a hybrid living shorelines technique that uses oyster restoration structures and wooden breakwalls to protect th e coast from boat induced shoreline erosion and to jump start oyster reefs in the intertidal zone Taking into account the recommendations from the oyster restoration review, this manual uses restoration materials that are locally available (oyster shell) and that area easy to assemble and deploy (BESE). In addition, future iterations of the wooden breakwalls will incorporate the use of alternative materials in the bottom 20 cm in order to prevent shipworm burrowing and to reduce maintenance costs. These re commendations are included in the manual in chapter 4 so that practitioners wishing to adopt our living shorelines approach into their restoration efforts can be aware that shipworms can affect the life span and functioning o f these breakwalls and can a dju st their design accordingly. Thus, chapter 4 presents an example of modern restoration efforts that incorporates lessons learned from past restoration projects and data obtained from region specific experimental studies. In this way, restoration is tailore d to a Although this thesis focuses on oyster reef restoration and shoreline protection, there are certain lessons that can be learned and broadly applied to other ecosystems. Mainly, these include (1) better reporting of past restoration projects is needed in or der for practitioners to learn from these efforts and build upon them when implementing new projects, (2) materials practitioners should aim to have durable, locally available substrates they can easily deploy, (3) better cost reporting can improve our

PAGE 75

75 adopt for restoring different ecosystems, (4) experimental field studies can greatly aid the design of restora tion projects by directly testing for specific variables of interest to the restoration project s or the area to be restored, and (5) integrating these lessons into the design of modern restoration projects and disseminating these findings in a format that practitioners can easily implement can help spread new restoration techniques and allow practitioners to easily build upon these new approaches and adjust the methods to their specific system. With these lessons, ecosystems can more effectively and efficie ntly be restored, and investments made in restoration projects will see higher returns in terms of the ecosystem services obtained

PAGE 76

76 APPENDIX A LIST OF SOURCES FOR OYSTER RESTORATION DATABASE Table A 1. Names and affi liations of restoration practitioners contacted by region. First Name Last Name Affiliation Region Jodi Baxter MD Department of Natural Resources Chesapeake Bay Dave Schulte US Army Corps of Engineers Chesapeake Bay Angie Sowers US Army Corps of Engineers Chesapeake Bay Stephanie Westby NOAA Chesapeake Bay Karl Willey Chesapeake Bay Foundation Chesapeake Bay Toby Baker Texas Commission on Environmental Quality Gulf Coast Jamie Boswell Charlotte Harbor National Estuary Program Gulf Coast Michael Ellis Louisiana Office of Coastal Protection and Restoration Gulf Coast Kyle Graham Louisiana Office of Coastal Protection and Restoration Gulf Coast Patrick Harper US Fish and Wildlife Services Gulf Coast Jason Herrmann AL Department of Conservation and Natural Resources Gulf Coast Phillip Hinesley AL Department of Conservation and Natural Resources Gulf Coast Jamie Miller MS Department of Marine Resources Gulf Coast Andrea Noel Northeast Florida Aquatic Preserves Department of Environmental Protection Gulf Coast Judy Ott Charlotte Harbor National Estuary Program Gulf Coast

PAGE 77

77 Table A 1. Continue d. First Name Last Name Affiliation Region James Pahl LA Coastal Protection and Restoration Authority Gulf Coast Rhonda Price MS Department of Marine Resources Gulf Coast Melody Ray Culp US Fish and Wildlife Services Gulf Coast Eric Sparks Mississippi Alabama Sea Grant Gulf Coast Cathy Drew The River Project Mid Atlantic Allison Fitzgerald New York/New Jersey Baykeeper Mid Atlantic Jeffrey Leviton Stony Brook University Mid Atlantic Jim Lodge The River Project Mid Atlantic Debbie Mans NY/NJ Baykeeper Mid Atlantic Amanda Muscavage US Army Corps of Engineers Mid Atlantic Joseph DeCrescenzo CT Department of Agriculture Northeast Matt Griffin Roger Williams University Northeast Dale Leavitt Roger Williams University Northeast Diane Murphy Woods Hole Oceanographic Institute Northeast David Patrick The Nature Conservancy Northeast Timothy Scott Roger Williams University Northeast Roxanna Smolowitz Roger Williams University Northeast RI Shellfish Management Program Northeast Thomas Bliss University of Georgia Southeast Joy Brown The Nature Conservancy Southeast Mary Conley The Nature Conservancy Southeast Carolyn Currin NOAA, University of North Carolina Southeast Anne Deaton NC Department of Environmental Quality Southeast Jane Harrison North Carolina Sea Grant Southeast

PAGE 78

78 Table A 1 Continued. First Name Last Name Affiliation Region Sharleen Johnson South Carolina Department of Natural Resources Southeast Peter Kingsley Smith South Carolina Department of Natural Resources Southeast Jan Mackinnon GA Department of Natural Resources Southeast Charles Peterson University of North Carolina Institute of Marine Sciences Southeast Linda Rimer US Environmental Protection Agency Southeast Mark Risse GA Sea Grant Southeast Denise Sanger SC Department of Natural Resources Southeast Deborah Scerno US Army Corps of Engineers Southeast Lisa Schiavinato NC Sea Grant Southeast Tracy Skrabal NC Coastal Federation Southeast Seth Theuerkauf North Carolina State University Southeast Linda Walters University of Central Florida Southeast Henry Wicker US Army Corps of Engineers Southeast Kevin Claridge FL Department of Environmental Protection Southeast/Gulf Coast Stephen Geiger FL Fish and Wildlife Conservation Commission Southeast/Gulf Coast Katie Konchar FL Fish and Wildlife Conservation Commission Southeast/Gulf Coast Kayleigh Michaelides FL Department of Environmental Protection Southeast/Gulf Coast Suzanne Simon Restore America's Estuaries Southeast/Gulf Coast

PAGE 79

79 Table A 1. Continued. First Name Last Name Affiliation Region Kent Smith FL Fish and Wildlife Conservation Commission Southeast/Gulf Coast Amber Whittle FL Fish and Wildlife Conservation Commission Southeast/Gulf Coast Dave Evans US Environmental Protection Agency Janine Harris NOAA, NMFS Office of Habitat Conservation Table A 2. List of databases accessed and the regions for which they had oyster restoration project data Database name Region NOAA Restoration Atlas All The Nature Conservancy Coastal Restoration Project Database All Chesapeake Bay Foundation Oyster Restoration Map Chesapeake Bay North Carolina Oyster Restoration Blueprint Southeast Coasts, O ceans, Ports, and Rivers Institute (COPRI) database Southeast, Gulf Coast Systems Approach to Geomorphic Engineering (SAGE) Project Database Northeast, Chesapeake Bay, Southeast Table A 3. List of publications detailing oyster restoration projects Publications Brown LA, Furlong JN, Brown KM, and Peyre MK La. 2014. Oyster reef restoration in the northern Gulf of Mexico: Effect of artificial substrate and age on nekton and benthic macroinvertebrate assemblage use. Restor Ecol 22 : 214 22. Brumbaugh RD, Sorabella LA, Oliveras Garcia C, et al. 2000. Making a case for community based oyster restoration: an example from Hampton Roads, Virginia, USA. J Shellfish Res 19 : 467 72. Dillon K, Peterson M, and May C. 2015. Functional equivalence of co nstructed and natural intertidal eastern oyster reef habitats in a northern Gulf of Mexico estuary. Mar Ecol Prog Ser 528 : 187 203.

PAGE 80

80 Table A 3. Continued Publications Dunn RP, Eggleston DB, and Lindquist NL. 2014. Effects of substrate type on demographic rates of eastern oyster (Crassostrea virginica). J Shellfish Res 33 : 177 85. Fodrie FJ, Rodriguez AB, Baillie CJ, et al. 2014. Classic paradigms in a novel environment : Inserting food web and productivity lessons from rocky shores and saltmarshes into biogenic reef restoration. J Appl Ecol : 1314 25. Furlong JN. 2012. Artificial Oyster Reefs in the Northern Gulf of Mexico: Management, Materials, and Faunal Effects. George LM, Santiago K De, Palmer T a., and Beseres Pollack J. 2014. Oyster reef restoration: effect of alternative substrates on oyster recruitment and nekton habitat use. J Coast Conserv 19 : 13 22. Grabowski JH, Hughes AR, Kimbro DL, and Dolan M a. 2005. How habitat setting influences restored oyster reef communities. Ecology 86 : 1926 35. Grizzle RE, Greene J, Jones S, et al. 2002. An oyster (Crassostrea virginica) reef restoration experiment in New Hampshire involving CROSBreed stock and native transplants. J Shellfish Res 21 : 430. Harding JM, Southworth MJ, Mann R, and James A. 2012. Comparison of Crassostrea virginica Gmel in (Eastern Oyster) Recruitment on Constructed Reefs and Adjacent Natural Oyster Bars over Decadal Time Scales. Northeast Nat 19 : 627 46. Humphries AT, Peyre MK La, Kimball ME, and Rozas LP. 2011. Testing the effect of habitat structure and complexity on nekton assemblages using experimental oyster reefs. J Exp Mar Bio Ecol 409 : 172 9. Peyre MK La, Humphries AT, Casas SM, and Peyre JF La. 2014. Temporal variation in development of ecosystem services from oyster reef restoration. Ecol Eng 63 : 34 44. Peyre M La, Furlong J, Brown LA, et al. 2014. Oyster reef restoration in the northern Gulf of Mexico: Extent, methods and outcomes. Ocean Coast Manag 89 : 20 8. Luckenbach MW, Coen LD, Ross PG, and Stephen JA. 2005. Oyster Reef Habitat ty Development Two Studies in Virginia and South Carolina. J Coast Res SI : 64 78. Marenghi F, Ozbay G, Erbland P, and Rossi Snook K. 2010. A comparison of the habitat value of sub tidal and floating oyster (Crassostrea virginica) aquaculture gear with a c reated Aquac Int 18 : 69 81.

PAGE 81

81 Table A 3. Continued Publications Meyer DL and Townsend EC. 2000. Faunal Utilization of Created Intertidal Eastern Oyster (Crassostrea virginica) Reefs in the Southeastern United States. Estuaries 23 : 34. Nelson K a, Leonard L a, Posey MH, et al. 2004. Using transplanted oyster (Crassostrea virginica) beds to improve water quality in small tidal creeks: a pilot study. J Exp Mar Bio Ecol 298 : 347 68. Oyster Crasso strea virginica on Created Oyster Reef Habitats in Chesapeake Bay. Restor Ecol 15 : 273 83. in constructed oyster reefs: oyster recruitment as a function of substrate typ e and tidal height. J Shellfish Res 19 : 387 95. Powers SP, Peterson CH, Grabowski JH, and Lenihan HS. 2009. Success of constructed oyster reefs in no harvest sanctuaries: implications for restoration. Mar Ecol Prog Ser 389 : 159 70. Ravit B, Comi M, Mans D, et al. 2012. Eastern Oysters (Crassostrea virginica) in the Hudson Raritan Estuary: Restoration Research and Shellfishery Policy. Environ Pract 14 : 110 29. Rossi Snook K, Ozbay G, and Marenghi F. 2010. Oyster (Crassostrea virginica) gardening Aquac Int 18 : 61 7. Scyphers SB, Powers SP, Heck KL, and Byron D. 2011. Oyster reefs as natural breakwaters miti gate shoreline loss and facilitate fisheries. PLoS One 6 Shorr J, Cervino J, Lin C, et al. 2012. Electrical Stimulation Increases Oyster Growth and Survival in Restoration Projects. : 151 9. Soniat TM and Burton GM. 2005. A comparison of the effectiveness of sandstone and limestone as cultch for oysters, Crassostrea virginica. J Shellfish Res 24 : 483 5. Swann L. 2008. The Use of Living Shorelines to Mitigate the Effects of Storm Events on Dau phin Island Alabama USA. Am Fish Soc Symp 64 : 11pp. Tallman JC and Forrester GE. 2007. Oyster Grow Out Cages Function as Artificial Reefs for Temperate Fishes. Trans Am Fish Soc 136 : 790 9.

PAGE 82

82 APPENDIX B SIZE OF CONSTRUCTED OYSTER REEF AREA BY REGION Table B 1. Average, minimum, and maximum sizes of constructed reefs in each region. Region Number of entries Number of entries with size data Average project size (Ha) Min. project size (Ha) Max. project size (Ha) Northeast 25 23 0.84 0.008 3.40 Mid Atlantic 25 19 1.18 0.00006 6.88 Chesapeake Bay 825 637 2.86 0.0001 122.67 Southeast 313 298 0.48 0.0004 18.41 Gulf Coast 580 201 15.85 0.0004 813.10 Total 1768 1181 4.4 0.00006 813.1

PAGE 83

83 APPENDIX C MONITORING SCHEDULE AND SAMPLE DATA SHEETS FOR LIVING SHORELINES PROJECT Monitoring Schedule November 2015 Site selection and preliminary shoreline monitoring This included shoreline position, vegetation, ground cover August 2016 Preliminary monitoring included same variables as previous monitoring date. May 2017 Monitoring immediately after wall installation and oyster restoration structure deployment. Monitoring covered same variables as before, but also included wall heights and biofouling on both walls and fence posts. September 2017 Monitoring was the same as May 2017. January 2018 Monitoring was the same as May 2017. Sample Data Sheets Figure C 1. Sample data sheet for monitoring wall height. Eight measurements are recorded for each wall on the seaward (front) and landward (back) sides. F igure C 2. Sample data sheet for monitoring biofouling on walls. Six measurements are taken on the seaward (front) and landward (back) sides.

PAGE 84

84 Figure C 3. Sample data sheet for vegetation and community monitoring. Figure C 4. Sample data sheet for per cent ground cover.

PAGE 85

85 Figure C 5. Sample data sheet for shoreline monitoring. DP indicates distance to the most seaward patch of remnant peat and DV indicates distance to most seaward vegetation.

PAGE 86

86 LIST OF REFERENCES Alleway HK and Connell SD. 2015. Loss of an ecological baseline through the eradication of oyster reefs from coastal ecosystems and human memory. Conserv Biol 29 : 795 804. Baggett LP, Powers SP, Brumbaugh RD, et al. 2015. Guidelines for evaluating performa nce of oyster habitat restoration. Restor Ecol 23 : 737 45. Bakker WT, Hulsbergen CH, Roelse P, et al. 1984. Permeable Groynes: Experiments and Practice in the Netherlands. In: Coastal Engineering. Houston, TX: American Society of Civil Engineers. Barrows A L. 1917. An unusual extension of the distribution of the shipworm in San Francisco Bay, California. Univ Calif Publ Zool 18 : 27 43. Bayraktarov E, Saunders MI, Abdullah S, et al. 2016. The cost and feasibility of marine coastal restoration. Ecol Appl 26 : 1 055 74. Beck MW, Brumbaugh RD, Airoldi L, et al. 2011. Oyster Reefs at Risk and Recommendations for Conservation, Restoration, and Management. Bioscience 61 : 107 16. Borges LMS, Cragg SM, and William s JR. 2003. Comparing the Resistance of a Number of Lesser Known Species of Tropical Hardwoods to the Marine Borer Limnoria Using a Short Term Laboratory Assay. Stockholm. Borges LMS. 2014. Biodegradation of wood exposed in the marine environment: Evaluati on of the hazard posed by marine wood borers in fifteen European sites. Int Biodeterior Biodegrad 96 : 97 104. Brown LA, Furlong JN, Brown KM, and Peyre MK La. 2014. Oyster re ef restoration in the northern G ulf of Mexico: Effect of artificial substrate and age on nekton and benthic macroinvertebrate assemblage use. Restor Ecol 22 : 214 22. Bulleri F and Chapman MG. 2010. The introduction of coastal infrastructure as a driver of change in marine environments. J Appl Ecol 47 : 26 35. Implementation Team) 2017. 2016 Oyster reef monitoring report: Analysis of data from large scale sanctuary oyster restoration projects in Maryland. NOA A Chesapeake Bay Office, Anapolis, MD 142 pp. https://chesapeakebay.noaa.gov/images/stories/pdf/2016oysterreefmonitoringreport.pdf Viewed 15 Jan 2018 Coen LD, Brumbaugh RD, Bushek D, et al. 2007. Ecosystem services related to oyster restoration. Mar Ecol Prog Ser 341 : 303 7.

PAGE 87

87 Colden AM, Latour RJ, and Lipcius RN. 2017. Reef height drives threshold dynamics of restored oyster reefs. Mar Ecol Prog Ser 582 : 1 13. Diederich S, Nehls G, Beusekom JE van, and Reise K. 2005. Introduced Pacific oysters (Crassostrea gigas) in the northern Wadden Sea: Invasion accelerated by warm summers? Helgol Mar Res 59 : 97 106. Dinnel PA, Peabody B, and Peter Contesse T. 2009. Rebuilding Olympia oysters, Ostrea lurida Carpenter 1864, in Fidalgo Bay, Washington. J Shellfish Res 28 : 79 85. FAO (Food and Agricultural Organization). 2014. The state of world fisheries and aquaculture: opportunities and challenges. Rome, Italy: Food and Agriculture Organization of the United Nations. Furlong JN. 2012. Artificial Oyster Reefs in the North ern Gulf of Mexico: Management, Materials, and Faunal Effects. Geraldi NR, Simpson M, Fegley SR, et al. 2013. Addition of juvenile oysters fails to enhance oyster reef development in Pamlico Sound. Mar Ecol Prog Ser 480 : 119 29. Gittman RK, Fodrie FJ, Popo wich AM, et al. analysis of shoreline hardening in the US. Front Ecol Environ 13 : 301 7. Gittman RK, Peterson CH, Currin CA, et al. 2016. Living shorelines can enhance the nursery role of threatened estuarin e habitats. Ecol Appl 26 : 249 63. Grabowski JH, Brumbaugh RD, Conrad RF, et al. 2012. Economic Valuation of Ecosystem Services Provided by Oyster Reefs. Bioscience 62 : 900 9. Grabowski JH, Hughes AR, Kimbro DL, and Dolan M a. 2005. How habitat setting infl uences restored oyster reef communities. Ecology 86 : 1926 35. Grave BH. 1928. Natural history of shipworm, Teredo navalis, at Woods Hole, Massachussetts. Biol Bull 55 : 260 82. Grave BH. 1942. T he Sexual Cycle of the Shipworm Teredo navalis. Biol Bull 82 : 438 45. Grizzle RE, Ward K, Lodge J, et al. 2011. Oyster Restoration Research Project (ORRP) technical report. New York. Grizzle RE and Ward K. 2016. Assessment of recent eastern oyster (Crassostrea virginica) reef restoration pr ojects in the Great Bay Est uary Durham, NH. Halpern BS, Walbridge S, Selkoe KA, et al. 2008. A global map of human impact on marine ecosystems. Science (80 ) 319 : 948 52. Hathway DE. 1958. Oak bark Tannins. Biochem J 70 : 34 42.

PAGE 88

88 Hilderbrand RH, Watts AC, and Randle AM. 2005. The myths of restoration ecology. Ecol Soc 10 Holbrook M and Putz FE. 2017. Infl uence of Neighbors on Tree Form : Effects of Lateral Shade and Prevention of Sway on the Allometry of Liquidambar styraciflua ( S weet Gum ) Author ( s ): N Michele Holbrook and Francis E. Putz Source : Americ an Journal of Botany, Vol 76, No 1. 76 : 1740 9. Hoppe KN. 2002. Teredo Navalis --the Cryptogenic Shipworm. In: Leppkoski E, Gollasch S, Olenin S (Eds). Invasive Aquatic S pecies of Europe. Distribution, Impacts and Management. Dordrecht: Springer Netherlands. Jackson JBC. 2008. Ecological extinction and evolution in the brave new ocean. Proc Natl Acad Sci U S A 105 : 11458 65. Kimura M and Wada H. 1989. Tannins in mangrove t ree roots and their role in the root environment. Soil Sci Plant Nutr 35 : 101 8. Kingsley Smith PR, Joyce RE, Arnott SA, et al. 2012. Habitat use of intertidal Eastern oyster (Crassostrea virginica) reefs by nekton in South Carolina Estuaries. J Shellfish Res 31 : 1009 21. Kirby MX. 2004. Fishing down the coast: historical expansion and collapse of oyster fisheries along continental margins. Proc Natl Acad Sci U S A 101 : 13096 9. Kuijper WJ and Turner H. 1992. Diet of a Roman centurion at Alphen aan den Rijn The Netherlands, in the first century AD. Rev Palaeobot Palynol 73 : 187 204. La Peyre MK Humphries AT, Casas SM, and Peyre JF La. 2014 b Temporal variation in development of ecosystem services from oyster reef restoration. Ecol Eng 63 : 34 44. La Peyre MK Furlong J, Brown LA, et al. 2014 a Oyster reef restoration in the northern Gulf of Mexico: Extent, methods and outcomes. Ocean Coast Manag 89 : 20 8. Laing I, Walker P, and Areal F. 2006. Return of the native is European oyster (Ostrea edulis) stock r estoration in the UK feasible ? Aquat Living Resour 19 : 283 7. Larsen PF, Wilson K a, and Morse D. 2013. Observations on the Expansion of a Relict Population of Eastern Oysters (Crassostrea virginica ) in a Maine Estuary : Implications for Climate Change and Restoration. Notes Northeast Nat 20 : 28 33. Lenihan HS and Peterson CH. 1998. How Habitat Degradation through Fishery Disturbance Enhances Impacts of Hypoxia on Oyster Reefs. Ecol Appl 8 : 128 40. Lopez Anido R, Michael AP, Goodell B, and Sandford TC. 2004. Assessment of Wood Pile Deterioration due to Marine Organisms. J Waterw Port, Coastal, Ocean Eng 130 : 70 6.

PAGE 89

89 Lotze HK, Lenihan HS, Bourque BJ, et al. 2006. Depletion, degradation, and recovery potential of estuaries and Coastal Seas. Science (80 ) 312 : 18 06 9. Lotze HK, Reise K, Worm B, et al. 2005. Human transformations of the Wadden Sea ecosystem through time: A synthesis. Helgol Mar Res 59 : 84 95. MacKenzie Jr. CL. 1996. History of oystering in the United States and Canada featuring the eight greatest oyster estuaries. Mar Fish Rev 58 : 1 78. Martin R. 2010. Commissioner Aims to Protect Public Health and Shellfish Industryhttp://www.nj.gov/dep/newsrel/2010/10_0053.htm. Viewed 18 Aug 2016. McGraw KA. 2009. The Olympia Oyster, Ostrea lurida Carpenter 1864 Along the West Coast of North America. J Shellfish Res 28 : 5 10. Millennium Ecosystem Assessment. 2005. Ecosystems and Human Well Being: desertification synthesis. National Academies of Sciences and Medicine E. 2017. Effective Monitoring to Evaluate Ecological Restoration in the Gulf of Mexico. Washington, DC: The National Academies Press. National Marine Fisheries Service. 201 5 Annual Commercial Landing Statistics. Silver Spring, MD: National Oceanic and Atmospheric Administration. http://www.st.nmfs.noaa.gov/commercial fisheries/commercial landings/annual landings/index. Viewed 21 Jul 201 7 National Research Council Division o n Engineering and Physical Sciences Commission on Engineering and Technical Systems Marine Board Water Science and Technology Board Committee on Coastal Erosion Zone Management. 1990. Managing Coastal Erosion. National Academies Press. Nelson DL. 2015. The Ravages of Teredo: The Rise and Fall of Shipworm in US History, 1860 1940. Environ Hist Durh N C 21 : 100 24. NOAA National Centers for Environmental information, Climate at a Glance: Divisional Time Series, published March 2018, retrieved on March 28, 201 8 from http://www.ncdc.noaa.gov/cag/ Paalvast P and Velde G van der. 2011. Distribution, settlement, and growth of first year individuals of the shipworm Teredo navalis L. (Bivalvia: Teredinidae) in the Port of Rotterdam area, the Netherlands. Int Biodeterior Biodegrad 65 : 1822 9. Paalvast P and Velde G van der. 2011. New threats of an old enemy: The distribution of the shipworm Teredo navalis L. (Bivalvia: Teredinidae) related to climate change in the Port of R otterdam area, the Netherlands. Mar Pollut Bull 62 : 1822 9.

PAGE 90

90 Pimentel D, Zuniga R, and Morrison D. 2005. Update on the environmental and economic costs associated with alien invasive species in the United States. Ecol Econ 52 : 273 88. Powell EN, Kraeuter JN and Ashton alcox KA. 2006. How long does oyster shell last on an Estuar Coast Shelf Sci 69 : 531 42. Powers SP, Peterson CH, Grabowski JH, and Lenihan HS. 2009. Success of constructed oyster reefs in no harvest sanctuaries: implications for restoration. Mar Ecol Prog Ser 389 : 159 70. Reyes G, Brown S, Chapman J, and Lugo AE. 1992. Wood densities of Tropical tree species. Rice SA, Johnson BR, and Estevez ED. 1990. Wood boring marine and estuarine animals in Florida. Rick TC, Reeder Myers LA, C ox CJ, et al. 2014. Shell Middens, Cultural Chronologies, and Coastal Settlement on the Rhode River Sub Estuary of Chesapeake Bay, Maryland, USA. Geoarchaeology 29 : 371 88. Robbins CT, Hanley TA, Hagerman AE, et al. 1987. Role of Tannins in Defending Plant s Ecology 68 : 98 107. Rossi Snook K, Ozbay G, and Marenghi F. 2010. Oyster (Crassostrea virginica) gardening Aquac Int 18 : 61 7. Saenger P. 2002. Mangrove Ecology, Silviculture and Conservation. Springer Science & Business Media. Scheltema RS. 1971. Dispersal of phytoplanktotrophic shipworm larvae (Bivalvia: Teredinidae) over long distances by ocean currents. Mar Biol 11 : 5 11. Scheltema RS and Tru itt RV. 1956. The Shipworm Teredo Navalis in Maryland Coastal Waters. Ecology 37 : 841 3. Schulte DM. 2017. History of the Virginia Oyster Fishery Chesapeake Bay, USA. Front Mar Sci 4 Schultz MP, Bendick JA, Holm ER, et al. 2018. Economic impact of biofo uling on a naval surface ship. 7014 SER ( Society for Ecological Restoration International Science & Policy Working Group ) 2004. The SER International Primer on Ecological Restoration. www.ser.org & Tucson: Society for Ecological Restoration International. Smyth D, Roberts D, and Browne L. 2009. Impacts of unregulated harvesting on a recovering stock of native oysters (Ostrea edulis). Mar Pollut Bull 58 : 916 22.

PAGE 91

91 Southworth MJ and Mann R. 1998. Oyster reef broodstock enhancement in the Great Wi comico River, Virginia. J Shellfish Res 17 : 1101 14. Steinmayer Jr. AG and Turfa JM. 1997. Shipworms and ancient Mediterranean warships a response. Int J Naut Archaeol 26 : 345 6. Steinmayer AG and Turfa JM. 1996. Effects of shipworm on the performance of ancient Mediterranean warships. Int J Naut Archaeol 25 : 104 21. Sundberg A. 2015. Molluscan Explosion: The Dutch Shipworm Epidemic of the 1730s. Environ Soc Portal 14 Therneau T, Atkinson B, and Ripley B. 2018. An Introduction to Recursive Partitioning U sing the RPART Routineshttps://cran.r project.org/web/packages/rpart/vignettes/longintro.pdf. Viewed Tuente U, Piepenburg D, and Spindler M. 2002. Occurrence and settlement of the common shipworm Teredo navalis (Bivalvia: Teredinidae) in Bremerhaven harbours, northern Germany. Helgol Mar Res 56 : 87 94. UNEP. 2006. Marine and coastal ecosystems and human w ell being: A synthesis report based on the findings of the Millennium Ecosystem Assessment. Wallace R, Rikard F, and Howe J. 2002. Optimum size for planting hatchery produced oyster seed: Final Technical Report. Weichbrodt F. 2008. Coastal Groynes in Germa ny. Die Kste 74 : 241 51. Westby SR, Bruce D, Slacum W, et al. 2016. 2015 Oyster Restoration Implementation Update Resto ration Progress in Harris Creek Little Choptank River and Tred Avon River Combined. White JM, Buhle ER, Ruesink JL, and Trimble AC. 2 009. Evaluation of Olympia oyster (Ostrea lurida Carpenter 1864) status and restoration techniques in Puget Sound, Washington, United States. J Shellfish Res 28 : 107 12. Worm B, Barbier EB, Beaumont N, et al. 2006. Impacts of biodiversity loss on ocean eco system services. Science (80 ) 314 : 787 90. Yang WJ, Neoh KG, Kang ET, et al. 2014. Polymer brush coatings for combating marine biofouling. Prog Polym Sci 39 : 1017 42. Yozzo DJ, Wilber P, and Will RJ. 2004. Beneficial use of dredged material for habitat c reation, enhancement, and restoration in New York New Jersey Harbor. J Environ Manage 73 : 39 52.

PAGE 92

92 Zobel BJ and Jett JB. 1995. The Importance of Wood Density (Specific Gravity) and Its Component Parts BT Genetics of Wood Production. In: Zobel BJ, Jett J B (Eds). Berlin, Heidelberg: Springer Berlin Heidelberg. Zu Ermgassen PSE, Spalding MD, Blake B, et al. 2012. Historical ecology with real numbers: past and present extent and biomass of an imperilled estuarine habitat. Proc Biol Sci 279 : 3393 400. Z u Erm gassen PSE, Grabowski JH, Gai r JR, and Powers SP. 2015. Quan ifying fish and mobile invertebrate production from a threatened nursery habitat. J Appl Ecol 53 : 596 606.

PAGE 93

93 BIOGRAPHICAL SKETCH Ada Cecilia Bersoza Hernndez was born in Monterrey, Mexico. She obtained her B .S. in biology with a focus in ecology and evolutionary b iology from Brown University in 2015.