Research on Marine Coastal Impacts to Promote Ecosystem-Based Management: Nonnative Species in Northeast Pacific Estuaries

RESEARCH ON MARINE COASTAL IMPACTS TO PROMOTE ECOSYSTEM-BASED MANAGEMENT: NONNATIVE SPECIES IN NORTHEAST PACIFIC ESTUARIES

by Megan Elizabeth Mach

B.Sc., The University of Washington, 2005 M.S., Boston University, 2007

A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY in THE FACULTY OF GRADUATE STUDIES (Resource Management and Environmental Studies)

THE UNIVERSITY OF BRITISH COLUMBIA (Vancouver)

August 2012

Abstract

Ecosystem-based management (EBM) offers a holistic evaluation of tradeoffs between human activities, but this offer rests upon a foundation of science. In this thesis, I assessed and advanced the knowledge-base for EBM in five ways, focusing on nonnative species in estuarine ecosystems.

In Chapter 2, I tested for the comprehensiveness of research that connects the impacts of anthropogenic activities to changes in ecosystem service production, employing a literature review of estuarine ecosystems. Research on these connections virtually never included the relationship of activities to ecosystem services production, presenting an impressive gap in research for evaluating tradeoffs using EBM.

I addressed the sufficiency of existing information regarding nonnative species in eelgrass beds in Chapter 3. I tested the relationship of nonnative species in British Columbia’s (BC) eelgrass beds with arrival pathways and environmental selection factors. There were few (12) nonnatives in BC’s eelgrass; all associated most commonly with aquaculture facilities and warm temperatures. Existing reports included the majority of nonnatives: only one species, the bamboo worm Clymenella torquata, represented a new record, as I described in Chapter 4.

Impacts of nonnatives are difficult to limit after invasion. In Chapter 5, I developed an approach for characterizing the potential economic impacts of nonnatives. I focused on European green crab, a nonnative species that has not yet arrived in Puget Sound, Washington. At a range of invasion densities and increasing calorie diets, I calculated a value-at-risk to shellfish harvest ranging from $1.6 - $41 million USD. Such calculations can aid in preparation for impending invasion by motivating prevention and mitigation efforts.

Nonnative management is often based on the available understanding of the impacts on native species. In Chapter 6, I assessed available research on the impact of nonnative seagrass, Zostera japonica, in northeast Pacific estuaries. My results suggested existing studies that quantitatively test Z. japonica impacts are insufficient to comprehensively assess the effects of this invasion. My dissertation research highlights the need for research to determine the ecosystem role of nonnatives in their invaded range through analysis of quantitative studies across broad scales.

Preface With the exception of Chapters 1 and 7, all chapters were originally prepared as stand- alone, peer-reviewed publications. I am the senior author and took primary responsibility for the development of research questions, research design and methods, analysis and writing of all co-authored chapters. Details of co-authorship for chapters 2 to 6 are outlined below.

Chapter 2. Development of this chapter stemmed from the literature review I completed for my comprehensive exam. My advisor, Dr. Kai Chan contributed to this chapter through the development of ideas and research questions. My other co-author, Dr. Rebecca Martone was key at implementing structure and testing hypotheses. Both co- authors contributed edits and comments to the manuscript.

Chapter 3. The overarching question in this chapter “what nonnative species are present in British Columbia’s ports?” was developed by my funding agency, the Canadian Aquatic Invasive Species Network. Beyond this I developed all research questions and sampling methods with the support of Dr. Colin Levings. I analyzed data and wrote the manuscript with detailed editing and comments from Dr. Levings and Dr. Kai Chan.

Chapter 4. Research question development, sampling methods and species identification originated from Chapter 3. In addition, my co-author, Dr. P. Sean MacDonald provided sampling results from habitats in Puget Sound, Washington as well as editing and comments on the final manuscript. I collected and analyzed all additional data and personal communications with researchers and aquaculture facilities in Washington State, and drafted the text. A version of this chapter reprinted with permission from Springer. Mach ME, CD Levings, PS McDonald, and KMA Chan. 2011. An Atlantic infaunal engineer is established in the Northeast Pacific: Clymenella torquata (polychaeta: maldanidae) on the British Columbia and Washington Coasts. Biological Invasions. 14(3):503-507

iii

Chapter 5. Dr. Kai Chan contributed to the initial idea and later to edits and comments on the manuscript. Dr. Robert Ahrens helped develop the model and sensitivity analysis used in this chapter. I collected all information, carried out all data management and analyses, and drafted the text.

Chapter 6. Dr. Sandy Wyllie-Echeverria contributed to the development of ideas and provided extensive edits and comments on the manuscript for this chapter. Dr. Kai Chan contributed suggestions and edits on numerous drafts, broadening the impact and application of the results. I collected all information, carried out all data management and analyses, and drafted the text.

Table of Contents

Abstract ...... ii

Preface ...... iii

Table of Contents ...... v

List of Tables ...... x

List of Figures ...... xi

List of Abbreviations ...... xiii

Acknowledgements ...... xiv

Epigraph ...... xvi

Chapter 1 Introduction ...... 1

1.1 Human reliance on coastal ecosystems ...... 1

1.2 Ecosystem-based management in marine systems ...... 1

1.3 Estuaries at the land-sea interface ...... 2

1.4 Impacts to estuaries ...... 4

1.5 Nonnative species in coastal ecosystems ...... 6

1.6 Seagrass and nonnative impacts in British Columbia, Canada ...... 9

1.7 Management of marine nonnative species ...... 10

1.8 Thesis overview ...... 12

Chapter 2 Research gaps in evaluating tradeoffs in ecosystem-based management14

2.1 Synopsis ...... 14

2.2 Introduction ...... 14

2.3 Methods ...... 21

2.4 Results ...... 23

2.5 Discussion ...... 25

2.5.1 Knowledge gaps for evaluating tradeoffs within EBM ...... 25

2.5.2 Why does understanding the complete impact-pathway matter? ...... 26

2.5.3 Linking science and management to inform EBM ...... 28

2.5.4 Limitations to informing the impact-pathway ...... 29

2.5.5 Conclusions ...... 30

Chapter 3 Nonnative species in British Columbian Zostera marina eelgrass beds arrive by aquaculture and stay for the mild climate ...... 32

3.1 Synopsis ...... 32

3.2 Introduction ...... 32

3.3 Methods ...... 36

3.3.1 Field sampling ...... 38

3.3.2 Vector and environmental data ...... 39

3.3.3 Nonnative abundance and distribution ...... 40

3.3.3.1 Relating vectors to nonnative richness and abundance ...... 41

3.3.3.2 Relating environmental variables nonnative richness and abundance ... 42

3.3.3.3 Influence of nonnative species composition and environment on site similarity ...... 44

3.4 Results ...... 45

3.4.1 Nonnatives’ frequency and distributions ...... 49

3.4.2 Aquaculture versus shipping vectors ...... 51

3.4.3 Model results for environment effect on nonnative richness and abundance 53

3.4.4 Differences in nonnative community composition ...... 55

3.5 Discussion ...... 57

3.5.1 Relationship to aquaculture vectors ...... 58

3.5.2 Climate influences nonnative abundance and distribution ...... 59

3.5.3 Nonnatives in BC eelgrass and their potential impacts ...... 61

3.5.4 Conclusions ...... 64

Chapter 4 An Atlantic infaunal engineer is established in the northeast Pacific: Clymenella torquata (Polychaeta: Maldanidae) on the British Columbia and Washington Coast ...... 66

4.1 Synopsis ...... 66

4.2 Introduction ...... 66

4.2.1 Clymenella in the northeast Pacific ...... 68

4.3 Methods ...... 69

4.4 Results and discussion ...... 70

4.4.1 Dispersal in the northeast Pacific ...... 71

4.4.2 Monitoring and management ...... 71

Chapter 5 Trading green backs for green crabs: evaluating the commercial shellfish harvest at risk to European green crab invasion ...... 73

5.1 Synopsis ...... 73

5.2 Introduction ...... 74

5.3 Methods ...... 77

5.3.1 Study site and organism: Carcinus maenas as a threat to Puget Sound ...... 77

5.3.2 Biomass and revenue of Puget Sound’s shellfish harvest ...... 80

5.3.3 Model of Carcinus maenas impacts on shellfish harvested in Puget Sound . 84

5.3.4 Estimating impacts to shellfish value in Puget Sound ...... 87

5.4 Results ...... 89

5.4.1 Commercial shellfish harvest ...... 89

5.4.2 Commercial shellfish harvest at risk from green crab invasion ...... 89

5.4.3 Value-at-risk associated with commercial shellfish harvest loss ...... 94

5.5 Discussion ...... 96

5.5.1 Additional ecosystem changes ...... 98

vii

5.5.2 Motivating prevention and mitigation of invasive impacts ...... 99

5.5.3 Incorporating uncertainty of future invasion impacts ...... 100

5.5.4 Future research directions ...... 101

5.5.5 Conclusions ...... 101

Chapter 6 Knowledge gaps may impede ecosystem management: environmental impacts of a nonnative seagrass, Zostera japonica, in the northeast Pacific ...... 103

6.1 Synopsis ...... 103

6.2 Introduction ...... 104

6.3 Methods ...... 107

6.3.1 Searched publications ...... 107

6.3.2 Assessed publications ...... 107

6.3.3 Analysis of the effects of Zostera japonica on the biotic and abiotic environments ...... 108

6.4 Results ...... 110

6.4.1 Space and time distribution ...... 110

6.4.2 Study foci ...... 112

6.4.3 Analysis of biotic impacts ...... 113

6.4.4 Analysis of abiotic impacts ...... 118

6.5 Discussion ...... 122

6.5.1 Logistical limitations ...... 122

6.5.2 Changes to populations of mudflat species ...... 123

6.5.3 Changes to the abiotic environment ...... 126

6.6 Gaps in research for management goals ...... 127

6.6.1 Conclusions ...... 128

Chapter 7 Conclusions ...... 130

7.1 Knowledge-base limitations for marine management ...... 130

viii

7.2 Advancing the knowledge-base for marine management ...... 131

7.3 Research across the US-Canada border ...... 132

7.4 Overcoming limitations and future directions for research ...... 133

7.5 Science to inform management ...... 136

References ...... 137

Appendices ...... 172

Appendix A Species accumulation curves ...... 172

Appendix B Species abundances in core, dredge and trap eelgrass bed samples .... 173

Appendix C Post hoc analysis of additional environmental variables not included in the AICc model test ...... 184

C.1 Methods ...... 184

C.2 Results ...... 186

C.3 Discussion ...... 187

Appendix D R code for the green crab consumption model ...... 188

List of Tables Table 2.1 Subset of activities and impacts analyzed as search terms in Chapter 2…... 21

Table 3.1 Nonnative and cryptogenic species sampled in BC eelgrass beds………… 46

Table 3.2 Variables used in the analysis of relationships between nonnative richness and abundance and the environmental conditions at each site……………………….. 48

Table 3.3 Spearman rank correlation and probability-value for benthic and epifaunal nonnative richness and abundance with nonnative vectors………………………...… 51

Table 3.4 Linear models of nonnative benthic and epifaunal richness and abundance compared to environmental conditions in the eleven eelgrass beds sampled ………... 54

Table 5.1 Harvest and total revenue of shellfish species by Puget Sound Partnership action area in Puget Sound……………………………………………………………. 90

Table 5.2 Shellfish harvest before and after green crab predation in Puget Sound…... 91

Table 5.3 2009 Puget Sound shellfish harvest revenue estimated for before and after green crab invasion…………………………………………………………………… 95

Table 6.1 The categories on which research on Z. japonica in the northeast Pacific has focused……………………………………..……………...…………………...... 113

Table 6.2 A summary of experimental and observational studies testing the effects of Z. japonica on native invertebrate and plant species……………………...... 114

Table 6.3 A summary of experimental and observational studies testing the effects of Z. japonica on the abiotic environment…………………………………………... 119

List of Figures Figure 2.1 Activity – Impact – Provider – Service (AIPS) EBM framework for understanding the impacts of multiple human activities on ecosystem services in coastal ecosystems………………………...... …………………………………………..…... 17

Figure 2.2 Impact-pathways for estuaries at the intersection of terrestrial, freshwater and marine ecosystems.…...……………...... ……………………………………...… 20

Figure 2.3 Map of global ecoregions demonstrating regions with complete impact- pathway studies.…...……………...... ……………………………………...... … 25

Figure 3.1 Map of the eleven eelgrass beds sampled in British Columbia...…………. 37

Figure 3.2 Presumed vectors of introduction for nonnatives in British Columbia across all marine habitats, in BC eelgrass beds, and reported in eelgrass beds in the northeast Pacific……………………...... ……..…………...... 49

Figure 3.3 Bar plots of mean nonnative benthic and epifaunal richness and abundance at each site sampled.……………………………...... … 50

Figure 3.4 Correlation of nonnative richness and vector data plotted with Spearman rank correlation.……………………………...... … 52

Figure 3.5 Correlation of nonnative abundance and vector data plotted with Spearman rank correlation.……………………………...... … 53

Figure 3.6 Native benthic and epifaunal richness as predicted by summer sea surface temperature.……………………………...... …wh 55

Figure 3.7 Nonmetric multidimensional scaling plot visualizing site dissimilarity of nonnative community composition...... … 57

Figure 4.1 Clymenella from Roberts Bank, BC, collected in 2008…...... … 67

Figure 4.2 Known distribution of Clymenella in the Pacific…………………………. 68

Figure 4.3 Abundances of Clymenella in Boundary Bay in 1980, and Roberts Bank and Campbell River in 2008, as estimated through sediment cores………………………. 70

Figure 5.1 The geographic potential of Carcinus maenas for the northeast Pacific coast from Northern California to Alaska, USA……………………………………………. 79

Figure 5.2 Puget Sound Partnership action area map (2012)………………………… 81

Figure 5.3 Washington State Department of Health (2010) commercial and recreational shellfish “Annual Inventory Growing Areas Map”………….……………………….. 82

Figure 5.4 Mean harvested shellfish biomass and value at increasing levels of green crab densities for four shellfish species at increasing levels of green crab density consumed by each green crab per year……………………………………………….. 92

Figure 5.5 Sensitivity analysis of the loss of shellfish harvest biomass (lbs year-1) as a function of calories consumed each year and the density of C. maenas km-2 in harvest areas…………………………………………………………………………………... 93

Figure 6.1 Map of northeast Pacific range of Zostera japonica.………………….….106

Figure 6.2 The mean number of studies per year conducted at each estuary in which Zostera japonica was studied each decade over the previous 40 years…………..…. 111

xii

List of Abbreviations

BC British Columbia C Celsius cm centimeter EBM ecosystem-based management EIA environmental impact assessment km kilometer m meter nMDS nonmetric multi-dimensional scaling ppt parts per thousand UBC University of British Columbia WA Washington State

xiii

Acknowledgements

Research during my dissertation was a collaborative effort involving many wonderful people who shared their advice, time and friendship while getting me to the end of this adventure. My advisor at the University of British Columbia, Dr. Kai Chan, put in many hours broadening my understanding of the application of my research. Working with Kai at the Institute for Resources, Environment and Sustainability altered my future trajectory, moving me from a scientist interested in applying her own work, to a scientist actively attempting to improve how science-at-large can be applied towards improving management of marine ecosystems. My advisor at the Department of Fisheries and Ocean Canada, Dr. Colin Levings, shared his vast knowledge of marine dynamics in British Columbia and helped me become part of the Canadian Aquatic Invasive Species Network (CAISN). Dr. Chris Harley welcomed me as a member of his lab, helping me to remain grounded in ecology. Thank you all for helping me down this path.

Field, laboratory, and data processing work contributing to my research, some of which made it in to this dissertation and some of which will follow, was done with boundless assistance. In British Columbia I received great help from Eva Bianca Corlett, Brendan Cowan, Robin Elahi, Trampus Goodman, Alex Hart, Dr. Katie Mach, Caitlin Millar, Greg McCullagh, Dr. Cathryn Clarke Murry, Yasha Podeswa, Kate Stanton. During field sampling along the coast of Nova Scotia I worked with the ever-entertaining Olivier d’Amour, Olivia Lacoste, and Karine Richer. I received additional assistance and taxonomic expertise from R. Eugene Ruff, Kathy Cowell, Dr. James Carlton Monica Bravo and Dr. Jeffrey Cordell who confirmed the identities of thousands of invertebrates I sampled during this research.

But what to do with all these data?! Thanks to many, I was finally able to beat the data into submission. For general ideas, edits, stats and critiques I would like to thank Dr. Robert Ahrens, Dr. Kevin Britton-Simmons, Dr. James Carlton, Dr. Megan Dethier, Robin Elahi, Ed Gregr, Lisa Mach, Dr. Rebecca Martone, Dr. Cathryn Clarke Murray,

xiv

Mark Plummer, Dr. Jennifer Rhode-Ward, Gerald Singh, Jordan Tam, Oscia Wilson, Dr. Sandy Wyllie-Echeverria

I had the chance to talk with folks involved in research and oyster aquaculture in Washington State who advised and assisted me with nonnative species south of the border. Thanks to Paul Blau, Steven Booth, Bill Dewey, Dr. Paul Dinnel, Dr. P. Sean McDonald, and Russel Rogers.

Research during my dissertation was supported by numerous sources. Thank you to CAISN, UBC, Washington Sea Grant, and for funding through Kai Chan from Packard, and NSERC Discovery Grant. I would especially like to thank the University of Washington’s Friday Harbor Laboratories (FHL). My first forays into marine science were taken as an undergraduate at FHL. Through my undergrad, Masters and PhD my time spent at FHL taught me the basics of how to be a marine scientist and more than 8 years later I hope I have done them proud. Thank you Aimee Urata, Stacy Markman, Kathy Cowell, Vicky Dauciunas, and Scott Schwinge, you all made FHL my home away from home.

Not surprisingly, spending five years working excessively and making minimum wage can only end well with amazing support from family and friends. Most importantly, my mom’s support through my PhD, really through all of my education and adventures, has been the most inspiring and guiding force. She leads by example and through her I have learned dedication, determination and creativity. Thank you mom. My lovely friends have stuck around despite my annoying habitat of picking work over spending time with them. I promise that will change! Thank you Oscia Timschell, Virginia Zachary, Kimberly Janos. And Robin Elahi, thanks for sticking it out, can’t wait to see what comes next! And to all the lovely ladies that I have had the chance to become friends with during these last dissertation years: Jennifer Jorve, Rebecca Kordas, Rebecca Martone, Meg O’Shea and Cathryn Clarke-Murray. I hope our friendship continues to grow as we all move forward in life. I love you all.

Epigraph

"If you want to catch beasts you don’t see every day, You have to go places quite out-of-the-way. You have to go places no others can get to. You have to get cold and you have to get wet, too."

- Dr. Seuss, If I Ran The Zoo

xvi 1.1 Human reliance on coastal ecosystems

Chapter 1 Introduction

1.1 Human reliance on coastal ecosystems As some of the most productive ecosystems in the world, coastal marine systems share an interdependent relationship with coastal human populations (Levin & Lubchenco, 2008). Human dependence on these ecosystems extends well beyond the economic benefit of fishing and harvesting, the common values accounted for in management and planning (Agardy & Alder, 2005). However, because many of these benefits are produced in part through ecological processes often out of human view, the importance of these services is often ignored (Barbier et al., 2011; Chan et al., 2012). To manage for ecosystem benefits the ways in which anthropogenic impacts alter coastal marine systems must be understood.

The objective of my dissertation is to investigate and advance the capacity of existing scientific knowledge on anthropogenic impacts to inform coastal ecosystem-based management.

1.2 Ecosystem-based management in marine systems Managing ecosystems requires an integrated approach that considers the entire ecosystem, including cumulative impacts, linkages between ecosystems and human systems, and the range of benefits (i.e., ecosystem services) provided to humans by these systems (Daily et al., 1997; Millennium Ecosystem Assessment, 2005; Browman & Stergiou, 2006; Ruttenberg & Granek, 2011). This integration is central to ecosystem- based management (EBM), considering the benefits from natural systems when evaluating trade-offs between management objectives. This approach differs from traditional approaches to marine management that consider only single sectors or species, such as managing a salmon fishery for a maximum sustainable yield by controlling only fishing rights. EBM involves managing the entire food-web that includes salmon, and managing the production of salmon in the context of other species. However, understanding the cumulative impacts that arise from multiple ecosystems with the

1 1.3 Estuaries at the land-sea interface potential to alter ecosystem values necessitates strategically focused research targeted at clarifying these complex relationships (Millennium Ecosystem Assessment, 2005; McLeod & Leslie, 2009).

EBM offers a more holistic evaluation of tradeoffs between impacts of human activities in coastal ecosystems, but this offer rests on a foundation of science. There is a major role for ecological research in decision-making using an EBM approach (Leslie & Mcleod, 2007; Tallis et al., 2010). Research that connects human activities to ecosystem services and their value can be used to select management objectives that consider the beneficiaries of those services. When these relationships are untested, management decisions are based on expert opinion or “best guess” scenarios, weakening the capability of management to accurately evaluate various objectives (Nelson et al., 2009; White et al., 2012). Many aspects of science to support EBM are understood and moving forward with this method is a priority (Price et al., 2009; Lester et al., 2010; Gregr & Chan, 2011). However, there is not a clear understanding of what gaps exist for research to support EBM. To improve the chances that future research will fill those research gaps, and to help managers develop realistic goals regarding questions answerable through research, lines of communication between researchers and management need to be improved (Groffman et al., 2006; Cheong, 2008). In this way we might develop effective management action at a multi-ecosystem, multi-species scale that connects human activities to their impacts in coastal ecosystems.

1.3 Estuaries at the land-sea interface The cumulative effects of cross-system impacts and valuable ecosystem services production make estuaries a model ecosystem for exploring the utility of EBM (Ruckelshaus et al., 2009; Wendt et al., 2009). Coastal marine estuaries are ecosystems with a disproportionate importance relative to their size for many resident and migratory species (Stoms et al., 2005). These tidal habitats and adjacent tidal wetlands are usually semi enclosed, receiving freshwater input from land but with partial access to the open ocean, forming at the intersection of terrestrial, freshwater and marine ecosystems (Cameron & Pritchard, 1963). Human derived materials move directly and indirectly into

2 1.3 Estuaries at the land-sea interface estuaries, predominately through water flow, from these neighboring systems (Reiners & Driese, 2001). The productivity of estuaries is maintained by those species that provide habitat and support biodiverse ecological processes. Coastal mangroves, seagrasses and shoregrasses protect against storms and floods (Adger et al., 2005), improve water quality, and filter nutrients from coastal runoff (Lee & Dunton, 1999); moreover, the organisms that use these habitats support fisheries, touristic value, and other cultural benefits (Agardy & Alder, 2005). For example, recreational fishing is associated with many values provided by estuaries: people enjoy fishing and find value in this activity that is not necessarily associated with keeping the fish for food and commercial businesses near estuaries receive indirect benefits from the purchases of food, supplies and lodging (Crutchfield, 1962).

Seagrasses, in particular, support complex food webs by virtue of their physical structure in near-shore sand and mud flats that are frequently devoid of upright structure and primary production (Hemminga & Duarte, 2000). Their blades, roots and rhizomes are habitat for the foraging and protection of numerous coastal species. The detrital nutrients from these plants form the base of a diverse and productive foodweb of bacteria and microorganisms, molluscs, crustaceans, fish, and waterfowl (Williams & Heck, 2001). Globally, productivity of seagrasses exceeds that of mangroves and coral reefs, some of the most productive coastal habitats (Duarte et al., 2008). In 2005 the US recorded commercial landings of several species that use seagrasses during at least part of their life cycle as totaling over $126 million USD (National Marine Fisheries Service, 2005; Hughes et al., 2008). These plants are also a key component in sediment stability that prevents coastal erosion, in the dampening of wave action that lessens storm impacts, in the reincorporation of nutrients into marine foodwebs, and in carbon sequestration and storage (Hemminga & Duarte, 2000; Thom et al., 2001; Barbier et al., 2011).

Despite the scientifically recognized importance and value of the species supported by seagrass, there is relatively little public awareness of seagrass’s productivity and contribution to estuarine ecosystems (Duarte et al., 2008). The focus of media reports in coastal ecosystems has been disproportionately concentrated on coral reefs over

3 1.4 Impacts to estuaries mangrove, seagrass and shoregrass habitats, with seagrasses receiving the least amount of attention. Less media attention may result in or signify a lack of public awareness that is out of proportion with the research published on each of these habitats or the estimated value of these ecosystems. In one coarse and controversial study, the annual value estimated for seagrass is $19,004 per hectare while coral reefs were valued at $6,075 per hectare (Costanza et al., 1997). The importance of seagrass is belied by its unobtrusive nature: it is not easy to see as it is submerged except at low tides, and many of its benefits are indirect and often not obvious outside of management and research studies.

1.4 Impacts to estuaries Estuaries are sensitive to changes in inputs from neighboring ecosystems (Stoms et al., 2005). Increased nutrient runoff enters estuaries from urban areas and agriculture, damming in freshwater systems alters freshwater flow (Brauman et al., 2007), and marine shipping and aquaculture introduce marine nonnative1 species (hereafter “nonnatives”). For example, the introduction of the Japanese wireweed, Sargassum muticum, via aquaculture, has reduced seagrass area and prevented reestablishment through competition for space in northwest Atlantic near-shore habitats (denHartog, 1997). In addition, stressors within estuaries, such as dredging and boating disturbance have reduced productive estuarine habitats. For example, approximately 90 percent of the original coastal wetland habitat in southern California was lost as a result of filling or dredging during the last century (National Oceanic and Atmospheric Administration (NOAA), 1990). In general, impacts to these important systems are cumulative and difficult to manage as a result of their position at the intersection of marine, freshwater and terrestrial ecosystems, and of their continued use and degradation as important access-points for shipping, freshwater, and food for human populations (Ruttenberg & Granek, 2011).

1 Nonnative species refers to those species that have been transported beyond their natural dispersal range. The term “invasive” specifically refers to those nonnative species that have a demonstrated negative impact on the habitats they are introduced to.

4 1.4 Impacts to estuaries

Human impacts to estuaries are increasingly reducing the production of ecosystem services (Levin & Lubchenco, 2008), but there are seldom direct and simple relationships between human impacts and ecosystem service production. For example, a number of stressors, (e.g., toxins, sedimentation, nonnative species) may directly or indirectly impact an ecosystem service provider (e.g., shellfish) and the many services it provides (e.g., filtration, habitat and food) (Ruesink et al., 2006). Or, a single stressor could have opposing impacts, for example, nutrient increases may result in increased growth of oysters while stimulating blooms of algae that reduce oxygen levels and kill those same oysters (Anderson et al., 2002). Discerning how the many impacts that converge in estuaries affect ecosystem services requires rigorous scientific exploration across ecosystem boundaries (Ruttenberg & Granek, 2011). At the same time, estuaries are one of the most difficult ecosystems to manage and study because of converging impacts from distant systems. In the Gulf of Mexico, a fisheries “dead zone” has formed at the mouth of the Mississippi River as a result of nutrient runoff from thousands of square kilometers of agriculture that border this river (Rabalais et al., 2002). Microalgae and bacteria blooms reduce the oxygen levels in the coastal waters, killing fish and invertebrate species and reducing the harvested biomass of local fisheries. To incorporate impacts from external ecosystems human activities must be managed while considering the dynamics of an estuary that could be hundreds of miles away (Diaz & Rosenberg, 2008).

For seagrass beds, which play a vital role in estuarine ecosystems worldwide (Hemminga & Duarte, 2000), the global net loss of areal extent is likely to have pronounced, widespread impacts (Orth et al., 2006; Hughes et al., 2008). Decline in seagrass area has resulted from many cross-system pressures, including the introduction of nonnative species (Short & Wyllie-Echeverria, 1996; Orth et al., 2006). Despite the role of nonnatives as important drivers of change, decreasing global biodiversity and altering food-web interactions (Wilcove et al., 1998; Didham et al., 2005; Hobbs, 2006), there remains a lack of understanding of nonnative marine species and their impacts in seagrass beds (Williams, 2007; Carlton, 2009). Nonnatives may alter seagrasses directly through habitat modification, for example the European green crab, Carcinus maenas, tears up

5 1.5 Nonnative species in coastal ecosystems seagrass when preying on infaunal bivalves (Davis et al., 1998), and these impacts have the potential to indirectly impact species associated with these habitats (Seymour et al., 2002), however few studies have explicitly tested these relationships.

1.5 Nonnative species in coastal ecosystems Increased global connectivity has led to the introduction of numerous nonnative marine species around the world, with harmful economic and ecological effects (Lockwood et al., 2005; Colautti et al., 2006). Nonnative species introductions have increased with increasing globalization of trade connecting regions of the world that until now have been relatively isolated (Wonham & Pachepsky, 2006) and studies have demonstrated that natural dispersal abilities of nonnatives do not account for their observed distributions (Cohen & Carlton, 1995; Carlton & Cohen, 2003; Cohen, 2004). Species introductions range from small crustaceans and molluscs to colonial tunicates and amphipods, arriving via shipping ballast and hull fouling, as intentional introductions for aquaculture fisheries, and as hitchhikers on oysters used for aquaculture (Wonham & Carlton, 2005; Molnar et al., 2008). While some nonnatives have been acknowledged for the benefits they bring to the regions they invade2, nonnative species are more frequently recognized for their competitive and predatory impacts on native communities for space and resources (Tilman, 1997; Stachowicz et al., 1999). The Pacific oyster Crassostrea gigas was intentionally introduced to Canada from Japan for aquaculture circa 1912 (Wonham & Carlton, 2005) and has since altered intertidal habitats and caused shifts in community composition (Ruesink et al., 2005). Perhaps of greater impact, the intentional introduction of C. gigas unintentionally introduced more than 60 new species of invertebrates and algae to the northeast Pacific (Wonham & Carlton, 2005).

Nonnatives frequently cause ecological changes to native environments that reduce ecosystem service provision and as a result, are economically costly (Hayes et al., 2005; deRivera et al., 2007a). Carcinus maenas, the European green crab, reduced northwest

2 Nonnative shoregrasses in Chesapeake Bay survive in wastewater runoff from urban areas, providing habitat to waterfowl and fish species when native shoregrasses die off from exposure to the same runoff (Hershner & Havens, 2008),

6 1.5 Nonnative species in coastal ecosystems

Atlantic soft shell clam biomass and have cost commercial fisheries an estimated 22.6 million USD per year (Lovell et al., 2007). While the costs of nonnatives can be great, successful prevention, control and eradication is rare (Hayes et al., 2005; deRivera et al., 2007a) (but see Myers et al., 2011). Eradication and control efforts in Washington State for invasive shoregrasses (Spartina spp.) have been underway since 1988, but in the five years between 1995 and 2000, shoregrass infestations increased by over 250% (Hedge et al., 2003).

The ecological and economic impacts of nonnatives have resulted in increased numbers of studies on how these species are transported and how they establish once they arrive in a new location (Rilov & Crooks, 2009). The risk of establishment is determined by a nonnative’s ability to arrive at the new region, survive environmental stressors, and their evolutionary predisposition for invasion (Shea & Chesson, 2002; Leung & Mandrak, 2007). First, the nonnative must be transported to a new location. The pathways by which nonnatives colonize new sites are still being studied, but shipping and aquaculture have been demonstrated as two of the most common vectors for transport of marine nonnatives (Wonham & Carlton, 2005; Molnar et al., 2008). The density and frequency of organisms’ arrivals, or propagule pressure, is an important determinant of what ultimately invades (Lockwood et al., 2005). The greater the number of introductions, the more likely a newly introduced species will settle and establish successfully (Wonham et al., 2001).

The susceptibility of the environment to invasion by new species, or its invasiblity, is an important factor in invasion success (Davis et al., 2000). Upon arrival in a new estuarine ecosystem, the abiotic environment frequently determines the success of the individual propagule: salinity (Mann & Harding, 2003; Powers et al., 2006; Miller et al., 2007), temperature (Stachowicz et al., 2002b; Clark & Johnston, 2005; Dafforn et al., 2009), and other abiotic variables must all be within the suitable range for the invading species to settle and survive (Dethier & Hacker, 2005; Incera et al., 2009; MacLeod et al., 2009). In addition, establishment is dependent on aspects of the biotic environment, such as resource availability (Moyle & Light, 1996; Davis et al., 2000; Shea & Chesson, 2002;

7 1.5 Nonnative species in coastal ecosystems

Maron & Marler, 2007) and the ability to escape local predators (Shea & Chesson, 2002). Disturbance is one method by which nonnatives take advantage of habitat space in their newly invaded environment (Glasby et al., 2007; Piola & Johnston, 2007). For example, when a site is disturbed, naturally (i.e., natural-fall log strikes in the intertidal) or human- mediated (i.e., habitat development, dredging), resources previously used by the local biological community become available and the site can become vulnerable to invasion (Davis et al., 2000). Resource and habitat availability may also be associated with high native diversity and reflect areas with increased niche-opportunity that may support regions of high nonnative diversity (Shea & Chesson, 2002) or may result in biotic- resistance, preventing the establishment by limiting resource availability for nonnative species (Tilman, 1997; Stachowicz et al., 1999). Invasion success is difficult to predict as abiotic and biotic conditions are stochastic in nature and vary across space and time (Menge & Sutherland, 1987).

Characteristics of the invader, such as diet, dispersal, and body size, are also important for a species’ survival at its new location. These characteristics are thought to contribute to the success of a nonnative in finding a niche in its new ecosystem (Kolar & Lodge, 2001). The speed of colonization after resources become available is an example of species characteristics contributing to invasion success. Nonnative tunicate species, such as Botrylloides spp., can have rapid colonization and growth rates that contribute to their ability to out-compete local tunicates (Stachowicz et al., 2002a). Facilitation, or a positive interaction between one nonnative and another, provides another mechanism assisting invading species to arrive and survive in their invaded ranges. The nonnative hull-fouling bryozoan, Watersipora subtorquata, is tolerant of many anti-fouling biocides and is commonly found on the hulls of ships arriving in Queensland, Australia. Floerl et al. (2004) demonstrated that areas colonized by this species provide a foundation for fouling organisms less-tolerant to biocides and facilitates the transport of other species at greater abundance than would otherwise be possible (Reviewed in Simberloff, 2006).

Understanding nonnative vectors and limitations to nonnative survival can improve targeted management action to prevent future invasions. For example, if species are

8 1.6 Seagrass and nonnative impacts in British Columbia, Canada arriving via ship ballast, management may be able to require ballast exchange for all ships entering coastal waters. This process would require that regulations were in-place that enable ballast exchange to occur in their jurisdiction, which would entail consideration of international, national and provincial legislation and regulations (Dahlstrom et al., 2011). Integrating research on nonnatives into coastal management would better support the development of this legislation, however, integrating research on nonnative interactions into management is a challenge as nonnative species are novel to the systems they invade and research is likely to lag behind the needs of managers for developing prevention and control efforts. Because of this lag it is important to consider lessons from other regions that have already dealt with similar species risks to better understand how nonnatives might affect services and how to prepare for and reduce these impacts (Occhipinti-Ambrogi & Savini, 2003).

1.6 Seagrass and nonnative impacts in British Columbia, Canada In British Columbia (BC), the native seagrass, Zostera marina (hereafter eelgrass), is important for its role as nursery grounds and habitat to commercially and recreationally important fisheries species (Hemminga & Duarte, 2000). Species, such as juvenile salmon, spawning herring, and Dungeness crab, represent a major portion of the seafood industry on which coastal communities throughout BC rely (GSGislason & Associates Ltd., 2007). Herring use eelgrass as spawning habitat (Haegele et al., 1981); herring roe alone had an annual value of no less than $40 million CAD between 1982 and 1993 (Department of Fisheries and Oceans Canada, 1994). Some legislative efforts have been made to protect eelgrass—for example the Canadian Fisheries Act protects fish habitats (section 35(2)), and encourages the use of natural habitats in the place of artificial, such as when maintaining seagrass for erosion control rather than replacing seagrass with retaining walls (Regional District of Nanaimo Policy B1.9). However implementation of these policies is difficult in BC where funds for protecting biodiversity and habitat will decrease by $60 million CAD between 2011 and 2012, from $148.4 to $89.1 million CAD, though available financial resources for Canada increased from $872.1 to $997.6 million CAD over this same time period (Environment Canada, 2012).

9 1.7 Management of marine nonnative species

The additional impacts of nonnative species are likely to exacerbate already threatened eelgrass habitats and reduce the benefits associated with these diverse communities (Hughes et al., 2008). Understanding the implications of nonnative species is important when considering coastal ecosystem services (Ruiz et al., 1997). However, the study of marine nonnative species and their impacts has been limited in BC’s eelgrass beds. Studies include the following: 1) Quayle (1964) reviewed the presence of the Manila clam Venerupis japonica, a nonnative species found in eelgrass, during his studies of marine molluscs in BC. However this author did not review any information on eelgrass specifically. 2) Observational studies of the biogeography and benthos of Roberts Bank and Boundary Bay in the southern Strait of Georgia that sampled nonnative molluscs, polychaetes, and seagrass, Zostera japonica, (Banse, 1981; Swinbanks & Murray, 1981) while Harrison’s (1979) work in the same region tested the interaction of nonnative seagrass Z. japonica and native invertebrates and eelgrass. And 3) more recently, a study in eelgrass beds on the west coast of Vancouver Island demonstrated that invasive tunicates are facilitated by the invasive alga, S. muticum, which grows on the hard substrate of native clam shells in these otherwise soft-sediment habitats (White & Orr, 2011). In BC, which species have invaded eelgrass beds, how they invaded, and their potential impacts are generally unknown.

1.7 Management of marine nonnative species Management of nonnative species in Canadian marine systems involves connecting regulations suggested at international and regional scales to national scale policies in place to prevent, reduce and manage the introduction of unwanted organisms. International policies have been established, such as for member countries of the World Trade Organization, that allow countries undergoing trade to establish health standards, surveillance programmes and arrival responses (World Trade Organization (WTO), 1995; Dahlstrom et al., 2011). In addition, regional policies, such as the North America Free Trade Agreement (NAFTA) promote environmental protection while increasing Canadian trade with the United States and Mexico (McLachlan, 2005). These trade regulations and their recommended risk-assessments that promote reduced species introductions are then implemented at national scales. Accordingly, if nations agree to the

10 1.7 Management of marine nonnative species international and regional regulations, the national policies developed must include those from international, regional and national levels.

Specific standards regarding biosecurity—the prevention of introduction of organisms that might result in environmental or economic harm—are enacted at the national scale to mitigate the risks of foreign trade. Risk-assessments have been developed that address concerns of and manage the international and regional policies and also specifically protect local waters. Canadian ports, for example, require ballast flushing before international shipping freighters can enter Canadian waters (Transport Canada, 2006), an arrival response that was established to prevent new introductions via ballast water vectors and that is supported by the WTO. However there are many barriers to implementation of risk assessments at the national scale. For example, few of the pertinent international regulations are legally binding, which makes it difficult to foster countries complying with recommendations regarding nonnative prevention. In addition, while most international, regional, and national policies encourage the development of a scientific understanding of invasive management, i.e., how unwanted organisms arrive and their potential impacts, little of this information has been included in developing risk assessments. In addition,risk assessments have not been frequently used to develop these policies (Dahlstrom et al., 2011).

To reduce this barrier, the Department of Fisheries and Oceans Canada (DFO), which develops marine policy and manages for coastal protection and national fisheries regulations at the federal level, established the Canadian Aquatic Invasive Species Network in 2006 (CAISN, 2011). CAISN has since developed risk assessment models for potential and existing nonnative species, and as a network, it improves the connection between the science of nonnative species and their management in Canada. By improving the quality of risk assessments for managing nonnative species, future species introductions will be limited, and when species are introduced Canada will be better prepared to prevent and mitigate their impacts.

11 1.8 Thesis overview

1.8 Thesis overview My dissertation research objective, to investigate and advance the capacity of existing scientific knowledge on anthropogenic impacts to inform coastal ecosystem-based management, was accomplished through quantitative literature analyses in Chapter 2 and 6 and by creating a baseline of data on nonnative species in BC seagrass beds – testing vectors and environmental limitations to nonnative spread (Chapters 3 and 4) and evaluating future impacts of the invasive European green crab (Chapter 5). In Chapter 7, I summarize my findings and discuss future directions for research on impacts to seagrass and the development of research questions for EBM.

In Chapter 2, I investigate the comprehensiveness of research available to inform the evaluation of human activities on natural ecosystems. I test for limitations to such research by reviewing the evidence for environmental-impact pathways that connect activities such as urban development or agriculture, to their impacts (or stressors), such as increased nutrient or waste inputs in near-shore ecosystems. These stressors affect ecosystem service providers (e.g., shellfish or seagrass), and associated ecosystem services (e.g., fisheries, erosion control, nutrient filtration, carbon sequestration). In the review I explicitly characterize the extent to which primary research (experimental and observational studies) test the effects of anthropogenic activities’ direct and indirect impacts on estuarine ecosystem providers and their production of services.

In Chapters 3 and 4, I report the first broad-scale study on nonnatives in seagrass in BC. In Chapter 3, to study the relative importance of vectors and environment on establishment of nonnatives in eelgrass beds, I conducted biological sampling of eelgrass beds along the BC coast. I compared richness and abundance of nonnative benthic, epifaunal and large mobile macroinvertebrates, and some algae to determine which of two arrival vectors (shipping and aquaculture) and environmental drivers (climate variables, human population density, and native richness and abundance) help explain the distribution of these species. In Chapter 4, I answer the question of whether any nonnative species sampled have expanded beyond their known ranges by describing the apparent expansion of the nonnative bamboo worm Clymenella torquata, last formally

12 1.8 Thesis overview reported from the northeast Pacific coast more than 30 years ago, collected during field sampling for Chapter 3.

In Chapter 5, I assess the potential impact of the invasive European green crab, Carcinus maenas, on shellfish harvest in Puget Sound, Washington State. Green crab has now invaded the outer coasts of Washington and Vancouver Island, BC and is predicted to invade Puget Sound and the Strait of Georgia in the near future via human vectors or natural dispersal. These crabs are voracious predators that feed on juvenile shellfish of several species and have severely reduced softshell clam fisheries in the western Atlantic. I assess the hardshell clam, mussel, and oyster harvest biomass and revenue value-at-risk to a set of plausible scenarios of green crab invasion, using benefit transfer methods to estimate the economic loss likely associated with this invasion. Furthermore, I characterize the extent to which variation in key parameters influenced the resulting value-at-risk.

In Chapter 6, I evaluate the extent to which research on impacts of nonnative species meets the needs of management. Management of nonnative species requires an understanding of the direction and magnitude of nonnative impacts on other species’ ecosystem structures, processes, and services. However, research that occurs in isolation of management goals may not provide understanding needed by policymakers and practitioners attempting to protect native species and human activities associated with ecosystem services that are threatened by invasions. To study knowledge gaps in nonnative species research for management, I reviewed the impact of the nonnative Japanese eelgrass, Zostera japonica, on local species and communities and on abiotic environments in the northeast Pacific.

13 2.1 Synopsis

Chapter 2 Research gaps in evaluating tradeoffs in ecosystem-based management

2.1 Synopsis There is increasing support for ecosystem-based management (EBM) to connect the multiple human activities to their cumulative impacts on ecosystems, which provide critical benefits to people through ecosystem services. To meet this challenge, managers must evaluate tradeoffs between various objectives, in part based on an understanding of the impacts to ecosystem services from multiple human activities and their associated stressors. Building on proposed research frameworks, I argue that the concept of ‘impact- pathways’ underpins the evaluation of tradeoffs (the pathway from activities to impacts, to ecosystem service providers, to ecosystem services, to benefits). In order to assess the evidence base for trade-off evaluation in EBM, I review the evidence for impact- pathways using estuaries as a case study, focusing on seagrass and shellfish. Keyword searches of peer-reviewed literature revealed 2380 studies for select impact pathways, but closer inspection demonstrated that the vast majority of these made connections only rhetorically, and only 4.6% (based on a subset of 250 studies) provided concrete evidence of complete impact-pathways from activities to ecosystem services. Furthermore, none of the subset of studies tested pathways based on metrics of ecosystem services that are specifically important to beneficiaries. I conclude that if an understanding of impacts of human activities on ecosystem services is necessary to appropriately evaluate trade-offs further research is required, though current research can support a risk-based, precautionary, adaptive approach critical for EBM.

2.2 Introduction Despite the importance and irreplaceability of coastal ecosystems and the benefits these ecosystems provide to people (i.e., ecosystem services; Daily, 1997; Millennium Ecosystem Assessment, 2005; Barbier et al., 2011), degradation and loss of ecosystems at the land-sea interface is intense and increasing worldwide (Worm et al., 2006; Halpern et al., 2008). Coastal ecosystems are faced with multiple and interacting environmental-

14 2.2 Introduction change drivers and the degradation of biodiversity and ecosystem functions of coastal habitats may contribute to loss of ecosystem services, such as reduction in viable fisheries, declining water quality, and decreased coastal protection from flooding and storm events (Hoffman et al., 1984; Carpenter et al., 1998; Valiela et al., 2001; Salomon et al., 2010). The need for an integrated approach to management that incorporates the entire ecosystem, including cumulative impacts, linkages between ecosystems and human systems, and the range of benefits provided to humans by these ecosystems has led to an increasing amount of support for EBM of these systems (McLeod et al., 2005; Levin & Lubchenco, 2008; Granek et al., 2010; Ruttenberg & Granek, 2011). EBM differs from current approaches that focus on single-species, ecosystems, or activities by integrating ecological productivity with human well-being and considering the cumulative impacts of many human activities on the production of ecosystem services, in an effort to maintain ecosystems in healthy, productive and resilient condition (McLeod et al., 2005). Given its place-based focus on ecological interactions, EBM offers an unprecedented major role for ecological research within decision-making.

While EBM is moving forward globally, its use varies according to the strength of governance and availability of data (Leslie & Mcleod, 2007; Tallis et al., 2010). Researchers agree that there is enough scientific understanding of ecosystem services and their value to begin integrated multi-species management across ecosystem boundaries (Price et al., 2009; Lester et al., 2010; Gregr & Chan, 2011). However, whether there is a scientific understanding of human activities’ impacts on ecosystems and their provision of ecosystem services, and interests of people (beneficiaries), is another question (Leslie & Mcleod, 2007; Granek et al., 2010). To develop effective management action at a multi-ecosystem scale, managers need ecological data to consider tradeoffs between gains from anthropogenic activities that impact ecosystem service providers and the resulting losses of ecosystem service supply, provision and endpoint values3 that matter to beneficiaries (Granek et al., 2010; Tallis et al., 2010; Tallis et al., 2012).

3 The value attributed to the ecosystem services provided by a given ecosystem.

15 2.2 Introduction

The impacts of anthropogenic activities frequently occur across ecosystem boundaries: nutrients run off into coastal estuaries from upstream agriculture and urban centres, sedimentation and water flow are altered by damming and logging, urbanization produces polluted waste-water with downstream impacts, challenging management across these systems (Jiao et al., 2007; Calabretta & Oviatt, 2008). Given the prevalence of cross- system linkages, through both biological and physical interactions, managing these impacts requires information on processes that span ecosystems traditionally managed and studied as independent systems (Ruttenberg & Granek, 2011). In the absence of data that support these ecosystem relationships, evaluation of management strategies that cross ecosystem boundaries may be unsubstantiated and uncertain (Kremen et al., 2007; Halpern et al., 2009).

Though it is essential to assess ecosystem service supply (the biophysical functions of an ecosystem to produce a service, regardless of whether humans use or value the particular outputs), service provision is determined by the value attributed to it by beneficiaries. In making decisions about tradeoffs between services, it is necessary to understand the preferences of beneficiaries in regards to service endpoint value (Daily, 1997). For example, if urban development releases waste-water to an estuarine ecosystem, and this results in increased nutrient levels and pathogens, it is possible that some service providers will benefit while others are negatively impacted, both increasing production of some ecosystem services and decreasing others (Hershner & Havens, 2008). How wastewater is managed will then be determined by the value that beneficiaries placed on the change in service provision. Without knowledge of the impacts of an anthropogenic activity on the benefits of an ecosystem service, it is not possible to suggest informed management alternatives for protecting that service (Granek et al., 2010).

Recent efforts to address impacts on service production include frameworks that integrate information on ecosystems, ecosystem service providers, and ecosystem service beneficiaries (Kremen, 2005; Granek et al., 2010; Rounsevell et al., 2010). An EBM framework can provide the dual utility of informing how managers might use science of ecosystem impacts to improve evaluation of tradeoffs, and apprising scientists in how

16 2.2 Introduction their research can apply to improving EBM. However, making these frameworks operational within the context of conservation and management remains a considerable challenge because of their perceived complexity (Tallis et al., 2010). To increase the functionality of earlier EBM frameworks, I developed a conceptual model that simplifies relationships between anthropogenic activities and impacts, production and valuation of services and management response for EBM at the scale of an ecosystem (Figure 2.1). This model is the structure by which I evaluate the comprehensiveness of current science of impact-pathway relationships for evaluation of tradeoffs in this review.

Figure 2.1 Activity-Impact-Provider-Service (AIPS) EBM framework for understanding the impacts of multiple human activities on ecosystem services in coastal ecosystems (adapted from Rounsevell et al., 2010). The AIPS framework connects anthropogenic activities exogenous and endogenous to an ecosystem to their direct and indirect impacts on ecosystem providers within that ecosystem. A change in the state of an ecosystem service provider results in a change in supply of its services, and these alter the provision of ecosystem services to people who value them, beneficiaries. Beneficiaries then motivate management and policy responses, based on their value of services, to regulate activities within the ecosystem, mitigate impacts the activities produce or support adaptation of the ecosystem service provider. Management within the ecosystem may additionally influence exogenous drivers of change that regulate activities both exogenous and endogenous to the ecosystem. These drivers may be management and policy responses, for example, local or state management at the scale of an ecosystem may have influence on regulatory decisions made at the federal level. Drivers of change may also be societal pressures that influence activities production of impacts or more long-term natural system drivers such as natural climate cycles (i.e., El Niño/Southern Oscillation, Pacific Decadal Oscillation) (Mantua & Hare, 2002; Stenseth et al., 2003).

17 2.2 Introduction

Estuaries are some of the most threatened coastal systems globally due in part to the position of estuaries at the interface between land and sea (Simenstad et al., 1992; Stoms et al., 2005; Halpern et al., 2009; Beger et al., 2010). These habitats are threatened by human activities from terrestrial, freshwater and marine sectors as well as those within the estuary itself, which degrade the myriad of services estuaries provide (Figure 2.2; Barbier et al., 2011). Despite this suite of impacts, estuaries are a centre of service provision due to their roles in nutrient cycling, water quality maintenance and as nursery grounds for coastal fish and invertebrate species (supporting fisheries provisioning), (Hemminga & Duarte, 2000; Granek et al., 2010; Barbier et al., 2011).

Thus, for decision-makers and managers to regulate impacts to estuarine systems they must engage across institutional jurisdictions to identify major drivers and activities. A comprehensive and consistent understanding of these estuarine ecosystems for EBM would specifically link activities (both exogenous and endogenous to an estuary) to the impacts those activities have on ecosystem service providers, and the subsequent change in services and endpoint values. Of course, ecological research can be of great value to EBM without characterizing impact-pathways in part or in full; the purpose here is not to assess the utility of such ecological research for EBM, but rather to aid in the evaluation of tradeoffs between ecosystem services, a task widely perceived as central to EBM (McLeod & Leslie, 2009; Lester et al., 2010).

Given the opportunities for improved EBM from understanding impact-pathways, the aim of this paper is to assess the comprehensiveness of research that connects the impacts of anthropogenic activities to changes in ecosystem service production and gaps in global distribution of this research. I tested for research comprehensiveness by conducting a literature review based on the AIPS EBM framework (Figure 2.1) using estuaries as a case study. This literature review specifically tested for studies that connect activities and impacts on service providers and their service supply, provision and value. In the review I explicitly considered primary research (experimental and observational studies) on anthropogenic activities that cause direct and indirect impacts to estuarine ecosystem providers. Providers can include habitats and whole communities however in this review

18 2.2 Introduction

I focus on single species, and their services. Because a comprehensive assessment was impractical, I selected representative activities (logging, dams, shipping, and aquaculture) from each of four sectors (terrestrial, freshwater, marine, and estuary) that have impacts within estuarine ecosystems (as described in Table 2.1). In addition, I limited the study to two commonly studied estuarine ecosystem service providers, shellfish and seagrass, and their production (supply, provision and endpoint value) of ecosystem services.

19 2.2 Introduction

Figure 2.2 Impact-pathways for estuaries at the intersection of terrestrial, freshwater and marine ecosystems. Anthropogenic activities in each ecosystem produce stressors that directly and indirectly impact ecosystem service providers, altering the supply of services and endpoint value. Not included in these pathways are natural and societal external drivers of change (i.e., natural climate cycles, climate change, grassroots scale management, etc.), the many additional activities, impacts, and service providers (which include microbes, birds, and other species that comprise estuarine ecosystems), and endpoint values represent only an example of possible ecosystem benefits. pbt= persistent bioaccumulative and toxic chemicals; seq’n = sequestration

20 2.2 Introduction

Table 2.1 Subset of activities and impacts from Figure 2.2 included as search terms in this review. An activity from each of the four sectors (terrestrial, marine, and freshwater, and from within estuaries) with impacts in estuaries.

Ecosystem Activity Impact Example of each activity’s impact stressor Estuary Aquaculture Habitat Modification Estuary converted to shrimp aquaculture (Paez-Osuna, 2001) Impact to wildstocks Transmission of pathogens from net-pen aquaculture to wild salmon (Naylor et al., 2005) Nonnative species Nonnative species vector (Cohen & Zabin, 2009) Nutrients Nutrient inputs from waste (Findlay et al., 1995) Organic/Inorg. Pollutants Increased pollutants and antibiotics (Tovar et al., 2000) Pathogens/Microbes/Disease Transmittance of MSX oyster pathogen (Ulrich et al., 2007) Terrestrial Logging Sedimentation Conversion of forest to agricultural land increased the rate of sediment runoff (Hewawasam et al., 2003). Runoff Altered waterflow (Kimmerer 2002) Temperature Runoff from logged riparian zones increased downstream temperature (Davies & Nelson, 1994) Marine Shipping Nonnative Species Nonnative species vector (Ruiz et al., 1997) Organic/Inorg. Pollutants Anti-fouling hull paint pollutants (Nehring, 2001) Dredging Dredging ports for large freighters (Short & Wyllie-Echeverria, 1996) Pathogens/Microbes/Disease Ships carry pathogens and microbes into estuaries (Drake et al., 2007) Freshwater Dams Runoff Altered freshwater flow, change outflow location (Sklar & Browder, 1998) Sedimentation Sedimentation decreased (Ibàñez et al., 1998)

Inorg. = Inorganic

21 2.3 Methods

2.3 Methods I conducted a global literature search to assess comprehensiveness in estuarine research on impact-pathways, which connected activities and their impacts, from terrestrial, marine, freshwater and estuarine ecosystems, to changes in providers and their services, and ultimately the effects on endpoint benefits (Figure 2.1). First, I selected one anthropogenic activity from each of the four ecosystems: logging (terrestrial), dams (freshwater), shipping (marine), shellfish aquaculture (estuary). I then searched across impacts (Table 2.1) caused by these activities and changes in the services of shellfish (search terms: shellfish, oyster, clam, mussel, crab) and seagrass (search terms: seagrass, eelgrass) ecosystem service providers. Keyword search for: estuar*, activity, impact, service provider, service; complete list of activities and impacts in Table 1 and the connection of those impacts to providers and services in Figure 2.2. Searches were all conducted in January 2011.

Each search was conducted twice. I used Web of Science and Biosis databases to search for articles that included keywords for impacts, providers and services (IPS). I then repeated the same search, this time including an activity (AIPS). The searches were only of peer-reviewed research and did not include grey-literature, such as environmental impact assessments and government reports that may quantify impact-pathways in estuaries. These two searches resulted in two suites of studies (AIPS a subset of IPS) that included keywords for linking IPS and AIPS studies in estuaries. For example, to search for studies that linked dams, sedimentation, seagrass and services (each of the 11 services) I 1) searched articles for: sediment* AND estuary*, subsearch for eelgrass AND seagrass, subsearch for each of the 11 services, resulting in all papers including IPS search terms, in this case sediment-seagrass-services, then 2) reran step 1) but added activity (the activity of each impact as listed in Table 1) to the keyword search. For example, for the impact “sedimentation” resulting from damming I searched for dam* AND sediment* AND estuary*, subsearch for eelgrass AND seagrass, subsearch for the 11 services, resulting in all papers including AIPS search terms, in this case dams- sediment-seagrass-services. Only those studies with abstracts were included in the

21 2.3 Methods database. Papers were included in the results as multiple studies if they included more than one activity, impact, or service provider.

Using the 2380 IPS and 253 AIPS studies that resulted from the search I processed articles for complete IPS impact-pathways by evaluating abstracts to ensure the research in each study actually made the intended AIPS or IPS connections and did not only discuss the search terms. For AIPS I processed all resulting studies. Each abstract was evaluated by ensuring the study was on the correct activity (aquaculture, shipping, dams, logging), that the activity caused the impact, and the impact was on the seagrass or shellfish provider.

For the IPS search I randomly selected a subset of 250 (~10%) of the 2380 results to process. I calculated a margin of error for the subset of IPS studies (sample size = n) for a 95% confidence interval (Levy & Lemeshow, 2008):

where z is the z-score from a standard normal distribution associated with a two-sided alpha level of 0.05 (i.e. 1.96), N is the size of the entire set of results (2380) from which the sample is to be drawn, P is the proportion of false positives assumed (estimated at 0.5). The margin of error (ε) was calculated as ± 12%.

I processed the abstracts of 250 of the IPS papers in the same manner as AIPS, randomly selected from the total pool of IPS studies. I ensured the impact was correct as defined by the search terms and the impact was on the seagrass or shellfish provider. Providers in both IPS and AIPS studies were then evaluated for a change in service supply and/or change in provision, and if that change in provision was linked to an endpoint value or impact to a beneficiary (as defined in Tallis et al., 2012). All studies that made full IPS and AIPS impact-pathways was termed accurate.

22 2.4 Results

Following this processing method I extracted additional information on study method (experiment, observation, review, model) and location of study, which I classified according to World Wildlife Federation ecoregions (Spalding et al., 2007). Papers that reviewed or modeled connections were noted but not included in the analysis, as these papers do not test a quantifiable change in the estuarine ecosystem.

To estimate total accurate IPS studies of the 2380 papers, I compared the ratio of accurate to inaccurate studies of the 250 processed shellfish and seagrass papers to the total papers. Then calculated the 12% margin of error for the proportion of accurate studies based on a 95% confidence interval from my earlier calculations of sample size. This extrapolation method likely underestimates rare IPS connections, thus I consider this a conservative estimate of IPS.

2.4 Results The search of estuarine studies resulted in 2380 studies connecting impacts to ecosystem service providers and their services (IPS); 1921 on shellfish and 874 on seagrass. When activity was considered, the search results were reduced to 253 AIPS studies; 231 shellfish and 40 seagrass studies. Of the 250 IPS papers processed (10% of total search), only 10 studies addressed IPS when considering change in supply of a service, two quantified a change in the service, and none addressed endpoint values.

For shellfish, 9 of 158 papers completed an IPS impact-pathway for a change in service supply (5.7% of the subsearch), of those, two quantify a change in service provision (1.3% of the subsearch). When I extrapolate to the total search results (1921 studies), there are 86 ± 28 papers that make the IPS connection for a change in service supply (supply of a service regardless of whether humans use or value it, i.e., oysters filter toxins), 24 ± 13 of which tested change in service provision (supply of a service at the scale at which humans use or value it, i.e., how much cleaner the water is when oysters filter toxins).

23 2.4 Results

For seagrass, 1 of 92 papers completed an IPS impact-pathway for a change in supply of a service (1.1% of the subsearch), and none of the papers quantified a change in service provision or endpoint values. By extrapolation, there are 10 ± 9 papers that make an IPS connection that include seagrass as a service provider. No papers processed test a change in service or endpoint value, thus I estimate that no studies from the total IPS batch make the connection, although this is a conservative estimate as I was unlikely to include extremely rare events.

The Atlantic and Gulf coasts of North America are the most commonly studied region for IPS studies, with six studies (60%) conducted in the NW Atlantic (Figure 3), two studies conducted in the NW Pacific, one in the NE Atlantic and one in the SW Atlantic.

None of the 253 papers resulting from the search terms connecting an activity to its impact on ecosystem service providers and their services (AIPS) completed an impact- pathway to a measured change in service supply, service provision or endpoint value. One study developed a framework that connected dams producing sedimentation and water flow impacts to shellfish and their services (Richter & Thomas 2007).

24 2.5 Discussion

Figure 2.3 Map of global ecoregions (numbered 1 – 62; Spalding et al., 2007) demonstrating regions with complete impact-pathway studies. Squares represent the number of studies including an impact on ecosystem provider and change in service (IPS) impact-pathway in each ecoregion (for scale, the largest squares are in regions 5 and 6, and each had 3 IPS studies), color represents ecosystem provider: red squares are shellfish studies, green squares are seagrass studies. Regions with IPS studies: Mediterranean Sea (4), Cold Temperate Northwest Atlantic (5), Warm Temperate Northwest Atlantic (6), Cold Temperate Northwest Pacific (8), Warm Temperate Southwestern Atlantic (47). No studies were found that connected the impact of a specific activity to an ecosystem service provider and a change in service (AIPS) impact- pathway.

2.5 Discussion 2.5.1 Knowledge gaps for evaluating tradeoffs within EBM Peer-reviewed research on impact-pathway relationships (Figure 2.2) in estuaries rarely included the entire pathway connecting cross-system impacts to changes in ecosystem service provision. No studies on seagrass or shellfish were found to connect impacts from activities to a change in service supply, provision or value. Because impacts can be the result of many activities and research spanning the land-sea interface is known to be difficult and limited (Ruttenberg & Granek, 2011), I simplified the search by removing activity and only evaluating impacts to service provision; yet, this reduced search still yielded limited results. There were no studies that evaluated a change in ecosystem service value based on an impact, though a few studies included the effect of an impact

25 2.5 Discussion on ecosystem service supply and provision. With so few studies connecting these pathways there are huge expanses of the world’s coasts where these relationships are not understood; and where understanding cannot be borrowed reliably from other systems (Connell & Irving, 2008; Hessing-Lewis et al., 2011). When considering the metrics of ecosystem services that actually matter to relevant beneficiaries, such as a value change, there were no studies that test these relationships in this review. Studies address metrics that may be salient for EBM, but do not necessarily connect to endpoint values.

2.5.2 Why does understanding the complete impact-pathway matter? In the absence of information on full pathways, researchers and managers may attempt to evaluate tradeoffs between estuarine impacts and ecosystem services by making generalizations based on studies of single-steps. Connecting a study on the activities’ impact on an ecosystem service provider to a different study addressing the change in that provider on ecosystem service provision (Thompson & Rueggeberg, 1989). There is a great potential for misalignment when estimations are made that connect single-step studies measuring impacts using various biometrics and across geographic regions (Halpern et al., 2009). For example, it may be possible to evaluate the effect of sedimentation on the nursery-habitat provision service of seagrass for herring by connecting a study of sedimentation effects on seagrass with a study of the habitat provision service of seagrass.

Efforts to connect an activity to an ecosystem service through multiple studies are likely to be compromised by the numerous metrics used to measure the impact to provider, such as a change in seagrass area (e.g., areal extent of seagrass), which may or may not correspond with the metrics used to measure the provider to service, such as seagrass in habitat provision studies (e.g., number of eelgrass shoots per meter). Misalignment of metrics is likely as most studies investigate only a small subset of potential metrics. These studies may also take place over different seasons or years and in different regions that may affect habitat use in unique ways. Accordingly, several great leaps must be made to characterize the effect of an impact on an ecosystem service.

26 2.5 Discussion

Despite biometric and regional differences, connecting single-step studies can be useful for risk-based estimates that incorporate uncertainty of the misalignment of studies (Casman et al., 1999). Previous research in Chesapeake Bay, USA has demonstrated that in regions with funding for extensive ecological research over long periods of time, ties between managers, science, and government have resulted in research that can connect these single-studies over an entire impact-pathway (Boesch & Goldman, 2009). However, when compared to regions where research exists only in piece-meal with little support from management, connections between single-studies can only be estimated from other regions (Tallis et al., 2010). Permitting general risk-estimates that do not include context-dependent differences between regions and biometrics of study (Connell & Irving, 2008; Kordas, 2009). This is not to say that these best-estimate scenarios are not valuable, only that they should be used cautiously, with the realization that they do not necessarily reflect the ecosystem-specific biotic and abiotic relationships. The assumption that studies align across biometrics can misrepresent the dynamic natural system and result in huge uncertainties, many of them implicit in ecosystem models (not acknowledged in outputs), which is extremely problematic for developing projections and evaluating tradeoffs for protecting ecosystem services (van Asselt & Rotmans, 2002).

Of course, research can be relevant for EBM without connecting all the way to changes in ecosystem services. For example, impacts on a provider species can pose a real risk to its production of services, in the absence of a demonstrated change in the provider. When sedimentation increases as a result of coastal development and seagrass beds are smothered, a net loss in services can be assumed even if researchers have not recorded that relationship. Given changes in industry, shifting preference in research and political support, and limited capacity for funding and time, filling these data gaps completely is likely impossible, and as such, management will always be operating under uncertainty (Oreskes, 2004). When uncertainty is too great to use single-linkage studies to evaluate tradeoffs between gains from activities and their impacts and losses in service provision, chains of single-linkage studies can be used to develop a “test bed” to examine the relative robustness of alternative management strategies (Casman et al., 1999). Such widespread and pervasive uncertainties are relatively less problematic in a risk-based

27 2.5 Discussion approach to management (as in Levin et al., 2009) than an approach that requires impacts to be demonstrated conclusively before management intervention.

Evaluating changes to ecosystem service providers is not equivalent to showing that changes to providers have implications for services (Tallis et al., 2012). One of the most difficult parts of evaluating how an ecosystem service provider contributes to a service is identifying the portion of the end service to which the provider contributes. Furthermore, simply stating that a provider produces a service is not sufficient, it is necessary to know what that change means at the pertinent ecological scale. For example, water filtration as a service is the net activity of a community not individual species, and the removal of the filtration capacity by a single organism or even a population of organisms does not necessarily translate into a loss of filtration capacity in a region, or more importantly, a noticeable decline in water quality from the perspective of human uses (e.g., swimming, fishing and eating shellfish). It is at this broad scale that EBM can then be used to evaluate tradeoffs between activities and their impacts on the network of ecosystem service providers and their influence on endpoints that people value.

2.5.3 Linking science and management to inform EBM Many of the studies evaluated during this review discuss the relationship of how an activity is likely producing an impact, proceed to test the impact on shellfish or seagrass ecosystems, and follow with a discussion that changes in service providers are likely to affect service provision (Kirby, 2004; Weimin et al., 2006; Korajkic et al., 2009; Lindqvist et al., 2009). In other words, papers in this review are discussing the topics but often not testing the linkages in a way that can inform impact to service studies. These frequent discussions of activity and service in terms of studying impacts to service providers suggest that understanding the full relationship is important to researchers. While not all estuarine research needs to contribute to application (Stokes, 1997), discussion of these studies’ further application demonstrates that researchers are interested in producing research that connects to anthropogenic impacts on ecosystem services.

28 2.5 Discussion

If coastal managers want to continue on a path to use EBM for evaluating trade-offs among objectives, it is necessary that at least some research be developed that integrates across activities occurring in interacting ecosystems and their impacts on ecosystem service provision. Furthermore, this research should be place-based, at a scale that takes into consideration context-dependent effects in the local system (Hessing-Lewis et al., 2011; Bulleri et al., 2012). Regional differences in ecological patterns and processes can create contradictory outcomes for EBM as ecosystem function varies over small spatial- scales resulting from differences in, for example, temperature, nutrient upwelling, and habitats (Menge & Sutherland, 1987). In regions where nutrient levels are low, an invasive species may compete for nutrients negatively impacting native species, alternatively in regions with upwelling or high nutrient runoff these same species may benefit native species by reducing nutrients and subsequent eutrophication (Hershner & Havens, 2008). Of the few impact-pathways sampled in this study most research took place in North America along the western Atlantic and Gulf of Mexico coast, thus most impact-pathways represent relationships studied in those geographic regions ( Figure 2.3). If tradeoffs are to be evaluated using informed impact-pathways, given the site-specific nature of many anthropogenic impacts, managers will likely benefit from increased research of local pathways in other regions of the world (Connell & Irving, 2008).

2.5.4 Limitations to informing the impact-pathway Why don’t these full impact-pathways exist? Determining which anthropogenic activities cause specific impacts is a challenge. One impact can be the result of many activities, as in nutrient runoff in an estuary. Nutrients may be the result of agriculture, wastewater and urbanization, and understanding how much the total nutrient flux in an estuary is due to these activities is necessary before the impact each those activities have on ecosystem services is understood. Difficulties also lie in creating scientific research plans that cross ecosystem boundaries. This cross-boundary research requires interdisciplinary collaboration or expertise in multiple fields of study that are challenging to facilitate. Future work does not necessarily need to start from scratch. In regions where single-steps between impact and provider, or a provider and service provision have been studied,

29 2.5 Discussion research can be organized to appropriately connect within those systems and metrics of data collection.

While the results of this review are limited to estuaries and include a limited set of search terms for seagrass and shellfish in the peer-reviewed literature, the connections selected for this review are some of the most critical and widely acknowledged for estuarine systems (Stoms et al., 2005; Tallis et al., 2008; Barbier et al., 2011) and are imperative for predicting changes in service provision for these coastal ecosystems. Though I expected there to be knowledge gaps when it came to some of these impact-pathways (such as those starting with dams and logging activities), surprisingly, data on many pathways do not even exist in the literature. It is possible that the imposed boundaries of my literature search resulted in a failed attempt to capture these relationships. For example, if this review included ecosystem impact assessments (EIA) not found in the peer-reviewed literature that evaluate human use of coastal ecosystems, there may have been more studies connecting human activities to ecosystem service providers and possibly service supply (International Association for Impact Assessment, 1999). However, EIAs are not necessarily designed to measure a change in ecosystem service provision that is necessary to complete impact-pathways for ground-truthing tradeoff estimates in real life scenarios.

2.5.5 Conclusions This review of the impacts on service provision associated with human activities demonstrates an impressive gap in research for evaluating tradeoffs using EBM. There are frequent calls for models of such impacts, and I agree with the importance of these, but my research reveals that modeling exercises currently can only be based on exceedingly sparse data knitted together with suites of assumptions that manifest in huge uncertainties, the vast majority of which is inaccessible to decision-makers. As managers attempt to evaluate tradeoffs made under various management scenarios they ignore the limited understanding of the whole system of relationships that begin at the level of an anthropogenic activity and end in the changing of an endpoint value. Instead fragments of studies are pieced together from around the world to establish a “best guess” (Nelson et

30 2.5 Discussion al., 2009; White et al., 2012). It is a great start, but testing tradeoffs based on estimated relationships is overly simplistic and ignores that the majority of the relationships being estimated are not well-understood (Casman et al., 1999).

So where does that leave EBM? Moving forward with EBM is a priority and many aspects of the science to support EBM are already available (Leslie & Mcleod, 2007; Lester et al., 2010). Though scientific uncertainty allows managers a margin of judgment in decision-making, to reduce uncertainty it is necessary that scientific research is directed by management context to fulfill specific needs (Daily et al., 2009). I have demonstrated that if an understanding of impacts of human activities on ecosystem services is necessary to appropriately evaluate trade-offs, more research is required. EBM may be better used as a precautionary, risk-based estimate that acknowledges the uncertainty inherent in our understanding of ecological systems when data is insufficient. Clarifying what additional research is necessary for supporting management will require collaborative relationships with interdisciplinary research goals to produce targeted cross- system answers (Leschine et al., 2003; Espinosa-Romero et al., 2011).

31 3.1 Synopsis

Chapter 3 Nonnative species in British Columbian Zostera marina eelgrass beds arrive by aquaculture and stay for the mild climate

3.1 Synopsis Nonnative species cause economic and ecological impacts in habitats they invade. In seagrass habitats, which play a vital role in estuarine ecosystems worldwide, there is little information on nonnative species. This is especially true for the eelgrass, Zostera marina, in coastal British Columbia (BC), Canada. I tested the relationships between nonnative species and both arrival vectors and environmental selection in northeast Pacific eelgrass. I compared richness and abundance of nonnative benthic, epifaunal and large mobile macroinvertebrates, and some algae to two arrival vectors (shipping and aquaculture) and environmental drivers (climate variables, human population density, and native richness and abundance). I found twelve nonnative species. Aquaculture activities were strongly correlated with benthic and epifaunal nonnative richness and abundance; shipping activity was not. Climate (temperature and salinity) was positively related to nonnative richness but not abundance and there was no relationship to native species richness and abundance or population density. Although there were few nonnative species detected, 50% have known negative impacts within eelgrass habitats. Results suggest that aquaculture activities are responsible for many primary introductions of nonnative species, and that temperature and salinity tolerances are responsible for post-introduction invasion success. While aquaculture and shipping vectors are becoming increasingly regulated to prevent further international spread of nonnative species it will be important to consider secondary spread from intraregional transport of nonnatives through local shellfish aquaculture and shipping.

3.2 Introduction Increased global connectivity has led to the introduction of numerous nonnative marine species (hereafter “nonnatives”) around the world, with harmful economic and ecological effects (Lockwood et al., 2005; Colautti et al., 2006). As nonnative introductions continue to increase (Wonham & Pachepsky, 2006) it is important to understand which

32 3.2 Introduction species have invaded and what factors might limit their success. Despite the role of nonnatives as important drivers of change globally (Wilcove et al., 1998; Didham et al., 2005), there remains a lack of understanding of nonnative marine species and their impacts in coastal marine habitats, such as seagrass beds (Williams, 2007; Carlton, 2009). High-impact nonnatives have been shown to directly alter seagrass through habitat modification, for example the European green crab, Carcinus maenas, tears up Zostera marina eelgrass when preying on infaunal bivalves (Davis et al., 1998) or by competing with seagrass for resources and preventing reestablishment, as seen in seagrass beds invaded by the Japanese wireweed, Sargassum muticum (denHartog, 1997). In seagrass habitats, which play a vital role in estuarine ecosystems worldwide (Hemminga & Duarte, 2000), nonnatives affect the health of seagrass plants and the communities that live within these habitats (Williams, 2007).

Zostera marina (hereafter eelgrass) the dominant soft-sediment seagrass in the northeast Pacific grows throughout BC on tidal flats and in the shallow waters fringing marine channels (Levings et al., 1983; Berry et al., 2003). Coastal development, habitat modification, and numerous other anthropogenic impacts have resulted in a net loss of eelgrass habitats (Short & Wyllie-Echeverria, 1996; Orth et al., 2006). The additional threat of nonnative introductions to these sensitive and productive systems is likely to be of great environmental concern (Waycott et al., 2009).

However, while nonnatives have increasingly become a focus of marine invertebrate research in the northeast Pacific with more than 62 species described, generally studies in coastal British Columbia (BC), Canada have either presented overviews of coastal nonnatives (Levings et al., 2002; Gillespie, 2007), focused on fouling species (Lu et al., 2007; Gartner, 2010; Clarke Murray et al., 2011), or sampled unvegetated mud and sand flats (Choi, 2011). Information is missing on nonnative communities within smaller-scale habitats such as eelgrass that make up the more than 25,300 kilometers of coastline in BC (Thompson, 1981). With an increase in the number of nonnatives in eelgrass beds globally over the last 10 years, now comprising almost 10% of total community richness (Williams, 2007), it is likely that BC eelgrass is host to numerous nonnative species.

33 3.2 Introduction

Many factors play a role in the increase of nonnatives in coastal marine systems, most notably the transport of organisms via shipping and aquaculture vectors (Carlton, 1987; Mineur et al., 2007; Cohen & Zabin, 2009; Haupt et al., 2010; Hewitt & Campbell, 2010; Ruiz et al., 2011b; Sylvester et al., 2011). Transport of nonnatives via international shipping to major ports is described as one of the greatest influences on the global dispersal of nonnatives (Ruiz et al., 2000b). For example, the majority of nonnative species described in San Francisco Bay, one of the most invaded bays in the world (Cohen & Carlton, 1998), have arrived with ballast water or hull fouling on ships arriving from overseas (Ruiz et al., 2011b). Similarly Port Metro Vancouver (Vancouver, BC), the fourth largest port in North America, is subject to a high frequency of international shipping traffic shown to transport nonnative species to BC (DiBacco et al., 2011; Sylvester et al., 2011; Lo et al., 2012).

Shellfish aquaculture, another important marine vector, transports both intentionally introduced and unintentional hitchhiking nonnatives (Levings et al., 2002; Cohen & Zabin, 2009). A global review of nonnatives in seagrass systems suggests the vector most commonly associated with nonnatives was shipping (32%) with aquaculture accounting for ~19% (Williams, 2007). However, shellfish aquaculture is a prominent vector for nonnatives in nearshore ecosystems (Carlton, 1987; Mineur et al., 2007; Cohen & Zabin, 2009; Haupt et al., 2010). In coastal BC, shellfish aquaculture covers over 3,700 hectares of coastline (Environment Canada, 2011) and previous studies of nonnatives suggest that oyster aquaculture is the dominant vector for introduction as compared to shipping (44% and 16% respectively; Levings et al., 2002; Gillespie, 2007). It is unknown whether eelgrass beds in BC receive nonnatives via the same aquaculture vectors as other marine and estuarine habitats in this region, or if they follow the global trend of arrival dominated by shipping vectors.

After arrival in a habitat outside their native range, establishment of nonnatives is dependent on species finding favorable environmental conditions (Incera et al., 2009; MacLeod et al., 2009). Abiotic factors, such as salinity (Mann & Harding, 2003; Powers

34 3.2 Introduction et al., 2006; Miller et al., 2007) and temperature (Stachowicz et al., 2002b; Clark & Johnston, 2005; Dafforn et al., 2009), may limit the success of nonnatives in newly invaded habitats. In the northeast Pacific, richness of nonnatives has been attributed to natural variation in temperature that occurs across latitudinal gradients (Ekman, 1953), with lower richness of nonnatives found in the colder waters of higher latitude regions (deRivera et al., 2005). Yet despite its high latitude, coastal BC’s unique geography results in some fjords and bays being warmer and less saline than the exposed coast, as a result of river runoff and limited exposure to the open ocean (Thompson, 1981). The warmer, less saline waters may allow greater numbers of nonnatives from warm and temperate regions to establish than would otherwise be expected due to the colder conditions of northern latitudes (deRivera et al., 2011).

Nonnatives must also tolerate changes in the biotic and abiotic environment, requiring sufficient habitat and resources (Moyle & Light, 1996; Shea & Chesson, 2002; Maron & Marler, 2007). Human disturbance causes release of resources such as space and nutrients through establishment of man-made structures, altering of natural habitats, and nutrient rich runoff (Glasby et al., 2007; Piola & Johnston, 2007; Martone & Wasson, 2008). In coastal BC, human population densities are at their highest in the southern regions of the Strait of Georgia (BCStats, 2006). Assuming disturbance increases with greater population densities and favors nonnative establishment, this region may support a greater richness and abundance of nonnatives than the less populated regions to the north (Altman & Whitlatch, 2007). Alternatively, in regions of high native species diversity biotic resistance may prevent the establishment of nonnatives by limiting resource availability (Tilman, 1997; Stachowicz et al., 1999). Human population densities and native species diversity may act as proxies for some aspects of resource availability that affect richness and abundance of nonnative species in eelgrass.

To study the relative importance of vectors and environment on establishment of nonnatives in eelgrass beds, I conducted biological sampling of eelgrass beds along the BC coast. Because eelgrass habitats comprise multiple environments that support a broad

35 3.3 Methods range of taxa, including benthic, epifaunal and large mobile invertebrates, I developed sampling methods that target each of these taxonomic groups.

I addressed the following research questions: 1. What is the frequency and abundance of nonnative species in BC eelgrass? 2. Do shipping frequency or distribution of aquaculture facilities explain variation in nonnative richness in eelgrass? 3. What is the relationship between the richness of nonnatives and local environmental conditions – climate variables, human population density, and native species richness?

3.3 Methods I sampled benthic, epifaunal and mobile species in 11 eelgrass beds on the coast of British Columbia (BC) in June – August 2008 (Figure 3.1). To represent a range of shipping activity levels, I selected sites by number of ship arrivals per year (data from Transport Canada Nov 2006 - Oct 2007) (Table 1.1; Lo et al., 2012). Of the 11 sites sampled, 8 were located in ports with shipping activity. To compare to sites influenced by shipping I selected sites with no ship arrivals at the northern (East Kaien Island), mid (Mud Bay) and southern (Fourth of July Beach, USA) ends of the sampling range. Fourth of July Beach is the only site outside BC, however the latitude of this site overlaps with the southern end of Vancouver Island and is located in waters contiguous with the Strait of Georgia. All sites were a minimum of 20 km apart. Site coordinates were measured using a Garmin eTrex Vista Cx GPS unit. Northern sites, Prince Rupert and East Kaien Island, are ~475 km north of the southern nine sites. These northern sites are considered to be located in a different biogeographic region than those sites to the south (Ricketts et al., 1982).

36 3.3 Methods

Figure 3.1 Map of British Columbia showing the 11 eelgrass beds (black circles) sampled. Insert map is a close up of sites in southern BC, Canada and Washington State, USA.

37 3.3 Methods

3.3.1 Field sampling In each intertidal eelgrass bed I collected invertebrates, fish and the Japanese wireweed alga, Sargassum muticum, over the course of 1-2 days (during the same low-tide cycle) using cores, dredge, traps and visual surveys to sample the full biotic assemblage present in each eelgrass bed (Short & Coles, 2001). Each method targeted a different species group. Core sampling targeted benthic and infaunal invertebrates (i.e., polychaetes, bivalves, gastropods; hereafter referred to as the benthic species group); dredges targeted epifaunal and sessile invertebrates that utilize eelgrass blades (i.e., amphipods, isopods, tunicates, limpets; hereafter referred to as the epifaunal species group). Traps targeted large mobile invertebrates (i.e., crab, sea stars; hereafter referred to as the mobile species group). Visual surveys targeted S. muticum (Asian wireweed) and Crassostrea gigas (Pacific Oyster) presence (hereafter referred to as S.mut/C.gigas).

Cores and Dredges: At each site I sampled benthic invertebrates using 6 sediment cores (10 cm diameter x 17 cm depth; Short & Coles, 2001) spaced 25 meters apart along two 50-meter transects. To sample epifaunal invertebrates I used five pulls of a naturalist dredge with a semi-circle mouth opening of 30 cm wide x 22 cm tall and a mesh size of 0.5 mm pulled 50 meters through the eelgrass against the water current, parallel to shore (flushing approximately 1036 m3 of water through the dredge). I stored each sample collected by core or dredge in a plastic bag until sampling for the day was completed. I then sieved the samples using a 1 mm sieve, removed plant matter and sediment by hand, and preserved the species in 95% ethanol the same day as collection.

Traps and Surveys: To sample mobile invertebrates I used a modified minnow trap, the opening of each end of the minnow traps was stretched to 10 cm in diameter to specifically target large invertebrates. Four pairs of traps were attached at 10-meter intervals along a line (4 to 8 traps per site) and anchored near the middle of the bed, parallel to shore. Traps were left in each bed for a 20 to 24 hour period after which I measured, identified or photographed (if the species was not identifiable in the field) and returned to the eelgrass bed all invertebrate and fish specimens. I trapped at 9 of the 11 sites (8 traps: Cowichan Bay, Campbell River, Esquimalt Lagoon, Fourth of July,

38 3.3 Methods

Nanaimo Estuary, Port Alberni; 4 traps: Prince Rupert, East Kaien Island, Tsawwassen; No traps: Stanley Park, Mud Bay). I completed visual walking surveys along a 100m x 2m transect parallel to shore and recorded S. muticum and C. gigas presence or absence but did not collect samples (3 visual transects per site).

I returned preserved collections from core and dredge samples to the lab and identified the species. Specimens that could not be identified in the lab were sent to Biologica (634 Humboldt St., Victoria, BC, V8W 1A4) for identification. More than 50% of amphipod and polychaete identifications were then re-identified by taxonomic experts to ensure correct identification (Taxonomists: Jeffery R. Cordell, University of Washington; R. Eugene Ruff, Ruff Systematics, Puyallup, WA). I classified organisms as native, cryptogenic, nonnative, or indeterminate. Nonnative species are those species that have been transported beyond their natural dispersal range. Cryptogenic species are those of unknown origin. ‘Indeterminate’ refers to organisms that I did not identify to a sufficient taxonomic resolution to determine their species identity or origin status.

3.3.2 Vector and environmental data To create an estimate of number and distance to shellfish aquaculture sites I calculated an ! aquaculture effect score = ! 4.8/aqdistance!, where n is the number of aquaculture sites associated with each sample site and aqdistancei is the distance (km) from aquaculture site i to the sample site (developed from Choi, 2011). The 4.8 constant is derived from the maximum larval dispersal distance of 480 km; it creates a decay function where the effect score will be ≤ 0.01 at distances greater than 480 km. This distance is estimated for Nuttalia obscurata, the varnish clam, as modeled from larval duration, sea surface circulation patterns, and stable environmental conditions (Dudas & Dower, 2006). Shellfish aquaculture site coordinates were collected from the British Columbia Ministry of Agriculture and Land, and Integrated Land Management Bureau (2005). Larger effect scores represent a greater density of aquaculture located close to the sample site. The number of ship arrivals for each site from Oct 2006 to Nov 2007 was calculated by Lo et al. (2012) using the Canadian Ballast Water Information System

39 3.3 Methods developed by the Department of Fisheries and Oceans Canada and Transport Canada and is recorded in Table 3.2.

Temperature and salinity data were not available for each site sampled, so I used an oceanographic climate model (pers. comm. with Mike Foreman, Department of Fisheries and Oceans Canada; Foreman et al., 2008) to generate data on sea surface temperature (SST) and salinity, averaged for the summer and winter. Population density (number of people per km) data were collected from the 2006 Canadian Census (BCStats, 2006) for the census region in closest proximity to the sample site.

The climate data for summer sea surface temperature (SST) was negatively correlated with summer salinity and winter salinity (for all correlations: r > -0.67, p < 0.02), likely a result of river runoff decreasing salinity levels in the warmer Strait of Georgia (Thompson 1981). I therefore chose summer SST as a proxy for summer and winter salinity climate data. In addition, to reduce deviations from normal distributions, I log- transformed distance-to-aquaculture effect scores, number of ship arrivals per year, and population density.

3.3.3 Nonnative abundance and distribution Using species accumulation curves I estimated total species richness (including native, cryptogenic, nonnative, and indeterminate) for core, dredge, and trap sampling methods (see Error! Reference source not found.). Native and indeterminate species are listed in Error! Reference source not found., while nonnative and cryptogenic species are listed in Table 3.1. The rate of species addition asymptotes after few samples for most sites suggesting sufficient sampling replication within each site (Gotelli & Colwell, 2001). Thus, I use richness as the number of distinct species from pooled samples for each method (between 3 and 8) at each site for the following analyses (units: benthic richness, total species in 6 cores, each 79 cm3; epifaunal richness, total species in 5 dredge pulls, each 1036 m3). Nonnative abundance data were calculated as the mean abundance across samples for each method at each site (units: mean benthic abundance,

40 3.3 Methods number of individuals per 79 m3, over six cores; mean epifaunal richness, number of individuals per 1036 m3, over six dredge pulls).

To ensure the independence between benthic and epifaunal analyses, species collected by multiple methods were only analyzed in the species group in which they were most commonly sampled (as in Table 3.1). For example, amphipods were primarily present in dredge samples, but they were also occasionally present in sediment cores; for these analyses I only included amphipod data from the dredge samples. Thus no species were analyzed in more than one species group. No tests were performed that might be affected by this assumption, and there were no cases of a species being found in the less-common sampling method at a given site but not the more-common method.

To test the difference in total nonnative richness and mean abundance of benthic and epifaunal species between sites I used a Kruskal-Wallis nonparametric test because data did not have a normal distribution (Sokal & Rohlf, 1995). Data was plotted with the function sciplot() using R software (Morales, 2011; R Development Core Team, 2012).

3.3.3.1 Relating vectors to nonnative richness and abundance To test the influence of shipping and aquaculture vectors on nonnative total richness and mean abundance, I compared these data to ship arrivals per year and distance from aquaculture effect scores using a nonparametric Spearman rank correlation (rho; Sokal & Rohlf, 1995). To demonstrate how nonnatives sampled in this study compare, in terms of vector, to provincial and regional patterns of nonnative species, I recorded predicted arrival vectors for species sampled in this study as compared to nonnatives described for coastal BC (review in Levings et al., 2002; Gillespie, 2007) and nonnatives previously identified in northeast Pacific seagrass beds (review in Williams, 2007). Species were attributed to either or both shellfish aquaculture and shipping vectors (which could include hull fouling or ballast water), with species attributed to multiple vectors referred to as polyvectic (Carlton & Ruiz, 2005). Only nonnatives from those phyla that were sampled in my study were included in the local (43 species) and regional (15 species) comparisons (Annelida, Arthropoda, Echinodermata, Mollusca and Nemertea). In

41 3.3 Methods addition, I identified Mytilus sp. only to genus level and did not include algal or plant species other than S. muticum.

3.3.3.2 Relating environmental variables nonnative richness and abundance The total richness and mean abundance of nonnatives in eelgrass beds is likely influenced by a complex of environmental factors which I tested using Generalized Linear Models (GLM) that assume a normal probability distribution. I used an information-theoretical approach to assess performance from the set of candidate models (Burnham & Anderson, 2002) for benthic and epifaunal nonnative richness and abundance. As the number of models should not exceed the number of sites sampled (Anderson, 2008) I selected a subset of three variables a priori to compare to benthic and epifaunal nonnative richness (summer temperature, population density, native richness) and abundance (summer temperature, population density, native abundance) (Error! Reference source not found.). Below I provide a brief rationale for each of the three environmental explanatory variables:

Summer Temperature: Nonnative richness and abundance have been positively correlated with summer temperature and negatively correlated with salinity in marine coastal studies of the northeast Pacific (Dethier & Hacker, 2005; deRivera et al., 2011) and temperature has been suggested to be one of the primary selection factors for nonnative establishment (Quayle, 1964).

Population Density: Anthropogenic disturbance, such as the building of new docks and piers, trampling of intertidal habitats, and anchor scars from boating, often releases resources and increases unoccupied habitat space, and has been positively correlated to nonnative establishment (Altman & Whitlatch, 2007). I predicted that nonnatives will be at their highest richness and abundance in the southeastern Strait of Georgia, BC, because human populations are at their highest densities in this region, and I assumed that population density correlates with general anthropogenic disturbance.

42 3.3 Methods

Native Richness and Abundance: Regions with high native species richness (for model analyses, native richness and abundance includes native, cryptogenic, and indeterminate species) may have “biotic resistance” to establishment of nonnative species, in these cases native richness has been found to have a negative correlation to nonnative richness (Tilman, 1997; Cohen & Carlton, 1998). On the other hand, in cases where variation in native species richness is explained largely by variation in productivity and resource availability, native richness has been found to correlate positively with nonnative richness (Shea & Chesson, 2002).

In the absence of any compelling theory or method to group variables into particular candidate models, I used all possible combinations of the three explanatory variables (7 models). I ranked models using the small-sample-size version of Akaike’s Information Criterion (AICc) by the likelihood of being the best model in the group using two ranking measures, as calculated by the R package ‘AICcmodavg’ (Mazerolle, 2012): 1) models with a difference of less than 3 units from the 1st-ranked model were considered to have substantial support and 2) models with high Akaike weights (i.e. the probability of a model being the best explanation of the given data) (Burnham & Anderson, 2002). For each of the benthic and epifaunal environmental model tests I inspected the residuals from the full model (all three explanatory variables) to ensure that the statistical assumptions of homoscedasticity and normality were satisfied (Sokal & Rohlf, 1995).

In order to explore additional relationships, I used post-hoc tests to assess the potential of the additional environmental variables (those sampled but not selected for model testing) for explaining variation in nonnative total richness and mean abundance (see Appendix C). In addition, though the assumptions of normality were not met, I conducted a final linear regression analyses of benthic and epifaunal richness in order to included both significant environmental and vector variables to test whether these variables together better explain nonnative distributions.

43 3.3 Methods

3.3.3.3 Influence of nonnative species composition and environment on site similarity To interpret differences in nonnative species composition between sites. I created a single matrix of site-by-species data of nonnative richness estimates at 8 sites for 12 nonnatives (presence or absence of benthic nonnatives (4 species), epifaunal nonnatives (6 species), S. muticum and C. gigas). I did not include the three sites with no nonnative species: Fourth of July, Prince Rupert, and E Kaien Island. Using the Kulczynski coefficient, which is appropriate for presence/absence species estimates, I then transformed the site- by-species data matrix into a distance matrix of the dissimilarities in nonnative composition between each pair of sites using the envfit() function in the R package vegan (Legendre & Legendre, 1998; Oksanen et al., 2012). Based on the site dissimilarity matrix, I then visualized nonnative composition using a cluster analysis and nonmetric multi-dimensional scaling (nMDS), an ordination technique that simplifies multidimensional relationships among sites to create a smaller number of dimensions that are easier to visualize and interpret (Kruskal & Wish, 1978). I plotted nMDS scores for nonnative composition against site. Each point represents the nonnative community for a given site, and the distance between points represents how the similarity of sites based on the composition of nonnative species.

To cross-check the nMDS site groupings I used a cluster analysis to perform hierarchic clustering of weighted averages calculated from average neighbor resemblances of site similarity. These groupings were then overlaid on the nMDS plot for sites with 50% and 70% similarity. To test if site differences in species composition can be explained by environmental variables, I compared the site dissimilarity matrix to average summer SST (Summer SST), human population density (Ln Population), and native species richness (Native Richness) using the envfit() function. Correlation significance was calculated using a permutation test (n = 11). The number of sites sampled limited the number of permutations possible. All statistical analyses were performed using R software (R Development Core Team, 2012).

44 3.4 Results

3.4 Results In eleven British Columbia (BC) eelgrass beds I sampled a total of 32,401 individuals from 181 different species; 12 were classified as nonnative, 8 as cryptogenic, 29 as indeterminate, and the remaining 132 were classified as native (Table 3.1). I sampled the 12 nonnatives in three of the four sampling methods: four benthic nonnatives in core samples, six epifaunal nonnatives in dredges, and Sargassum muticum and Crassostrea gigas in visual surveys. I did not record any nonnatives in trap sampling.

45 3.4 Results

Table 3.1 Nonnative (Table A) and cryptogenic (Table B; those of unknown origin) species sampled in British Columbia Z. marina beds, habitat in which they were sampled (Hab), percent of sites occupied (% sites), which sites the species was found (listed from south to north), and likely vector of introduction. Quotations around species are unresolved taxa and the value for each species at each site is mean abundance across samples collected at each site. A ‘+’ represents species that were present but abundance data not collected. Native richness of species (includes indeterminate species) in Annelid, Arthropod and Mollusca phyla and ship arrivals are the number of ships arriving at each port between Nov 2006 and Oct 2007 (Lo et al., 2012). Site abbreviations: Esquimalt Lagoon (EL), Fourth of July (FJ), Cowichan Bay (CB), Tsawwassen (Ts), Port Alberni (PA), Nanaimo Estuary (NE), north of Stanley Park (SP), Mud Bay (MB), Campbell River (CR), Prince Rupert (PR), East Kaien Island (EKI).

Table A. Nonnative species % Taxonomic Native Species Taxa Hab Sites EL FJ CB Ts PA NE SP MB CR PR EKI Authority Range Vector Ampithoe valida Amphipod E 64 1 0.4 0.6 15.4 4 4.6 0.7 Smith, 1873 NWAtl A/S Batillaria attramentaria Gastropod B 9 1.3 Sowerby II, 1855 NWPac A Clymenella torquata Polychaete B 18 2.8 0.5 Leidy, 1855 NWAtl A Crassostrea gigas Bivalve V 18 + + Thunberg, 1793 NWPac N Grandidierella japonica Amphipod E 9 0.2 Stephensen, 1938 NWPac A/S Melita nitida Amphipod E 9 1.6 Smith, 1873 NWAtl A/S Monocorophium acherusicum Amphipod E 36 0.14 0.2 1 0.2 Costa, 1857 NEAtl A/S Monocorophium insidiosum Amphipod E 9 0.8 Crawford, 1937 NAtl A/S Mya arenaria Bivalve B 27 0.2 0.4 2.7 Linnaeus, 1758 NWAtl N Sargassum muticum Algae V 27 + + + Fensholt, 1955 NWPac A Sinelobus sp. Tanaid E 9 Sieg, 1980 Unk P Venerupis philippinarum Bivalve B 18 0.6 0.3 0.6 Adams and Reeve, 1850 NWPac A Ship Arrivals 3 0 14 177 22 14 1328 0 2 159 0 Native Richness 33 35 33 36 9 24 35 45 42 25 32

Habitat: B = Benthic, E = Epifauna, V = Visual survey of S. muticum and C. gigas presence, no nonnative mobile macroinvertebrates; Native range: Atl = Atlantic, Pac = Pacific, Unk = Unknown; Vector: A = aquaculture, S = ship hull or ballast water, N = natural dispersal from aquaculture introduction, P = polyvectic

46 3.4 Results

Table B. Cryptogenic species % Taxonomic Species Taxa Hab sites EL FJ CB Ts PA NE SP MB CR PR EKI Authority Ampithoe lacertosa Amphipod E 18 0.6 1 Bate, 1858 Capitella telata (?) Polychaete B 9 1 Fabricius, 1780 Eumida "sanguinea" Polychaete B 18 0.17 Oersted, 1843 Harmothoe "imbricata" Polychaete B 27 9.8 0.3 0.5 Linnaeus, 1767 Harpacticus uniremis group Copepod E 18 0.3 38.5 Holmes, 1900 Leptochelia "dubia" Tanaid E 64 0.2 4.5 1.6 0.17 2.83 2.17 8.3 Krøyer, 1842 Prionospio "steenstrupi" Polychaete B 9 0.33 Malmgren, 186 Spiophanes "bombyx" Polychaete B 9 0.33 Claparède, 1970

Habitat: B = Benthic, E = Epifauna, V = Visual survey of S. muticum and C. gigas presence, no nonnative mobile macroinvertebrates

47 3.4 Results

Only one of the 12 nonnatives found in this study is a new record in the Strait of Georgia, BC when compared to previous nonnative reviews (Levings et al., 2002; Gillespie, 2007). Previously recorded in Boundary Bay, BC (Latitude: 49.05 N, Longitude: 122.97 W) in 1981, the bamboo worm, Clymenella torquata, had not been sampled in the Strait of Georgia since that time (Mach et al., 2012). Response and explanatory variables used in the following analyses are presented in Table 3.2.

Table 3.2 Variables used in the analysis of relationships between nonnative total richness and mean abundance and the environmental conditions at each site. Mean (units described in section 3.2.2 and 3.2.3), standard deviation (S.D.), minimum and maximum values of each response and explanatory variable.

Variable Mean S.D. Min. Max. Response variables Benthic Nonnative Richness 0.73 0.79 0 2 Epifaunal Nonnative Richness 1.36 1.29 0 4 Benthic Nonnative Abundance 0.80 1.20 0 3.2 Epifaunal Nonnative Abundance 2.82 5.06 0 17.2

Explanatory variables Arrival Vector Ln Shellfish Aquaculture Score 3.43 1.15 1.79 5.96 Ln Ship Arrivals 2.59 2.41 0 7.19 Environment Summer Temperature 13.92 2.29 10.78 17.25 Ln Population Density per Km 3.21 2.29 0 6.54 Benthic Native Richness 17.73 7.66 8 34 Epifaunal Native Richness 24.45 7.48 11 39 Benthic Native Abundance 4.82 7.17 0 19 Epifaunal Native Abundance 14.18 25.26 0 86 Ln = Log normal transformation to the data before analysis

Of the nonnatives in this study 50% were classified as arriving via aquaculture and 50% as polyvectic, likely arriving via aquaculture or shipping vectors. No species sampled had shipping as their only vector (Table 3.1; Figure 3.5). Six of these 12 species have their native range in the north Atlantic, while five are from the north Pacific and one is unknown (Table 3.1). In comparison, review studies in BC found nonnative species to be 54% arrivals from aquaculture, 20% from polyvectic sources, and 17% via shipping vectors. In the larger northeast Pacific region, nonnatives previously documented in eelgrass beds were dominated by 44% aquaculture introductions, 19% polyvectic, and 25% arriving from shipping vectors (Figure 3.2).

48 3.4 Results

Aquaculture Polyvectic Shipping Unknown 0.50

0.25 Proportion of Total Nonnatives Proportion

0.00 BC-All Habitats BC-Z.marina NE Pacific-Z.marina

Figure 3.2 Presumed vectors of introduction for nonnatives in British Columbia across all marine habitats (N = 46 nonnatives; Gillespie 2007; Levings et al. 2002), in BC eelgrass beds (N = 12; this study), and reported in eelgrass beds in the northeast Pacific (N = 16; Williams 2007).

3.4.1 Nonnatives’ frequency and distributions Nonnatives in eelgrass beds were not particularly widespread or abundant; one or more nonnatives were present in 8 of 11 eelgrass beds. The greatest richness of nonnatives (six species) was found at both Mud Bay and Nanaimo estuary, and no nonnatives were found at the two northernmost sites (East Kaien Island and Prince Rupert) or Fourth of July Beach, USA. The greatest abundance of benthic nonnatives was found at Mud Bay, and the greatest abundance of epifaunal nonnatives at Nanaimo estuary.

The total richness of benthic and epifaunal nonnatives was significantly different across 2 2 sites (Figure 3.3; benthic, chi 10=29.28, P = 0.001; epifaunal: chi 10=43.96, P <0 .0001) 2 2 as was mean abundance (benthic: chi 10=29.77, P = 0.0009; epifaunal: chi 10=44.686, P < 0.0001). For benthic nonnatives, Mya arenaria had the greatest frequency at Mud Bay while at Tsawwassen C. torquata was the most abundant species. Epifaunal nonnatives were greatest at Nanaimo estuary, driven by the frequency and abundance of the most commonly sampled nonnative in this study, the amphipod Ampithoe valida. It was found at seven of the eleven sites and it made up 11% of native and nonnative epifaunal species

49 3.4 Results abundance in the Nanaimo estuary (sampled in all 5 dredges). All other nonnatives were found at one to four sites and at low abundances relative to A. valida. Sargassum muticum (sampled at 3 sites) and C. gigas (sampled at 2 sites) were present at fewer than half of sites sampled (Table 3.1).

2 a) Benthic 4 b) Epifauna

1 2

Nonnative Richness Nonnative Richness Nonnative 1

EL FJ CB Ts PA NE SP MB CR PR EKI EL FJ CB Ts PA NE SP MB CR PR EKI

6 c) Benthic 25 d) Epifauna

5 20

4 15

10 2

Nonnative Abundance Nonnative Abundance Nonnative 5 1

EL FJ CB Ts PA NE SP MB CR PR EKI EL FJ CB Ts PA NE SP MB CR PR EKI

Figure 3.3 Bar plots (with standard error) of total nonnative benthic richness (a; total species per 6 x 79 m3) and epifaunal richness (b; total species per 5 x 1036 m3), mean nonnative benthic abundance (c; number of individuals per 79 m3, over 6 cores) and epifaunal abundance (d; number of individuals per 1036 m3, over 5 dredge pulls) in each eelgrass bed. Black circles represent total richness. Sites are in order of latitude from south to north. Kruskal-Wallis nonparametric tests of site differences are significant for nonnative richness 2 2 and abundance: benthic richness (chi = 29.28, P = 0.001) and abundance (chi = 29.77, P = 0.0009), 2 2 epifaunal richness (chi = 43.96, P < 0.0001) and abundance (chi = 44.686, P < 0.0001).

50 3.4 Results

3.4.2 Aquaculture versus shipping vectors Sites with high aquaculture scores had both higher benthic nonnative richness and abundance in Spearman’s rank correlations (Table 3.3; N = 11; richness: rho = 0.82, P = 0.002, Figure 3.4a; abundance: rho = 0.763, P = 0.006, Figure 3.5a). Similarly sites with high aquaculture scores had higher epifaunal nonnative richness and abundance (rho = 0.75, P = 0.008, Figure 3.4c; rho = 0.743, P = 0.009, Figure 3.5c). However, neither benthic nor epifaunal nonnative richness or abundance was significantly related to ship arrivals (Benthic: rho = 0.06, P > 0.05, Figure 3.4b; rho = -0.014p, P > 0.05, Figure 3.5d. Epifauna: rho = 0.293, P > 0.05, Figure 3.4d; rho = 0.181, P > 0.05, Figure 3.5d).

Table 3.3 Spearman rank correlation (rho) and probability-value (P) for benthic and epifaunal nonnative richness and abundance with distance to and frequency of shellfish aquaculture sites (Aquaculture) and the number of ship arrivals at each port (Ship Arrivals). Significant P-values in bold.

Vector SR P SR P Richness Abundance Benthic Nonnatives Aquaculture 0.82 0.002 0.76 0.006 Ship Arrivals -0.06 0.86 -0.01 1.00 Epifaunal Nonnatives Aquaculture 0.75 0.008 0.74 0.009 Ship Arrivals 0.29 0.38 0.18 0.59

51 3.4 Results

a) b) 2

0 Benthic Nonnative Richness Nonnative Benthic

c) d) 4

1 Epifaunal Nonnative Richness Nonnative Epifaunal 0

3 4 5 6 0 2 4 6 Ln Aquaculture Score Ln Ship Arrivals

Figure 3.4 Correlation of nonnative richness and vector variables. Spearman rank correlation results were the following: nonnative benthic richness (total species per 6 x 79 m3) compared to transformed aquaculture effect score (a; rho = 0.823, P = 0.002) and number of ship arrivals (b; rho = -0.06, P = 0.862) and nonnative epifaunal richness (total species per 5 x 1036 m3) compared to aquaculture effect score (c; rho = 0.75, P = 0.008) and number of ship arrivals (d; rho= 0.293, P = 0.382). Relationships significant in the Spearman rank test plotted as dashed line.

52 3.4 Results

a) 3 b)

Benthic Nonnative Abundance Nonnative Benthic 0

c) d) 16

0 Epifaunal Nonnative Abundance Nonnative Epifaunal 3 4 5 6 0 2 4 6 Ln Aquaculture Score Ln Ship Arrivals

Figure 3.5 Nonnative abundance and vector data plotted with Spearman rank correlation test results. Comparison of nonnative benthic abundance (number of individuals per 6 x 79 m3, over 6 cores) to the transformed aquaculture effect score (a; rho = 0.763, P = 0.006) and number of ship arrivals (b; rho = - 0.014, P = 0.97) and epifaunal abundance (number of individuals per 1036 m3, over 5 dredge pulls) to aquaculture effect score (c; rho = 0.743, P = 0.008) and number of ship arrivals (d; rho = 0.181, P = 0.59). Relationships significant in the Spearman rank test plotted as dashed line.

3.4.3 Model results for environment effect on nonnative richness and abundance Nonnative richness of both benthic and epifaunal communities is best explained by summer SST alone (the optimal model for total nonnative richness based on AICc scores and Akaike weights; Table 3.4). Models that included the other two variables (population density and native richness or abundance) were non-significant. When summer SST was compared to benthic and epifaunal nonnative richness in a Pearson correlation test both are significant (P < 0.05) with r2 values of 0.73 and 0.84, respectively, suggesting summer SST, or variables correlated with summer SST such as salinity, predicted 73% of

53 3.4 Results the variance in benthic nonnative richness and 84% of epifaunal nonnative richness (Figure 3.6). The best model representing mean nonnative abundance contained only the y-intercept and no explanatory variables, suggesting that the variables selected for this analysis do not significantly explain trends in nonnative abundance (Burnham & Anderson 2001). While these data were non-normal, the high significance of only one variable for nonnative richness supports the accuracy of this model-test despite the right- skew distribution of the data.

The results of the linear regression analysis of significant vector and environmental variables, summerSST and aquaculture, together revealed that both contribute to explaining the variation of benthic and epifaunal nonnative richness, (benthic: r2 = 0.62, 2 F8 = 9.13, P = 0.009; epifauna: r = 0.63, F8 = 9.86, P = 0.007). However, the relationship of these two variables to nonnative richness is less than each explanatory variable alone.

Table 3.4 Linear models of nonnative benthic and epifaunal total richness and mean abundance in the eleven eelgrass beds sampled. The three best-ranked candidate models of the full factorial of possible models (Full model: y~ SummerSST + LnPopulation + Native; 7 models tested) are listed with the number of parameters (K), corrected AIC (AICc), the difference in AIC between the candidate model and the best model (deltai), Akaike weights (wi), and log-likelihood (LL). The ‘best’ models are in bold. “Intercept” are those models without any of the three variables.

Model K AICc Deltai wi LL Benthic ANS Richness SummerSST 3 26.03 0 0.68 -8.30 SummerSST + Benthic.Native 4 29.23 3.20 0.14 -7.28 Intercept 2 30.38 4.34 0.08 -12.44

Epifaunal ANS Richness SummerSST 3 31.49 0 0.79 -11.03 SummerSST + Epifauna.Native 4 34.97 3.48 0.14 -10.15 SummerSST + LnPopulation 4 36.58 5.07 0.06 -10.96

Benthic ANS Abundance Intercept 2 39.61 0 0.41 -17.06 SummerSST 3 39.72 0.10 0.39 -15.14 LnPopulation 3 43.31 3.70 0.06 -16.94

Epifaunal ANS Abundance Intercept 2 71.34 0 0.43 -32.92 SummerSST 3 72.11 0.77 0.30 -31.34 Epifauna.Native 3 73.57 2.23 0.14 -32.07

54 3.4 Results

a) r2 = 0.73, P = 0.011 2

0 Benthic Nonnative Richness Nonnative Benthic

b) r2 = 0.84, P = 0.001 4

Summer SST Summer 1 Epifaunal Nonnative Richness Nonnative Epifaunal 0

12 14 16 Summer SST

Figure 3.6 Nonnative benthic (a) and epifaunal (b) richness as predicted by SummerSST (y ~ SumTemp, the ‘best’ AICc model). Relationships significant in the Pearson coorelation test plotted as dashed line.

3.4.4 Differences in nonnative community composition The differences in nonnative assemblage at each site were interpreted using nMDS ordination space (Figure 3.7). The Port Alberni site, located on an estuary on the west coast of Vancouver Island, was separated from other sites (negative nMDS values along the y-axis) as a result of three unique arthropod species not found at other sites: Sinelobus sp., M. nitidia, and G. japonica. Nanaimo, Mud Bay, and Cowichan Bay are located on

55 3.4 Results mid-Vancouver Island in the Strait of Georgia and were characterized by three nonnative molluscs: Venerupis philippinarum, C. gigas, Batillaria attramentaria, and Monocorophium amphipods separating them to the right of the nMDS (positive nMDS values along x-axis). These sites had species compositions that were more than 50% similar to one another in cluster analyses. Other sites with similar species composition: Tsawwassen, Stanley Park, Esquimalt, and Campbell River are separated by the Strait of Georgia from east to west and spread between the northern and southern ends of the Strait suggesting site groupings are not driven by spatial auto-correlation. These four sites also have species compositions that are 50% similar and group together in the negative nMDS values along the x-axis.

Native richness and human population density are positively correlated with sites positioned with negative values on the x-axis and positive values on the y-axis (Figure 1.7): Campbell River, Tsawwassen, Stanley Park and Esquimalt Lagoon, and explain 28% and 20% of site differences, respectively, while the summer SST vector explains 25% of variance between sites and is negatively related to these same sites, pointing to the opposite corner of the nMDS plot. Environmental variables did not significantly relate to species assemblage, though this may result from the limited number of sites and species richness across those sites, and does not necessarily indicate whether environmental variables tested are important for species at these sites.

56 3.5 Discussion

Figure 3.7 nMDS plot visualizing site dissimilarity of nonnative community composition with polygons from cluster analysis, (50% similarity = grey, 70% similarity = white). Location of the centre of species names (in italics) in relation to site names on the nMDS figure demonstrates how each nonnative influences the relationship of sites to one another. Includes only those sites with nonnatives. Relationship of environmental variables to the site dissimilarities is depicted in the bottom right of the plot. The length of vectors represents the difference in r2 values between each variable: SummerSST (r2 = 0.25), Ln Population (r2 = 0.20), and Native Richness (r2 = 0.28).

3.5 Discussion There are few nonnative species in British Columbia (BC) eelgrass when compared to the more than 62 nonnative species recorded in all coastal BC habitats (Levings et al., 2002; Gillespie, 2007); only 12 species were detected, three of which were introduced intentionally for aquaculture. This low number of nonnatives suggests there may be limited niche availability for macrofauna in eelgrass habitats and that habitat and resources are already in use by native species (Shea & Chesson, 2002; Duffy et al., 2007). Nonnatives that use eelgrass habitats may also be less likely to enter ballast tanks,

57 3.5 Discussion foul on ship hulls, or are not those best equipped for successful invasion: generalist predators, species with high reproduction rate, and species with highly mobile propagules (Ricciardi & Rasmussen, 1998; Kolar & Lodge, 2001; Wonham et al., 2001). Alternatively, newly introduced species from aquaculture may not be species that would utilize soft-sediment eelgrass habitats, as most nonnatives are arriving as fouling organisms that require hard substrates and may not associate with eelgrass beds.

Nonnative species composition between sites differed greatly; no two sites had the same set of nonnatives. Intraregional transport of goods, recreational boating, and local aquaculture trade have the potential to more than double the number of nonnatives sampled during this study, at any given site (Cohen & Carlton, 1998; Davidson et al., 2008; Clarke Murray et al., 2011). The rarity of nonnatives suggests that secondary spread is a major concern for the future, as regional dispersal will increase introductions to not yet invaded sites.

3.5.1 Relationship to aquaculture vectors Nonnative species distributions in coastal marine systems are largely a product of arrival vectors and selective environmental conditions (Colautti & MacIsaac, 2004; Blackburn et al., 2011). In BC, the majority of nonnatives have been hypothesized to arrive via ballast and hull fouling on shipping freighters and through import of aquaculture species (Quayle, 1964; Levings et al., 2002; Gillespie, 2007). As such, I expected a strong positive relationship between vectors important for nonnatives in eelgrass and nonnative richness. However, if secondary spread and natural dispersal of nonnatives had continued since initial establishment, there would be no significant relationship with either vector (Mineur et al., 2010). My data on nonnatives in eelgrass had a significant positive relationship between the richness and abundance of nonnatives and the distance and quantity of shellfish aquaculture, while neither richness nor abundance of nonnatives was significantly related to the number of ships arriving at the nearest port. The relationship of nonnatives to aquaculture but not shipping suggests introductions from aquaculture to be the most important vector explaining the current distribution of nonnative species in BC eelgrass beds.

58 3.5 Discussion

These results differ from regional patterns of invasion that have shown shipping to be an important vector in the introduction of mollusc, arthropod and annelid species to other seagrass beds in the northeast Pacific (Williams, 2007). They also differ from results in other habitats in BC where nonnative establishment was associated in part with shipping vectors (Figure 3.2; Levings et al., 2002; Gillespie, 2007). Accordingly, my findings were somewhat unexpected in showing that BC eelgrass beds were dominated by species arriving via aquaculture and contain no species specifically linked to shipping vectors.

A common limitation in studies sampling marine nonnatives, including this study, is the scale at which sampling is conducted (Carlton, 2009). Sampling methods regularly include macrofauna (Wyatt et al., 2005; Dafforn et al., 2009; Vermonden et al., 2010), but less commonly consider meio- or microfauna scale organisms (but see Kask et al., 1982; Cordell et al., 2007). It is possible these species are arriving via shipping vectors but were not sampled in this study.

3.5.2 Climate influences nonnative abundance and distribution Benthic and epifaunal nonnatives were related to environmental selection factors. Species richness was significantly correlated with summer SST variation across sites, but abundance was not. These findings may be the result of species establishment being constrained by the local marine climate but population growth being limited by other factors such as available resources (Elahi & Sebens, 2012). Similarly, in a study of subtidal fouling nonnative species on hard substrates in BC, the richness of nonnatives was related to temperature, however human population density was also important in explaining distribution patterns of these species (Clarke Murray, 2012). The relationship of fouling species and population density is likely the result of artificial habitat and disturbance increasing hard substrate for settlement; in reference to the current study, increased population density did not benefit nonnatives in soft-sediment eelgrass beds. The richness and abundance of native species were also not significantly related to nonnatives. This suggests temperature and/or salinity are more important selection factors

59 3.5 Discussion than the availability of habitat and resources or biotic resistance by native species for nonnative establishment in BC’s eelgrass habitats.

The Strait of Georgia receives fresh water input from the Fraser River leading to lower surface salinities while increased residence time of these protected waters and lack of coastal upwelling results in a higher temperature than other regions of the exposed BC coast (Thompson, 1981). This unique environment may support a community of nonnatives unable to exist in the colder more saline conditions of other sites. This combination of factors, increased chance of nonnative arrivals and unique environmental conditions, may have influenced the greater richness of nonnatives for seagrass beds in this region of BC. However Port Alberni, not located in the Strait of Georgia, had the lowest summer salinity (less than 6 parts per thousand) and warmest summer temperatures. This site had extremely low native species richness (9 natives) and was one of the most invaded sites (5 nonnatives) with the highest observed proportion of nonnative to native species (36% of species were nonnative). This observation aligns with theory that species-poor communities are more vulnerable to nonnative species than species-rich communities (Tilman, 1997; Cohen & Carlton, 1998). In addition, as many nonnative species have been demonstrated as limited by cold temperatures (Stachowicz et al., 2002b; Clark & Johnston, 2005; Dafforn et al., 2009), the warm temperature at this site may have resulted in increased susceptibility to nonnative colonization.

The lack of nonnatives in the two northernmost sites may be the result of cold temperatures limiting nonnative establishment upon arrival (however see Sloan & Bartier, 2004). Additionally, because of the geographic distance and because trading routes are different for ships arriving to northern BC, it is possible the nonnatives found further south have yet to arrive to Prince Rupert, though secondary spread of nonnatives between the Strait of Georgia and the northern coast has been demonstrated (Piercey et al., 2000; DiBacco et al., 2011). While some nonnatives may be limited from northward spread, I had expected but failed to find Mya arenaria at my northern sites, where it has been previously sampled (Quayle, 1964).

60 3.5 Discussion

3.5.3 Nonnatives in BC eelgrass and their potential impacts Most nonnatives sampled in this study were well-established species found in other BC coastal ecosystems with known impacts to their invaded habitats (Quayle, 1964; Levings et al., 2002; Gillespie, 2007). The clams, Mya areneria and Venerupis philippinarum, were introduced in the 1800s for aquaculture and have subsequently invaded intertidal habitats along coastal BC (Newcombe, 1891). Studies suggest the bivalves may inhibit rhizome growth, reducing eelgrass biomass (Reusch & Williams, 1998). Similarly, the Pacific oyster, Crassostrea gigas, spread from aquaculture to the natural environment and was found in Nanaimo Estuary and Mud Bay eelgrass beds. Generally Pacific oysters are thought to grow attached to a hard rocky substrate, however C. gigas is an ecosystem engineer that can grow independently in soft-sediment environments, physically altering these habitats (Lejart & Hily, 2011). As a result this species may reduce Z. marina density through competition for space and has the potential to alter species presence within the eelgrass bed with the addition of hard substrate. However, studies on the effects of bivalves on seagrass have suggested that increased water flow from these species can increase eelgrass area by increasing the depth at which eelgrass can grow by improving clarity (Newell & Koch, 2004).

Batillaria attramentaria was the only nonnative gastropod sampled and was only found at Nanaimo Estuary, though other studies have found B. attramentaria at Tsawwassen (Levings & Coustalin, 1975; Swinbanks & Murray, 1981). At Tsawwassen, my samples were collected towards the middle of the large eelgrass-covered mudflat and B. attramentaria are frequently found at the shore-edge of the Z. marina bed; accordingly, my sampling may have missed a present-day population of B. attramentaria (Swinbanks & Murray, 1981; Wonham et al., 2005). The invasion of B. attramentaria in eelgrass has increased the invasibility of eelgrass beds as it has done in other mudflats in this region (Wonham et al., 2005) by increasing hard substrate in soft-sediment mudflats, grazing, and bioturbating, which may facilitate invasion by the nonnative Japanese eelgrass Z. japonica (a species present at Tsawwassen though not recorded in this study, personal observation; Harrison & Bigley, 1982b).

61 3.5 Discussion

The bamboo worm, Clymenella torquata was sampled for the first time in the northern Strait of Georgia during this study. Of the two sites where C. torquata was sampled, highest abundances were found at Tsawwassen, perhaps the result of an initial introduction to Tsawwassen from Boundary Bay (Banse, 1981) followed by a subsequent introduction event to Campbell River. Clymenella torquata has limited planktonic dispersal, thus northward spread in the Strait of Georgia is likely the result of human- mediated transfer (Mach et al., 2012). In addition the tube-forming polychaete causes sediments to become spongy, this habitat alteration has the potential to deleteriously affect seagrass growth and abundance through competition for space and disturbance as has been demonstrated with other infaunal bioturbators (Dumbauld & Wyllie-Echeverria, 2003).

Crustaceans were the most abundant nonnative taxon in this study; the amphipod Amipithoe valida was the most abundant species, and was most abundant in the Nanaimo Estuary, mid-Vancouver Island. There have been no studies on the impacts of this species in its invaded range (Ruiz et al., 2011a). Other nonnative crustaceans include the amphipods Monocorphium acherusicum, M. insidiosum, Melita nitida, Grandidierella japonica and the tanaid, Sinelobus sp. Despite the invasion of these nonnatives on mudflats across the northeast Pacific, the impacts of only two of these crustaceans have been examined (review in Ruiz et al., 2011a). M. acherusicum has the potential to alter the native invertebrate community in seagrass beds in BC through predation and competition for resources and space (Barnard, 1958; Onbe, 1966; Talman et al., 1999). In contrast, the effect of G. japonica on estuarine food webs has been studied yet no negative impacts documented (West et al., 2003; Whitcraft et al., 2008). This range of impacts and general lack of understanding makes it difficult to predict how the other nonnative crustacean species might alter eelgrass habitats.

The invasive alga, Sargassum muticum, was identified in Tsawwassen, Stanley Park and Mud Bay eelgrass beds; all sites located in the mid-Strait of Georgia region. This species has a holdfast that requires hard substrate, heavy enough to weigh down a juvenile or adult plant that is buoyed by many air vesicles (White & Orr, 2011). In Tsawwassen, a

62 3.5 Discussion mud flat comprised of fine sediments, S. muticum was found growing on large Tresus capax clamshells, one of the only hard substrates in this eelgrass bed. S. muticum has been shown to shade out Z. marina plants in the NW Atlantic, reducing shoot densities where it invades (denHartog, 1997).

There were several nonnatives I anticipated finding, but were not present. For example, Carcinus maenas, the European green crab, is currently found on the outer coast of Vancouver Island from Barkley Sound to Kyuquot Sound (Yamada & Gillespie, 2008). It has not yet been found in the Strait of Georgia or on mainland BC (Gillespie et al., 2007), and mandatory ballast exchange is currently in place to help prevent human mediated transfer into the Strait of Georgia (Lo et al., 2012); however, natural dispersal is a likely method for invasion in the near future (Yamada & Gillespie, 2008). The trapping method used in this study successfully caught C. maenas during a similar study along the outer coast of Nova Scotia, Canada (personal observations). Thus, if C. maenas were present in sufficient abundance, I would have expected to sample it. This species is of concern for eelgrass in BC because C. maenas is a destructive consumer that rips up eelgrass when it feeds at high densities (Davis et al., 1998).

Interestingly, no nonnative tunicate species were sampled in this study. Nonnative tunicates are more frequently found near shipping and marina docks (Carman et al., 2009) and several species are already present on the BC coast (Clarke Murray et al., 2011). Given that I sampled at nine major ports in BC, I expected to find fouling nonnative tunicates. That I did not find such species could be due to my dredging and core sampling methods, which may not have been optimal for targeting species that have been shown to foul the stalk and blades of eelgrass in other regions of the northeast Pacific (Carman & Grunden, 2010). However, dredging methods in this study were successful at collecting native epiphytic species such as the bryozoan, Membranipora membranacea, which were present on eelgrass blades, and during visual surveys the presence of native kelps and the epiphytic red alga, Smithora naiadum, were recorded. In addition, core samples included whole eelgrass plants, which were searched for

63 3.5 Discussion invertebrate species when sorting samples. Accordingly, I would have expected to sample solitary or colonial tunicates if they had been common in my selected BC eelgrass beds.

3.5.4 Conclusions While shipping remains a common source of introduced species globally, this vector may be less important in BC now that recent ballast flushing regulations require mid-ocean exchange of ballast water before ships arrive at ports (Transport Canada, 2006), though introductions still occur (DiBacco et al., 2011). With aquaculture transport already slowing internationally as a result of reduced shellfish imports and improved regulations (International Council for the Exploration of the Sea (ICES), 2005; Canadian Food Inspection Agency, 2011), the chances of a new introduction are similarly reduced. However, at a local-scale aquaculture is likely to remain an important vector through secondary transport of nonnatives via the movement of shellfish and equipment, until preventative management action reduces further spread (Cohen & Zabin, 2009; Clarke Murray et al., 2011).

Nonnatives in BC eelgrass have successfully passed the first two selection filters for nonnative establishment: arrival vector and environment (Colautti & MacIsaac, 2004). The most important environmental factor for explaining nonnatives in BC eelgrass is variation in climate—temperature in particular. Temperature and salinity differences are clear across the BC coast, with the warmer and less saline waters of the protected inner channels apparently promoting more nonnative survival than regions along the exposed coast. In addition, these protected regions contain the greatest density of vectors (shipping, aquaculture leases, recreational boats, etc.). The low number of nonnatives in eelgrass beds as compared to all habitats on the BC coast may indicate limited habitat and resources availability in eelgrass or that newly introduced species are not those that would use eelgrass habitats for grazing and shelter (Heck & Orth, 1980; Duffy & Harvilicz, 2001). Further tests of nonnative introductions in eelgrass should aim to investigate species characteristics, dispersal limitations and environmental limitations to clarify why nonnatives are relatively rare in these habitats.

64 3.5 Discussion

Growing evidence demonstrates that seagrass meadows are experiencing worldwide decline primarily as a result of human disturbances, such as direct physical damage and deterioration of water quality (Short & Wyllie-Echeverria, 1996; Hemminga & Duarte, 2000). With more than half of nonnatives found in BC eelgrass having known negative impacts on eelgrass habitats, the threats of these species in conjunction with anthropogenic pressures has the potential to dramatically impact the health of BC eelgrass beds (Ban & Alder, 2008).

65 4.1 Synopsis

Chapter 4 An Atlantic infaunal engineer is established in the northeast Pacific: Clymenella torquata (Polychaeta: Maldanidae) on the British Columbia and Washington Coast1

4.1 Synopsis The Northwest Atlantic bamboo worm Clymenella torquata, believed to have been imported with commercial oyster culture, was last formally reported from the American Pacific coast more than 30 years ago from a single location. I report here that it is broadly distributed in British Columbia, Canada and is now established in Washington State, USA. In Samish Bay, Washington, this tubiculous infaunal worm creates a spongy, porous substrate that has proved detrimental to commercial oyster farms by causing the oysters to sink into unconsolidated sediments and suffocate. Little is known about the ecological or economic impacts of this invasion in the northeast Pacific.

4.2 Introduction The introduction of nonnative species may pose a serious economic and ecological impact to coastal ecosystems (Pimentel et al., 2000; Hoagland & Jin, 2006). While many marine invertebrate invasions have been documented, introduction of marine polychaete worms have received less attention compared to other groups (Eno et al., 1997; Ruiz et al., 2000a). I document here the expansion in the northeast Pacific of a well-known north Atlantic polychaete, last formally reported more than three decades ago.

Clymenella torquata Leidy, 1855 (hereafter Clymenella, Figure 4.1) commonly known as the bamboo worm, a member of the family Maldanidae, is native along the Atlantic and

1 A version of this chapter has been published: Mach, M.E., Levings, C.D., McDonald, P.S., & Chan, K.M.A. (2012) An Atlantic infaunal engineer is established in the Northeast Pacific: Clymenella torquata (polychaeta: maldanidae) on the British Columbia and Washington Coasts. Biological Invasions 14(3):503- 507.

66 4.2 Introduction

Gulf coasts of North America (Mangum, 1964; Fauchald et al., 2009). The bamboo worm is a filter feeding polychaete which constructs a cylindrical tube about 20 cm long perpendicular to the surface (Mangum, 1964). In its native range these tubes can be extremely numerous, occurring in densities up to 675 m-2 on intertidal sandy mud flats (Mangum, 1964), and on rare occasion in much higher numbers (150,000 m-2; Sanders et al. 1962), creating a "spongy" effect on the sediment (Sanders et al., 1962). Like all bamboo worms, Clymenella torquata are aptly named for their truncate ends and long, cylindrical segments giving them a jointed appearance. Within the family, the genus Clymenella is distinguished by the deep membranous collar on the anterior margin of setiger (segment) four that extends up over part of setiger three. Clymenella torquata (hereafter Clymenella) differs from its west coast congeners by lacking acicular spines in the anterior neuropodia, and by possessing a total of 18 rather than 21+ setigers (Banse, 1981; R. E. Ruff, personal communication).

Figure 4.1 Clymenella from Roberts Bank, BC, collected in 2008. Clymenella has a membranous collar on the anterior margin of setiger four which extends over part of setiger three (white arrow) and rostrate hooks on setiger one (black arrow). Scale bar = 2mm (Photo: T. Goodman) © BC Biodiversity Lifedesk, 2011, by permission.

Non-native populations of Clymenella have been previously reported outside North America only in England (Newell, 1949; Eno et al., 1997), its introduction a consequence of the American oyster trade between 1870 and 1936 (Eno et al., 1997; Wolff & Reise, 2002). Past studies have described the British Clymenella as unusually large (maximum length 15 cm) when compared to American populations (11 cm) (Mangum, 1964;

67 4.2 Introduction

Pilgrim, 1965), however samples from Massachusetts suggest American Clymenella are capable of surpassing this length (19.5 cm) (Rankin, 1946).

4.2.1 Clymenella in the northeast Pacific Clymenella was last formally recorded on the northeast Pacific coast in Boundary Bay, British Columbia (Figure 4.2) based upon specimens collected in 1980 in eelgrass (Zostera marina) beds (Banse, 1981). Banse reported densities averaging 416 m-2 and ranging from 16 to 1680 m-2 and suggested the species had been introduced in the 1930s with importations of the Atlantic oyster Crassostrea virginica from the North American Atlantic coast. There have been no subsequent reports of Clymenella in British Columbia.

Figure 4.2 Known distribution of Clymenella in the Pacific. Site number and year of record: 1) Boundary Bay, BC (1980); 2) Samish Bay, WA (2007); 3) Roberts Bank (2008); 4) Campbell River (2008). Black circles and open circles are sites where infaunal cores were taken but no Clymenella found. Shaded area: Clymenella not present (see text).

68 4.3 Methods

In the mid-2000s Clymenella was reported in Samish Bay, Washington (Figure 1), about 75 km south of Boundary Bay (Figure 4.1; Eissinger, 2008, as "Puget Sound", but sample identified from Samish Bay, A. Eissinger, personal communication; Hancock et al., 2008; P. Blau of Blau Oyster (Samish Bay), personal communication), but these records have not benefited from formal publication. Clymenella is now abundant in Samish Bay (P.A. Dinnel, R.E. Rogers, W. Dooey, and R.E. Ruff, personal communications). Samish Bay mudflats support aquaculture of the nonnative Pacific oyster Crassostrea gigas. Around 2006, the first populations of Clymenella were found on the northeast side of Samish Island on lands leased to Blau Oyster. The worms are reported to have since spread patchily throughout Samish Bay, negatively affecting many farmed tidelands. Oysters in this area are typically grown using “on-bottom” culture methods; thus when Clymenella densities are high, the oysters sink into the porous sediment and suffocate (Rogers, 2007; P. Dinnel, personal communication). The decline in sediment firmness associated with high worm densities is similar to what has been reported within its native range (unpublished data; Sanders et al., 1962). I investigated the expanded range of Clymenella north into the Strait of Georgia, British Columbia, and south into Puget Sound, Washington to clarify its current population status in the northeast Pacific.

4.3 Methods As part of the baseline study of native and nonnative species in seagrass beds described in Chapter 3, infaunal samples were collected at 9 sites along the British Columbia coast during the summer of 2008 (Figure 4.2, black circles and points 3 & 4). At each site, 6 sediment cores (10 cm diameter x 17 cm depth) spaced 25 meters apart along two randomly placed 50-meter transects were collected in Zostera marina eelgrass beds. Cores were sieved at 1 mm and specimens preserved in 95% ethanol for identification. In Puget Sound, additional haphazard sampling was done at 9 sites between 2008 and 2011 (open circles and point 2); all Puget Sound sampling targeted preferred habitat of Clymenella (i.e., substrate types and tidal elevations; unpublished data) in areas currently or historically used for oyster aquaculture.

69 4.4 Results and discussion

4.4 Results and discussion I found Clymenella at sites in British Columbia: Roberts Bank (N 49.018o lat, W 123.119o long) and Campbell River, mid-Vancouver Island (N 50.057o lat, W 125.262o long) (Figure 4.1). Roberts Bank and Campbell River are 20 km and 240 km, respectively, northwest of Boundary Bay, yet Clymenella was not found at sampling sites located between these sites. Abundance of Clymenella at Campbell River (367/m2, SD +/- 367, sampled in 3 of 6 cores) was almost the same as that measured in 1980 in Boundary Bay, while Clymenella at Roberts Bank was more than four times as dense (2056/m2, SD +/-1681, sampled in 6 of 6 cores) (Figure 4.3). Identification of Clymenella was confirmed by R. Eugene Ruff (Ruff Systematics, Puyallup, WA) and reference specimens have been deposited in the collections of the University of British Columbia’s Beaty Biodiversity Museum and digitized on the Marine Biodiversity of British Columbia LifeDesks webpage (Nelson & Goodman, 2010).

Figure 4.3 Abundances (and standard deviations) of Clymenella in Boundary Bay in 1980 (Banse 1981) and Roberts Bank and Campbell River in 2008, as estimated through sediment cores.

No Clymenella were found at Puget Sound sampling sites outside of Samish Bay (Site 2, Figure 1), including Padilla Bay (1 site), Thorndyke Bay (1 site), Case Inlet (3 sites), Hartstene Island (1 site), Eld Inlet (1 site), and Totten Inlet (1 site). As of 2011, Clymenella has not been reported in central or southern Puget Sound (personal

70 4.4 Results and discussion observation; M. Dethier, personal communication) nor has the species been found in coastal embayments of Washington State, such as Willapa Bay (personal observation; S. Booth, personal communication).

4.4.1 Dispersal in the northeast Pacific Clymenella is believed to have a benthic larval phase (Newell, 1951), presumably not amenable to ballast water transport or long distance dispersal in ocean currents. Although first reported in Boundary Bay, where large numbers of Atlantic oysters were transplanted in the 1930s, the site of initial introduction in the northeast Pacific is not known, nor is it known if there were multiple separate introductions with different oyster importations. Oyster farmers in the Samish Bay region believe Clymenella arrived on the hull of a barge from north of Samish Bay, possibly from Boundary Bay (Paul Blau, personal communication). As Clymenella is not a fouling organism, it more likely moved south, if it was not historically present in low numbers in Samish Bay (the species had been overlooked in Boundary Bay for nearly 50 years if it was originally introduced with Atlantic oysters), by other means, such as the movement of oysters and aquaculture equipment. Similar oyster translocations may explain its presence far to the north in Campbell River assuming the population present at this location is not due to an early but long-undetected introduction. Molecular analyses could be used to resolve the population structure in Clymenella as it has for numerous other nonnative species, such as Carcinus maenas, Littorina littorea, and Gemma gemma (Roman, 2006; Blakeslee et al., 2008; Hoos et al., 2010)

4.4.2 Monitoring and management In Samish Bay various control strategies for Clymenella have been tested (P. Dinnel, unpublished data). Physical methods, including tilling (using rotating blades to break up the sediment and worms) and application of shell pavement, significantly decreased worm biomass and number of worm tubes, while increasing sediment firmness, in experimental plots (Hancock et al., 2008). Oyster farms in Samish Bay are attempting to employ these methods at a local level while adapting their practices to limit further spread of Clymenella in the area (B. Dewey, personal communication). In some cases,

71 4.4 Results and discussion growers have discontinued operations where density of Clymenella is high (P. Blau, personal communication).

Given the importance of Puget Sound and the Strait of Georgia to oyster aquaculture (PSAT 2003) and the potential for continued and increasing economic costs of Clymenella's impact on oyster farms, it will be critical to monitor populations of Clymenella and further explore control methods. Data on habitat selection and modification, reproduction, and species interactions will also be important for predicting the relative impact of this invasion should it spread further south into Puget Sound or increase its range in the Strait of Georgia.

72 5.1 Synopsis

Chapter 5 Trading green backs for green crabs: evaluating the commercial shellfish harvest at risk to European green crab invasion

5.1 Synopsis Nonnative species represent a threat to native biodiversity and can have immense impacts on biological communities, altering ecosystem function and services. How much value is at risk from high-impact invasive species, and which parameters determine variation in that value, is critical knowledge for directing both management and research, but it is knowledge rarely available. I evaluated the value of commercial shellfish harvest at risk in nearshore ecosystems of Puget Sound, Washington State, from the European green crab, Carcinus maenas. I assessed this value using a simple static ecological model combined with an economic model using data from Puget Sound’s shellfish harvest and revenue and the relationship between C. maenas abundance and commercial shellfish harvest and biomass. The model incorporates a range in C. maenas diet preference, calories consumed per year, and crab densities. C. maenas is likely to prey on commercially harvested hardshell clam, oyster, and mussel which would likely reduce the total value, including processing and distribution revenue, and the number of jobs associated with these fisheries.

The model suggests possible harvest losses of these species across increasing calorie diets ranging from $1.03 million to $23.8 million USD (2.2 - 56% loss) and increasing invasion densities ranging from $2.5 million to $23.8 million USD (2.5 – 56%), with secondary losses up to $17.3 million USD (48%) and a loss of up to 383 job positions each year associated with a range of plausible parameter values. The broad range of values reflects the uncertainty in key factors underlying impacts, factors which are highly variable across invaded regions. Future research evaluating species invasions can reduce the uncertainty of impacts by characterizing several key parameters: density-dependent predation rates, switching prey species, and limited prey recruitment limitations of prey. This study therefore provides direction for research to inform more accurate estimates of

73 5.2 Introduction harvest, revenue and value at-risk, and it provides substantial motivation for strong measures to prevent, monitor, and manage the possible invasion of C. maenas.

5.2 Introduction In coastal ecosystems, preventing and mitigating the spread and impacts of nonnative species has become a global priority (Millennium Ecosystem Assessment, 2005; Lodge et al., 2006). While many nonnative species have little to no impact on their invaded regions, a few have caused great economic and ecosystem harm (Pimentel et al., 2005; Colautti et al., 2006). The impacts from these few invasive species can affect ecosystem function and thereby reduce the benefits ecosystems provide for people (Hobbs & Huenneke, 1992; Wilcove et al., 1998; Grosholz, 2002; Allendorf & Lundquist, 2003; Altman & Whitlatch, 2007). For example, zebra mussel, Dreissena polymorpha, invasions in the Laurentian Great Lakes have fouled industrial water intake pipes, enhanced macrophyte growth and reduced water turbidity altering habitat conditions for native fish (MacIsaac, 1996). In the USA the cost of invasive species impacts has been estimated at over $120 billion per year (Lockwood et al., 2005; Pimentel et al., 2005). Invasive species are difficult to eradicate, suggesting preventing invasion as the best option for limiting impacts. With limited funds to manage and research coastal ecosystems, calculating the value-at-risk (the losses that might accompany establishment of a high-impact invasive species prior to introduction) for areas not yet invaded may justify the allocation of resources to prevent, mitigate, or further understand invasive impacts (Shogren, 2000; Keller et al., 2007).

The concept of value-at-risk to invasive species should be distinguished clearly from prediction. Rather than a statement of what is expected to occur, value-at-risk can provide decision-makers with a sense of what might plausibly be lost to invasion without prevention or mitigation. Whereas in finance, value-at-risk is often a monetary quantity subject to loss with a given probability (Jorion, 2006), the concept might also be useful in settings where data limitations restrict the explicit assignment of probabilities. For example, a decision-maker faced with the decision whether to fund action to prevent biological invasion, or to put in place mechanisms to mitigate losses should invasion

74 5.2 Introduction occur, might only wish to know what value—in revenues, jobs, etc.—might plausibly be subject to loss, based on our current understanding and its limitations. For such decisions, which are faced everyday, waiting for better data or even a more sophisticated model may not be an option. Just as this quantity of value-at-risk can motivate management action, it can also motivate research to improve estimates of value-at-risk, and perhaps even enable prediction.

Economic value-at-risk can be seen as a product of two components: ecological (plausible ecological changes that might result from the introduction of a known invader) and economic (the economic costs that might be associated with the above ecological changes). Ecological consequences have been assessed mainly as projected post-invasion impacts or pre-invasion estimated ecosystem changes (Kolar & Lodge, 2001; Colnar & Landis, 2007; Pejchar & Mooney, 2009). For example, ecological impact estimates of the European green crab, Carcinus maenas, suggest a future loss of valuable habitats and native species abundance in invaded areas, while larvae of the green crab may also provide food resources to migrating salmon in the northeast Pacific (Colnar & Landis, 2007). The loss or gain of benefits human societies derive from natural ecosystems, or ecosystem services, (Daily et al., 1997; Millennium Ecosystem Assessment, 2005), resulting from invasion can have important economic effects.

The economic consequences of invasive species can be estimated through damages to resources (Reinhardt et al., 2003; Pimentel et al., 2005; Colautti et al., 2006). However, many studies that evaluate economic costs do so with considerable simplification of important ecological processes, potentially misrepresenting costs (Barbier, 2001; Born et al., 2005; Knowler, 2005). For example, damages have been measured as reduced fishery values since time of invasion (Campbell, 1993; Lafferty & Kuris, 1996), but these estimates do not consider other potential causal factors or variation in key traits of the native or invading species, such as varying invasion densities, predation rates or varying spatial distribution of native species. Economic valuations can better represent impacts of nonnative species if they incorporate specific information on both the native and invasive populations and their potential interactions (Knowler, 2005).

75 5.2 Introduction

There are great uncertainties in predicting impacts of species invasions (Perrings et al., 2000; Perrings, 2005), in part because impacts vary across space and time and are otherwise context-dependent (Williamson & Fitter, 1996; Ruiz et al., 1997; Mack et al., 2000; Padilla, 2010; Thomsen et al., 2011). To be useful, evaluations made before invasions occur should incorporate economic costs and uncertainties associated with the invasion’s possible ecological consequences (Barbier, 2001; Leung et al., 2002). If value- at-risk estimates are structured to enable explicit assessment of the uncertainties associated with key parameters, even a coarse understanding of potential impacts can yield useful assessments. Ecological-economic models offer the opportunity to estimate economic changes that include these uncertainties (Born et al., 2005). Though the value- at-risk of predicted invasions has not frequently been calculated, preemptive management strategies that incorporate ecological modeling could have considerable economic benefits (Born et al., 2005; Keller et al., 2007). For example, if early preventative efforts had been funded to prevent invasion of the rusty crayfish, Orconectes rusticus, Vilas County, Wisconsin would have protected $6 million USD in fisheries harvest revenue during the 30 years since invasion (Keller et al., 2007). These calculations serve a critically important purpose of aligning future research and management (Perrings, 2005).

I assessed the value-at-risk for commercial shellfish harvest in nearshore ecosystems of Puget Sound, Washington State, from the green crab, C. maenas. I characterize this value through an ecological model combined with an economic model using existing data on the relationship between C. maenas abundance and commercial shellfish harvest and biomass, on Puget Sound’s current shellfish harvest and revenue, and on the secondary economic effects from a similar neighbouring region (British Columbia, Canada). Furthermore, I characterized the extent to which variation in key parameters influenced the resulting value-at-risk.

76 5.3 Methods

5.3 Methods 5.3.1 Study site and organism: Carcinus maenas as a threat to Puget Sound Nonnative species with broad physiological tolerances and diverse diets, such as C. maenas, are well suited to take advantage of available resources and out-compete native species (Snyder & Evans, 2006). C. maenas is already regarded as a threat to ecosystems in Puget Sound, WA where extensive mudflats, eelgrass beds and warmer inland waters provide optimal habitats (deRivera et al., 2007b; Yamada & Gillespie, 2008). Puget Sound is a large coastal estuary that has been drastically altered by anthropogenic activities (Puget Sound Partnership, 2008a). A risk assessment from Puget Sound and field experiment in the northeast Atlantic have demonstrated green crabs negatively affect natural habitats; Zostera marina eelgrass habitat and associated food webs are disrupted during green crab digging activities (Davis et al., 1998; Colnar & Landis, 2007). The loss of these habitats could have secondary impacts on many of the benefits eelgrass provides to near-shore estuaries, such as nursery grounds for many commercially important fish, shellfish and juvenile crab and essential spawning habitat for herring (Thayer & Philips, 1977; Simenstad, 1994). It is important to note that not all impacts of green crab are likely to be negative: Coho salmon (Oncorhynchus kisutch) were predicted to benefit from invasion by feeding on the larvae of green crab, although this projection is associated with great uncertainty (Colnar & Landis, 2007) and is not included in the current model.

Carcinus maenas is a generalist predator that, in one study, was found to consume species from at least 104 families and 158 genera within 14 animal and 5 plant and protozoan phyla (Cohen et al., 1995), though it most commonly feeds on bivalves (Grosholz & Ruiz, 1996; Grosholz et al., 2000). In feeding trials C. maenas was able to feed more generally and consume a greater biomass than several native northeast Pacific Cancer crab species (Cohen et al., 1995). Green crab has destroyed artificial shellfish beds and consumed juvenile bivalves and Cancer crabs throughout northern New England and the Canadian Maritimes (review in Lafferty & Kuris, 1996), reducing profits from the northwest Atlantic shellfish industry by as much as $22.6 million each year since introduction (Lovell et al., 2007).

77 5.3 Methods

Native to northern Europe, C. maenas has established populations in North America (Carlton & Cohen, 2003), South Africa (Le Roux et al., 1990), Japan (Geller et al., 1997), Argentina (Hidalgo et al., 2006) and Australia (Fulton & Grant, 1900). In North America, green crab was first found on the northern Atlantic coast in 1899 where it has since expanded to cover 1000 kilometers of coast from Virginia in the south to a still- expanding range on Prince Edward Island (Carlton & Cohen, 2003; Miron et al., 2005). In 1989, C. maenas reached San Francisco Bay in the northeast Pacific via ballast water from populations on the northwest Atlantic coast (Elston, 1997).

Secondary spread of green crab north along the northeast Pacific coastline has been correlated with increased seawater temperatures and north-running coastal currents during the 1998 El Nino event, making it particularly likely that future climate change will allow the crab to invade new areas of coastline (Huyer et al., 1998; Yamada et al., 2005). Green crab is limited by temperature and salinity, surviving in temperatures ranging from 0o to 30o C and salinities of 4 to 34 ppt, although reproduction and larval survival occur in a more limited range than the adults (review in Cohen et al., 1995). Predictions based on environmental limitations suggest that under current conditions, green crab will continue expanding northward from its current northern extent of Vancouver Island until it reaches the Aleutian Islands (Carlton & Cohen, 2003) and may enter the contiguous waters of Puget Sound and the Strait of Georgia, British Columbia either by secondary introduction events from ballast release from large shipping freighters or through natural larval dispersal during one of the next El Nino events (Figure 5.1) (Jamieson et al., 2002).

78 5.3 Methods

Figure 5.1 The geographic potential of Carcinus maenas for the northeast Pacific coast from Northern California to Alaska, USA. The current nonnative distribution along the coast is indicated by a broad, stippled polygon, while the potential distribution of the species is plotted in black. Figure and Maxtent potential distribution model from deRivera et al. (2011), figure altered to clarify C. maenas’ absence in Puget Sound. © Diversity and Distributions, 2011, adapted by permission.

Studies from C. maenas previous invasions suggest Puget Sound’s commercially harvested bivalve species are a likely target for green crab predation (Lafferty & Kuris, 1996; Jamieson et al., 1998; Lovell et al., 2007). Cost predictions of green crab’s future predation impact in the northeast Pacific has been estimated at $844,000/year once green crab expands into Puget Sound and up to the Aleutian Islands (Lovell et al., 2007). This coarse estimate may be low as ie does not include regional variation in feeding rate or all harvest shellfish species at risk, though it does estimate costs across increasing invasion densities. Green crab invasion in Puget Sound is predicted to impact economically important hardshell clam, oyster and mussel species (Grosholz et al., 2000) all of which bring in millions of dollars in direct and indirect revenue to the Puget Sound region (Dethier, 2006; NOAA Fisheries, 2008).

79 5.3 Methods

5.3.2 Biomass and revenue of Puget Sound’s shellfish harvest To estimate commercial harvest for Puget Sound commercial shellfish industry, I obtained commercial harvest data for 2009 from PacFIN on May 26, 2010 (Pacific Fisheries Information Network, 2009). Data were apportioned to individual Puget Sound Partnership (PSP) action areas (Figure 5.2; Puget Sound Partnership, 2008b) by intersecting these with Washington Department of Fish and Wildlife (WDFW) Shellfish Management and Aquaculture growing areas using ArcGIS 9 Intersect tool (Figure 5.3; completed by Mark Plummer, NOAA). Data were assigned to PSP areas assuming harvest occurs uniformly throughout the growing area. Commercial data included harvest (pounds per year) and total revenue (USD per year) for hardshell clam (Venerupis philipinarum and Protothaca staminea), oyster (Crassostrea virginica and C. gigas) and mussel (Mytilus spp.) in six PSP action areas (hereafter, harvest areas). Shellfish biomass within each species group was summed to create a total estimated biomass for each harvest area. This biomass was then used to calculate an average cost per pound of shellfish (USD lb-1). These data were used as the baseline estimate for current shellfish harvest and revenue before green crab invasion.

80 5.3 Methods

Figure 5.2 Puget Sound Partnership action area map (2012). Each color represents a unique watershed and harvest region within Puget Sound. Included in this study are data on shellfish harvests from all shellfish harvest areas except South Central Puget Sound (Olive Green). © Puget Sound Partnership, 2012, by permission.

81 5.3 Methods

Figure 5.3 Washington State Department of Health (2010) commercial and recreational shellfish “Annual Inventory Growing Areas Map”. Green polygons represent approved commercial shellfish growing areas. © Washington State Department of Health, 2010, by permission.

82 5.3 Methods

The six harvest areas of Puget Sound with complete commercial harvest data for each taxa group were Hood Canal, North Central Puget Sound, Strait of Juan de Fuca, Whidbey Island, and South Central Puget Sound. South Central Puget Sound only represents 0.4% of total harvest area and was not included in the analyses.

Shellfish species in Puget Sound are commercially harvested from mudflats or grown in aquaculture farms. All species evaluated in this study spend a portion of their life-cycle in the near-shore where they are susceptible to green crab predation:

a) Hardshell clam – Manila (V. philipinarum) and native Littleneck (P. staminea) – are harvested on tidal flats throughout Puget Sound at sediment depths of less than 15 cm. Hardshell clams are either raked off beaches where they grow naturally or their beds are “seeded” (seed clams are sown onto beaches leased from Washington State). These two species make up 98% of total hardshell clam harvest.

b) Oyster – European (C. virginica) and Pacific oyster (C. gigas) – these species are harvested from populations that grow without assistance in the high subtidal/low intertidal and on aquaculture farms where they are grown directly on mudflats, on racks sitting on the bottom substrate, or suspended under floating rafts.

c) Blue mussel (Mytilus spp.) grow on rocks in the high subtidal/low intertidal and are harvested on state approved beaches and on aquaculture farms grown on racks sitting on the bottom substrate or suspended under floating rafts.

83 5.3 Methods

5.3.3 Model of Carcinus maenas impacts on shellfish harvested in Puget Sound To model C. maenas’s predation impact on shellfish in Puget Sound I applied the following simple linear model to estimate the total pounds of each of three species of shellfish (hardshell clam, oyster, mussel) consumed each year by C. maenas (Consumption):

Consumption = (Area * Den * Cal * Diet) / Cal lb-1

The variables Area, Den, Cal, Diet, and Cal lb-1 are defined below.

Area: Commercial shellfish harvest area (km2) in Puget Sound (Area) was estimated as 523.71 km2 by tracing polygons around the shellfish harvest areas in the Washington State Department of Health “Annual Inventory Growing Areas Map” (Figure 5.3) (2010) within each of the five PSP action areas (Figure 5.2) (2008b) using Image J (Table 5.1; Rasband, 2009).

Den: Density of adult C. maenas invading the harvest areas (crabs km-2) was represented as a range of possible invasion densities. The high density estimate, 100,000 km-2 (High), represents maximum adult green crab densities from one study in C. maenas’s native range, in Sweden (Pihl & Rosenberg, 1982). The high density estimate for population density averaged over time is justifiable given that densities are sometimes considerably higher in invaded ranges due to lower predation pressures and greater available resources (Shea & Chesson, 2002; Lambrinos, 2004). The low density estimate, one order of magnitude less than high at 10,000 km-2 (Low), is equivalent to average adult green crab densities in their native range (Pihl & Rosenberg, 1982). I selected a medium density estimate of 50,000 km-2 (Medium), roughly midway between high and low estimates. I refer to studies of densities from the native range because studies from invaded regions use catch per unit effort (CPUE) to estimate invasion densities (Baeta et al., 2006; Hidalgo et al., 2006; Tremblay et al., 2006; Yamada & Gillespie, 2008) and CPUE is difficult to translate to a density of individuals per area because studies use different traps

84 5.3 Methods or dredges, set them using different methods, and leave them to catch crab for differing lengths of time.

Cal: The number of calories consumed by each adult C. maenas per year was estimated by extrapolation from laboratory diet studies as the number of calories mg-1 (ash-free dry weight, AFDW) of shellfish meat (clams, 6.15 cal mg-1; oysters, 4.85 cal mg-1; mussels 5.47 cal mg-1; as reviewed by Beukema 1997). I then multiplied the number of calories of shellfish meat by the number of mg per day a C. maenas (25 to 32 cm carapace) was found to consume. However, green crab prefers to prey on juvenile shellfish (Mascaro & Seed, 2001; Walton et al., 2002) while shellfish harvest biomass is of adult shellfish, estimates based on harvested adult may overestimate predation impacts. In addition, this model assumes all harvested shellfish biomass is accessible to green crab predation though aquaculture using racks suspended above the bottom sediment to grow shellfish will limit or prevent predation, and clams may burrow deeper than green crab can dig through the sediment reducing predation on these species.

Diet: The proportion of C. maenas diet consisting of each shellfish species group being modeled was estimated as ranging from 0.20 to 0.35. Grosholz and Ruiz (1996) demonstrated that green crab diet is similarly dominated by bivalves in each region it has invaded. My estimate range for diet assumes that 60% - 100% of C. maenas’s diet will be harvestable hardshell clam, oyster, and mussel, thus allowing for other species to comprise up to 40% of C. maenas’s diet (Grosholz & Ruiz, 1996; Yamada et al., 2010). Estimates of Diet do not include prey-switching by green crab, which may occur as preferred shellfish biomass is reduced during invasion.

Cal lb-1: I used calories mg-1 AFDW (ash free dry weight) of shellfish meat (as in Cal; reviewed by Beukema 1997 (as in Cal; reviewed by Beukema, 1997), to calculate the number of calories per pound of shellfish (Cal lb-1). AFDW was converted to wet weight (WW), the unit of biomass for shellfish harvested in Puget Sound, using conversions reviewed in Ricciardi & Bourget (1998). Conversion estimates were different for each species group and include a 95% confidence interval for all estimates that were made

85 5.3 Methods using more than one study. I used AFDW/WW conversion estimates of M. edulis for mussels (2.5 - 6.7), C. virginica for oysters (1.7), and the average conversion estimates of all bivalves for hardshell clams (5.2 - 6.4; combined for the two hardshell clam species).

I accounted for uncertainty in these parameters in two ways. First, where the range in potential data for the variable is large, as described below for crab density (Den) and calorie diet for each crab (Cal), I set three broad estimates (high, medium, and low) across the range of data from the literature and modeled these as separate scenarios for C. maenas consumption. Second, where the range in data uncertainty was smaller, I represented the uncertainty through randomization within scenarios defined as above, by choosing values across a uniform distribution for harvest area (Area), the proportion of each crab’s diet that is the shellfish being analyzed (Diet), and the number of calories per pound of each shellfish species (Cal lb-1). Den and Cal parameters are arguably not independent, certain values of one have the potential to constrain values of the other. This is an important possibility for which there is insufficient data to model the relationship given the presence of other complicating factors, such as limited recruitment of harvestable shellfish (see 5.5 Discussion).

To implement the randomization for Area, Diet, and Cal lb-1, I used a Markov Chain Monte Carlo (MCMC) algorithm using R software (R Development Core Team, 2012). Data were generated by resampling 10,000 times within the constraints described below for each parameter in the consumption model. Due to the specific values for estimating Area and Diet I used the runif() function in R. Runif creates a flat line between two values along which samples are randomly estimated. For Area I assumed this range to be relatively small, +/- 75 km2, though some error was possible as a result of apportioning of WDFW Shellfish Management and Aquaculture areas to PSP action areas, and the measurement of harvest areas using Image J. The Diet proportion was evenly sampled between 0.20 and 0.35 for each shellfish species. Considering C. maenas has not yet invaded Puget Sound nor had its predation preference tested for any of these shellfish species, precise estimates of its diet preference with the range of species it will encounter in the Puget Sound region were not available. In addition, these values likely vary

86 5.3 Methods depending on availability of shellfish and ease of predation, thus I chose diet preferences values evenly across a range of proportions. I calculated the range in number of calories per pound of each shellfish species using the range of conversion estimates between AFDW and WW for each shellfish species group (Cal lb-1). An example of the R code used for these analyses is available in Appendix D.

I estimated green crab impact on shellfish in harvest areas for high, med, and low invasion densities at high, med, and low calorie diets by calculating the error for mean consumption at each density (95% confidence interval from randomizations within each scenario). This calculation resulted in an upper and lower estimate of consumption of total shellfish biomass in harvest areas for each combination of invasion density and calorie intake. Impact on annual harvest of each shellfish species was then estimated as the total annual harvest minus the upper and lower consumption estimates. These methods were repeated for each of the three shellfish species groups: hardshell clam, oyster and mussel.

In order to represent parameter combinations that fall between the consumption scenarios for Den and Cal, I performed a subjective sensitivity analysis of parameters in the model (Hamby, 1994). In this analysis, I allowed Den and Cal to range freely from zero to high estimates, with a uniform distribution, and held the other parameters to the same constraints as described above, except Diet of harvested shellfish was estimated as 60% to 100% of green crab diet to include all three harvested shellfish species groups. Data were again sampled with the MCMC algorithm using R software (R Development Core Team, 2012), with 10,000 replicate samples within the parameter constraints.

5.3.4 Estimating impacts to shellfish value in Puget Sound I considered primary economic value of shellfish harvest in terms of existing harvest revenue (landed value) and secondary economic value in terms of processing and distribution value and direct impacts from primary and secondary value in terms of labour income and employment. To evaluate the primary economic value-at-risk to green crab predation on shellfisheries in Puget Sound, I estimated the loss of existing harvested

87 5.3 Methods revenue from the total revenue of hardshell clam, oyster and mussel for high, medium, and low densities of green crab and across high, medium, and low calorie diets. Loss of shellfish harvest revenue, which is calculated as USD per lb, was assumed to decrease in parallel with the loss of harvested shellfish biomass to green crab predation as estimated by the consumption model.

Further estimates of the secondary economic value direct impacts of Puget Sound shellfisheries: processing, distribution, and labour values, were made using benefit transfer methods as data on secondary value were not available for shellfish species in Puget Sound. Thus, I compared the known revenue of Puget Sound’s shellfish harvest to an analysis done on the value of shellfish harvest in British Columbia (BC), an adjacent region to Puget Sound (Table 3; GSGislason & Associates Ltd., 2007). GSGislason & Associates Ltd. (2007) estimated the value of all of BC’s shellfisheries (in CAD) in 2005 for harvesting of shellfish, which involves the use of beach harvest, diving and other gear and aquaculture of shellfish from seed to market size. Secondary economic values were estimated for all fisheries species (including both fish and shellfish). To calculate direct impacts of only shellfisheries I assumed the proportion of each secondary value as compared to all fisheries harvesting value was the same for harvesting value of shellfisheries.

Secondary values were calculated for 1) the processing margin, which includes transportation from sea to processing plants and the processing of raw shellfish, 2) the distribution margin, defined as the delivery of these processed shellfish products to consumers through wholesale and retail food channels, 3) the direct impacts of the seafood industry on labour income, which includes wages, salaries, and employer contributions to health and dental plans, pension plans, etc., and 4) employment years (EY) in persons per year (Statistics Canada, 2005). I then assumed the secondary economic value of shellfish harvest as compared to known harvest revenue is proportionally the same as for shellfisheries in Puget Sound as in BC and calculated the ratio of Puget Sound shellfish harvest revenue to BC shellfish harvest revenue (Table 5.3). This ratio was then assumed to be the same for the secondary value and direct

88 5.4 Results impact of commercial shellfish processing, distribution, labour income and employment for Puget Sound. I then estimated C. maenas impact on these secondary economic values at high, medium, and low invasion densities and high, medium, and low calorie diets. I assumed the loss of shellfish harvest revenue and secondary values decreased at the same rate as more green crabs invade and consume increasing calories per crab as estimated by the consumption model.

5.4 Results 5.4.1 Commercial shellfish harvest The baseline for total shellfish harvest in Puget Sound for 2009 for hardshell clam, oyster and mussel recorded by PacFIN (Pacific Fisheries Information Network, 2009) was 2.84 million lbs of shellfish, with a landed harvest value of $37.26 million (Table 5.1). Hardshell clams had the highest biomass harvested out of the three species groups (7.5 million lbs) and the greatest associated revenue, $18.3 million USD even though oyster species are valued per pound at almost twice that of hardshell clam, $4.54 lb-1 compared to $2.09 lb-1 respectively.

5.4.2 Commercial shellfish harvest at risk from green crab invasion

Using the consumption model, I estimated harvested shellfish biomass and total shellfish harvest revenue associated with scenarios of low, medium and high calorie diets and densities for green crab: the medium-medium (Cal-Den) scenario yielded a value-at-risk estimate of 1.2 million lbs and $3.72 million USD in harvest revenue; the low-low scenario suggested a minor loss of only 0.04 million lbs and $0.08 million USD, and medium-high, high-medium, and high-high scenarios suggesting losses of at least 2.18 million lbs and $6.76 million USD (Table 5.2 and Figure 5.4). Mussels were the only shellfish to reach an estimated value loss of 100% for the high-high scenario (Figure 5.4). Note that these losses pertain to losses of harvested biomass, not of all prey organisms, accounting for the possibility that green crab might eliminate harvestable biomass without causing local extirpation (see 5.5 Discussion).

89 5.4 Results

Table 5.1 Harvest of shellfish species a) hardshell clam, b) oyster, and c) mussel, by PSP action area in Puget Sound, size of harvested area (km2), pounds of shellfish (in millions of pounds), total revenue (in millions of dollars, USD) and average price per pound (USD/Lb) (Pacific Fisheries Information Network, 2009).

Shellfish Species Action Area Area (km2) Lbsa Revenueb Avg. $/Lb a. Hardshell Clam Hood Canal 126 1.94 $4.65 $2.40 North Central Puget Sound 41 0.10 $0.18 $1.83 San Juan Islands 54 0.42 $0.85 $2.02 South Puget Sound 108 4.67 $11.82 $2.53 Strait of Juan de Fuca 104 0.09 $0.13 $1.37 Whidbey Island 90 0.30 $0.71 $2.39 Puget Sound Total 524 7.52 $18.34 $2.09

b. Oyster Hood Canal 126 1.25 $4.73 $3.78 North Central Puget Sound 41 0.01 $0.01 $2.82 San Juan Islands 54 0.08 $0.43 $5.70 South Puget Sound 108 1.53 $8.35 $5.46 Strait of Juan de Fuca 104 0.03 $0.11 $4.13 Whidbey Island 90 0.08 $0.43 $5.37 Puget Sound Total 524 2.97 $14.06 $4.54

c. Mussel Hood Canal 126 0.002 $.004 $1.96 San Juan Islands 54 0.004 $0.01 $3.50 South Puget Sound 108 0.98 $2.05 $2.09 Strait of Juan de Fuca 104 0.003 $0.01 $2.47 Whidbey Island 90 0.85 $2.78 $1.50 Puget Sound Total 483 2.84 $4.85 $2.30

Grand Total 13.33 $37.26 $2.98 a Millions of pounds b Millions of dollars (USD)

90 5.4 Results

Table 5.2 Shellfish harvest before and after green crab predation in Puget Sound. The baseline shellfish biomass harvest (millions) and direct harvest revenue value (millions, 2009 USD) are compared to biomass of the shellfish harvest at increasing green crab densities (low, medium, high) and increasing calories consumed by green crab each year (a) low, b) medium, c) high calorie diet). % loss is the percent change in shellfish biomass from the baseline. Standard deviations are not presented, as they are all less than 10,000. a Millions of pounds; b Millions of dollars (USD) Crab Densities a. Low Calorie Diet (% loss) Baseline (0) Low (0.5%) Medium (0.6%) High (2.4%) Species lbsa $b lbs a $b lbs a $b lbs a $b Hardshell Clam 7.52 $18.30 7.5 $18.31 7.49 $18.29 7.37 $17.97 Oyster 2.97 $14.10 2.94 $14.04 2.94 $13.93 2.91 $13.80 Mussel 2.84 $4.80 2.83 $4.83 2.82 $4.81 2.72 $4.64 Total 13.33 $37.26 13.27 $37.18 13.25 $37.03 13 $36.41 Difference from Baseline 0.08 $0.08 0.08 $0.23 0.33 $1.03

b. Medium Calorie Diet (% loss) Baseline (0) Low (4.1%) Medium (9.0%) High (40.5%) Species lbsa $b lbs a $b lbs a $b lbs a $b Hardshell Clam 7.52 $18.30 7.27 $17.74 7.15 $17.44 5.03 $12.28 Oyster 2.97 $14.10 2.88 $13.62 2.51 $11.88 2.05 $9.69 Mussel 2.84 $4.80 2.64 $4.51 2.47 $4.22 0.85 $1.46 Total 13.33 $37.26 12.79 $35.87 12.13 $33.54 7.93 $23.43 Difference from Baseline 0.54 $1.39 1.2 $3.72 5.4 $13.83

c. High Calorie Diet (% loss) Baseline (0) Low (7.4%) Medium (16.4%) High (73.8%) Species lbsa $b lbs a $b lbs a $b lbs a $b Hardshell Clam 7.52 $18.30 7.06 $17.24 6.85 $16.70 3.01 $7.35 Oyster 2.97 $14.10 2.8 $13.27 2.13 $10.10 1.29 $6.11 Mussel 2.84 $4.80 2.48 $4.23 2.17 $3.70 0 $0.00 Total 13.33 $37.26 12.34 $34.74 11.15 $30.50 4.3 $13.46 Difference from Baseline 0.99 $2.52 2.18 $6.76 9.84 $9.03

91 5.4 Results

Figure 5.4 Mean harvested shellfish biomass (lbs) and value (USD) at increasing levels of green crab densities for four shellfish species: a) hardshell clam, b) oyster, c) mussel at three increasing levels (light grey, low; dark grey, medium; black, high of calories consumed) by each green crab year-1.

92 5.4 Results

The Monte Carlo sensitivity analysis demonstrated the relationship between density and calorie consumption rate on loss of shellfish harvest biomass (Figure 5.5). If crabs invade at high densities but consume calories at a low rate, or vise-versa, they have a limited effect on shellfish harvest. However, at high densities and high calorie intake rates the slope of the shellfish harvest loss is steep and green crab will greatly reduce shellfish harvest biomass.

Figure 5.5 Sensitivity analysis of the loss of shellfish harvest biomass (lbs year-1) as a function of calories consumed each year and the density of C. maenas km-2 in harvest areas. Assumes 60% to 100% of green crab diet is made up of hardshell clam, oyster and mussel.

93 5.4 Results

5.4.3 Value-at-risk associated with commercial shellfish harvest loss

The harvest revenue associated with the total biomass for these three species of shellfish in Puget Sound is $37.26 million USD (26.6% of British Columbia’s revenue). Estimated loss at low, medium and high green crab calorie diets was up to 3%, 47% and 64% at the highest green crab density of 100,000 green crab per km2. Specific values for change in known harvesting values and estimated processing margin and distribution margin values, as well as potential change in labour income, are presented in Table 5.3. Employment associated with shellfish harvest and farming, processing, and distribution is estimated around 687 persons per year (PYs). Green crab invasion at highest densities is estimated to reduce these PYs to 671 at the lowest calorie diet, and to 384 at the high calorie diets (Table 5.3).

94 5.4 Results

Table 5.3 2009 Puget Sound shellfish harvest value data were used to estimate % shellfish loss (in parentheses) at each of three green crab invasion densities (low, medium, high) and green crab calorie intake estimated at three increasing diets: a) low, b) medium, and c) high, to primary harvest value. Original shellfish harvest value (capture and aquaculture) was estimated in British Columbia in 2005 by GSGialason (2007). These values were used as benefit transfer estimates of i) secondary value and ii) direct impacts for total shellfish harvest value. The baseline value, pre-green crab invasion, were calculated as the percent difference of harvest & farm level value between the Puget Sound baseline and the BC 2005 value. Estimates of value loss resulting from green crab invasion at increasing densities were calculated as the baseline multiplied by the % shellfish loss. Crab Densities Puget a. Low Calorie Diet BC Sound Low Medium High i. Primary & Secondary Value $ millions (0.2%) (0.4%) (1.7%) Harvesting 139 37.3 37.2 37.0 36.2 Processing Margin 100 26.7 26.7 26.6 26.1 Distribution Margin 35 9.5 9.5 9.4 9.3 Total 274 73.5 73.4 73.0 71.6

ii. Direct Impacts Labour Income $ millions 94 25.3 25.2 25.2 24.7 Employment PYs 2562 687 685 684 675

Puget b. Medium Calorie Diet BC Sound Low Medium High i. Primary & Secondary Value $ millions (3.7%) (6.5%) (30.2%) Harvesting 139 37.3 35.9 33.6 23.4 Processing Margin 100 26.7 25.7 25.0 18.7 Distribution Margin 35 9.5 9.1 8.9 6.6 Total 274 73.5 70.7 67.5 48.7

ii. Direct Impacts Labour Income $ millions 94 25.3 24.3 23.7 17.7 Employment PYs 2562 687 661 642 479

Puget c. High Calorie Diet BC Sound Low Medium High i. Primary & Secondary Value $ millions (6.8%) (12.2%) (55.9%) Harvesting 139 37.3 34.7 30.5 13.5 Processing Margin 100 26.7 24.9 23.5 11.8 Distribution Margin 35 9.5 8.8 8.3 4.2 Total 274 73.5 68.4 62.3 29.5

ii. Direct Impacts Labour Income $ millions 94 25.3 23.6 22.2 11.1 Employment PYs 2562 687 640 603 384

95 5.5 Discussion

5.5 Discussion By assessing the value-at-risk for commercial shellfish harvest in nearshore ecosystems of Puget Sound, Washington, from the green crab, C. maenas I estimated a range of possible losses that included, at highest invasion densities across increasing calorie diets, up to 0.33 – 9.8 million lbs of shellfish worth $1.03 – $23.8 million USD, a loss of 2.2% – 56%. While at highest calorie diets across increasing invasion densities shellfish harvest loss estimates ranged from 0.99 – 9.03 million lbs worth $2.5 – $23.8 million (6.8% - 56%). This corresponds to a total loss of revenue, including processing and distribution margins, of $1.6 – $41 million USD (19% – 48% loss) and a loss of 15 – 383 jobs (employment in person years). Estimated value-at-risk is likely to vary from site to site within Puget Sound harvest areas as there are likely to be refuges for shellfish populations and the distribution of green crab is unlikely to evenly spread across harvest areas with preferred habitats varied across these regions. Green crab may have the potential to reduce shellfish biomass enough to stop the shellfish harvest industry completely at the highest invasion densities or highest calorie diet rate, depending on those shellfish population densities that result in shellfish harvest area closures.

The reduction of harvested shellfish biomass and shellfish harvest values only apply to harvested shellfish, not the entire natural population that lives on these mudflats. In Puget Sound, Manila clams harvest limits are set at 33% and native Littleneck harvest is limited to 25% of the total population, suggesting there is a total biomass of 29.7 million pounds of Manila clam, and 0.26 million pounds of native Littleneck clams in harvest areas. If green crab significantly contributes to reducing biomass, hardshell clam populations may be reduced below allowable harvest biomass. Thus, reducing populations of harvested shellfish to zero within the consumption model does not suggest there are no shellfish remaining, but rather that when shellfish populations are sufficiently reduced there is no more commercial harvest of shellfish. In addition, green crab is an opportunistic feeder that is likely to feed on other species when preferred bivalve abundance becomes low. So although green crab may decrease shellfish population densities, they are likely to feed on whatever species are the easiest to access, maintaining bivalve populations at low densities but not completely decimating the populations (Baeta et al., 2006). The

96 5.5 Discussion reduction of shellfish biomass also has the potential to limit recruitment of shellfish by reducing the population of reproductive adult shellfish and by preying on newly settled juvenile shellfish. These negative feedbacks may result in greater losses to shellfish then I have estimated.

Although the range in values is broad it appropriately represents the uncertainty in key factors associated with green crab invasion impacts, factors which are highly variable across regions. As such, the most extreme estimates of shellfish harvest loss cannot be excluded, as these estimates are based on current understanding of green crab populations in other regions. Lower calorie estimates are likely more similar to impacts from previous invasions than medium-high calorie because when green crab invade at high densities, it is unlikely to have unlimited access to shellfish. High densities of green crab invasion will reduce harvested shellfish biomass through predation, however if there is intraspecific competition for prey high crab densities are likely to reduce the number of calories consumed per crab (Mansour & Licpcius, 1991). The converse should also be considered; optimal feeding conditions where the highest number of calories are consumed is less likely to occur under heavy competition with conspecifics for prey. Thus, it is unlikely that the highest crab invasion densities will consume calories at the highest rates. The lower calorie estimates demonstrate a limited access scenario and therefore present a more realistic change in biomass. The medium estimates of shellfish harvest loss are reached both at high calorie diets but medium crab densities, and at high crab densities but medium calorie diets, suggesting that there are a number of invasion scenarios that could result in a loss to shellfish harvest, and the range of estimates of green crab impacts should all be considered equally plausible.

In the northeast Atlantic the shellfish industry was estimated to have lost 9.9 million lbs of biomass and $22.6 million in revenue each year from green crab, though this estimate did not include oyster species, potentially underestimating total costs, and did not describe what proportion of the total fishery was lost to green crab (Lovell et al., 2007). Without knowing what the total biomass of this northeast Atlantic fishery it is not possible to compare estimates of value-at-risk to green crab invasion estimated in this

97 5.5 Discussion study. This same study (Lovell et al., 2007) estimated green crab’s impact to Washington State shellfisheries, however shellfish harvest value was underestimated as $17.2 million for the entire state (not including oysters), while in this study I demonstrated that shellfish in Puget Sound alone were worth $37.26 million. The lack of previous data on economic loss to green crab makes comparing my estimates value-at-risk in Puget Sound to previous estimates difficult. Additional value-at risk estimates for regions already invaded by green crab would improve estimates for not yet invaded regions.

5.5.1 Additional ecosystem changes The introduction of C. maenas is a threat to biodiversity and ecosystem function for Puget Sound’s near-shore food web (Lafferty & Kuris, 1996; Jamieson et al., 1998; Yamada & Randall, 2006; Lovell et al., 2007). The dietary preference for bivalves and resulting ecological impact has been relatively similar across invaded regions (Grosholz & Ruiz, 1996). Assuming this remains true for Puget Sound, green crab removal of bivalves may have indirect effects on shorebird populations by removing their prey as seen in other sites in the northeast Pacific (Grosholz & Ruiz, 1996). Green crab bivalve predation also may result in a shift in the bivalve community if bivalve species that are less-preferred by green crab replace those being consumed by green crab (Grosholz, 2005). When this occurred in Bodega Harbor, California, green crab suppressed the native clam, Nutricola spp., which had an indirect positive effect on the nonnative clam, Gemma gemma. Green crab also has the potential to increase productivity in Puget Sound. Though its impacts on shellfish may be great, as larvae green crab provide a prey resource to fish species (Colnar & Landis, 2007) and as adults are likely prey for birds, seals, and fish that normally feed on local crab species.

Recreational shellfisheries in Puget Sound are also likely to be directly affected by loss of shellfish biomass. As number of recreational harvest days per year decreases, harvesters are likely to experience reduced shellfish harvest and loss of cultural benefits, such as family engagement and traditional harvesting by native coastal tribes. Indirect effects may result when local human communities located near harvest beaches experience reduced benefits as fewer harvesters buy supplies, permits, food and lodging. The social

98 5.5 Discussion and environmental-engagement value of recreational shellfish harvest might be difficult to quantify but is an important aspect of harvest value that is often ignored (Bergstrom et al., 2004).

Shellfish threatened by this invasion provide more than just commercial and recreational harvest revenue; by filtering toxins and nutrients from Puget Sound, shellfish help to increase oxygen levels and reduce toxic poisoning in other organisms. Eutrophication in Hood Canal and southern Puget Sound are likely to increase with a decline in shellfish populations (Diaz & Rosenberg, 1995). Because these monetary and nonmonetary values are not incorporated in model estimates, this study likely underestimates the total value lost to green crab invasion.

5.5.2 Motivating prevention and mitigation of invasive impacts Managers can prepare for major losses of harvest and revenue by initiating strong preventative measures (Perrings, 2005; Pimentel et al., 2005; Keller et al., 2007). Current efforts in Puget Sound to prevent green crab invasion include restricting out-of-state imports of shellfish, encouraging commercial shellfish harvesters to inspect their equipment before transferring gear between invaded and non-invaded regions, requiring ballast exchange before entering Puget Sound, and instituting a detection program that incorporates community volunteer groups and paid specialists (Washington Department of Fish and Wldlife (WDFW), 2012a). These efforts are useful but could be improved if more funding was allocated to preventative efforts. For example, funds could be used to better enforce gear inspection and ballast exchange. At this time there is no-ballast exchange required between Oregon, Washington, and British Columbia, though green crab is already present along the outer coast of each of these states and provinces (DiBacco et al., 2011; Washington Department of Fish and Wldlife (WDFW), 2012b). In addition, increasing the number of paid specialists sampling for green crab would increase the chances of catching green crab invasion early and prevent further spread in Puget Sound (Myers et al., 2011).

99 5.5 Discussion

Despite current efforts to limit human introduction of green crab into Puget Sound, climate change resulting in warming sea waters and changing current flow will likely result in larval transport into Puget Sound with no further human assistance (Yamada & Gillespie, 2008; deRivera et al., 2011). Plans to mitigate the impact of green crab will be most effective if they are in place before green crab invasion occurs. Managers can prepare commercial harvesters to take preemptive measures in reducing green crab’s impact by altering their current methods of shellfish aquaculture. Oyster and mussel racks suspended off the bottom substrate may limit green crab’s access to commercial culture of bivalves and is already a common practice in neighboring BC (BC Shellfish Grower's Association, 2012). Anti-predator netting has been used successfully in the northeast Atlantic, reducing loss of clam biomass to predation by 13% to 55% (Beal & Kraus, 2002). Netting may be used to minimized predation on Puget Sound clams and has been tested in BC, though this method has detrimental side-effects on other infaunal species also netted in the mudflat (Munroe & McKinley, 2007). A comprehensive post-invasion plan would combine mitigating impacts with monitoring and control efforts (Horan et al., 2002; Perrings, 2005).

5.5.3 Incorporating uncertainty of future invasion impacts It is not possible to precisely estimate invasion impacts of green crab and other invading species because of temporal and spatial variation and context-dependent differences across invaded regions (Williamson & Fitter, 1996; Ruiz et al., 1997; Padilla, 2010; Thomsen et al., 2011). However, though few data are available about specific impacts before invasion, researchers can refine estimates of value-at-risk by incorporating a range of potential invasion parameters: density of individuals, number of arrivals, potential predation and competition interactions, and economic impacts, these general estimates can produce more accurate assessments of what is known regarding the species that is invading (Shogren, 2000; Perrings, 2005). If estimates are too specific or not made at all there is little motivation to rationalize economic spending to monitor, prevent or mitigate for species invasions before invasion occurs (Born et al., 2005).

100 5.5 Discussion

5.5.4 Future research directions Future research using the consumption model presented in this chapter can improve upon these estimates of green crab impact by incorporating greater detail on predation rates, shellfish recruitment effects, and estimates of total biomass of shellfish in Puget Sound. The impact to each shellfish species could take into account both green crab preference for these prey species, their probability of encountering each species in the field, and density effects on predation rates. Future estimates may also benefit from considering the decrease in predation rate as total shellfish biomass in all of Puget Sound declines due to biomass loss and reduced recruitment of juvenile shellfish.

Green crab invasion densities in Puget Sound were estimated from density measurements in its native range because studies from invaded regions use catch per unit effort (CPUE) (Baeta et al., 2006; Hidalgo et al., 2006; Tremblay et al., 2006; Yamada & Gillespie, 2008) to estimate density. While CPUE is useful for comparing densities within and between regions in a single study, measurements are difficult to translate into a density of crab per area and results may vary with trap type and deployment methods used. To improve future estimates of green crab invasion, studies that sample existing populations of green crab should include crab densities measured in a metric that is repeatable across regions, such as crab density per square meter measured using dredges or quadrats.

5.5.5 Conclusions I have estimated the value-at-risk to green crab invasion into Puget Sound to describe the potential range of impacts this species may have on future shellfish harvests. At high densities or high calorie predation rates green crab may have the potential to reduce revenue from shellfish harvest and processing by as much as $1.6 - $41 million. I provided a range of potential crab invasion consumption rates to account for the uncertainty of estimating impacts of an invasive species when its impacts are not realized. Value lost to shellfish harvest is dependent on the density of crabs that invade harvest areas and the actual calories consumed by each crab, which is likely to be a product of individual rate of predation and accessibility to prey. Loss of shellfish has implications for recreational shellfish harvest and the potential for reduced filtration rates that may

101 5.5 Discussion lead to increased eutrophication in already threatened coastal habitats (Officer et al., 1982). Preventing or reducing the effect of high-impact species invasions should be a priority as these invasions have direct economic impacts and a range of indirect effects on ecosystem function (Hedgpeth, 1993; Puget Sound Partnership, 2008a). By incorporating uncertainty when estimating impacts from invasions, management plans can describe the range of potential costs of invasion and motivate preventative action in preparation for future ecosystem damage even when local impacts are still unknown (Shogren, 2000; Horan et al., 2002; Perrings, 2005; Keller et al., 2007).

102 6.1 Synopsis

Chapter 6 Knowledge gaps may impede ecosystem management: environmental impacts of a nonnative seagrass, Zostera japonica, in the northeast Pacific

6.1 Synopsis Effective management necessitates science integration, though many barriers – science communication, management processes that lack science incorporation, and lack of scientific data – may impede integration efforts. In situations where managers actively seek out scientific support, potential lack of research applicable to management objectives remains the prominent barrier. I use management of a nonnative species to explore science as a barrier to management. Management of nonnative species requires an understanding of the direction and magnitude of invasive impacts on other species’ ecosystem structures, processes, and services. However, research that occurs in isolation of management goals may not provide understanding needed by policymakers and practitioners attempting to protect native species and human activities associated with ecosystem services that are threatened by invasions. To study knowledge gaps in nonnative species research for management, I reviewed the impact of the nonnative seagrass, Zostera japonica on biotic and abiotic environments in the northeast Pacific estuaries. I found clear documentation that when Z. japonica invades, it reduces growth of species that live on unvegetated mudflats and alters the physical environment. However, my results suggest that existing studies that quantitatively test Z. japonica impacts are insufficient to comprehensively assess the effects of this invasion. I base this claim on the (1) paucity of study sites in the invaded range: fifty percent of studies were completed in only two estuaries in the northeast Pacific, (2) lack of repeated or long-term sampling: only two studies collected data across multiple years, and (3) poor coverage of several important trophic groups: few limited studies on epifauna, fish, and migrating birds. My investigation highlights the need for research objectives to determine the ecosystem role of nonnative species in their invaded range through analysis of quantitative studies across a broad geographic and temporal scale, and, in order to

103 6.2 Introduction mitigate the lack of science applicable to management, I suggest scientists conduct research that integrates management objectives.

6.2 Introduction Management of nonnative species requires an understanding of the magnitude and direction of a species’ impacts on invaded habitats. First, this allows managers to determine if the species is likely to negatively impact the economy, environment, or human health (Clinton, 1999), and second, it enables an integrative policy response based on a scientific measurement of those impacts. Because impact assessment data may evolve as managers evaluate potential damage over time (Buckley, 2008), reliable, timely, rigorously collected data are critical. Yet, information derived from studies conducted post-introduction is frequently limited in temporal, spatial, and ecological scope and so unlikely to fulfill managers’ information needs (Britton et al., 2011). Even in cases where nonnative species have been present for long periods of time, research performed in isolation of management goals is unlikely to fulfill managers’ specific needs (Neff, 2011). Here, I review the studies available on impacts of a nonnative seagrass species in the northeast Pacific Zostera japonica Aschers. & Graebner, to assess if existing knowledge can inform management of Z. japonica’s effects on the native environment.

There are seven seagrass species in the northeast Pacific (Wyllie-Echeverria & Ackerman, 2003). Of these, Z. japonica (formerly Z. nana, Z. noltii and Z. americana) was probably introduced to this region via propagules (either seeds or vegetative fragments) that were transported to two sites in Washington State (Samish Bay and Willapa Bay), USA through the oyster trade with Japan (Harrison & Bigley, 1982a). Since its introduction, Z. japonica has spread north to Campbell River, British Columbia, Canada and south to Humboldt Bay, California, USA (Figure 6.1; Shafer et al., 2008; Mach et al., 2010). While research globally and regionally has focused on a native seagrass in the genus Zostera, Z. marina, a congener in the northeast Pacific, little is understood regarding Z. japonica’s effects on the biotic and abiotic environment in its invaded range (Mach et al., 2010).

104 6.2 Introduction

From 1957, when Z. japonica was first discovered in Willapa Bay, to 2010, research in the northeast Pacific (Mach et al., 2010) was likely conducted in isolation of direct nonnative management goals. Until 2010 both Zostera species were protected in Washington, the native Z. marina and the introduced Z. japonica (Pawlak & Olson, 1995), thus the political will to fund research regarding impacts of the nonnative species was lacking. However, early in 2011 motivation to control the spread of Z. japonica was influenced by stakeholder concerns regarding the economic impacts of this plant to coastal fisheries and aquaculture (Anderson, 2011). Pressure to remove protection of the plant in Willapa Bay, an estuary important for shellfish aquaculture, was followed by the listing of Z. japonica as a Class C Noxious Weed by the end of the year, with removal permitted on commercial properties across Washington (Washington State Noxious Weed Control Board, 2011). While removal is already being attempted at its southern extent in Northern California (Schlosser, 2011), at its northern extent in British Columbia there are no specific laws regarding protection or removal.

Efforts to integrate science into management decision-making is evidenced by a recent Washington State SeaGrant workshop developed to assemble scientific expertise (Mach et al., 2010) and the written findings of the Washington State Noxious Weed Control Board (2011) that reviewed known studies of Z. japonica before listing it as a noxious weed. To assess environmental impacts of Z. japonica in the northeast Pacific I first conducted a bibliometric analysis (as in Duarte, 1999) and then summarized studies that compare the biotic and abiotic environment in Z. japonica invaded habitat with uninvaded native habitat (either unvegetated mudflat or its congener, the native Z. marina). As the motivation to manage Z. japonica stems largely from concern of fisheries stakeholders, I specifically address research on Z. japonica’s impacts to important fisheries, such as salmon and shellfish aquaculture.

105 6.2 Introduction

Figure 6.1 Map of northeast Pacific range of Zostera japonica with circles representing the total number of field studies completed at each site, and color-tone representing the decade in which those studies occurred. Open circles, Tofino and Campbell River, are the known northern extent of Z. japonica’s range in British Columbia where there have been no studies (C. Durance pers comm. Tofino, 2008; Campbell River, 2006), while Humboldt Bay represents the southern extent in California. Seven studies included research performed at multiple sites.

106 6.3 Methods

6.3 Methods 6.3.1 Searched publications Following quantitative review methods established by Duarte (1999), I searched the biological literature for studies on Z. japonica from British Columbia to northern California, its northeast Pacific introduced range (Figure 6.1; (Mach et al., 2010). I conducted literature searches in the ISI Web of Science database, Aquatic Sciences and Fisheries Abstracts, Proquest – Dissertations and Theses using all Latin binomials and common names: Zostera japonica, Zostera americana, Zostera nana, Zostera noltii, Nanozostera japonica, dwarf eelgrass, Asian eelgrass, duck grass, and Japanese eelgrass. Additional requests were made for any literature on Z. japonica research in the northeast Pacific to scientists and managers attending the “Zostera japonica Workshop” on September 23 – 24, 2010 (Mach et al., 2010). My search of the literature ended August 2011.

6.3.2 Assessed publications Each publication was reviewed to ensure the study involved research or a review of Z. japonica in the northeast Pacific and to confirm it was the most recent report or journal article published on the research. When two instances of the same study by the same author were found (e.g., a journal article and one of the following: a report, dissertation or meeting abstract), the journal article was used (in all cases this was the most recent). For dissertations, chapters not included in the journal article published from the dissertation were still included in my analysis. Data were summarized by publication type, journal, method, research focus, and year.

Studies were then categorized by methodology (i.e., laboratory study, field study, review, and biophysical modeling) and the type of research conducted (observation, experiment, or both where applicable). Location, length of study, and study focus were also recorded. Study foci were classified into the following categories: distribution and abundance; the impact of the abiotic and biotic environment and pulse and press stressors (e.g., Collins et al., 2011) on Z. japonica; the impact of Z. japonica on biotic (species Z. japonica interacts with in the northeast Pacific) and abiotic (physical factors; such as nutrients,

107 6.3 Methods sediment, etc.) environments in the northeast Pacific; and the effect of resource management programs on Z. japonica presence and distribution (further definitions in Table 6.1). Because a particular study could have more than one focus or location, a database was developed to analyze data from the studies.

The number of studies conducted at each study location (14 sites) during the previous 4 decades, 1972 – 2011, were compared using two-way ANOVA (JMP 7.0, 2007) after data were confirmed to meet assumptions of the test. When a significant interaction between site and decade was detected, multiple comparisons of means among decades were conducted for each site separately by Tukey’s test, using the mean square of residuals, followed by multiple comparisons of means for the site over the entire 40-year period.

6.3.3 Analysis of the effects of Zostera japonica on the biotic and abiotic environments To summarize the effect of Z. japonica on the biotic and abiotic environments in its introduced range, I assessed all observational studies (comparing non-manipulated invaded and uninvaded sites) and experiments. My selection criteria required studies to quantitatively compare a species, community, or abiotic variable in both invaded Z. japonica and control plots, either Z. marina or unvegetated mudflat. When a response variable was measured at different time intervals (e.g., across seasons or years) I used the final measured response in my summary.

Of the observational studies that fulfill these requirements, none compared the studied habitat before and after invasion; all compared invaded areas relative to uninvaded areas. This sampling design implies limitations on the interpretation of Z. japonica’s impacts because of the possibility that invaded and uninvaded habitats differed in environmental characteristics other than the presence of Z. japonica (as described in Vila et al., 2011).

108 6.3 Methods

Quantitative studies (experimental and observational) designed to determine potential biotic impacts measured 11 different response variables (e.g. growth rate, abundance, recruitment, etc.) across 11 species in 12 studies (one study had both an experiment and an observation; Table 6.2), and studies of abiotic impacts compared 35 different abiotic response variables in 7 studies (Table 6.3). The lack of repeated testing of specific species or abiotic variables prevented a statistically meaningful analysis of measured effect size (Gates, 2002). Therefore, I conducted a simplified assessment of impact to the response variables studied across the 11 species identified in Table 6.2 and the 35 abiotic environmental variables identified in Table 6.3, as follows.

For biotic interactions, if a study examined more than one response variable for a single species within the same experiment, such as leaf growth or biomass, each variable was considered separately, as they represent different ecological impacts. The change in each response variable was scored: 1 = statistically significant increase, 0 = no statistically significant effect, or -1 = statistically significant decrease. Scores were then averaged for the species within an individual experiment or observation and this final score was then considered a single ‘replicate’ representing the entire experiment to prevent pseudoreplication of sampled response variables (see Hurlbert, 1984). Replicates, therefore, are the number of independent tests of each interacting species or community. There were no cases where I averaged across positive and negative scores, thus the final score represents an absolute impact. When community composition was recorded as a significant change, but there was no reference corresponding increase or decrease in species richness, abundance or biomass, this interaction was coded as a ‘significant change’. I executed this scoring technique for each response variable measured in all 12 studies (Table 6.2) For each species or community that was replicated in two or more tests, the score of each replicate was averaged across all studies testing the effect of Z. japonica on that species. If a significant change was recorded, it was noted but not included in this averaging.

To score abiotic studies I coded each impact to a single abiotic variable, such as a change in nutrient concentration or rate of production, as: 1 = statistically significant increase, 0

109 6.4 Results

= no statistically significant effect, or -1 = statistically significant decrease, in native habitat (unvegetated mudflat or one populated with Z. marina) verses Z. japonica invaded habitat. Each score was considered a single ‘replicate’ -- the number of independent tests of the change in abiotic response variable. Using this technique I scored each response variable measured in all 7 studies displayed in Table 6.3. Scores were then averaged across variables tested in more than one experiment or observation.

6.4 Results Most of the 80 studies focused on Z. japonica in the northeast Pacific found in this review were published in peer-reviewed journals (64%; 51 articles). Other publication outlets included state (or provincial) and federal government reports (16), graduate theses (6), books (4), and meeting abstracts (3). Of the 51 peer-reviewed articles, 37% were originally published in a different format including government reports (8), graduate theses (6), and meeting abstracts (5). Research on Z. japonica in the northeast Pacific has been published in 24 journals; a quarter of published papers (13) appeared in Aquatic Botany.

6.4.1 Space and time distribution The first published record of Z. japonica presence in the northeast Pacific appeared in Hitchcock et al. (1969) and consisted of a reference to an herbarium specimen (WTU- 208020), archived at the University of Washington Herbarium (WTU), collected in Willapa Bay, Washington in 1957. From then, the number of publication doubled each decade until 2000 when publications quadrupled (60% of pubs). The number of studies for site location and decade, and the interaction between site and decade were significant

(two-way ANOVA, F3,13 = 2.34, P ≤ 0.0001). Although the regional distribution of Z. japonica ranges from Southern British Columbia to Northern California, the majority of the 56 field studies were completed in two of the 14 studied estuaries: Padilla Bay, WA (15), and Willapa Bay, WA (15), posthoc tests showed these two sites were significantly more studied than the 7 least studied estuaries, for the previous 40 years of research, 1972-2011 (Q = 3.370 P ≤ 0.05). Each decade (except the 70s) had a specific estuary or estuaries in which there averaged significantly more studies: in the 80s Roberts Bank,

110 6.4 Results

BC, in the 90s Padilla Bay, WA, and in the 2000s Willapa Bay, WA, Padilla Bay, WA and Yaquina Bay, OR averaged more studies than other sites in their respective decade (Q = 4.072 P ≤ 0.05; Figure 6.2 ).

Figure 6.2 The mean number of studies per year (SE) conducted at each estuary in which Zostera japonica was studied each decade (a. 1972-81; b.1982-91; c. 1992-2001; d. 2002-11) over the previous 40 years. Sites in order of latitude from left to right. Sites with an asterisk (*) were significantly different from all other sites (but not one another) as detected by Tukey post-hoc comparisons

111 6.4 Results

6.4.2 Study foci Study foci included distribution and abundance (48), changes to the biotic environment (26), impacts of the biotic environment on distribution and/or abundance (18), changes to the abiotic environment resulting from Z. japonica invasion (16), physiological (15) or life history (12) responses to in situ or manipulated environments, effect of policy on distribution and abundance (4), and response to environmental stressors (2). Research papers (61 articles) reported either observational (59%) or experimental (33%) results but rarely both in the same study (7%).

Of the 20 papers with experimental studies testing changes resulting to or from Z. japonica, 13 were focused on biotic interactions and 12 on abiotic variables; five of these papers included both biotic and abiotic studies. Study foci, such as the number of studies of impacts to or from Z. japonica, are given in Table 6.1. There were 11 studies focused only on the distribution and abundance of Z. japonica populations (all observations), while most distribution and abundance data (28 studies) resulted from before and after measurements taken for experiments and observations of other study foci. Four studies experimentally tested the impacts of management control efforts, such as herbicides, light deprivation and removal, on Z. japonica presence and abundance. Four biophysical models were published: three based on data from field studies and one reviewed the literature to build a seagrass ecosystem model. Research effort was limited geographically (primarily two locations; Padilla and Willapa Bay) and temporally (Figure 6.2 ). There were no comprehensive peer-reviewed reviews of the impacts of Z. japonica in the northeast Pacific.

112 6.4 Results

Table 6.1 The eight categories on which research on Z. japonica has focused, the number of field and lab studies that fall into these categories and the percentage of these that were experiments (% Exp), and the number of papers that review some aspect of Z. japonica research but are not primary studies. 33 of 80 studies fall into more than one research focus category.

Research Focus Definition Studies (% Exp) Review Distribution & Studies describing the spatial extent of Z. 39 9 Abundance japonica both along the eastern Pacific coast or within an estuary Impact of the environment on Zostera japonica Biotic The impact of other species on Z. japonica 12 (75%) 3 Life History Changes to the population (regional and local) 14 (14%) 1 status, including but not limited to shoot growth, seed development, flowering due to the abiotic environment Physiology Z. japonica’s physiological response to the 10 (60%) 3 abiotic environment Management The effect of resource management programs 4 (100%) 6 on Z. japonica presence and distribution Stressors Studies that observed or tested the impact of a 2 (100%) 2 pulse (short-term disturbance) or press (long- a term gradual change) stress. Zostera japonica impact on the environment Biotic The impact of Z. japonica on other species 19 (42%) 7 Abiotic Change in the abiotic environment caused by Z. 11 (27%) 4 japonica a As in Collins et al. 2011

6.4.3 Analysis of biotic impacts Within the nine studies experimentally testing the impact of Z. japonica on other biota, were 23 tests of its effect on a single species or multi-species community (Table 6.2). Four observational studies statistically compared biotic response variables, such as growth and abundance, in areas with and without Z. japonica. Of the experimental and observational studies, six included measures of multiple response variables of a single species. All other tests measured a single response variable. Over all measured response variables in all studies (29), 14 demonstrated a significant reduction, 8 determined a neutral impact (no significant change), and four found a significant increase (Table 6.2). Three studies on species assemblages report significant changes of individual species with no reference to increase or decrease in species abundance, diversity or richness. Length of the studies ranged from one week to 15 months.

113 6.4 Results

Table 6.2 A summary of experimental and observational studies testing the community effects of Z. japonica on native invertebrate and plant species. The table lists the publication reference, publication type, study location, study method (experiment (Exp) or observation (Obs)), the number of independently replicated tests within a study (Ind. Test; each number (1, 2, 3) represents the number of independent experiments or observations that were conducted within a publication), the interacting species or community whose response variable was measured, measured response variable, change in response variable, the length of time the experiment or observation was conducted before the variables were measured or the number of times the variable was measured (ex, 7x = 7 times), and start date when available. All studies on invertebrates compared the mudflat with Z. japonica to unvegetated mudflat unless otherwise marked, while all studies on Z. marina compared ‘Z. marina bed mixed with Z. japonica’ to ‘Z. marina bed monoculture.’ The change to the response variable is depicted as: ‘+' = significant increase in the response variable, '0' = no significant change in response variable, '-' = significant decrease in the response variable, and ‘△ ’ = a significant change in the community when the paper did not describe whether there was an increase or decrease.

Ind. Reference Type Site Method Interacting species Response Variable Change Timeline Test Harrison (1987) J RB 1 Exp N. californiensis Number of burrows - 23 wks 2 Exp N. californiensis Number of burrows - 8 wks 3 Exp P. gracilis Abundance - 12 wks

Posey (1988) J CB 1 Exp Infaunal community Richness + 9 mnths 2 Obs Infaunal community Richness + 2 yrs, 9x Infaunal community Abundance +/0 2 yrs, 9x

Nomme and Harrison (1991) J RB 1 Obs Z. marina Shoot density 0 5 mnths, 7x

Merrill (1995) R PB 1 Exp Z. marina Shoot density - 8 wks Z. marina Leaf growth (cm) - 8 wks

Thom et al. (1995) J PB 1 Obs Idotea, Lacuna, Caprellid Abundance, Zj vs Zm 0 9 mnths, 7x

Publication type: J = Journal, R= Report, T = Thesis; Site: RB = Roberts Bank, BC, CB = Coos Bay, OR, PB = Padilla Bay, WA, WB = Willapa Bay, WA, NB = Netarts Bay, OR; Interacting species: Neotrypaea californiensis, Praxillella gracilis, Zostera marina, Haminoea vesicula, Spartina alterniflora, Oncorhynchus tshawytscha, Ruditapes philippinarum; Zj vs Zm = Compared the measured factor of the interacting species in Z. japonica verses Z. marina; Zj vs C. gigas = Compared the measured factor of the interacting species in Z. japonica verses a C. gigas oyster flat

114 6.4 Results

Table 6.2 continued Ind. Reference Type Site Method Interacting species Response Variable Change Timeline Test Hahn (2003b) T PB 1 Exp Z. marina Shoot density - ~15 mnths Z. marina Leaf growth (cm) - ~15 mnths Z. marina Biomass - ~15 mnths 2 Exp H. vesicula Number of egg masses, Zj vs Zm - ~15 mnths 3 Exp Infaunal community Species assemblage, Zj vs Zm ! ~15 mnths Infaunal community Abun/Rich/Div, Zj vs Zm 0 ~15 mnths

Hahn (2003a) J PB 1 Exp Micobial community Species assemblage, Zj vs Zm ! 4 wks Micobial community Abun, Zj vs Zm 0 4 wks

Bando (2005) T WB 1 Exp S. alterniflora Seedling condition - 3 mnths 2 Exp S. alterniflora Above ground biomass - 4 mnths

Bando (2006) J WB 1 Exp Z. marina Dry weight biomass - 47 d 2 Exp Z. marina Leaf growth 0 17 d

Berkenbusch et al. (2007) J NB 1 Exp N. californiensis Number of burrows - 20 wks 2 Exp Infaunal community Dissimilarity factor, Zj vs Zm ! 20 wks

Semmens (2008) J WB 1 Obs O. tshawytscha Swimming Speed, Zj vs Zm - 1 wk 2 Obs O. tshawytscha Swimming Speed, Zj vs C. gigas 0 1 wk

Tsai et al. (2010) J WB 1 Exp R. philippinarum Meat weight (clam condition) - 2 mnths R. philippinarum Shell growth 0 2 mnths 2 Exp R. philippinarum Recruitment 0 4 mnths 3 Exp "All clams" Recruitment + 4 mnths

115 6.4 Results

Only two species, Z. marina, and the ghost shrimp Neotrypaea californiensis, were tested in more than two studies. The impact of Z. japonica on Z. marina was the most commonly tested community interaction (5 of 23 tests). In three studies Z. marina growth, biomass, and shoot density were significantly reduced in the presence of Z. japonica, while two found Z. japonica had no impact on shoot density and leaf growth. Studies of the interactions between Z. japonica and N. californiensis, the second most common species studied (3 of 23 tests), showed that the number of burrows in a given area were significantly reduced in the presence of Z. japonica. Z. japonica was also found to reduce growth of the invasive shoregrass Spartina alterniflora (2 of 23 tests). The effect of Z. japonica on the ‘infaunal community’ was measured in four studies with differing results: two tests demonstrated an increase in richness and abundance while the third had no effect. The fourth test was measured as a significant “change” in community composition and was not included in analyses. The effect of Z. japonica on the clam Ruditapes philippinarum was measured in two experiments (2 of 23 tests) as a reduction in clam soft-tissue but not clam weight and the second demonstrated no effect on clam recruitment. No other tests of Z. japonica’s impact on an interacting species or community were repeated (7 of 23 tests; Table 6.2).

Mean estimates of measured effects on response variables were made in all cases where there was more than one test of an interacting species. In all other cases, the estimated effect was based on the only available test for that interacting species. All studies of individual infaunal invertebrates species compared unvegetated mudflat to Z. japonica invaded mudflat (4 species), while all other studies measured the difference in the response variable of the interacting species in Z. marina versus Z. japonica.

Mean estimates demonstrated that Z. japonica generally reduced growth and biomass of other marine plants, Z. marina (-0.6 from 5 tests) and Sparitina alterniflora (-1 from 2 tests; Figure 6.3 a). In studies comparing Z. japonica and Z. marina, the nonnative seagrass had significantly fewer epifaunal egg masses of Haminoea vesicula (-1 from 1

116 6.4 Results

test) and Chinook salmon (Oncorhynchus tshawytscha) swimming speed was significantly slower (-1 from 1 test) than through Z. marina canopy, but otherwise had a neutral effect (0) on the microbial community (1 test), epifaunal grazers (1 test), and infaunal community (1 test). In studies comparing Z. japonica to unvegetated mudflat (Figure 6.3 b), the nonnative seagrass significantly reduced the response measured for individual infaunal invertebrates: N. californiensis (-1 from 3 tests), the polychaete Praxillella gracilis (-1 from 1 test), and R. philippinarum (-0.33 from 1 test). However, in a study that did not differentiate clam species, the presence of Z. japonica did not reduce clam abundance (+1 from 1 test), and in another the infaunal community response variables were significantly increased (+1 from 2 tests). There was no significant difference (1 test) in Chinook salmon swimming speed between Z. japonica and unvegetated oyster flats.

Figure 6.3 Mean effect of Zostera japonica on the measured response variables of interacting species when compared to growth in or of (a) marine plants: Z. marina and Spartina alterniflora, or (b) unvegetated and Crassostrea gigas oyster mudflats (data from Table 6.2). Effect score is measured as: 1 = significant increase, 0 = no significant change, -1 = significant decrease. Numbers to the right of the figure represent the number of replicated tests of each variable. Infaunal and microbial communities that recorded only a ‘significant change’ were not included in this summary. Standard error was included for those species tested three or more times (N. californiensis, error bar = 0). Common name: Eelgrass, Z. marina; Chinook Salmon, Oncorhynchus tshawytscha; Bubble Snail, Haminaea vesicular; Smooth Cordgrass, S. alterniflora; Manila Clam, Ruditapes philippinarum; Ghost Shrimp, Neotrypaea californiensis; polychaete, Praxillella gracilis

117 6.4 Results

6.4.4 Analysis of abiotic impacts Of the 11 studies evaluating Z. japonica’s impact on the abiotic environment, seven studies tested these changes through five observational and four experimental field studies, and the remaining four studies lacked statistical analysis of Z. japonica’s impact on the abiotic environment and thus were not included Table 6.3. All included studies measured more than one abiotic variable, testing the difference between the environment on unvegetated mudflats (23 variables) or within Z. marina (13 variables), to invaded Z. japonica habitat. Length of the studies ranged from three months to two years.

Abiotic response variables differed significantly in native seagrass habitat, Z. marina, and invaded Z. japonica habitat for 6 of the 13 tested variables Table 6.3. There were only two studies comparing these habitats and no response variables were repeated across these studies. Zostera japonica had no effect on porewater nutrients ([dissolved reactive phosphate], [nitrite], [nitrate]), sediment particulate organic nutrients (carbon, nitrogen, phosphorous; % dry weight), or nitrate fluctuation, though Z. japonica had less ammonium and phosphate fluctuation than the native Z. marina. There was a significant increase in porewater ammonium and sediment grain size and heterogeneity when present, and decomposition rate of these plants (blade mass/day) was significantly faster than Z. marina (Figure 6.4a).

118 6.4 Results

Table 6.3 Summary of experimental and observational studies testing the change Z. japonica has on abiotic environment. The table shows the publication reference, publication type, study location, number of independently replicated tests within a study (Ind. Test; each number (1, 2, 3) represents the number of independent experiments or observations that were conducted within a publication), study method (experiment (Exp) or observation (Obs)), habitat type to which Z. japonica is compared (Habitat), measured abiotic variable, change in the abiotic variable, length of time the experiment or observation was conducted during which the variables were measured or the number of times the variable was measured (ex, 7x = 7 times), and the start date. The change to the abiotic variable is depicted as: ‘+' = significant increase, '0' = no significant change, '-' = significant decrease. Reference Type Site Ind. Test Method Habitat Abiotic Variable Change Timeline Hahn (2003a) J Padilla Bay, WA 1 Exp Z. marina Decomposition rate + 27-39 days

Larned (2003) J Yaquina Bay, OR 1 Obs Unveg [DRP] flux 0 1 year, 2x [NH4] flux + [NO3] flux + Pore [DRP] 0 Pore [NH4] - Pore [NO2] - Pore [NO3] - Sed [Chla] - Sed grain size 0 Sed heterogeneity 0 Sed POC 0 Sed PON 0 Sed PP 0

2 Obs Unveg [DRP] flux 0 4 months, 1x [NH4] flux + [NO3] flux +

3 Obs Z. marina [DRP] flux - 5 months, 1x [NH4] flux - [NO3] flux 0 Pore [DRP] 0 Pore [NH4] + Pore [NO2] 0 Pore [NO3] 0

119 6.4 Results

Table 6.3 continued

Reference Type Site Ind. Test Method Habitat Abiotic Variable Change Timeline Larned (2003) continued Sed grain size + Sed heterogeneity + Sed POC 0 Sed PON 0 Sed PP 0

Posey (1988) J Coos Bay, OR 1 Obs Unveg Sed grain size - 2 years, 9x Sed volatile organics +

Tsai et al. (2010) J Willapa Bay, WA 1 Exp Unveg Pore [NH4] 0 3 months, 1x Pore [NO2] 0 Pore [NO3] 0 Pore [PO4] 0 Water flow - 2 months, 2x

Turnbull (2009; ch.3) T Coos Bay, OR 1 Exp Unveg [NO3] flux 0 10 months, 4x Sed [NH4] 0/- Sed [NO3] 0 Sed CO2 resp. rates 0 Sed grain size 0 Sed pH 0 Sed temp 0 Sed rate - 2 years

Turnbull (2009; ch.4) T Coos Bay, OR 1 Obs Unveg Sed CO2 resp, rates + 10 months, 4x Sed POC 0 Sed PON 0 Sed grain size 0 Sed moisture 0 Sed pH 0 Sed rate 0 2 years

Publication type: J = Journal, T = Thesis; DRP = Dissolved Reactive Phosphorous; NO2 = Nitrite; NO3 = Nitrate; NH4 = Ammonium; PO4 = Orthophosphate; if more than one value given as a change, both results found; POC = Particulate Organic Carbon; PON = Particulate Organic Nitrogen; PP = Particulate Phosphorous; Water flow = dissolution block dissolve rate; Sed = Sediment; Pore = Porewater; resp. = respiration

120 6.4 Results

Twenty-three different response variables were tested to compare the change in the abiotic environmental characteristics between Z. japonica and unvegetated mudflat Table 6.3. Only sediment grain size was tested in more than two studies, 9 variables were measured in two studies and the remaining 13 variables were only tested one time. Of these, 11 were found to have no significant difference between the two habitats. Z. japonica did not cause a change in particular sediment characteristics (particulate organic nutrients (carbon, nitrogen, phosphorous; % dry weight), nitrate, pH, temperature, moisture and heterogeneity), porewater phosphate, or phosphate nutrient fluctuation, though the plant did increase ammonium and nitrate fluctuation, sediment volatile organics and sediment CO2 respiration rate. Z. japonica beds cause a reduction in other sediment characteristics ([Chl a], ammonium, sedimentation rate, and grain size), porewater nutrients ([dissolved reactive phosphate], [nitrite], [nitrate]), and water flow measured using a dissolution block (Figure 6.4b).

Figure 6.4 Abiotic response variables found to change significantly (reduction or increase) in Zostera japonica invaded habitat verses native habitat: (a) Z. marina bed (6 of 13 had a significantly change) or (b) unvegetated mudflat (12 of 23 had a significantly change; data from Table 6.3). Numbers to the right of the figure represent the number of replicated tests of each variable. Standard error was included for the variable tested three or more times.

121 6.5 Discussion

6.5 Discussion The functions of Z. japonica in intertidal mudflats include some of those for which native seagrasses are protected (Hemminga & Duarte, 2000). However, Z. japonica is also moving into unvegetated mudflats, a habitat that serves important ecosystem functions and is also the valued location for shellfish aquaculture. Here I reviewed the temporal and spatial limitations of Z. japonica research and the impacts of Z. japonica on the biotic and abiotic environments to discuss changes it causes to northeast Pacific ecosystems and describe information gaps in current research.

6.5.1 Logistical limitations Prediction of future expansion is difficult because research on Z. japonica’s past expansion through the northeast Pacific has been geographically and temporally limited. More than fifty percent of all field research in the invaded range of Z. japonica has been conducted in two estuaries in Washington State (Figure 6.1). The influence of commercial aquaculture in Willapa Bay and the added attention of the National Estuarine Research Reserve in Padilla Bay are likely causes for the focus on these two estuaries (Bulthuis, 1995; Mach et al., 2010). While these data contribute to an understanding of Z. japonica’s effect on the native environment, these two large flat estuaries are relatively unique in the northeast Pacific where much of the invaded habitat is fringing the intertidal edges of channels (Willapa Bay, Hazen, 1996; Puget Sound, Berry et al., 2003). Combined with the fact that few studies have compared or contrasted research findings between estuaries, the ability to generalize results throughout the entire northeast Pacific and elsewhere on the Pacific Coast of North America is limited.

While studies most commonly took place in Padilla Bay and Willapa Bay, even in these estuaries research has not been conducted with temporal consistency over the last 40 years (Figure 6.1). From 1982 to 1991 research on Z. japonica was most common in Roberts Bank, BC followed by a peak in Padilla Bay research from 1992 to 2001. From 2002 to 2011 research continued to be common in Padilla Bay and peaked in Willapa Bay and Yaquina Bay (Figure 6.2 ). This temporal irregularity suggests that when

122 6.5 Discussion considering research on Z. japonica, results may be restricted to the time period in which they were collected and may not apply across all four decades.

Research has occurred in 14 estuaries where Z. japonica is present, but only one study in Washington State (Gaeckle et al., 2009) has attempted to quantify where it is not yet located. Understanding where Z. japonica is absent is necessary for predicting the rate of spread and future habitat preferences in the coastal estuaries. Finally, of the 15 quantitative experiments and observations included in this review, only two collected data across multiple years, limiting the ability to make long-term predictions. Even where recent floral surveys exist, sampling methods and species interactions tested usually differ among sites and over time.

6.5.2 Changes to populations of mudflat species In many estuaries where Z. japonica is present, the plants have invaded an upper intertidal band of unvegetated mudflat (Shafer, 2007). Introduced macrophytes can often have negative competitive effects on infaunal species competing for the same space, as well as epifauna water column dwellers, in comparison to native macrophytes (Thomsen et al., 2009). In contrast, other studies of nonnative macrophytes document higher diversity and abundance of associated organisms relative to unvegetated areas (Neira et al., 2005; McKinnon et al., 2009).

In my review I learned that Z. japonica invading unvegetated mudflats both increased and decreased the abundance, biomass, and growth of infaunal invertebrates (Figure 6.3 ). Native species N. californiensis (burrowing shrimp) and P. gracilis (polychaete) competed for space with the below-ground architecture of roots and rhizomes (Harrison, 1987; Berkenbusch et al., 2007), while the whole infaunal community increased in diversity and abundance (Posey, 1988). Similarly, the cultured Manila clam, R. philippinarum, had reduced meat weight in one measurement, while others did not find an effect on shell growth or recruitment (Tsai et al., 2010). In addition, this study found an increase in the overall recruitment of clams in areas with Z. japonica. There was no preference detected in Chinook salmon swimming speeds between Z. japonica and

123 6.5 Discussion unvegetated oyster flats (Semmens, 2008). This was the only northeast Pacific study to compare megafauna in unvegetated habitats to those invaded by Z. japonica. In addition, no studies compared epifaunal species use between these habitats, though in its native range epifaunal species richness and abundance have been found to be higher in Z. japonica beds than in adjacent unvegetated areas (Lee et al., 2001). This uneven coverage of taxonomic groups has the potential to suggest overall negative results on other species where the story may be considerably more complex when considering the whole biotic community. Some unvegetated mudflat fauna are likely to compete for space with roots and rhizomes of this seagrass, but other species may benefit from the cover and leaf area in the water column associated with its presence.

Though the low intertidal to high subtidal range of Z. marina and the high intertidal range of Z. japonica often limits direct competition, when these species do compete for space both species experience a reduction in shoot density, leaf growth, and biomass (Hahn, 2003b; Bando, 2006). Competition was also found to reduce seedling condition and biomass in a nonnative saltmarsh grass, S. alterniflora, a potentially beneficial effect from a management perspective, as eradication efforts of S. alterniflora are ongoing in the northeast Pacific (Taylor & Hastings, 2004).

While habitat structure provided by intertidal seagrass species is functionally similar (Hemminga & Duarte, 2000), organisms that utilize habitat provided by Z. marina do not necessarily populate Z. japonica with the same abundance or diversity. The same study testing Chinook salmon swimming speed in unvegetated habitats also found that while juvenile Chinook salmon did not prefer or avoid Z. japonica, they slow their swimming speed in Z. marina (Semmens, 2008). It is possible that the juvenile fish from this study (mean length: 115 mm) were too large to effectively use Z. japonica with its smaller vertical structure, as these fish are larger than the migrating fry (entering estuaries at 65 - 75 cm) that use these estuaries for juvenile rearing (Healey, 1991). Thom et al. (1995) found epibenthic invertebrates important to the diets of economically and ecologically important fish species, such as surf smelt (Hypomesus pretiosus), Pacific herring (Clupea pallasi), Pacific sand lance (Ammodytes hexapterus), and Chinook salmon (O.

124 6.5 Discussion tshawytscha), occur in Z. japonica and Z. marina with equal abundance and diversity. This suggests Z. japonica may have an indirect benefit for these fish species when occupying unvegetated mudflats where Z. marina is absent, though this is unclear because there are no studies comparing epifaunal species in invaded verses uninvaded unvegetated mudflats. Studies on infaunal communities found that Z. japonica’s effect on the abundance and assemblage of the community depended on the study (Hahn, 2003a; Berkenbusch et al., 2007), though no change was found in the abundance of the microbial community between Z. marina and Z. japonica habitat (Hahn, 2003b).

Zostera japonica’s impacts on the mudflat community cannot be characterized by a single aggregate measure, as some infaunal species derive a positive effect from food and habitat resources, while other species are negatively impacted in terms of population density or performance, and some have no response. In addition, native habitat type appears to influence what types of species comparisons are selected by investigators for study. When studying the invasion of Z. japonica into Z. marina, research focused on epifaunal and plant species, but when studying invasion into unvegetated mudflat, the focus was almost entirely on infaunal species and community composition. Consequently, assessment of the overall impact of Z. japonica is not possible with present data, as the ‘overall’ effect has not been studied. The increase in habitat and nutrients Z. japonica provides may increase ecosystem benefits, as in Boundary Bay, BC, where Baldwin and Lovvorn (1994) found up to 84% of migratory waterfowl diet consisted of Z. japonica. In Padilla Bay, WA, Dinnel et al. (1986) found more Dungeness crab young-of-the-year in Z. japonica stands. Other nonnative plant species, such as Phragmites australis (Hershner & Havens, 2008) and Tamarix spp. (Paxton et al., 2011) have demonstrated increased provision of some ecosystem services in their invaded ranges. Alternatively Z. japonica may in fact reduce shellfish production and increase harvesting difficulty and costs (Anderson, 2011), a common result of plant invasions in Willapa Bay (Taylor & Hastings, 2004).

125 6.5 Discussion

6.5.3 Changes to the abiotic environment As with other seagrass species, Z. japonica is both known and expected to alter abiotic properties of intertidal flats, particularly in comparison to unvegetated areas (Hemminga & Duarte, 2000). Based on the substrate, structure, and biogeochemical cycling that Z. japonica provides, it functions similarly to Z. marina in estuarine environments, though at different intertidal elevations (e.g. Fonseca et al., 1982; Pellikaan & Nienhuis, 1988). However, only two studies have statistically compared Z. japonica to Z. marina in the northeast Pacific (Hahn, 2003b; Larned, 2003). Interestingly, the rates of nutrient removal from the water column for Z. marina are significantly higher than Z. japonica (Larned, 2003), while the decomposition rate of Z. japonica (1.65% of total mass lost per day) is significantly faster than Z. marina (1.35% of total mass lost per day) (Hahn, 2003b). This pattern was also found in terrestrial plants (Grout et al., 1997). This suggests that nonnative seagrass beds in the northeast Pacific are both producing more particulate and dissolved organic matter (POM; DOM) as well as removing fewer nutrients from the water column. These additional nutrients are then available for consumption by organisms such as filter feeders and grazers, however the effect of this additional material in the food chain has not been quantified or modeled. These differences could lead to differing rates of decomposition and nutrient retention as well as alter the interaction between microbes and intertidal vegetation that in turn, affect higher trophic levels. This cascading effect (Carpenter et al., 1985) has the potential to enrich estuarine productivity and biodiversity in the northeast Pacific, though it is important to note that many abiotic environmental variables, such as porewater nutrient concentrations and sediment particulate organic nutrients (PON) do not differ between these two environments (Table 6.3).

Nutrient uptake within a bed of Z. japonica has been found to be higher than on unvegetated mudflats, so this plant has the potential to limit nitrogen levels in estuaries that are already nitrogen-limited (Larned, 2003), though this relationship can vary by estuary and season (Turnbull, 2009). For example, Tsai et al. (2010) showed that there were no differences in ammonium levels in porewater where Z. japonica was present vs. removed, and Kaldy (2006) showed extremely high levels of porewater ammonium in

126 6.6 Gaps in research for management goals sediments occupied by Z. japonica, although a comparison to unvegetated mudflat or Z. marina was not done. Other studies comparing PON did not find any difference between these habitats (Larned, 2003; Turnbull, 2009). Conflicting evidence of nutrient use by Z. japonica makes a discussion about potential effects on nitrogen cycling difficult. The vertical canopy structure resulting from Z. japonica invasion alters an unvegetated mudflat and slows water flow by up to 40% (Tsai et al., 2010). Though these physical changes did not manifest in observed increases in sedimentation rates or effects on sediment pH or temperature (Turnbull, 2009). Changes in water flow can affect how organisms use mudflats, where species requiring high water flow for respiration or filter feeding may be negatively impacted by Z. japonica.

Whether the ecosystem impacts of Z. japonica are positive or negative can only be determined by the ecological and societal priorities of managers and stakeholders in the estuaries it invades. Examples of nonnative plants establishing and providing important structure and function to their invaded range are becoming more common; in some cases these nonnative plants are more capable of dealing with the increased stressors in anthropogenically impacted ecosystems (Hershner & Havens, 2008; Schlaepfer et al., 2011). Alternatively, the altered habitat and increased nutrient cycling in estuaries already nutrient limited could have detrimental impacts on other species using these mudflats (Kaldy, 2006; Bolton & Brooks, 2010). As many countries have management policies that require the protection of native species from nonnative impacts based on national, regional and international agreements, even if nonnatives have a positive impact, obligations may require management action that does not consider a nonnative’s impact on ecosystems, as regulation at the national or higher scale will not necessarily consider local processes.

6.6 Gaps in research for management goals Zostera japonica has been shown to alter the intertidal mudflats and native seagrass beds it invades. When compared to other exotic species, the body of research could be considered quite extensive, with 57 field studies across its invaded range. Yet there have been an insufficient number of quantitative studies to conclusively target Z. japonica as

127 6.6 Gaps in research for management goals the cause of ecological change, especially in the case of fisheries species. For example, research has only focused on Z. japonica impacts to three species of economic importance: Manila clam, Chinook salmon, and Dungeness crab. Of these, only one study demonstrated a potential impact (Tsai et al., 2010; Manila Clam), otherwise studies have not determined any effect. This finding does not mean Z. japonica has no impact, as these three isolated studies are temporally and spatially limited.

With so few studies on Z. japonica, it is impossible to disentangle how its environmental impacts vary across ecological contexts, including species diversity and composition, physical disturbance, and resource availability. The magnitude of Z. japonica’s impacts are likely to be context-dependent, as with other ecological effects (Salomon et al., 2010). Furthermore, an apparent negative (or positive) impact according to the few studied metrics (e.g., nutrient fluctuation or species richness) could be offset by positive (or negative) impacts that have not been considered at this time. Finally, with Z. japonica studied primarily in two estuaries in the northeast Pacific and 60% of studies occurring since 2000, it is not clear how its impacts to native species and habitats might change across this range over time. In addition, research on Z. japonica has been largely observational, with few experimental and synthetic studies. This pattern of investigation is common for seagrass species and is at odds with the general tendency towards experimentally driven research in the field of ecology (Duarte, 1999). Observational studies can be effective for establishing a baseline and documenting potential impacts in invaded regions, but they cannot specifically demonstrate a causal role for the nonnative. Furthermore, only four of seven observational studies of Z. japonica impacts established a difference statistically; the remaining three only discussed the changes qualitatively. Consequently my ability to assess changes Z. japonica has in habitats it invades is limited, further restricting predictions of the impact of further range expansion.

6.6.1 Conclusions My study of existing research on Z. japonica demonstrates that in the absence of management or policy guidance, research is of some, but limited, utility in assessing the

128 6.6 Gaps in research for management goals impact of Z. japonica in its invaded range. The sum of the individual research projects is insufficiently comprehensive and conclusive to address emerging and unforeseen management objectives, such as the decision to list Z. japonica as a noxious weed at sites in Washington (those where its presence is in conflict with shellfish aquaculture). These information gaps hamper effective policy decisions and limit the ability to anticipate future expansion and its effects (Ruiz et al., 2000a; Ruiz & Hewitt, 2002). Understanding the environmental implications of Z. japonica’s invasion is necessary if control measures are to be effective (Williams & Grosholz, 2008). A conceptual framework of information necessary for policy makers and managers regarding nonnative species management may help guide future research development (Groffman et al., 2006; Cheong, 2008).

Once the effect of a nonnative species is understood, a risk-assessment to evaluate whether the environmental changes that may arise from the invasion are positive or negative may help management efforts. In the case of Z. japonica, increasing seagrass area in the northeast Pacific may be viewed as a positive benefit in light of current net loss of native Z. marina (Waycott et al., 2009) by replacing some of the habitat and nutrient cycling functions of the native eelgrass. However, other stakeholders whose clam and oyster aquaculture harvests may be impaired by Z. japonica may view the nonnative as a negative impact (Washington State Noxious Weed Control Board, 2011). Evaluating the value gained or lost to Z. japonica may assist managers in maximizing total ecosystem value while protecting ecosystem function.

129 7.1 Knowledge-base limitations for marine management

Chapter 7 Conclusions

The research completed during my dissertation contributes to an improved understanding of the utility and comprehensiveness of research for marine ecosystem-based management (EBM). In this conclusion I present a summary of my research: contributions and implications for future research.

7.1 Knowledge-base limitations for marine management I achieved my objective, to investigate the capacity of existing scientific knowledge on anthropogenic impacts to inform coastal EBM, in Chapters 2 and 6. I accomplished this by developing a framework in Chapter 2 that simplified ecosystem impact-pathways, impacts from anthropogenic activities connect to changes in ecosystem service providers and their services. By an analysis of the literature, I found there is very little research connecting complete impact-pathways, and of those papers that do form impact-pathways none measure impacts on endpoint values. This finding has implications for evaluating tradeoffs between management objectives to maximize ecosystem service value with EBM if tradeoffs are to be informed by science. My work in this chapter is the first to attempt to quantitatively demonstrate gaps in impact-pathways for evaluating tradeoffs between EBM objectives, though much previous work has addressed research gaps in science for conservation of resources (Fausch et al., 2002; Sale et al., 2005; Gleason et al., 2006). Taken together, data limitations suggest (1) that future research in support of EBM tradeoffs should be targeted towards the fulfillment of management objectives, as undirected research will be unlikely to satisfy these objectives; and (2) that alternative risk-based management approaches, which do not require evaluation of cost-benefit tradeoffs, may be crucial ways forward given the scope and nature of science available.

I specifically addressed research gaps for management of nonnative species in Chapter 6. Communication of science is frequently faulted as the limitation of science integration into management. Using a nonnative seagrass in the northeast Pacific as a case study I demonstrated that in one case in which management agencies are seeking out research to

130 7.2 Advancing the knowledge-base for marine management inform decision-making, there is a considerable limitation to the integration of science into decision-making from adequate data. I found that at this stage in nonnative research management develops regulations based on both expert opinion and scientific research to more accurately assess the positive and negative interactions of nonnative species.

7.2 Advancing the knowledge-base for marine management I achieved the second half of my objective, to advance the capacity of existing scientific knowledge on anthropogenic impacts to inform coastal EBM, in Chapters 3, 4, and 5. My work with nonnative species fills an important gap in understanding nonnative species introductions and impacts, especially considering the limited research on nonnatives in seagrass. I established the first baseline species assessment of nonnative macroinvertebrates in eelgrass beds of British Columbia (BC), Canada and demonstrated likely vectors and environmental selection factors in Chapter 3. Shipping and aquaculture are well known vectors of marine invasion, however I was able to demonstrate that for nonnatives in eelgrass, aquaculture is likely the dominant vector. Nonnative species richness in BC eelgrass beds is low, especially when compared to the total number of nonnative species in BC. This low richness may signify limited habitat and resource availability in eelgrass or that many newly introduced species are not those that would utilize eelgrass. Considering that half of nonnatives in BC eelgrass have known impacts to the plant, it will be important to understand the dispersal and environmental limitations to potential future invasions to better regulate these species.

Through this work I sampled a species not previously described in the northern Strait of Georgia, BC (Mach et al., 2012). In Chapter 4, I present on the range expansion of the bamboo worm, Clymenella torquata. This species was last described in Boundary Bay and Tsawwassen, BC in the southern Strait of Georgia in 1981 (Banse, 1981; Swinbanks & Murray, 1981). The bamboo worm is now broadly distributed in the Strait of Georgia, BC and established in Washington State, north of Puget Sound. Sampling conducted in Puget Sound by my collaborator, P. Sean MacDonald, and phone interviews with oyster aquaculture facilities in Puget Sound and along the outer coast in Grays Harbor and Willapa Bay, demonstrate a likely southern range limit in Samish Bay, Washington. This

131 7.3 Research across the US-Canada border site is only 75 km south of where it was first sampled in Boundary Bay, BC, while its northern range now extends 240 km from this site. Whereas sampling of nonnative species can provide information on species range, most research on nonnatives demonstrates the presence, and few studies are sufficiently rigorous and comprehensive to demonstrate an absence (Carlton, 2009). In this context, my study represents a rare strong case for the absence of the bamboo worm from Puget Sound.

In Chapter 5, I followed my analysis of vector and environmental selection factors for nonnative establishment with a theoretical analysis estimating harvested shellfish value- at-risk to an invasive species not yet recorded in Puget Sound, Washington. My study of the European green crab, Carcinus maenas, demonstrates that, should green crab invade, shellfish harvest in Puget Sound may decrease dramatically. I estimated a shellfish biomass value-at-risk of 0.33 to 9.8 million lbs, a 3% – 47% loss of harvest, across a range of green crab invasion densities and increasing calorie diets. The loss of revenue associated with reduced biomass for harvest, processing and distribution of shellfish was $1.6 - $41 million USD and an associated loss of 15 to 383 jobs (employment person years). Although the ranges in crab density and calorie diet are broad, they appropriately represent the uncertainty associated with green crab invasion impacts. Loss of shellfish has implications for recreational shellfish harvest and the potential for reduced filtration rates that may lead to increased eutrophication in already threatened coastal habitats (Officer et al., 1982). It is not possible to precisely estimate invasion impacts of green crab and other invading species because of temporal and spatial variation and context- dependent differences across invaded regions (Williamson & Fitter, 1996; Ruiz et al., 1997; Padilla, 2010; Thomsen et al., 2011). Research such as my value-at-risk estimate, that incorporate uncertainty when estimating economic costs resulting from ecological impacts, may be more informative and more accurately assess the range of losses than a more precise estimate that ignores the variability of species invasions.

7.3 Research across the US-Canada border As a student in Vancouver, BC and a US citizen from Seattle, Washington, a city only 150 km south of the Canadian border, I have been presented with a unique opportunity to

132 7.4 Overcoming limitations and future directions for research engage in cross-border research of nonnative species in the region where I grew up. While the ecological barriers that limit species distributions and ecosystem interactions are not present at the US-Canada border, federal and institutional support of research often stops at this political boundary (deRivera et al., 2005; but see Wonham & Carlton, 2005). In Chapters 3, 4, 5 and 6, I worked on nonnative species that span the international border between BC, Canada and Washington State, USA. This international research resulted in international collaborations with Washington State Seagrant (Chapter 6), Department of Fisheries and Oceans Canada (Chapter 3 and 4), NOAA in Washington State (Chapter 5), collaborators from the University of Washington and University of British Columbia (Chapter 2, 3, 4, 5, and 6), and researchers in the Canadian Aquatic Invasive Species Network with whom I presented my findings, collaborated, and conducted field work (Chapters 3 and 4). These collaborative research efforts increase our understanding of species interactions across political barriers that frequently limit the spatial extent of research and improve opportunities for future cross-border research.

7.4 Overcoming limitations and future directions for research My dissertation, exploring and filling knowledge gaps in research towards advancing marine management, was an important step in advancing an understanding of how science might be developed to improve its incorporation into management. As is often the case, this research has resulted in more questions than answers. For example, Chapter 2 provided us with information regarding gaps in research for evaluating tradeoffs in EBM. Future work should explicitly demonstrate where and how assumptions are made when drawing conclusions from multiple studies to explain single impact-pathways and the uncertainty that results from these assumptions. Such research would better inform the utility and applicability of impact-pathway research for evaluating tradeoffs.

In developing research that expanded the understanding of nonnative species in northeast Pacific estuaries, I chose to study selection factors that improve or inhibit invasion success and the impacts that arise from invasion. My work on nonnatives in BC seagrass was extremely detailed in sampling of small-scale habitats within seagrass beds. These habitats host benthic, epibenthic and mobile macro-invertebrates and are an essential part

133 7.4 Overcoming limitations and future directions for research of the productivity of seagrass beds. Because of my detailed work within each seagrass bed, a further extension of this work should be replicated in more eelgrass beds to strengthen statistical testing of environmental variables that may be important selection factors in determining species composition, but which I was unable to test with sufficient strength. Increasing sample sites is also likely to increase the number of nonnative species found in BC eelgrass, as most species are rare and species composition is different amongst sites. In addition, context-dependent effects, such as seasonal temperature change, influx of spring freshwater snow melt altering salinity levels, and change in tidal timing between summer and winter, make it likely that nonnative species richness and abundance fluctuate across seasons and years. Future research should consider temporal shifts in community composition and sample across them to understand the changing community.

Future researchers that seek to create a baseline of nonnative species should consider the importance of a taxonomic skillset. I am not a taxonomist, and as such struggled during the process of identifying species. While I overcame this barrier during my studies by reaching out to a small set of skilled taxonomists in the northeast Pacific, I would recommend the training of future taxonomists, who are indispensible to future research and management of nonnative species (Khuroo et al., 2007; Carlton, 2009). And finally, while observational studies can provide this baseline of information and hint at the relationship of nonnatives to vectors and the environment, these relationships remain unconfirmed until experimental studies can test their causality in influencing species distributions. Further tests of nonnative introductions in eelgrass should aim to disentangle species characteristics, dispersal limitations and environmental limitations to clarify why nonnatives are relatively rare in these habitats.

During sampling and interviews with oyster aquaculture regarding Clymenella, the bamboo worm that I found to have recently expanded north in the Strait of Georgia, BC, it became apparent that research on Clymenella’s impact and possible control mechanisms would be beneficial. Clymenella creates a spongy, porous substrate that has proved detrimental to commercial oyster farms. Oysters in the region of Samish Bay, WA

134 7.4 Overcoming limitations and future directions for research are typically grown using “on-bottom” culture methods; thus when Clymenella densities are high, the oysters sink into the porous sediment and suffocate (Rogers, 2007; P. Dinnel, personal communication). Given the importance of oyster aquaculture to Puget Sound and the Strait of Georgia (PSAT 2003), and the potential for continued and increasing economic costs of Clymenella's impact on oyster farms, it will be important to monitor populations of Clymenella and further explore control methods. Data on habitat selection and modification, reproduction, and species interactions will also aid predictions of this invasion’s future impact should it spread further south into Puget Sound or increase its range in the Strait of Georgia.

Theoretical modeling of nonnative impacts can inform and prepare managers for imminent invasions, allowing for ample time to strengthen preventative regulations that may hold nonnatives at bay. These tests may benefit from comparisons to other regions, such as the northeast Atlantic, where nonnatives have already invaded. For the European green crab, C. maenas, in Puget Sound, ground-truthing of consumption rate on harvest shellfish was not possible as the species has yet to invade. Accordingly, model accuracy would improve with additional investigations. Future exploration regarding the impact of green crab could use mesocosm and field experiments to test the feeding rates and ecosystem level impacts, such as eelgrass destruction, on species that live in Puget Sound and the Strait of Georgia. These results would improve the quality and applicability of the green crab consumption model I developed. In the future, research that estimates the total potential economic damages resulting from green crab invasion could include the costs associated with various management options and estimate the damages with and without these management plans (Hoagland & Jin, 2006).

Nonnatives are often managed based on the available understanding of their impacts on native species. While understanding the gaps in research on Z. japonica can inform future research efforts, specifying what research is sufficient for management of a nonnative species will be necessary for better informing both researchers and managers of future research directions. Because current research efforts on Z. japonica have dominantly focused on benthic species groups that are likely competing with this species for space,

135 7.5 Science to inform management most results have revealed negative impacts of this species. The potential benefits of Z. japonica, such as erosion control and habitat structure, should also be considered before control efforts are funded. The benefits of nonnatives are often overshadowed by their economic and ecological impacts (Schlaepfer et al., 2011).

7.5 Science to inform management Growing evidence demonstrates that seagrass meadows are experiencing worldwide decline primarily as a result of human disturbances, such as direct physical damage and deterioration of water quality (Short & Wyllie-Echeverria, 1996; Hemminga & Duarte, 2000). With more than half of nonnatives found in BC eelgrass having known negative impacts on eelgrass, the threats of these species in conjunction with anthropogenic pressures has the potential to dramatically impact the health of BC eelgrass beds (Ban & Alder, 2008). Understanding how nonnatives establish and impact eelgrass may better inform management actions that prevent establishment and spread (Morgan & Richardson, 2012).

Moving forward with an informed management strategy in coastal ecosystems is a priority and many aspects of the science to support management are already available (Leslie & Mcleod, 2007; Lester et al., 2010). While there will never be research sufficient to answer to all the needs of managers (Oreskes, 2004), scientific uncertainty allows managers a margin of judgment in decision-making. To reduce this uncertainty it is necessary that scientific research is directed by management to fulfill those specific needs (Daily et al., 2009). A decision-making framework will be an important next step for developing objectives that integrate the goals and values of stakeholders and beneficiaries, and provides a consistent evaluation method to choose between alternative management scenarios (Groffman et al., 2006; Cheong, 2008; Espinosa-Romero et al., 2011). Increasing the quantity and scope of research in alignment with management goals should be a priority especially given other impending impacts to ecosystems around the world, including sea-level rise, climate change, and habitat development, and the need to manage under these various future scenarios.

136 References

References Adger, W.N., Hughes, T.P., Folke, C., Carpenter, S.R. & Rockström, J. (2005) Social- ecological resilience to coastal disasters. Science, 309, 1036-1039.

Agardy, T. & Alder, J. (2005) Chapter 19: Coastal Systems. Millennium Ecosystem Assessment - Ecosystems and human well-being: current state and trends (ed. by R. Hassan, R. Scholes and N. Ash), pp. 513-549. Island Press, Washington, D.C.

Allendorf, F.W. & Lundquist, L.L. (2003) Introduction: population biology, evolution, and control of invasive species. Conservation Biology, 17, 24-30.

Altman, S. & Whitlatch, R.B. (2007) Effects of small-scale disturbance on invasion success in marine communities. Journal of Experimental Marine Biology and Ecology, 342, 15-29.

Anderson, D.M., Glibert, P.M. & Burkholder, J.M. (2002) Harmful algal blooms and eutrophication: nutrient sources, composition, and consequences. Estuaries and Coasts, 25, 704-726.

Anderson, D.R. (2008) Model based inference in the life sciences: primer on evidence. Springer, New York.

Anderson, P. (2011) Letter to Chair Brian Blake regarding amendments to the Priority Habitats and Species (PHS) listing of Japanese Eelgrass, Zostera japonica. Department of Washington State Fish and Wildlife, Olympia, Washington. February 8, 2011. pp. 2.

Baeta, A., Cabral, H.N., Marques, J.C. & Pardal, M.A. (2006) Feeding ecology of the green crab, Carcinus maenas (L., 1758) in a temperate Estuary, Portugal. Crustaceana, 79, 1181-1193.

Baldwin, J.R. & Lovvorn, J.R. (1994) Expansion of seagrass habitat by the exotic Zostera japonica, and its use by dabbling ducks and brant in Boundary Bay, British Columbia. Marine Ecology Progress Series, 103, 119-127.

Ban, N. & Alder, J. (2008) How wild is the ocean? Assessing the intensity of anthropogenic marine activities in British Columbia, Canada. Aquatic Conservation: Marine and Freshwater Ecosystems, 18, 55-85.

Bando, K.J. (2006) The roles of competition and disturbance in a marine invasion. Biological Invasions, 8, 755-763.

137 References

Banse, K. (1981) On some Cossuridae and Maldanidae (Polychaeta) from Washington and British Columbia. Canadian Journal of Fisheries and Aquatic Sciences, 38, 633-637.

Barbier, E.B. (2001) A note on the economics of biological invasions. Ecological Economics, 39, 197-202.

Barbier, E.B., Hacker, S.D., Kennedy, C., Koch, E.W., Stier, A.C. & Silliman, B.R. (2011) The value of estuarine and coastal ecosystem services. Ecological Monographs, 81, 169-193.

Barnard, J.L. (1958) Amphipod crustaceans as fouling organisms in Los Angeles-Long Beach Harbors, with reference to the influence of seawater turbidity. California Fish & Game 44, 161–170.

BC Shellfish Grower's Association (2012) Shellfish industry encyclopedia. Available at: http://bcsga.ca/about/industry-encyclopedia/ (accessed June 2012).

BCStats (2006) 2006 Census, Statistics Canada. Available at: http://www12.statcan.gc.ca/census-recensement/2006/index-eng.cfm (accessed Jan 20, 2010).

Beal, B.F. & Kraus, M.G. (2002) Interactive effects of initial size, stocking density, and type of predator deterrent netting on survival and growth of cultured juveniles of the soft-shell clam, Mya arenaria L., in eastern Maine. Aquaculture, 208, 81-111.

Beger, M., Grantham, H.S., Pressey, R.L., Wilson, K.A., Peterson, E.L., Dorfman, D., Mumby, P.J., Lourival, R., Brumbaugh, D.R. & Possingham, H.P. (2010) Conservation planning for connectivity across marine, freshwater, and terrestrial realms. Biological Conservation, 143, 565-575.

Bergstrom, J.C., Dorfman, J.H. & Loomis, J.B. (2004) Estuary management and recreational fishing benefits. Coastal Management, 32, 417-432.

Berkenbusch, K., Rowden, A.A. & Myers, T.E. (2007) Interactions between seagrasses and burrowing ghost shrimps and their influence on infaunal assemblages. Journal of Experimental Marine Biology and Ecology, 341, 70-84.

Berry, H.D., Sewell, A.T., Wyllie-Echeverria, S., Reeves, B.R., Mumford Jr., T.F., Skalski, J., Zimmerman, R.C. & Archer, J. (2003) Puget Sound submerged vegetation monitoring project: 2000-2002 monitoring report. Nearshore Habitat Program, Washington State Department of Natural Resources, Olympia, WA. pp. 60.

138 References

Beukema, J.J. (1997) Caloric values of marine invertebrates with an emphasis on the soft parts of marine bivalves. Oceanography and Marine Biology: an Annual Review, 35, 387-414.

Blackburn, T.M., Pysek, P., Bacher, S., Carlton, J.T., Duncan, R.P., Jarosik, V., Wilson, J.R.U. & Richardson, D.M. (2011) A proposed unified framework for biological invasions. Trends in Ecology and Evolution, 1-7.

Blakeslee, A.M.H., Byers, J.E. & Lesser, M.P. (2008) Solving cryptogenic histories using host and parasite molecular genetics: The resolution of Littorina littorea’s North American origin. Molecular Ecology, 17, 3684-96.

Boesch, D.F. & Goldman, E.B. (2009) Chapter 15: Chesapeake Bay, USA. Ecosystem- based management for the oceans (ed. by K. Mcleod and H. Leslie). Island Press, Washington, D.C.

Bolton, R.M. & Brooks, R.J. (2010) Impact of the seasonal invasion of Phragmites australis (Common Reed) on turtle reproductive success. Chelonian Conservation and Biology, 9, 238-243.

Born, W., Rauschmayer, F. & Brauer, I. (2005) Economic evaluation of biological invasions--a survey. Ecological Economics, 55, 321-336.

Brauman, K.A., Daily, G.C., Duarte, T.K. & Mooney, H.a. (2007) The nature and value of ecosystem services: An overview highlighting hydrologic services. Annual Review of Environmental Resources, 32, 67-98.

Britton, J.R., Gozlan, R.E. & Copp, G.H. (2011) Managing non-native fish in the environment. Fish and Fisheries, 12, 256-274.

Browman, H. & Stergiou, K. (2006) Politics and socio-economics of ecosystem-based management of marine resources - Theme Issue. Marine Ecology Progress Series, 300, 241-291.

Buckley, Y.M. (2008) The role of research for integrated management of invasive species, invaded landscapes and communities. Journal of Applied Ecology, 45, 397-402.

Bulleri, F., Russell, B.D. & Connell, S.D. (2012) Context-dependency in the effects of nutrient loading and consumers on the availability of space in marine rocky environments. PLoS Biology, 7, e33825.

139 References

Bulthuis, D.A. (1995) Distribution of seagrasses in a north Puget Sound estuary: Padilla Bay, Washington, USA. Aquatic Botany, 50, 99-105.

Burnham, K.P. & Anderson, D.R. (2002) Model selection and multimodel inference: a practical information-theoretic approach, 2nd edn. Springer, New York.

CAISN (2011) Canadian Aquatic Invasive Species Network 2006-2011 Final Report. Available at: http://www.caisn.ca/uploads/CAISN_FinalReport20112a.pdf (accessed August 2012).

Calabretta, C.J. & Oviatt, C.A. (2008) The response of benthic macrofauna to anthropogenic stress in Narragansett Bay, Rhode Island: a review of human stressors and assessment of community conditions. Marine Pollution Bulletin, 56, 1680-1695.

Cameron, W.M. & Pritchard, D.W. (1963) Estuaries. The Sea: Ideas and Observations on the Progress in the Study of the Seas. Vol. 2 (ed. by M.N. Hill), pp. 306-324. Interscience Publishers, John Wiley and Sons Ltd, New York and London.

Campbell, D. (1993) Socio-economic study of the Black Sea fisheries. General Fisheries Council for the Mediterranean: Report of the Second Technical Consultation on Stock Assessment in the Black Sea, Ankara, Turkey, 15–19 February, pp. 102- 129. FAO Fisheries Report No. 495, Rome.

Canadian Food Inspection Agency (2011) Aquatic Animal Imports. Available at: http://www.inspection.gc.ca/animals/aquatic- animals/imports/eng/1299156741470/1320599337624 (accessed May 2012).

Carlton, J.T. (1987) Patterns of transoceanic marine biological invasions in the Pacific Ocean. Bulletin of Marine Science, 41, 452-465.

Carlton, J.T. (2009) Chapter 2. Deep invasion ecology and the assembly of communities in historical time. Biological Invasions in Marine Ecosystems (ed. by G. Rilov and J.A. Crooks), pp. 13-56. Springer-Verlag, Berlin

Carlton, J.T. & Cohen, A.N. (2003) Episodic global dispersal in shallow water marine organisms: the case history of the European shore crabs Carcinus maenas and C. aestuarii. Journal of Biogeography, 30, 1809-1820.

Carlton, J.T. & Ruiz, G.M. (2005) Vector science and integrated vector management in bioinvasion ecology: conceptual frameworks. Invasive alien speices: a new synthesis (ed. by H.A. Mooney, R.N. Mack, J.A. Mc Neely, L.E. Neville, P.J. Schei and J.K. Waage), pp. 36-58. Island Press, Covelo, California.

140 References

Carman, M. & Grunden, D. (2010) First occurrence of the invasive tunicate Didemnum vexillum in eelgrass habitat. Aquatic Invasions, 5, 23-29.

Carman, M.R., Hoagland, K.E., Green-Beach, E. & Grunden, D.W. (2009) Tunicate faunas of two North Atlantic-New England islands: Martha’s Vineyard, Massachusetts and Block Island, Rhode Island. Aquatic Invasions, 4, 65-70.

Carpenter, S.R., Kitchell, J.F. & Hodgson, J.R. (1985) Cascading trophic interactions and lake productivity. Bioscience, 35, 634-639.

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

Carr, L.A., Boyer, K.E. & Brooks, A.J. (2010) Spatial patterns of epifaunal communities in San Francisco Bay eelgrass (Zostera marina) beds. Marine Ecology, 32, 88- 103.

Casman, E.A., Morgan, M.G. & Dowlatabadi, H. (1999) Mixed levels of uncertainty in complex policy models. Risk Analysis, 19, 33-42.

Chan, K.M.A., Satterfield, T. & Goldstein, J. (2012) Rethinking ecosystem services to better address and navigate cultural values. Ecological Economics, 74, 8-18.

Cheong, S.M. (2008) A new direction in coastal management. Marine Policy, 32, 1090- 1093.

Choi, F.M. (2011) Assessing intertidal marine non-indigenous species in Canadian ports. M.Sc. thesis, University of British Columbia, Vancouver.

Clark, G.F. & Johnston, E.L. (2005) Manipulating larval supply in the field: a controlled study of marine invasibility. Marine Ecology Progress Series, 298, 9-19.

Clarke Murray, C. (2012) The role of recreatinal boating in the introduction and spread of marine invasive species. Ph.D thesis, University of British Columbia, Vancouver.

Clarke Murray, C., Pakhomov, E.A. & Therriault, T.W. (2011) Recreational boating: a large unregulated vector transporting marine invasive species. Diversity and Distributions, 17, 1161-1172.

Clinton, W.J. (1999) Executive Order 13112 of February 3, 1999. Federal Register 64 (Doc no. 25) 64:6183-6186.

141 References

Cohen, A.N. (2004) An exotic species detection program for Puget Sound. Puget Sound Action Team, Olympia, Washington. pp. 60.

Cohen, A.N. & Carlton, J.T. (1995) Nonindigenous aquatic species in a United States estuary: a case study of the biological invasions of the San Francisco Bay and Delta. United States Fish and Wildlife Service, Washington D.C., and the National Sea Grant College Program Connecticut Sea Grant.

Cohen, A.N. & Carlton, J.T. (1998) Accelerating invasion rate in a highly invaded estuary. Science, 279, 555-558.

Cohen, A.N. & Zabin, C.J. (2009) Oyster shells as vectors for exotic organisms. Journal of Shellfish Research, 28, 163-167.

Cohen, A.N., Carlton, J.T. & Fountain, M.C. (1995) Introduction, dispersal and potential impacts of the green crab Carcinus maenas in San Francisco Bay, California. Marine Biology, 122, 225-237.

Colautti, R.I. & MacIsaac, H.J. (2004) A neutral terminology to define invasive species. Diversity and Distributions, 10, 135-141.

Colautti, R.I., Bailey, S.A., van Overdijk, C.D.A., Amundsen, K. & MacIsaac, H.J. (2006) Characterised and projected costs of nonindigenous species in Canada. Biological Invasions, 8, 45-59.

Collins, S.L., Carpenter, S.R., Swinton, S.M., Orenstein, D.E., Childers, D.L., Gragson, T.L., Grimm, N.B., Grove, J.M., Harlan, S.L., Kaye, J.P., Knapp, A.K., Kofinas, G.P., Magnuson, J.J., McDowell, W.H., Melack, J.M., Ogden, L.A., Philip Robertson, G.P., Smith, M.D. & Whitmer, A.C. (2011) An integrated conceptual framework for long-term social-ecological research. Frontiers in Ecology and the Environment, 9, 351-357.

Colnar, A.M. & Landis, W.G. (2007) Conceptual model development for invasive species and a regional risk assessment case study: The European green crab, Carcinus maenas, at Cherry Point, Washington, USA. Human and Ecological Risk Assessment, 13, 120-155.

Connell, S.D. & Irving, A.D. (2008) Integrating ecology with biogeography using landscape characteristics: a case study of subtidal habitat across continental Australia. Journal of Biogeography, 35, 1608-1621.

142 References

Cordell, J.R., Draheim, R. & Sytsma, M. (2007) First record of the harpacticoid genus Harpacticella in the Pacific Northwest, USA: another probable introduction. Aquatic Biology, 1, 17-20.

Costanza, R., d'Arge, R., de Groot, R., Farber, S., Grasso, M., Hannon, B., Limburg, K., Naeem, S., O'Neill, R.V., Paruelo, J., Raskin, R.G., Sutton, P. & van den Belt, M. (1997) The value of the world's ecosystem services and natural capital. Nature, 387, 253-260.

Crutchfield, J.A. (1962) Valuation of fishery resources. Land Economics, 38, 145-154.

Dafforn, K.A., Glasby, T.M. & Johnston, E.L. (2009) Links between estuarine condition and spatial distributions of marine invaders. Diversity and Distributions, 15, 807- 821.

Dahlstrom, A., Hewitt, C.L. & Campbell, M.L. (2011) A review of international, regional and national biosecurity risk. Marine Policy, 35, 208-217.

Daily, G.C. (1997) Nature's services: societal dependence on natural ecosystems. Island Press, Washington, D.C. pp. 362.

Daily, G.C., Polasky, S., Goldstein, J., Kareiva, P.M., Mooney, H.A., Pejchar, L., Ricketts, T.H., Salzman, J. & Shallenberger, R. (2009) Ecosystem services in decision making: time to deliver. Frontiers in Ecology and the Environment, 7, 21-28.

Daily, G.C., Alexander, S., Ehrlich, P.R., Goulder, L., Lubchenco, J., Matson, P.A., Mooney, H.a., Pstel, S., Schneider, S.H., Tilman, D. & Woodwell, G.M. (1997) Ecosystem services: benefits supplied to human societies by natural ecosystems. Issues in Ecology, 2, 1-18.

Davidson, I.C., McCann, L.D., Fofonoff, P.W., Sytsma, M.D. & Ruiz, G. (2008) The potential for hull-mediated species transfers by obsolete ships on their final voyages. Diversity and Distributions, 14, 518-529.

Davis, M.A., Grime, J.P. & Thompson, K. (2000) Fluctuating resources in plant communities: a general theory of invasibility. Journal of Ecology, 88, 528-534.

Davis, R.C., Short, F.T. & Burdick, D.M. (1998) Quantifying the effects of green crab damage to eelgrass transplants. Restoration Ecology, 6, 297-302.

143 References

Demes, K.W., Kordas, R.L. & Jorve, J.P. (2012) Ferry wakes increase seaweed richness and abundance in a sheltered rocky intertidal habitat. Hydrobiologia, Online First, DOI: 10.1007/s10750-012-1082-4. denHartog, C. (1997) Is Sargassum muticum a threat to eelgrass beds? Aquatic Botany, 58, 37-41.

Department of Fisheries and Oceans Canada (1994) Herring geographical bulletin: herring spawn. Available at: http://www.pac.dfo-mpo.gc.ca/science/species- especes/pelagic-pelagique/herring-hareng/herspawn/pages/indicator-eng.htm (accessed May 2012). deRivera, C., Ruiz, G., Crooks, J., Wasson, K., Lonhart, S., Fofonoff, P., Steves, B., Rumrill, S., Brancato, M.S., Pegau, S. & Bulthuis, D. (2005) Broad-scale nonindigenous species monitoring along the west coast in National Marine Sanctuaries and National Estuarine Research Reserves. Report to the National Fish and Wildlife Foundation. pp. 126. deRivera, C.E., Steves, B.P., Fofonoff, P.W., Hines, A.H. & Ruiz, G.M. (2011) Potential for high-latitude marine invasions along western North America. Diversity and Distributions 17, 1198-1209. deRivera, C.E., Grosholz, E.D., Ruiz, G.M., Larson, A.A., Kordas, R. & Sytsma, M.D. (2007a) Green crab management: Reduction of an invasive population. Proceedings of Coastal Zone 07, Portland, Oregon. pp. 6. deRivera, C.E., Hitchcock, N.G., Teck, S.J., Steves, B.P., Hines, A.H. & Ruiz, G.M. (2007b) Larval development rate predicts range expansion of an introduced crab. Marine Biology, 150, 1275-1288.

Dethier, M.N. (2006) Native shellfish in nearshore ecosystems of Puget Sound. Puget Sound Nearshore Partnership, Olympia, Washington. pp. 26.

Dethier, M.N. & Hacker, S.D. (2005) Physical factors vs. biotic resistance in controlling the invasion of an estuarine marsh grass. Ecological Applications, 15, 1273-1283.

Diaz, R.J. & Rosenberg, R. (1995) Marine benthic hypoxia: a review of its ecological effects and the behavioural responses of benthic macrofauna. Oceanography and Marine Biology, 33, 245-303.

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

144 References

DiBacco, C., Humphrey, D.B., Nasmish, L.E. & Levings, C.D. (2011) Ballast water transport of non-indigenous zooplankton to Canadian ports. ICES Journal of Marine Science, 10.1093/icesjms/fsr133.

Didham, R.K., Tylianakis, J.M., Hutchison, M.A., Ewers, R.M. & Gemmell, N.J. (2005) Are invasive species the drivers of ecological change? Trends in Ecology & Evolution, 20, 470-474.

Dinnel, P., McMillan, R., Armstrong, D., Wainwright, T., Whiley, A., Burge, R. & Bumgarner, R. (1986) Padilla Bay Dungeness crab, Cancer magister, habitat study. Report to NOAA/OCRM/MEMD by Univ. Wash., Fisheries Research Institute, Seattle, WA. Padilla Bay National Estuarine Research Reserve Reprint Series No. 3, 199078.

Duarte, C.M. (1999) Seagrass ecology at the turn of the millennium: challenges for the new century. Aquatic Botany, 65, 7-20.

Duarte, C.M., Dennison, W.C., Orth, R.J. & Carruthers, T.J.B. (2008) The charisma of coastal ecosystems: addressing the imbalance. Estuaries Coasts, 31, 233-238.

Dudas, S.E. & Dower, J.F. (2006) Reproductive ecology and dispersal potential of the varnish clam Nuttallia obscurata, a recent invader in the Northeast Pacific. Marine Ecological Progress Series, 320, 195-205.

Duffy, J.E. & Harvilicz, A.M. (2001) Species-specific impacts of grazing amphipods in an eelgrass-bed community. Marine Ecological Progress Series, 223, 201-211.

Duffy, J.E., Cardinale, B.J., France, K.E., McIntyre, P.B., Thebault, E. & Loreau, M. (2007) The functional role of biodiversity in ecosystems: incorporating trophic complexity. Ecology Letters, 10, 522-538.

Dumbauld, B.R. & Wyllie-Echeverria, S. (2003) The influence of burrowing thalassinid shrimps on the distribution of intertidal seagrasses in Willapa Bay, Washington, USA. Aquatic Botany, 77, 27-42.

Eissinger, A. (2008) Puget Sound marine invasive species identification guide. Nahkeeta Northwest Wildlife Services, Bow, WA & Washington Department of FIsh and Wildlife, Aquatic Nuisance Species Program, Olympia, WA.

Ekman, S. (1953) Zoogeography of the Sea. Sidgwick and Jackson Limited, London. pp. 440.

145 References

Elahi, R. & Sebens, K. (2012) Consumers mediate natural variation between prey richness and resource use in a benthic marine community. Marine Ecological Progress Series, 452, 131-143.

Elston, R. (1997) Pathways and management of marine nonindigenous species in the shared waters of British Columbia and Washington. For: Puget Sound Water Quality Authority, US Environmental Protection Agency and Region 10 Department of Fisheries and Oceans, Canada, Carlsborg, Washington. pp. 50.

Eno, N.C., Clark, R.A. & Sanderson, W.G. (1997) Non-native marine species in British waters: a review and directory. Joint Nature Conservation Committee, Peterbourough, UK. pp. 136.

Environment Canada (2011) Shellfish aquaculture statistics for British Columbia. Available at: http://www.env.gov.bc.ca/omfd/fishstats/aqua/shellfish.html (accessed March 21 2012).

Environment Canada (2012) Reports on plans and priorities for 2012-2013: section I - overview. Available at: http://www.tbs-sct.gc.ca/rpp/2012-2013/inst/doe/doe00- eng.asp (accessed May 2012).

Espinosa-Romero, M.J., Chan, K.M.A., McDaniels, T. & Dalmer, D.M. (2011) Structuring decision-making for ecosystem-based management. Marine Policy, 35, 575-583.

Fauchald, K., Granados-Barba, A. & Solís-Weiss, V. (2009) Polychaeta (Annelids) of the Gulf of Mexico. Gulf of Mexico - Origins, Waters, Biota: Biodiversity (ed. by D.L. Felder and D.K. Camp), pp. 751-788. Texas A&M Press, College Station, Texas.

Fausch, K.D., Torgersen, C.E., Baxter, C.V. & Li, H.W. (2002) Landscapes to riverscapes: bridging hte gap between research and conservation of stream fishes. BioScience, 52, 483-498.

Floerl, O., Pool, T.K. & Inglis, G.J. (2004) Positive interactions between nonindigenous species facilitate transport by human vectors. Ecological Applications, 14, 1724- 1736.

Fonseca, M.S., Fisher, J.S., Zieman, J.C. & Thayer, G.W. (1982) Influence of the seagrass Zostera marina L., on current flow. Estuarine, Coastal and Shelf Science, 15, 351-364.

146 References

Foreman, M.G.G., Crawford, W.R., Cherniawsky, J.Y. & Galbraith, J. (2008) Dynamic ocean topography for the northeast Pacific and its continental margins. Geophysical Research Letters, 35, 1-5.

Fulton, S.W. & Grant, F.E. (1900) Note on the occurrence of the European crab, Carcinus maenas, Leach, in Port Phillip. Victorian Naturalist, 27, 147-148.

Gaeckle, J., Dowty, P., Berry, H. & Ferrier, L. (2009) Puget Sound submerged vegetation monitoring project: 2008 monitoring report. Nearshore Habitat Program, Washington State Department of Natural Resources, Olympia, WA. pp. 57.

Gartner, H.N. (2010) Subtitdal invertebrate fouling communities of the British Columbian coast. M.Sc. thesis, University of Victoria, Victoria.

Gates, S. (2002) Review of methodology of quantitative reviews using meta-analysis in ecology. Journal of Animal Ecology, 71, 547-557.

Geller, J.B., Walton, E.D., Grosholz, E.D. & Ruiz, G.M. (1997) Cryptic invasions of the crab Carcinus detected by molecular phylogeography. Molecular Ecology, 6, 901- 906.

Gillespie, G., Phillips, A., Paltzat, D. & Therriault, T. (2007) Status of the European green crab, Carcinus maenas, in British Columbia - 2006. Department of Fisheries and Oceans, Canadian Technical Report of Fisheries and Aquatic Sciences, 270049.

Gillespie, G.E. (2007) Distribution of non-indigenous intertidal species on the Pacific Coast of Canada. Nippon Suisan Gakkaishi, 73, 1133-1137.

Glasby, T.M., Connell, S.D., Holloway, M.G. & Hewitt, C.L. (2007) Nonindigenous biota on artificial structures: could habitat creation facilitate biological invasions? Marine Biology, 151, 887-895.

Gleason, M.G., Merrifield, M.S., Cook, C.W., Davenport, A.L. & Shaw, R. (2006) Assessing gaps in marine conservation in California. Frontiers in Ecology and the Environment, 4, 249-258.

Gotelli, N.J. & Colwell, R.K. (2001) Quantifying biodiversity: procedures and pitfalls in the measurement and comparison of species richness. Ecology Letters, 4, 379- 391.

Granek, E.F., Polasky, S., Kappel, C.V., Reed, D.J., Stoms, D.M., Koch, E.W., Kennedy, C.J., Cramer, L.A., Hacker, S.D. & Barbier, E.B. (2010) Ecosystem services as a

147 References

common language for coastal ecosystem‐based management. Conservation Biology, 24, 207-216.

Gregr, E.J. & Chan, K.M.A. (2011) Making science relevant to marine ecosystem-based management. Biological Conservation, 144, 670-671.

Groffman, P.M., Baron, J.S., Blett, T., Gold, A.J., Goodman, I., Gunderson, L.H., Levinson, B.M., Palmer, M.A., Peaerl, H.W. & Peterson, G.D. (2006) Ecological thresholds: the key to successful environmental management or an important concept with no practical application? Ecosystems, 9, 1-13.

Grosholz, E.D. (2002) Ecological and evolutionary consequences of coastal invasions. Trends in Ecology & Evolution, 17, 22-27.

Grosholz, E.D. (2005) Recent biological invasion may hasten invasional meltdown by accelerating historical introductions. Proceedings of the National Academy of Sciences of the United States of America, 102, 1088-1091.

Grosholz, E.D. & Ruiz, G.M. (1996) Predicting the impact of introduced marine species: Lessons from the multiple invasions of the European green crab Carcinus maenas. Biological Conservation, 78, 59-66.

Grosholz, E.D., Ruiz, G.M., Dean, C.A., Shirley, K.A., Maron, J.L. & Connors, P.G. (2000) The impacts of a nonindigenous marine predator in a California bay. Ecology, 81, 1206-1224.

Grout, J., Levings, C.D. & Richardson, J.S. (1997) Decomposition rates of purple loosestrife (Lythrum salicaria) and Lyngbyei’s sedge (Carex lyngbyei) in the Fraser River estuary Estuaries, 20, 96-102.

GSGislason & Associates Ltd. (2007) Economic contribution of the oceans sector in British Columbia. For: Canada/British Columbia Oceans Coordinating Committee, Vancouver, British Columbia. pp. 136.

Haegele, C.W., Humphreys, R.D. & Hourston, A.S. (1981) Distribution of eggs by depth and vegetation type in Pacific herring (Clupea harengus pallasi) spawnings in Southern British Columbia. Canadian Journal of Fisheries and Aquatic Sciences, 38, 381-386.

Hahn, D.R. (2003a) Changes in community composition and ecosystem processes associated with biological invasions: Impacts of Zostera japonica in the marine intertidal zone. Ph.D Thesis, University of Washington, Seattle, WA.

148 References

Hahn, D.R. (2003b) Alteration of microbial community composition and changes in decomposition associated with an invasive intertidal macrophyte. Biological Invasions, 5, 45-51.

Halpern, B.S., Ebert, C.M., Kappel, C.V., Madin, E.M.P., Micheli, F., Perry, M., Selkoe, K.A. & Walbridge, S. (2009) Global priority areas for incorporating land-sea connectionsin marine conservation. Conservation Letters, 2, 189-196.

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

Hamby, D.M. (1994) A review of techniques for parameter sensitivity analysis of environmental models. Environmental Monitoring and Assessment, 32, 135-154.

Hancock, L.P., McDonald, P.S., Goetz, F.E. & Dinnel, P. (2008) The bamboo worm invasion of Samish Bay: ecology and control of Clymenella torquata in a northeastern Pacific estuary. In: Technical Abstracts of the 61st Annual Conference of the National Shellfisheries Association (Oct 9-12, 2007), Mountain Welches, Oregon. Journal of Shellfish Research, 27, 465-473.

Harrison, P.G. (1979) Reproductive strategies in intertidal populations of two co- occurring seagrasses (Zostera spp.). Canadian Journal of Botany, 57, 2635-2638.

Harrison, P.G. (1987) Natural expansion and experimental manipulation of seagrass Zostera spp. abundance and the response of infaunal invertebrates. Estuarine Coastal and Shelf Science, 24, 799-812.

Harrison, P.G. & Bigley, R.E. (1982a) The recent introduction of the seagrass Zostera japonica Aschers. and Graebn. to the Pacific coast of North America. Aquatic Science, 39, 1642-1648.

Harrison, P.G. & Bigley, R.E. (1982b) The recent introduction of the seagrass Zostera japonica Aschers. & Graebn. to the Pacific coast of North America. Canadian Journal of Fisheries and Aquatic Sciences, 39, 1642-1648.

Haupt, T.M., Griffiths, C.L., Robinson, T.B. & Tonin, A.F.G. (2010) Oysters as vectors of marine aliens, with notes on four introduced species associated with oyster farming in South Africa. African Zoology, 45, 52-62.

149 References

Hayes, K., Cannon, R., Neil, K. & Inglis, G.J. (2005) Sensitivity and cost considerations for the detection and eradication of marine pests in ports. Marine Pollution Bulletin, 50, 823-834.

Hazen, C.J. (1996) The mapping of eelgrass in Willapa Bay, WA and an evaluation of C- Cap mapping guidelines. Masters Thesis, Oregon State University, Corvallis, Oregon.

Healey, M.C. (1991) The life history of chinook salmon (Oncorhynchus tshawytscha). Life history of Pacific salmon (ed. by C. Groot and L. Margolis), pp. 311-393. Univ. BC Press, Vancouver, BC.

Heck, K.L. & Orth, R.J. (1980) Seagrass habitats: the roles of habitat complexity, competition and predation in structuring associated fish and motile macroinvertebrate assemblages. Estuarine Perspectives (ed. by V.S. Kennedy), pp. 449–464. Academic Press, New York.

Hedge, P., Kriwoken, L.K. & Patten, K. (2003) A review of Spartina management in Washington State, US. Journal of Aquatic Plant Management, 41, 82-90.

Hedgpeth, J.W. (1993) Foreign invaders. Science, 261, 34-35.

Hemminga, M.A. & Duarte, C.M. (2000) Seagrass Ecology. Cambridge University Press, Cambridge, UK.

Hershner, C. & Havens, K. (2008) Managing invasive aquatic plants in a changing system: strategic consideration of ecosystem services. Conservation Biology, 22, 544-550.

Hessing-Lewis, M.L., Hacker, S.D., Menge, B.A. & Rumrill, S.S. (2011) Context- dependent eelgrass–macroalgae interactions along an estuarine gradient in the Pacific Northwest, USA. Estuaries and Coasts, 34, 1169-1181.

Hewitt, C.L. & Campbell, M.L. (2010) Relative contribution of vectors to the introduction and translocation of marine invasive species. Australian Government Department of Agriculture, Fisheries and Forestry, Canberra City.

Hidalgo, F.J., Baron, P.J. & Orensanz, J.M. (2006) A prediction come true: the green crab invades the Patagonian coast. Biological Invasions, 7, 547-552.

Hitchcock, C.L., Cronquist, A., Ownbey, M. & Thompson, J.W. (1969) Vascular plants of the Pacific Northwest, Part 1: Vascular cryptogams, gymnosperms, and monocotyledons. Univ. Washington Press, Seattle, WA.

150 References

Hoagland, P. & Jin, D. (2006) Science and economics in the management of an invasive species. Bioscience, 56, 931-935.

Hobbs, R.J. (2006) Novel ecosystems: theoretical and management aspects of the new ecological world order. Global Ecology and Biogeography, 15, 1-7.

Hobbs, R.J. & Huenneke, L.F. (1992) Disturbance, diversity, and invasion: Implications for conservation. Conservation Biology, 6, 324-337.

Hoffman, E.J., Mills, G.L., Latimer, J.S. & Quinn, J.G. (1984) Urban runoff as a source of polycyclic aromatic hydrocarbons to coastal waters. Environmental Science & Technology, 18, 580-587.

Hoos, P.M., Miller, A.W., Ruiz, G.M., Vrijenhoek, R.C. & Geller, J.B. (2010) Genetic and historical evidence disagree on likely sources of the Atlantic amethyst gem clam Gemma gemma (Totten, 1834) in California. Diversity and Distributions, 16, 582-592.

Horan, R.D., Perrings, C., Lupi, F. & Bulte, E.H. (2002) Biological pollution prevention strategies under ignorance: the case of invasive species. American Journal of Agriculture Economics, 84, 1303-1310.

Hughes, A.R., Williams, S.L., Duarte, C.M., Heck, K.L. & Waycott, M. (2008) Associations of concern: declinging seagrasses and threatened dependent species. Frontiers in Ecology and the Environment, 7, 242-246.

Hurlbert, S.H. (1984) Pseudoreplication and the design of ecological field experiments. Ecological Monographs, 54, 187-211.

Huyer, A., Barth, J.A., Fleischbein, J. & Kosro, P.M. (1998) The coastal ocean off Oregon and Northern California during the 1997-8 El Nino, part 1: temperature, salinity and geostropic velocity fields. Transactions of the American Geophysical Union, 79, F485.

Incera, M., Olbarria, C., Troncoso, J.S. & Lopez, J. (2009) Response of the invader Sargassum muticum to variability in nutrients supply. Marine Ecological Progress Series, 377, 91-101.

International Association for Impact Assessment (1999) Principle of environmental impact assessment best practice. Available at: http://www.iaia.org/publicdocuments/special-publications/ (accessed June 2012).

151 References

International Council for the Exploration of the Sea (ICES) (2005) ICES code of practice on the introductions and transfers of marine organisms. ICES, Copenhagen, Denmark.

Jamieson, G., Grosholz, E., Armstrong, D. & Elner, R. (1998) Potential ecological implications from the introduction of the European green crab, Carcinus maenas (Linneaus), to British Columbia, Canada, and Washington, USA. Journal of Natural History, 32, 1587-1598.

Jamieson, G.S., Foreman, M.G.G., Cherniqwsky, J. & Levings, C.D. (2002) European green crab (Carcinus maenas) dispersal: the Pacific experience. Crabs in cold water regions: biology, management, and economics (ed. by A.J. Paul, E.G. Dawe, R. Elner, G.S. Jamieson, G.H. Kruse, R.S. Otto, B. Seainte-Marie, T.C. Shirley and D. Dwoodby). University of Alaska Sea Grant (AK-SC-02-01), Fairbanks, Alaska.

Jiao, N.Z., Zhang, Y., Zeng, Y.H., Gardner, W.D., Mishonov, A.V., Richardson, M.J., Hong, N., Pan, D.L., Yan, X.H., Jo, Y.H., Chen, C.T.A., Wang, P.X., Chen, Y.T., Hong, H., Bai, Y., Chen, X.H., Huang, B.Q., Deng, H., Shi, Y., Yang, D.C., Jiao, N., Zhang, Y., Zeng, Y., Gardner, W.D., Mishonov, A.V., Richardson, M.J., Hong, N., Pan, D., Yan, X.H., Jo, Y.H., Chen, C.A., Wang, P., Chen, Y., Hong, H., Bai, Y., Chen, X., Huang, B., Deng, H., Shi, Y. & Yang, D. (2007) Ecological anomalies in the East China Sea: Impacts of the three gorges dam? Water Research, 41, 1287-1293.

JMP 7.0 (2007) JMP. SAS Institute Inc.

Jorion, P. (2006) Value at risk: the new benchmark for managing financial risk (3rd ed.). McGraw-Hill, Maidenhead, UK.

Kaldy, J.E. (2006) Production ecology of the non-indigenous seagrass, dwarf eelgrass (Zostera japonica Ascher. & Graeb.), in a Pacific Northwest estuary, USA. Hydrobiologia, 533, 201-217.

Kask, B.A., Sibert, J.R. & Windecker, B. (1982) A check list of marine and brackish water harpacticoid copepods from the Nanaimo estuary, southwestern British Columbia. Syesis, 15, 25-38.

Keller, R.P., Frang, K. & Lodge, D.M. (2007) Preventing the spread of invasive species: economic benefits of intervention guided by ecological predictions. Conservation Biology, 22, 80-88.

152 References

Khuroo, A.A., Dar, G.H., Khan, Z.S. & Malik, A.H. (2007) Exploring an inherent interface between taxonomy and biodiversity: current problems and future challenges. Journal for Nature Conservation, 15, 256-261.

Kirby, M.X. (2004) Fishing down the coast: historical expansion and collapse of oyster fisheries along continental margins. Proceedings of the National Academy of Sciences, 101, 13096-13099.

Knowler, D. (2005) Reassessing the costs of biological invasion: Mnemiopsis leidyi in the Black sea. Ecological Economics, 52, 187-199.

Kolar, C.S. & Lodge, D.M. (2001) Progress in invasion biology: predicting invaders. Trends in Ecology & Evolution, 16, 199-204.

Korajkic, A., Badgley, B.D., Brownell, M.J. & Harwood, V.J. (2009) Application of microbial source tracking methods in a Gulf of Mexico field setting. Journal of Applied Microbiology, 107, 1518-1527.

Kordas, R.L. (2009) Modeling variation in interaction strength between barnacles and fucoids. Oecologia, 158, 717-731.

Kremen, C. (2005) Managing ecosystem services: what do we need to know about their ecology? Ecology Letters, 8, 468-479.

Kremen, C., Williams, N.M., Aizen, M.A., Gemmill-Herren, B., LeBuhn, G., Minckley, R., Packer, L., Potts, S.G., Roulston, T.a.a., Steffan-Dewenter, I., Vazquez, D.P., Winfree, R., Adams, L., Crone, E.E., Greenleaf, S.S., Keitt, T.H., Klein, A.-M., Regetz, J. & Ricketts, T.H. (2007) Pollination and other ecosystem services produced by mobile organisms: a conceptual framework for the effects of land- use change. Ecology Letters, 10, 299-314.

Kruskal, J.B. & Wish, M. (1978) Multidimensional Scaling. Sage University Paper series on Quantitative Aquantitative Applications in the Social Sciences, 07-001 (ed. by E.M. Uslaner). Sage Publications, Inc., Beverly Hills and London.

Lafferty, K.D. & Kuris, A.M. (1996) Biological control of marine pests. Ecology, 77, 1989-2000.

Lambrinos, J.G. (2004) How interactions between ecology and evolution influence contemporary invasion dynamics. Ecology, 85, 2061-2070.

153 References

Larned, S.T. (2003) Effects of the invasive, nonindigenous seagrass Zostera japonica on nutrient fluxes between the water column and benthos in a NE Pacific estuary. Marine Ecology Progress Series, 254, 69-80.

Le Roux, P.J., Branch, G.M. & Joska, M.A.P. (1990) On the distribution, diet, and possible impact of the invasive European shore crab Carcinus maenas (L.) along the South African coast. South African Journal of Marine Science, 9, 85-93.

Lee, K.S. & Dunton, K.H. (1999) Inorganic nitrogen acquisition in the seagrass Thalassia testudinum: development of a whole plant nitrogen budget. Limnology and Oceanography, 44, 1204-1215.

Lee, S.Y., Fong, S.W. & Wu, R.S.S. (2001) The effects of seagrass (Zostera japonica) canopy structure on associated fauna: a study using artificial seagrass units and sampling of natural beds Journal of Experimental Marine Biology and Ecology, 259, 23-50.

Legendre, P. & Legendre, L. (1998) Numerical ecology, 2nd edn. Elsevier Science B.V., Amsterdam.

Leidy, J. (1855) Contributions towards a knowledge of the marine invertebrates of the coasts of Rhode Island and New Jersey. Journal of the Academy of Natural Sciences of Philadelphia, 3, 135-158.

Lejart, M. & Hily, C. (2011) Differential response of benthic macrofauna to the formation of novel oyster reefs (Crassostrea gigas, Thunberg) on soft and rocky substrate in the intertidal of the Bay of Brest, France. Journal of Sea Research, 65, 84-93.

Leschine, T.M., Ferriss, B.E., Bell, K.P., Bartz, K.K., MacWilliams, S., Pico, M. & Bennett, A.K. (2003) Challenges and strategies for better use of scientific information in the management of coastal estuaries. Estuaries, 26, 1189-1204.

Leslie, H.M. & Mcleod, K.L. (2007) Confronting the challenges of implementing marine ecosystem-based management. Frontiers in Ecology and the Environment, 5, 540- 548.

Lester, S.E., Mcleod, K.L., Tallis, H., Ruckelshaus, M., Halpern, B.S., Levin, P.S., Chavez, F.P., Pomeroy, C., Mccay, B.J. & Costello, C. (2010) Science in support of ecosystem-based management for the US West Coast and beyond. Biological Conservation, 143, 576-587.

154 References

Leung, B. & Mandrak, N. (2007) The risk of establishment of aquatic invasive species: Joining invasibility and propagule pressure. Proceedings of the Royal Society B, 274, 2733-2739.

Leung, B., Lodge, D.M., Finnoff, D., Shogren, J., Lewis, M.A. & Lamberti, G. (2002) An ounce of prevention or a pound of cure: bioeconomic risk analysis of invasive species. Proceedings of the Royal Society B, 269, 2407-2413.

Levin, P.S., Fogarty, M.J., Murawski, S.A. & Fluharty, D.L. (2009) Integrated ecosystem assessments: developing the scientific basis for ecosystem-based management of the ocean. PLoS Biology, 7, 23-28.

Levin, S.A. & Lubchenco, J. (2008) Resilience, robustness and marine ecosystem-based mangement. Bioscience, 58, 27-32.

Levings, C.D. & Coustalin, J.B. (1975) Zonation of intertidal biomass and related benthic data, Sturgeon and Roberts Banks, Fraser River estuary, British Columbia. Fisheries Marine Service Research Devision Technical Report 468. pp. 138.

Levings, C.D., Foreman, R.E. & Tunnicliffe, V.J. (1983) A review of the benthos of the Strait of Georgia and contiguous fjords. Canadian Journal of Fisheries and Aquatic Sciences, 40, 1120-1141.

Levings, C.D., Kieser, D., Jamieson, G.S. & Dudas, S. (2002) Marine and estuarine alien species in the Strait of Georgia, British Columbia. Alien invaders in Canada's waters, wetlands, and forests (ed. by R. Claudi, P. Nantel and E. Muckle-Jeffs), pp. 113-131. Canadian Forest Service, Natural Resoures Canada, Ottawa.

Levy, P.S. & Lemeshow, S. (2008) Sampling of populations: methods and applications. John Wiley & Sons, Inc., Hoboken, New Jersey.

Lindqvist, S., Norling, K. & Hulth, S. (2009) Biogeochemistry in highly reduced mussel farm sediments during macrofaunal recolonization by Amphiura filiformis and Nephtys sp. Marine Environmental Research, 67, 136-145.

Lo, V.B., Levings, C.D. & Chan, K.M.A. (2012) Quantifying potential propagule pressure of aquatic invasive species from the commercial shipping industry in Canada. Marine Pollution Bulletin, 64, 295-302.

Lockwood, J.L., Cassey, P. & Blackburn, T. (2005) The role of propagule pressure in explaining species invasions. Trends in Ecology & Evolution, 20, 223-228.

155 References

Lodge, D.M., Williams, S.L., MacIsaac, H.J., Hayes, K.R., Leung, B., Reichard, S., Mack, R.N., Moyle, P.B., Smith, M., Andow, D.A., Carlton, J.T. & McMichael, A. (2006) Biological invasions: recommendations for U.S. policy and management. Ecological Applications, 16, 2035-2054.

Lovell, S., Besedin, E. & Grosholz, E. (2007) Modeling economic impacts of the European green crab. American Agricultural Economics Association Annual Meeting, Portland, OR. pp. 40.

Lu, L., Levings, C.D. & Piercey, G. (2007) Preliminary investigation on aquatic invasive species of marine and estuarine macrobenthic invertebrates on floating structures in five British Columbia harbours. Canadian Manuscript Report of Fisheries and Aquatic Sciences, 281430.

Mach, M.E., Wyllie-Echeverria, S. & Ward, J.R. (2010) Distribution and potential effects of a non-native seagrass in Washington State: Zostera japonica workshop. Washington State Department of Natural Resources and Washington Sea Grant, Friday Harbor, Washington. pp. 41.

Mach, M.E., Levings, C.D., McDonald, P.S. & Chan, K.M.A. (2012) An Atlantic infaunal engineer is established in the Northeast Pacific: Clymenella torquata (Polychaeta: Maldanidae) on the British Columbia and Washington Coasts. Biological Invasions, 14, 503-507.

MacIsaac, H.J. (1996) Potential abiotic and biotic effects of zebra mussels on inland waters of North America. American Zoologist 36, 287-299.

Mack, R.N., Simberloff, D., Lonsdale, W.M., Evans, H., Clout, M. & Bazzaz, F.A. (2000) Biotic invasions: causes, epidemiology, global consequences, and control. Ecological Applications, 10, 689-710.

MacLeod, C.J., Newson, S.E., Blackwell, G. & Duncan, R.P. (2009) Enhanced niche opportunities: can they explain the success of New Zealand's introduced bird species? Diversity and Distributions, 15, 41-49.

Mangum, C.P. (1964) Studies on speciation in maldanid polychaetes of the North American Atlantic coast. II. Distribution and competitive interaction of five sympatric species. Limnology and Oceanography, 9, 12-26.

Mann, R. & Harding, J.M. (2003) Salinity tolerance of larval Rapana venosa: implications for dispersal and establishment of an invading predatory gastropod on the North American Atlantic coast. Biological Bulletin, 204, 96-103.

156 References

Mansour, R.A. & Licpcius, R.N. (1991) Density-dependent foraging and mutual interference in blue crabs preying upon infaunal clams. Marine Ecological Progress Series, 72, 239-246.

Mantua, N.J. & Hare, S.R. (2002) The Pacific Decadal Oscillation. Journal of Oceanography, 58, 35-44.

Maron, J.L. & Marler, M. (2007) Native plant diversity resists invasion at both low and high resource levels. Ecology, 88, 2651-2661.

Martone, R.G. & Wasson, K. (2008) Impacts and interactions of multiple human perturbations in a California salt marsh. Oecologia, 158, 151-163.

Mascaro, M. & Seed, R. (2001) Foraging behavior of juvenile Carcinus maenas (L.) and Cancer pagurus (L.). Marine Biology, 139, 1135-1145.

Mazerolle, M.J. (2012) AICcmodavg: model selection and multimodel inference based on (Q)AIC(c). R package version 1.24 Available at: http://CRAN.R- project.org/package=AICcmodavg (accessed February 2012).

McKinnon, J.G., Gribben, P.E., Davis, A.R., Jolley, D.F. & Wright, J.T. (2009) Differences in soft-sediment macrobenthic assemblages invaded by Caulerpa taxifolia compared to uninvaded habitats. Marine Ecology Progress Series, 380, 59-71.

McLachlan, C. (2005) The principle of systemic integration and Article 31(3)(c) of the Vienna convention. International and Comparative Law Quarterley, 54

McLeod, K., Lubchenco, J., Palumbi, S.R. & Rosenberg, A.A. (2005) Communication Partnership for Science and the Sea (COMPASS): scientific consensus statement on marine ecosystem-based management. Available at: http://www.compassonline.org/resources (accessed May 2012).

McLeod, K.L. & Leslie, H.M. (2009) Chapter 1: Why ecosystem-based management? Ecosystem-based management for the oceans (ed. by K. Mcleod and H. Leslie). Island Press, Washington, D.C.

Menge, B.A. & Sutherland, J.P. (1987) Community regulation: variation in disturbance, competition, and predation in relation to environmental stress and recruitment. The American Naturalist, 130, 730-757.

Millennium Ecosystem Assessment (2005) Ecosystems and human well-being: synthesis. Island Press, Washington, D.C. pp. 137.

157 References

Miller, A.W., Ruiz, G.M., Minton, M.S. & Ambrose, R.F. (2007) Differentiating successful and failed mollusan invaders in estuarine ecosystems. Marine Ecology Progress Series, 332, 41-51.

Mineur, F., Belsher, T., Johnson, M.P., Maggs, C.A. & Verlaque, M. (2007) Experimental assessment of oyster transfers as a vector for macroalgal introductions. Biological Conservation, 137, 237-247.

Mineur, F., Davies, A.J., Maggs, C.A., Verlaque, M. & Johnson, M. (2010) Fronts, jumps and secondary introducitons suggested as different invasion patterns in marine species, with an increase in spread rates over time. Proceedings of the Royal Society B, 277, 2693-2701.

Miron, G., Audet, D., Landry, T. & Moriyasu, M. (2005) Predation potential of the invasive green crab (Carcinus maenas) and other common predators on commercial bivalve species found on Prince Edward Island. Journal of Shellfish Research, 24, 579-586.

Molnar, J.L., Gamboa, R.L., Revenga, C. & Spalding, M.D. (2008) Assessing the global threat of invasive species to marine biodiversity. Frontiers in Ecology and the Environment, 6, 485-492.

Morales, M. (2011) Sciplot: scientific graphing functions for factorial designs. R package version 1.0-9, Available at: http://CRAN.R-project.org/package=sciplot (accessed April 2012).

Morgan, E.H. & Richardson, C.A. (2012) Capricious bioinvasions versus uncoordinated management strategies: how the most unlikely invaders can prosper under the current UK legislation framework. Aquatic Conservation: Marine and Freshwater Ecosystems, 22, 87-103.

Moyle, P.B. & Light, T. (1996) Fish invasions in California: do abiotic factors determine success? Ecology, 77, 1666-1670.

Munroe, D. & McKinley, R.S. (2007) Effect of predator netting on recruitment and growth of manila clams (Venerupis philippinarum) on soft substrate intertidal plots in British Columbia, Canada. Journal of Shellfish Research, 26, 1065-1044.

Myers, J.H., Simberloff, D., Kuris, A.M. & Carey, J.R. (2011) Eradication revisited: dealing with exotic species. Trends in Ecology & Evolution, 15, 316-320.

National Marine Fisheries Service (2005) Annual commercial landing statistics. Available at:

158 References

http://www.st.nmfs.noaa.gov/st1/commercial/landings/annual_landings.html (accessed April 2012).

National Oceanic and Atmospheric Administration (NOAA) (1990) Estuaries of the United States, statistics of a national resource base. Strategic Assessment Branch, Office of Oceanography and Marine Assessment, Washington, DC. pp. 79.

Neff, M. (2011) What research should be done and why? Four competing visions among ecologists. Frontiers in Ecology and the Environment, 9, 462-469.

Neira, C., Levin, L.A. & Grosholz, E.D. (2005) Benthic macrofaunal communities of three sites in San Francisco Bay invaded by hybrid Spartina, with comparison to uninvaded habitats. Marine Ecology - Progress Series, 292, 111-126.

Nelson, E., Mendoza, G., Regetz, J., Polasky, S., Tallis, H., Cameron, D., Chan, K.M.A., Daily, G.C., Goldstein, J., Kareiva, P.M., Lonsdorf, E., Naidoo, R., Ricketts, T.H. & Shaw, M. (2009) Modeling multiple ecosystem services, biodiversity conservation, commodity production, and tradeoffs at landscape scales. Frontiers in Ecology and the Environment, 7, 4-11.

Nelson, J. & Goodman, T. (2010) Clymenella torquata (Leidy, 1855). Available at: http://bcbiodiversity.lifedesks.org/pages/18062 (accessed October 2010).

Newcombe, C.F. (1891) Report on the marine shells of British Columbia. Paper Communications of the Natural History Society of B.C., 1, 31-72.

Newell, G.E. (1949) Occurance of a species of Clymenella Verrill (Polychaetea, fam. Maldanidae) on the North Kent coast. Nature, 163, 648.

Newell, G.E. (1951) The life-history of Clymenella torquata (Leidy). (Polychaeta.). Journal of Zoology, 121, 561-586.

Newell, R.I.E. & Koch, E.W. (2004) Modeling seagrass density and distribution in response to changes in turbidity stemming from bivalve filtration and seagrass sediment stabilization. Estuaries and Coasts, 27, 793-806.

NOAA Fisheries (2008) Commercial fishery landings. Available at: http://www.st.nmfs.noaa.gov/st1/commercial/index.html (accessed June 2008).

Occhipinti-Ambrogi, A. & Savini, D. (2003) Biological invasions as a component of global change in stressed marine ecosystems. Marine Pollution Bulletin, 46, 542- 551.

159 References

Officer, C.B., Smayda, T.J. & Mann, R. (1982) Benthic filter feeding: a natural eutrophication control. Marine Ecological Progress Series, 9, 203-210.

Oksanen, J., Blanchet, F.G., Kindt, R., Legendre, P., Minchin, P.R., O'Hara, R.B., Simpson, G.L., Solymos, P., Hentry, M., Stevens, H. & Wagner, H. (2012) Vegan: community ecology package. R package version 2.0-3. Available at: http://CRAN.R-project.org/package=vegan (accessed February 2012).

Onbe, T. (1966) Observations on the tubicolous amphipod, Corophium acherusicum, in Fukuyama harbor area. Journal of the Faculty of Fisheries and Animal Husbandry Hiroshima University, 6, 323–338.

Oreskes, N. (2004) Science and public policy: what's proof got to do with it? Environmental Science & Policy, 7, 369-383.

Orth, R.J., Carruthers, T.J.B., Dennison, W.C., Duarte, C.M., Fourqurean, J.W., Heck, K.L., Hughes, A.R., Kendrick, G.A., Kenworthy, W.J., Olyarnik, S., Short, F.T., Waycott, M. & Williams, S.L. (2006) A global crisis for seagrass ecosystems. Bioscience, 56, 987-996.

Pacific Fisheries Information Network (2009) Available at: http://pacfin.psmfc.org/pacfin_pub/data.php (accessed january 2010).

Padilla, D.K. (2010) Context-dependent impacts of non-native ecosystem engineers, the Pacific oyster Crassostrea gigas. Integrative and Comparative Biology, 50, 213- 225.

Pawlak, B.T. & Olson, A.M. (1995) Policies and management practices of Washington State agencies as they pertain to the seagrasses Zostera marina and Zostera japonica: an ecological analysis. Puget Sound Water Quality Authority, Olympia, Washington. pp. 516-522.

Paxton, E.H., Theimer, T.C. & Sogge, M.K. (2011) Tamarisk biocontrol using Tamarisk beetles: potential consequences for riparian birds in the Southwestern United States. The Condor, 113, 255-265.

Pejchar, L. & Mooney, H.A. (2009) Invasive species, ecosystem services and human well-being. Trends in Ecology & Evolution, 24, 497-504.

Pellikaan, G.C. & Nienhuis, P.H. (1988) Nutrient uptake and release during growth and decompostion of eelgrass, Zostera marina L., and its effects on the nutrient dynamics of Lake Grevelingen. Aquatic Botany, 30, 189-214.

160 References

Perrings, C. (2005) Mitigation and adaptation strategies for the control of biological invasions. Ecological Economics, 52, 315-325.

Perrings, C., Williamson, M. & Dalmazzone, S. (2000) The economics of biological invasions. Edward Elgar, Cheltenham, UK.

Piercey, G.E., Levings, C.D., Elfert, M., Galbraith, M. & Waters, R. (2000) Invertebrate fauna in ballast water collected in vessels arriving in British Columbia ports, especially those from the western North Pacific. Can Data Rep Fish Aquat Sci 106050.

Pihl, L. & Rosenberg, R. (1982) Production, abundance, and biomass of mobile epibenthic marine fauna in shallow waters, Western Sweden. Journal of Experimental Marine Biology and Ecology, 57, 273-301.

Pilgrim, M. (1965) The functional anatomy and histology of the alimentary canal of the Maldanid polychaetes Clymenella torquata and Euclymene oerstedi. Journal of Zoology, 147, 387-405.

Pimentel, D., Zuniga, R. & Morrison, D. (2005) Update on the environmental and economic costs associated with alien-invasive species in the United States. Ecological Economics, 52, 273-288.

Pimentel, D., Lach, L., Zuniga, R. & Morrison, D. (2000) Environmental and economic costs of nonindigenous species in the United States. Bioscience, 50, 53-65.

Piola, R.F. & Johnston, E.L. (2007) Pollution reduces native diversity and increases invader dominance in marine hard-substrate communities. Diversity and Distributions, 14, 329-342.

Posey, M.H. (1988) Community changes associated with the spread of an introduced seagrass, Zostera japonica. Ecology, 69, 974-983.

Powers, S.P., Bishop, M.A., Grabowski, J.H. & Peterson, C.H. (2006) Distribution of the invasive bivalve Mya arenaria L. on intertidal flats of southcentral Alaska. Journal of Sea Research, 55, 207-216.

Price, K., Roburn, A. & MacKinnon, A. (2009) Ecosystem-based management in the Great Bear Rainforest. Forest Ecology and Management 258, 495-503.

Puget Sound Action Team (PSAT) (2003) Shellfish economy: treasures of the tidelands. Available at:

161 References

http://www.psparchives.com/publications/our_work/waste/shellfish/fact_sheets/ec onomy_web1.pdf (accessed September 2008).

Puget Sound Partnership (2008a) Action agenda. Available at: http://www.psp.wa.gov/aa_action_agenda.php (accessed May 2012).

Puget Sound Partnership (2008b) Action areas. Available at: http://www.psp.wa.gov/aa_action_areas.php (accessed May 2012).

Puget Sound Partnership (2012) Action area map. Puget Sound Partnership, Olympia, Washington.

Quayle, D. (1964) Distribution of introduced marine mollusca in British Columbia waters. Journal of the Fisheries Research Board of Canada, 21, 1155-1181.

R Development Core Team (2012) A language and environment for statistical computing. R Foundation for Statistical Computing.

Rabalais, N.N., Turner, R.E. & Scavia, D. (2002) Beyond science into policy: Gulf of Mexico hypoxia and the Mississippi River. Bioscience, 52, 129-142.

Rankin, J.S. (1946) Notes on the ecology of the polychaete annelid Clymenella torquata (Leidy), with particular reference to color variation. Ecology, 27, 262-264.

Rasband, W.S. (2009) ImageJ 1.43u. Available at: http://imagej.nih.gov/ij/ (accessed September 2008).

Reiners, W.A. & Driese, K.L. (2001) The propagation of ecological influences through heterogeneous environmental space. Bioscience, 51, 939-950.

Reinhardt, F., Herle, M., Bastiansen, F. & Streit, B. (2003) Okonomische Folgen der Ausbreitung von gebietsfremden Organismen in Deutschland. Umweltbundesamt, Berlin. pp.

Reusch, T.B.H. & Williams, S.L. (1998) Variable responses of native eelgras Zostera marina to non-indigenous bivalve Musculista senhousia. Oecologia, 113, 428- 441.

Ricciardi, A. & Rasmussen, J.B. (1998) Predicting the identity and impact of future biological invaders: a priority for aquatic resource management. Canadian Journal of Fisheries and Aquatic Sciences, 55, 1759-1765.

162 References

Ricciardi, A. & Bourget, E. (1998) Weight-to-weight conversion factors for marine benthic macroinvertebrates. Marine Ecological Progress Series, 163, 245-251.

Ricketts, E.F., Calvin, J., Hedgpeth, J.W. & Phillips, D.W. (1982) Between Pacific Tides, 5th edn. Stanford University Press, Stanford.

Rilov, G. & Crooks, J.A. (2009) Marine bioinvasions: conservation hazards and vehicles for ecological understanding. Biological Invasions in Marine Ecosystems (ed. by G. Rilov and J.A. Crooks). Springer-Verlag, Berlin

Rogers, R. (2007) Bamboo worm workshop. Washington Department of Fish and WIldlife and the Pacific Shellfish Institute, Mount Vernon, Washington. pp.

Roman, J. (2006) Diluting the founder effect: cryptic invasions expand a marine invader's range. 2453-2459.

Rounsevell, M.D.A., Dawson, T.P. & Harrison, P.A. (2010) A conceptual framework to assess the effects of environmental change on ecosystem services. Biodiversity and Conservation, 19, 2823-2842.

Ruckelshaus, M., Essington, T. & Levin, P. (2009) Chapter 12: Puget Sound, Washington, USA. Ecosystem-based management for the oceans (ed. by K. Mcleod and H. Leslie). Island Press, Washington, D.C.

Ruesink, J.L., Feist, B.E., Harvey, C.J., Hong, J.S., Trimble, A.C. & Wisehart, L.M. (2006) Changes in productivity associated with four introduced species: ecosystem transformation of a 'pristine' estuary. Marine Ecology Progress Series, 311, 203-215.

Ruesink, J.L., Lenihan, H.S., Trimble, A.C., Heiman, K.W., Micheli, F., Byers, J.E. & Kay, M.C. (2005) Introduction of non-native oysters: ecosystem effects and restoration implications. Annual Review of Ecology Evolution and Systematics, 36, 643-689.

Ruiz, G., Carlton, J.T., Grosholz, E. & Hines, A.H. (1997) Global invasions of marine and estuarine habitats by non-indigenous species: mechanisms, extent, and consequences. American Zoologist, 37, 621.

Ruiz, G., Fofonoff, P., Steves, B. & Dahlstrom, A. (2011a) Chapter 6: Marine crustacean invasions in North America: a synthesis of historical records and documented impacts. In the wrong place - alien marine crustaceans: distribution, biology and impacts (ed. by B.S. Galil, P.F. Clark and J.T. Carlton), pp. 215-250. Springer Netherlands, Dordrecht.

163 References

Ruiz, G.M. & Hewitt, C.L. (2002) Toward understanding patterns of coastal marine invasions: a prospectus. Invasive aquatic species of Europe (ed. by E. Leppakoski, S. Oleniin and S. Gollasch), pp. 529-547. Kluwer Academic Publishers, Dordrect.

Ruiz, G.M., Fofonoff, P.W., Carlton, J.T., Wonham, M.J. & Hines, A.H. (2000a) Invasion of coastal marine communities in North America: apparent patterns, processes, and biases. Annual Review of Ecology and Systematics, 31, 481-531.

Ruiz, G.M., Fofonoff, P.W., Carlton, J.T., Wonham, M.J. & Hines, A.H. (2000b) Invasion of coastal marine communities in North America: apparent patterns, processes, and biases. Annual Reviews in Ecology and Systematics, 31, 481-531.

Ruiz, G.M., Fofonoff, P.W., Steves, B., Foss, S.F. & Shiba, S.N. (2011b) Marine invasion history and vector analysis of California: a hotspot for western North America. Diversity and Distributions, 17, 362-373.

Ruttenberg, B. & Granek, E. (2011) Bridging the marine–terrestrial disconnect to improve marine coastal zone science and management. Marine Ecology Progress Series, 434, 203-212.

Sale, P.F., Cowen, R.K., Danilowicz, B.S., Jones, G.P., Kritzer, J.P., Linderman, K.C., Planes, S., Polunih, N.V.C., Russ, G.R., Sadovy, Y.J. & Steneck, R.S. (2005) Critical science gaps impede use of no-take fishery reserves. Trends in Ecology & Evolution, 20, 74-80.

Salomon, A.K., Gaichas, S.K., Shears, N.T., Smith, J.E., Madin, E.M.P. & Gaines, S.D. (2010) Key features and context-dependence of fishery-induced trophic cascades. Conservation Biology, 24, 382-394.

Sanders, H.L., Goudsmit, E.M., Mills, E.L. & Hampson, G.E. (1962) A study of the intertidal fauna of Barnstable Harbor, Massachusetts. Limnology and Oceanography, 7, 63-79.

Schlaepfer, M.A., Sax, D.F. & Olden, J.D. (2011) The potential conservation value of non-native species. Conservation Biology, 25, 428-437.

Schlosser, S. (2011) Humboldt Bay cooperative eelgrass project: dwarf eelgrass monitoring and eradication. Final report to the National Fish and Wildlife Foundation6.

164 References

Semmens, B.X. (2008) Acoustically derived fine-scale behaviors of juvenile Chinook salmon (Oncorhynchus tshawytscha) associated with intertidal benthic habitats in an estuary. Canadian Journal of Fisheries and Aquatic Sciences, 65, 2053-2062.

Seymour, N.R., Miller, A.G. & Garbary, D.J. (2002) Decline of Canada geese (Branta canadensis) and common goldeneye (Bucephala clangula) associated with a collapse of eelgrass (Zostera marina) in a Nova Scotia estuary. Helgoland Marine Research, 56, 198-202.

Shafer, D.J. (2007) Physiological factors affecting the distribution of the non-indigenous seagrass Zostera japonica along the Pacific coast of North America. Ph.D Dissertation, University of South Alabama, Mobile, AL.

Shafer, D.J., Wyllie-Echeverria, S. & Sherman, T.D. (2008) The potential role of climate in the distribution and zonation of the introduced seagrass Zostera japonica in North America. Aquatic Botany, 89, 297-302.

Shea, K. & Chesson, P. (2002) Community ecology theory as a framework for biological invasions. Trends in Ecology & Evolution, 17, 170-176.

Shogren, J.F. (2000) Risk reduction strategies against the ‘explosive invader'. The Economics of Biological Invasions (ed. by C. Perrings, M. Williamson and D. S.), pp. 56-69. Edward Elgar, Cheltenham.

Short, F.T. & Wyllie-Echeverria, S. (1996) Natural and human-induced disturbance of seagrasses. Environmental Conservation, 23, 17-27.

Short, F.T. & Coles, R.G. (2001) Global seagrass research methods. Elsevier Science, Amsterdam, The Netherlands.

Simberloff, D. (2006) Invasional meltdown 6 years later: important phenomenon, unfortunate metaphor, or both? Ecology Letters, 9, 912-919.

Simenstad, C.A. (1994) Faunal associations and ecological interactions in seagrass communities of the Pacific Northwest coast. Seagrass science and policy in the Pacific Northwest: Proceedings of a seminar series (SMA 94-1), EPA 910/R-94- 004 (ed. by S. Wyllie-Echeverria, A.M. Olson and M.J. Hershman), pp. 11-18.

Simenstad, C.A., Jay, D.A. & Sherwood, C.R. (1992) Impacts of watershed management on land-margin ecosystems: the Columbia River estuary. Watershed management: balancing sustainability and evironmental change (ed. by R.J. Naiman), p. 542. Springer, New York.

165 References

Sloan, N.A. & Bartier, P.M. (2004) Introduced marine species in the Haida Gwaii (Queen Charlotte Islands) region, British Columbia. Canadian Field-Naturalist, 118, 77- 84.

Snyder, W.E. & Evans, E.W. (2006) Ecological effects of invasive arthropod generalist predators. Annual Review of Ecology Evolution and Systematics, 37, 95-122.

Sokal, R.R. & Rohlf, F.J. (1995) Biometry: the principals and practice of statistics in biological research, 3rd edn. W.H. Freeman and Company, New York.

Spalding, M., Fox, H.E., Halpern, B.S., McManus, M.A., Molnar, J., Allen, G.R., Davidson, N., Jorge, Z.A., Lombana, A.L., Lourie, S.A., Martin, K.D., McManus, E., Recchia, C.A. & Robertson, J. (2007) Marine ecoregions of the world: A bioregionalization of coastal and shelf areas. Bioscience, 57, 573-583.

Stachowicz, J., Whitlatch, R. & Osman, R. (1999) Species diversity and invasion resistance in a marine ecosystem. Science, 286, 1577.

Stachowicz, J.J., Fried, H., Osman, R.W. & Whitlatch, R.B. (2002a) Biodiversity, invasion resistance, and marine ecosystem function: reconciling pattern and process. Ecology, 83, 2575-2590.

Stachowicz, J.J., Terwin, J.R., Whitlatch, R.B. & Osman, R.W. (2002b) Linking climate change and biological invasions: ocean warming facilitates nonindigenous species invasions. Proceedings of the National Academy of Sciences of the United States of America, 99, 15497-15500.

Statistics Canada (2005) Provincial GDP by Industry and Sector, Cat. No. 15-209-XCB.

Stenseth, N.C., Ottersen, G., Hurrell, J.W., Mysterud, A., Lima, M., Chan, K.S., Yoccoz, N.G. & Adlandsvik, B. (2003) Studying climate effects on ecology through the use of climate indices: the North Atlantic Oscillation, El Niño Southern Oscillation and beyond. Proceedings of the Royal Society B, 270, 2087-2096.

Stokes, D.E. (1997) Pasteur's quadrant: basic science and technological innovation. The Brookings Institution, Washington D.C.

Stoms, D.M., Davis, F.W., Andelman, S.J., Carr, M.H., Gaines, S.D., Halpern, B.S., Hoenicke, R., Leibowitz, S.G., Leydecker, A., Madin, E.M., Tallis, H. & Warner, R.R. (2005) Integrated coastal reserve planning: making the land-sea connection. Frontiers in Ecology and the Environment, 3, 429-436.

166 References

Swinbanks, D.D. & Murray, J.W. (1981) Biosedimentological zonation of Boundary Bay tidal flats, Fraser River Delta, British Columbia. Sedimentology, 28, 201-237.

Sylvester, F., Kalaci, O., Leung, B., Lacoursiere-Roussel, A., Clarke Murray, C., Choi, F.M., Bravo, M.A., Therriault, T.w. & MacIsaac, H.J. (2011) Hull fouling as an invasion vector: can simple models explain a complex problem? Journal of Applied Ecology, 48, 415-423.

Tallis, H., Ferdana, Z. & Gray, E. (2008) Linking terrestrial and marine conservation planning and threats analysis. Conservation Biology, 22, 120-133.

Tallis, H., Levin, P.S., Ruckelshaus, M., Lester, S.E., Mcleod, K.L., Fluharty, D.L. & Halpern, B.S. (2010) The many faces of ecosystem-based management: making the process work today in real places. Marine Policy, 34, 340-348.

Tallis, H., Lester, S.E., Ruckelshaus, M., Plummer, M., McLeod, K., Guerry, A., Andelman, S., Caldwell, M.R., Conte, M., Copps, S., Fox, D., Fujita, R., Gaines, S.D., Gelfenbaum, G., Gold, B., Kareiva, P., Kim, C., Lee, K., Papenfus, M., Redman, S., Silliman, B., Wainger, L. & White, C. (2012) New metrics for managing and sustaining the ocean's bounty. Marine Policy, 36, 303-306.

Talman, S., Bite, S.J., Holloway, M., McArthur, M., Ross, D.J. & Storey, M. (1999) Impacts of some introduced marine species found in Port Phillip Bay. Marine biological invasions of Port Phillip Bay, Victoria, center for research on introduced marine pests (ed. by C.L. Hewitt, M.L. Campbell, R.E. Thresher and R.B. Martin). Commonwealth Scientific and Industrial Research Organisation (CSIRO) Marine Research, Hobart, Tasmania.

Taylor, C.M. & Hastings, A. (2004) Finding optimal control strategies for invasive species: a density-structured model for Sparina alterniflora. Journal of Applied Ecology, 41, 1049-1057.

Thayer, G.W. & Philips, R.C. (1977) Importance of eelgrass beds in Puget Sound. Marine Fisheries Review Paper 1271, 39, 18-22.

Thom, R.M., Miller, B. & Kennedy, M. (1995) Temporal patterns of grazers and vegetation in a temperate seagrass system. Aquatic Botany, 50, 201-205.

Thom, R.M., Blanton, S.L., Woodruff, D.l., Wiliams, G.D. & Borde, A.B. (2001) Carbon sinks in nearshore marine vegetated ecosystems. First National Conference on Carbon Sequestration, p. 20. Washington D.C.

167 References

Thompson, A.R. & Rueggeberg, H. (1989) Science, inference, and the law: the ultimate fish story. Proceedings of the National Workshop on Effects of Habitat Alteration on Salmonid Stocks (ed. by C.D. Levings, L.B. Holtby and M.A. Henderson), pp. 164-172. Canadian Special Publication in Fisheries and Aquatic Sciences 105.

Thompson, R.E. (1981) Oceanography of the British Columbia coast. Canadian Special Publication of Fisheries and Aquatic Sciences, 56291.

Thomsen, M.S., Wernberg, T., Silliman, B.R. & Tuya, F. (2009) Evidence for impacts of non-indigenous macroalgae: a meta-analysis of experimental field studies. Journal of Phycology, 45, 812-819.

Thomsen, M.S., Wernberg, T., Olden, J.D., Griffin, J.N. & Silliman, B.R. (2011) A framework to study the context-dependent impacts of marine invasions. Journal of Experimental Marine Biology And Ecology, 400, 322-327.

Tilman, D. (1997) Community invisibility, recruitment limitation, and grassland biodiversity. Ecology, 78

Transport Canada (2006) Ballast water control and management regulations. Available at: http://laws.justice.gc.ca/en/SOR-2006-129/. (accessed May 2012).

Tremblay, M.J., Thompson, A. & Paul, K. (2006) Recent trends in the abundance of the invasive green crab (Carcinus maenas) in Bras d’Or Lakes and Eastern Nova Scotia based on trap surveys. Canadian Technical Report of Fisheries and Aquatic Sciences 2673, Fisheries and Oceans Canada, Dartmouth, NS. pp. 32.

Tsai, C., Yang, S., Trimble, A.C. & Ruesink, J.L. (2010) Interactions between two introduced species: Zostera japonica (dwarf eelgrass) facilitates itself and reduces condition of Ruditapes philippinarum (Manila clam) on intertidal flats. Marine Biology, 157, 1929-1936.

Turnbull, L. (2009) Changes to nutrient and carbon cycling, soil properties, and ecosystem processes by the invasive plants Phalaris arundinacea and Zostera japonica. Ph.D. Thesis. University of Oregon, Eugene, Oregon.

Valiela, I., Bowen, J.L. & York, J.K. (2001) Mangrove forests, one of the world's threatend major tropical environments. Bioscience, 51, 807-815. van Asselt, M. & Rotmans, J. (2002) Uncertainty in integrated assessment modelling - from positivism to pluralis. Climatic Change, 54, 75-105.

168 References

Vermonden, K., Leuven, R.S.E.W. & van der Velde, G. (2010) Environmental factors determining invasibility of urban waters for exotic macroinvertebrates. Diversity and Distributions, 16, 1009-1021.

Vila, M., Espinar, J.L., Hejda, M., Hulme, P.E., Jorosik, V., Maron, J.L., Pergl, J., Schaffner, U., Sun, Y. & Pysek, P. (2011) Ecological impacts of invasive alien plants: a meta-analysis of their effects on species, communities and ecosystems. Ecology Letters, 14, 702-708.

Walton, W.C., MacKinnon, C., Rodriguez, L.F., Proctor, C. & Ruiz, G.A. (2002) Effect of an invasive crab upon a marine fishery: green crab, Carcinus maenas, predation upon a venerid clam, Katelysia scalarina, in Tasmania (Australia). Journal of Experimental Marine Biology and Ecology, 272, 171-189.

Washington Department of Fish and Wldlife (WDFW) (2012a) Carcinus maenas (European green crab). Available at: http://wdfw.wa.gov/ais/carcinus_maenas/ (accessed June 2012).

Washington Department of Fish and Wldlife (WDFW) (2012b) Ballast water program. Available at: http://wdfw.wa.gov/ais/ballast/ (accessed June 2012).

Washington State Department of Health (2010) Annual inventory growing areas map. Available at: http://www.doh.wa.gov/CommunityandEnvironment/Shellfish.aspx. (accessed 2012).

Washington State Noxious Weed Control Board (2011) Zostera japonica class C noxious weed listing on commercially leased lands in Washington State, November 2011. Available at: http://www.nwcb.wa.gov/detail.asp?weed=173 (accessed December 2011).

Waycott, M., Duarte, C.M., Carruthers, T.J.B., Orth, R.J., Dennison, W.C., Olyarnik, S., Calladine, A., Fourqurean, J.W., Heck, K.L., Hughes, A.R., Kendrick, G.A., Kenworthy, W.J., Short, F.T. & Williams, S.L. (2009) Accelerating loss of seagrasses across the globe threatens coastal ecosystems. Proceedings of the National Academy of Sciences, 106, 12377-12381.

Weimin, Q., Xinqiang, S., Minbo, L. & Yaqu, C. (2006) Ecological function and restoration measures of oyster reef in estuaries. Shengtaixue Zazhi, 25, 1234- 1239.

Wendt, D.E., Pendletton, L. & Maruska, D. (2009) Chapter 11: Morro Bay, California, USA. Ecosystem-based management for the oceans (ed. by K. Mcleod and H. Leslie). Island Press, Washington, D.C.

169 References

West, J.M., Williams, G.D., Madon, S.P. & Zedler, J.B. (2003) Integrating spatial and temporal variability into the analysis of fish food web linkages in Tijuana Estuary. Environmental Biology of Fishes, 67, 297–309.

Whitcraft, C.R., Levin, L.A., Talley, D. & Crooks, J.A. (2008) Utilization of invasive tamarisk by salt marsh consumers. Oecologia, 158, 259–272.

White, C., Halpern, B.S. & Kappel, C.V. (2012) Ecosystem service tradeoff analysis reveals the value of marine spatial planning for multiple ocean uses. Proceedings of the National Academy of Sciences of the United States of America, 109, 4696- 4701.

White, L. & Orr, L. (2011) Native clams facilitate invasive species in an eelgrass bed. Marine Ecology Progress Series, 424, 87-95.

Wilcove, D.S., Rothstein, D., Dubow, J., Phillips, A. & Losos, E. (1998) Quantifying threats to imperiled species in the United States. BioScience, 48, 607-615.

Williams, S.L. (2007) Introduced species in seagrass ecosystems: status and concerns. Journal of Experimental Marine Biology and Ecology, 350, 89-110.

Williams, S.L. & Heck, K.L. (2001) Seagrass communities. Marine community ecology (ed. by M.D. Bertness, S.D. Gaines and M.E. Hay), pp. 317-337. Sinauer Associates, Sunderland, MA.

Williams, S.L. & Grosholz, E.D. (2008) The invasive species challenge in estuarine and coastal environments: marrying management and science. Estuaries and Coasts, 31, 3-20.

Williamson, M. & Fitter, A. (1996) The varying success of invaders. Ecology, 77, 1661- 1666.

Wolff, W.J. & Reise, K. (2002) Oyster imports as a vector for the introduction of alien species into northern and western European coastal waters. Invasive aquatic species of Europe: distribution, impacts, and management, 152 pp. (ed. by E. Leppakoski, S. Gollasch and S. Olenin), pp. 193-205. Kluwer Academic Publishers, Netherlands.

Wonham, M.J. & Carlton, J.T. (2005) Trends in marine biological invasions at local and regional scales: The Northeast Pacific Ocean as a model system. Biological Invasions, 7, 369-392.

170 References

Wonham, M.J. & Pachepsky, E. (2006) A null model of temporal trends in biological invasion records. Ecology Letters, 9, 663-672.

Wonham, M.J., O'Connor, M. & Harley, C.D.G. (2005) Positive effects of a dominant invader on introduced and native mudflat species. Marine Ecology-Progress Series, 289, 109-116.

Wonham, M.J., Walton, W.C., Ruiz, G.M., Frese, A.M. & Galil, B.S. (2001) Going to the source: role of the invasion pathway in determining potential invaders. Marine Ecology Progress Series, 215, 1-12.

World Trade Organization (WTO) (1995) Agreement on the application of sanitary and phytosanitary measures. Available at: http://www.wto.org/english/docs_e/legal_e/15sps_01_e.htm (accessed August 2012).

Worm, B., Barbier, E.B., Beaumont, N., Duffy, J.E., Faolke, C., Halpern, B.S., HJackson, J.B.C., Lotze, H.K., Micheli, F., Palumbi, S.R., Sala, E., Selkoe, K.A., Stachowicz, J.J. & Watson, R. (2006) Impacts of biodiversity loss on ocean ecosystem services. Science, 314, 787-790.

Wyatt, A.S.J., Hewitt, C.L., Walker, D.I. & Ward, T.J. (2005) Marine introductions in the Shark Bay World Heritage Property, Western Australia: a preliminary assessment. Diversity and Distributions, 11, 33-44.

Wyllie-Echeverria, S. & Ackerman, J.D. (2003) The seagrasses of the Pacific coast of North America. World atlas of seagrasses (ed. by E.P. Green and F.T. Short), pp. 199-206. University of California Press, Berkeley and Los Angeles, California.

Yamada, S.B. & Randall, A. (2006) PSMFC: Status of European Green Crab in OR and WA Estuaries. REPORT, 38.

Yamada, S.B. & Gillespie, G.E. (2008) Will the European green crab (Carcinus maenas) persist in the Pacific Northwest? ICES Journal of Marine Science, 65, 725-729.

Yamada, S.B., Davidson, T.M. & Fisher, S. (2010) Claw morphology and feeding rates of introduced European green crabs (Carcinus maenas L, 1758) and native Dungeness crabs (Cancer magister Dana, 1852). Journal of Shellfish Research, 29, 471-477.

Yamada, S.B., Dumbauld, B.R., Kalin, A., Hunt, C.E., Figlar-Barnes, R. & Randall, A. (2005) Growth and persistence of a recent invader Carcinus maenas in estuaries of the northeastern Pacific. Biological Invasions, 7, 309-321.

171 Appendix A

Appendices Appendix A Species accumulation curves Total species accumulation curves of total species richness for core, dredge and trap samples at each site from data collected in Chapter 3. Sites listed from in order from northern to southern latitudes. 40 E Kaien Island Prince Rupert

30 Campbell River Mud Bay Port Alberni

20 Stanley Park Tsawwassen Nanaimo

10 Cowichan Bay Species Richness Species Fourth of July Esquimalt 0

1 2 3 4 5 6 7 Core Sample 40 30 20 10 Species Richness Species 0

1 2 3 4 5 6 7 Dredge Sample 6 5 4 3 2 Species Richness Species 1 0

1 2 3 4 5 6 7 8 Trap Sample

172 Appendix B

Appendix B Species abundances in core, dredge and trap eelgrass bed samples The abundance and standard deviation of native (A), indeterminate (B), cryptogenic (C), and nonnative (D) species sampled using infaunal cores (Table C.1), epifaunal dredges (Table C.2), and traps sampling mobile macrofauna (Table C.3), as well as a visual survey for presence or absence of Crassostrea gigas and Sargassum muticum (Table C.4), in British Columbia eelgrass beds at each site (listed from south to north, acronyms of site names in Table 3.1 legend) the species was found. Table C.1. Core samples (1 of 5)

No. Species Sites EL FJ CB Ts PA NE SP MB CR PR EKI A. Native Annelida Polychaeta Ampharete labrops 3 0.17 ± 0.41 1.17 ± 1.83 1 ± 0.89 Armandia brevis 1 0.17 ± 0.41 Dipolydora quadrilobata 1 0.33 ± 0.82 Eteone californica 2 0.33 ± 0.82 0.17 ± 0.41 Glycera americana 1 0.17 ± 0.41 Glycera nana 1 0.33 ± 0.52 Glycinde picta 4 0.67 ± 0.52 0.17 ± 0.41 0.5 ± 0.84 1.17 ± 1.6 Hemipodia simplex 4 1.83 ± 1.17 0.17 ± 0.41 0.17 ± 0.41 0.33 ± 0.52 Leitoscoloplos pugettenis 7 0.17 ± 0.41 1.33 ± 1.03 0.5 ± 0.55 0.67 ± 0.52 0.33 ± 0.82 8.5 ± 8.46 3.33 ± 1.03 Lepidonotus squamatus 2 1.17 ± 2.04 0.33 ± 0.52 Lysippe labiata 1 5.57 ± 13.88 Magelona hobsonae 1 0.17 ± 0.41 Mediomastus californiensis 1 0.5 ± 0.84 Neanthes brandti 1 0.5 ± 0.55 Nephtys caeca 2 0.17 ± 0.41 0.17 ± 0.41 Nephtys californiensis 3 0.17 ± 0.41 0.33 ± 0.52 0.5 ± 0.55 Nereis procera 2 0.67 ± 0.52 0.17 ± 0.41 Notomastus tenuis 4 4 ± 5.48 6.67 ± 4.23 14.17 ± 8.59 2.17 ± 3.25 Odontosyllis phosphorea 1 0.17 ± 0.41 Onuphis elegans 1 0.33 ± 0.52 Ophelina acuminata 1 1.33 ± 1.03 Ophiodromus pugettensis 1 0.5 ± 0.55 Owenia collaris 5 0.5 ± 0.84 0.5 ± 0.55 4.83 ± 2.04 2.83 ± 2.23 0.33 ± 0.52 Owenia johnsoni 5 0.17 ± 0.41 0.83 ± 1.17 0.33 ± 0.82 0.17 ± 0.41 0.33 ± 0.82 Pectinaria granulata 1 0.5 ± 0.84 Pholoe glabra 2 0.33 ± 0.52 0.33 ± 0.82 Site abbreviations: Esquimalt Lagoon (EL), Fourth of July (FJ), Cowichan Bay (CB), Tsawwassen (Ts), Port Alberni (PA), Nanaimo Estuary (NE), north of Stanley Park (SP), Mud Bay (MB), Campbell River (CR), Prince Rupert (PR), East Kaien Island (EKI).

173 Appendix B

Table C.1. Core samples continued (2 of 5)

No. Species Sites EL FJ CB Ts PA NE SP MB CR PR EKI Phyllodoce medipapillata 1 0.17 ± 0.41 Platynereis bicanaliculata 7 0.33 ± 0.52 0.5 ± 0.84 0.67 ± 1.03 0.33 ± 0.52 0.86 ± 1.07 0.33 ± 0.82 1 ± 1.1 Spio filicornis 1 0.17 ± 0.41 Syllides longocirrata 1 0.17 ± 0.41 Arthropoda Malacostraca Americorophium salmonis 3 0.17 ± 0.41 0.67 ± 1.63 3.67 ± 3.98 Americorophium spinicorne 1 0.14 ± 0.38 Anisogammarus pugettensis 2 1 ± 1.55 0.17 ± 0.41 Aoroides inermis 1 0.33 ± 0.82 Caprella laeviuscula 2 1.83 ± 3.54 0.17 ± 0.41 Eobrolgus chumashi 1 0.33 ± 0.82 Eogammarus confervicolus 2 0.17 ± 0.41 0.17 ± 0.41 Gnorimosphaeroma oregonense 1 1.71 ± 2.87 Hemigrapsus oregonensis 1 0.33 ± 0.52 Idotea resecata 1 0.17 ± 0.41 Lamprops triserratus 1 0.17 ± 0.41 Neotrypaea californiensis 3 0.5 ± 0.84 0.33 ± 0.52 0.17 ± 0.41 Orchomene minutus 1 2.5 ± 2.07 Pagurus armatus 1 0.17 ± 0.41 Paracalliopiella pratti 1 0.17 ± 0.41 Photis brevipes 2 0.17 ± 0.41 1.5 ± 1.64 Pontogeneia rostrata 4 1.5 ± 3.67 0.67 ± 0.52 0.67 ± 0.82 0.17 ± 0.41 Protohyale frequens 1 0.17 ± 0.41 Rhepoxynius pallidus 1 1.17 ± 0.98 Scleroplax granulata 2 0.17 ± 0.41 0.17 ± 0.41 Synidotea consolidata 1 0.5 ± 0.84 Maxillopoda Balanus glandula 2 0.14 ± 0.38 0.17 ± 0.41 Chthamalus dalli 1 2.17 ± 4.83 Leimia vaga 1 0.17 ± 0.41 Chordata Actinopterygii Pholis ornata 1 0.86 ± 0.9 Echinodermata Echinoidea Strongylocentrotus sp. 1 0.17 ± 0.41 Ophiuroidea Amphiodia occidentalis 2 0.17 ± 0.41 0.33 ± 0.52

174 Appendix B

Table C.1. Core samples continued (3 of 5)

No. Species Sites EL FJ CB Ts PA NE SP MB CR PR EKI Mollusca Bivalvia Axinopsida serricata 1 0.5 ± 0.84 Clinocardium nuttalli 3 0.17 ± 0.41 0.33 ± 0.82 1 ± 1.55 Epilucina californica 3 0.33 ± 0.82 0.83 ± 1.6 2.17 ± 3.06 Leukoma staminea 2 0.17 ± 0.41 0.17 ± 0.41 Liocyma fluctuosum 1 0.5 ± 1.22 Macoma balthica 7 0.83 ± 1.6 49.71 ± 36.51 1.33 ± 1.63 0.5 ± 0.84 0.17 ± 0.41 0.67 ± 0.82 0.17 ± 0.41 Macoma inquinata 3 0.33 ± 0.52 0.33 ± 0.82 0.67 ± 0.82 Macoma nasuta 9 42 ± 54.05 0.83 ± 1.33 2 ± 1.79 1 ± 1.1 1.5 ± 1.52 0.33 ± 0.82 1.83 ± 1.83 1.33 ± 1.75 0.17 ± 0.41 Nutricola tantilla 2 0.17 ± 0.41 1.5 ± 2.07 Parvilucina tenuisculpta 1 0.17 ± 0.41 Rochefortia tumida 1 0.67 ± 1.63 Saxidomus gigantea 1 0.67 ± 1.03 Turtonia minuta 1 1.17 ± 1.94 Gastropoda Alia carinata 1 0.17 ± 0.41 Fartulum orcutti 1 0.33 ± 0.82 Haminoea vesicula 2 0.33 ± 0.52 0.33 ± 0.52 Iselica ovoidea 1 0.17 ± 0.41 Lacuna marmorata 2 0.17 ± 0.41 0.33 ± 0.52 Lacuna sp. 1 0.33 ± 0.52 Lacuna unifasciata 5 3.83 ± 4.54 0.17 ± 0.41 0.33 ± 0.52 0.33 ± 0.52 0.33 ± 0.52 Lacuna variegata 1 0.67 ± 1.03 Littorina sitkana 1 0.17 ± 0.41 Nemertea Anopia Carinoma mutabilis 1 0.5 ± 0.84 Phoronida Phoronis psammophila 1 1 ± 1.55

175 Appendix B

Table C.1. Core samples continued (4 of 5) No. Species Sites EL FJ CB Ts PA NE SP MB CR PR EKI B. Indeterminate Annelida Oligochaete Naididae (Family) 2 6.83 ± 11.92 2.33 ± 2.25 Tectidrilus sp. 1 0.67 ± 1.63 Polychaete Aphelochaeta sp. 2 0.83 ± 1.33 0.17 ± 0.41 Lumbrineris sp. 1 0.33 ± 0.52 Owenia sp. 1 0.33 ± 0.82 Arthropoda Malacostraca Ischyrocerus sp. 1 0.33 ± 0.82 Maxillopoda Cyclopoida (Order) 1 0.17 ± 0.41 Poecilostomatoida (Order) 1 0.17 ± 0.41 Mollusca Bivalvia Macoma sp. 6 0.33 ± 0.52 8.33 ± 4.59 0.17 ± 0.41 0.5 ± 1.22 1.67 ± 1.97 3.17 ± 4.36 Mytilus sp. complex 4 0.83 ± 0.98 6 ± 9.24 0.17 ± 0.41 0.17 ± 0.41 Gastropoda Doridacea (Order) 1 0.17 ± 0.41 Nemertea Anopia Cerebratulus sp. 1 0.17 ± 0.41

C. Cryptogenic Annelida Polychaete Capitella capitata complex 1 1 ± 1.67 Eumida sanguinea 1 0.17 ± 0.41 Harmothoe imbricata 1 0.5 ± 0.84 Prionospio steenstrupi 1 0.33 ± 0.82 Spiophanes bombyx 1 0.33 ± 0.52

176 Appendix B

Table C.1. Core samples continued (5 of 5) No. Species Sites EL FJ CB Ts PA NE SP MB CR PR EKI Arthropoda Malacostraca Leptochelia dubia 6 0.17 ± 0.41 4.5 ± 7.29 0.17 ± 0.41 2.83 ± 3.25 2.17 ± 5.31 8.33 ± 5.89 Maxillopoda Harpacticus uniremis group 1 0.17 ± 0.41

D. Exotic Arthropoda Malacostraca Ampithoe valida 3 0.17 ± 0.41 0.17 ± 0.41 0.17 ± 0.41 Melita nitida 1 1.57 ± 3.31 Monocorophium acherusicum 1 0.17 ± 0.41 Monocorophium insidiosum 1 0.5 ± 0.84 Sinelobus sp. 1 0.14 ± 0.38 Annelida Polychaete Clymenella torquata 2 2.83 ± 2.32 0.5 ± 0.55 Mollusca Bivalvia Mya arenaria 3 0.17 ± 0.41 0.43 ± 0.79 2.67 ± 5.13 Venerupis philippinarum 1 0.33 ± 0.52 Gastropoda Batillaria attramentaria 1 1.33 ± 1.63

177 Appendix B

Table C.2. Dredge samples (1 of 5) No. Species Sites EL FJ CB Ts PA NE SP MB CR PR EKI A. Native Annelida Polychaeta Ampharete labrops 1 0.2 ± 0.45 Armandia brevis 2 1.29 ± 2.63 0.2 ± 0.45 Eteone californica 1 0.43 ± 1.13 Leitoscoloplos pugettenis 2 2 ± 1.73 0.6 ± 1.34 Lepidonotus squamatus 1 1 ± 1.22 Nephtys cornuta 1 0.6 ± 0.89 Ophiodromus pugettensis 3 47.2 ± 34.36 0.2 ± 0.45 0.4 ± 0.89 Pectinaria granulata 1 0.43 ± 0.79 Platynereis bicanaliculata 4 0.2 ± 0.45 0.2 ± 0.45 1 ± 0.71 0.8 ± 0.84 Arthropoda Malacostraca Allorchestes angusta 1 12.6 ± 12.78 Americorophium salmonis 1 0.8 ± 0.84 Americorophium spinicorne 1 5.4 ± 7.37 Anisogammarus pugettensis 6 2.2 ± 2.28 141.4 ± 83.93 35.2 ± 42.76 3.8 ± 2.77 1.33 ± 2.31 19.6 ± 16.2 Aoroides inermis 3 21.2 ± 14.5 3 ± 2.83 2.8 ± 6.26 Calliopius carinatus 3 3.2 ± 2.05 0.2 ± 0.45 0.33 ± 0.82 Caprella californica 1 0.2 ± 0.45 Caprella kennerlyi 3 6.8 ± 10.89 0.6 ± 0.89 0.2 ± 0.45 Caprella laeviuscula 9 463.8 ± 366.27 259.2 ± 31.98 0.43 ± 0.79 149 ± 148.48 1.6 ± 2.51 1.4 ± 1.34 4.5 ± 4.68 8 ± 8.54 523.4 ± 236.35 Crangon alaskensis 3 10.43 ± 10.69 35.33 ± 38.55 15.6 ± 9.56 Crangon franciscorum 7 0.4 ± 0.89 12.8 ± 8.14 0.2 ± 0.45 6.2 ± 12.76 2.8 ± 1.92 12.8 ± 9.44 0.33 ± 0.82 Cumella vulgaris 7 0.8 ± 0.84 13.14 ± 15.82 5.6 ± 5.37 3 ± 2.92 0.4 ± 0.89 0.4 ± 0.89 3.67 ± 4.32 Eogammarus confervicolus 7 0.2 ± 0.45 1.71 ± 2.56 2.2 ± 2.39 13.8 ± 11.14 1 ± 1 0.8 ± 0.84 0.17 ± 0.41 Gnorimosphaeroma oregonense 3 0.2 ± 0.45 4.2 ± 3.19 0.17 ± 0.41

Site abbreviations: Esquimalt Lagoon (EL), Fourth of July (FJ), Cowichan Bay (CB), Tsawwassen (Ts), Port Alberni (PA), Nanaimo Estuary (NE), north of Stanley Park (SP), Mud Bay (MB), Campbell River (CR), Prince Rupert (PR), East Kaien Island (EKI).

178 Appendix B

Table C.2. Dredge samples continued (2 of 5) No. Species Sites EL FJ CB Ts PA NE SP MB CR PR EKI Hemigrapsus oregonensis 2 5.2 ± 2.86 0.2 ± 0.45 Heptacarpus brevirostris 2 3.2 ± 4.38 10.67 ± 18.48 Heptacarpus paludicola 4 0.2 ± 0.45 0.29 ± 0.49 4 ± 6.08 55.2 ± 28.23 Heptacarpus stimpsoni 2 6 ± 8.69 0.86 ± 2.27 Hippolyte clarki 4 12.2 ± 9.04 0.2 ± 0.45 2.6 ± 2.7 0.5 ± 0.84 Idotea fewkesi 6 1.2 ± 2.68 6 ± 8.51 0.2 ± 0.45 0.2 ± 0.45 6.67 ± 5.75 9.6 ± 8.88 Idotea resecata 6 4 ± 3 1.6 ± 2.07 18.8 ± 7.19 0.8 ± 1.3 0.83 ± 1.33 16 ± 5.61 Idotea rufescens 5 3.6 ± 3.51 0.14 ± 0.38 7.2 ± 4.55 0.67 ± 1.15 23.8 ± 20.17 Idotea wosnesenskii 1 0.2 ± 0.45 Lamprops augustinensis 2 0.29 ± 0.76 0.6 ± 0.55 Lamprops triserratus 1 4 ± 8.47 Pagurus armatus 6 0.4 ± 0.55 4.2 ± 6.02 0.4 ± 0.55 7 ± 6.44 2.33 ± 2.52 0.2 ± 0.45 Pagurus hirsutiusculus 1 0.14 ± 0.38 Pancolus californiensis 1 0.2 ± 0.45 Paracalliopiella pratti 3 0.2 ± 0.45 1.6 ± 2.07 1331.33 ± 985.67 Parathemisto pacifica 1 0.17 ± 0.41 Photis brevipes 1 1 ± 1.22 Pontogeneia rostrata 8 0.2 ± 0.45 0.14 ± 0.38 123.6 ± 85.62 0.4 ± 0.89 1.8 ± 1.48 14.4 ± 12.99 4 ± 6.32 5.4 ± 3.13 Pugettia gracilis 3 2.2 ± 1.3 0.2 ± 0.45 0.33 ± 0.52 Synidotea consolidata 2 6.4 ± 7.5 0.2 ± 0.45 Telmessus cheiragonus 2 0.8 ± 0.45 2.2 ± 2.28 Thorlaksonius brevirostris 4 0.14 ± 0.38 0.67 ± 0.82 1 ± 1 1.2 ± 1.1 Maxillopoda Balanus glandula 2 0.2 ± 0.45 0.2 ± 0.45 Diosaccus spinatus 7 6 ± 11.77 932.2 ± 155.07 0.29 ± 0.76 36.8 ± 47.02 1.2 ± 2.17 1 ± 1 2.6 ± 3.13 Eurytemora americana 1 0.14 ± 0.38 Chordata Actinopterygii Gasterosteus aculeatus 3 0.29 ± 0.76 0.2 ± 0.45 0.2 ± 0.45 Myoxocephalus polyacanthocephalus 5 1 ± 1.22 1.4 ± 0.89 0.6 ± 0.89 0.17 ± 0.41 0.2 ± 0.45 Pholis ornata 7 1.8 ± 0.45 0.6 ± 0.89 1 ± 1.22 1.2 ± 1.3 0.6 ± 0.55 3.4 ± 2.97 1.83 ± 2.48 Syngnathus leptorhynchus 6 0.14 ± 0.38 3.2 ± 1.92 0.4 ± 0.55 0.4 ± 0.89 0.4 ± 0.89 0.33 ± 0.58

179 Appendix B

Table C.2. Dredge samples continued (3 of 5) No. Species Sites EL FJ CB Ts PA NE SP MB CR PR EKI Echinodermata Echinoidea Strongylocentrotus sp. 1 0.4 ± 0.55 Mollusca Bivalvia Clinocardium nuttalli 3 0.2 ± 0.45 0.17 ± 0.41 0.67 ± 1.15 Epilucina californica 2 1.33 ± 1.15 0.2 ± 0.45 Liocyma fluctuosum 1 0.6 ± 0.89 Macoma nasuta 1 0.57 ± 0.79 Saxidomus gigantea 1 0.2 ± 0.45 Gastropoda Alia carinata 1 57 ± 60.22 Alia gausapata 1 0.33 ± 0.58 Amphissa columbiana 1 0.2 ± 0.45 Crepipatella lingulata 1 0.2 ± 0.45 Fartulum orcutti 1 0.4 ± 0.89 Haminoea vesicula 3 6 ± 8 38.86 ± 39.69 7 ± 10.63 Hermissenda crassicornis 1 0.2 ± 0.45 Kurtziella crebricostata 1 0.4 ± 0.89 Lacuna marmorata 3 1 ± 1.73 12.8 ± 8.32 2.33 ± 3.67 Lacuna porrecta 1 1.8 ± 3.03 Lacuna sp. 2 5.4 ± 5.46 110.5 ± 95.03 Lacuna unifasciata 10 0.2 ± 0.45 27.4 ± 6.8 0.71 ± 0.95 0.8 ± 0.84 0.2 ± 0.45 1.8 ± 2.05 4.2 ± 4.92 0.5 ± 1.22 568.67 ± 497.26 96 ± 51.91 Lacuna variegata 3 0.6 ± 0.55 0.2 ± 0.45 23 ± 17.5 Lirobittium attenuatum 1 0.4 ± 0.89 Lirularia parcipicta 1 1.6 ± 3.58 Lirularia sp. 1 0.6 ± 1.34 Lirularia succincta 1 0.2 ± 0.45 Littorina sitkana 6 0.2 ± 0.45 0.8 ± 1.3 0.2 ± 0.45 112.4 ± 144.9 6.67 ± 7.64 5 ± 4.47 Lottia persona 2 0.2 ± 0.45 1.2 ± 0.84 Phyllaplysia taylori 1 0.2 ± 0.45 Nemertea 1 Enopia Paranemertes peregrina 1 0.2 ± 0.45

180 Appendix B

Table C.2. Dredge samples continued (4 of 5) No. Species Sites EL FJ CB Ts PA NE SP MB CR PR EKI B. Indeterminate Arthropoda Malacostraca Aoroides sp. 1 0.2 ± 0.45 Cymothoida (Suborder) 1 0.2 ± 0.45 Heptacarpus sp. 6 1 ± 2.24 5.8 ± 2.39 1.4 ± 2.07 0.8 ± 1.1 28.8 ± 30.24 2.17 ± 1.94 Ischyrocerus sp. 4 0.8 ± 1.3 38 ± 30.01 15 ± 13.23 6.6 ± 3.05 Isopoda (Order) 1 Nebalia sp. 1 0.4 ± 0.55 Photis sp. 2 0.2 ± 0.45 0.2 ± 0.45 Maxillopoda Calanoida (Order) 2 25.6 ± 19.79 Peltidiidae (Family) 1 1 ± 2 Scutellidium sp. 1 0.2 ± 0.45 Chordata Actinopterygii Gobiesocidae (Family) 1 0.2 ± 0.45 Pleuronectiformes (Order) 1 0.2 ± 0.45 Mollusca Bivalvia Macoma sp. 3 0.6 ± 0.89 0.4 ± 0.89 0.17 ± 0.41 Mytilus sp. complex 7 0.6 ± 0.55 4.8 ± 6.5 0.2 ± 0.45 0.2 ± 0.45 0.17 ± 0.41 1 ± 0 1 ± 1 Gastropoda Calyptraeidae (Family) 1 16.6 ± 34.9 Doridacea (Order) 2 0.2 ± 0.45 0.4 ± 0.89

C. Cryptogenic Annelida Polychaete Harmothoe imbricata 2 9.8 ± 13.83 0.33 ± 0.58 Arthropoda Malacostraca Ampithoe lacertosa 2 0.6 ± 0.89 1 ± 1 Leptochelia dubia 4 0.4 ± 0.89 1.6 ± 2.19 0.17 ± 0.41 0.6 ± 0.55 Maxillopoda Harpacticus uniremis group 2 0.29 ± 0.76 38.5 ± 35.02

181 Appendix B

Table C.2. Dredge samples continued (5 of 5) No. Species Sites EL FJ CB Ts PA NE SP MB CR PR EKI D. Exotic Arthropoda Malacostraca Ampithoe valida 7 1 ± 1.22 0.4 ± 0.89 0.6 ± 0.89 15.4 ± 12.12 4 ± 3.24 4.6 ± 5.46 0.67 ± 0.82 Grandidierella japonica 1 0.2 ± 0.45 Melita nitida 1 1.6 ± 2.61 Monocorophium acherusicum 4 0.14 ± 0.38 0.2 ± 0.45 1 ± 0 0.2 ± 0.45 Monocorophium insidiosum 1 0.8 ± 1.3 Sinelobus sp. 1 0.2 ± 0.45 Mollusca Bivalvia Mya arenaria 1 0.4 ± 0.89 Venerupis philippinarum 1 0.6 ± 1.34

182 Appendix B

Table C.3. Trap samples (1 of 1) No. Species Sites EL FJ CB Ts PA NE CR PR EKI Native Arthropoda Malacostraca Crangon alaskensis 1 0.75 ± 0.5 Hemigrapsus oregonensis 1 17.25 ± 10.05 Pagurus hirsutiusculus 3 0.13 ± 0.35 0.2 ± 0.45 2.33 ± 1.53 Pandalus danae 1 4 ± 2.94 Chordata Actinopterygii Gasterosteus aculeatus 2 0.25 ± 0.46 0.13 ± 0.35 Icelinus fimbriatus 1 0.25 ± 0.46 Myoxocephalus polyacanthocephalus 9 9 ± 2.93 0.63 ± 0.74 4.63 ± 2.13 0.8 ± 0.84 2.25 ± 1.49 5.25 ± 4.56 0.13 ± 0.35 0.33 ± 0.58 0.25 ± 0.5 Pholis ornata 8 0.13 ± 0.35 1.63 ± 0.92 0.75 ± 0.71 0.2 ± 0.45 0.5 ± 0.76 0.25 ± 0.46 0.13 ± 0.35 0.25 ± 0.5 Ronquilus jordani 1 0.5 ± 0.76 Syngnathus leptorhynchus 1 0.13 ± 0.35 Echinodermata Asteroidea Pycnopodia helianthoides 1 1.13 ± 1.36 Mollusca Gastropoda Haminoea vesicula 1 1.25 ± 1.75 Lacuna vincta 1 0.38 ± 0.74

Indeterminate Cnidaria Scyphozoa Semaeostomeae (Order) 1 0.13 ± 0.35

Table C.4. Visual survey (1 of 1) No. Species Sites EL FJ CB Ts PA NE SP MB CR PR EKI Exotic Mollusca Gastropoda Crassostrea gigas 2 + + Algae Phaeophyta Sargassum muticum 3 + + +

183 Appendix C

Appendix C Post hoc analysis of additional environmental variables not included in the AICc model test Post hoc analysis of environmental variables collected in Chapter 3 for environmental analyses but not selected for testing in the chapter because AICc model tests were run with only 3 variables to reduce the total number of models (less models than sample sites).

C.1 Methods Eelgrass plant density and length were collected along the same 50-m transects from the core sampling. Quadrats (.25m x .25m) were placed across the transect line from each core, all plants were counted in each of the 6 quadrats and the length of the longest blades from three randomly selected plants were measured and averaged for each quadrat. Sediment samples were collected at each end of the 50-m transect (4 per eelgrass bed). Dried sediment samples were sieved using a Ro-Tap® Test Sieve Shaker (W.S. Tyler Industrial Group, Inc., Mentor, Ohio) to identify 8 sediment size classes (2 mm – 63 microns) and each size class weighed to determine the grain size distribution. The resulting multivariate data of grain size distribution was collapsed using principle components analysis (PCA) to create a smaller subset of dimensions that captured the dominant gradients. The first two axes explain 91.8% of the total variation; negative PCA 1 values were most found at sites with fine pebbles (grain sizes > 2 mm) while positive values were found at sites with fine sand (grain sizes between 0.25 and 0.13 mm), PCA 2 values were more negative at sites with coarse to fine sand (grain sizes between 0.50 and .06 mm), positive PCA 2 values cannot be attributed to a specific grain type.

The shore angle was calculated at low-tide by measuring the distance between the waterline at the upshore edge and the middle of the eelgrass bed along the surface of the water, measuring the depth of the water at that mid-point, then calculating the sine of the angle between the bottom (hypotenuse) and the water surface. A mean shore angle was then calculated from each of three angles measured at each site. Eelgrass beds with a slope below 2 degrees were categorized as flats while beds with a slope of 2 or greater

184 Appendix C were categorized as a fringe. In most cases fringe eelgrass beds were found along channels while flat beds were found in large mudflats (Campbell River was categorized as a flat but was found along a channel). Winter temperature data were not available for each site sampled, thus I used an oceanographic climate model (pers. comm. with Mike Foreman, Department of Fisheries and Oceans Canada; Foreman et al., 2008) to generate data on sea surface temperature (SST), averaged for the winter.

To reduce deviations from a normal distribution, I log-normal transformed eelgrass density and shore angle. Environmental variable data are presented in S2 Table 1.

Table C.1: Variables used in the analysis of the relationships between species richness and environmental conditions. Mean (units described in B.1 Methods), standard deviation (S.D.), minimum and maximum values of each response and explanatory variable.

Environmental Variable Mean S.D. Min. Max. Ln Grass Density 4.67 0.71 3.56 5.96 Mean Grass Length 84.33 43.36 29.59 161.40 Sediment Grain Size PCA axis 1 0 0.31 -0.64 0.41 Sediment Grain Size PCA axis 2 0 0.19 -0.27 0.30 Winter SST 6.98 0.58 6.13 7.66 Ln Shore Angle 0.28 1.30 -2.30 1.94 Latitude 50.03 2.14 48.43 54.29 Ln = log-normal transformation

Explanation of Predicted Relationship: Density of Eelgrass: I expect a greater richness and abundance of nonnative species with more shelter from greater densities of eelgrass plants (Carr et al., 2010). Length of Eelgrass: The length of eelgrass is likely to be positively correlated to nonnative epifaunal richness because it increases habitat structure (Carr et al., 2010)). Sediment Grain Size: Benthic community composition has been associated with sediment grain size (Calabretta & Oviatt, 2008). Sediment PCA axis 1 explains 67% of the variation of grain size while Sediment PCA axis 2 explains 25%. Winter SST: Cold temperatures are negatively correlated with survival and reproduction on nonnative species (Stachowicz et al., 2002b; Clark & Johnston, 2005; Dafforn et al., 2009).

185 Appendix C

Shore Angle: Species richness and abundance are likely to be greater in regions with more wave energy; where shore angle is steep and more exposed to water currents and waves (Demes et al., 2012). Latitude: With lower richness of nonnatives found in the colder waters of higher latitudes (deRivera et al., 2005), I would predict a greater richness and abundance of nonnatives at lower latitudes.

To test the relationship of these five environmental explanatory variables on nonnative total richness and mean abundance, I compared these data using a nonparametric Spearman rank correlation (rho; Sokal & Rohlf, 1995).

C.2 Results Sites with low sediment PCA1 values had both higher epifaunal nonnative richness and abundance in Spearman’s rank correlations (S2 Table 2; N = 11; richness: rho = -0.59, P = 0.05; abundance: rho = -0.81, P = 0.003). Shore angle correlation with epifaunal nonnative richness (rho = 0.55, P = 0.08) and grass density correlation with epifaunal nonnative abundance (rho = 0.53, P = 0.09) were mildly significant (< 0.01). No other variables were significantly related to nonnative epifaunal richness or abundance. No variables were significantly related to nonnative benthic richness or abundance.

186 Appendix C

Table C.2: Post hoc tests of Spearman rank correlations (rho) and probability-values (P) for benthic and epifaunal nonnative richness and abundance with additional environmental variables not included in model tests. Significant P-values in bold.

Vector rho P rho P Richness Abundance Benthic Nonnatives Ln Grass Density 0.19 0.57 0.27 0.42 Mean Grass Length -0.13 0.70 -0.06 0.87 Sediment PCA 1 -0.27 0.41 -0.16 0.63 Sediment PCA2 0.07 0.83 0.11 0.74 Winter SST 0.01 0.98 -0.02 0.94 Ln Shore Angle 0.16 0.63 0.15 0.65 Latitude -0.01 0.96 0.02 0.96 Epifaunal Nonnatives Ln Grass Density 0.53 0.09 0.36 0.28 Mean Grass Length -0.01 0.97 0.16 0.65 Sediment PCA 1 -0.59 0.05 -0.81 0.003 Sediment PCA 2 0.21 0.53 -0.15 0.67 Winter SST 0.04 0.92 0.08 0.81 Ln Shore Angle 0.35 0.29 0.55 0.08 Latitude -0.21 0.54 -0.10 0.77

C.3 Discussion Environmental variables historically demonstrated as important for determining nonnative species distributions were collected during this study and tested in post hoc analyses. Of all environmental variables only sediment grain size was correlated to nonnative species. Sites with larger grain sizes had a greater richness and abundance of epifaunal nonnatives. Interestingly benthic nonnatives were not related to sediment. For nonnatives in BC eelgrass, initial summer temperature and salinity filters, described in the results (section 3.3) and discussion (section 3.4) of this chapter, drive species richness while environmental variables included in this post hoc analysis had little to no effect on nonnatives.

187 Appendix D

Appendix D R code for the green crab consumption model #Green Crab Consumption Model #MAY 16, 2012 #R code used for the consumption model, example below contains data for a low calorie (40000 cal year-1), low density (10000 crabs km-2) model of oyster consumption. R- syntax text in blue colour.

Ntrials=10000 oarea=runif(Ntrials,450,600) #Harvest area for the shellfish species in Puget Sound, km2. Calculated as 523.71 km2 for commercial shellfish +/- ~ 75 km) Num=10000 #number of crab per km2 Cal=40000 #calories consumed per crab individual in a year odiet=runif(Ntrials,.2,.35) # % of green crab diet that is this species of shellfish ocallbADFW=2199922.995 #calories per pound ash free dry weight (AFDW) of the shellfish species poAFDWtoWW=runif(Ntrials,1.7,1.7) #convert AFDW to wet weight (to compare to biomass weight for harvested biomass), oyster conversion estimate did not include a standard deviation pocallbWW=pocallbAFDW/poAFDWtoWW

#consumption of pacific oysters ocon=oarea*Num*Cal*odiet/ pocallbWW #consumption of pacific oysters

#notes: #rnorm = a normal curve, used for MCMC sampling when less certain about accuracy of your values (example for Num above, the average number of individual crabs expected per km2 is 10000 with a standard deviation of 500, this means most samples will occur near the peak at 10000 but will also sample up to 500 away from 10000). #runif= a flat line between the two values (example for oarea above, will sample an even number of times between 450 and 600), used when you are confident in your data and range of values

188 Appendix D

#To calculate 95% confidence intervals - calculates an error for the mean and find the CI by adding and subtracting the error from the mean oerror <- qt(0.975,df=length(ocon)-1)*sd(ocon)/sqrt(length(ocon)) #error of the mean to calculate the 05% probability that the true mean is within the confidence interval oleft<- mean(ocon)-oerror #mean is between this lower value oright<- mean(ocon)+oerror # and this upper value

#Biomass & revenue values for 6 harvest areas in Puget Sound obiomass = 2968366 #pounds of oyster biomass harvested ovalue = 14062126.68 #revenue of harvested oyster

#mean pounds of oyster biomass consumed mean(pocon)

#Harvested oyster biomass after green crab invasion lowbio = obiomass-oleft highbio = obiomass-oright lowbio highbio

#Percent of original biomass that is the oyster harvest biomass after invasion perlow=lowbio/obiomass perhigh=highbio/obiomass

#estimate of change of revenue after green crab invasion (does not account for change in market value) lowvalue=ovalue*perlow highvalue=ovalue*perhigh lowvalue highvalue

189