Community-level effects of nonindigenous aquatic ecosystem engineers

Jessica MacKay Ward

Department of Biology

McGill University

Montreal, Canada

January 2010

A thesis submitted to McGill University in partial fulfillment of the requirements

of the degree of Doctor of Philosophy

© Jessica MacKay Ward 2010

Library and Archives Bibliothèque et Canada Archives Canada

Published Heritage Direction du Branch Patrimoine de l’édition

395 Wellington Street 395, rue Wellington Ottawa ON K1A 0N4 Ottawa ON K1A 0N4 Canada Canada

Your file Votre référence ISBN: 978-0-494-68563-1 Our file Notre référence ISBN: 978-0-494-68563-1

NOTICE: AVIS:

The author has granted a non- L’auteur a accordé une licence non exclusive exclusive license allowing Library and permettant à la Bibliothèque et Archives Archives Canada to reproduce, Canada de reproduire, publier, archiver, publish, archive, preserve, conserve, sauvegarder, conserver, transmettre au public communicate to the public by par télécommunication ou par l’Internet, prêter, telecommunication or on the Internet, distribuer et vendre des thèses partout dans le loan, distribute and sell theses monde, à des fins commerciales ou autres, sur worldwide, for commercial or non- support microforme, papier, électronique et/ou commercial purposes, in microform, autres formats. paper, electronic and/or any other formats. . The author retains copyright L’auteur conserve la propriété du droit d’auteur ownership and moral rights in this et des droits moraux qui protège cette thèse. Ni thesis. Neither the thesis nor la thèse ni des extraits substantiels de celle-ci substantial extracts from it may be ne doivent être imprimés ou autrement printed or otherwise reproduced reproduits sans son autorisation. without the author’s permission.

In compliance with the Canadian Conformément à la loi canadienne sur la Privacy Act some supporting forms protection de la vie privée, quelques may have been removed from this formulaires secondaires ont été enlevés de thesis. cette thèse.

While these forms may be included Bien que ces formulaires aient inclus dans in the document page count, their la pagination, il n’y aura aucun contenu removal does not represent any loss manquant. of content from the thesis.

Abstract

The identification of generalities regarding which biological invasions are most likely to cause major ecological disruptions is a central goal of invasion ecology, having both theoretical and applied importance. Introduced ecosystem engineers (organisms that modify, create and/or destroy habitats) can exert strong impacts on recipient communities, altering not only species composition but also the physical structure of ecosystems. In this thesis, I examine variation in the impacts of nonindigenous ecosystem engineers on aquatic macroinvertebrate communities, with emphasis given to Dreissena spp. because they are well- studied exotic ecosystem engineers of both ecological and economic concern. The objectives of this research were to identify general patterns and test existing theories with respect to the community-level impacts of exotic ecosystem engineers. To meet these objectives, I employed a combination of field experiments and statistical syntheses of published data. In Chapters 2 and 3, I examined the reliance of ecosystem engineering impacts on a community‘s evolutionary experience with the engineer or with similar species (i.e. ecological naiveté). Meta-analyses revealed that impacts on community diversity and abundance do not systematically differ among native and exotic ecosystem engineers (Chapter 2), and suggested that introduced ecosystem engineers with distinct functional traits compared to native species generally promote facilitation (Chapter 3). I focused on Dreissena spp. in Chapters 4 and 5, wherein I examined sources of variation in the community-level impacts of these exotic mussels. A meta-analysis revealed that Dreissena generally increases macroinvertebrate density and richness, and that the strength of these effects varies with sediment particle size (Chapter 4). In Chapter 5, I tested the effect of Dreissena patch topography on benthic diversity by experimentally manipulating the area and perimeter-to-area ratio of mussel patches. My results indicated that Dreissena patch topography may be an important determinant of species richness at small spatial scales, and suggested that the impact of Dreissena is modified in the presence of another habitat-forming species, the native alga Cladophora. This

i work contributes to our understanding of the context-dependence of invasion impacts; that is, how the direction or magnitude of impact depends on the characteristics of invaders and invaded systems.

ii Résumé

L'identification de généralités en ce qui concerne les invasions biologiques d‘qui ont les plus grandes chances de provoquer des perturbations écologiques importantes est un but central de l'écologie des invasions, ayant de l'importance théorique et pratique. Les ingénieurs d'écosystèmes introduits (les organismes qui modifient, créent et-ou détruisent des habitats) peuvent exercer des impacts importants sur les communautés envahies, en transformant non seulement la composition d'espèces, mais aussi la structure physique des écosystèmes. Dans cette thèse, j'examine la variation dans les impacts d'ingénieurs d'écosystème nonindigènes sur les communautés des macroinvertébrés aquatiques, avec l'accentuation donnée à Dreissena spp. parce qu'elles sont bien étudiées en tant qu'ingénieurs d'écosystèmes qui ont des impacts écologiques et économiques importants. Les objectifs de cette recherche sont d'identifier des généralités et d‘évaluer des théories en ce qui concerne les impacts d‘ingénieurs d'écosystèmes exotiques au niveau de la communauté. Pour atteindre ces objectifs, j'ai employé une combinaison d‘expériences sur le terrain et de synthèses statistiques de données publiées. Dans les chapitres 2 et 3, j'ai examiné la dépendance des impacts d'ingénieurs d‘écosystèmes sur l'expérience évolutive d'une communauté soit avec la présence de l'ingénieur ou bien avec des espèces semblables (c'est-à- dire la naïveté écologique). Les meta-analyses ont démontré que les impacts sur la diversité et l'abondance communautaires ne diffèrent pas systématiquement parmi les ingénieurs d'écosystème indigènes et exotiques (chapitre 2), et ont suggéré que les ingénieurs d'écosystème exotiques qui possèdent des traits fonctionnels distinctifs comparés aux espèces indigènes favorisent généralement la facilitation (chapitre 3). Je me suis concentrée sur les Dreissena spp. dans les chapitres 4 et 5, où j'ai examiné les sources de variation dans les impacts de ces moules exotiques au niveau communautaire. Une meta-analyse a démontré que Dreissena augmente généralement la densité et la richesse des macroinvertébrés, et que l‘ampleur de ces effets varie selon la granulométrie des particules de sédiment (chapitre 4). Dans le chapitre 5, j'ai évalué l'effet de la topographie de

iii parcelles de Dreissena sur la biodiversité des macroinvertébrés benthiques en manipulant expérimentalement la surface occupée et le rapport du périmètre et de la surface des parcelles de moules. Mes résultats ont indiqué que la topographie des parcelles de Dreissena est peut-être un déterminant important de la richesse des espèces aux échelles spatiales petites et ont suggéré que l'impact de Dreissena est modifié en présence d'une autre espèce créateur d‘habitats, l'algue indigène Cladophora. Ce travail contribue à notre compréhension de la dépendance vis-à- vis du contexte des impacts d'invasions; c'est-à-dire la manière dont la direction ou l‘ampleur d'impact dépend des caractéristiques des envahisseurs et des systèmes envahis.

iv Acknowledgements

I am above all grateful to my supervisor, Dr. Anthony Ricciardi, for his frequent council and creative input, enthusiasm for research, lively and stimulating discussions, and support throughout my development from an undergraduate student to a young scientist. I am also grateful to the members of my supervisory committee, Dr. Frédéric Guichard and Dr. Brian Leung, for their constructive input, questions and advice. I would also like to thank Dr. Andrew Gonzalez for constructive input and suggestions, Dr. Andrew Hendry for thoughtful questions, and Dr. Brian McGill for advice on statistical analyses. I am sincerely grateful to Michelle Palmer, who provided invaluable training and assistance with field research. I also thank Lisa Jones, Anneli Jokela, Rebekah Kipp, Åsa Kestrup, Robin Tiller, James Snider and Taïca Replansky for their assistance in the field and help processing experimental substrates in the lab. For assistance with invertebrate identification, I thank Rebekah Kipp, Arthis Sowmithran, Niloufar Bayani, and Hanie Seo. For assistance with data entry, I thank Suzanne Ellwood, Katie O‘Sullivan and Ahdia Hassan. I am also grateful to past and current members of the Ricciardi lab for stimulating discussions, creating a friendly work environment, and providing humour and support. I am genuinely grateful to my family and friends, especially James Snider, Barbara Ward, and Patricia MacKay for their continuous support and encouragement. I dedicate this thesis to them. In conducting this work, I was supported financially by a Canada Graduate Scholarships from the National Research and Engineering Council of Canada, a Vineberg Family Fellowship from the Department of Biology, and a Dr. and Mrs. Milton Leong Graduate Student Award from McGill‘s Graduate and Postdoctoral Studies Office. Travel grants provided by the Redpath Museum and McGill‘s Department of Biology allowed me to present portions of this research at national and international meetings and conferences.

v Contributions of Authors

This thesis consists of a collection of papers (Chapters 2 through 5) individually prepared as manuscripts for publication in peer-reviewed scientific journals. I designed and executed the research and wrote the manuscripts in full, under the supervision of Dr. Anthony Ricciardi, who contributed to the development of hypotheses and provided technical and editorial advice, as well as financial and logistical support required to conduct this research. I am the primary author and Dr. Ricciardi is co-author of each of these manuscripts.

vi Claim to Originality

This thesis represents the first comprehensive, quantitative treatment of the community-level impacts of nonindigenous ecosystem engineers in aquatic systems. For the first time, meta-analyses of literature data are used to relate variation in the community-level impacts of exotic ecosystem engineers to biotic and abiotic characteristics of invaded systems. Chapters 4 and 5 of this thesis contain peer-reviewed papers for Diversity and Distributions (published) and Freshwater Biology (in press), respectively. At the time of submission, these papers presented original work not previously published in the primary literature. The manuscripts presented in Chapter 2 and 3 have been prepared for submission to peer-reviewed journals.

The major novel findings of this thesis are:

The effect of benthic ecosystem engineering, particularly bioconstruction, on macroinvertebrate diversity and abundance does not systematically differ among native and exotic ecosystem engineers (Chapter 2).

Impacts of exotic ecosystem engineers on native macroinvertebrates are correlated with their phylogenetic relatedness to native species in the invaded region, although not in the direction previously observed for predatory and competitive interactions (Chapter 3).

The impacts of Dreissena spp. on macroinvertebrate communities vary with the biotic (e.g. presence of native bioconstructors, Chapter 5) and abiotic (e.g. substrate particle size, Chapter 4) characteristics of invaded sites.

Dreissena patch topography influences benthic diversity at small spatial scales (Chapter 5).

vii Table of Contents

Abstract ...... i

Résumé ...... iii

Acknowledgements ...... v

Contributions of Authors ...... vi

Claim to Originality ...... vii

Table of Contents ...... viii

List of Tables ...... xii List of Figures ...... xiii

Chapter 1 – General Introduction ...... 1

1.1 Impacts of biological invasions – a review ...... 1 1.1.1 Factors explaining variation in invasion impacts ...... 2 1.1.2 Community-level impacts ...... 4 1.1.3 Exotic ecosystem engineers...... 5 1.2 Research objectives and approach ...... 9 1.3 References cited ...... 11

Connecting statement 1–2 ...... 21

Chapter 2 – Community-level effects of native and exotic ecosystem engineers: a meta-analysis ...... 22

2.1 Abstract ...... 23 2.2 Introduction ...... 24 2.3 Methods...... 27 2.3.1 Data collection ...... 27 2.3.2 Statistical analysis ...... 29 2.4 Results ...... 30 2.4.1 Hypotheses 1, 2 and 3 ...... 31 2.4.2 Hypotheses 4, 5 and 6 ...... 33

viii 2.5 Discussion ...... 38 2.5.1 Bioconstruction versus bioturbation ...... 38 2.5.2 Ecological naiveté ...... 40 2.5.3 Selection regime modification and invasional meltdown ...... 41 2.5.4 Implications for theory and management ...... 42 2.5.5 Conclusions and future directions ...... 43 2.6 Acknowledgements ...... 43 2.7 References cited ...... 44

Connecting statement 2–3 ...... 50

Chapter 3 – The impacts of exotic ecosystem engineers are mediated by their phylogenetic distinctiveness ...... 51

3.1 Abstract ...... 52 3.2 Introduction ...... 53 3.3 Methods...... 55 3.3.1 Data collection ...... 55 3.3.2 Statistical analysis ...... 56 3.4 Results ...... 59 3.5 Discussion ...... 65 3.5.1 The ecological naiveté hypothesis ...... 65 3.5.2 Interaction strength and environmental stress models ...... 69 3.5.3 Facilitation by distinct engineers...... 72 3.5.4 Time since invasion ...... 74 3.5.5 Conclusions ...... 74 3.6 Acknowledgements ...... 75 3.7 References cited ...... 75

Connecting statement 3–4 ...... 82

Chapter 4 – Impacts of Dreissena invasions on benthic macroinvertebrate communities: a meta-analysis ...... 83

4.1 Abstract ...... 84

ix 4.2 Introduction ...... 85 4.3 Methods...... 87 4.4 Results ...... 93 4.4.1 Variation in effects on the numerical density and taxonomic richness of macroinvertebrate communities ...... 93 4.4.2 Variation in effects on taxonomic and functional groups ...... 96 4.5 Discussion ...... 100 4.5.1 Predictable patterns of Dreissena impact...... 100 4.5.2 Caveats ...... 105 4.5.3 Comparisons with other introduced mussel species ...... 105 4.6 Acknowledgements ...... 108 4.7 References cited ...... 108

Connecting statement 4–5 ...... 116

Chapter 5 – Community level effects of co-occurring native and exotic ecosystem engineers ...... 117

5.1 Abstract ...... 118 5.2 Introduction ...... 119 5.3 Methods...... 121 5.3.1 Study sites...... 121 5.3.2 Experiment 1 ...... 122 5.3.3 Experiment 2 ...... 123 5.3.4 Experiment 3 ...... 124 5.3.5 Statistical analysis ...... 124 5.4 Results ...... 126 5.4.1 Experiment 1 ...... 126 5.4.2 Experiment 2 ...... 131 5.4.3 Experiment 3 ...... 136 5.5 Discussion ...... 139 5.5.1 Effect of Dreissena patch topography on benthic diversity ...... 139 5.5.2 Effect of Cladophora on benthic macroinvertebrates ...... 141

x 5.5.3 Is Cladophora obscuring the effect of Dreissena? ...... 143 5.5.4 Alternative explanations for reduced Dreissena effects at Pointe-du-Moulin ...... 146 5.6 Acknowledgements ...... 149 5.7 References cited ...... 149

Connecting statement 5–6 ...... 156

Chapter 6 – General conclusions and future research ...... 157

6.1 Summary ...... 157 6.2 Conceptual framework ...... 158 6.3 Future research ...... 162 6.4 References cited ...... 163

Appendices ...... 166

Appendix A. Comment on ―Opposing Effects of Native and Exotic Herbivores on Plant Invasions‖ ...... 167 Appendix B. Data sources for meta-analysis of the effects of native and exotic ecosystem engineers (Chapter 2)...... 172 Appendix C. Data sources for meta-analysis of the effects of exotic ecosystem engineers (Chapter 3)...... 194 Appendix D. Data sources for meta-analysis of Dreissena effects (Chapter 4)...... 202 Appendix E. Macroinvertebrate taxa collected from experimental substrata (Chapter 5)...... 208

xi List of Tables

1.1. Summary of ecological concepts involving physical habitat modification by organisms ...... 6 3.1. Fisher Exact tests comparing the frequency of large positive (>80% increase) and negative (>80% reduction) impacts on native species among exotic ecosystem engineers belonging to shared or distinct taxa...... 64 3.2. Univariate weighted meta-regression analyses for the relationship between moderator variables and the impacts of exotic engineers on native macroinvertebrates...... 66 4.1. Univariate regression models of Dreissena effect on macroinvertebrate numerical density vs. study site characteristics and methodological variables ...... 89 5.1. Similarity percentages analysis for Experiment 1 ...... 130 5.2. Univariate analyses of macroinvertebrate community response variables for Experiment 2 ...... 133 5.3. Permutational analysis of variance for Experiment 2 ...... 134 5.4. Similarity percentages analysis for Experiment 2 ...... 135 5.5. Univariate analyses of macroinvertebrate community response variables for Experiment 3 ...... 137 5.6. Permutational analysis of variance for Experiment 3 ...... 138 5.7. Comparison of my experimental results with those of published experiments examining the effect of Dreissena shells on benthic macroinvertebrate communities...... 147 B.1. Summary of effect size data sources for Chapter 2 ...... 172 C.1. Summary of effect size data sources for Chapter 3 ...... 194 D.1. Summary of effect size data sources for Chapter 4 ...... 202 E.1. Macroinvertebrate taxa collected from experimental substrata in Chapter 5 ...... 208

xii List of Figures

2.1. Mean effect of bioconstruction by native and exotic ecosystem engineers on macroinvertebrate community response variables...... 32 2.2. Mean effect of bioconstruction and bioturbation by native and exotic ecosystem engineers on macroinvertebrate numerical density ...... 34 2.3. Mean effect of bioconstruction by native and exotic ecosystem engineers on macroinvertebrate communities in marine, estuarine and freshwater systems ...... 35 2.4. Mean effect of bioconstruction by native and exotic ecosystem engineers on the total taxonomic richness and numerical density of native and exotic macroinvertebrates...... 37 3.1. Mean effects (Hedges‘ d) of exotic ecosystem engineers on native species for engineer species representing shared or distinct taxa in the invaded region...... 60 3.2. Mean effects (lnR) of exotic ecosystem engineers on native species for engineer species representing shared or distinct taxa in the invaded region...... 61 3.3. Mean positive and negative effects (Hedges‘ d) of exotic ecosystem engineers on native species for engineer species representing shared or distinct taxa in the invaded region...... 62 3.4. Mean positive and negative effects (lnR) of exotic ecosystem engineers on native species for engineer species representing shared or distinct taxa in the invaded region...... 63 3.5. Variation in the effects of exotic ecosystem engineers on native species with time since invasion...... 67 3.6. Possible patterns of variation in the magnitude of facilitative and competitive interaction components in relation to the functional distinctiveness of exotic ecosystem engineers ...... 71 4.1. Mean response of macroinvertebrate communities to Dreissena...... 94

xiii 4.2. Relationship between effect of Dreissena on macroinvertebrate numerical abundance or taxonomic richness and Dreissena density or substrate particle size ...... 95 4.3. Mean effects of Dreissena on the numerical density of major groups of benthic macroinvertebrates...... 97 4.4. Relationship between Dreissena effects on the numerical density of either epifaunal or infaunal macroinvertebrate taxa and substrate particle size ...... 98 4.5. Relationships between effect of Dreissena on the numerical density of macroinvertebrate taxa belonging to different functional feeding groups and substrate particle size ...... 99 4.6. Relationship between Dreissena effect on the numerical density of gastropod taxa and gastropod shell size ...... 101 5.1. Total non-dreissenid macroinvertebrate abundance, taxonomic richness, diversity and evenness on Dreissena shell treatments in Experiment 1. ....127 5.2. Non-metric MDS ordinations for Experiments 1, 2 and 3 ...... 128 5.3. Mean Cladophora density on Dreissena shell treatments in Experiments 2 and 3...... 132 5.4. Schematic illustration of the biotic interactions between Dreissena, Cladophora and epibenthic macroinvertebrates ...... 142 6.1. Hypothetical framework for the community-level impacts of exotic ecosystem engineers...... 159 A.1. Effects of native herbivores on exotic plant abundance and survival in 18 experimental studies...... 169

xiv Chapter 1 – General Introduction

1.1 Impacts of biological invasions – a review

Exotic species introductions are at once an unprecedented form of global anthropogenic change and an ongoing biogeographical experiment in ecology and evolutionary biology. The biological invasion process consists of a sequence of stages that include the transport of a species to a region beyond its historical geographical range (i.e. outside the region in which it evolved), formation of a self-sustaining population (i.e. establishment), and impact (Williamson 1996; Kolar & Lodge 2001). Human activities associated with transportation and economic globalization are accelerating the rate of biological invasions, particularly in aquatic systems (Ribera & Boudouresque 1995; Cohen & Carlton 1998; Leppäkoski & Olenin 2000; Ricciardi 2001), where nonindigenous (or nonnative or exotic) species currently pose a major threat to native biodiversity and ecosystem function (Wilcove et al. 1998; Kats & Ferrer 2003; Clavero & García-Berthou 2005; Dextrase & Mandrak 2006). The capacity of exotic species to have extensive ecological impacts, in addition to significant societal and economic consequences (Vitousek et al. 1996; Pimentel 2002), has stimulated the growth of invasion ecology as an emerging field in recent decades. These investigations have increased our understanding of the forces that structure ecological communities (Lodge 1993; Moyle & Light 1996; Lockwood et al. 1997; Callaway & Maron 2006; Sax et al. 2007). However, while much research has focused on predicting the successful establishment of nonindigenous species (e.g. Kolar & Lodge 2001; Lester 2005; Ruesink 2005; Moyle & Marchetti 2006; García-Berthou 2007; Hayes & Barry 2008), a predictive understanding of impact remains elusive. Still, the development of generalizations regarding impact is a central goal of invasion ecology, given that (i) many nonindigenous species appear not to have substantial ecological impacts (Williamson 1996; Parker et al. 1999; García-Berthou et al. 2005); and (ii) the impact of the same species can vary across its invaded range,

1 presumably as a result of site-to-site variation in the physical or biological characteristics of recipient systems (e.g. D‘Antonio et al. 2000; Ross et al. 2003; Ricciardi & Kipp 2008). Generalizations regarding which invasions cause the largest ecological disruptions are therefore important for the effective allocation of limited management resources (Lodge et al. 1998; Parker et al. 1999; Byers et al. 2002). Moreover, such generalizations would provide insight into the general ecological question of which species are most important for the maintenance of biodiversity and ecosystem function (Paine 1966; Mills et al. 1993). Here, I define ecological impact as a measurable change in a property of an ecosystem. This basic definition recognizes that all invasions will have some ecological effect on the recipient system. Furthermore, I consider the impact of an exotic species to be the outcome of its interaction with its new biotic and abiotic environment, and thus context-dependent. This context-dependence has led some to the opinion that the outcomes of biological invasions are inherently unpredictable (e.g. Gilpin 1990; Williamson 1999). Yet, a burgeoning literature of published case studies presently provides an opportunity to synthesize existing data in order to develop and test general models of impact.

1.1.1 Factors explaining variation in invasion impacts

Several frameworks have been advanced to facilitate quantification, comparison and prediction of invasion impacts. Parker et al. (1999) proposed a simple linear equation wherein the impact of an exotic species (I) is the product of the area of its invaded range (R), its average abundance per unit area (A) and its per capita (or unit biomass) effect (E):

I = R x A x E

Subsequently, Ricciardi (2003) put forward a framework wherein the impact of an exotic species (I) is a function of its abundance (A), its ecological function or per capita effect (F) and the composition of the recipient community (C). Although originally represented as a linear equation (I = A x F x C), this model is more accurately formalized as:

2 I = ƒ (A, F, C)

These equations highlight key ecological variables expected to moderate the severity of invasion impacts either among exotic species or across invaded sites. Perhaps the most intuitive of these is that an invader‘s impact will be some positive function of its abundance. For example, the impact of exotic carp (Cyprinus carpio) on aquatic vegetation is proportional to carp biomass (Crivelli 1983). One advantage of relating an invader‘s impact to its abundance is that impact may be predicted from models that relate abundance to environmental parameters (Ricciardi 2003). However, invasion impacts are often assumed, but rarely demonstrated, to be correlated with observed abundances; most studies that have quantified ecological impact report a total effect that combines abundance (A) and per capita effect (E) (Parker et al. 1999). Moreover, the degree to which variation in the physical or biological properties of invaded systems (i.e. context- dependence) may override variation in abundance is largely unknown. Ecological impacts of biological invasions have been observed to vary across environmental gradients such as salinity (Alcaraz et al. 2008), light availability (Flory et al. 2007), calcium concentration (Jokela & Ricciardi 2008), tidal elevation (Krassoi et al. 2008), habitat complexity (Stuart-Smith et al. 2007), and eutrophication (Byers 2000). Such variation in local environmental conditions may moderate an invader‘s impact by regulating its local abundance (Parker et al. 1999; Ricciardi 2003) or by changing the nature or intensity of its interactions with other species (i.e. its per capita effect), irrespective of its abundance (e.g. Alcaraz et al. 2008; Krassoi et al. 2008). In addition, processes that modulate the abundance or per capita effect of exotic species on temporal scales may contribute to variation in impacts across sites invaded for different lengths of time (Strayer et al. 2006). Several lines of reasoning suggest that invasion impacts should depend on the biological characteristics of the recipient community (C). Firstly, an invader‘s impact will depend on the presence of natural enemies that can regulate its abundance (i.e. biotic resistance; Elton 1958; Levine et al. 2004; Parker et al. 2006). In addition, several authors have postulated that the impact of an exotic species will be larger when it is ecologically different from resident species in the

3 invaded community; exotic species that introduce new attributes to an invaded community are likely to play a novel functional role and may therefore alter nutrient cycling and energy flow pathways (Vitousek 1990; Simberloff 1991; Ruesink et al. 1995; Simon & Townsend 2003; Dukes & Mooney 2004). As well, invasion impacts are expected to be most severe when the recipient community lacks a common evolutionary history with species belonging to the same guild or functional group as the invader (Diamond & Case 1986; Case & Bolger 1991; Callaway & Ridenour 2004; Ricciardi & Atkinson 2004; Cox & Lima 2006). Native species may be naïve to the effects of exotic species possessing functional traits with which they lack evolutionary experience, for example the root exudates of introduced plants (i.e. the novel weapons hypothesis; Callaway & Ridenour 2004) or the novel hunting tactics of introduced predators (i.e. the ecological naiveté hypothesis; Diamond & Case 1986; Ricciardi & Atkinson 2004; Cox & Lima 2006). Numerous authors have recognized that predators tend to have large impacts when introduced to insular systems (e.g. Elton 1958; Atkinson 1985; Diamond & Case 1986; Blackburn et al. 2004; Cox & Lima 2006). Additional evidence for this line of reasoning comes from pathogens and parasites that are relatively benign in their native range but have devastating effects in invaded regions where native hosts lack immunity, such as chestnut blight fungus, Cryphonectria parasitica (Anagnostakis 1987) and the emerald ash borer, Agrilus planipennis (Rebek et al. 2008). Finally, the presence of multiple exotic species in many invaded communities, and the potential for their impacts to be synergistic (Simberloff & Von Holle 1999; Blackburn et al. 2005), is an additional factor that may contribute to variation in invasion impacts.

1.1.2 Community-level impacts

The ecological impacts of biological invasions have been measured at the individual, population, community and ecosystem level (Parker et al. 1999; Simon & Townsend 2003; Lockwood et al. 2007). Here, I focus on community- level impacts, which I consider to be a measurable change in a characteristic of a

4 community (e.g. species richness, composition or functional diversity). Existing knowledge of the community-level impacts of exotic species mainly involves negative effects on biodiversity mediated through predation or competition (Parker et al. 1999; Lockwood et al. 2007). Some of the most notorious examples of introduced consumers include the introduction of Nile perch (Lates niloticus) to Lake Victoria, the invasion of Guam by the brown tree snake (Boiga irregularis), and the introduction of mammalian predators such as rats (Rattus rattus and R. norvegicus) and herbivores such as goats (Capra hircus) to oceanic islands throughout the world. Examples of competitive exclusion of native species by exotics abound in plant communities, where many exotic pests are known to form extensive monocultures (Vilà & Weiner 2004), but also include animals such as the red imported fire ant (Solenopsis invicta) in the United States. Although less often studied, exotic species introductions can also produce large community-level impacts by modifying the physical structure of habitats, which may result in positive or negative effects on community abundance and diversity (Crooks 2002; Rodriguez 2006).

1.1.3 Exotic ecosystem engineers

Invaders that modulate the availability of resources to other species by building and modifying habitats can potentially be defined by a suite of related terms (Table 1.1). Throughout this thesis, I use the term ecosystem engineer (sensu Jones et al. 1994, 1997) to describe such organisms. While generalizations regarding the community-level impacts of exotic ecosystem engineers are lacking (but see Crooks 2002), these organisms have become widespread in a variety of freshwater and marine systems (Crooks 2002; Wallentinus & Nyberg 2007), providing an opportunity to test hypotheses related to community-level invasion impacts. Numerous authors have suggested that ecosystem engineering organisms will be high-impact invaders (Vitousek 1990; Crooks 2002; Wallentinus & Nyberg 2007; Crooks 2009). Indeed, ecosystem engineers are likely to have large

5 Table 1.1. Summary of ecological concepts involving physical habitat modification by organisms. Concept Definition Reference Edificator A plant species that plays a crucial role in structuring the environment and providing Braun-Blanquet niches for associated dependant organisms. (1928)

Foundation species A single species that defines much of the structure of a community by creating locally Dayton (1972) stable conditions for other species, and by modulating and stabilizing fundamental ecosystem processes.

Keystone modifier A species that affects habitat features without necessarily having direct trophic effects Mills et al. (1993) on other species, if the modified habitat affects the survival of many other species.

Ecosystem engineer An organism that modifies, maintains or creates habitats; an organism that directly or Jones et al. (1994); indirectly controls the availability of resources to other organisms by causing physical Jones et al. (1997) state changes in biotic or aboitic materials.

Structural species Species that create or provide the physical structure of the environment, produce Huston (1994) variability in physical (e.g. microclimatic) conditions, provide resources, and in general create the habitat used by many other, generally smaller organisms.

Transformer An invasive plant that changes the character, condition, form or nature of an ecosystem Richardson et al. over a substantial area relative to the extent of that ecosystem (e.g. by donating limiting (2000) resources, promoting or suppressing fires, stabilizing or destabilizing sediments).

6 community-level impacts because their changes to the physical environment affect a broad range of species (Vitousek 1990). Such invaders can disrupt the local adaptation of native species to their environment (Byers 2002, 2009). Invasions by ecosystem engineers may also be especially difficult to control and manage. Recent theoretical models suggest that exotic ecosystem engineers are more likely to quickly colonize new habitats (i.e. are more invasive) than non- engineering exotic species (Cuddington & Hastings 2004; Gonzalez et al. 2008). Managing these invasions may also require additional effort, since simply removing an exotic engineer species might not effectively restore the invaded system if its engineering effects are not also managed (Byers et al. 2006). The effects of ecosystem engineers on other species are likely to be context-dependent. These effects are expected to scale with factors determining the abundance of the engineer or the engineered habitat (e.g. population density, regional spatial distribution, formation rate of constructs or artifacts and their durability; Jones et al. 1994, 1997), as well as biotic and abiotic properties of the system. For example, the effect of an ecosystem engineer is predicted to vary with the number of other species that depend on the resource it controls (Jones et al. 1997), its relative body size (Huston 1994; Jones & Gutiérrez 2007), and the presence of other organisms that perform the same engineering function (Dayton 1972; Mills et al. 1993). The effects of ecosystem engineers may also vary across gradients in habitat complexity (Crooks 2002), productivity (Wright & Jones 2004), resource availability (Gutiérrez et al. 2003; Wright et al. 2006), and physical stress (Bertness & Callaway 1994; Hacker & Gaines 1997; Bruno et al. 2003). These factors are likely to contribute to context-dependence in the ecological impacts of engineer species introduced to new regions. Given that environmental change alters natural selection pressures (Endler 1986), organisms that modify the environment can affect the evolution, as well as the distribution, of other species. Ecologists and evolutionary biologists are increasingly recognizing the phenomenon of niche construction, whereby habitat- modifying organisms can alter their own selective environment, as well as selection pressures acting on populations of other species (Odling-Smee 1988,

7 Odling-Smee et al. 2003). Recent evidence suggests that genetic variation in a habitat-forming species can have important consequences for entire communities (Whitham et al. 2006; Ungerer et al. 2008). Biological invasions, which involve the introduction of species into communities with which they may share no evolutionary history, provide a unique opportunity to test hypotheses regarding the role of prior evolutionary experience in determining the community- structuring effects of species interactions. Recent analyses suggest that impacts on native species are generally stronger for exotic predators and herbivores than for native consumers (McCarthy et al. 2006; Parker et al. 2006; Salo et al. 2007) and tend to be larger when exotic predators or competitors are dissimilar to native species in the recipient community (Ricciardi & Atkinson 2004; Strauss et al. 2006). These findings are consistent with ecological theory, which predicts that co-evolution reduces the intensity of negative species interactions (Connell 1980; Case & Bolger 1991; Callaway & Ridenour 2004; Cox & Lima 2006; Bruno et al. 2005). However, the effect of prior evolutionary experience on facilitative interactions is less clear (Bruno et al. 2005). Since habitat formation by exotic species is frequently associated with positive effects on native species (Crooks 2002; Rodriguez 2006), invasions by ecosystem engineers may provide insight into the role of prior evolutionary experience in determining the frequency or intensity of interspecific interactions, including facilitation, mediated by physical habitat modification. Two chapters of this thesis focus on the ecological impacts of some of the most well-studied exotic ecosystem engineers, the zebra mussel Dreissena polymorpha and quagga mussel D. bugensis. The zebra mussel is a small (< 5cm) bivalve native to the Black, Caspian and Azov seas (i.e. the Ponto-Caspian region; Stanczykowska 1977). Its spread through Europe was facilitated by the construction of shipping canals, and the mussel later traversed the Atlantic Ocean to North America, probably as veliger larvae in the ballast water of a cargo ship (McMahon 1996). It was first discovered in Lake St. Clair in 1988 (Hebert et al. 1989), but was present in the Great Lakes perhaps as early as 1986 (Carlton 2008). The quagga mussel, a small (< 5cm) bivalve native to the Dneiper River

8 drainage of Ukraine and Ponto-Caspian seas, was first detected in the Great Lakes in 1989 (Mills et al. 1996). Zebra and quagga mussels are morphologically and ecologically similar, although the quagga mussel is better able to colonize offshore soft sediments and can competitively displace the zebra mussel (Stoeckmann 2003). These mussels have invaded lakes and rivers throughout North America (US Geological Survey 2009) and Europe (DAISIE European Invasive Alien Species Gateway 2009), where they have been linked to dramatic changes in native communities, nutrient dynamics and energy flow pathways (Mellina et al. 1995; Vanderploeg et al. 2002; Ricciardi 2003; Strayer 2009).

1.2 Research objectives and approach

In this thesis, I examine the impacts of exotic species on aquatic macroinvertebrate communities via their alteration of benthic habitats. My objectives were to identify general patterns and test theories regarding the community-level impacts of exotic ecosystem engineers. Specifically, I develop and test models that incorporate invader-community interactions and use meta- analytical tools to link variation in impacts to characteristics of exotic organisms and recipient systems. I focus on invasion impacts on benthic macroinvertebrate communities because they are well-studied, and because changes in the abundance and community composition of macroinvertebrates affect their availability to consumers and the efficiency of aquatic ecosystem functions (Wallace & Webster 1996), and are commonly used as indicators in environmental monitoring (Rosenberg & Resh 1993). While my focus is primarily on community-level impacts, I also consider population-level impacts (e.g. changes in abundance or distribution). The inclusion of impacts at lower levels of ecological organization is justified given their value in developing a mechanistic understanding of the community-level impacts of exotic species (Simon & Townsend 2003). The development and evaluation of general models of impact requires techniques to quantitatively compare impacts across species and systems. To this

9 end, meta-analysis is a powerful tool for revealing generalizations about invasion impacts and for investigating patterns in their variation. By integrating investigations carried out under disparate conditions across multiple experiments or multiple sites, we can test hypotheses and explore large-scale patterns in community response that would not be feasible in a primary study. Despite this, meta-analysis has only recently been used to study invasion ecology (Levine et al. 2004; Agrawal et al. 2005, Parker et al. 2006; Salo et al. 2007; Liao et al. 2008), and rarely with regard to impact (but see McCarthy et al. 2006; Salo et al. 2007). These analyses have revealed interesting and sometimes counterintuitive patterns of invasion success and impact (e.g. Levine et al. 2004; Ricciardi & Atkinson 2004; Parker et al. 2006; Ricciardi & Ward 2006 – Appendix A). While the identification of general patterns of impact requires a broad, synthetic approach, a mechanistic understanding of impact ultimately requires experimental manipulation. I therefore utilized a combination of meta-analytical techniques and field experiments to address the following core questions:

1) Do native and exotic ecosystem engineers impact communities differently? (Chapters 2 & 5)

2) Does the impact of an exotic ecosystem engineer depend on whether it is functionally distinct in the invaded community? (Chapter 3)

3) How does the community-level impact of Dreissena vary with biotic and abiotic characteristics of invaded sites? (Chapter 4)

4) Does mussel patch topography affect the response of benthic macroinvertebrate communities to Dreissena shells? (Chapter 5)

It is the central premise of this thesis that variation in the impacts of exotic ecosystem engineers can be explained by considering the ecological characteristics or function of the invader in relation to the biotic and abiotic characteristics of the recipient system; that is, by accounting for variation in the traits of exotic ecosystem engineers and the biological (e.g. community

10 composition, evolutionary experience) and physical characteristics of invaded systems.

1.3 References cited

Agrawal, A.A., Kotanen, P.M., Mitchell, C.E., Power, A.G., Godsoe, W. & Klironomos, J. (2005). Enemy release? An experiment with congeneric plant pairs and diverse above- and belowground enemies. Ecology, 86, 2979-2989. Alcaraz, C., Bisazza, A. & García-Berthou, E. (2008). Salinity mediates the competitive interactions between invasive mosquitofish and an endangered fish. Oecologia, 155, 205-213. Anagnostakis, S.L. (1987). Chestnut blight: the classical problem of an introduced pathogen. Mycologia, 79, 23-37. Atkinson, I.A.E. (1985). The spread of commensal species of Rattus to oceanic islands and their effects on island avifauna. In P.J. Moore (Ed.), Conservation of Island Birds: Case Studies for the Management of Threatened Island Species (pp. 35-81). Cambridge, UK: International Council for Bird Preservation. Bertness, M.D. & Callaway, R. (1994). Positive interactions in communities. Trends in Ecology and Evolution, 9, 191-193. Blackburn, T.M., Cassey, P., Duncan, R.P., Evans, K.L. & Gaston, K.J. (2004). Avian extinction and mammalian introductions on oceanic islands. Science, 305, 1955-1958. Blackburn, T.M., Petchey, O.L., Cassey, P. & Gaston, K.J. (2005). Functional diversity of mammalian predators and extinction in island birds. Ecology, 86, 2916-2923. Braun-Blanquet, J. (1928). Pflanzensoziologie. Berlin: Springer-Verlag. Bruno, J.F., Fridley, J.D., Bromberg, K. & Bertness, M.D. (2005). Insights into biotic interactions from studies of species invasions. In D.F. Sax, S.D. Gaines & J.J. Stachowicz (Eds.), Species Invasions: Insights into Ecology, Evolution and Biogeography (pp. 13-40). Sunderland, MA: Sinauer.

11 Bruno, J.F., Stachowicz, J.J. & Bertness, M.D. (2003). Inclusion of facilitation into ecological theory. Trends in Ecology and Evolution, 18, 119-125. Byers, J.E. (2000). Differential susceptibility to hypoxia aids estuarine invasion. Marine Ecology Progress Series, 203, 123-132. Byers, J.E. (2002). Impact of non-indigenous species on natives enhanced by anthropogenic alteration of selection regimes. Oikos, 97, 449-458. Byers, J.E. (2009). Invasive animals in marshes: biological agents of change. In B. Silliman, E. Grosholz & M. Bertness (Eds.), Human Impacts on Salt Marshes: A Global Perspective (pp. 41-56). Berkeley, CA: University of California Press. Byers, J.E., Cuddington, K., Jones, C.G., Talley, T.S., Hastings, A., Lambrinos, J.G., Crooks, J.A. & Wilson, W.G. (2006). Using ecosystem engineers to restore ecological systems. Trends in Ecology and Evolution, 21, 493-500. Byers, J.E., Reichard, S., Randall, J.M., Parker, I.M., Smith, C.S., Lonsdale, W.M., Atkinson, I.A.E., Seastedt, T.R., Williamson, M., Chornesky, E. & Hayes, D. (2002). Directing research to reduce the impacts of nonindigenous species. Conservation Biology, 16, 630-640. Callaway, R.M. & Maron, J.L. (2006). What have exotic plant invasions taught us over the past 20 years? Trends in Ecology and Evolution, 21, 369-374. Callaway, R.M. & Ridenour, W.M. (2004). Novel weapons: invasive success and the evolution of increased competitive ability. Frontiers in Ecology and the Environment, 2, 436-443. Carlton, J.T. (2008). The zebra mussel Dreissena polymorpha found in North America in 1986 and 1987. Journal of Great Lakes Research, 34, 770-773. Case, T.J. & Bolger, D.T. (1991). The role of introduced species in shaping the abundance and distribution of island reptiles. Evolutionary Ecology, 5, 272- 290. Clavero, M. & García-Berthou, E. (2005). Invasive species are a leading cause of animal extinctions. Trends in Ecology and Evolution, 20, 110. Cohen, A.N. & Carlton, J.T. (1998). Accelerating invasion rate in a highly invaded estuary. Science, 279, 555-558.

12 Connell, J.H. (1980). Diversity and the coevolution of competitors, or the ghost of competition past. Oikos, 35, 131-138. Cox, J.G. & Lima, S.L. (2006). Naiveté and an aquatic-terrestrial dichotomy in the effects of introduced predators. Trends in Ecology and Evolution, 21, 674- 680. Crivelli, A.J. (1983). The destruction of aquatic vegetation by carp. Hydrobiologia, 106, 37-41. Crooks, J.A. (2002). Characterizing ecosystem-level consequences of biological invasions: the role of ecosystem engineers. Oikos, 97, 153-166. Crooks, J.A. (2009). The role of exotic marine ecosystem engineers. In G. Rilov & J.A. Crooks (Eds.), Biological Invasions in Marine Ecosystems: Ecological, Management, and Geographic Perspectives (pp. 287-304). Berlin: Springer- Verlag. Cuddington, K. & Hastings, A. (2004). Invasive engineers. Ecological Modelling, 178, 335-347. DAISIE European Invasive Alien Species Gateway (2009). Dreissena polymorpha and Dreissena bugensis pages. Available at: http://www.europe- aliens.org/. Accessed 24 November 2009. D‘Antonio, C.M., Tunison, J.T. & Loh, R.K. (2000). Variation in the impact of exotic grasses on native plant composition in relation to fire across an elevation gradient in Hawaii. Austral Ecology, 25, 507-522. Dayton, P.K. (1972). Toward an understanding of community reilience and the potential effects of enrichments to the benthos at McMurdo Sound, Antarctica. In B.C. Parker (Ed.), Proceedings of the Coloquium on Conservation Problems in Antarctica (pp. 81-96). Lawrence, KS: Allen Press. Dextrase, A.J. & Mandrak, N.E. (2006). Impacts of alien invasive species on freshwater fauna at risk in Canada. Biological Invasions, 8, 13-24. Diamond, J. & Case, T.J. (1986). Overview: introductions, extinctions, exterminations, and invasions. In J. Diamond & T.J. Case (Eds.), Community Ecology (pp. 65-79). New York, NY: Harper & Row.

13 Dukes, J.S. & Mooney, H.A. (2004). Disruption of ecosystem processes in western North America by invasive species. Revista Chilena de Historia Natural, 77, 411-437. Elton, C.S. (1958). The Ecology of Invasions by Animals and Plants. London: Methuen. Endler, J.A. (1986). Natural Selection in the Wild. Princeton, NJ: Princeton University Press. Flory, S.L., Rudgers, J.A. & Clay, K. (2007). Experimental light treatments affect invasion success and the impact of Microstegium vimineum on the resident community. Natural Areas Journal, 27, 124-132. García-Berthou, E. (2007). The characteristics of invasive fishes: what has been learned so far? Journal of Fish Biology, 71, 33-55. García-Berthou, E., Alcaraz, C., Pou-Rovira, Q., Zamora, L., Coenders, G. & Feo, C. (2005). Introduction pathways and establishment rates of invasive aquatic species in Europe. Canadian Journal of Fisheries and Aquatic Sciences, 62, 453-463. Gilpin, M. (1990). Ecological prediction. Science, 248, 88-89. Gonzalez, A., Lambert, A. & Ricciardi, A. (2008). When does ecosystem engineering cause invasion and species replacement? Oikos, 117, 1247-1257. Gutiérrez, J.L., Jones, C.G., Strayer, D.L., & Iribarne, O.O. (2003). Mollusks as ecosystem engineers: the role of shell production in aquatic habitats. Oikos, 101, 79-90. Hacker, S.D. & Gaines, S.D. (1997). Some implications of direct positive interactions for community species diversity. Ecology, 78, 1990-2003. Hayes, K.R. & Barry, S.C. (2008). Are there any consistent predictors of invasion success? Biological Invasions, 10, 483-506. Hebert, P.D.N., Muncaster, B.W. & Mackie, G.L. (1989). Ecological and genetic studies on Dreissena polymorpha (Pallas): a new mollusc in the Great Lakes. Canadian Journal of Fisheries and Aquatic Sciences, 46, 1587-1591. Huston M.A. (1994). Biological Diversity: The Coexistence of Species on Changing Landscapes. Cambridge, UK: Cambridge University Press.

14 Jokela, A. & Ricciardi, A. (2008). Predicting zebra mussel fouling on native mussels from physiochemical variables. Freshwater Biology, 53, 1845-1856. Jones, C.G. & Gutiérrez, J.L. (2007). On the purpose, meaning and usage of the physical ecosystem engineering concept. In K. Cuddington, J.E. Byers, W.G. Wilson & A. Hastings (Eds.), Ecosystem Engineers: Plants to Protists (pp. 3- 24). San Diego, CA: Elsivier. Jones, C.G., Lawton, J.H. & Shachak, M. (1994). Organisms as ecosystem engineers. Oikos, 69, 373-386. Jones, C.G., Lawton, J.H. & Shachak, M. (1997). Positive and negative effects of organisms as physical ecosystem engineers. Ecology, 78, 1946-1957. Kats, L.B. & Ferrer, R.P. (2003). Alien predators and amphibian declines: review of two decades of science and the transition to conservation. Diversity and Distributions, 9, 99-110. Kolar, C.S. & Lodge, D.M. (2001). Progress in invasion biology: predicting invaders. Trends in Ecology and Evolution, 16, 199-204. Krassoi F.R., Brown K.R., Bishop M.J., Kelaher B.P. & Summerhayes, S. (2008). Condition-specific competition allows coexistence of competitively superior exotic oysters with native oysters. Journal of Animal Ecology, 77, 5-15. Leppäkoski, E. & Olenin, S. (2000). Non-native species and rates of spread: lessons from the brackish Baltic Sea. Biological Invasions, 2, 151–163. Lester, P.J. (2005). Determinants for the successful establishment of exotic ants in New Zealand. Diversity and Distributions, 11, 279-288. Levine, J.M., Adler, P.B. & Yelenik, S.G. (2004). A meta-analysis of biotic resistance to exotic plant invasions. Ecology Letters, 7, 975-989. Liao, C., Peng, R., Luo, Y., Zhou, X., Wu, X., Fang, C., Chen, J. & Li, B. (2008). Altered ecosystem carbon and nitrogen cycles by plant invasion: a meta- analysis. New Phytologist, 177, 706-714. Lockwood, J.L., Hoopes, M.F. & Marchetti, M.P. (2007). Invasion Ecology. Malden, MA: Blackwell Publishing. Lockwood, J.L., Powell, R.D., Nott, M.P. & Pimm, S.L. (1997). Assembling ecological communities in time and space. Oikos, 80, 549-553.

15 Lodge, D.M. (1993). Biological invasions: lessons for ecology. Trends in Ecology and Evolution, 8, 133-136. Lodge, D.M., Stein, R.A., Brown, K.M., Covich, A.P., Brönmark, C., Garvey, J.E. & Klosiewski, S.P. (1998). Predicting impact of freshwater exotic species on native biodiversity: challenges in spatial scaling. Australian Journal of Ecology, 23, 53-67. McCarthy, J.M., Hein, C.L., Olden, J.D. & Vander Zanden, M.J. (2006). Coupling long-term studies with meta-analysis to investigate the impacts of non-native crayfish on zoobenthic communities. Freshwater Biology, 51, 224-235. McMahon, R.F. (1996). The physiological ecology of the zebra mussel, Dreissena polymorpha, in North America and Europe. American Zoologist, 36, 339-363. Mellina, E., Rasmussen, J.B. & Mills, E.L. (1995). Impact of mussel (Dreissena polymorpha) on phosphorus cycling and chlorophyll in lakes. Canadian Journal of Fisheries and Aquatic Sciences, 52, 2553-2573. Mills, E.L., Rosenberg, G., Spidle, A.P., Ludyanskiy, M., Pligin, Y. & May, B. (1996). A review of the biology and ecology of the quagga mussel (Dreissena bugensis), a second species of freshwater Dreissenid introduced to North America. American Zoologist, 36, 271-286. Mills, L.S., Soulé, M.E. & Doak, D.F. (1993). The keystone-species concept in ecology and conservation. BioScience, 43, 219-224. Moyle, P.B. & Light, T. (1996). Biological invasions of fresh water: empirical rules and assembly theory. Biological Conservation, 78, 149-161. Moyle, P.B. & Marchetti, M.P. (2006). Predicting invasion success: freshwater fishes in California as a model. BioScience, 56, 515-524. Odling-Smee, F.J. (1988). Niche constructing phenotypes. In H.C. Plotkin (Ed.), The Role of Behavior in Evolution (pp. 73-132). Cambridge, MA: MIT Press. Odling-Smee, F.J., Laland, K.N. & Feldman, M.W. (2003). Niche Construction: The Neglected Process in Evolution. Monographs in Population Biology 37. Princeton: Princeton University Press. Paine, R.T. (1966). Food web complexity and species diversity. The American Naturalist, 100, 65-74.

16 Parker, I.M., Simberloff, D., Lonsdale, W.M., Goodell, K., Wonham, M., Kareiva, P.M., Williamson, M.H., Von Holle, B., Moyle, P.B., Byers, J.E. & Goldwasser L. (1999). Impact: toward a framework for understanding the ecological effects of invaders. Biological Invasions, 1, 3-19. Parker, J.D., Burkepile, D.E. & Hay, M.E. (2006). Opposing effects of native and exotic herbivores on plant invasions. Science, 311, 1459-1461. Pimentel, D. (Ed.). (2002). Biological Invasions: Economic and Environmental Costs of Alien Plant, Animal, and Microbe Species. Boca Raton, FL: CRC Press. Rebek, E.J., Herms, D.A. & Smitley, D.R. (2008). Interspecific variation in resistance to Emerald ash borer (Coleoptera: Buprestidae) among North American and Asian ash (Fraxinus spp.). Environmental Entomology, 37, 242- 246. Ribera, M.A. & Boudouresque, C.F. (1995). Introduced marine plants, with special reference to macroalgae: mechanisms and impact. Progress in Phycological Research, 11, 188–268. Ricciardi, A. (2001). Facilitative interactions among aquatic invaders: is an "invasional meltdown" occurring in the Great Lakes? Canadian Journal of Fisheries and Aquatic Sciences, 58, 2513-2525. Ricciardi, A. (2003). Predicting the impacts of an introduced species from its invasion history: an empirical approach applied to zebra mussel invasions. Freshwater Biology, 48, 972-981. Ricciardi, A. & Atkinson, S.K. (2004). Distinctiveness magnifies the impact of biological invaders in aquatic ecosystems. Ecology Letters, 7, 781-784. Ricciardi, A. & Kipp, R. (2008). Predicting the number of ecologically harmful species in an aquatic system. Diversity and Distributions, 14, 374-380. Ricciardi, A. & Ward, J.M. (2006). Comment on "Opposing effects of native and exotic herbivores on plant invasions". Science, 313, 298. Richardson, D.M., Pyšek, P., Rejmánek, M., Barbour, M.G., Panetta, F.D. & West, C.J. (2000). Naturalization and invasion of alien plants: concepts and definitions. Diversity and Distributions, 6, 93-107.

17 Rodriguez, L.F. (2006). Can invasive species facilitate native species? Evidence of how, when, and why these impacts occur. Biological Invasions, 8, 927-939. Rosenberg, D.M. & Resh, V.H. (Eds.). (1993). Freshwater Biomonitoring and Benthic Macroinvertebrates. New York, NY: Chapman & Hall. Ross, D.J., Johnson, C.R. & Hewitt, C.L. (2003). Variability in the impact of an introduced predator (Asterias amurensis: Asteroidea) on soft-sediment assemblages. Journal of Experimental Marine Biology and Ecology, 288, 257- 278. Ruesink, J.L. (2005). Global analysis of factors affecting the outcome of freshwater fish introductions. Conservation Biology, 19, 1883-1893. Ruesink, J.L., Parker, I.M., Groom, M.J. & Kareiva, P.M. (1995). Reducing the risks of nonindigenous species introductions: guilty until proven innocent. BioScience, 45, 465-477. Salo, P., Korpimäki, E., Banks, P.B., Nordström, M. & Dickman, C.R. (2007). Alien predators are more dangerous than native predators to prey populations. Proceedings of the Royal Society B: Biological Sciences, 274, 1237-1243. Sax, D.F., Stachowicz, J.J., Brown, J.H., Bruno, J.F., Dawson, M.N., Gaines, S.D., Grosberg, R.K., Hastings, A., Holt, R.D., Mayfield, M.M., O‘Connor, M.I. & Rice, W.R. (2007). Ecological and evolutionary insights from species invasions. Trends in Ecology and Evolution, 22, 465-471. Simberloff, D. (1991). Keystone species and community effects of biological introductions. In L.R. Ginzburg (Ed.), Assessing Ecological Risks of Biotechnology (pp. 1-19). Boston, MA: Butterworth-Heinemann. Simberloff, D. & Von Holle, B. (1999). Positive interactions of nonindigenous species: invasional meltdown? Biological Invasions, 1, 21-32. Simon, K.S. & Townsend, C.R. (2003). Impacts of freshwater invaders at different levels of ecological organisation, with emphasis on salmonids and ecosystem consequences. Freshwater Biology, 48, 982-994. Stanczykowska, A. (1977). Ecology of Dreissena polymorpha (Pallas) (Bivalvia) in lakes. Polish Archives of Hydrobiology, 24, 461-530.

18 Stoeckmann, A. (2003). Physiological energetics of Lake Erie dreissenid mussels: a basis for the displacement of Dreissena polymorpha by Dreissena bugensis. Canadian Journal of Fisheries and Aquatic Sciences, 60, 126-134. Strauss, S.Y., Webb, C.O. & Salamin, N. (2006). Exotic taxa less related to native species are more invasive. Proceedings of the National Academy of Sciences of the United States of America, 103, 5841-5845. Strayer, D.L. (2009). Twenty years of zebra mussels: lessons from the mollusk that made headlines. Frontiers in Ecology and the Environment, 7, 135-141. Strayer, D.L., Eviner, V.T., Jeschke, J.M. & Pace, M.L. (2006). Understanding the long-term effects of species invasions. Trends in Ecology and Evolution, 21, 645-651. Stuart-Smith, R.D., Stuart-Smith, J.F., White, R.W.G. & Barmuta, L.A. (2007). The impact of an introduced predator on a threatened galaxiid fish is reduced by the availability of complex habitats. Freshwater Biology, 52, 1555-1563. USGS (US Geological Survey). (2009). Zebra and quagga mussel page. Available at: http://nas.er.usgs.gov/taxgroup/mollusks/zebramussel/. Viewed 24 November 2009. Ungerer, M.C., Johnson, L.C. & Herman, M.A. (2008). Ecological genomics: understanding gene and genome function in the natural environment. Heredity, 100, 178-183. Vanderploeg H.A., Nalepa T.F., Jude D.J., Mills E.L., Holeck K.T., Liebig J.R., Grigorovich I.A. & Ojaveer H. (2002). Dispersal and emerging ecological impacts of Ponto-Caspian species in the Laurentian Great Lakes. Canadian

Journal of Fisheries and Aquatic Sciences, 59, 1209-1228. Vilà, M. & Weiner, J. (2004). Are invasive plant species better competitors than native plant species? – evidence from pair-wise experiments. Oikos, 105, 229- 238. Vitousek, P.M. (1990). Biological invasions and ecosystem processes: towards an integration of population biology and ecosystem studies. Oikos, 57, 7-13.

19 Vitousek, P.M., D‘Antonio, C.M., Loope, L.L. & Westbrooks, R. (1996). Biological invasions as global environmental change. American Scientist, 84, 468-478. Wallace, J.B. & Webster, J.R. (1996). The role of macroinvertebrates in stream ecosystem function. Annual Review of Entomology, 41, 115-139. Wallentinus, I. & Nyberg, C.D. (2007). Introduced marine organisms as habitat modifiers. Marine Pollution Bulletin, 55, 323-332. Whitham, T.G., Bailey, J.K., Schweitzer, J.A., Shuster, S.M., Bangert, R.K., LeRoy, C.J., Lonsdorf, E.V., Allan, G.J., DiFazio, S.P., Potts, B.M., Fischer, D.G., Gehring, C.A., Lindroth, R.L., Marks, J.C., Hart, S.C., Wimp, G.M. & Wooley, S.C. (2006). A framework for community and ecosystem genetics: from genes to ecosystems. Nature Reviews Genetics, 7, 510-523. 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. Williamson, M. (1996). Biological invasions. London, UK: Chapman and Hall. Williamson, M. (1999). Invasions. Ecography, 22, 5-12. Wright, J.P. & Jones, C.G. (2004). Predicting effects of ecosystem engineers on patch-scale species richness from primary productivity. Ecology, 85, 2071- 2081. Wright, J.P., Jones, C.G., Boeken, B. & Shachak, M. (2006). Predictability of ecosystem engineering effects on species richness across environmental variability and spatial scales. Journal of Ecology, 94, 815-824.

20 Connecting statement 1–2

In Chapter 1, I briefly reviewed the ecological literature concerning the quantification of invasion impacts, with emphasis given to community-level impacts mediated by physical habitat modification or ecosystem engineering. This review highlighted a gap in our knowledge regarding the degree to which the impact of an ecosystem engineer might depend on a community‘s evolutionary experience with the engineer species or with functionally similar organisms (i.e. ecological naiveté). I address this general question in Chapters 2 and 3. In Chapter 2, I contrast the impacts of native and exotic ecosystem engineers on benthic macroinvertebrate communities. This is the first quantitative synthesis of native and exotic ecosystem engineering effects on community abundance and diversity across a broad range of aquatic habitats and taxa.

21

Chapter 2 – Community-level effects of native and exotic ecosystem

engineers: a meta-analysis

22 2.1 Abstract

Although it is generally well recognized that ecosystem engineering by plants and animals can have important consequences for ecological communities, the extent to which these consequences are shaped by a community‘s evolutionary experience with the ecosystem engineer is unknown. We conducted a meta- analysis of 130 studies that investigated the effects of 88 aquatic ecosystem engineers on benthic macroinvertebrate communities from shores and inland waters worldwide, and determined if the direction and magnitude of these effects depend on whether the engineer is native or exotic to the community. In general, ecosystem engineers that modified benthic habitats by adding physical structures (bioconstructors) had positive effects on the taxonomic richness and abundance of macroinvertebrates, while ecosystem engineers that modified benthic habitats by destabilizing sediments or removing epibenthic structures (bioturbators) had negative effects on macroinvertebrate numerical density. We found no significant differences between native and exotic ecosystem engineers regarding the direction and magnitude of these effects, nor did we find evidence of selection regime modification by exotic ecosystem engineers. However, habitat modification by exotic bioconstructors was associated with increased richness and numerical density of other exotic species, suggesting that invading bioconstructors can cause an invasional meltdown. This meta-analysis contributes to a growing understanding of when and how biotic interactions involving exotic species differ from those involving native species.

23 2.2 Introduction

Ecologists have long sought to understand the extent to which species‘ abundances and distributions are regulated by interactions with other species. Early theoretical (e.g. Lotka 1925; Volterra 1926) and empirical (e.g. Gause 1934) studies focused on competition and predation as the dominant biotic interactions structuring communities, and these remain a major focus of ecological research today. However, a growing body of evidence points to habitat modification by organisms — or ecosystem engineering (sensu Jones et al. 1994, 1997) — as an important determinant of population- and community-level variables (Wright & Jones 2006; Cuddington et al. 2007). In aquatic systems, native ecosystem engineers create habitats such as coral reefs, seagrass meadows and kelp forests that commonly support high levels of local biodiversity, and conservation ecologists have drawn attention to the community-level consequences of their ongoing loss (Coleman & Williams 2002; Boogert et al. 2006). Concurrently, researchers have highlighted the threat posed by the introduction of nonindigenous ecosystem engineers, which can dramatically transform the physical structure of invaded systems, thereby altering community structure and ecosystem function (Crooks 2002; Wallentinus and Nyberg 2007; Crooks 2009). Quantifying and comparing the effects of native and exotic ecosystem engineers is therefore of critical importance in order to predict, prevent and mitigate the harmful effects of declining native ecosystem engineers and the impacts of ecosystem engineering exotic species. Two forms of ecosystem engineering that often have important and perhaps predictable consequences for aquatic communities are bioconstruction and bioturbation (Reise 2002; Meysman et al. 2006). The ecological significance of bioconstruction — broadly defined as the formation of physical structures by organisms — has been recognized in the development of the foundation species (Dayton 1972) and structural species (Huston 1994) concepts, which both describe organisms that create physical habitat upon which many other species depend. Herein, we consider bioconstruction to include autogenic ecosystem

24 engineering (i.e. habitat modification via the presence of body structures; Jones et al. 1997) by plants and sessile invertebrates, as well as a subset of allogenic ecosystem engineering (i.e. habitat modification via mechanical means; Jones et al. 1994) such as the active accretion of particulate matter by tube-building polychaete worms. It is hypothesized bioconstruction generally has positive effects on local species richness, owing to increased habitat complexity or sediment stability (Crooks 2002; Reise 2002). Bioturbation — broadly defined as the biological reworking of sediments — was the topic of Charles Darwin‘s final book, on earthworms (Darwin 1881). This subset of allogenic ecosystem engineering (sensu Jones et al. 1994) results from activities such as burrowing by mobile infauna and feeding by bulldoze grazers or animals that laterally ‗plough‘ the sediment surface (Reise 2002; Meysman et al. 2006). It is hypothesized that bioturbation generally inhibits benthic communities by removing biogenic epibenthic structures, and thus reducing habitat complexity, or by destabilizing sediments (Crooks 2002; Reise 2002). We focus our meta-analysis on these two forms of ecosystem engineering. Several predictions in the ecological literature pertain to how the effect of an ecosystem engineer on other species might depend on whether it is native or exotic. Jones et al. (1997) suggested that dependence of other species on the habitats or resources provided by an ecosystem engineer might increase over evolutionary time. Alternatively, the absence of evolutionary experience (i.e. ecological naiveté) has been posited to explain the vulnerability of oceanic islands and freshwater systems to the harmful effects of exotic species, given their evolutionary isolation relative to continental and marine systems (Diamond & Case 1986; Ricciardi & Atkinson 2004; Cox & Lima 2006). Although this line of reasoning has focused on the naiveté of native prey species to the novel hunting tactics of introduced predators (Cox & Lima 2006), it may logically extend to other mechanisms of invasion impact, provided they are agents of natural selection. The capacity for ecosystem engineering by organisms to exert strong selection pressures on populations of other species is increasingly recognized (e.g. niche construction sensu Odling-Smee 1988; Olding-Smee et al. 2003), as is its

25 probable role in shaping macroevolutionary patterns (Thayer 1979; Erwin 2008). Therefore, based on the premise that evolutionary experience reduces the strength of negative interactions (Levin et al. 1982; Case & Bolger 1991) and may promote facilitation (Jones et al. 1997), we predict that exotic ecosystem engineers will have more negative effects than native ecosystem engineers on community diversity and abundance. Given that naiveté to the effects of exotic species is expected to be more pronounced in insular systems, where communities lack evolutionary experience with a functionally diverse array of native species (Cox & Lima 2006), we further predict that the effects of exotic ecosystem engineers are more severe in freshwater versus marine systems. The selection regime modification hypothesis states that habitat alteration by exotic ecosystem engineers can disrupt the local adaptation of native species to their environment and thereby reduce a competitive advantage of native species over other exotic species in habitats modified by exotic ecosystem engineers (Byers 2002, 2009). Accordingly, we predict that habitat modification by exotic species disproportionately benefits other exotic species at the expense of native species. Direct or indirect positive interactions among exotic species may be common in aquatic systems (Ricciardi 2001) and are hypothesized to accelerate the accumulation of established exotic species and amplify their impacts – i.e. produce an invasional meltdown (Simberloff & Von Holle 1999; Parker et al. 1999). The questions of whether and how invasional meltdown occurs are important for conservation because this concept implies that each prevented introduction may have disproportionate benefits (Parker et al. 1999). Therefore, we determined whether habitat modification by exotic ecosystem engineers generally facilitates other exotic species. The objectives of our study were to (i) identify general patterns in the community-level effects of aquatic ecosystem engineers, (ii) quantitatively compare these effects among native and exotic engineers, and (iii) test hypotheses from the ecological literature with respect to the community-level impacts of exotic ecosystem engineers. We tested the following hypotheses:

26 1) Ecosystem engineers that create structurally complex habitats (bioconstructors) generally increase local species richness and abundance.

2) Ecosystem engineers that disturb or destabilize sediments via their movement or feeding activities (bioturbators) generally decrease local species richness and abundance.

3) Within each of these groups (bioconstructors and bioturbators), native ecosystem engineers have more positive (or less negative) effects on local species richness and abundance than do exotic ecosystem engineers.

4) Exotic ecosystem engineers have more deleterious impacts in freshwater systems than in marine systems.

5) Other exotic species are disproportionately favoured, relative to native species, in habitats modified by exotic ecosystem engineers (selection regime modification hypothesis).

6) Exotic ecosystem engineers generally facilitate other exotic species (invasional meltdown hypothesis).

We focus on aquatic ecosystem engineers because they encompass a wide variety of organisms in freshwater and marine systems where they often produce large community-level effects, and because much invasion ecology theory has been developed in terrestrial systems and involves predatory or competitive interactions, leaving other areas ripe for exploration.

2.3 Methods

2.3.1 Data collection

To test these hypotheses, we conducted a meta-analysis of published studies designed to measure the effects of habitat modification by aquatic ecosystem engineers on benthic macroinvertebrate communities. Relevant studies were identified by searching the online databases ISI Web of Science and BIOSIS

27 Previews using combinations of the following keywords: exotic, nonindigenous, introduced, invasi*, alien, ecosystem engineer*, habitat engineer*, habitat modif*, bioconstruct*, bioturbat*, macroinvertebrate*, invertebrate*, diversity, species richness, addition, removal, uninvaded, experiment*, effect*, and impact*. Additional studies were located by examining the references of included studies and review articles (e.g. Crooks 2002; Gutiérrez et al. 2003; Wallentinus & Nyberg 2007). We included studies that measured the effect of habitat modification by an ecosystem engineer either by experimentally manipulating its presence or by comparing communities in engineer-modified (e.g. mussel bed, oyster reef, seagrass meadow) and unmodified habitats. Ecosystem engineers were not necessarily identified as such by the primary authors; rather, our inclusion criterion was that the study measured the effect of habitat modification on the benthic macroinvertebrate community. Each study was scrutinized to determine whether the measured effect was predominantly the result of bioconstruction or bioturbation (as defined above), and studies were excluded if they did not measure the effect of either form of ecosystem engineering. Included studies also met the following requirements for computation of effect sizes: (1) contain replicate samples of macroinvertebrate communities in the engineered habitat (e.g. mussel bed) and in the unmodified habitat (e.g. ambient substrata), and (2) report at least one of the following community response variables (i.e. excluding the ecosystem engineer species) in both habitats: total taxonomic richness (# taxa), taxa density (# taxa/unit area), numerical density (# individuals/unit area), biomass density (biomass/unit area), diversity indices (e.g. Shannon index) or evenness indices (e.g. Simpson‘s evenness). We considered multiple results within a single paper to be independent if they involved different ecosystem engineers or study sites. When a paper reported both experimental and correlational data for a given ecosystem engineer, we included only the experimental data. When an experiment involved multiple treatment densities of an ecosystem engineer, we included the density closest to its ambient density at the site. When data from multiple sampling dates were reported, we included only the final sampling date.

28 Ecosystem engineer origin (native or exotic) was identified using information provided in primary papers and confirmed using additional sources: the United States Geological Survey‘s Nonindigenous Aquatic Species database (http://nas.er.usgs.gov); the IUCN/SSC Invasive Species Specialist Group‘s Global Invasive Species Database (http://www.issg.org/database); the DAISIE European Invasive Alien Species Gateway (http://www.europe-aliens.org); and published surveys (e.g. Leppäkoski et al. 2002). We excluded ecosystem engineer species whose origin was unknown or contentious (i.e. cryptogenic species). Because fewer studies involving exotic (versus native) ecosystem engineers were available, we included native engineers that were taxonomically similar (e.g. belonging to the same genus or family) to the exotic engineers in our dataset whenever possible; this was done in order to avoid taxonomic biases in the dataset. Hence, our meta-analysis is not an exhaustive synthesis of all available studies involving the effects of native ecosystem engineers. In addition, we recorded whether effect size estimates were calculated using experimental data as well as the duration of experiments, and determined whether studies were conducted in freshwater, estuarine or marine systems. We also determined whether each responding macroinvertebrate species was native, exotic or cryptogenic for the subset of studies that reported species lists for both habitats or changes in species-level abundances.

2.3.2 Statistical analysis

Ecosystem engineer effects on macroinvertebrate communities were quantified using the loge response ratio, ln(X+E/X-E), where X+E was the mean value of the response variable in patches modified by the ecosystem engineer and X-E was the mean value of the response variable in unmodified patches (Hedges et al. 1999). Thus, positive values indicate an increase in a community response variable, and negative values a decrease, in response to the ecosystem engineer. This effect size metric was selected because it has a straightforward biological interpretation in terms of the proportional change in a response variable, as well as desirable

29 statistical properties for use in meta-analysis (Hedges et al. 1999; Lajeunnesse & Forbes 2003), and because its ease of application in theoretical modeling may facilitate theory development (e.g. Shurin & Seabloom 2005). Impacts on the species richness and numerical density of native and exotic macroinvertebrates were calculated by tallying the number of native and exotic species present in species lists or by adding the numerical densities of individual species identified as native or exotic, respectively, in each habitat. We tested our hypotheses using nonparametric meta-analysis with statistical significance evaluated using randomization tests and generated bias- corrected bootstrapped 95% confidence intervals with 9999 random iterations (Adams et al. 1997; Rosenberg et al. 2000). Analyses were performed on unweighted effect size estimates because the information required to compute weights (sample sizes and variance estimates) was inconsistently reported, and because weights were incalculable for one of our community response variables, total taxonomic richness (Hedges et al. 1999). Unweighted meta-analysis has an increased Type II error rate compared to weighted meta-analysis, but this may be offset by increased statistical power resulting from the inclusion of a larger number of effect size estimates (Gurevitch & Hedges 1999; Lajeunnesse & Forbes 2003). To ensure that our results were not driven by a few well-studied engineer species, we also compared the effects of native and exotic ecosystem engineers using a restricted dataset that contained only one effect size estimate per native or exotic ecosystem engineer; this was the mean effect of ecosystem engineers for which we had more than one effect size estimate for a given response variable. We carried out additional analyses to evaluate the degree to which methodological variables (study design and experimental duration) may have influenced our results. All analyses were conducted using MetaWin v.2 (Rosenberg et al. 2000).

2.4 Results

Our meta-analysis included estimates of the community-level effects of 88

30 ecosystem engineers derived from 130 publications (Table B.1 in Appendix B). Of these, 51 publications reported the effects of exotic ecosystem engineers, 64 reported the effects of native ecosystem engineers, and 15 reported the effects of both native and exotic ecosystem engineers. Of the ecosystem engineers in our dataset, 33% (29 species) were studied in their introduced range, 64% (56 species) were studied in their native range, and 3% (3 species) were studied in both native and introduced ranges. Most studies investigated the effects of bioconstruction (119 studies, 74 ecosystem engineers), while only 14 studies investigated the effects of bioturbation (15 ecosystem engineers). Effects on macroinvertebrate communities were most often measured in terms of numerical density (111 studies, 72 ecosystem engineers), total taxonomic richness (69 studies, 54 ecosystem engineers), and taxa density (69 studies, 51 ecosystem engineers), while biomass density (26 studies, 23 ecosystem engineers), diversity indices (41 studies, 36 ecosystem engineers), and evenness indices (32 studies, 32 ecosystem engineers), or sufficient information to compute such indices, were less commonly reported.

2.4.1 Hypotheses 1, 2 and 3

Among the studies involving bioconstruction, the community-level effects of exotic ecosystem engineers were not significantly different from those of native ecosystem engineers (Fig. 2.1). Native and exotic bioconstructors had indistinguishable effects on total taxonomic richness (P = 0.63), taxonomic density (P = 0.39), numerical density (P = 0.77), biomass density (P = 0.76), diversity indices (P = 0.12), and evenness indices (P = 0.20). On average, habitat modification by bioconstructors increased benthic macroinvertebrate total taxonomic richness by 31%, taxa density by 25%, numerical density by 85% and biomass density by 197% relative to unmodified habitats. Mean effects on diversity and evenness indices were not significantly different from zero. Analyses conducted on a restricted dataset that included only one effect size estimate per native or exotic ecosystem engineer yielded generally concordant

31 ) 3.0 Native 2.5 24 Exotic

ln +E/-E ln

2.0 18

1.5

1.0 100 80

46 62 56 0.5 72 45 20 34 16 0.0

on macroinvertebrate communities macroinvertebrate on

Ecosystem engineer effect ( effect engineer Ecosystem -0.5 Total Taxa Numerical Biomass Diversity Evenness richness density density density

Community response variable

Figure 2.1. Mean effect (ln+E/-E) of bioconstruction by native and exotic ecosystem engineers on macroinvertebrate community response variables. Error bars are 95% bias-corrected bootstrapped confidence intervals; effects are significant if these do not overlap zero. Numbers indicate sample sizes.

32 results, although the mean effect of exotic ecosystem engineers on total taxonomic richness was nonsignificant (n = 18, mean = 0.067, 95% CI = -0.236 to 0.298). Our dataset contained a sufficient number of effect size estimates to allow only a statistical comparison of the effects of native and exotic bioturbators on macroinvertebrate numerical density. The effects of native and exotic bioturbators on numerical density did not differ significantly (P = 0.19), although the mean effect of exotic bioturbators was negative, while the mean effect of native bioturbators was not significantly different from zero (Fig. 2.2). This result did not change when only one effect size estimate per engineer species was included in the analysis (P = 0.23). Native bioturbators had nonsignificant mean effects on total taxonomic richness (n = 6, mean = 0.043, 95% CI = -0.293 to 0.334), taxa density (n = 8, mean = -0.129, 95% CI = -0.463 to 0.235), and biomass density (n = 4, mean = -0.434, 95% CI = -1.705 to 0.699). When data from native and exotic ecosystem engineers were pooled, the effects of bioconstruction (mean = 0.615, 95% CI = 0.450 to 0.775) and bioturbation (mean = -0.588, 95% CI = -1.099 to -0.101) on the numerical density of benthic macroinvertebrates were significantly different (P = 0.0003).

2.4.2 Hypotheses 4, 5 and 6

Because studies involving bioturbation were limited, we tested our final three hypotheses only among the studies involving bioconstruction. The effects of exotic bioconstructors on total taxonomic richness (P = 0.0003) and taxa density (P = 0.01) varied significantly among marine, estuarine and freshwater systems, both metrics being positive in freshwater systems but not significantly different from zero in marine and estuarine systems (Fig. 2.3); freshwater systems did not differ from marine and estuarine systems after the effects of Dreissena spp. were excluded from the analyses (P > 0.4). The effects of exotic bioconstructors on numerical density (P = 0.24), biomass density (P = 0.10), diversity indices (P =

33 2 Native Exotic

) on) the 1 100 80

ln +E/-Eln 10

0 7

-1

Ecosystem engineer effect ( engineer effect Ecosystem

numerical density of benthic macroinvertebrates numerical density -2 Bioconstruction Bioturbation

Dominant mode of ecosystem engineering

Figure 2.2. Mean effect (ln +E/-E) of bioconstruction and bioturbation by native and exotic ecosystem engineers on macroinvertebrate numerical density. Error bars are 95% bias-corrected bootstrapped confidence intervals; effects are significant if these do not overlap zero. Numbers indicate sample sizes.

34 1.5 1.5 a) Taxonomic richness b) Taxa density 17 15 1.0 1.0 26 32 0.5 13 18 26 3 0.5 43 15 0.0 14 10

0.0 -0.5 Native Exotic -1.0 -0.5

6 2 c) Numerical density d) Biomass density 24 3 7 43 4 45 29 21 1 27 11 4 2 7

) on macroinvertebrate communities 0 0

-2 -1

ln +E/-E -4 -2 1.0 1.0 e) Diversity indices f) Evenness indices 5 0.5 0.5 24 22 4 9 11 20 5 10 0.0 3 0.0 -0.5

-0.5 -1.0

Ecosystem engineer effect ( Ecosystem -1.0 -1.5 Marine Estuarine Freshwater Marine Estuarine Freshwater

Aquatic system Figure 2.3. Mean effect (ln+E/-E) of bioconstruction by native and exotic ecosystem engineers on macroinvertebrate communities in marine, estuarine and freshwater systems. Community-level effects were quantified as changes in (a) total taxonomic richness, (b) taxonomic density, (c) numerical density, (d) biomass density, (e) diversity indices and (f) evenness indices. Error bars are 95% bias-corrected bootstrapped confidence intervals; effects are significant if these do not overlap zero. Numbers indicate sample sizes.

35 0.22), and evenness indices (P = 0.23) did not differ among marine, estuarine and freshwater systems. Different results were obtained for native bioconstructors. Their effects on numerical density varied among aquatic systems (P = 0.003), being positive in marine and estuarine systems but not significantly different from zero in freshwater systems (Fig. 2.3). Their effects on total taxonomic richness (P = 0.28), taxa density (P = 0.24), biomass density (P = 0.79), diversity indices (P = 0.11), and evenness indices (P = 0.75) did not differ among aquatic systems. Contrary to our fifth hypothesis (selection regime modification), exotic bioconstructors did not appear to facilitate exotic macroinvertebrates more than native macroinvertebrates; they had equally positive effects on native and exotic taxonomic richness (P = 0.79) and numerical density (P = 0.75; Fig. 2.4). Likewise, bioconstruction by native ecosystem engineers had equally positive effects on the taxonomic richness (P = 0.63) and numerical density (P = 0.33) of native and exotic macroinvertebrates (Fig. 2.4). The statistical significance of these comparisons did not differ when analyses were conducted using the restricted dataset, although some of the 95% confidence intervals overlapped zero. The positive effects of exotic ecosystem engineers on the taxonomic richness and abundance of other exotic species supports our sixth hypothesis (invasional meltdown). Methodological differences among the studies did not appear to influence our results. For the studies involving bioconstruction, none of the community response variables were related to experimental duration (P > 0.2), nor did they depend on whether the ecosystem engineer was an autotroph or a heterotroph (P > 0.1). Although positive effects on total taxonomic richness (P = 0.04) and numerical density (P = 0.004) were significantly larger for correlational studies that compared areas with and without ecosystem engineers than for studies that experimentally manipulated the presence of engineers, the effects of native and exotic ecosystem engineers on these response variables did not differ when separate analyses were conducted for correlational and experimental studies (P >

36 Native Exotic ecosystem engineers ecosystem engineers 0.8 P = 0.63 P = 0.79

0.6

0.4 19 19 22 0.2 22

Total richnessTotal

) on macroinvertebrate 0.0

ln +E/-E 5 P = 0.33 P = 0.75 4

3

2 16

1 16 33 33

Numerical density 0

Ecosystem engineer effect ( Ecosystem -1 Native Exotic Native Exotic

Origin of responding species

Figure 2.4. Mean effect (ln+E/-E) of bioconstruction by native and exotic ecosystem engineers on the total taxonomic richness (e.g. # species) and numerical density (# individuals/unit area) of native and exotic macroinvertebrates. Error bars are 95% bias-corrected bootstrapped confidence intervals; effects are significant if these do not overlap zero. Numbers indicate sample size.

37 0.3). All other community response variables were not influenced by study design (P > 0.1).

2.5 Discussion

2.5.1 Bioconstruction versus bioturbation

We found that bioconstruction by native and exotic ecosystem engineers generally increased the local taxonomic richness and abundance of benthic macroinvertebrates (Fig. 2.1). These findings are consistent with the expectation that bioconstructors generally have facilitative effects on other species and drive increases in local species richness, resulting from increased habitat complexity or sediment stability (Crooks 2002; Reise 2002; hypothesis 1). Positive effects of bioconstructors on macroinvertebrate richness and abundance may also reflect the amelioration of physical or physiological stress, provision of competitor- or predator-free space, or increased resource availability in biogenic habitats (Bruno et al. 2003; Gutiérrez et al. 2003). Our meta-analysis further suggests that the direction and magnitude of these effects are equivalent for native and exotic ecosystem engineers. However, positive effects on local species richness do not indicate that bioconstructors facilitate all macroinvertebrates. Indeed, many of the exotic bioconstructors included in this analysis (e.g. Dreissena polymorpha, Musculista senhousia) have been implicated in the severe decline or extirpation of some native macroinvertebrates (Crooks 2002; Ricciardi 2003). As well, many of the exotic plants in our meta-analysis have been shown to negatively impact native flora (e.g. Williams & Smith 2007). Exotic bioconstructors can also reduce habitat complexity at landscape scales by replacing a mosaic of different habitats, and may therefore reduce species richness at these larger spatial scales (Jones et al. 1997; Crooks 2002). Furthermore, while we found that bioconstructors generally had positive effects on local taxonomic richness, their effects on diversity and evenness indices were nonsignificant, indicating that increases in the number of taxa were offset by reductions in evenness (Fig. 2.1); hence, some

38 species appear to benefit from the modified habitats more than others. This variation in response across macroinvertebrate taxa does not appear to reflect differences in their evolutionary experience with ecosystem engineering organisms (Fig. 2.4), and instead is probably driven by differences in species traits. For example, several of the studies included in our meta-analysis noted differences in the response of macroinvertebrates according to their functional feeding group (e.g. Everett 1994; Hily & Bouteille 1999), habitat preferences (e.g. Bouma et al. 2009; Buschbaum et al. 2009), mobility (e.g. Castilla et al. 2004; Kimbro & Grosholz 2006), or mode of development (e.g. Crooks et al. 1998; Crooks & Khim 1999). Future work could examine whether the characteristics of responding species that are associated with facilitative effects differ across groups of ecosystem engineers (e.g. native versus exotic, plants versus animals) or aquatic systems. In contrast to bioconstructors, ecosystem engineers that destabilize sediments or remove epibenthic structures (i.e. bioturbators) are hypothesized to generally have inhibitory effects on benthic communities and drive reductions in local species richness (Crooks 2002; Reise 2002; hypothesis 2). Our meta- analysis supported this prediction for effects on macroinvertebrate numerical density (Fig. 2.2). Impacts of exotic bioturbators on macroinvertebrate communities are not commonly quantified, but their effects on other ecosystem components are more frequently studied; for example, several exotic bioturbators, such as the burrowing isopod Sphaeroma quoyanum and the catadromous Chinese mitten crab Eriocheir sinensis, have accelerated shoreline and riverbank erosion, with negative consequence for native vegetation (Talley et al. 2001; Wallentinus & Nyberg 2007). Alternatively, bioturbation may play an important role in maintaining biodiversity and ecosystem function through the irrigation and aeration of sediments (Widdicombe et al. 2004; Volkenborn et al. 2007). In our meta-analysis, some native bioturbators had positive effects on macroinvertebrate taxonomic richness (e.g. the burrowing bivalves Mulinia edulis and Venus antiqua) while others had inhibitory effects (e.g. the burrowing ghost shrimp Neotrypaea californiensis), resulting in nonsignificant overall effects. This

39 suggests that the effect of bioturbation varies greatly across engineer taxa or communities, making generalizations difficult.

2.5.2 Ecological naiveté

Two of our hypotheses (3 and 4) considered the impacts of exotic ecosystem engineers from an evolutionary perspective, both of which tested aspects of ecological naiveté. Contrary to our third hypothesis, the average magnitude of community-level effects did not differ among native and exotic ecosystem engineers for either bioconstructors or bioturbators, although limited availability of data precluded statistical comparison of multiple community response variables for the latter. We expected native ecosystem engineers to have larger positive effects (or smaller negative effects) on species richness and abundance as a result of shared evolutionary history with the macroinvertebrate community. Our findings contrast those of two meta-analytical comparisons of the effects of native and exotic species. Parker et al. (2006) found that exotic generalist herbivores cause greater reductions in the abundance and richness of native plants than do native generalist herbivores. Likewise, Salo et al. (2007) found that exotic mammalian and avian predators have more severe impacts than native predators on native prey species. These differences in the intensity of consumer effects were interpreted to reflect native species‘ lack of evolutionary experience with exotic consumers. In contrast, our results suggest that impacts of exotic species mediated by changes to the physical structure of benthic habitats are not contingent on prior evolutionary experience with the ecosystem engineering organism. However, an alternative explanation is that most macroinvertebrate communities have acquired experience with functionally similar ecosystem engineers or ecosystem engineer archetypes (i.e. a set of ecosystem engineers that use similar morphology or behaviour in modifying habitats) and are therefore not naïve to their effects. It has been observed that freshwater systems are more vulnerable to the effects of exotic predators than are marine systems, possibly because the

40 evolutionary isolation of freshwater systems has precluded adaptation to a diverse array of predator archetypes (Cox & Lima 2006). On this basis, we predicted that exotic ecosystem engineers would have more detrimental effects in freshwater compared to marine systems (hypothesis 4). Our meta-analysis did not support this prediction with respect to the community-level impacts of exotic bioconstructors. Instead, exotic bioconstuctors had larger positive effects on local species richness in freshwater systems (Fig. 2.3). This pattern was driven by large positive effects of exotic fouling bivalves, zebra and quagga mussels (Dreissena spp.), on macroinvertebrate richness in freshwater systems that, unlike marine systems, have no native fouling bivalves. This suggests that the introduction of an ecosystem engineer representing a novel functional group may be associated with more facilitative effects on native species, at least for bioconstructors.

2.5.3 Selection regime modification and invasional meltdown

It is hypothesized that habitat modification by exotic ecosystem engineers may eliminate a competitive advantage of native species that are otherwise adapted to their local environment, thus enhancing the establishment and population growth of other exotic species at the expense of native species (Byers 2002, 2009). In our meta-analysis, exotic bioconstructors did not appear to disproportionately facilitate other exotic species (hypothesis 5). Instead, exotic and native ecosystem engineers facilitated both native and exotic species equally (Fig. 2.4). These positive effects may therefore reflect changes to habitat conditions that benefit macroinvertebrates generally (e.g. increased resources or microhabitat complexity), regardless of evolutionary experience with the habitat-modifying species. Nonetheless, positive effects of exotic ecosystem engineers on the taxonomic richness and numerical density of other exotic species could produce an invasional meltdown (i.e. accelerated rate of invasion or amplified impacts of other exotic species in the recipient system; Simberloff & Von Holle 1999; Ricciardi 2001; hypothesis 6). For example, elevated densities of exotic species inside engineered habitats could accelerate the establishment and population

41 growth of exotics by increasing their chances of overcoming Allee effects (i.e. positive density-dependence), and are expected to enhance impacts on native species given that impacts are correlated with the abundance of exotic species (Parker et al. 1999; Ricciardi 2003). Habitat-forming exotic species may also facilitate the establishment and spread of other exotic species by harbouring them during transport (Florl et al. 2004), or by disrupting the relationship between resource availability (e.g. habitat space) and invasion success (Stachowicz & Byrnes 2006). While our dataset was insufficient to evaluate these hypotheses with respect to the effects of bioturbators, one study found that sediment disturbance by native bioturbators (heart urchins) can reduce the colonization success of exotic species (Lohrer et al. 2008). It would be interesting to discover whether the enhancement of biotic resistance to invasion by native bioturbators is a general phenomenon.

2.5.4 Implications for theory and management

Our results suggest that native and exotic ecosystem engineers, particularly bioconstructors, have parallel effects on benthic macroinvertebrate diversity and abundance. Therefore, the same models may explain variation in the strength of their community-level effects. Our results further suggest that exotic ecosystem engineers will have more dramatic impacts on recipient communities when they invade areas not previously occupied by a similar native habitat-modifying species (e.g. nonindigenous Spartina spp. invading previously unvegetated mudflats in Europe and North America) or when they replace a native engineer species that modifies habitats in a different way (e.g. an introduced bioturbator replacing a native bioconstructor — such as the removal of submersed vegetation by introduced carp, Cyprinus carpio). While general patterns of the community- level effects of exotic bioturbators remain undiscovered, our findings suggest that exotic bioturbators have more inhibitory effects than exotic bioconstructors on local macroinvertebrate communities and may therefore be a priority for management (Fig. 2.2). However, these impacts must be considered alongside

42 impacts on other ecosystem components (e.g. native vegetation) and biotic resistance to future invasions (Fig. 2.4). In addition, the increased richness and density of exotic species inside habitats created by native and exotic bioconstructors suggests that these habitats may be useful target areas for the early detection of new invasions.

2.5.5 Conclusions and future directions

Compared to trophic interactions, our understanding of ecosystem engineer- mediated species interactions is still in its infancy. Our meta-analysis suggests that the community-level impacts of ecosystem engineers, particularly those resulting from bioconstruction, are not generally contingent on a community‘s prior evolutionary experience with the engineer species, in contrast to the impacts of predators and herbivores (Salo et al. 2007; Parker et al. 2006; Cox & Lima 2006). Future research could examine variation in the effects of native and exotic ecosystem engineers across different organisms and ecosystem components (e.g. plant communities, sediment properties). From a management perspective, it would be useful to determine (i) whether these impacts are correlated; (ii) whether habitats created by native and exotic ecosystem engineers provide the same ecosystem services and functions (e.g. nursery habitat for fish, productivity, nutrient cycling); and (iii) how the impacts of ecosystem engineers vary with species traits, community composition or habitat characteristics.

2.6 Acknowledgements

The authors thank S. Ellwood, K. O‘Sullivan and A. Hassan for help with data entry. Funding for this research was provided by a Canada Graduate Scholarship to J.M. Ward by NSERC Canada.

43 2.7 References cited

Adams, D.C., Gurevitch, J. & Rosenberg, M.S. (1997). Resampling tests for meta-analysis of ecological data. Ecology, 78, 1277-1283. Boogert, N.J., Paterson, D.M. & Laland, K.N. (2006). The implications of niche construction and ecosystem engineering for conservation biology. BioScience, 56, 570-578. Bouma, T.J., Ortells, V. & Ysebaert, T. (2009). Comparing biodiversity effects among ecosystem engineers of contrasting strength: macrofauna diversity in Zostera noltii and Spartina anglica vegetations. Helgoland Marine Research, 63, 3-18. Bruno, J.F., Stachowicz, J.J. & Bertness, M.D. (2003). Inclusion of facilitation into ecological theory. Trends in Ecology and Evolution, 18, 119-125. Buschbaum, C., Dittmann, S., Hong, J.S., Hwang, I.S., Strasser, M., Thiel, M., Valdivia, N., Yoon, S.P. & Reise, K. (2009). Mytilid mussels: global habitat engineers in coastal sediments. Helgoland Marine Research, 63, 47-58. Byers, J.E. (2002). Impact of non-indigenous species enhanced by anthropogenic alteration of selection regimes. Oikos, 97, 449-458. Byers, J.E. (2009). Invasive animals in marshes: biological agents of change. In B. Silliman, E. Grosholz & M. Bertness (Eds.), Human Impacts on Salt Marshes: A Global Perspective (pp. 41-56). Berkeley, CA: University of California Press. Case, T.J. & Bolger, D.T. (1991). The role of introduced species in shaping the abundance and distribution of island reptiles. Evolutionary Ecology, 5, 272- 290. Castilla, J.C., Lagos, N.A. & Cerda, M. (2004). Marine ecosystem engineering by the alien ascidian Pyura praeputialis on a mid-intertidal rocky shore. Marine Ecology Progress Series, 268, 119-130. Coleman, F.C. & Williams, S.L. (2002). Overexploiting marine ecosystem engineers: potential consequences for biodiversity. Trends in Ecology and Evolution, 17, 40-44.

44 Cox, J.G. & Lima, S.L. (2006). Naiveté and an aquatic-terrestrial dichotomy in the effects of introduced predators. Trends in Ecology and Evolution, 21, 674-680. Crooks, J.A. (1998). Habitat alteration and community-level effects of an exotic mussel, Musculista senhousia. Marine Ecology Progress Series, 162, 137-152. Crooks, J.A. (2002). Characterizing ecosystem-level consequences of biological invasions: the role of ecosystem engineers. Oikos, 97, 153-166. Crooks, J.A. (2009). The role of exotic marine ecosystem engineers. In G. Rilov & J.A. Crooks (Eds.), Biological Invasions in Marine Ecosystems: Ecological, Management, and Geographic Perspectives (pp. 287-304). Berlin: Springer- Verlag. Crooks, J.A. & Khim, H.S. (1999). Architectural vs. biological effects of a habitat- altering, exotic mussel, Musculista senhousia. Journal of Experimental Marine Biolology and Ecology, 240, 53-75. Cuddington, K., Byers, J.E., Wilson, W.G. & Hastings, A. (Eds.). (2007). Ecosystem Engineers: Plants to Protists. San Diego, CA: Elsevier. Darwin, C. (1881). The Formation of Vegetable Mould Through the Action of Worms, With Observations on their Habits. London: John Murray. Dayton, P.K. (1972). Toward an understanding of community resilience and the potential effects of enrichments to the benthos at McMurdo Sound, Antarctica. In B.C. Parker (Ed.), Proceedings of the Coloquium on Conservation Problems in Antarctica (pp.81-96). Lawrence, KS: Allen Press. Diamond, J. & Case, T.J. (1986). Overview: introductions, extinctions, exterminations, and invasions. In J. Diamond & T.J. Case (Eds.), Community Ecology (pp. 65-79). New York, NY: Harper & Row. Erwin, D.H. (2008). Macroevolution of ecosystem engineering, niche construction and diversity. Trends in Ecology and Evolution, 23, 304-310. Everett, R.A. (1994). Macroalgae in marine soft-sediment communities: effects on benthic faunal assemblages. Journal of Experimental Marine Biology and Ecology, 175, 253-274.

45 Florl, O., Pool, T.K. & Inglis, G.J. (2004). Positive interactions between nonindigenous species facilitate transport by human vectors. Ecological Applications, 14, 1724-1736. Gause, G.F. (1934). The Struggle for Existence. Baltimore, MD: Williams and Wilkens. Gurevitch, J. & Hedges, L.V. (1999). Statistical issues in ecological meta-analysis. Ecology, 80, 1142-1149. Gutiérrez, J.L., Jones, C.G., Strayer, D.L., & Iribarne, O.O. (2003). Mollusks as ecosystem engineers: the role of shell production in aquatic habitats. Oikos, 101, 79-90. Hedges, L.V., Gurevitch, J. & Curtis, P.S. (1999). The meta-analysis of response ratios in experimental ecology. Ecology, 80, 1150-1156. Hily, C. & Bouteille, M. (1999). Modifications of the specific diversity and feeding guilds in an intertidal sediment colonized by an eelgrass meadow (Zostera marina) (Brittany, France). Life Sciences, 322, 1121-1131. Huston, M.A. (1994). Biological Diversity: The Coexistence of Species on Changing Landscapes. Cambridge, UK: Cambridge University Press. Jones, C.G., Lawton, J.H. & Shachak, M. (1994). Organisms as ecosystem engineers. Oikos, 69, 373-386. Jones, C.G., Lawton, J.H. & Shachak, M. (1997). Positive and negative effects of organisms as physical ecosystem engineers. Ecology, 78, 1946-1957. Kimbro, D.L. & Grosholz, E.D. (2006). Disturbance influences oyster community richness and evenness, but not diversity. Ecology, 87, 2378-2388. Lajeunnesse, M.J. & Forbes, M.R. (2003). Variable reporting and quantitative reviews: a comparison of three meta-analytical techniques. Ecology Letters, 6, 448-454. Leppäkoski, E., Gollasch, S. & Olenin, S. (Eds.). (2002). Invasive Aquatic Species of Europe: Distribution, Impacts and Management. Dordrecht, The Netherlands: Kluwer Academic Publishers. Levin, B.R., Allison, A.C., Bremermann, H.J., Cavalli-Sforza, L.L., Clarke, B.C., Frentzel-Beyme, R., Hamilton, W.D., Levin, S.A., May, R.M. & Thieme, H.R.

46 (1982). Evolution of parasites and hosts. In R.M. Anderson & R.M. May (Eds.), Population Biology of Infectious Diseases (pp. 213-243). Berlin: Springer-Verlag. Lohrer, A.M., Chiaroni, L.D., Hewitt, J.E. & Thrush, S.F. (2008). Biogenic disturbance determines invasion success in a subtidal soft-sediment system. Ecology, 89, 1299-1307. Lotka, A.J. (1925). Elements of Physical Biology. Baltimore, MD: Williams and Wilkens. Meysman, F.J.R., Middelburg, J.J. & Heip, C.H.R. (2006). Bioturbation: a fresh look at Darwin‘s last idea. Trends in Ecology and Evolution, 21, 688-695. Odling-Smee, F.J. (1988). Niche constructing phenotypes. In H.C. Plotkin (Ed.), The Role of Behavior in Evolution (pp. 73-132). Cambridge, MA: MIT Press. Odling-Smee, F.J., Laland, K.N. & Feldman, M.W. (2003). Niche Construction: The Neglected Process in Evolution. Monographs in Population Biology, 37. Princeton: Princeton University Press. Parker, J.D., Burkepile, D.E. & Hay, M.E. (2006). Opposing effects of native and exotic herbivores on plant invasions. Science, 311, 1459-1461. Parker, I.M., Simberloff, D., Lonsdale, W.M., Goodell, K., Wonham, M., Kareiva, P.M., Williamson, M.H., Von Holle, B., Moyle, P.B., Byers, J.E. & Goldwasser L. (1999). Impact: toward a framework for understanding the ecological effects of invaders. Biological Invasions, 1, 3-19. Reise, K. (2002). Sediment mediated species interactions in coastal waters. Journal of Sea Research, 48, 127-141. Ricciardi, A. (2001). Facilitative interactions among aquatic invaders: is an "invasional meltdown" occurring in the Great Lakes? Canadian Journal of Fisheries and Aquatic Sciences, 58, 2513-2525. Ricciardi, A. (2003). Predicting the impacts of an introduced species from its invasion history: an empirical approach applied to zebra mussel invasions. Freshwater Biology, 48, 972-981. Ricciardi, A. & Atkinson, S.K. (2004). Distinctiveness magnifies the impact of biological invaders in aquatic ecosystems. Ecology Letters, 7, 781-784.

47 Rosenberg, M.S., Adams, D.C. & Gurevitch, J. (2000). MetaWin: Statistical Software for Meta-Analysis. Version 2. Sunderland, MA: Sinauer Associates. Salo, P., Korpimäki, E., Banks, P.B., Nordström, M. & Dickman, C.R. (2007). Alien predators are more dangerous than native predators to prey populations. Proceedings of the Royal Society of London B: Biological Sciences, 274, 1237- 1243. Shurin, J.B. & Seabloom, E.W. (2005). The strength of trophic cascades across ecosystems: predictions from allometry and energetics. Journal of Animal Ecology, 74, 1029-1038. Simberloff, D & Von Holle, B. (1999). Positive interactions of nonindigenous species: invasional meltdown? Biological Invasions, 1, 21-32. Stachowicz, J.J. & Byrnes, J.E. (2006). Species diversity, invasion success, and ecosystem functioning: disentangling the influence of resource competition, facilitation, and extrinsic factors. Marine Ecology Progress Series, 311, 251- 262. Talley, T.S., Crooks, J.A. & Levin, L.A. (2001). Habitat utilization and alteration by the invasive burrowing isopod, Sphaeroma quoyanum, in California salt marshes. Marine Biology, 138, 561-573. Thayer, C.W. (1979). Biological bulldozers and the evolution of marine benthic communities. Science, 203, 458-461. Volkenborn, N., Polerecky, L., Hedtkamp, S.I.C., van Beusekom, J.E.E. & de Beer, D. (2007). Bioturbation and bioirrigation extend the open exchange regions in permeable sediments. Limnology and Oceanography, 52, 1898- 1909. Volterra, V. (1926). Variations and fluctuations of the numbers of individuals in animal species living together. Reprinted in R.N. Chapman (1931), Animal Ecology (pp. 409-448). New York, NY: McGraw-Hill. Wallentinus, I. & Nyberg, C.D. (2007). Introduced marine organisms as habitat modifiers. Marine Pollution Bulletin, 55, 323-332. Widdicombe, S., Austen, M.C., Kendall, M.A., Olsgard, F., Schaanning, M.T., Dashfield, S.L. & Needham, H.R. (2004). Importance of bioturbators for

48 biodiversity maintenance: indirect effects of fishing disturbance. Marine Ecology Progress Series, 275, 1-10. Williams, S.L. & Smith, J.E. (2007). A global review of the distribution, taxonomy, and impacts of introduced seaweeds. Annual Review of Ecology, Evolution and Systematics, 38, 327-359. Wright, J.P. & Jones, C.G. (2006). The concept of organisms as ecosystem engineers ten years on: progress, limitations, and challenges. BioScience, 56, 203-209.

49 Connecting statement 2–3

In Chapter 2, I examined the reliance of ecosystem engineering impacts on a community‘s evolutionary experience with the engineering organism by comparing the impacts of native and exotic ecosystem engineers. A meta-analysis confirmed that bioconstruction by ecosystem engineers generally increases aquatic macroinvertebrate abundance and diversity, and revealed that the magnitude of these effects does not differ among native and exotic engineer species. In Chapter 3, I investigate another aspect of the role of prior evolutionary experience in governing the impacts of exotic ecosystem engineers. Here, I test a hypothesis that relates variation in the impacts of introduced species to their functional distinctiveness from native species in the invaded community.

50

Chapter 3 – The impacts of exotic ecosystem engineers are mediated by their

phylogenetic distinctiveness

51 3.1 Abstract

The ecological naiveté hypothesis posits that exotic species have larger detrimental impacts on native species that lack evolutionary experience with functionally similar organisms. Although often invoked to explain the disproportionate impacts of introduced predators and pathogens, the influence of naiveté on other species interactions is less understood. Using phylogenetic relatedness to native species as a proxy for functional similarity, we conducted a meta-analytical test of naiveté to the impacts of exotic ecosystem engineers. Ecosystem engineers are potentially high-impact exotic species because they can profoundly alter the physical structure of invaded habitats. Contrary to the ecological naiveté hypothesis, we found that exotic engineers belonging to distinct genera or families generally had more facilitative effects on native species than did exotic engineers having native congeneric or confamilial species in the invaded region. This result was driven by differences in the number and magnitude of negative effects on native species: declines in native species abundances were fewer and less severe in response to habitat modification by phylogenetically distinct engineers, while the average magnitude of positive effects did not vary with the phylogenetic relatedness of exotic engineers to native species. These patterns may result from the creation of novel resources or refugia by distinct engineers. Our results suggest that the impact of an invading ecosystem engineer is contingent on its phylogenetic relationship with the invaded community, and reveal a general pattern in the context-dependence of invasions that may inform risk assessment.

52 3.2 Introduction

Biological invasions pose a global threat to the conservation of aquatic systems (Carlton 1996; Ruiz et al. 1999; Clavero & García-Berthou 2005; Dextrase & Mandrak 2006). However, the ecological impacts of exotic species are highly variable; some invasions profoundly alter recipient communities, while others appear to have weak or negligible impacts (Williamson 1996; García-Berthou et al. 2005). Effective conservation requires a predictive understanding of invasions that are likely to have a large detrimental effect on native species. One promising approach links an exotic species‘ impact to its functional distinctiveness within the invaded community, recognizing that some exotic species are more alien to the community than others. The ecological naiveté hypothesis posits that exotic species have stronger and more deleterious effects on recipient communities that lack evolutionary experience with functionally similar organisms (Diamond & Case 1986; Ricciardi & Atkinson 2004; Cox & Lima 2006). Ecological naiveté is an intriguing yet scarcely tested hypothesis; most of the existing evidence is derived from observations regarding the vulnerability of species in insular systems (e.g. oceanic islands, freshwater lakes) to introduced predators (Case & Bolger 1991; Cox & Lima 2006). For example, Atkinson (1985) noted an apparent reduction in the vulnerability of avifauna to predation by introduced rats on islands inhabited by native rats or land crabs compared to islands lacking functionally similar native predators. More recently, the hypothesis has received support from studies using the phylogenetic relatedness of exotic species to native species as a proxy for functional similarity (Ricciardi & Atkinson 2004; Strauss et al. 2006a). Although species-depauperate systems likely exhibit naiveté to a broader array of exotic species, introduced species with novel traits have produced large impacts even in species-rich systems (e.g. the speciose chiclid fishes of Lake Victoria were naïve to the effects of the predatory Nile Perch, Lates niloticus; Ricciardi & Atkinson 2004). Theory suggests that native species‘ prior evolutionary experience with functionally similar species can reduce the magnitude of the exploitative or competitive impacts of an exotic

53 species (e.g. through the evolution of anti-predator defenses or tolerance to allelopathic chemicals; Cox & Lima 2006; Callaway & Ridenour 2004), but its effect on the strength of facilitative interactions is less understood (Bruno et al. 2005). Moreover, tests of the ecological naiveté hypothesis have so far been concerned with negative effects on native species, and no quantitative analysis involving the prevalence or magnitude of facilitative effects has yet been made. Among the introduced species considered to have large impacts on native communities are ecosystem engineers (sensu Jones et al. 1994, 1997), which can profoundly modify the physical characteristics of invaded habitats (Vitousek 1990; Crooks 2002; Wallentinus & Nyberg 2007; Bouma et al. 2009; Byers 2009). The ecological significance of habitat creation by native animals and plants, upon which entire communities often depend, has been underscored by several ecological concepts including foundation species (Dayton 1972), keystone modifier (Mills et al. 1993), structural species (Huston 1994), and physical ecosystem engineering (Jones et al. 1994, 1997). Likewise, the niche construction concept in evolutionary biology has highlighted the potential for ecosystem engineering to have frequent and important evolutionary consequences (Odling-Smee 1988; Odling-Smee et al. 2003). Like native ecosystem engineers, exotic engineer species can facilitate natives via the provision of structurally complex habitat and spatial refugia, and by modulating abiotic processes (Crooks 2002; Rodriguez 2006). However, unlike ecosystem engineering by native species, habitat conversion by exotic engineers may produce a novel condition to which native species are not adapted, altering selection regimes to a condition beyond the evolutionary experience of the recipient community (Byers 2002, 2009). Because introduced ecosystem engineers that are functionally distinct from native species are particularly likely to do this, we expect such exotic species to have more deleterious effects on native communities. The objective of this study was to test the ecological naiveté hypothesis with respect to the impacts of exotic ecosystem engineers in benthic habitats. We hypothesized that exotic engineers are associated with larger declines in native species abundances in communities in which the exotic engineer species is

54 phylogenetically distinct, based on the assumption that phylogenetic distinctiveness is correlated with functional distinctiveness (Harvey & Pagel 1991; Webb et al. 2002; Cavender-Bares et al. 2004). We focus on exotic species that modify benthic habitats by adding physical structures (e.g. aquatic plants, mussels, tube-building polychaete worms). Such species often exert a significant influence on the composition of aquatic communities (Crooks 2002; Reise 2002; Wallentinus & Nyberg 2007; Bouma et al. 2009; Byers 2009), and have both positive and negative effects on native species, allowing us to examine the role of naiveté in both facilitative and antagonistic interactions. Our original intention was to contrast the impacts of these ecosystem engineers with the impacts of engineers that disturb sediments (such as the burrowing isopod Sphaeroma quoyanum; Talley et al. 2001), which may have opposing effects on benthic communities (Crooks 2002; Reise 2002; Chapter 2), but too few studies involving the latter are presently available to allow a meaningful comparison.

3.3 Methods

3.3.1 Data collection

We conducted a literature search for studies designed to measure the impact of habitat modification by exotic ecosystem engineers on benthic macroinvertebrate communities. Computerized searches were carried out in Science Citation Index Expanded (SCI-Expanded; 1899 – April 2009) and BIOSIS Previews (1969 – April 2009) using combinations of the following terms: exotic, nonindigenous, introduced, invasi*, alien, ecosystem engineer*, habitat engineer*, habitat modif*, macroinvertebrate*, invertebrate*, addition, removal, uninvaded, experiment*, effect* and impact*. Additional data sources were located by examining the references of included studies, as well as those of review articles (e.g. Crooks 2002; Wallentinus & Nyberg 2007). We included studies that compared benthic macroinvertebrate communities in ecosystem engineer-modified and unmodified habitats (i.e. in the presence and absence of an exotic ecosystem engineer) and

55 reported the information required to compute effects sizes (see below) for at least five invertebrate taxa identified to at least the genus level. In order for a study to be included, the habitat modifying species did not need to be identified as an ‗ecosystem engineer‘ by the primary authors; rather, our inclusion criterion was that the study measured the effect of habitat creation or modification by an exotic species on the benthic macroinvertebrate community. Because our hypothesis relates an exotic engineer‘s impact to the prior evolutionary experience of native species, we excluded impacts on exotic or cryptogenic taxa (i.e. species of unknown origin). We also excluded studies that compared macroinvertebrate assemblages in habitats created by native and exotic ecosystem engineers, as well as studies investigating the effects of aquaculture or the impact of management or control efforts rather than the exotic species itself. To reduce pseudoreplication in our dataset, we included only one impact (effect size) estimate per native species per invaded site for each exotic ecosystem engineer species. We used experimental (versus observational) data when available. When experiments used multiple treatment densities of the engineer species, we included the treatment that most closely resembled ambient density at the site. When data from multiple sampling dates were available for the same site, we included only the most recent sampling date. Finally, when at least five experimental studies were available for the same exotic engineer species, we avoided including additional observational studies for that species so as to prevent well-studied exotic species from dominating the dataset. Following this procedure, our dataset contained information for a total of 462 impacts by 16 exotic ecosystem engineer species derived from 30 publications (Table C.1 in Appendix C). For each invasion, we determined whether the exotic ecosystem engineer belonged to a genus or family shared by native species in the recipient region (references provided in Appendix C).

3.3.2 Statistical analysis

Two steps were taken to reduce potential confounding of our results. First, we

56 tested the ecological naiveté hypothesis using different effect size metrics and both parametric and non-parametric analyses. Second, we used multiple meta- regression (Lipsey & Wilson 2001) to statistically control for the potentially confounding effects of several factors that were associated with the impacts of exotic ecosystem engineers.

We used the loge response ratio (lnR) to quantify the impacts of exotic ecosystem engineers on native species for studies that reported the mean density (e.g. individuals/m2) of native species in engineer-modified and unmodified habitats (Hedges et al. 1999). Because an abundance of zero in one habitat may be associated with an ecologically significant effect but renders this effect size metric incalculable, we added the smallest detectable difference (e.g. 0.01) to both numerator and denominator prior to loge transformation. We also quantified impacts using Hedges‘ d (Hedges and Olkin 1985) for studies that reported either (1) mean densities, together with estimates of their variability (e.g. standard error) in engineer-modified and unmodified habitats, or (2) sufficiently detailed results of statistical tests comparing species abundances in these two habitats. The loge response ratio measures an engineer‘s effect as the relative change in a native species‘ abundance and Hedges‘ d as the standardized mean difference. For both metrics, positive values indicate increased native species abundance and negative values decreased abundance associated with the exotic ecosystem engineer. Although the two metrics were correlated (r = 0.69, P < 0.00001, n = 257), variable reporting of results by primary studies meant that both could not always be calculated (Table C.1 in Appendix C). Therefore, following Englund et al. (1999), we report results for both metrics to prevent our conclusions from being reliant on methodological choices. For analyses involving Hedges‘ d, we used parametric mixed-effects meta- analysis (Gurevitch & Hedges 2001) to compare the impacts of exotic engineers belonging to a taxon (genus or family) shared with native species in the invaded region to the impacts of engineers belonging to a distinct taxon. Given that prior experience with functionally similar species may influence the strength of facilitative and antagonistic interactions differently, we conducted additional

57 analyses comparing the average magnitude of positive (d > 0) and negative effects (d < 0). Effect sizes of zero magnitude (i.e. native species occurred in equal abundance in engineer-modified and unmodified habitats) were excluded from these analyses. Each effect size was weighted by its inverse variance. Analogous analyses for the lnR dataset were conducted on unweighted effect size estimates using nonparametric meta-analysis, with statistical significance evaluated and confidence intervals generated using resampling techniques with 9999 random iterations (Adams et al. 1997). To facilitate comparison of our results with those of previous studies, we also determined whether the frequency of impacts involving a greater than 80% increase or decrease in a native species‘ abundance differed between exotic engineer species belonging to shared and distinct genera or families using Fisher Exact tests (Sokal & Rohlf 1995). Meta-analyses of studies that examined the responses of multiple species are likely to avoid biases that may be present in studies measuring impacts on particular species. Nonetheless, we checked for publication bias by visually inspecting funnel plots (Rosenberg et al. 2000). In addition to tests of our primary hypothesis, we conducted analyses to rule out the potential confounding effects of three moderator variables expected to influence engineer impacts: (1) the length of time (years) between the establishment of the exotic ecosystem engineer and the date on which the study was conducted (Strayer et al. 2006), (2) whether the investigation was limited to infaunal invertebrates dwelling within the sediments (Bouma et al. 2009), and (3) whether the introduced ecosystem engineer was a heterotroph (invertebrate) or an autotroph (alga or vascular plant). Potential confounding effects of moderator variables significantly related to impact (Hedges‘ d) in univariate analyses were evaluated using inverse variance weighted multiple meta-regression with stepwise removal of non-significant terms (Lipsey & Wilson 2001). Time since invasion spanned two orders of magnitude and therefore was log10(x+1) transformed prior to analysis. Parametric analyses were conducted using R v.2.9.1 (R Core Development Team 2009), whereas nonparametric analyses were conducted using MetaWin v.2 (Rosenberg et al. 2000).

58 3.4 Results

The impact (Hedges‘ d) of exotic ecosystem engineers on the abundance of native macroinvertebrate taxa differed significantly among engineers belonging to shared and distinct taxa (P < 0.0001 for genera and P< 0.00001 for families; Fig. 3.1). On average, the effect of exotic engineers belonging to genera or families not represented in the native community was positive, while the effect of engineers belonging to taxa shared by native species was not significantly different from zero. Similar results were obtained for analyses involving lnR at the genus (P = 0.0001, Fig. 3.2a) and family level, however the effect of engineers belonging to shared families was significant and positive in this case (P = 0.0001, Fig. 3.2b). Engineers belonging to shared and distinct families increased native species abundances by a mean factor of 1.5 and 5.5, respectively. Each engineer species in our dataset had both positive and negative effects on native species, and more than half were associated with large negative effects. Analyses of the average magnitude of positive effects revealed mixed results. For the Hedges‘ d dataset, the average magnitude of positive effects did not differ significantly among engineers representing shared and distinct genera (P = 0.31, Fig. 3.3a), while engineers representing distinct families had larger positive effects (P = 0.02, Fig. 3.3b). For the lnR dataset, the magnitude of positive effects did not depend on the relatedness of exotic engineers to native species at either the genus (P = 0.15, Fig. 3.4a) or family level (P = 0.77, Fig. 3.4b). Fisher Exact tests failed to reject the null hypothesis that exotic engineers belonging to shared and distinct taxa are equally likely to have large positive impacts (>80% increase) on native species, although the test conducted at the level of genera was marginally insignificant (distinct genera tended to have more large positive effects; P = 0.06, Table 3.1). The average magnitude of negative effects was consistently larger for exotic engineers that were more closely phylogenetically related to native species. This result was obtained in analyses involving Hedges‘ d conducted at the genus (P = 0.02, Fig. 3.3c) and family level (P = 0.003, Fig. 3.3d), as well as analyses

59 1.5

) A P < 0.0001

d 1.0

0.5 216

0.0 74

-0.5

-1.0

on native species (Hedges'

Exotic ecosystem engineerExotic effect ecosystem -1.5 Shared Distinct Engineer Genus

1.5

) B P < 0.00001

d 1.0 76

0.5

214 0.0

-0.5

-1.0

on native species (Hedges'

Exotic ecosystem engineerExotic effect ecosystem -1.5 Shared Distinct

Engineer Family Figure 3.1. Mean effects (Hedges‘ d) of exotic ecosystem engineers on native species for engineer species representing shared (native) or distinct taxa in the invaded region. Error bars are 95% confidence intervals; effects are significant if these do not overlap zero. Confidences intervals and significance tests for the difference between shared and distinct taxa were generated using a weighted mixed-effects model. Numbers indicate sample sizes.

60 2 A P = 0.0001

) 1 308

lnR

0 121

-1

on native species (

Exotic ecosystem engineerExotic effect ecosystem -2 Shared Distinct Engineer Genus

3 B P = 0.0001

) 2

lnR 92

1

337 0

-1

on native species (

Exotic ecosystem engineerExotic effect ecosystem -2 Shared Distinct Engineer Family

Figure 3.2. Mean effects (lnR) of exotic ecosystem engineers on native species for engineer species representing shared (native) or distinct taxa in the invaded region. Error bars are 95% bias-corrected bootstrapped confidence intervals; effects are significant if these do not overlap zero. Confidence intervals and significance tests for the difference between shared and distinct taxa were generated using resampling methods with 9999 iterations. Numbers indicate sample sizes.

61 2.0 A P = 0.31 B P = 0.02

)

d 1.5

59 1.0 140 35 116

0.5

0.0

-0.5

native species native (Hedges'

Positive effects of engineers on Positive -1.0 1.0 C P = 0.02 D P = 0.003

)

d 0.5

0.0 16

-0.5 69 92 39 -1.0

-1.5

native species (Hedges'native

Negative effects of engineers of effects on Negative -2.0 Shared Distinct Shared Distinct

Engineer Genus Engineer Family

Figure 3.3. Mean positive and negative effects (Hedges‘ d) of exotic ecosystem engineers on native species for engineer species representing shared (native) or distinct taxa in the invaded region. Error bars are 95% confidence intervals; effects are significant if these do not overlap zero. Confidences intervals and significance tests for the difference between shared and distinct taxa were generated using a weighted mixed-effects model. Numbers indicate sample size.

62 4 A P = 0.15 B P = 0.77

) 3

lnR 202 185 75 2 58

1

0

on speciesnative ( -1

Positive effects of engineers

-2 2 C P = 0.0003 D P = 0.07

) 1

lnR 0

-1 16 98 -2 142 60

on speciesnative ( -3

Negative effectsNegative of engineers -4 Shared Distinct Shared Distinct

Engineer Genus Engineer Family

Figure 3.4. Mean positive and negative effects (lnR) of exotic ecosystem engineers on native species for engineer species representing shared (native) or distinct taxa in the invaded region. Error bars are 95% bias-corrected bootstrapped confidence intervals; effects are significant if these do not overlap zero. Confidence intervals and significance tests for the difference between shared and distinct taxa were generated using resampling methods with 9999 iterations. Numbers indicate sample sizes.

63 Table 3.1. Results from two-tailed Fisher Exact tests comparing the frequency of large positive and negative impacts on native species among exotic ecosystem engineers belonging to native (shared) or distinct taxa.

Large impacts (> 80% change) Small impacts (< 80% change) Shared Distinct Shared Distinct P-value I. Positive impacts Genus 29 103 92 205 0.06 Family 97 35 240 57 0.1 II. Negative impacts Genus 39 35 82 273 <0.0001 Family 70 4 267 88 <0.0001

64 using lnR conducted at the genus (P = 0.0003, Fig. 3.4c) and family level, although the latter test was marginally insignificant (P = 0.07, Fig. 3.4d). Likewise, exotic engineers that belonged to native genera or families were more likely to have large negative impacts (>80% reduction) on native species (P < 0.0001, Table 3.1). The overall pattern (Fig. 3.1 and 3.2) was generated by these differences in the size of negative effects, together with differences in their relative number: engineers belonging to native taxa had near 1:1 ratios of positive to negative effects, while those belonging to distinct taxa had 2–5 times as many positive effects compared with negative effects (Fig. 3.3 and 3.4). In univariate meta-regression analyses, engineer impacts were significantly related to time since invasion, whether the engineer was a heterotroph or an autotroph, and whether studies were limited to infauna (Table 3.2). However, only time since invasion and phylogenetic distinctiveness were retained by the multiple meta-regression. This result was obtained regardless of whether engineer distinctiveness was considered at the genus or family level, although the multiple meta-regression analysis involving engineer genus may have been influenced by multicollinearity. All of the engineer species representing shared genera in our dataset were autotrophs. Removal of heterotrophic engineers from the comparison of shared and distinct genera did not change the direction or significance of our results (Hedges‘ d: P = 0.004; lnR: P = 0.0002). Impacts of exotic ecosystem engineers on native species were negatively correlated with the length of time the exotic engineer had been present at a site (Fig. 3.5).

3.5 Discussion

3.5.1 The ecological naiveté hypothesis

The impacts of exotic ecosystem engineers on native macroinvertebrates were correlated with their phylogenetic relatedness to native species, but surprisingly not in the direction predicted by the ecological naiveté hypothesis. Instead, exotic

65 Table 3.2. Results from univariate weighted meta-regression analyses for the relationship between moderator variables and engineer impacts (Hedges‘ d). The table gives the number of effect sizes (n), significance level (P), and the slope ( SE) and intercept ( SE) of the regression model. Moderator variable n P Slope Intercept Engineer organisma 290 0.003 0.38 (0.13) 0.08 (0.10) Infauna onlyb 290 0.02 -0.34 (0.14) 0.40 (0.07) Years since invasionc 290 <0.00001 -0.80 (0.15) 1.25 (0.19) a Coded as autotroph (0) or heterotroph (1). b Coded as yes (1) or no (0). c log10(x+1) transformed.

66

Figure 3.5. Variation in the effects of exotic ecosystem engineers on native species with time since invasion. Regression line fitted by inverse variance weighted meta-regression, using a mixed-effects model (P < 0.00001). Circle size is proportional to the weight of effect size estimates (n = 290). Years since invasion is log10(x + 1) transformed.

67 engineers belonging to distinct taxa had more facilitative effects overall (Fig. 3.1 and 3.2), and their negative impacts were fewer and less severe (Fig. 3.3 and 3.4, Table 3.1). These results were robust to choices of effect size metric and analysis, and persisted after accounting for the influence of several potentially confounding moderator variables. Because researchers are more likely to study nonindigenous species suspected of having large effects, the subset of exotic ecosystem engineers included in our meta-analysis is probably biased toward high-impact species; 56% of these species had at least one large negative impact (i.e. >80% change) on a native species, compared to 11% of exotic species in the analysis by Ricciardi & Atkinson (2004). Such a bias would make our statistical tests more conservative and is unlikely to account for our findings. Moreover, the sizeable proportion of exotic ecosystem engineers in our meta-analysis with large negative impacts supports the view that ecosystem engineers are often high-impact exotic species (Vitousek 1990; Crooks 2002; Wallentinus & Nyberg 2007; Byers 2009). Indeed, the average magnitude of negative impacts (Fig. 3.3 and 3.4) by exotic ecosystem engineers in our meta-analysis exceeded that reported for the impacts of exotic predators on native prey species in another recent meta-analysis (lnR = -1.4; Salo et al. 2007). The frequency of exotic engineers causing substantial declines in some native species, together with their large negative effects compared to exotic predators, strengthens the argument for prioritizing management efforts to prevent invasions by habitat modifying species. Our results contrast those of two previous studies that have tested the ecological naiveté hypothesis using phylogenetic relatedness as a proxy for functional similarity. Strauss et al. (2006a) observed that introduced noxious grasses (e.g. those that spread into natural areas where they displace native species) in California were less closely phylogenetically related to native species than were introduced non-noxious grasses. Similarly, Ricciardi & Atkinson (2004) found that aquatic nonindigenous species were more likely to cause a large (> 80%) decline in a native population if they belonged to genera unshared by native species in the recipient waterbody. Their investigation incorporated the impacts of introduced fishes, invertebrates, algae and vascular plants in seven

68 freshwater and marine systems, wherein large negative impacts were mediated predominantly by predation and competition. Importantly, these aquatic systems were invaded by many of the ecosystem engineers included in our meta-analysis, and several of these species are known to have large negative impacts on native species via other types of interactions (e.g. interference competition between Dreissena polymorpha and native unionid mussels in North America; Ricciardi 2003). Taken together, these differences suggest that invasion impacts associated with habitat modification — specifically, the creation of physical structures in benthic habitats — do not conform to the same pattern as impacts mediated through other types of species interactions, with respect to the phylogenetic distinctiveness of exotic species.

3.5.2 Interaction strength and environmental stress models

Our results have some intriguing implications when considered in the context of environmental stress models, which make predictions about how the strength of facilitation and other biotic interactions vary across gradients of environmental stress or resource availability (Menge & Sutherland 1987; Bruno et al. 2003; Crain & Bertness 2006). Interspecific interactions involving habitat-modifying organisms are often thought to be comprised of both positive (facilitative) and negative (competitive) components (Bruno et al. 2003). For example, shrubs or ‗nurse plants‘ in deserts facilitate local germination of other plants by reducing soil surface temperatures and increasing the moisture content of adjacent soil, but inhibit their growth via shading (Holzapfel & Mahall 1999). Habitat modification by the exotic species in our meta-analysis is likely to impact native species through a variety of mechanisms including changes to habitat availability and complexity, flow velocity, light availability, abiotic (e.g. hydrodynamic, desiccation) and biotic stressors (e.g. refugia from predation), substrate stability, sedimentation rate, sediment properties (e.g. particle size, organic matter content), dissolved oxygen concentration and temperature (Crooks 2002; Reise 2002; Gutiérrez et al. 2003; Wallentinus & Nyberg 2007). Because habitat modification

69 affects multiple species, exotic engineers are also likely to indirectly impact natives through changes in the abundance of their predators, competitors, or prey. Finally, ecosystem engineers also have non-engineering (trophic) effects. Disentangling the relative importance of these mechanisms of impact and determining whether and how they vary with local environmental conditions is a challenging area of research (e.g. Bruno et al. 2003; Neira et al. 2007; Bouma et al. 2009). The complex nature of this type of species interaction complicates interpretation of our results, because the effect sizes calculated in our analysis represent net effects. Consequently, the more facilitative net effects of distinct engineers could be generated by smaller competitive components (Fig. 3.6a) or larger facilitative components (Fig. 3.6b) of their interactions with native macroinvertebrates. However, if prior evolutionary experience with functionally similar species reduces the magnitude of negative interaction components (Connell 1980; Callaway & Ridenour 2004; Cox & Lima 2006), then it must also and more dramatically reduce the magnitude of facilitative interaction components in order for distinct engineers to have more facilitative net effects (Fig. 3.6c). An important exception might be interactions between exotic species and closely related native species, which are expected to compete for shared resources (e.g. as predicted by Darwin‘s naturalization hypothesis; Darwin 1859; Daehler 2001). None of the impacts in our meta-analysis were measured on native congeners of the exotic ecosystem engineers, but our dataset did include two impacts on confamilial native species by the introduced mussel Mytilus galloprovincialis: the native mussel Aulacomya ater occurred in higher densities while Choromytilus meridionalis occurred in lower densities in the presence of this exotic ecosystem engineer (Robinson & Griffiths 2002). Whether the phylogenetic relatedness of an introduced ecosystem engineer to its recipient community promotes or constrains its invasion success and impact is likely to vary along environmental stress gradients. Phylogenetic relatedness with native species should confer an advantage to species invading communities that are structured by physical processes as opposed to biotic interactions, provided that physiological tolerances are phylogenetically conserved (Webb et

70 (a) + Facilitative component

0

(b) +

0

Competitive component

Direction and magnitude and ofmagnitude impact Direction (c) +

0

Low High Functional distinctiveness

Figure 3.6. Interactions between exotic ecosystem engineers and native species likely consist of both competitive and facilitative components. A positive relationship between net impact (on native species density) and the functional distinctiveness of exotic ecosystem engineers could be driven by differences in the magnitude of (a) facilitative components, (b) competitive components or (c) both facilitative and competitive components of the interaction. Figure modified after Bruno et al. (2003).

71 al. 2002; Cavender-Bares et al. 2004). Conversely, in physically benign environments where biotic interactions are the dominant drivers of community structure (Menge and Sutherland 1987), exotic species that are closely related to natives are more likely to face intense competition and may be especially vulnerable to the natural enemies of their native relatives (Strauss et al. 2006a and references therein). Here, native competitors or natural enemies could reduce the impacts of exotic engineers that resemble natives by suppressing their local density (Parker et al. 1999; Ricciardi 2003). While this scenario is consistent with the overall pattern we found (Figs. 3.1 and 3.2), it does not explain differences in the prevalence and magnitude of negative impacts on native species (Figs. 3.3 and 3.4, Table 3.1). Alternatively, exotic ecosystem engineers with native congeners may possess traits that predispose them to resist native generalist consumers (Ricciardi & Ward 2006; Appendix A). The relative importance of facilitation via different ecosystem engineering mechanisms is also expected to vary across environmental stress gradients (Bruno et al. 2003; Crain & Bertness 2006). Under harsh environmental conditions, the facilitative effects of ecosystem engineers are probably mediated through amelioration of physical stress, while provision of competitor- or predator-free space should be more important in physically benign environments (Bruno et al. 2003; Crain & Bertness 2006). Therefore, if phylogenetically distinct engineers are more likely than engineers with native relatives to successfully invade physically benign environments, where competition is relatively more important (consistent with Fig. 3.6c), their facilitative effects are likely to be mediated through the amelioration of competition or predation pressures.

3.5.3 Facilitation by distinct engineers

Both ecological and evolutionary processes might result in phylogenetically distinct ecosystem engineers generating more facilitative effects on native macroinvertebrates. Several authors have suggested that the strength of ecosystem-level impacts will increase with the disparity between the traits of the

72 exotic species and those of resident species (e.g. Vitousek 1990; Chapin et al. 1996; Dukes & Mooney 2004). In particular, exotic species that are dissimilar to natives in resource acquisition or use can have pronounced effects on resource availability for other species (e.g. nitrogen-fixing plants invading nitrogen-poor soils; Vitousek 1990). Moreover, the foundation species (Dayton 1972) and keystone modifier (Mills et al. 1993) concepts both describe a single habitat- modifying species with important consequences for community structure, placing an emphasis on the functional distinctiveness of a particular ecosystem engineering species. Phylogenetically distinct ecosystem engineers might have more facilitative effects on native species because they add novel habitats or resources, whereas introduced engineers that resemble native species are likely to provide habitats or resources that are already available and not necessarily limiting in the invaded system. These impacts need not depend on prior evolutionary experience. For example, aquatic macroinvertebrates generally take advantage of structurally complex habitats, be they native or exotic (Crooks 2002; Chapter 2). Even so, the magnitude of impacts of exotic ecosystem engineers on macroinvertebrates might still depend on the evolutionary experience of native species in the invaded community. Habitats created by distinct engineers could provide superior spatial refugia from predation if native predators are more efficient at foraging in engineered habitats that resemble those created by native species to which they are adapted; such habitats are more likely to be created by engineers that closely resemble natives. Dreissenid mussels introduced to North American inland waters, which lack functionally similar native species, have been associated with reduced foraging success in a number of native fish and invertebrate predators; some native predators have been observed to avoid Dreissena-covered substrates (Beekey et al. 2004; Dieterich et al. 2004; McCabe et al. 2006). If this explanation is true, higher macroinvertebrate densities in habitats modified by distinct engineers could reflect negative impacts on native predators. Large positive effects of distinct engineers on benthic macroinvertebrates may in this way represent a perturbation with adverse ramifications for aquatic ecosystems.

73 3.5.4 Time since invasion

Given that introduced species can exert strong selection pressures on native species (Strauss et al. 2006b), we might expect negative impacts (as measured here) to become less severe or frequent over time as native species adapt or vulnerable species are eliminated from the invaded community. However, we found a negative correlation between the impacts of exotic ecosystem engineers and the number of years since their establishment (Fig. 3.5). Several factors may have contributed to this result. First, the characteristic boom-and-bust cycle of exotic species abundance (Sakai et al. 2001) could drive this relationship, if earlier investigations were conducted during the boom phase and subsequent investigations during the bust phase of invasions. Second, this pattern could reflect the evolution of increased competitive ability in introduced engineer species (e.g. via higher resource allocation to competitive traits in the absence of specialist natural enemies; Strayer et al. 2006). Third, different mechanisms of engineering impacts are likely to occur over different timescales. While the effect of provision of structurally complex habitat is likely to be facilitative and immediate (Crooks 2002), other impacts resulting from cumulative changes to the environment – such as changes to the nutrient content or redox potential of sediments – develop over longer timescales (Strayer et al. 2006). For example, in San Francisco Bay, the diversity of macroinvertebrates associated with hybrid Spartina alterniflora has declined over time, apparently in response to increasing pore-water sulfide concentrations and hypoxia (Neira et al. 2007). Consistent with our hypothesis related to refugia from predation, native predators may eventually adapt to forage more efficiently in habitats created by exotic species. Finally, aquatic systems with a longer invasion history may presently be more anthropogenically altered or degraded, which could enhance the negative effects of exotic engineers on native species (Byers 2002).

3.5.5 Conclusions

Invasions by ecosystem engineering species can exert strong impacts on recipient

74 communities, altering not only species composition but also the physical structure of invaded ecosystems. Here, we have shown that the impacts of habitat modification by exotic species on native macroinvertebrates are correlated with their phylogenetic relatedness to native species in the invaded community. Exotic engineers that are dissimilar to natives appear particularly likely to have facilitative effects, perhaps by providing novel resources or refugia from predation. Still, the ubiquity and magnitude of negative effects on native macroinvertebrates underscores the need to prevent invasions by species that can substantially alter the physical structure of aquatic habitats. We also found a negative relationship between the impacts of exotic ecosystem engineers and time since invasion; an intriguing implication of this result is that introduced engineers that facilitate native species during the initial stages of an invasion may have reduced or even inhibitory effects at later stages. Experimental tests are needed to explore these findings, but they reveal a general pattern in the context-dependence of impact and highlight the potential application of phylogeny to the risk assessment of invasions.

3.6 Acknowledgements

The authors thank J. Kochmann for generously providing raw data from experiments, and K. O‘Sullivan and A. Hassan for help with data entry. Funding for this research was provided by a Canada Graduate Scholarship awarded to J.M. Ward by NSERC Canada.

3.7 References cited

Adams, D.C., Gurevitch, J. & Rosenberg, M.S. (1997). Resampling tests for meta-analysis of ecological data. Ecology, 78, 1277-1283. Atkinson, I.A.E. (1985). The spread of commensal species of Rattus to oceanic islands and their effects on island avifauna. In P.J. Moore (Ed.), Conservation

75 of Island Birds: Case Studies for the Management of Threatened Island Species (pp. 35-81). Cambridge, UK: International Council for Bird Preservation. Beekey, M.A., McCabe, D.J. & Marsden, J.E. (2004). Zebra mussels affect benthic predator foraging success and habitat choice on soft sediments. Oecologia, 141, 164-170. Bouma, T.J., Olenin, S., Reise, K. & Ysebaert, T. (2009). Ecosystem engineering and biodiversity in coastal sediments: posing hypotheses. Helgoland Marine Research, 63, 95-106. Bruno, J.F., Fridley, J.D., Bromberg, K. & Bertness, M.D. (2005). Insights into biotic interactions from studies of species invasions. In D.F. Sax, S.D. Gaines & J.J. Stachowicz (Eds.), Species Invasions: Insights into Ecology, Evolution and Biogeography (pp. 13-40). Sunderland, MA: Sinauer. Bruno, J.F., Stachowicz, J.J. & Bertness, M.D. (2003). Inclusion of facilitation into ecological theory. Trends in Ecology and Evolution, 18, 119-125. Byers, J.E. (2002). Impact of non-indigenous species on natives enhanced by anthropogenic alteration of selection regimes. Oikos, 97, 449-458. Byers, J.E. (2009). Invasive animals in marshes: biological agents of change. In B. Silliman, E. Grosholz & M. Bertness (Eds.), Human Impacts on Salt Marshes: A Global Perspective (pp. 41-56). Berkeley, CA: University of California Press. Callaway, R.M. & Ridenour, W.M. (2004). Novel weapons: invasive success and the evolution of increased competitive ability. Frontiers in Ecology and the Environment, 2, 436-443. Carlton, J.T. (1996). Marine bioinvasions: the alteration of marine ecosystems by nonindigenous species. Oceanography, 9, 36-43. Case, T.J. & Bolger, D.T. (1991). The role of introduced species in shaping the abundance and distribution of island reptiles. Evolutionary Ecology, 5, 272- 290. Cavender-Bares, J., Ackerly, D.D., Baum, D.A. & Bazzaz, F.A. (2004). Phylogenetic overdispersion in Floridian oak communities. The American Naturalist, 163, 823-843.

76 Chapin, F.S., III, Reynolds, H.L., D‘Antonio, C.M. & Eckhart, V.M. (1996). The functional role of species in terrestrial ecosystems. In B. Walker & W. Steffen (Eds.), Global Change and Terrestrial Ecosystems (pp. 403-428). Cambridge: Cambridge University Press. Clavero, M. & García-Berthou, E. (2005). Invasive species are a leading cause of animal extinctions. Trends in Ecology and Evolution, 20, 110. Connell, J.H. (1980). Diversity and the coevolution of competitors, or the ghost of competition past. Oikos, 35, 131-138. Cox, J.G. & Lima, S.L. (2006). Naiveté and an aquatic-terrestrial dichotomy in the effects of introduced predators. Trends in Ecology and Evolution, 21, 674- 680. Crain, C.M. & Bertness, M.D. (2006). Ecosystem engineering across environmental gradients: implications for conservation and management. BioScience, 56, 211-218. Crooks, J.A. (2002). Characterizing ecosystem-level consequences of biological invasions: the role of ecosystem engineers. Oikos, 97, 153-166. Daehler, C.C. (2001). Darwin‘s naturalization hypothesis revisited. The American Naturalist, 158, 324-330. Darwin, C. (1859). On the Origins of Species by Means of Natural Selection. London, UK: John Murray. Dayton, P.K. (1972). Toward an understanding of community reilience and the potential effects of enrichments to the benthos at McMurdo Sound, Antarctica. In B.C. Parker (Ed.), Proceedings of the Coloquium on Conservation Problems in Antarctica (pp. 81-96). Lawrence, KS: Allen Press. Dextrase, A.J. & Mandrak, N.E. (2006). Impacts of alien invasive species on freshwater fauna at risk in Canada. Biological Invasions, 8, 13-24. Diamond, J. & Case, T.J. (1986). Overview: introductions, extinctions, exterminations, and invasions. In J. Diamond & T.J. Case (Eds.), Community Ecology (pp. 65-79). New York, NY: Harper & Row. Dieterich, A., Mörtl, M. & Eckmann, R. (2004). The effects of zebra mussels (Dreissena polymorpha) on the foraging success of Eurasian perch (Perca

77 fluviatilis) and ruffe (Gymnocephalus cernuus). International Review of Hydrobiology, 89, 229-237. Dukes, J.S. & Mooney, H.A. (2004). Disruption of ecosystem processes in western North America by invasive species. Revista Chilena de Historia Natural, 77, 411-437. Englund, G., Sarnelle, O. & Cooper, S. (1999). The importance of data selection criteria: meta-analysis of stream predation experiments. Ecology, 80, 1132- 1141. García-Berthou E., Alcaraz C., Pou-Rovira Q., Zamora L., Coenders G. & Feo C. (2005). Introduction pathways and establishment rates of invasive aquatic species in Europe. Canadian Journal of Fisheries and Aquatic Sciences, 62, 453-463. Gurevitch, J. & Hedges, L.V. (2001). Meta-analysis: combining the results of independent experiments. In S.M. Scheiner & J. Gurevitch (Eds.), Design and Analysis of Ecological Experiments (pp. 347-369). New York, NY: Oxford University Press. Gutiérrez, J.L., Jones, C.G., Strayer, D.L., & Iribarne, O.O. (2003). Mollusks as ecosystem engineers: the role of shell production in aquatic habitats. Oikos, 101, 79-90. Harvey, P.H. & Pagel, M.D. (1991). The Comparative Method in Evolutionary Biology. Oxford: Oxford University Press. Hedges, L.V., Gurevitch, J. & Curtis, P.S. (1999). The meta-analysis of response ratios in experimental ecology. Ecology, 80, 1150-1156. Hedges, L.V. & Olkin, I. (1985). Statistical Methods for Meta-Analysis. Orlando, FL: Academic Press. Holzapfel, C. & Mahall, B.E. (1999). Bidirectional facilitation and interference between shrubs and annuals in the Mojave desert. Ecology, 80, 1747-1761. Huston M.A. (1994). Biological Diversity: The Coexistence of Species on Changing Landscapes. Cambridge, UK: Cambridge University Press. Jones, C.G., Lawton, J.H. & Shachak, M. (1994). Organisms as ecosystem engineers. Oikos, 69, 373-386.

78 Jones, C.G., Lawton, J.H. & Shachak, M. (1997). Positive and negative effects of organisms as physical ecosystem engineers. Ecology, 78, 1946-1957. Lipsey, M.W. & Wilson, D.B. (2001). Practical Meta-Analysis. Thousand Oaks, CA: Sage. McCabe, D.J., Beekey, M.A., Mazloff, A. & Marsden, J.E. (2006). Negative effect of zebra mussel on foraging and habitat use by lake sturgeon (Acipenser fulvescens). Aquatic Conservation: Marine and Freshwater Ecosystems, 16, 493-500. 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. Mills, L.S., Soulé, M.E. & Doak, D.F. (1993). The keystone-species concept in ecology and conservation. BioScience, 43, 219-224. Neira, C., Levin, L.A., Grosholz, E.D. & Mendoza, G. (2007). Influence of invasive Spartina growth stages on associated macrofaunal communities. Biological Invasions, 9, 975-993. Odling-Smee, F.J. (1988). Niche constructing phenotypes. In H.C. Plotkin (Ed.), The Role of Behavior in Evolution (pp. 73-132). Cambridge, MA: MIT Press. Odling-Smee, F.J., Laland, K.N. & Feldman, M.W. (2003). Niche Construction: The Neglected Process in Evolution. Monographs in Population Biology, 37. Princeton, NJ: Princeton University Press. Parker, I.M., Simberloff, D., Lonsdale, W.M., Goodell, K., Wonham, M., Kareiva, P.M., Williamson, M.H., Von Holle, B., Moyle, P.B., Byers, J.E. & Goldwasser, L. (1999). Impact: toward a framework for understanding the ecological effects of invaders. Biological Invasions, 1, 3-19. Reise, K. (2002). Sediment mediated species interactions in coastal waters. Journal of Sea Research, 48, 127-141. Ricciardi, A. (2003). Predicting the impacts of an introduced species from its invasion history: an empirical approach applied to zebra mussel invasions. Freshwater Biology, 48, 972-981.

79 Ricciardi, A. & Atkinson, S.K. (2004). Distinctiveness magnifies the impact of biological invaders in aquatic ecosystems. Ecology Letters, 7, 781-784. Ricciardi, A. & Ward, J.M. (2006). Comment on "Opposing effects of native and exotic herbivores on plant invasions". Science, 313, 298. Robinson, T.B. & Griffiths, C.L. (2002). Invasion of Langebaan Lagoon, South Africa, by Mytilus galloprovincialis – effects on natural communities. African Zoology, 37, 151-158. Rodriguez, L.F. (2006). Can invasive species facilitate native species? Evidence of how, when, and why these impacts occur. Biological Invasions, 8, 927-939. Rosenberg, M.S., Adams, D.C. & Gurevitch, J. (2000). MetaWin: Statistical Software for Meta-Analysis. Version 2. Sunderland, MA: Sinauer Associates. Ruiz, G.M., Fofonoff, P., Hines, A.H. & Grosholz, E.D. (1999). Non-indigenous species as stressors in estuarine and marine communities: assessing invasion impacts and interactions. Limnology and Oceanography, 44, 950-972. Sakai, A.K., Allendorf, F.W., Holt, J.S., Lodge, D.M., Molofsky, J., With, K.A., Baughman, S., Cabin, R.J., Cohen, J.E., Ellstrand, N.C., McCauley, D.E., O'Neil, P., Parker, I.M., Thompson, J.N. & Weller, S.G. (2001). The population biology of invasive species. Annual Review of Ecology and Systematics, 32, 305-332. Salo, P., Korpimäki, E., Banks, P.B., Nordström, M. & Dickman, C.R. (2007). Alien predators are more dangerous than native predators to prey populations. Proceedings of the Royal Society B: Biological Sciences, 274, 1237-1243. Sokal, R.R. & Rohlf, F.J. (1995). Biometry: The Principles and Practices of Statistics in Biological Research. San Francisco, CA: Freeman. Strauss, S.Y., Lau, J.A. & Carroll, S.P. (2006b). Evolutionary responses of natives to introduced species: what do introductions tell us about natural communities? Ecology Letters, 9, 357-374. Strauss, S.Y., Webb, C.O. & Salamin, N. (2006a). Exotic taxa less related to native species are more invasive. Proceedings of the National Academy of Sciences, USA, 103, 5841-5845.

80 Strayer, D.L., Eviner, V.T., Jeschke, J.M. & Pace, M.L. (2006). Understanding the long-term effects of species invasions. Trends in Ecology and Evolution, 21, 645-651. Talley, T.S., Crooks, J.A. & Levin, L.A. (2001). Habitat utilization and alteration by the invasive, burrowing isopod, Sphaeroma quoyanum, in California salt marshes. Marine Biology, 138, 561-573. Vitousek, P.M. (1990). Biological invasions and ecosystem processes: towards an integration of population biology and ecosystem studies. Oikos, 57, 7-13. Wallentinus, I. & Nyberg, C.D. (2007). Introduced marine organisms as habitat modifiers. Marine Pollution Bulletin, 55, 323-332. Webb, C.O., Ackerly, D.D., McPeek, M.A. & Donoghue, M.J. (2002). Phylogenies and community ecology. Annual Review of Ecology and Systematics, 33, 475–505. Williamson, M. (1996). Biological Invasions. London, UK: Chapman and Hall.

81 Connecting statement 3–4

In Chapter 3, I tested the ecological naiveté hypothesis with respect to the impacts of exotic ecosystem engineers on native macroinvertebrates in aquatic systems. A meta-analysis showed that introduced bioconstructors are more likely to facilitate native species, and their negative effects are less severe, when they belong to a taxon (genus or family) not represented by native species in the recipient system. These results support the view that the impact of an introduced species is correlated with its functional distinctiveness relative to native species in the recipient community. In Chapters 4 and 5, I focus on two exotic ecosystem engineers that have no native ecological analog (byssally-attached epibenthic mussels) throughout much of their invaded range, the zebra mussel (Dreissena polymorpha) and quagga mussel (D. bugensis). These species have invaded a variety of freshwater locations where they have well documented impacts on benthic macroinvertebrate communities, permitting a more detailed analysis of variation in their ecological effects across invaded sites.

82

Chapter 4 – Impacts of Dreissena invasions on benthic macroinvertebrate

communities: a meta-analysis

A version of this chapter appears as: Ward, J.M. & Ricciardi, A. (2007). Impacts of Dreissena invasions on benthic macroinvertebrate communities: a meta-analysis. Diversity and Distributions, 13, 155-165.

Reprinted with permission from John Wiley and Sons. http://www3.interscience.wiley.com/journal/118507128/home

83 4.1 Abstract

Dreissenid mussels (the zebra mussel Dreissena polymorpha and the quagga mussel Dreissena bugensis) have invaded lakes and rivers throughout North America and Europe, where they have been linked to dramatic changes in benthic invertebrate community diversity and abundance. Through a meta-analysis of published data from 47 sites, we developed statistical models of Dreissena impact on benthic macroinvertebrates across a broad range of habitats and environmental conditions. The addition of Dreissena was generally associated with increased benthic macroinvertebrate density and taxonomic richness, and with decreased community evenness (excluding Dreissena). However, the strength of these effects was context-dependent, as they varied systematically with sediment particle size across sites. The effects of Dreissena differed among taxonomic and functional groups of macroinvertebrates, with positive effects on the densities of scrapers and predators, particularly leeches (Hirudinea), flatworms (Turbellaria) and mayflies (Ephemeroptera). Gastropod densities increased in the presence of Dreissena, but large-bodied snail taxa tended to decline. Dreissena was associated with declines in the densities of sphaeriid clams and other large filter- feeding taxa, as well as burrowing amphipods (Diporeia spp.), but had strong positive effects on gammarid amphipods. These patterns are robust to variation in the methodology of primary studies. The effects of Dreissena are remarkably concordant with those of ecologically-similar taxa, suggesting universality in the interactions between introduced byssally-attached mussels and other macroinvertebrates.

84 4.2 Introduction

A central goal of invasion ecology is the development of predictive models for the ecological impacts of introduced species (Parker et al. 1999; Ruiz et al. 1999). A major challenge to prediction is context-dependence, because impacts of the same species may vary across different communities (Ricciardi 2003; Ross et al. 2003; Ricciardi & Atkinson 2004). By integrating results from multiple studies, synthetic statistical approaches such as meta-analysis offer powerful tools for detecting predictable patterns; indeed, recent syntheses have discovered new patterns explaining variation in invasion success and impact (e.g. Levine et al. 2004; Ricciardi & Atkinson 2004; Parker et al. 2006; Ricciardi & Ward 2006 – Appendix A). Despite its promising potential, meta-analysis has only just begun to be applied to study invasions (Levine et al. 2004; Agrawal et al. 2005; Parker et al. 2006), and has thus far been used primarily to address questions about species establishment rather than the ecological consequences of invasion (but see McCarthy et al. 2006). Among the introduced species that cause major ecological impacts are bivalves, whose activities can alter resource availability and ecosystem functioning (Crooks & Khim 1999; Gutiérrez et al. 2003). Dreissenid mussels (the zebra mussel Dreissena polymorpha Pallas and the quagga mussel Dreissena bugensis (Andrusov)) have invaded lakes and rivers throughout North America and Europe, where they are the only bivalves that attach to hard substrates. These invasions have been linked to dramatic changes in benthic invertebrate diversity and abundance (Karatayev et al. 1997; Ricciardi 2003). Their effects are largely positive, but vary in magnitude across studies (Ricciardi 2003), and may vary spatially even within a single lake (Mayer et al. 2002) or river (Ricciardi et al. 1997). Negative effects on individual taxa are relatively rare, but usually involve gastropods and filter-feeding taxa (Wisenden & Bailey 1995; Ricciardi et al. 1997; Haynes et al. 2005). Although increases in species richness have been observed at multiple sites (Ricciardi 2003), changes in diversity indices vary in significance across studies (e.g. Horvath et al. 1999; Nalepa et al. 2003). A

85 burgeoning literature on the community-level effects of Dreissena provides an opportunity for quantitative synthesis, allowing us to broadly test hypotheses and explore the generality of community responses. The objectives of this study were to develop statistical models of the impacts of Dreissena invasions on benthic macroinvertebrate communities, and to identify predictable patterns across a broad range of habitats and environmental conditions. We treat both species as a single taxon, because both mussels have similar attributes and commonly occur together in mixed colonies (Jones & Ricciardi 2005). Given that benthic macroinvertebrates are valuable indicators of water quality (Washington 1984; Gabriels et al. 2005), a predictive understanding of the impact of Dreissena is necessary so that observed changes to invaded communities are not erroneously attributed to watershed management (e.g. nutrient abatement) or other environmental stressors. Mussels, like other ecosystem engineers (Jones et al. 1994, 1997), often have large community-level effects resulting from their ability to modulate the availability of resources to other species by building and modifying habitats. These impacts may be predictable; engineer species that increase the structural complexity of benthic habitats tend to cause increases in the abundance and species richness of benthos (Crooks 2002; Chapters 2 and 3). Byssally-attached mussels create spatially complex patches that can support diverse assemblages of associated organisms (Suchanek 1986; Seed 1996). These patches provide gastropods and other benthic herbivores with additional grazing area and spatial refugia from predators (Griffiths et al. 1992; Ricciardi et al. 1997). However, in marine systems, mussel shells limit the size of interstitial spaces – and thus the size of colonizing organisms (Jacobi 1987; Griffiths et al. 1992). Mussels also affect nutrient availability via biodeposition of feces and pseudofeces (Izvekova & Lvova-Katchanova 1972; Sephton et al. 1980), increasing microhabitat complexity on rocky substrata through biodeposition and sediment focusing in pits (Yager et al. 1993). Their effects on resident species should vary depending on whether the resources they provide are limiting in the local environment. Hence, we derived the following predictions:

86 1) The effects of Dreissena on macroinvertebrate abundance and taxonomic richness are positively correlated with Dreissena density.

2) Dreissena colonization increases densities of gastropods, but limits their mean body size and displaces large-bodied taxa.

3) The effects of Dreissena vary across functional feeding groups, with strong positive effects on scrapers and deposit-feeding taxa and strong negative effects on filter-feeding taxa.

4) The effects of Dreissena vary with the characteristics of the underlying sediments. Effects on burrowing and epilithic taxa are positively and negatively correlated, respectively, with sediment particle size.

Because some studies have already summarized patterns of impact on native mussels (Ricciardi et al. 1995; Schloesser et al. 1996), we have restricted our study to Dreissena‘s effects on non-mussel taxa.

4.3 Methods

We searched the literature for studies that compared benthic macroinvertebrate communities in the presence and absence of Dreissena and reported effects on total non-dreissenid macroinvertebrate abundance or diversity. Relevant studies were located by searching Aquatic Sciences and Fisheries Abstracts (ASFA, 1971-June 2006), Ecology Abstracts (1982-June 2006) and Science Citation Index Expanded (SCI-Expanded, 1900-June 2006). Search terms included all possible combinations of (1) zebra mussel*, Dreissena, dreissenid, (2) communit*, benth*, macroinvertebrate*, and (3) abundance, diversity, richness, impact*. Additional studies were located by examining the reference lists of primary studies and narrative reviews, and by forward reference tracking in SCI-Expanded. In order to limit problems of non-independence (Gurevitch & Hedges 1999), we used only one effect size estimate per study site in each analysis. We considered multiple study sites from a single publication to be independent if the authors reported results separately for each site or group of sampling stations (i.e.

87 we used the authors‘ definition of independence). When multiple publications reported Dreissena effects for the same site, we included a single effect size estimate for that site, according to the following protocol. When authors compared macroinvertebrate communities both experimentally and observationally (inside and outside mussel beds, or before and after invasion) at the same site, we included the experimental data. When multiple Dreissena density treatments were reported in an experiment, we used the density treatment that most closely resembled ambient Dreissena density at the site. For studies where Dreissena treatments were crossed with other treatments (e.g. predators), we used the Dreissena and control treatments that most resembled conditions in the single factor experiments. When data from multiple sampling dates were available for a given site, we used the final sampling date. Because Dreissena effects on macroinvertebrates are driven largely by the characteristics of clustered mussels (Botts et al. 1996; Ricciardi et al. 1997), we excluded sites where the mean Dreissena density was less than 100 m-2, as clusters are rare at such densities (personal observation). We also excluded pre- and post-invasion comparisons that were separated by more than 20 years. Our search identified data from a total of 47 study sites that met all our predefined inclusion criteria. To investigate patterns of context-dependence, we also recorded information about several study site characteristics chosen a priori (Table 4.1). These were variables hypothesized to explain variation in Dreissena impacts across sites, as well as methodological variables that could potentially bias or confound our results. Sediment particle size was calculated by taking the weighted average phi

(-log2 particle diameter, in mm) of the sediments sampled, or by using the diameter of a sphere with surface area equal to that of experimental substrates. Negative phi values denote coarse (hard) substrates, while positive phi values denote increasingly fine sediments. We examined the responses of benthic macroinvertebrate communities to Dreissena using standard meta-analytical techniques (Cooper & Hedges 1994; Osenberg et al. 1999; Gurevitch et al. 2001). The community response variables used in these analyses were measures of non-Dreissena macroinvertebrate

88 Table 4.1. Univariate regression models of Dreissena effect on macroinvertebrate numerical density vs. study site characteristics and methodological variables recorded for each site. Factor n Slope (β) P r Dreissena density (No. m-2)* 31 0.524 0.106 0.30 Depth (m)* 34 -1.196 <0.001 -0.66 Phi 28 -0.083 0.012 -0.47 Flow environment† 36 0.259 0.649 0.08 Time since invasion (years)* 26 -0.831 0.151 -0.29 Continent‡ 36 0.755 0.214 0.21 Type of comparison 1§ 36 0.584 0.339 0.16 Type of comparison 2¶ 36 -0.812 0.056 -0.32 Sampling date (days) 30 -0.014 0.019 -0.43 Sieve aperture size (mm) 34 0.100 0.605 0.09 Multisite data** 34 -0.780 0.236 -0.21

* Factor was log10-transformed. † Coded as lentic (0) or lotic (1). ‡ Site was located in North America (0) or Europe (1). § Comparison between communities was experimental (0), spatial (1), or temporal (0). ¶ Comparison between communities was experimental (0), spatial (0), or temporal (1). ** Effect size calculated with single site (0) or multi-site (1) data.

89 abundance and diversity. We also explored patterns of response among different macroinvertebrate groups, measured as changes in the numerical density of individual taxa. For each response variable, the effect size in our meta-analysis was the loge response ratio, ln(X+D/X–D), calculated using measures of that response variable in the presence (X+D) and absence (X–D) of a dense Dreissena colony. Thus, a positive ratio indicates that macroinvertebrate density or diversity increased in the presence of Dreissena. This effect size metric was chosen for several reasons. Unlike other commonly used metrics, it does not confound differences in variance between studies with differences in the size of effect (Osenberg et al. 1997), nor does it depend on the magnitude of the average value of the response variable in a study (Englund et al. 1999). It also meets the assumptions of parametric analyses and has a clear biological interpretation as the

(loge-transformed) proportionate change in a response variable associated with

Dreissena invasion (Hedges et al. 1999). The use of the loge response ratio assumes that experimental durations are sufficiently long for equilibrium conditions to be reached (Hedges et al. 1999). This assumption was supported by available experimental time series data (Slepnev et al. 1994) and by some authors‘ use of pilot studies to determine the appropriate experimental duration (Stewart et al. 1998). We did not weight the effect size values by the inverse of their sampling variance, because of incomplete reporting in primary studies. Many studies did not report sampling variances and sample sizes for the response variables measured, rendering the corresponding weights incalculable. Exclusion of these studies would greatly reduce the sample size of our data set and could possibly introduce biases into our analyses (Englund et al. 1999). Instead, following the advice of Gurevitch and Hedges (1999) for synthesis of poorly reported data, we used unweighted standard parametric statistical tests. We caution that our estimated P values may be less precise and our tests less powerful as compared to weighted analyses (Gurevitch & Hedges 1999). Nonetheless, by including a greater number of sites we have reduced probabilities of both Type I and Type II errors in our meta-analysis (Lajeunesse & Forbes 2003).

90 Community-level effects of Dreissena were measured as changes in non- Dreissena macroinvertebrate numerical density, biomass density, taxonomic richness, taxonomic density, Simpson‘s diversity index and Simpson‘s evenness index. We attempted to avoid the ‗apples and oranges‘ problem (Gurevitch et al. 2001) by only combining effects measured in the same units in any given analysis (e.g. we did not combine numerical and biomass densities). We compiled diversity and evenness indices for studies that either reported them directly or contained sufficient information to allow their calculation. Simpson‘s diversity 2 was calculated as D = 1/∑( pi) where pi was the proportion of individuals th accounted for by the i taxon, and evenness was calculated as ED = D/Dmax, where

Dmax was the number of taxa (excluding Dreissena). For each community response variable, we calculated the mean effect size across all study sites and its 95% confidence interval. To assess the most important factors explaining variation in the response of benthic macroinvertebrate communities to Dreissena, we created a backward stepwise multiple regression model for the most widely reported community response variable: numerical density (Lipsey & Wilson 2001). Because incomplete reporting of study site characteristics resulted in missing data in our data matrix, we used univariate regression to pre-select variables for inclusion in the multiple regression, including only those variables significant in univariate tests (P < 0.05). Prior to regression analyses, Dreissena density, depth and time since invasion were log10-transformed, and categorical variables were dummy coded as follows: flow environment was lentic (0) or lotic (1), the site was located in North America (0) or Europe (1), and the effect size was calculated using single site (0) or multi-site (1) data. We created two dummy variables in order to tests whether effects varied among experimental (0,0), spatial (1,0) and temporal (0,1) comparisons of macroinvertebrate communities. Here we infer significant relationships only for those variables retained by the backward stepwise multiple regression using a significance level of 0.05 for variable entry and retention. Furthermore, we determined whether the factors retained by the stepwise regression explained variation in the response of macroinvertebrate taxonomic

91 richness (the second most commonly reported community response variable), using least squares regression analyses. We tested whether the magnitude of Dreissena effects on macroinvertebrates varied across taxonomic and functional groups, using the subset of studies reporting numerical density data for individual taxa in the presence and absence of Dreissena. For each major taxonomic group, we calculated within-site effects by summing across taxa before calculating the mean effect across sites (n ≥ 5). When the taxonomic resolution of primary studies permitted, we assigned each taxon to a functional feeding group, and identified infauna and epifauna, according to Merritt and Cummins (1984), Klemm (1985), Pennak (1989), Peckarsky et al. (1990), and Thorp and Covich (1991), and using the following definitions. For the purpose of our study, ―filter-feeder‖ refers to organisms carrying out all forms of suspension feeding. Infauna were organisms that spend the majority of their aquatic life cycle within soft sediments, while epifauna were organisms whose normal habit is occupying stable, solid surfaces including rocky substrates and submerged vegetation (e.g. epilithic and epiphytic taxa, respectively). For each of these functional groups, we conducted least squares regressions on the factors previously identified as most important in determining the response of total macroinvertebrate numerical density to Dreissena. Finally, we tested the effect of Dreissena colonization on gastropod density and body size using the subset of studies that reported data for gastropod numerical densities (No. m-2) to at least the family level. We calculated Dreissena effects on each gastropod family for a given site and determined the shell size for each family by averaging the maximum shell dimensions from all species having a distribution that includes the Great Lakes – St. Lawrence region, using data from Clarke (1981). We tested the relationship with gastropod shell size using least squares regression. All analyses were carried out using SAS release 8.02 (SAS institute, Cary, North Carolina, USA). See Table D.1 in Appendix D for data sources used in this meta-analysis.

92 4.4 Results

Macroinvertebrate communities responded positively to the presence of Dreissena in terms of their total non-dreissenid numerical density (No. m-2), biomass density (g m-2), taxonomic richness (No. taxa) and taxa density (No. taxa/sample), when data from all study sites were considered together (Fig. 4.1). In each of these variables, the mean response increased by a factors of 1.5 to 2.7. By contrast, Simpson‘s diversity was 1.2-fold lower in the presence of Dreissena (although this change was not significant) and Simpson‘s evenness declined by a mean factor of 1.3 in the presence of Dreissena.

4.4.1 Variation in effects on the numerical density and taxonomic richness of macroinvertebrate communities

Impacts on the numerical density of macroinvertebrates varied significantly with substrate particle size, but not with Dreissena density (Fig. 4.2). Macroinvertebrate numerical density tended to increase with increasing Dreissena density, but this trend was not significant (Fig. 4.2a). Depth, sediment particle size, and sampling date were significantly related to effects of Dreissena on macroinvertebrate numerical density in univariate analyses (Table 4.2). However, substrate particle size (phi) was the only variable retained by the backwards stepwise multiple regression. The magnitude of positive effects of Dreissena on the numerical density of macroinvertebrates declined with decreasing substrate particle size (i.e. increasing phi) (Fig. 4.2b). The magnitude of taxonomic richness response varied significantly with both substrate particle size and Dreissena density. Dreissena had larger positive effects on taxonomic richness with increasing Dreissena density (Fig. 4.2c) and in areas of fine sediment (Fig. 4.2d).

93 36 Numerical Density

13 Biomass Density

24 Total Richness

16 Taxa Density

14 Simpson's Diversity

14 Simpson's Evenness

-1.0 -0.5 0.0 0.5 1.0 1.5 2.0 Mean Dreissena effect (ln X /X ) +D -D

Figure 4.1. Mean response of benthic macroinvertebrate communities to Dreissena. All abundance and diversity calculations exclude Dreissena. Error bars are 95% confidence intervals. Numbers are sample sizes (number of study sites).

94 3 3 (a) (b) ) 2 -D 2

X

/

+D 1

ln X ln 1 0

effect ( effect 0 -1

-1 -2

Dreissena P = 0.1 P = 0.01 -3 -2

on macroinvertebrate numerical density numerical macroinvertebrate on 102 103 104 105 -10 -8 -6 -4 -2 0 2 4 6 8 0.6 2.0 (c) (d)

)

-D 0.4 1.5

X

/

+D

ln X ln 0.2 1.0

effect ( effect 0.0 0.5

-0.2 0.0

Dreissena P = 0.02 P = 0.0006 -0.4 -0.5 2 3 4 5 on macroinvertebrate taxonomic richness taxonomic macroinvertebrate on 10 10 10 10 -10 -8 -6 -4 -2 0 2 4 6 8 Dreissena density (No./m2) Phi (-log particle diameter, in mm) 2

Figure 4.2. Top panels: Relationship between community-level effects of Dreissena on the numerical density of benthic macroinvertebrates (excluding Dreissena) and (a) Dreissena density (r = 0.30, n = 31, P = 0.1), or (b) decreasing substrate particle size (increasing phi values) of the site where macroinvertebrates were sampled (r = -0.47, n = 28, P = 0.01). Bottom panels: Relationship between community-level effects of Dreissena on the taxonomic richness of benthic macroinvertebrates (excluding Dreissena) and (c) Dreissena density (r = 0.63, n = 14, P = 0.02), or (d) decreasing substrate particle size (r = 0.66, n = 23, P = 0.0006). Lines were generated by least-squares regression.

95 4.4.2 Variation in effects on taxonomic and functional groups

The average magnitude of Dreissena effects on the numerical density of macroinvertebrates varied across major taxonomic groups (Fig. 4.3). Overall, the presence of Dreissena was associated with significant increases in the density of leeches (Hirudinea), flatworms (Turbellaria), gastropods and mayflies (Ephemeroptera). Isopod crustaceans also tended to be more abundant in the presence of Dreissena, although this effect was not significant. In general, Dreissena effects on the density of amphipods, ostracods, dipteran flies, oligochaetes, caddisflies (Trichoptera) and nematodes were not significantly different from zero. The density of sphaeriid clams was generally lower in the presence of Dreissena. Epifaunal taxa were more abundant in the presence of

Dreissena, increasing by a mean factor of 2.5 (ln X+D/X-D = 0.92, n = 154), while the change in density of infaunal taxa was negative, decreasing by a mean factor of 1.4 (ln X+D/X-D = -0.33, n = 140). The overall response to Dreissena also differed between functional feeding guilds, with scrapers being the most positively affected in terms of numerical density (ln X+D/X-D = 0.98, n= 57), followed by predators (ln X+D/X-D = 0.63, n = 53), increasing by mean factors of

2.7 and 1.9, respectively. Deposit-feeding (ln X+D/X-D = 0.12, n = 161) and filter- feeding taxa were, on average, not significantly affected by Dreissena (ln X+D/X-D = -0.52, n = 36). Variation in the magnitude of Dreissena effects on macroinvertebrate numerical density across a substrate particle size gradient differed among functional groups of macroinvertebrate taxa. The magnitude of Dreissena effects on epifaunal and infaunal macroinvertebrates was related to substrate particle size (Fig. 4.4). Positive effects on epifauna declined with decreasing substrate particle size (Fig. 4.4a), while effects on the numerical density of infaunal (burrowing) taxa tended to be positive on hard substrates and negative on fine sediments (Fig. 4.4b). Dreissena effects on the numerical density of deposit-feeders (Fig. 4.5a) and predators (Fig. 4.5c) were also negatively correlated with decreasing substrate particle size. Effects on filter-feeders tended to decrease with decreasing

96 Hirudinea 8 Turbellaria 10 Gastropoda 16 Isopoda 8 Ephemeroptera 11 Amphipoda 23 Ostracoda 7 Diptera 23 Oligochaeta 23 Trichoptera 11 Nematoda 10 Sphaeriidae 17

-3 -2 -1 0 1 2 3 Mean Dreissena effect (ln X /X ) +D -D

Figure 4.3. Mean effects of Dreissena on the numerical density of major groups of benthic macroinvertebrates (excluding Dreissena). Error bars are 95% confidence intervals. Numbers are sample sizes (study sites).

97

6 (a) Epifauna

4

2

0

-2

) on the the ) on -4

-D

/X P = 0.03 +D -6 -10 -8 -6 -4 -2 0 2 4 6 8

ln X ln

6 (b) Infauna

effect ( effect 4

2

Dreissena

0

numerical density of benthic macroinvertebrate taxa macroinvertebrate benthic of density numerical -2

-4

P = 0.003 -6 -10 -8 -6 -4 -2 0 2 4 6 8 Phi (-log particle diameter, in mm) 2

Figure 4.4. Relationship between taxa-level effects of Dreissena on the numerical density of (a) epifaunal macroinvertebrates (r = -0.18, P = 0.03, n = 145) or (b) infaunal macroinvertebrates (r = -0.29, P = 0.003, n = 110) and decreasing substrate particle size (increasing phi values) of the site where macroinvertebrates were sampled. Lines were generated by least-squares regression.

98 6 6 (a) Deposit-feeders (b) Filter-feeders 4 4

2 2

0 0

-2 -2

) on the the ) on

-D -4 -4

/X

+D P < 0.0001 P = 0.10 -6 -6

ln X ln -10 -8 -6 -4 -2 0 2 4 6 8 -10 -8 -6 -4 -2 0 2 4 6 8 6 6 (c) Predators (d) Scrapers

effect ( effect 4 4

2 2

0 0

Dreissena

-2 -2

numerical density of macroinvertebrate taxa macroinvertebrate of density numerical

-4 -4 P = 0.003 P = 0.90 -6 -6 -10 -8 -6 -4 -2 0 2 4 6 8 -10 -8 -6 -4 -2 0 2 4 6 8

Phi (-log2 particle diameter, in mm)

Figure 4.5. Relationship between taxa-level effects of Dreissena on the numerical density of (a) deposit-feeding (r = -0.40, P < 0.0001, n = 129), (b) filter-feeding (r = -0.32, P = 0.10, n = 28), (c) predatory (r = -0.41, P = 0.003, n = 51) or (d) scraping macroinvertebrates (P = 0.90, n = 56) and decreasing substrate particle size (increasing phi values) of the site where macroinvertebrates were sampled. Lines were generated by least-squares regression.

99 substrate particle size, although this trend was not significant (Fig. 4.5b). The change in the numerical density of scrapers (primarily gastropods) was not related to substrate particle size, although there were few effect estimates for this group on soft substrates (Fig. 4.5d). As predicted, the positive effects of Dreissena on the numerical density of gastropods decreased with increasing gastropod shell size (Fig. 4.6). However, this result was due to strong negative effects on pleurocerid snails and the relationship was no longer significant when the effects on this family were removed (n = 36, P = 0.19).

4.5 Discussion

4.5.1 Predictable patterns of Dreissena impact

The magnitude of Dreissena impacts on the density and richness of benthic macroinvertebrate communities were positively and negatively correlated, respectively, with sediment particle size. The relationship between sediment particle size and the effect of Dreissena on total numerical density reflects changes in the abundances of multiple functional groups, including deposit- feeding, predatory and filter-feeding taxa. These effects were not significantly related to Dreissena density, even though the latter varies negatively with sediment particle size (Jones & Ricciardi 2005). Dreissena effects on the taxonomic richness of macroinvertebrates showed the opposite pattern across a sediment particle size gradient, with effects being greatest on fine sediments. This result is driven by differences in the occurrence of epifaunal and infaunal taxa among sediment environments. Dreissena-associated communities on soft sediments frequently include organisms more typical of rocky substrata, such as gammarid amphipod and isopod crustaceans, hydroids, flatworms, leeches and snails (e.g. Karatayev et al. 1997; Bially & MacIsaac 2000; Beekey et al. 2004). Similarly, in marine sedimentary environments, epifauna normally unable to occupy soft sediments are restricted in their occurrence to the mussel bed (Dittmann 1990; Robinson & Griffiths 2002). But this is not true for infauna taxa

100 6

Hy Ph

)

-D 4

/X An

+D Ly Bi Pla

ln X ln 2 Ple

effect ( effect 0

-2

Dreissena

on gastropod numerical density numerical gastropod on

-4 0 5 10 15 20 25 30 35

Gastropod shell size (mm)

Figure 4.6. Relationship between Dreissena effect on the numerical density of gastropod taxa and the average shell size across gastropod species within a family. The line was generated by least-squares regression (r = -0.45, P = 0.003, n = 43). Hy: Hydrobiidae, An: Ancylidae, Bi: Bithyniidae, Pla: Planorbidae, Ph: Physidae, Ly: Lymnaeidae, Ple: Pleuroceridae.

101 richness in Dreissena colonies on rocky substrates. Sedimentary microhabitats within mussel beds on hard substrates develop via passive and active deposition (Jacobi 1987; Yager et al. 1993), which will vary spatially and temporally among mussel colonies and sites. Further, while highly mobile epifauna may quickly colonize mussel patches on soft sediments (Mörtl & Rothhaupt 2003), many infaunal species are less mobile (Merritt & Cummins 1984). These factors may limit the ability of infauna to colonize and persist in Dreissena colonies on rocky substrates. We tested whether Dreissena effects on macroinvertebrate numerical density were correlated with study site characteristics that were either not significant in univariate analyses or not retained by the multiple regression model. Primary studies have contrasted the effects of Dreissena among shallow and deep sites (Strayer et al. 1998; Nalepa et al. 2003). While sampling depth was highly significant in a univariate regression, it was not retained by the multiple regression. Depth is correlated with sediment particle size and, while depth by itself explained a greater amount of variation, particle size explained more variation independent of this correlation. This pattern describes variation in the local-scale effects of Dreissena in dense mussel beds, while negative effects at deepwater sites have generally been measured outside of dense Dreissena beds and are attributable to the distal effects of Dreissena filtration via the transfer of energy from pelagic to benthic food webs (Strayer et al. 1998; Ricciardi 2003). Time since invasion may be an important factor moderating the impacts of an exotic species within sites (Strayer et al. 2006), but this variable did not explain variation in Dreissena effects on macroinvertebrate density. We also found no difference in community response between study sites located in North America and in Europe, although the latter group was limited in size (n = 6). Dreissenid mussels are alien to both regions (apart from southeastern Europe within the Black Sea basin) and so may similarly affect communities with which they do not share an evolutionary history. Finally, impact magnitude did not differ between lentic and lotic sites, perhaps because of spatial heterogeneity within large systems (e.g. fluvial lakes in the St. Lawrence River).

102 Macroinvertebrate abundance has been correlated with Dreissena density within sites (e.g. Horvath et al. 1999; Kuhns & Berg 1999; Strayer & Smith 2000; Mayer et al. 2002), but this relationship did not emerge across sites in our meta- analysis. Variation in site-specific environmental conditions (e.g. species richness, productivity, resource limitation) may modify the effect of Dreissena density. Alternatively, biomass or size frequency distribution may be more important in determining impact magnitude, the latter being a more accurate predictor for a mussel population‘s filtering capacity and pseudofeces production rate (Young et al. 1996). Thus, differences in the population structure of Dreissena across sites might render mussel population densities incomparable. Limited reporting of Dreissena biomass and population size structure prevented us from investigating this further. It is also difficult to compare correlations involving numerical density and taxonomic richness (Fig. 4.2a and 4.2c), because of differences in sample size and in the range of effect sizes. However, our results suggest that increases in taxonomic richness for a given Dreissena density may be more predictable across sites than increases in the numerical density of macroinvertebrates. While Dreissena colonization generally increases the taxonomic richness of benthic macroinvertebrates, the dominance structure of the community changes even when the most obvious dominant, Dreissena, is excluded from evenness calculations. A tendency towards declining Simpson‘s diversity even with increased richness, and a significant overall decline in Simpson‘s evenness index in the presence of Dreissena, indicate that some taxa are much more positively affected by Dreissena than others. Dreissena colonization has been associated with dramatic increases in the densities of gammarid amphipods and predatory macroinvertebrates (Ricciardi 2003). In particular, leeches, flatworms and snails were more abundant in the presence of Dreissena, while the abundance of sphaeriid clams was typically reduced. The response of scrapers and predatory taxa was also typically positive, the latter likely reflecting increased prey availability in the form of Dreissena and associated invertebrates. Some taxa, such as caddisfly larvae (Trichoptera) had a mixture of positive and negative

103 responses to Dreissena. This may, in part, reflect the responses of small filter- feeding caddisflies (e.g. Brachycentrus) that can exploit filtration currents generated by the mussels, in contrast with the competitive exclusion of large net- spinning caddisflies from Dreissena-colonized substrates (Ricciardi et al. 1997). Similarly, the overall non-significant effect of Dreissena on amphipods was driven by a combination of large positive effects on epibenthic gammarid amphipods, in particular Gammarus fasciatus, and large negative effects on burrowing amphipods (Diporeia spp.). While both are deposit-feeders, these amphipods differ in their ability to feed on Dreissena fecal deposits, and the latter may instead be out-competed by Dreissena for particulate food (Lozano et al. 2001). The decline of large snails on substrates colonized by Dreissena has been reported by several authors (e.g. Dusoge 1966; Wisenden & Bailey 1995; Ricciardi et al. 1997) and is concordant with observations of the competitive exclusion of large grazers from marine mussel beds (e.g. Griffiths et al. 1992; Lohse 1993; Tokeshi & Romero 1995). Our study found limited support for the hypothesized exclusion of families of large-bodied gastropods. We classified gastropod families on the basis of the maximum shell size obtained but were unable to account for variation in size among individuals. With the exception of pleurocerid snails, these families may be too small relative to the size of Dreissena to be competitively excluded and may persist on secondary substrata provided by mussel shells. In fact, some evidence suggests that gastropods grow more quickly on Dreissena colonized substrates (Greenwood et al. 2001). Furthermore, mussel beds may serve as gastropod nursery grounds, harboring juveniles of taxa in which the large adults are found only on bare rock (c.f. Tokeshi & Romero 1995); marine mussels may thereby competitively exclude large limpets but enhance recruitment of juveniles (Griffiths et al. 1992). Differences in the size-at-maturity of gastropods may also modify the outcome of their interactions with Dreissena, as observed for marine mussels (Branch & Steffani 2004).

104 4.5.2 Caveats

Several notes of caution are required due to the nature of our analysis and limitations of the available data. First, meta-analyses of variation in effect sizes are inherently correlational; explanatory variables were generally not controlled or manipulated across studies. Nonetheless, we take a first step in explaining large-scale variation in Dreissena impacts and provide a means for making quantitative predictions. Second, the available literature contained a limited number of experimental manipulations so we included observational data that likely have additional sources of variation; increases or declines in macroinvertebrates may have been caused by concomitant factors, such as changes in water quality. In addition, responses measured over different timescales may be differentially influenced by indirect or feedback effects of Dreissena (e.g. through changes in benthophagous fish populations or algal species composition). However, some strong patterns emerge despite this heterogeneity. Ideally, we would have divided the available data and used a subset to test our models, but limited availability of data made this unfeasible. Third, relatively few of the sites were located in Europe so our results may be biased towards describing North American macroinvertebrate communities. Fourth, while we made efforts to reduce problems associated with non- independence, our dataset was too limited to allow a rigorous assessment of such effects. Finally, our meta-analysis could not calculate effect sizes with zero values, and we are aware of no satisfactory method to deal with this problem, therefore we omitted these values.

4.5.3 Comparisons with other introduced mussel species

Facilitation of benthic organisms may be a general effect of introduced habitat engineers that create physical structures or increase habitat complexity (Crooks & Khim 1999; Crooks 2002; Rodriguez 2006). Overall, this appears to be an accurate description of the effects of introduced byssally-attached mussels. Significant increases in the abundance or taxonomic richness of benthic

105 macroinvertebrates have been reported for several introduced mussel species, including Mytilus galloprovincialis (Lamarck) on sand (Griffiths et al. 1992; Robinson & Griffiths 2002; Branch & Steffani 2004), Musculista senhousia (Benson) on soft sediments (Crooks 1998; Crooks & Khim 1999; Mistri 2002), and Limnoperna fortunei (Dunker) on rocky substrates (Darrigran et al. 1998). However, declines in the abundance and richness of macroinvertebrates in the presence of introduced Musculista senhousia have also been reported (Creese 1997). Modulation of facilitative interactions between habitat engineers and benthic macroinvertebrates by environmental gradients may help to explain some of this variation (Cummings et al. 2001; Norkko et al. 2006). Similar types of organisms may respond in comparable ways to the presence of introduced mussels. Like the effects of Dreissena on sphaeriid clams, filter-feeding infaunal bivalves tend to be negatively affected by mats of Musculista senhousia (Creese 1997; Mistri 2002). Soft sediments underlying introduced mussel beds may become inhospitable for burrowing taxa through smothering and/or anoxia (Creese 1997; Robinson & Griffiths 2002). Also paralleling the effects of Dreissena, invasion by Mytilus galloprovincialis has been associated with the replacement of naturally occurring sandbank communities with species more typical of rocky shores (Robinson & Griffiths 2002). Increased densities of surface-feeding taxa, including amphipods and small gastropods, have been noted among beds of introduced M. senhousia (Crooks 1998; Crooks & Khim 1999; Mistri 2002) and M. galloprovianicialis on soft sediments (Robinson & Griffiths 2002). Predatory macroinvertebrates were also more abundant inside introduced mussel mats (Crooks & Khim 1999). Consistent with effects of Dreissena on hard substrates, densities of both epifaunal and infaunal taxa were higher in beds of introduced Limnoperna fortunei compared to bare rock (Darrigran et al. 1998). Facilitative effects of introduced engineer species are predicted to vary such that the size of the effect is dependent upon whether the resource provided by the engineer is limiting (Crooks 2002; Rodriguez 2006). In the case of Dreissena, several mechanisms of its effect on benthic macroinvertebrates may

106 vary in importance across sites. First, the importance of habitat provision for epifaunal and infaunal taxa is expected to vary with the local availability of suitable rocky and sedimentary habitats. Our examination of variation in the effects of Dreissena on the numerical densities of macroinvertebrate taxa across a range of substrate particle sizes supported this prediction for infaunal taxa, but not for epifaunal taxa. Conversely, changes in taxonomic richness appear to show the opposite pattern, with large positive effects on epifaunal taxa richness on fine sediments and smaller effects on infaunal taxa richness on hard substrates. However, low taxonomic resolution in primary studies prevented us from explicitly comparing changes in epifaunal and infaunal taxa richness. Depositional environments may represent sub-optimal conditions for epilithic taxa (e.g. through interference with feeding; Stewart et al. 1999), thereby preventing them from attaining high densities. Second, the importance of nutrient enrichment is expected to vary with the degree of nutrient limitation. We found that positive effects of Dreissena on deposit-feeding taxa increased with sediment particle size, although the same pattern was observed for other feeding groups. This pattern is consistent with a decreasing role of biodeposition as the organic matter content of sediments increases, also noted for the suspension-feeding pinnid bivalve Atrina zelandica (Gray) (Hewitt et al. 2002). Third, the importance of spatial refugia from biotic and abiotic stresses is expected to vary along environmental stress gradients (not examined here). Alternatively, the size of facilitative effects may be determined, or at least constrained, by local species pools.

In summary, although macroinvertebrate response is highly correlated with Dreissena density at a given site, or across sites within a given region, these correlations do not necessarily hold on a global scale. However, increased benthic macroinvertebrate density and taxonomic richness, and decreased community evenness, are generally associated with the addition of Dreissena. The strength of these effects is context-dependent, as they vary systematically with sediment particle size. Nevertheless, these patterns are robust to

107 heterogeneity in the methodology of primary studies. Interestingly, the effects of Dreissena are generally consistent with those of ecologically-similar taxa, suggesting universality in the interactions between introduced byssally-attached mussels and other macroinvertebrates.

4.6 Acknowledgements

Funding for this research was provided by a Canada Graduate Scholarship awarded to J.M. Ward by NSERC Canada.

4.7 References cited

Agrawal, A.A., Kotanen, P.M., Mitchell, C.E., Power, A.G., Godsoe, W. & Klironomos, J. (2005). Enemy release? An experiment with congeneric plant pairs and diverse above- and belowground enemies. Ecology, 86, 2979-2989. Beekey, M.A., McCabe, D.J. & Marsden, J.E. (2004). Zebra mussel colonisation of soft sediments facilitates invertebrate communities. Freshwater Biology, 49, 535-545. Bially, A. & MacIsaac, H.J. (2000). Fouling mussels (Dreissena spp.) colonize soft sediments in Lake Erie and facilitate benthic invertebrates. Freshwater Biology, 43, 85-97. Botts, P.S., Patterson, B.A. & Schloesser, D.W. (1996). Zebra mussel effects on benthic invertebrates: physical or biotic? Journal of the North American Benthological Society, 15, 179-184. Branch, G.M. & Steffani, C.N. (2004). Can we predict the effects of alien species? A case-history of the invasion of South Africa by Mytilus galloprovincialis (Lamarck). Journal of Experimental Marine Biology and Ecology, 300, 189-215. Clarke, A.H. (1981). Les Mollusques d’eau douce du Canada. Ottawa, Canada: Musée national des sciences naturelles.

108 Cooper, H.M. & Hedges, L.V. (Eds.) (1994). The Handbook of Research Synthesis. New York, NY: Russell Sage Foundation. Creese, R., Hooker, S., De Luca, S. & Wharton, Y. (1997). Ecology and environmental impact of Musculista senhousia (Mollusca: Bivalvia: Mytilidae) in Tamaki Estuary, Aukland, New Zealand. New Zealand Journal of Marine and Freshwater Research, 31, 225-236. Crooks, J.A. (1998). Habitat alteration and community-level effects of an exotic mussel, Musculista senhousia. Marine Ecology Progress Series, 162, 137-152. Crooks, J.A. (2002). Characterizing ecosystem-level consequences of biological invasions: the role of ecosystem engineers. Oikos, 97, 153-166. Crooks, J.A. & Khim, H.S. (1999). Architectural vs. biological effects of a habitat-atlering, exotic mussel, Musculista senhousia. Journal of Experimental Marine Biology and Ecology, 240, 53-75. Cummings, V.J., Thrush, S.F., Hewitt, J.E. & Funnell, G.A. (2001). Variable effects of a large suspension-feeding bivalve on infauna: experimenting in a complex system. Marine Ecology Progress Series, 209, 159-175. Darrigran, G., Martin, S.M., Gullo, B. & Armendariz, L. (1998). Macroinvertebrates associated with Limnoperna fortunei (Dunker, 1857) (Bivalvia, Mytilidae) in Río de la Plata, Argentina. Hydrobiologia, 367, 223- 230. Dittmann, S. (1990). Mussel beds – amensalism or amelioration for intertidal fauna? Helgoland Marine Research, 44, 335-352. Dusoge, K. (1966). Composition and interrelations between macrofauna living on stones in the littoral of Mikolajskie Lake. Ekologia Polska Seria A, 14, 755- 762. Englund, G., Sarnelle, O. & Cooper, S. (1999). The importance of data-selection criteria: meta-analysis of stream predation experiments. Ecology, 80, 1132- 1141. Gabriels, W., Goethals, P.L.M. & De Pauw, N. (2005). Implications of taxonomic modifications and alien species on biological water quality assessment as exemplified by the Belgian Biotic Index method. Hydrobiologia, 542, 137-150.

109 Greenwood, K.S., Thorp, J.H., Summers, R.B., Guelda, D.L. (2001). Effects of an exotic bivalve mollusc on benthic invertebrates and food quality in the Ohio River. Hydrobiologia, 462, 169-172. Griffiths, C.L., Hockey, P.A.R., Schurink, C. van E. & Le Roux, P.J. (1992). Marine invasive aliens on South African shores: implications for community structure and trophic functioning. South African Journal of Marine Science, 12, 713-722. Gurevitch, J., Curtis, P.S. & Jones, M.H. (2001). Meta-analysis in ecology. Advances in Ecological Research, 32, 199-247. Gurevitch, J. & Hedges, L.V. (1999). Statistical issues in ecological meta- analyses. Ecology, 80, 1142-1149. Gutiérrez, J.L., Jones, C.G., Strayer, D.L., & Iribarne, O.O. (2003). Mollusks as ecosystem engineers: the role of shell production in aquatic habitats. Oikos, 101, 79-90. Haynes, J.M., Tisch, N.A., Mayer, C.M. & Rhyne, R.S. (2005). Benthic macroinvertebrate communities in southwestern Lake Ontario following invasion of Dreissena and Echinogammarus: 1983 to 2000. Journal of the North American Benthological Society, 24, 148-167. Hedges, L.V., Gurevitch, J. & Curtis, P.S. (1999). The meta-analysis of response ratios in experimental ecology. Ecology, 80, 1150-1156. Hewitt, J.E., Thrush, S.F., Legendre, P., Cummings, V.J. & Norkko, A. (2002). Integrating heterogeneity across spatial scales: interactions between Atrina zelandica and benthic macrofauna. Marine Ecology Progress Series, 239, 115- 128. Horvath, T.G., Martin, K.M. & Lamberti, G.A. (1999). Effect of zebra mussels, Dreissena polymorpha, on macroinvertebrates in a lake-outlet stream. The American Midland Naturalist, 142, 340-347. Izvekova, E.I. & Lvova-Katchanova, A.A. (1972). Sedimentation of suspended matter by Dreissena polymorpha Pallas and its subsequent utilization by Chironomidae larvae. Polish Archives of Hydrobiology, 19, 203-210. Jacobi, C.M. (1987). The invertebrate fauna associated with intertidal beds of the

110 brown mussel Perna perna (L.) from Santos, Brazil. Studies on Neotropical Fauna and Environment, 22, 57-72. Jones, C.G., Lawton, J.H. & Shachak, M. (1994). Organisms as ecosystem engineers. Oikos, 69, 373-386. Jones, C.G., Lawton, J.H. & Shachak, M. (1997). Positive and negative effects of organisms as physical ecosystem engineers. Ecology, 78, 1946-1957. Jones, L.A. & Ricciardi, A. (2005). Influence of physicochemical factors on the distribution and biomass of invasive mussels (Dreissena polymorpha and Dreissena bugensis) in the St. Lawrence River. Canadian Journal of Fisheries and Aquatic Sciences, 62, 1953-1962. Karatayev, A.Y., Burlakova, L.E. & Padilla, D.K. (1997). The effects of Dreissena polymorpha (Pallas) invasion on aquatic communities in Eastern Europe. Journal of Shellfish Research, 16, 187-203. Klemm, D.J. (1985). A Guide to the Freshwater Annelida (Polychaeta, Naidid and Tubificid Oligochaeta, and Hirudinea) of North America. Dubuque, IA: Kendall/Hunt Publishing Company. Kuhns, L.A. & Berg, M.B. (1999). Benthic invertebrate community responses to round goby (Neogobius melanostomus) and zebra mussel (Dreissena polymorpha) invasion in southern Lake Michigan. Journal of Great Lakes Research, 25, 910-917. Lajeunesse, M.J. & Forbes, M.R. (2003). Variable reporting and quantitative reviews: a comparison of three meta-analytical techniques. Ecology Letters, 6, 448-454. Levine, J.M., Adler, P.B. & Yelenik, S.G. (2004). A meta-analysis of biotic resistance to exotic plant invasions. Ecology Letters, 7, 975-989. Lipsey, M.W. & Wilson, D.B. (2001). Practical Meta-Analysis. Thousand Oaks, CA: Sage Publications. Lohse, D.P. (1993). The importance of secondary substratum in a rocky intertidal community. Journal of Experimental Marine Biology and Ecology, 166, 1-17. Lozano, S.J., Scharold, J.V. & Nalepa, T.F. (2001). Recent declines in benthic macroinvertebrate densities in Lake Ontario. Canadian Journal of Fisheries

111 and Aquatic Sciences, 58, 518-529. Mayer, C.M., Keats, R.A., Rudstam, L.G. & Mills, E.L. (2002). Scale-dependent effects of zebra mussels on benthic invertebrates in a large eutrophic lake. Journal of the North American Benthological Society, 21, 616-633. McCarthy, J.M., Hein, C.L., Olden, J.D. & Vander Zanden, M.J. (2006). Coupling long-term studies with meta-analysis to investigate impacts of non-native crayfish on zoobenthic communities. Freshwater Biology, 51, 224-235. Merritt, R.W. & Cummins, K.W. (1984). An Introduction to the Aquatic Insects of North America (2nd ed.). Dubuque, IA: Kendall/Hunt Publishing Company. Mistri, M. (2002). Ecological characteristics of the invasive Asian date mussel, Musculista senhousia, in the Sacca di Goro (Adriatic Sea, Italy). Estuaries and Coasts, 25, 431-440. Mörtl, M. & Rothhaupt, K.-O. (2003). Effects of adult Dreissena polymorpha on settling juveniles and associated macroinvertebrates. International Review of Hydrobiology, 88, 561-569. Nalepa, T.F., Fanslow, D.L., Lansing, M.B. & Lang, G.A. (2003). Trends in the benthic macroinvertebrate community of Saginaw Bay, Lake Huron, 1987 to 1996: responses to phosphorus abatement and the zebra mussel, Dreissena polymorpha. Journal of Great Lakes Research, 29, 14-33. Norkko, A., Hewitt, J.E., Thrush, S.F. & Funnell, G.A. (2006). Conditional ourcomes of facilitation by a habitat-modifying subtidal bivalve. Ecology, 87, 226-234. Osenberg, C.W., Sarnelle, O. & Cooper, S.D. (1997). Effect size in ecological experiments: the application of biological models in meta-analysis. The American Naturalist, 150, 798-812. Osenberg, C.W., Sarnelle, O., Cooper, S.D. & Holt, R.D. (1999). Resolving ecological questions through meta-analysis: goals, metrics and models. Ecology, 80, 1105-1117. Parker, J.D., Burkepile, D.E. & Hay, M.E. (2006). Opposing effects of native and exotic herbivores on plant invasions. Science, 311, 1459-1461.

112 Parker, I.M., Simberloff, D., Lonsdale, W.M., Goodell, K., Wonham, M., Kareiva, P.M., Williamson, M.H., Von Holle, B., Moyle, P.B., Byers, J.E. & Goldwasser L. (1999). Impact: toward a framework for understanding the ecological effects of invaders. Biological Invasions, 1, 3-19. Peckarsky, B.L., Fraissinet, P.R., Penton, M.A. & Conklin, D.J. Jr. (1990). Freshwater Macroinvertebrates of Northeastern North America. Ithica, NY: Cornell University Press. Pennak, R.W. (1989). Fresh-water Invertebrates of the United States: Protozoa to Mollusca (3rd ed.). New York, NY: John Wiley & Sons. Ricciardi, A. (2003). Predicting the impacts of an introduced species from its invasion history: an empirical approach applied to zebra mussel invasions. Freshwater Biology, 48, 972-981. Ricciardi, A. & Atkinson, S.K. (2004). Distinctiveness magnifies the impact of biological invaders in aquatic ecosystems. Ecology Letters, 7, 781-784. Ricciardi, A. & Ward, J.W. (2006). Comment on ―Opposing effects of native and exotic herbivores on plant invasions‖. Science, 313, 298. Ricciardi, A., Whoriskey, F.G. & Rasmussen, J.B. (1995). Predicting the intensity and impact of Dreissena infestation on native unionid bivalves from Dreissena field density. Canadian Journal of Fisheries and Aquatic Sciences, 52, 1449- 1461. Ricciardi, A., Whoriskey, F.G. & Rasmussen, J.B. (1997). The role of the zebra mussel (Dreissena polymorpha) in structuring macroinvertebrate communities on hard substrata. Canadian Journal of Fisheries and Aquatic Sciences, 54, 2596-2608. Robinson, T.B. & Griffiths, C.L. (2002). Invasion of Langebaan Lagoon, South Africa, by Mytilus galloprovincialis – effects on natural communities. African Zoology, 37, 151-158. Rodriguez, L.F. (2006). Can invasive species facilitate native species? Evidence of how, when, and why these impacts occur. Biological Invasions, 8, 927-939. Ross, D.J., Johnson, C.R., & Hewitt, C.L. (2003). Variability in the impact of an introduced predator (Asterias amurensis: Asteroidea) on soft-sediment

113 assemblages. Journal of Experimental Marine Biology and Ecology, 288, 257- 278. Ruiz, G.M., Fofonoff, P., Hines, A.H. & Grosholz, E.D. (1999). Non-indigenous species as stressors in estuarine and marine communities: assessing invasion impacts and interactions. Limnology and Oceanography, 44, 950-972. Schloesser, D.W., Nalepa, T.F. & Mackie, G.L. (1996). Zebra mussel infestation of unionid bivalves (Unionidae) in North America. American Zoologist, 36, 300-310. Seed, R. (1996). Patterns of biodiversity in the macro-invertebrate fauna associated with mussel patches on rocky shores. Journal of the Marine Biological Association of the United Kingdom, 76, 203-210. Sephton, T.W., Paterson, C.G. & Fernando, C.H. (1980). Spatial interrelationships of bivalves and nonbivalve benthos in a small reservoir in New Brunswick, Canada. Canadian Journal of Zoology, 58, 852-859. Slepnev, A.Y., Protasov, A.A. & Videnina, Y.L. (1994). Development of a Dreissena polymorpha population under experimental conditions. Hydrobiological Journal, 30, 24-30. Stewart, T.W., Miner, J.G. & Lowe, R.L. (1998). Macroinvertebrate communities on hard substrates in western Lake Erie: structuring effects of Dreissena. Journal of Great Lakes Research, 24, 868-879. Stewart, T.W., Miner, J.G. & Lowe, R.L. (1999). A field experiment to determine Dreissena and predator effects on zoobenthos in a nearshore, rocky habitat of western Lake Erie. Journal of the North American Benthological Society, 18, 488-498. Strayer, D.L., Eviner, V.T., Jeschke, J.M. & Pace, M.L. (2006). Understanding the long-term effects of species invasions. Trend in Ecology and Evolution, 21, 645-651. Strayer, D.L. & Smith, L.C. (2000). Macroinvertebrates of a rocky shore in the freshwater tidal Hudson River. Estuaries and Coasts, 23, 359-366. Strayer, D.L., Smith, L.C. & Hunter, D.C. (1998). Effects of the zebra mussel (Dreissena polymorpha) invasion on the macrobenthos of the freshwater tidal

114 Hudson River. Canadian Journal of Zoology, 76, 419-425. Suchanek, T.H. (1986). Mussels and their rôle in structuring rocky shore communities. In P.G. Moore & R. Seed (Eds.), The Ecology of Rocky Coasts (pp. 70-96). New York, NY: Columbia University Press. Thorp, J.H. & Covich, A.P. (1991). Ecology and Classification of North American Freshwater Invertebrates (2nd ed.). San Diego, CA: Academic Press. Tokeshi, M. & Romero, L. (1995). Filling a gap: dynamics of space occupancy on a mussel-dominated subtropical rocky shore. Marine Ecology Progress Series, 119,167–176. Washington, H.G. (1984). Diversity, biotic and similarity indices: a review with special relevance to aquatic ecosystems. Water Research, 18, 653-694. Wisenden, P.A. & Bailey, R.C. (1995). Development of macroinvertebrate community structure associated with zebra mussel (Dreissena polymorpha) colonization of artificial substrates. Canadian Journal of Zoology, 73, 1438- 1443. Yager, P.L., Nowell, A.R.M. & Jumars, P.A. (1993). Enhanced deposition to pits: a local food source for benthos. Journal of Marine Research, 51, 209-236. Young, B.L., Padilla, D.K., Schneider, D.W. & Hewett, S.W. (1996). The importance of size-frequency relationships for predicting ecological impact of zebra mussel populations. Hydrobiologia, 332, 151–158.

115 Connecting statement 4–5

In Chapters 2 through 4, I have quantified the impacts of exotic ecosystem engineers by comparing macroinvertebrate communities in habitats modified by a dominant ecosystem engineering organism to habitat patches where such organisms are absent. Chapter 5 highlights how these patterns may be complicated by the presence of a native autogenic ecosystem engineer in the unmodified habitat. This research adds a hitherto unstudied aspect to our understanding of the community-level impacts of Dreissena spp., namely that co- occurring ecosystem engineers (in this case the native filamentous green alga Cladophora) can modify the impact that the mussels exert on the community structure of associated organisms. This is also the first study to demonstrate that Dreissena patch topography influences benthic macroinvertebrate diversity, implying that, at small spatial scales, a patchy mussel distribution matters.

116

Chapter 5 – Community level effects of co-occurring native and exotic

ecosystem engineers

A version of this chapter appears as: Ward, J.M. & Ricciardi, A. (2010). Community-level effects of co-occurring native and exotic ecosystem engineers. Freshwater Biology. In press.

Reprinted with permission from Wiley-Blackwell. http://www.wiley.com/bw/journal.asp?ref=0046-5070

117 5.1 Abstract

Nonindigenous ecosystem engineers can substantially affect native biodiversity by transforming the physical structure of habitats. In the Great Lakes–St. Lawrence River system, introduced dreissenid mussels (Dreissena polymorpha and D. bugensis) and the native benthic macroalga Cladophora act as ecosystem engineers by increasing substratum complexity and providing interstitial habitat for benthic macroinvertebrates. We manipulated the topography and perimeter- to-area ratio of patches of dreissenid mussels in a series of colonization experiments conducted at two sites in the St. Lawrence River. Experimental substrata were variably colonized by Cladophora, prompting us to examine (i) how the topography of Dreissena patches affects benthic macroinvertebrate diversity, and (ii) the extent to which the effects of Dreissena are altered by the presence of another habitat-modifying organism (Cladophora). The results of our first experiment suggested that a patchy distribution of dreissenid mussels is an important driver of benthic diversity at small spatial scales. The results of our second and third experiments suggested that a native habitat engineer, Cladophora, modifies the impact of Dreissena on benthic macroinvertebrate communities. While macroalgal blooms have been linked to the large-scale impacts of Dreissena on light and nutrient availability, Dreissena shells inhibited Cladophora growth at our experimental scale. These findings demonstrate that the interactions between habitat-modifying species can complicate efforts to predict the community-level effects of an invasion.

118 5.2 Introduction

Since their introduction to North America in the 1980s, dreissenid mussels (Dreissena polymorpha (Pallas) and Dreissena bugensis Andrusov) have become a dominant component of benthic communities in many lakes and rivers. Dreissena spp. can act as ecosystem engineers (sensu Jones et al. 1997) by altering resource availability at lake-wide scales; filtration by Dreissena can substantially increase water clarity, thereby stimulating benthic algal growth, and redirects nutrients and energy from pelagic to littoral benthic communities (Lowe & Pillsbury 1995; Strayer et al. 1998; Mayer et al. 2002). Dreissena spp. also act as ecosystem engineers at local scales by creating structurally complex interstitial habitat between clumped mussels and by altering resource availability (e.g. deposited organic material) within mussel beds (Ricciardi et al. 1997; Stewart et al. 1998; Mörtl & Rothhaupt 2003). Clumped mussel shells alter benthic surface topography, hydrodynamic regime and passive sedimentation, and may serve as spatial refugia from predation and physical stresses for other invertebrates (Gutiérrez et al. 2003). For these reasons, dense aggregations of Dreissena generally cause local increases in the total density and taxonomic richness of smaller-bodied benthic macroinvertebrates (Ward & Ricciardi 2007; Chapter 4). Experiments comparing the relative contribution of physical effects (empty shell treatments) and biological effects (filtration and biodeposition by living mussels) have attributed most of these impacts to the physical characteristics of mussel aggregations (Ricciardi et al. 1997; Stewart et al. 1998; Horvath et al. 1999; Mörtl & Rothhaupt 2003). Although Dreissena generally facilitates other benthic macroinvertebrates, negative impacts have also been reported. Of these, the most conspicuous are dramatic declines in unionid clams, which may be outcompeted for food or smothered by Dreissena fouling their shells (Ricciardi et al. 1998 and references therein). Dreissena has also been associated with the decline of sphaeriid clams (Strayer et al. 1998; Ward & Ricciardi 2007 – Chapter 4) and crashes in populations of the burrowing amphipod Diporeia spp. in soft sediment habitats in

119 the Great Lakes (Nalepa et al. 1998; Lozano et al. 2001; Nalepa et al. 2003). On rocky substrata, Dreissena may competitively exclude other primary space occupiers, such as net-spinning caddisflies and large snails (Wisenden & Bailey 1995; Ricciardi et al. 1997; Ward & Ricciardi 2007 – Chapter 4). Furthermore, a reduction in evenness has also been associated with Dreissena invasion, indicating that some macroinvertebrate species benefit disproportionately from the modified habitat (Ward & Ricciardi 2007; Chapter 4). The impact of Dreissena on benthic macroinvertebrate communities has typically been quantified by comparing communities in the presence and absence of dense mussel colonies, or by measuring changes in macroinvertebrate communities along a density gradient of Dreissena. However, mussels are often patchily distributed in space, and studies from marine systems have shown that the spatial distribution or patch topography of mussel beds can influence benthic macroinvertebrate abundance and diversity at local scales (Tsuchiya & Nishihira 1985; Tanaka & Magalhães 2002). For example, large patches of marine mussels support more species than small patches (Tsuchiya & Nishihira 1985), and the abundance of some mobile grazers increases with the availability of patch edge habitat, which they may occupy preferentially as a defence against predators or physical stressors (Tanaka & Magalhães 2002). Moreover, mussels and other ecosystem engineers are predicted to enhance species richness at large (e.g. landscape) spatial scales (that include both engineer-modified and unmodified habitat patches) owing to increased habitat heterogeneity (Jones et al. 1997; Wright et al. 2002; Badano et al. 2006). We conducted a series of field experiments to investigate the effect of Dreissena patch topography on freshwater benthic diversity. More specifically, we manipulated habitat heterogeneity, including (1) the number of Dreissena- modified and unmodified (bare substratum) habitat patches, and (2) the perimeter- to-area ratio of Dreissena patches, while holding spatial scale constant. We hypothesized that macroinvertebrate diversity would increase with both types of habitat heterogeneity.

120 Some of our experimental substrata were heavily colonized by native filamentous green algae, predominately Cladophora. The co-occurrence of Dreissena and Cladophora is becoming increasingly common, as recent and widespread macroalgal blooms throughout the Great Lakes are linked to the large- scale impacts of Dreissena (Haynes et al. 1999; Higgins et al. 2008; Malkin et al. 2008). In particular, Cladophora appears to be stimulated by increased water clarity and light availability, and may also benefit from phosphorus inputs from mussel biodeposits (Higgins et al. 2008; Malkin et al. 2008). Like Dreissena, Cladophora can act as a physical ecosystem engineer by creating structurally complex habitat and spatial refugia, as well as reducing current velocity, enhancing sedimentation and shading the substratum (Hart 1992; Holomuzki et al. 1999; Feminella & Resh 1991; Dodds & Gudder 1992). Variable colonization of our experimental substrata by Cladophora allowed an additional, unplanned comparison in our study and enabled us to determine whether this native species modifies the effects of the exotic ecosystem engineer, Dreissena, on benthic macroinvertebrate communities.

5.3 Methods

5.3.1 Study sites

Experiments were conducted in shallow water (<3m depth) at two sites located in Lake St. Louis, a fluvial lake in the upper St. Lawrence River, Quebec, Canada. The first experiment was carried out at a site near Lachine, Quebec (73°41′11.5′′ W, 45°26′01′′ N) in a small bay. The benthic habitat at this site consists mainly of abundant cobbles and large boulders colonized by Dreissena and benthic algae (Cladophora) overlying a thick layer of silt. Zebra mussels (D. polymorpha) dominated the Dreissena assemblage, while quagga mussels (D. bugensis) constituted a minor component (<1%). The main fish species present were yellow perch (Perca flavescens (Mitchill)) and rock bass (Ambloplites rupestris (Rafinesque)). Aquatic plant management (removal of macrophytes by dredging)

121 occurred during the experiment and probably contributed to the high turbidity at this site. The second and third experiments were conducted at Pointe-du-Moulin, Île Pérrot (73°51′15′′ W, 45°21′55′′ N), a historical site where human activities are restricted near the shore. This site has a faster flowing current and is less turbid than the Lachine site. Sediments mainly consist of limestone bedrock with scattered large boulders. Zebra and quagga mussels were patchily distributed and approximately equally abundant. The site had abundant macrophyte beds (predominantly Vallisneria americana Michx.). Common macroinvertebrate predators included yellow perch, rock bass, smallmouth bass (Micropterus dolomieu Lacepède) and crayfish (Orconectes spp.).

5.3.2 Experiment 1

Experimental substrata consisted of solid rectangular cement bricks (individually measuring 19 cm x 9 cm x 5.7 cm), which were subject to one of three Dreissena shell treatment levels (n = 10 replicates per treatment). These consisted of bricks with their upper surface (19 x 9 cm) entirely covered in a single layer of Dreissena shells (full shell treatment, mean 47 ± 15 mussels brick-1), bricks with one half of their upper surface covered in Dreissena shells (half shell treatment, mean 25 ± 5 mussels brick-1), and control bricks to which no Dreissena shells were attached. Dreissenid mussels were collected at the site from a depth of 1-3 m. Mussels were boiled, eviscerated and rinsed to remove all mussel tissue and attached invertebrates, and the shells were dried at ambient temperature for several days. Shell valves were then glued together and attached to experimental substrata at their base (anteroventral region) using non-toxic silicone aquarium glue. The experimental substrata were deployed at Lachine from 1 September to 6 October 2003 (35 days). Bricks were collected by carefully placing them in plastic freezer bags while underwater, and transported to the laboratory where all macroinvertebrates retained on a 500µm sieve were preserved in 70% ethanol and

122 were later identified, usually to family, following Merrit & Cummins (1984), Pennak (1989), Thorp & Covich (1991) and Witt et al. (1997). One full shell brick was not recovered.

5.3.3 Experiment 2

Experimental substrata were subjected to one of four Dreissena shell treatment levels (n = 20 replicates per treatment), three of which were identical to those in Experiment 1 (full shell treatment, mean 46.9 ± 0.5 mussels brick-1; half shell treatment, 23.3 ± 0.5; no shell treatment). An additional shell treatment level was included to investigate the effects of mussel patch area-to-perimeter ratio on benthic biodiversity. As in the half shell treatment, this consisted of bricks with one half of their upper surface covered in Dreissena shells, but this time arranged in two patches of equal size (checker: mean 23.5 ± 0.5 mussels brick-1). In the checker shell treatment, mean Dreissena density and total patch area were equal to that of the half shell treatment (85.5 cm2), while total patch perimeter was equal to that of the full shell treatment (56 cm). These treatment densities were within the range of Dreissena densities found naturally at the study site. Experimental substrates were prepared following the same methodology as in Experiment 1. Experimental substrata were deployed at Pointe-du-Moulin from 12 July to 20 September 2004 (70 days). Colonizing macroinvertebrates were collected and preserved following the methods of Experiment 1, and were identified to the lowest possible taxonomic level, usually to species or genus, following the references cited above, as well as Resh (1976) and Klemm (1985) to identify Ceraclea caddisflies and annelids, respectively, to at least the genus level. One control brick was not recovered. Experimental substrata were heavily colonized by attached filamentous macroalgae (predominantly Cladophora and hereafter referred to as such), which sometimes overgrew Dreissena patches. Because Cladophora was suspected to influence macroinvertebrate colonization, it was removed from 40 bricks (n=10 replicates from each Dreissena treatment) by scraping Dreissena shell and brick surfaces with forceps. Cladophora samples

123 were dried at 70ºC for 24 hours, and the dry weight was measured with a Denver Instrument (Denver, Colorado, USA) APX-602 electronic balance.

5.3.4 Experiment 3

Due to heavy fouling of Experiment 2 by Cladophora, the experiment was replicated in 2006. The study site, experimental design and methodology for Experiment 3 were identical to that for Experiment 2 except that substrata were exposed for 17 days less. Experimental substrata (n=10 replicates per treatment; full, mean 56.1 ± 1.1 mussels brick-1; half, 28.2 ± 0.9; checker, 26.4 ± 1.3) were deployed from 20 July to 11 September 2006 (53 days). All bricks were recovered.

5.3.5 Statistical analysis

Univariate and multivariate analyses were used to test the general hypothesis that macroinvertebrate assemblages differed among Dreissena shell treatments and, for Experiments 2 and 3, to assess the extent to which Cladophora modified Dreissena effects. The duration and scheduling of the experiments did not allow extensive colonization of our substrata by Dreissena (mean number of live mussels per brick ± SE = 0.67 ± 0.08); those few individuals found on the substrata were excluded from analysis because we were not concerned with testing the influence of Dreissena shells on Dreissena recruitment. Univariate analyses were conducted on four community response variables: total macroinvertebrate density, taxonomic richness, Simpson‘s diversity index (D) and Simpson‘s evenness index (ED). For Experiment 1, the effect of Dreissena shell treatment on each community response variable was analysed using an analysis of variance (ANOVA), followed by a posteriori Tukey‘s HSD tests when appropriate. For Experiments 2 and 3, the effects of experimental Dreissena shell treatment and unmanipulated Cladophora gradient on each community response variable were assessed using an analysis of

124 covariance (ANCOVA) with Cladophora biomass as a covariate. These models were reduced to ANOVAs for Dreissena treatment effects by sequentially removing the interaction term and covariate when appropriate, as indicated by the significance of a general linear test. For Experiments 2 and 3, Cladophora biomass was compared among experimental Dreissena treatments using ANOVA. Univariate analyses were performed on square root transformed macroinvertebrate abundance and Cladophora biomass, and were conducted using the R (v. 2.8.1) software package (R Development Core Team, Vienna, Austria). The effects of Dreissena and Cladophora on non-dreissenid macroinvertebrate community composition were analysed using multivariate routines in the PRIMER (v. 6) software package (Clarke & Gorley 2006). For each experiment, a matrix of Bray-Curtis dissimilarities between samples was produced using square root transformed, non-standardized abundances. We tested for differences among Dreissena treatment levels using analysis of similarity (ANOSIM; Experiment 1), and among Dreissena treatment levels and along Cladophora biomass gradient using permutational multivariate analysis of variance (PERMANOVA; Anderson et al. 2008; Experiments 2 and 3). The significance of each term was tested with 9999 random permutations of the relevant units. Cladophora biomass was square root transformed prior to these analyses. When ANOSIM or PERMANOVA detected a significant effect of Dreissena treatment, the similarity percentages (SIMPER) procedure was used to identify which taxa contributed most to the Bray-Curtis dissimilarity between Dreissena treatment levels. When PERMANOVA detected a significant effect of Cladophora, the relationship between Cladophora biomass and the abundance of individual macroinvertebrate taxa was examined using Spearman‘s rank correlation and significance was assessed by t-tests. Non-metric multidimensional scaling (nMDS) ordination was performed on the Bray-Curtis dissimilarity matrix for each experiment in order to generate a graphical representation of the multivariate data.

125 5.4 Results

5.4.1 Experiment 1

A total of 20 non-dreissenid macroinvertebrate taxa were collected from experimental substrata. All 20 taxa were present on the half shell bricks, while a total of 13 and 14 taxa occurred on the control and full bricks, respectively (Table E.1 in Appendix E). The most abundant taxon was the amphipod Gammarus fasciatus Say, followed by chironomids, flatworms (Planariidae), Hydracarina and planorbid snails. Total macroinvertebrate density increased significantly with successive increases in Dreissena shell cover (ANOVA, F = 54.74, P < 0.0001; Fig. 5.1a). Mean taxonomic richness was higher in both the half and full shell treatments than in the control treatment but did not differ significantly between half and full shell bricks (ANOVA, F = 19.21, P < 0.0001; Fig. 5.1b). Simpson‘s diversity index was significantly lower on full than on either control or half shell bricks, while the latter two did not differ (ANOVA, F = 6.32, P = 0.006; Fig. 5.1c). Simpson‘s evenness index was significantly higher in the control treatment than in either the half or full shell treatments (ANOVA, F = 34.45, P < 0.0001; Fig. 5.1d). Multivariate analyses indicated that the macroinvertebrate assemblage structure differed across the Dreissena shell treatments (ANOSIM, Global R = 0.42, P = 0.01; Fig. 5.2a). Pair-wise comparisons indicated that there were significant differences in assemblage structure between the control and full shell bricks (R = 0.59, P = 0.01) and between the control and half shell bricks (R = 0.49, P = 0.01), but not between the half and full shell bricks (R = 0.27, P = 0.1). Assemblages on control bricks were also more dissimilar than those in either the half or full shell treatments; the average within-treatment Bray-Curtis similarity was 44.24 for control bricks, compared to 68.75 and 74.96 for half and full shell bricks, respectively. These patterns are reflected in the nMDS plot (Fig. 5.2a). The amphipod Gammarus fasciatus accounted for most of the Bray-Curtis dissimilarity between Dreissena treatment levels (SIMPER, Table 5.1). Other

126

Figure 5.1. a) Total non-dreissenid macroinvertebrate abundance, b) taxonomic richness, c) Simpson‘s diversity index and d) Simpson‘s evenness index on the three Dreissena shell treatments in Experiment 1. Data are expressed as mean (± SE) per brick. Letters denote significant differences between treatment levels (Tukey‘s tests, P < 0.05).

127 Figure 5.2. Non-metric MDS ordination for a) Experiment 1, b) Experiment 2 and c) Experiment 3, based on Bray-Curtis dissimilarity applied to square root transformed species abundances. Labels indicate control (Z), half (H), chequer (C) and full (F) Dreissena shell treatments. Cladophora biomass is indicated by the size of the ‗bubbles‘ in b) and c).

128 A) 2D Stress: 0.14 Z Z

Z H H F H F F Z F H F H F H H F H Z Z F F H H Z

Z Z

Z

B) 2D Stress: 0.22 Z

C H F H Z F F H F Z Z Z Z F F C H F F F H C H C H C C H C H Z C F C H Z C Z

C) 2D Stress: 0.25 Z Z

F Z

F Z F F C F Z H Z H Z H

F C C H Z H H H

C H H F C C F F C C C Z C F H Z

129 Table 5.1. Similarity percentages (SIMPER) analysis for Experiment 1, showing which taxa made the greatest percent contribution ( %) to the Bray-Curtis dissimilarity between Dreissena shell treatment levels. Only taxa contributing >3% are shown.

Control vs. Full Control vs. Half

Average dissimilarity = 61.05 % Average dissimilarity = 56.37 %

Gammarus fasciatus 49.88 Gammarus fasciatus 31.99 Caecidotea 6.66 Planariidae 7.99 Caenidae 5.95 Caenidae 7.55 Planariidae 5.84 Planorbidae 7.21 Chironomidae 5.37 Chironomidae 6.25 Planorbidae 5.29 Caecidotea 5.84 Hydracarina 5.03 Hydracarina 4.85 Hydrobiidae 3.75 Hydrobiidae 4.66 Leptoceridae 3.17 Naididae 4.58 Total = 90.94 Glossiphoniidae 3.92 Leptoceridae 3.39 Total = 88.24

130 taxa that made an important contribution to dissimilarity between treatment levels include isopods (Caecidotea), caenid mayflies, flatworms (Planariidae), chironomids and planorbid snails.

5.4.2 Experiment 2

A total of 63 non-dreissenid macroinvertebrate taxa were identified from experimental substrata (Table E.1 in Appendix E), with 45, 43, 43 and 46 taxa collected from control, half, checker, and full bricks, respectively. The most abundant taxon was the native amphipod Gammarus fasciatus, followed by the nonindigenous Eurasian amphipod Echinogammarus ischnus (Stebbing), chironomids, the snail Gyraulus deflectus (Say) and the flatworm Dugesia tigrina Girard. Mean Cladophora biomass did not differ significantly among Dreissena treatment levels (Fig. 5.3a). In univariate analyses, the Dreissena x Cladophora interaction term did not explain a significant amount of variation in any of the community response variables, nor did Cladophora biomass when it was included as a covariate (Table 5.2). Dreissena treatment did not explain a significant amount of variation in total macroinvertebrate density or Simpson‘s evenness index. When Cladophora was removed from the model, the difference in Simpson‘s diversity index across Dreissena treatments was marginally insignificant (ANOVA, P = 0.06). This was probably driven by differences between control and full shell bricks, which had a mean Simpson‘s diversity of 4.31 (± 0.38) and 5.55 (± 0.29), respectively. Multivariate analyses indicated that the macroinvertebrate assemblage differed significantly among Dreissena shell treatments but this was not related to Cladophora biomass (PERMANOVA, Table 5.3; Fig. 5.2b). Pairwise comparisons detected a significant difference between the control and checker shell treatments (P = 0.009). When Cladophora was removed from the model, the difference between control and full bricks became significant (P = 0.009). Five taxa consistently made large contributions to the dissimilarity between these Dreissena treatment levels (SIMPER, Table 5.4). These were the amphipods

131

A)

B)

Figure 5.3. Mean (± SE) Cladophora density on the four Dreissena shell treatments in a) Experiment 2 and b) Experiment 3. Letters denote significant differences between treatment levels (Tukey‘s tests, P < 0.05).

132 Table 5.2. Univariate analyses of community response variables for Experiment 2. Cladophora biomass and total macroinvertebrate abundance were square root transformed prior to analysis.

Source of variation df Total abundance Richness Diversity (D) Evenness (ED)

MS F P MS F P MS F P MS F P

Full model (ANCOVA) Dreissena (T) 3 6.44 0.62 0.61 15.39 1.51 0.23 3.40 1.70 0.19 0.007 0.58 0.63 Cladophora (Cov) 1 0.72 0.07 0.79 14.87 1.46 0.24 4.63 2.31 0.14 0.002 0.20 0.66 T x Cov 3 9.90 0.95 0.43 2.30 0.23 0.88 3.07 1.53 0.23 0.017 1.36 0.27 Residual 31 10.43 10.21 2.00 0.012 Total 38

Reduced model (ANOVA) Dreissena (T) 3 4.86 0.51 0.68 18.30 2.24 0.09 5.65 2.65 0.06 0.010 0.96 0.42 Residual 75 9.59 8.17 2.13 0.010 Total 78

133 Table 5.3. Permutational analysis of variance (PERMANOVA) for Experiment 2, based on Bray-Curtis dissimilarities using square root transformed taxon abundances. P-values generated by permutation (9999 iterations).

Source of variation df MS Pseudo-F P Pair-wise comparisons P

Cladophora (Cov) 1 489.28 0.951 0.470 Control vs. Checker 0.009 Dreissena (T) 3 890.65 1.732 0.009 Control vs. Half 0.309 Cov x T 3 549.61 1.069 0.365 Control vs. Full 0.078 Residual (R) 31 514.40 Checker vs. Half 0.078 Total 38 Checker vs. Full 0.357 Half vs. Full 0.200

134 Table 5.4. Similarity percentages (SIMPER) analysis for Experiment 2, showing which taxa made the greatest percent contribution ( %) to the Bray-Curtis dissimilarity between Dreissena shell treatment levels. Only taxa contributing >3% are shown.

Control vs. Checker Control vs. Full

Average dissimilarity = 35.00 % Average dissimilarity = 30.30 %

Gammarus fasciatus 14.09 Gammarus fasciatus 12.22 Echinogammarus ischnus 10.99 Echinogammarus ischnus 10.82 Chironomidae 8.46 Dugesia tigrina 9.32 Dugesia tigrina 6.84 Chironomidae 7.62 Gyraulus deflectus 4.97 Gyraulus deflectus 5.19 Stagnicola catascopium 4.11 Naididae 3.39 Valvata tricarinata 3.95 Prostoma graecense 3.28 Agraylea 3.33 Stagnicola catascopium 3.08 Amnicola limosa 3.32 Hydracarina 3.07 Hydropsyche 3.14 Valvata tricarinata 3.05 Hydracarina 3.09 Total = 61.03 Total = 66.29

135 Gammarus fasciatus and Echinogammarus ischnus, the flatworm Dugesia tigrina, chironomids, and the snail Gyraulus deflectus. The average Bray-Curtis dissimilarities between Dreissena treatment levels were lower than in Experiment 1.

5.4.3 Experiment 3

A total of 45 non-dreissenid macroinvertebrate taxa were identified from experimental substrata, with a total of 32 taxa collected from both control and half shell bricks, and a total of 34 taxa collected from both checker and full shell bricks (Table E.1 in Appendix E). Chironomids were the most abundant taxon, followed by Gammarus fasciatus, Gyraulus deflectus, Dugesia tigrina and Echinogammarus ischnus. Mean Cladophora biomass varied significantly among Dreissena treatment levels (ANOVA, F = 3.64, P = 0.02), being significantly higher on control bricks than on all other Dreissena shell treatments (Fig. 5.3b). In univariate analyses, none of the Dreissena x Cladophora interaction terms explained a significant amount of variation in any of the community response variables, nor did Cladophora biomass when it was included as a covariate (Table 5.5). Regardless of whether these terms were included in the model, Dreissena treatment did not explain a significant amount of variation in total macroinvertebrate density, taxonomic richness, Simpson‘s diversity index or Simpson‘s evenness index. Multivariate analyses indicated that the macroinvertebrate assemblage structure was related to Cladophora biomass but did not differ significantly among Dreissena shell treatments after accounting for this relationship (PERMANOVA, Table 5.6; Fig. 5.2c). This result was probably driven by negative correlations between Cladophora biomass and the abundance of two species, the amphipods Gammarus fasciatus (Spearman‘s rank correlation, rs = -

0.381, P = 0.04) and Echinogammarus ischnus (rs = -0.396, P = 0.03), as well as positive correlations between Cladophora biomass and the abundance of three species: the snail Physa gyrina (Say) (rs = 0.449, P = 0.01), the leech

136 Table 5.5. Univariate analyses of community response variables for Experiment 3. Cladophora biomass and total macroinvertebrate abundance were square root transformed prior to analysis.

Source of variation df Total abundance Richness Diversity (D) Evenness (ED)

MS F P MS F P MS F P MS F P

Full model (ANCOVA) Dreissena (T) 3 11.72 1.32 0.29 6.76 0.91 0.45 2.03 0.59 0.62 0.016 0.93 0.44 Cladophora (Cov) 1 1.49 0.17 0.69 0.80 0.11 0.75 1.48 0.43 0.52 0.009 0.53 0.47 T x Cov 3 7.11 0.80 0.50 11.20 1.50 0.23 1.13 0.33 0.80 0.006 0.34 0.79 Residual 32 8.91 7.47 3.42 0.017 Total 39

Reduced model (ANOVA) Dreissena (T) 3 11.72 1.37 0.27 6.76 0.89 0.46 2.03 0.64 0.59 0.016 0.99 0.41 Residual 36 8.55 7.60 3.17 0.016 Total 39

137 Table 5.6. Permutational analysis of variance (PERMANOVA) for Experiment 3, based on Bray-Curtis dissimilarities using square root transformed taxon abundances. P-values generated by permutation (9999 iterations).

Source of variation df MS Pseudo-F P

Cladophora (Cov) 1 1288.80 2.141 0.018 Dreissena (T) 3 799.41 1.328 0.115 Cov x T 3 446.70 0.742 0.889 Residual (R) 32 601.90 Total 39

138 Batracobdella phalera (Graf) (rs = 0.434, P = 0.02) and the oligochaete Stylaria lacustris (L.) (rs = 0.371, P = 0.04). After sequential Bonferroni correction (Rice 1989), none of these correlations remain significant. However, the risk of Type II error associated with this procedure argues against its use (Moran 2003).

5.5 Discussion

5.5.1 Effect of Dreissena patch topography on benthic diversity

In our first experiment, substrata completely covered with Dreissena shells had higher total macroinvertebrate density and species richness, as well as lower Simpson‘s diversity and evenness indices, compared to control bricks lacking Dreissena. These findings are consistent with previous studies (Ward & Ricciardi 2007; Chapter 4) and support the hypothesis that some Dreissena impacts on macroinvertebrate communities are predictable across invaded sites. Despite higher species richness, full shell bricks had lower diversity and evenness indices than control bricks, owing to increased dominance by the amphipod Gammarus fasciatus, which accounted for more than 80% of all macroinvertebrates collected from full shell bricks. Gammarid amphipods frequently dominate macroinvertebrate assemblages on Dreissena-covered substrata, probably in response to mussel-generated interstitial habitat as well as to the provision of refugia from predation and hydrodynamic stress (Stewart & Haynes 1994; Ricciardi et al. 1997; Stewart et al. 1998). Simpson‘s diversity remained relatively high in the half shell treatment, which contained patches of both Dreissena-modified and unmodified habitats. This reflected both high taxonomic richness and lower dominance by G. fasciatus. Interestingly, mean taxonomic richness tended to be highest on the half shell bricks, which were also colonized by the greatest total number of macroinvertebrate taxa (Fig. 5.1b, Table E.1 in Appendix E). This result contrasts with the assumption that the effect of Dreissena on benthic communities is a simple linear function of mussel density (e.g. Strayer & Smith 2000; Mayer et al. 2002), and further suggests that the

139 spatial arrangement of mussel patches, along with mussel abundance (Horvath et al. 1999; Strayer & Smith 2000; Mayer et al. 2002; Ricciardi 2003) and size frequency distribution (Young et al. 1996), modifies the impact. Factors that affect the arrangement of Dreissena patches, such as changes in settlement patterns or disturbance (e.g. mortality caused by sponge overgrowth; Ricciardi et al. 1995) may therefore be important. That the same pattern was not observed in Experiments 2 and 3 indicates either that this result was site-specific, or the result is general but was obscured in subsequent experiments by Cladophora. We believe the latter explanation is more likely. It has been predicted that ecosystem engineers should generally increase species richness at large (e.g. landscape) spatial scales that include both engineer- modified and unmodified habitat patches (Jones et al. 1997; Wright et al. 2002). This is purported to result from enhanced landscape-scale habitat heterogeneity, which should increase species richness if some species occur only in one habitat type (Wright et al. 2002, Badano et al. 2006). In our study, spatial scale was held constant while habitat heterogeneity (the number of different habitat types) was manipulated. Our results from the half shell treatment, where both modified and unmodified habitats were represented, support a positive relationship between habitat heterogeneity and species richness. This pattern arose in part because of the mechanism above: five taxa were present in the half shell treatment that otherwise occurred in either the control or full shell treatment but not in both (E. ischnus, heptageniid mayflies, polycentropodid caddisflies, tubificid oligochaetes and Bithynia tentaculata L.). An additional four taxa occurred only in the half shell treatment: valvatid snails and brachycentrid, hydrophychid and hydroptilid caddisflies. This suggests an added contribution of habitat diversity (generated by ecosystem engineers) to species richness at small spatial scales, arising from the preferential colonization of heterogeneous (or patchy) habitats by some species. However, these four taxa were relatively rare, and it is unclear whether this pattern would hold at larger spatial scales or with greater sampling intensity.

140 5.5.2 Effect of Cladophora on benthic macroinvertebrates

A review of the literature on the effects of Cladophora on benthic invertebrates reveals mixed results (Fig 5.4). In one study, Cladophora density was negatively correlated with diversity (H‘) and evenness (J‘) of stream macroinvertebrates, although it was positively correlated with Baetis, Simulium and chironomid abundance (Ellsworth 2000). Elsewhere, Cladophora was associated with increased total density and taxonomic richness of stream invertebrates, although it negatively affected some dipteran taxa (Blepharicera and large-bodied Simulium), apparently via competition for substratum (Dudley et al. 1986). Similarly, Hart (1992) found positive correlations between Cladophora cover and the abundance of chironomids, the mayfly Stenonema, and the stoneflies Taeniopteryx and Isoperla, as well as negative correlations with grazing caddisflies (Leucotrichia and Psychomyia) and the tipulid Antocha. Harrison & Hildrew (2001) reported a positive association between Cladophora and tube-dwelling chironomids (Glyptotendipes pallens (Meigen) and Microtendipes pedellus (de Geer)) in a lake littoral. In Lake Ontario, Cladophora was positively associated with gammarid amphipods and naidid oligochaetes, and negatively associated with chironomids (Barton & Hynes 1978). Some of the macroinvertebrates that commonly increase in abundance following Dreissena invasion, such as gammarid amphipods and snails (Ward & Ricciardi 2007; Chapter 4), are also expected to respond to Cladophora. In particular, a positive association has been documented between Gammarus fasciatus and Cladophora, which may provide the amphipod with both food resources and interstitial habitat (Barton & Hynes 1978; Stewart & Haynes 1994; Van Overdijk et al. 2003; Palmer & Ricciardi 2004). We detected a negative correlation between Cladophora and G. fasciatus in Experiment 3, but this was driven by the amphipod‘s positive response to Dreissena rather than avoidance of Cladophora (ANCOVA, effect of Dreissena: F = 4.48, P = 0.01, effect of Cladophora: F = 1.03, P = 0.32). We also found a negative association between Echinogammarus ischnus and Cladophora, which is consistent with the observation that this amphipod prefers substrata covered by Dreissena rather than

141

Figure 5.4. Schematic illustration of the biotic interactions between Dreissena, Cladophora and epibenthic macroinvertebrates. Details of these interactions are discussed by Feminella and Resh (1991), Dodds and Gudder (1992), Vanderploeg et al. (2002), and Higgins et al. (2008).

142 Cladophora (Van Overdijk et al. 2003, Palmer & Ricciardi 2004). Snails and other macroinvertebrates that graze on diatoms may avoid Cladophora (Haynes et al. 1999). The abundance of benthic diatoms, an important food source for grazers, declined with increasing biomass of filamentous green algae in Lake Huron following Dreissena invasion (Lowe & Pillsbury 1995). Cladophora may inhibit the growth of epilithic microalgae by reducing light penetration (Feminella & Resh 1991). In comparison to diatoms, Cladophora has lower nutritional value and is generally not a preferred food for freshwater grazers (Dodds & Gudder 1992). However, epiphytic algae growing on Cladophora may be important for some grazers (Dodds 1990). In our study, Cladophora biomass did not explain variation in total macroinvertebrate density, taxonomic richness, Simpson‘s diversity index or Simpson‘s evenness index. However, Cladophora biomass explained a significant amount of variation in community composition in Experiment 3, when it was present at relatively low densities, but not in Experiment 2, when it occurred at relatively high densities. This suggests that the response of macroinvertebrates to Cladophora is nonlinear. Indeed, previous studies have demonstrated that macroinvertebrate responses to increasing habitat structure can be nonlinear (e.g. Stewart et al. 2003). Furthermore, Cladophora has been shown to have significant effects on benthic macroinvertebrates at relatively low abundances; for example, Holomuzki et al. (1999) found that 4-13% cover of Cladophora on rocks was enough to nullify the effects of predatory stoneflies (Plecoptera) on drift of their larval hydropsychid prey. The Cladophora cover on substrata in Experiment 2 was typically much higher than 13% and reached 100% on some bricks (J. M. Ward, personal observation), which may have exceeded the range of Cladophora densities within which variation in its effect is most pronounced.

5.5.3 Is Cladophora obscuring the effect of Dreissena?

If the impact of Dreissena on benthic macroinvertebrates occurs mostly through

143 the physical effects of interstitial habitat among the shells (an assumption supported by previous experiments comparing the effects of Dreissena shells to living mussels: Ricciardi et al. 1997; Stewart et al. 1998; Horvath et al. 1999; Mörtl & Rothhaupt 2003) then it follows that this effect will be reduced in systems where structurally complex habitat is not limiting. That is, in communities with abundant benthic algae such as Cladophora, the effect of the mussels should be less than in systems where structurally complex habitats are scarce (Gutiérrez et al. 2003). These habitats may not have been limiting on control substrata colonized by Cladophora, where the alga likely provided some of the same benefits as mussel-engineered habitat to benthic fauna (Fig 5.4), such as interstices and refugia from predation (Hart 1992; Holomuzki et al. 1999). Our experiments were not designed to examine the effect of Cladophora; consequently, we lacked an adequate experimental control for both Cladophora and Dreissena effects. Neither Experiment 2 nor Experiment 3 had a treatment lacking both Dreissena and Cladophora, which we believe contributed to the lack of statistically significant effects on community response variables. In Experiment 2, control bricks (as well as other treatments) were heavily colonized by Cladophora, which may have reduced variation in habitat heterogeneity and structural complexity across Dreissena treatments. However, the significant effect of Dreissena shells on community structure, and the tendency for Simpson‘s diversity index to be higher on full versus control bricks in Experiment 2, indicates that some taxa responded to the habitat provided by Dreissena shells even in the presence of heavy Cladophora fouling. Here, Dreissena restructured the macroinvertebrate community without changing its total abundance. Cladophora and associated accumulation of fine sediment sometimes clogged the spaces between Dreissena shells and may have reduced the amount of shell- generated habitat available to macroinvertebrates (J. M. Ward, personal observation). Similarly, Jacobi (1987) found the density of macroinvertebrates associated with the brown mussel Perna perna (L.) was reduced when sediment covered the mussel shells and filled interstitial spaces.

144 In Experiment 3, Cladophora was most abundant on control bricks, possibly as a result of shading by mussel shells. It is also possible that grazing invertebrates inhibited Cladophora growth on Dreissena shells. Cladophora may deter grazers by chemical defences when it is mature but lacks these defences during its initial growth, and grazing insects have previously been shown to inhibit Cladophora establishment in streams (Dudley & D'Antonio 1991; Hart 1992) and lakes (Harrison & Hildrew 2001). However, we found no negative associations between the abundance of grazing macroinvertebrates and Cladophora and one grazer, the snail Physa gyrina, appeared to be positively associated with Cladophora. We also found positive associations between Cladophora and both the deposit-feeding oligochaete Stylaria lacustris and the predatory leech Batracobdella phalera. These species were probably responding to habitat structure and possibly to increased availability of food or prey. Several studies in marine systems have compared benthic macroinvertebrate communities in habitats created by introduced bivalves and native macrophytes. Ferraro & Cole (2007) found that benthic macroinvertebrate communities inhabiting introduced oyster (Crassostrea gigas (Thunberg)) and native eelgrass (Zostera marina L.) habitats in Willapa Bay, Washington, were indistinguishable in terms of their total abundance, taxonomic richness and Bray- Curtis similarity. In another study, total macroinvertebrate density was higher in Z. marina beds compared to C. gigas beds, while the density of meiofauna did not differ between these habitats (Hosack et al. 2006). On biogenic reefs created by the native tube-building polychaete Sabellaria alveolata (L.), species richness was higher on reefs colonized by both oysters (C. gigas) and a green alga (Ulva armoricana Dion, Reviers & Coat) compared to reefs colonized by only the alga (Dubois et al. 2006). Elsewhere, the richness and identity of macroinvertebrates inhabiting a native mussel (Aulacomya ater (Molina)) bed and the macroalgae on its shell closely resembled that found in association with an exotic mussel (Mytilus galloprovincialis Lamarck) that was free of macroalgae, although the total density of associated macroinvertebrates was higher for the latter (Griffiths et al. 1992).

145 One additional line of evidence supports our hypothesis that Cladophora is responsible for the paucity of observed community-level impacts by Dreissena in Experiments 2 and 3. This comes from a survey of natural substrata (rocks, n = 30) spanning a range of Dreissena densities (0 to 2,468 m-2; mean = 1,086 m-2), which was conducted concurrently and at the same site as Experiment 1. Unlike the artificial substrata in Experiment 1, on which Cladophora was essentially absent, natural substrata at the site were heavily colonized by Cladophora (range: 0 to 381 g dry weight m-2; mean = 91 g dry weight m-2). Linear regression analyses detected no statistically significant relationship between Dreissena density (square root transformed) and the total density (fourth root transformed, R2 = 0.01, P = 0.62), taxonomic richness (fourth root transformed, R2 = 0.08, P = 0.14), Simpson‘s diversity (R2 = 0.08, P = 0.13), or Simpson‘s evenness (R2 = 0.01, P = 0.58) of macroinvertebrates on the rocks. Thus, while Dreissena had highly significant effects on these community response variables on experimental substrata lacking Cladophora (Fig. 5.1) no effect was apparent on surrounding natural substrata that had abundant Cladophora cover.

5.5.4 Alternative explanations for reduced Dreissena effects at Pointe-du-Moulin

An alternative explanation for the results of our second and third experiments is that we lacked sufficient statistical power to detect Dreissena effects. Another possibility is that the time allowed for macroinvertebrates to colonize the experimental substrata was insufficient, preventing them from reaching equilibrium densities. However, the sample size and experimental duration in Experiments 2 and 3 equaled or surpassed that used in previously published experiments, as well as in our first experiment, which have consistently found significant effects of Dreissena shells on macroinvertebrate communities (Table 5.7). These explanations are therefore unlikely to account for our findings. Similarly, time since invasion, which could explain variation across sites if there are time lags in the impacts of Dreissena (Strayer et al. 2006), is not a satisfactory explanation. Significant Dreissena effects were detected in Experiment 1, which

146 Table 5.7. Comparison with other experiments that examined the effect of Dreissena shells on benthic macroinvertebrate communities. Results are for control (C) and Dreissena shell (S) treatments, the latter represented by the full shell treatment in this study (Experiments 1, 2 and 3). Taxonomic richness was standardized to the lowest level of identification common to all studies for each taxon.

Ricciardi Stewart Horvath Mörtl & This study et al. (1997) et al. (1998) et al. (1999) Rothhaupt Expt 1 Expt 2 Expt 3 (2003)

Study characteristics No. replicates (n) 10 5 7 9 10 20 10 Duration (days) 63 37 28 27 35 70 53 Years post-invasion 4 8 3 33 12 13 15 Results Total macroinvertebrate abundance (no. m-2) C = 2,152 C = 7,700 C = 2,967 C = 7,796 C = 284 C = 3,401 C = 1,738 S = 5,930 S = 33,212 S = 7,473 S = 15,133 S = 2,032 S = 3,811 S = 2,732 Taxonomic richness (standardized) C = 20 C = 11 C = 19 C ≥ 7 C = 13 C = 29 C = 22 S = 22 S = 13 S = 24 S ≥ 7 S = 13 S = 28 S = 25

147 was conducted 1-2 years earlier at a nearby site in the same system. Finally, some other aspect of the system besides Cladophora may have differed between the two sites. Anthropogenic disturbance differed greatly between Lachine and Pointe- du-Moulin, and may have influenced our results; in particular, this may have contributed to the relatively low abundance and diversity of macroinvertebrates in Experiment 1 (Table 5.7). However, the two experiments with anomalous results were conducted at a site where anthropogenic disturbance was limited. Here, macroinvertebrate densities on control substrata were comparable to those reported for other experiments conducted at lotic sites (Ricciardi et al. 1997; Horvath et al. 1999), and taxonomic richness equaled or surpasses that reported in other experimental studies (Table 5.7). Furthermore, the published studies summarized here were conducted in a diverse assortment of aquatic habitats, with different substrata, current velocities and fauna.

In conclusion, the results from our first experiment suggest that a patchy distribution of dreissenid mussels may be an important driver of benthic diversity at small spatial scales. If we had interpreted the results of Experiments 2 and 3 in the absence of other studies of the effects of Dreissena on benthic macroinvertebrates, we would have erroneously concluded that it has no effect. However, the results of many other studies, including some on the same system, corroborate a significant effect of Dreissena. This highlights the context- dependent nature of invasion impacts. Most studies of invasion examine the effects of an introduced species in isolation. As ecosystems become increasingly invaded, the co-occurrence of introduced engineers is likely to become more common, and there is a need for studies investigating their interactive effects (Jones et al. 1997; Altieri et al. 2007; Ferraro & Cole 2007). The introduced ecosystem engineer, Dreissena, commonly occurs with native macroalgae such as Cladophora, and both are habitat modifiers that often comprise an interacting system (Fig. 5.4). The potential alteration of the effect of Dreissena by Cladophora illustrates the complexity of invasion impacts on biotic communities.

148 Our results demonstrate that the interactions of habitat-modifying species can complicate efforts to predict the community-level effects of an exotic species.

5.6 Acknowledgements

The authors thank M. Palmer, R. Kipp, Å. Kestrup, L. Jones, A. Jokela, J. Snider, R. Tiller, T. Replansky, H. Seo, A. Sowmithran and N. Bayani for their help in the field and in the lab, and two anonymous reviews for providing helpful comments. Funding for this research was provided by NSERC Canada, and by a Dr. and Mrs. Milton Leong Graduate Student Award and Vineberg Family Fellowship awarded to J. M. Ward.

5.7 References cited

Altieri A.H., Silliman B.R. & Bertness M.D. (2007). Hierarchical organization via a facilitation cascade in intertidal cordgrass bed communities. The American Naturalist, 169, 195-206. Anderson M.J., Gorley R.N. & Clarke K.R. (2008). PERMANOVA+ for PRIMER: Guide to Software and Statistical Methods. Plymouth, UK: PRIMER-E. Badano E.I., Jones C.G., Cavieres L.A. & Wright J.P. (2006). Assessing impacts of ecosystem engineers on community organization: a general approach illustrated by effects of a high-Andean cushion plant. Oikos, 115, 369-385. Barton D.R. & Hynes H.B.N. (1978). Wave-zone macrobenthos of the exposed Canadian shores of the St. Lawrence Great Lakes. Journal of Great Lakes Research, 4, 27-45. Clarke K.R. & Gorley R.N. (2006). PRIMER v6: User Manual/Tutorial. Plymouth, UK: PRIMER-E. Dodds W.K. (1990). Community interactions between the filamentous alga Cladophora glomerata (L.) Kuetzing, its epiphytes, and epiphyte grazers. Oecologia, 85, 572-580.

149 Dodds W.K. & Gudder D.A. (1992). The ecology of Cladophora. Journal of Phycology, 28, 415-427. Dubois S., Commito J.A., Olivier F. & Retière C. (2006). Effects of epibionts on Sabellaria alveolata (L.) biogenic reefs and their associated fauna in the Bay of Mont Saint-Michel. Estuarine, Coastal and Shelf Science, 68, 635-646. Dudley T.L., Cooper S.D. & Hemphill N. (1986). Effects of macroalgae on a stream invertebrate community. Journal of the North American Benthological Society, 5, 93-106. Dudley T.L. & D'Antonio C.M. (1991). The effects of substrate texture, grazing, and disturbance on macroalgal establishment in streams. Ecology, 72, 297-309. Ellsworth S.D. (2000). Influence of substrate size, Cladophora, and caddisfly pupal cases on colonization of macroinvertebrates in Sagehen Creek, California. Western North American Naturalist, 60, 311-319. Feminella J.W. & Resh V.H. (1991). Herbivorous caddisflies, macroalgae, and epilithic microalgae: dynamic interactions in a stream grazing system. Oecologia, 87, 247-256. Ferraro S.P. & Cole F.A. (2007). Benthic macrofauna–habitat associations in Willapa Bay, Washington, USA. Estuarine, Coastal and Shelf Science, 71, 491-507. Griffiths C.L., Hockey P.A.R., Schurink C. van E. & Le Roux P.J. (1992). Marine invasive aliens on South African shores: implications for community structure and trophic functioning. South African Journal of Marine Science, 12, 713- 722. Gutiérrez J.L., Jones C.G., Strayer D.L. & Iribarne O.O. (2003). Mollusks as ecosystem engineers: the role of shell production in aquatic habitats. Oikos, 101, 79-90. Harrison, S.S.C. & Hildrew, A.G. (2001). Epilithic communities and habitat heterogeneity in a lake littoral. Journal of Animal Ecology, 70, 692-707. Hart D.D. (1992). Community organization in streams: the importance of species interactions, physical factors, and chance. Oecologia, 91, 220-228.

150 Haynes J.M., Stewart T.W. & Cook G.E. (1999). Benthic macroinvertebrate communities in southwestern Lake Ontario following invasion of Dreissena: continuing change. Journal of Great Lakes Research, 25, 828-838. Higgins S.N., Malkin S.Y., Howell E.T., Guildford S.J., Campbell L., Hiriart- Baer V. & Hecky R.E. (2008). An ecological review of Cladophora glomerata (Chlorophyta) in the Laurentian Great Lakes. Journal of Phycology, 44, 839- 854. Holomuzki J.R., Pillsbury R.W. & Khandwala S.B. (1999). Interplay between dispersal determinants of larval hydropsychid caddisflies. Canadian Journal of Fisheries and Aquatic Sciences, 56, 2041-2050. Horvath T.G., Martin K.M. & Lamberti G.A. (1999). Effect of zebra mussels, Dreissena polymorpha, on macroinvertebrates in a lake-outlet stream. American Midland Naturalist, 142, 340-347. Hosack G.R., Dumbauld B.R., Ruesink J.L. & Armstrong D.A. (2006). Habitat associations of estuarine species: comparisons of intertidal mudflat, seagrass (Zostera marina), and oyster (Crassostrea gigas) habitats. Estuaries and Coasts, 29, 1150-1160. Jacobi C.M. (1987). The invertebrate fauna associated with intertidal beds of the brown mussel Perna perna (L.) from Santos, Brazil. Studies on Neotropical Fauna and Environment, 22, 57-72. Jones C.G., Lawton J.H. & Shachak M. (1997). Positive and negative effects of organisms as physical ecosystem engineers. Ecology, 78, 1946-1957. Klemm D.J. (1985) A Guide to the Freshwater Annelida (Polychaeta, Naidid and Tubificid Oligochaeta, and Hirudinea) of North America. Dubuque, IA: Kendall/Hunt Publishing Company. Lowe R.L. & Pillsbury R.W. (1995). Shifts in benthic algal community structure and function following the appearance of zebra mussels (Dreissena polymorpha) in Saginaw Bay, Lake Huron. Journal of Great Lakes Research, 21, 558-566.

151 Lozano S.J., Scharold J.V. & Nalepa T.F. (2001). Recent declines in benthic macroinvertebrate densities in Lake Ontario. Canadian Journal of Fisheries and Aquatic Sciences, 58, 518-529. Malkin S.Y., Guildford S.J. & Hecky R.E. (2008). Modeling the growth response of Cladophora in a Laurentian Great Lake to the exotic invader Dreissena and to lake warming. Limnology and Oceanography, 53, 1111-1124. Mayer C.M., Keats R.A., Rudstam L.G. & Mills E.L. (2002). Scale-dependent effects of zebra mussels on benthic invertebrates in a large eutrophic lake. Journal of the North American Benthological Society, 21, 616-633. Merritt R.W. & Cummins K.W. (1984). An Introduction to the Aquatic Insects of North America. (2nd ed.). Dubuque, IA: Kendall/Hunt Publishing Company. Moran, M.D. (2003). Arguments for rejecting the sequential Bonferroni in ecological studies. Oikos, 100, 403-405. Mörtl M. & Rothhaupt K.-O. (2003). Effects of adult Dreissena polymorpha on settling juveniles and associated macroinvertebrates. International Review of Hydrobiology, 88, 561-569. Nalepa T.F., Fanslow D.L., Lansing M.B. & Lang G.A. (2003). Trends in the benthic macroinvertebrate community of Saginaw Bay, Lake Huron, 1987 to 1996: responses to phosphorus abatement and the zebra mussel, Dreissena polymorpha. Journal of Great Lakes Research, 29, 14-33. Nalepa T.F., Hartson D.J., Fanslow D.L., Lang G.A. & Lozano S.J. (1998). Declines in benthic macroinvertebrate populations in southern Lake Michigan, 1980-1993. Canadian Journal of Fisheries and Aquatic Sciences, 55, 2402- 2413. Palmer M.E. & Ricciardi A. (2004). Physical factors affecting the relative abundance of native and invasive amphipods in the St. Lawrence River. Canadian Journal of Zoology, 82, 1886-1893. Pennak R.W. (1989). Fresh-water Invertebrates of the United States: Protozoa to Mollusca (3rd ed.). New York , NY: John Wiley & Sons.

152 Resh V.H. (1976). The biology and immature stages of the caddisfly genus Ceraclea in eastern North America (Trichoptera: Leptoceridae). Annals of the Entomological Society of America, 69, 1039-1061. Ricciardi A. (2003). Predicting the impacts of an introduced species from its invasion history: an empirical approach applied to zebra mussel invasions. Freshwater Biology, 48, 972-981. Ricciardi A., Neves R.J. & Rasmussen J.B. (1998). Impending extinctions of North American freshwater mussels (Unionoida) following zebra mussel (Dreissena polymorpha) invasion. Journal of Animal Ecology, 67, 613-619. Ricciardi A., Snyder F.L., Kelch D.O. & Reiswig H.M. (1995). Lethal and sublethal effects of sponge overgrowth on introduced dreissenid mussels in the Great Lakes–St. Lawrence River system. Canadian Journal of Fisheries and Aquatic Sciences, 52, 2695-2703. Ricciardi A., Whoriskey F.G. & Rasmussen J.B. (1997). The role of the zebra mussel (Dreissena polymorpha) in structuring macroinvertebrate communities on hard substrata. Canadian Journal of Fisheries and Aquatic Sciences, 54, 2596-2608. Rice, W.R. (1989). Analyzing tables of statistical tests. Evolution, 43, 223-225. Stewart T.W. & Haynes J.M. (1994). Benthic macroinvertebrate communities of southwestern Lake Ontario following invasion of Dreissena. Journal of Great Lakes Research, 20, 479-493. Stewart T.W., Miner J.G. & Lowe R.L. (1998). Macroinvertebrate communities on hard substrates in western Lake Erie: structuring effects of Dreissena. Journal of Great Lakes Research, 24, 868-879. Stewart T.W., Shumaker T.L. & Radzio T.A. (2003). Linear and nonlinear effects of habitat structure on composition and abundance in the macroinvertebrate community of a large river. American Midland Naturalist, 149, 293-305. Strayer D.L., Eviner V.T., Jeschke J.M. & Pace M.L. (2006). Understanding the long-term effects of species invasions. Trends in Ecology and Evolution, 21, 645-651.

153 Strayer D.L. & Smith L.C. (2000). Macroinvertebrates of a rocky shore in the freshwater tidal Hudson River. Estuaries and Coasts, 23, 359-366. Strayer D.L., Smith L.C. & Hunter D.C. (1998). Effects of the zebra mussel (Dreissena polymorpha) invasion on the macrobenthos of the freshwater tidal Hudson River. Canadian Journal of Zoology, 76, 419-425. Tanaka M.O. & Magalhães C.A. (2002). Edge effects and succession dynamics in Brachidontes mussel beds. Marine Ecology Progress Series, 237, 151-158. Thorp J.H. & Covich A.P. (1991). Ecology and Classification of North American Freshwater Invertebrates. San Diego, CA: Academic Press. Tsuchiya M. & Nishihira M. (1985). Islands of Mytilus as a habitat for small intertidal animals: effect of island size on community structure. Marine Ecology Progress Series, 25, 71-81. Van Overdijk C.D., Grigorovich I.A., Mabee T., Ray W.J., Ciborowski J.J.H. & MacIsaac H.J. (2003). Microhabitat selection by the invasive amphipod Echinogammarus ischnus and native Gammarus fasciatus in laboratory experiments and in Lake Erie. Freshwater Biology, 48, 567-578. Vanderploeg H.A., Nalepa T.F., Jude D.J., Mills E.L., Holeck K.T., Liebig J.R., Grigorovich I.A. & Ojaveer H. (2002). Dispersal and emerging ecological impacts of Ponto-Caspian species in the Laurentian Great Lakes. Canadian

Journal of Fisheries and Aquatic Sciences, 59, 1209-1228. Ward J.M. & Ricciardi A. (2007). Impacts of Dreissena invasions on benthic macroinvertebrate communities: a meta-analysis. Diversity and Distributions, 13, 155-165. Wisenden P.A. & Bailey R.C. (1995). Development of macroinvertebrate community structure associated with zebra mussel (Dreissena polymorpha) colonization of artificial substrates. Canadian Journal of Zoology, 73, 1438- 1443. Witt J.D.S., Hebert P.D.N. & Morton, W.B. (1997). Echinogammarus ischnus: another crustacean invader in the Laurentian Great Lakes basin. Canadian Journal of Fisheries and Aquatic Sciences, 54, 264–268.

154 Wright J.P., Jones C.G. & Flecker A.S. (2002). An ecosystem engineer, the beaver, increases species richness at the landscape scale. Oecologia, 132, 96- 101. Young B.L., Padilla D.K., Schneider D.W. & Hewett S.W. (1996). The importance of size-frequency relationships for predicting ecological impact of zebra mussel populations. Hydrobiologia, 332, 151–158.

155 Connecting statement 5–6

In Chapter 5, I presented evidence suggesting that the impact of Dreissena on benthic macroinvertebrate communities is reduced in the presence of another habitat-modifying species (Cladophora), which I interpret to reflect the provision of similar resources (structurally complex microhabitats) by these two ecosystem engineers. In Chapter 6, I integrate this hypothesis into a conceptual framework for the community-level impacts of nonindigenous ecosystem engineers and briefly discuss potential avenues for future research.

156 Chapter 6 – General conclusions and future research

6.1 Summary

An improved understanding of how biological invasions impact ecological communities is at once academically informative and necessary in light of ongoing global change (Chapin et al. 2000; Byers et al. 2002). Generalizations regarding the ecological impacts of nonindigenous species — especially those that apply broadly across taxonomic groups and ecosystems — are therefore crucial, yet there exist few quantitative tests. To this end, I employed meta-analytical techniques to model variation in the impacts of exotic ecosystem engineers using available empirical data, in order to identify general patterns and test predictions drawn from the invasion ecology and community ecology literature. In Chapters 2 and 3, I examined the reliance of ecosystem engineering impacts on a community‘s prior evolutionary experience with the engineer or with similar species (i.e. ecological naiveté). In Chapter 2, I synthesized published data involving the community-level impacts of native and exotic ecosystem engineers that modify benthic habitats via bioconstruction or bioturbation. This analysis confirmed a generally positive relationship between the effect of exotic ecosystem engineers on habitat complexity (via bioconstruction) and their impacts on community richness and abundance, as previously hypothesized in narrative reviews (Crooks 2002). It further revealed that the magnitude of these community-level impacts does not systematically differ among native and exotic ecosystem engineers, particularly bioconstructors. I tested the ecological naiveté hypothesis in Chapter 3 by determining whether the impacts of exotic ecosystem engineers on native macroinvertebrates are correlated with their phylogenetic relatedness to native species in the invaded region. My analysis revealed that these are indeed correlated, however not in the direction previously observed for predator-prey and competitive interactions (Ricciardi & Atkinson 2004; Strauss et al. 2006), and suggested that the functional distinctiveness of exotic ecosystem engineers generally promotes facilitation.

157 Chapters 4 and 5 focused on the community-level impacts of one of the most notorious exotic ecosystem engineers in aquatic communities, dreissenid mussels (Strayer 2009). In Chapter 4, I quantitatively tested hypotheses from invasion ecology (e.g. the impact of an exotic species is a function of its density; Parker et al. 1999; Ricciardi 2003) and community ecology (e.g. the impact of an ecosystem engineer depends on the ambient availability of the resources it controls; Gutiérrez et al. 2003) with respect to Dreissena impacts on macroinvertebrate communities across multiple invaded sites. In Chapter 5, I described a series of field experiments designed to examine the effect of Dreissena patch topography on benthic diversity. The results of these experiments suggest that the impact of Dreissena is modified in the presence of another dominant ecosystem engineer, Cladophora, and that Dreissena patch topography may be an important determinant of species richness at small spatial scales. Because invasion impact theory has, thus far, largely been developed by studying competitive or predator-prey interactions, the results presented here provide some insight into how our theoretical understanding of impact might be fine-tuned with regard to impacts mediated by physical habitat modification or ecosystem engineering.

6.2 Conceptual framework

This work contributes to our understanding of the conditionality or context- dependence of invasion impacts; that is, how the direction or magnitude of impact depends on the characteristics of the exotic species and the biotic and abiotic properties of invaded systems (Fig. 6.1). Existing frameworks for understanding the ecological impact of invasions correlate impact to the abundance and per capita effect of exotic species (Parker et al. 1999; Ricciardi 2003), each of which can vary with physical environmental conditions and the composition of the invaded community. An invader‘s abundance, and therefore impact, can be moderated by the physical environment either directly, because the match between local conditions and the invader‘s physiological tolerances determine its

158

Figure 6.1. Hypothetical framework for the community-level impacts of exotic ecosystem engineers. Solid arrows indicate general relationships. Dashed arrows indicate relationships specific to ecosystem engineers. The impact of an exotic species on its invaded community (5) is a function of its abundance (4) and its per capita effect (9). Local environmental conditions can regulate the abundance of an exotic species directly (1) or indirectly by structuring the local community (2), which can in turn constrain (via biotic resistance) or enhance (e.g. invasional meltdown) invader abundance (3). Likewise, the physical environment can moderate the per capita effect of an exotic species (e.g. its competitive ability; 8). The per capita effect of an exotic species is also expected to depend on its functional distinctiveness (e.g. its traits relative to those of species in the recipient community; 6, 7). Ecosystem engineering by exotic (10) and native species (11) results in feedback loops wherein species modify the physical environment, which in turn affects community composition and impact. The impact of an exotic ecosystem engineer will also depend on the background environmental conditions, which determine the extent of its habitat modification (e.g. change in resource availability) relative to the unmodified state (12). See text for examples.

159 local abundance (e.g. calcium or salinity; Jokela & Ricciardi 2008; Alcaraz & García-Berthou 2007), or indirectly, because the environment structures the local community. Invader abundance might be constrained by natural enemies in the invaded community (i.e. via biotic resistance; Elton 1958), or enhanced by facilitator species, be they native (e.g. pollinators; Richardson et al. 2000) or exotic (i.e. invasional meltdown; Simberloff & Von Holle 1999). In addition, an invader‘s susceptibility to native natural enemies, and hence the strength of biotic resistance, can vary with local environmental conditions (e.g. sediment composition; Byers 2002). Likewise, the per capita effect of an invader can depend on local environmental conditions; for example, resource availability can influence an invader‘s relative competitive ability or predation rate (Kestrup & Ricciardi 2009), and habitat complexity can affect the vulnerability of native species to an introduced predator (Stuart-Smith et al. 2007). An invader‘s per capita effect is also expected to depend on its functional traits in relation to the traits of species in the invaded community, which will determine its functional distinctiveness. This dependence may result from either ecological (Vitousek 1990; Dukes & Mooney 2004) or evolutionary processes (i.e. ecological naiveté; Callaway & Ridenour 2004; Cox & Lima 2006). While the aspects of context-dependence described above are common to introduced species in general, additional complexities arise upon incorporation of ecosystem engineering (Fig 6.1). Exotic ecosystem engineers modify local environmental conditions, thereby creating feedback loops which may enhance their own abundance and impact (Cuddington & Hastings 2004; Gonzalez et al. 2008). For example, invasions by fire-promoting grasses can alter the frequency and intensity of fires, which in turn promotes their dominance (Brooks et al. 2004). In addition, because the impact of physical ecosystem engineering arises via the engineer‘s control of abiotic factors, the functional distinctiveness of an exotic ecosystem engineer will also depend on the physical (abiotic) properties of the invaded system; that is, the magnitude of engineering effects will be greater when the conditions produced by a exotic ecosystem engineer (i.e. in the modified

160 habitat) are distinct from those produced by other biotic or abiotic processes operating in the recipient system (Gutiérrez et al. 2003). Here, I have shown that the community-level effects of exotic ecosystem engineers are context-dependent: the direction and magnitude of their impacts depend on the background availability of the habitats or resources provided by the engineer. This background availability is determined by a combination of biotic and abiotic factors; in this case, whether another ecosystem engineer that fills a similar engineering function is present in the invaded region (Chapter 3) or in uninvaded patches of the invaded site (Chapter 5), or the degree to which the engineer controls the availability of a resource relative to physical (abiotic) processes (e.g. provision of colonizable hard or soft substrate habitats by Dreissena in rocky and sedimentary environments; Chapter 4). My results further suggest that the functional distinctiveness of an ecosystem engineer, defined relative to other biotic and abiotic factors that control the physical structure of the environment, is more important than its native or exotic origin in determining the magnitude of its impacts on other species. Evidence for this argument includes (i) no difference in the community-level effects of native and exotic ecosystem engineers (Chapters 1 and 5), and (ii) exotic ecosystem engineers have larger effects when they are phylogenetically distinct (Chapter 3) or when the resources they provide are otherwise limited in the physical environment (Chapter 4). My results further suggest that the loss of functionally (or phylogenetically) distinct native ecosystem engineers, particularly bioconstructors, will be especially damaging for local biodiversity. Jones et al. (1997) questioned whether the effects of ecosystem engineers on local diversity are predictable. The findings presented in this thesis contribute to an emerging framework that relates the impact of an ecosystem engineer on local diversity to its effect on factors known to correlate with species richness. These factors include habitat complexity (Crooks 2002), sediment stability (Reise 2002), resource availability (Gutiérrez et al. 2003), primary productivity (Wright & Jones 2004), and the intensity or frequency of disturbance (Bruno et al. 2003). My results further suggest that an ecosystem engineer‘s effect on these factors is

161 more important than its native or exotic origin in determining the magnitude of a community‘s response to habitat modification by that organism.

6.3 Future research

This body of work has identified several interesting research questions that warrant further attention, including:

1. How do invasions by exotic ecosystem engineers alter ecosystem function? What is the relationship between their community-level and ecosystem-level impacts?

2. Why do exotic bioconstructors have more facilitative effects on native species when they are phylogenetically distinct in the invaded region? Does this pattern extend to other forms of ecosystem engineering (e.g. bioturbation)?

3. Do the patterns of impact described here for Dreissena (e.g. variation in impact with substrate particle size) apply to other introduced mussels?

4. What is the role that macroalgae play in modifying the impact of autogenic engineers like Dreissena? With respect to the Cladophora –Dreissena tandem (Chapter 5), this might be investigated by manipulating the presence and density of Cladophora, or by using Cladophora mimics.

In general, future application of meta-analysis to the study of biological invasions could allow quantitative comparisons of impact magnitude among species, ecosystems and types of biotic interactions. This synthetic approach has the potential to improve our theoretical understanding of invasion impacts (e.g. what is the proportion of exotic species that have large or extreme impacts? What is the distribution of impact magnitude?) and enhance our power to predict them.

162 6.4 References cited

Alcaraz, C. & García-Berthou, E. (2007). Life history variation of invasive mosquitofish (Gambusia holbrooki) along a salinity gradient. Biological Conservation, 139, 83-92. Brooks, M.L., D‘Antonio, C.M., Richardson, D.M., Grace, J.B., Keeley, J.E., DiTomaso, J.M., Hobbs, R.J., Pellant, M. & Pyke, D. (2004). Effects of invasive alien plants on fire regimes. BioScience, 54, 677-688. Bruno, J.F., Stachowicz, J.J. & Bertness, M.D. (2003). Inclusion of facilitation into ecological theory. Trends in Ecology and Evolution, 18, 119-125. Byers, J.E. (2002). Physical habitat attribute mediates biotic resistance to non- indigenous species invasion. Oecologia, 130, 146-156. Byers, J.E., Reichard, S., Randall, J.M., Parker, I.M., Smith, C.S., Lonsdale, W.M., Atkinson, I.A.E., Seastedt, T.R., Williamson, M., Chornesky, E. & Hayes, D. (2002). Directing research to reduce the impacts of nonindigenous species. Conservation Biology, 16, 630-640. Callaway, R.M. & Ridenour, W.M. (2004). Novel weapons: invasive success and the evolution of increased competitive ability. Frontiers in Ecology and the Environment, 2, 436-443. Chapin, F.S. III, Zavaleta, E.S., Eviner, V.T., Naylor, R.L., Vitousek, P.M., Reynolds, H.L., Hooper, D.U., Lavorel, S., Sala, O.E., Hobbie, S.E., Mack, M.C. & Díaz, S. (2000). Consequences of changing biodiversity. Nature, 405, 234-242. Cox, J.G. & Lima, S.L. (2006). Naiveté and an aquatic-terrestrial dichotomy in the effects of introduced predators. Trends in Ecology and Evolution, 21, 674- 680. Crooks, J.A. (2002). Characterizing ecosystem-level consequences of biological invasions: the role of ecosystem engineers. Oikos, 97, 153-166. Cuddington, K. & Hastings, A. (2004). Invasive engineers. Ecological Modelling, 178, 335-347.

163 Dukes, J.S. & Mooney, H.A. (2004). Disruption of ecosystem processes in western North America by invasive species. Revista Chilena de Historia Natural, 77, 411-437. Elton, C.S. (1958). The Ecology of Invasions by Animals and Plants. London, UK: Methuen. Gonzalez, A., Lambert, A. & Ricciardi, A. (2008). When does ecosystem engineering cause invasion and species replacement? Oikos, 117, 1247-1257. Gutiérrez J.L., Jones C.G., Strayer D.L. & Iribarne O.O. (2003). Mollusks as ecosystem engineers: the role of shell production in aquatic habitats. Oikos, 101, 79-90. Jones, C.G., Lawton, J.H. & Shachak, M. (1997). Positive and negative effects of organisms as physical ecosystem engineers. Ecology, 78, 1946-1957. Jokela, A. & Ricciardi, A. (2008). Predicting zebra mussel fouling on native mussels from physicochemical variables. Freshwater Biology, 53, 1845-1856. Kestrup, Å.M. & Ricciardi, A. (2009). Environmental heterogeneity limits the local dominance of an invasive freshwater crustacean. Biological Invasions, 11, 2095-2105. Parker, I.M., Simberloff, D., Lonsdale, W.M., Goodell, K., Wonham, M., Kareiva, P.M., Williamson, M.H., Von Holle, B., Moyle, P.B., Byers, J.E. & Goldwasser L. (1999). Impact: toward a framework for understanding the ecological effects of invaders. Biological Invasions, 1, 3-19. Reise, K. (2002). Sediment mediated species interactions in coastal waters. Journal of Sea Research, 48, 127-141. Ricciardi, A. (2003). Predicting the impacts of an introduced species from its invasion history: an empirical approach applied to zebra mussel invasions. Freshwater Biology, 48, 972-981. Ricciardi, A. & Atkinson, S.K. (2004). Distinctiveness magnifies the impact of biological invaders in aquatic ecosystems. Ecology Letters, 7, 781-784. Richardson, D.M., Allsopp, N., D‘Antonio, C.M., Milton, S.J. & Rejmánek, M. (2000). Plant invasions – the role of mutualisms. Biological Reviews of the Cambridge Philosophical Society, 75, 65-93.

164 Simberloff, D. & Von Holle, B. (1999). Positive interactions of nonindigenous species: invasional meltdown? Biological Invasions, 1, 21-32. Strauss, S.Y., Webb, C.O. & Salamin, N. (2006). Exotic taxa less related to native species are more invasive. Proceedings of the National Academy of Sciences of the United States of America, 103, 5841-5845. Strayer, D.L. (2009). Twenty years of zebra mussels: lessons from the mollusk that made headlines. Frontiers in Ecology and the Environment, 7, 135–141. Stuart-Smith, R.D., Stuart-Smith, J.F., White, R.W.G. & Barmuta, L.A. (2007). The impact of an introduced predator on a threatened galaxiid fish is reduced by the availability of complex habitats. Freshwater Biology, 52, 1555-1563. Vitousek, P.M. (1990). Biological invasions and ecosystem processes: towards an integration of population biology and ecosystem studies. Oikos, 57, 7-13. Wright, J.P. & Jones, C.G. (2004). Predicting effects of ecosystem engineers on patch-scale species richness from primary productivity. Ecology, 85, 2071-2081

165

Appendices

166 Appendix A. Comment on “Opposing Effects of Native and Exotic

Herbivores on Plant Invasions”

A version of this manuscript appears as: Ricciardi, A. & Ward, J.M. (2006). Comment on ―Opposing Effects of Native and Exotic Herbivores on Plant Invasions‖. Science, 313, 298.

Reprinted with permission from the American Association for the Advancement of Science (AAAS). http://www.sciencemag.org/

Abstract Parker et al. (Reports, 10 March 2006, p. 1459) showed that native herbivores suppress exotic plants more than native plants. Further analysis reveals that the effect of native herbivores is reduced on exotic plant species that are closely related to native species in the invaded region. Exotic plants may share traits with native congeners that confer similar resistance to resident herbivores. ______

Through a meta-analysis of published experimental data, Parker et al. (1) demonstrated that native herbivores typically reduce the survival and abundance of introduced exotic plants, but tend to have weak positive effects on co-occurring native plants. It was concluded that plants are particularly susceptible to generalist herbivores that they have not been selected to resist. Here, we expand the analysis of Parker et al. by considering the phylogenetic relationship of the exotic and native plants. Given that genetic divergence decreases as taxonomic relatedness increases, evolutionary logic suggests that species of the same genus are more likely to be functionally similar (2, 3). Indeed, congeneric plants do tend to have similar herbivore defenses (4, 5). Therefore, exotic plants that share a genus with native plants in the invaded range might be similarly susceptible to

167 native herbivores, while those that belong to a novel genus would likely have differential susceptibility – potentially affecting their ability to persist and spread in their new environment. We tested this hypothesis using the dataset of native herbivore impacts on exotic plants, compiled by Parker et al. (1). For each experiment in the dataset, we determined if a native plant of the same genus as the exotic plant was historically present in the region in which the experiment was performed, by consulting native species lists provided by the Flora Europaea database (http://rbg-web2.rbge.org.uk/FE/fe.html) and the U.S. Native Plant Information Network of the Lady Bird Johnson Wildflower Center (http://www.wildflower2.org/NPIN/plants/plant.html). We then performed an unweighted, fixed-effects model meta-analysis using Meta-Win 2.1 (6), following the same procedure as Parker et al. We generated 95% bias-corrected bootstrap confidence intervals and tested for significant differences between herbivore effects on native and exotic genera using a randomized resampling technique for meta-analysis with 9999 iterations (7). Our analysis revealed that the negative effects of native herbivores are more pronounced on introduced plants belonging to exotic genera (Fig. 1). The mean loge-transformed effect on exotic genera exceeded that on native genera by a factor of 5.8. We obtained similar results using a mixed-effects model. More than 83% of the dataset we used were measurements of plant survival. Because Parker et al. found that vertebrate herbivores exerted a greater negative effect than invertebrate herbivores on exotic plant survival, we considered that our results might be biased by differences in the proportions of vertebrate herbivores in studies involving native and exotic plant genera, respectively. However, when we ran replicate meta-analyses on the exclusive effects of vertebrate and invertebrate herbivores, respectively, the same result was obtained as for the combined dataset. This finding supports the view that variation in an invader's success and impact is, at least in part, explained by the invaded community's prior experience with functionally similar species (8). Factors proposed to explain the variation in the

168 1.0 P=0.0001

0.5

0.0

39 -0.5

-1.0

on exotic onplants exotic

-1.5 52

Native herbivore effect (ln +H/-H) herbivore effect Native -2.0 Native Exotic

Plant genus

Fig. A.1. Effects of native herbivores on exotic plant abundance and survival in 18 experimental studies reported by Parker et al. (1). Effects are weakest on plants belonging to native genera, i.e. those that share a genus with a native species in the invaded region. Symbols show means ± 95% confidence intervals, which were calculated using a bias-corrected bootstrapping technique with 9999 randomized iterations (7). Numbers to the right of the symbols are the number of experiments contributing to the mean. P value indicates difference in effects on exotic (unshared) versus native (shared) genera.

169 success of exotic species include the number of introduced propagules (e.g. seeds, eggs, individuals), reproductive capacity, environmental tolerance limits, prior disturbance in the recipient community, and release from natural enemies (9–11). To date, few studies have examined invasion success as a function of the phylogenetic relationship between the introduced species and members of the recipient assemblage. Darwin hypothesized that introduced plants are more successful in colonizing areas that do not contain native species of the same genus because they would compete with their close relatives and encounter herbivores that could more easily exploit them (12, 13). Our study rejects this hypothesis, and suggests that exotic plants are pre-adapted to conditions of herbivory experienced by congeneric native species. This provides further support for the view that generalist herbivores should have greater effects on exotic species with which they have not shared any evolutionary experience. Our results, together with those of Parker et al., demonstrate the inadequacy of 'enemy release' models that simply relate the success of an invader to the absence of its natural predators in the invaded region (11).

References and notes

1. Parker, J.D., Burkepile, D.E. & Hay, M.E. (2006). Opposing effects of native and exotic herbivores on plant invasions. Science, 311, 1459-1461. 2. Webb, C.O., Ackerly, D.D., McPeek, M.A. & Donahue, M.J. (2002). Phylogenies and community ecology. Annual Review of Ecology and Systematics, 33, 475-505. 3. Thorpe, J.P. (1982). The molecular clock hypothesis: biochemical evolution, genetic differentiation and systematics. Annual Review of Ecology and Systematics, 13, 139-168. 4. Berenbaum, M. (1981). Patterns of furanocoumarin distribution and insect herbivory in the Umbelliferae: plant chemistry and community structure. Ecology, 62, 1254-1266. 5. Harborne, J.B. (1993). Introduction to Ecological Biochemistry. London: Academic Press.

170 6. Rosenberg, M.S., Adams, D.C. & Gurevitch, J. (2000). MetaWin: Statistical Software for Meta-Analysis. Version 2. Sunderland, MA: Sinauer Associates. 7. Adams, D.C., Gurevitch, J. & Rosenberg, M.S. (1997). Resampling tests for meta-analysis of ecological data. Ecology, 78, 1277-1283. 8. Ricciardi, A & Atkinson, S.K. (2004). Distinctiveness magnifies the impact of biological invaders in aquatic ecosystems. Ecology Letters, 7, 781-784. 9. Kolar, C.S. & Lodge, D.M. (2001). Progress in invasion biology: predicting invaders. Trends in Ecology and Evolution, 16, 199-204. 10. Levine, J.M., Adler, P.B. & Yelenik, S.G. (2004). A meta-analysis of biotic resistente to exotic plant invasions. Ecology Letters, 7, 975-989. 11. Colautti, R.I., Ricciardi, A., Grigorovich, I.A. & MacIsaac, H.J. (2004). Is invasion success explained by the enemy release hypothesis? Ecology Letters, 7, 721-733. 12. Darwin, C. (1859). On the Origins of Species by Means of Natural Selection. London, UK: John Murray. 13. Rejmánek, M. (1996). A theory of seed plant invasiveness: the first sketch. Biological Conservation, 78, 171-181. 14. Supported by the Natural Sciences and Engineering Research Council of Canada.

171 Appendix B. Data sources for meta-analysis of the effects of native and exotic ecosystem engineers (Chapter 2).

Table B.1. List of native and exotic ecosystem engineers included in the meta-analysis and data sources for community response variables: total taxonomic richness (S), taxa density (TD), numerical density (ND), biomass density (BD), and diversity (D) and evenness (E) indices.

Ecosystem engineer species Location Response variables Data sources

I. BIOCONSTRUCTORS Animalia Ascidiacea Exotic Botrylloides diegensis Long Island Sound, USA ND Osman & Whitlatch (1995) Botryllus schlosseri Long Island Sound, USA ND Osman & Whitlatch (1995) Ciona intestinalis San Francisco Bay, USA S, TD Blum et al. (2007) Didemnum vexillum Long Island Sound, USA TD, ND, D, E Mercer et al. (2009) Pyura praeputialis Antofagasta Bay, Chile S Castilla et al. (2004) Styela clava Port Phillip Bay, Australia TD, ND, D, E Ross et al. (2007)

Native Molgula pedunculata King George Island, Antarctica S, ND Sahade et al. (1998) Pyura stolonifera Cape Banks, Australia TD Monteiro et al. (2002) Styela plicata Gulf of Mexico, USA ND Young (1989)

Mollusca Exotic Crassostrea gigas Arcachon Bay, France ND, BD Castel et al. (1989) Bahia Anegada, Argentina S, ND Escapa et al. (2004) Wadden Sea, Germany S, TD, ND Kochmann et al. (2008)

172 S, ND Markert et al. (2010) Willapa Bay, USA S, TD, ND, BD Ferraro & Cole (2007) ND Hosack et al. (2006) Dreissena spp. Christiana Creek, USA S, TD, ND, D, E Horvath et al. (1999) Lake Champlain, USA TD, ND Beekey et al. (2004) Lake Constance, Germany ND Mörtl & Rothhaupt (2003) Lake Erie, USA S, TD, ND Bially & MacIssac (2000) ND Botts et al. (1996) ND, BD Dermott et al. (1993) S, TD, ND, BD, D, E Stewart et al. (1998) Lake Michigan, USA ND, BD Kuhns & Berg (2003) S Reed et al. (2004) Lake St. Claire, USA BD Thayer et al. (1997) Lake St. Louis, Canada S, TD, ND, D, E Ricciardi et al. (1997) Mikolajskie Lake, Poland ND Dusoge (1966) Oneida Lake, USA ND Mayer et al. (2002) Limnoperna fortunei Parana River delta, Argentina TD, ND, BD, D, E Sardiña et al. (2008) S, ND, BD Sylvester et al. (2007) Maoricolpus roseus Twofolds Bay, Australia ND Nicastro et al. (2009) Musculista senhousia Adriatic Sea, Italy S, TD, ND Mistri (2002) S, TD, ND Munari (2008) Mission Bay, USA TD, ND Crooks (1998) S, TD, ND Crooks & Khim (1999) Tamaki Estuary, New Zealand S, TD, ND Creese et al. (1997) Tyrrhenian Sea, Sardinia S, TD, ND Munari (2008) Mytilus galloprovincialis Langebaan Lagoon, South Africa S, ND, BD Robinson & Griffiths (2002) S, TD, ND, BD, D, E Robinson et al. (2007) Sydney Harbour, Australia S, TD, ND Chapman et al. (2005)

173 Native Atrina zealandica Jamieson Bay, New Zealand S, TD, ND Norkko et al. (2001) TD, ND Norkko et al. (2006) Mahurangi Harbour, New Zealand TD, ND Cummings et al. (2001) Martins Bay, New Zealand TD, ND, D, E Cummings et al. (1998) TD, ND Cummings et al. (2001) S, TD, ND Norkko et al. (2001) TD, ND Norkko et al. (2006) Te Kapa Inlet, New Zealand TD, ND, D, E Cummings et al. (1998) TD, ND Cummings et al. (2001) S, TD, ND Norkko et al. (2001) TD, ND Norkko et al. (2006) Austrovenus stutchburyi Otago Harbour, New Zealand S, TD, ND, D, E Mouritsen & Poulin (2005) Crassostrea virginica Back Sound, North Carolina, USA ND Grabowski et al. (2005) Long Island Sound, USA ND Osman & Whitlatch (1995b) Modiolus metcalfi Gulf of Thailand D, E Tsuchiya (2002) Modiolus modiolus Isles of Shoals, USA S, ND, D, E Witman (1985) Musculista senhousia Yellow Sea, South Korea S, TD, D, E Buschbaum et al. (2009) Mytilus chilensis Golfo de Reloncaví, Chile S, TD, ND, D, E Duarte et al. (2006) Mytilus edulis Baltic Sea, Germany TD, D Dürr & Wahl (2004) English Channel, France TD Davoult et al. (1998) Maine, USA S, TD, ND Commito et al. (2005) North Sea, Germany S, TD, ND Kochmann et al. (2008) Skagerrak, Sweden S, TD, ND, BD Norling & Kautsky (2007) Wadden Sea, Germany S, TD, D, E Buschbaum et al. (2009) S, TD, ND, D, E Dittmann (1990) S, TD, ND, D, E Günther (1996) S, TD, ND Kröncke (1996) S, ND Markert et al. (2010)

174 Ythan Estuary, Scotland S, TD, ND Ragnarsson & Raffaelli (1999) Ostreola conchaphila Tomales Bay, USA TD, D, E Kimbro & Grosholz (2006) Perumytilus purpuratus Bahai Totoralillo, Chile TD, ND Valdivia & Thiel (2006) Semimytilus algosus Ancon Bay, Peru S, ND Tokeshi & Romero (1995) Tagelus plebeius Mar Chiquita lagoon, Argentina S, ND Gutiérrez & Iribarne (1999) Unionidae Cambridgeshire & Somerset, UK TD Aldridge et al. (2007) Ouachita Highlands, USA ND Vaughn & Spooner (2006) Xenostrobus inconstans Coffin Bay, Australia S, TD, D, E Buschbaum et al. (2009)

Polychaeta (tube-building) Exotic Ficopomatus enigmaticus Mar Chiquita, Argentina S Schwindt & Iribarne (2000) Sabella spallanzanii Port Phillip Bay, Australia ND Holloway & Keough (2002a) ND O'Brien et al. (2006) TD, ND, D, E Ross et al. (2007) Outer Adelaide Harbour, Australia ND Holloway & Keough (2002a) ND Holloway & Keough (2002b)

Native Chaetopterus variopedatus Chesapeake Bay, USA TD, ND, D, E Schaffner (1990) Diopatra cuprea Indian River lagoon, Florida, USA ND Ban & Nelson (1987) South Carolina, USA ND Luckenbach (1984) Lanice conchilega Gower Peninsula, Wales S, TD, ND Callaway (2006) North Sea, Belgium TD, ND Rabaut et al. (2007) Wadden Sea, Germany S, TD, ND, D, E Zühlke (2001) Macroclymene zonalis Chesapeake Bay, USA TD, ND, D, E Schaffner (1990) Pygospio elegans Firth of Forth, North Sea, Scotland S, ND Bolam & Fernandes (2003)

Plantae Angiospermae

175 Exotic Glyceria maxima West Gippsland, Victoria, Australia TD, ND Clarke et al. (2004) Hydrilla verticilatta Chesapeake Bay, USA S, ND Posey et al. (1993) Starkville, Mississippi, USA S, TD, ND, BD Theel et al. (2008) Spartina alterniflora Jiangsu coast, China S, ND, BD, D, E Zhou et al. (2009) Willapa Bay, USA S, TD, ND, BD Ferraro & Cole (2007) S. alterniflora x foliosa San Francisco Bay, USA S, ND, BD, D Brusati & Grosholz (2006) ND, BD Levin et al. (2006) S, TD, ND Neira et al. (2005) S, TD, ND, D, E Neira et al. (2007) Spartina anglica Arcachon Bay, France S, ND Cottet et al. (2007) Dublin Bay, Ireland S, ND McCorry & Otte (2001) Tasmania, Australia S, ND Hedge & Kriwoken (2000) Westerschelde estuary, the Netherlands S, ND, BD, D, E Bouma et al. (2009) Trapa natans Hudson River, USA S, ND Strayer et al. (2003) Urochloa mutica Kakadu National Park, Australia S, TD, ND Douglas & O'Connor (2003) Zostera japonica Coos Bay, USA S, ND Posey (1988)

Native Cymodocea nodosa Adriatic Sea, Italy S, ND, BD, D Pranovi et al. (2000) Enhalus acoroides Unguja Island, Tanzania S, ND, BD Eklöf et al. (2005) Halophila ovalis St. George Basin, Australia S, TD, ND McKinnon et al. (2009) Heterozostera tasmanica Western Port, Victoria, Australia S, TD, ND Edgar et al. (1994) Hymenachne acutigluma Kakadu National Park, Australia S, TD, ND Douglas & O'Connor (2003) Oryza meridionalis Kakadu National Park, Australia S, TD, ND Douglas & O'Connor (2003) Ruppia maritima Åland Islands, Baltic Sea, Finland S, ND Boström & Bonsdorff (2000) Cape Cod, USA S, ND, BD, D, E Heck et al. (1995) Salicornia virginica Mission Bay, USA TD, ND, BD, D, E Whitcraft & Levin (2007) Salicornia spp. Dublin Bay, Ireland S, ND McCorry & Otte (2001) Spartina alterniflora Newport River estuary, USA TD, ND Levin et al. (1996)

176 Paranaguá Bay, Brazil S, TD, ND, D, E Lana & Guiss (1991) ND Pagliosa & Lana (2005) Spartina foliosa Bolinas Lagoon, USA S, ND, BD, D Brusati & Grosholz (2006) Drakes Estero, USA S, ND, BD, D Brusati & Grosholz (2006) Mission Bay, USA TD, ND Levin et al. (1998) TD, ND, BD, D, E Whitcraft & Levin (2007) San Francisco Bay, USA S, ND, BD, D Brusati & Grosholz (2006) San Pablo Bay, USA S, ND, BD, D Brusati & Grosholz (2006) Tomales Bay, USA S, ND, BD, D Brusati & Grosholz (2006) Thalassia hemprichii Unguja Island, Tanzania S, ND, BD Eklöf et al. (2005) Vallisneria americana Hudson River, USA S, TD Strayer et al. (2003) Zostera capensis Langebaan Lagoon, South Africa TD, ND, D, E Siebert & Branch (2007) Zostera capricorni Coromandel Peninsula, Zew Zealand TD, ND, BD, D, E van Houte-Howes et al. (2004) Zostera japonica Crooked Harbour, Hong Kong S, TD, ND, D, E Lee et al. (2001) Zostera marina Åland Islands, Baltic Sea, Finland S, ND Boström & Bonsdorff (2000) Brittany, France TD, ND, BD, D, E Hily & Bouteille (1999) Cape Cod, USA S, ND, BD, E Heck et al. (1995) Devon, Britain TD, ND Turner & Kendall (1999) Damariscotta River estuary, USA S, ND, D Mattila et al. (1999) San Diego Bay, USA TD, ND, D, E Reed & Hovel (2006) Willapa Bay, USA S, TD, ND, BD Ferraro & Cole (2007) ND Hosack et al. (2006) Zostera muelleri Baker Inlet-Port River estuary, Australia ND, BD Connolly (1997) Z. muelleri & Halophila Western Port, Victoria, Australia S, ND Edgar et al. (1994) australis Zostera noltii Arcachon Bay, France ND, BD Castel et al. (1989) S, ND Cottet et al. (2007) Westerschelde estuary, the Netherlands S, ND, BD, D, E Bouma et al. (2009)

177 Zostera spp. & Adelaide, Australia ND Connolly (1994) Heterozostera spp.

Chlorophyta Exotic Caulerpa racemosa Ligurian Sea, Italy S Piazzi & Balata (2008) Caulerpa taxifolia St. George Basin, Australia S, TD, ND McKinnon et al. (2009) Codium fragile Cranberry Cove, Canada S, D Schmidt & Scheibling (2007)

Native Caulerpa prolifera Bay of Algeciras, Spain D, ND Sánchez-Moyano et al. (2001) Cladophora glomerata & Santa Barbara County, USA TD, ND Dudley et al. (1986) Nostoc sp. Ulva expansa Sonoma County, USA S, TD, ND, D, E Everett (1994)

Phaeophyta Exotic Sargassum muticum Langstone Harbour, Britain TD, ND, D, E Strong et al. (2006) San Juan Islands Marine Preserve, USA TD Britton-Simmons (2004) Strangford Lough, Ireland TD, ND, D, E Strong et al. (2006) Undaria pinnatifida Lyttelton Harbour, New Zealand TD Forrest & Taylor (2002)

Native Ecklonia radiata Gulf of St. Vincent, Australia S, TD Goodsell & Connell (2005) Homosira banksii Kaikoura & Moeraki, New Zealand S, TD, D, E Lilley & Schiel (2006) Laminaria spp. Cranberry Cove, Canada S, D Schmidt & Scheibling (2007) Macrocystis pyrifera Channel Islands National Park, USA S Graham (2004)

II. BIOTURBATORS

Chordata Exotic Cyprinus carpio Lake Kasumigaura, Japan ND Matsuzaki et al. (2007)

178 ND Matsuzaki et al. (2009) Native Dorosoma cepedianum Lake Texoma, USA ND Gido (2003) Parodon apolinari Rio Las Marias, Venezuela ND Flecker & Taylor (2004)

Crustacea Exotic Procambarus clarkii Lake Kasumigaura, Japan ND Matsuzaki et al. (2009) Native Neotrypaea californiensis Willapa Bay, USA S, TD, ND, BD Ferraro & Cole (2007) Pacifastacus leniusculus Coast Range Mountains, Canada TD, ND, BD Zhang et al. (2004) Upogebia pugettensis Willapa Bay, USA ND Posey et al. (1991) S, TD, ND, BD Ferraro & Cole (2007) Echinodermata Native Echinocardium sp. Mahurangi Harbour, New Zealand TD, D Lohrer et al. (2008)

Mollusca Exotic Corbicula fluminea Lake Constance, Germany ND Werner & Rothhaupt (2008) Littorina littorea Mount Hope Bay, USA S, ND Bertness (1984) Native Austrovenus stutchburyi Otago Harbour, New Zealand S, TD, ND, D, E Mouritsen & Poulin (2005) Mulinia edulis Compu, Chile S, TD, ND Jaramillo et al. (2007) Venus antiqua Compu, Chile S, TD, ND Jaramillo et al. (2007)

Polychaeta Exotic Marenzelleria spp. The Isefjord, Denmark ND, D Olsen et al. (2008) Native Hediste diversicolor Ria Formosa lagoon, Portugal S, TD, ND, BD Carvalho et al. (2007)

Total # ecosystem engineers: 88 Total # studies: 130

179 Data sources Aldridge, D.C., Fayle, T.M. & Jackson, N. (2007). Freshwater mussel abundance predicts biodiversity in UK lowland rivers. Aquatic Conservation: Marine and Freshwater Ecosystems, 17, 554-564. Ban, S.M. & Nelson, W.G. (1987). Role of Diopatra cuprea Bosc (Polychaeta: Onuphidae) tubes in structuring a subtropical infaunal community. Bulletin of Marine Science, 40, 11-21. Beekey, M.A., McCabe, D.J. & Marsden, J.E. (2004). Zebra mussel colonisation of soft sediments facilitates invertebrate communities. Freshwater Biology, 49, 535-545. Bertness, M.D. (1984). Habitat and community modification by an introduced herbivorous snail. Ecology, 65, 370-381. Bially, A. & MacIsaac, H.J. (2000). Fouling mussels (Dreissena spp.) colonize soft sediments in Lake Erie and facilitate benthic invertebrates. Freshwater Biology, 43, 85-97. Blum, J.C., Chang, A.L., Liljesthröm, M., Schenk, M.E., Steinberg, M.K. & Ruiz, G.M. (2007). The non-native solitary ascidian Ciona intestinalis (L.) depresses species richness. Journal of Experimental Marine Biolology and Ecology, 342, 5-14. Bolam, S.G. & Fernandes, T.F. (2003). Dense aggregations of Pygospio elegans (Claparède): effect on macrofaunal community structure and sediments. Journal of Sea Research, 49, 171-185. Boström, C. & Bonsdorff, E. (2000). Zoobenthic community establishment and habitat complexity – the importance of seagrass shoot-density, morphology and physical disturbance for faunal recruitment. Marine Ecology Progress Series, 205, 123-138. Botts, P.S., Patterson, B.A. & Schloesser, D.W. (1996). Zebra mussel effects on benthic invertebrates: physical or biotic? Journal of the North American Benthological Society, 15, 179-184. Bouma, T.J., Ortells, V. & Ysebaert, T. (2009). Comparing biodiversity effects among ecosystem engineers of contrasting strength: macrofauna diversity in

180 Zostera noltii and Spartina anglica vegetations. Helgoland Marine Research, 63, 3-18. Britton-Simmons, K.H. (2004). Direct and indirect effects of the introduced alga Sargassum muticum on benthic, subtidal communities of Washington State, USA. Marine Ecology Progress Series, 277, 61-78. Brusati, E.D. & Grosholz, E.D. (2006). Native and introduced ecosystem engineers produce contrasting effects on estuarine infaunal communities. Biological Invasions, 8, 683-695. Buschbaum, C., Dittmann, S., Hong, J.S., Hwang, I.S., Strasser, M., Thiel, M., Valdivia, N., Yoon, S.P. & Reise, K. (2009). Mytilid mussels: global habitat engineers in coastal sediments. Helgoland Marine Research, 63, 47-58. Callaway, R. (2006). Tube worms promote community change. Marine Ecology Progress Series, 308, 49-60. Carvalho, S., Barata, M., Gaspar, M.B., Pousão-Ferreira, P. & Cancela da Fonseca, L. (2007). Enrichment of aquaculture earthen ponds with Hediste diversicolor: consequences for benthic dynamics and natural productivity. Aquaculture, 262, 227-236. Castel, J., Labourg, P.J., Escaravage, V., Auby, I. & Garcia, M.E. (1989). Influence of seagrass beds and oyster parks on the abundance and biomass patterns of meio- and macrobenthos in tidal flats. Estuarine, Coastal and Shelf Science, 28, 71-85. Castilla, J.C., Lagos, N.A. & Cerda, M. (2004). Marine ecosystem engineering by the alien ascidian Pyura praeputialis on a mid-intertidal rocky shore. Marine Ecology Progress Series, 268, 119-130. Chapman, M.G., People, J. & Blockley, D. (2005). Intertidal assemblages associated with natural corralina turf and invasive mussel beds. Biodiversity and Conservation, 14, 1761-1776. Clarke, A., Lake, P.S. & O‘Dowd, D.J. (2004). Ecological impacts on aquatic macroinvertebrates following upland stream invasion by a ponded pasture grass (Glyceria maxima) in southern Australia. Marine and Freshwater Research, 55, 709-713.

181 Commito, J.A., Celano, E.A., Celico, H.J., Como, S. & Johnson, C.P. (2005). Mussel matter: postlarval dispersal dynamics altered by a spatially complex ecosystem engineer. Journal of Experimental Marine Biology and Ecology, 316, 133-147. Connolly, R.M. (1994). Removal of seagrass canopy: effects on small fish and their prey. Journal of Experimental Marine Biology and Ecology, 184, 99-110. Connolly, R.M. (1997). Differences in composition of small, motile invertebrate assemblages from seagrass and unvegetated habitats in a southern Australian estuary. Hydrobiologia, 346, 137-148. Cottet, M., de Montaudouin, X., Blanchet, H. & Lebleu, P. (2007). Spartina anglica eradication experiment and in situ monitoring assess structuring strength of habitat complexity on marine macrofauna at high tidal level. Estuarine, Coastal and Shelf Science, 71, 629-640. Creese, R., Hooker, S., de Luca, S. & Wharton, Y. (1997). Ecology and environmental impact of Musculista senhousia (Mollusca: Bivalvia: Mytilidae) in Tamaki Estuary, Auckland, New Zealand. New Zealand Journal of Marine and Freshwater Research, 31, 225-236. Crooks, J.A. (1998). Habitat alteration and community-level effects of an exotic mussel, Musculista senhousia. Marine Ecology Progress Series, 162, 137-152. Crooks, J.A. & Khim, H.S. (1999). Architectural vs. biological effects of a habitat-altering, exotic mussel, Musculista senhousia. Journal of Experimental Marine Biology and Ecology, 240, 53-75. Cummings, V.J., Thrush, S.F., Hewitt, J.E. & Funnell, G.A. (2001). Variable effect of a large suspension-feeding bivalve on infauna: experimenting in a complex system. Marine Ecology Progress Series, 209, 159-175. Cummings, V.J., Thrush, S.F., Hewitt, J.E. & Turner, S.J. (1998). The influence of the pinnid bivalve Atrina zealandica (Gray) on benthic macroinvertebrate communities in soft-sediment habitats. Journal of Experimental Marine Biology and Ecology, 228, 227-240.

182 Davoult, D., Dewarumez, J.M. & Migné, A. (1998). Long-term changes (1979- 1994) in two coastal benthic communities (English Channel): analysis of structural developments. Oceanologica Acta, 21, 609-617. Dermott, R., Mitchell, J., Murray, I. & Fear, E. (1993). Biomass and production of zebra mussels (Dreissena polymorpha) in shallow waters of northeastern Lake Erie. In T.F. Nalepa & D.W. Schloesser (Eds.), Zebra Mussels: Biology, Impacts, and Control (pp. 399-413). Boca Raton, FL: Lewis Publishers. Dittmann, S. (1990). Mussel beds – amensalism or amelioration for intertidal fauna? Helgölander Meeresuntersuchungen, 44, 335-352. Douglas, M.M. & O‘Connor, R.A. (2003). Effects of the exotic macrophyte, para grass (Urochloa mutica), on benthic and epiphytic macroinvertebrates of a tropical floodplain. Freshwater Biology, 48, 962-971. Duarte, C., Jaramillo, E., Contreras, H. & Figueroa, L. (2006). Community structure of the macrofauna in the sediments below an intertidal mussel bed (Mytilus chilensis (Hupe)) of southern Chile. Revista Chilena de Historia Natural, 79, 353-368. Dudley, T.L., Cooper, S.D. & Hemphill, N. (1986). Effects of macroalgae on a stream invertebrate community. Journal of the North American Benthological Society, 5, 93-106. Dürr, S. & Wahl, M. (2004). Isolated and combined impacts of blue mussels (Mytilus edulis) and barnacles (Balanus improvisus) on structure and diversity of a fouling community. Journal of Experimental Marine Biology and Ecology, 306, 181-195. Dusoge, K. (1966). Composition and interrelations between macrofauna living on stones in the littoral of Mikolajskie Lake. Ekologia Polska – Seria A, 39, 1-8. Edgar, G.J., Shaw, C., Watson, G.F. & Hammond, L.S. (1994). Comparisons of species richness, size-structure and production of benthos in vegetated and unvegetated habitats in Western Port, Victoria. Journal of Experimental Marine Biology and Ecology, 176, 201-226.

183 Eklöf, J.S., Castro, M.T., Adelsköld, L., Jiddawi, N.S. & Kautsky, N. (2005). Differences in macrofaunal and seagrass assemblages in seagrass beds with and without seaweed farms. Estuarine, Coastal and Shelf Science, 63, 385-396. Escapa, M., Isacch, J.P., Daleo, P., Alberti, J., Iribarne, O., Borges, M., Santos, E.P.D., Gagliardini, D.A. & Lasta, M. (2004). The distribution and ecological effects of the introduced Pacific Oyster Crassostrea gigas (Thunberg, 1793) in Northern Patagonia. Journal of Shellfish Research, 23, 765-772. Everett, R.A. (1994). Macroalgae in marine soft-sediment communities: effects on benthic faunal assemblages. Journal of Experimental Marine Biology and Ecology, 175, 253-274. Ferraro, S.P. & Cole, F.A. (2007). Benthic macrofauna – habitat associations in Willapa Bay, Washington, USA. Estuarine, Coastal and Shelf Science, 71, 491-507. Flecker, A.S. & Taylor, B.W. (2004). Tropical fishes as biological bulldozers: density effects on resource heterogeneity and species diversity. Ecology, 85, 2267-2278. Forrest, B.M. & Taylor, M.D. (2002). Assessing invasion impact: survey design considerations and implications for management of an invasive marine plant. Biological Invasions, 4, 375-386. Gido, K.B. (2003). Effects of gizzard shad on benthic communities in reservoirs. Journal of Fish Biology, 62, 1392-1404. Goodsell, P.J. & Connell, S.D. (2005). Historical configuration of habitat influences the effects of disturbance on mobile invertebrates. Marine Ecology Progress Series, 299, 79-87. Grabowski, J.H., Hughes, A. R., Kimbro, D.L. & Dolan, M.A. (2005). How habitat setting influences restored oyster reef communities. Ecology, 86, 1926- 1935. Graham, M.H. (2004). Effects of local deforestation on the diversity and structure of southern California giant kelp forest food webs. Ecosystems, 7, 341-357. Günther, C.P. (1996). Development of small Mytilus beds and its effects on resident intertidal macrofauna. Marine Ecology, 17, 117-130.

184 Gutiérrez, J. & Iribarne, O. (1999). Role of Holocene beds of the stout razor clam Tagelus plebeius in structuring present benthic communities. Marine Ecology Progress Series, 185, 213-228. Heck, K.L., Jr., Able, K.W., Roman, C.T. & Fahay, M.P. (1995). Composition, abundance, biomass, and production of macrofauna in a New England estuary: comparison among eelgrass meadows and other nursery habitats. Estuaries, 18, 379-389. Hedge, P. & Kriwoken, L.K. (2000). Evidence for effects of Spartina anglica invasion on benthic macrofauna in Little Swanport estuary, Tasmania. Austral Ecology, 25, 150-159. Hily, C. & Bouteille, M. (1999). Modifications of the specific diversity and feding guilds in an intertidal sediment colonized by an eelgrass meadow (Zostera marina) (Brittany, France). Life Sciences, 322, 1121-1131. Holloway, M.G. & Keough, M.J. (2002a). An introduced polychaete affects recruitment and larval abundance of sessile invertebrates. Ecological Applications, 12, 1803-1823. Holloway, M.G. & Keough, M.J. (2002b). Effects of an introduced polychaete, Sabella spallanzanii, on the development of epifaunal assemblages. Marine Ecology Progress Series, 236, 137-154. Horvath, T.G., Martin, K.M. & Lamberti, G.A. (1999). Effect of zebra mussels, Dreissena polymorpha, on macroinvertebrates in a lake-outlet stream. American Midland Naturalist, 142, 340-347. Hosack, G.R., Dumbauld, B.R., Ruesink, J.L., Armstrong, D.A. (2006). Habitat associations of estuarine species: comparisons of intertidal mudflat, seagrass (Zostera marina), and oyster (Crassostrea gigas) habitats. Estuaries and Coasts, 29, 1150-1160. Jaramillo, E., Contreras, H. & Duarte, C. (2007). Community structure of the macroinfauna inhabiting tidal flats characterized by the presence of different species of burrowing bivalves in Southern Chile. Hydrobiologia, 580, 85-96. Kimbro, D.L. & Grosholz, E.D. (2006). Disturbance influences oyster community richness and evenness, but not diversity. Ecology, 87, 2378-2388.

185 Kochmann, J., Buschbaum, C., Volkenborn, N. & Reise, K. (2008). Shift from native mussels to alien oysters: differential effects of ecosystem engineers. Journal of Experimental Marine Biology and Ecology, 364, 1-10. Kröncke, I. (1996). Impact of biodeposition on macrofaunal communities in intertidal sandflats. Marine Ecology, 17, 159-174. Kuhns, L.A. & Berg, M.B. (2003). Benthic invertebrate community responses to round goby (Neogobius melanostomus) and zebra mussel (Dreissena polymorpha) invasion in southern Lake Michigan. Journal of Great Lakes Research, 25, 910-917. Lana, P.C. & Guiss, C. (1991). Influence of Spartina alterniflora on structure and temporal variability of macrobenthic associations in a tidal flat of Paranaguá Bay (southeastern Brazil). Marine Ecology Progress Series, 73, 231-244. Lee, S.Y., Fong, C.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. Levin, L.A., Neira, C. & Grosholz, E.D. (2006). Invasive cordgrass modifies wetland trophic function. Ecology, 87, 419-432. Levin, L.A., Talley, D. & Thayer, G. (1996). Succession of macrobenthos in a created salt marsh. Marine Ecology Progress Series, 141, 67-82. Levin, L.A, Talley, T.S. & Hewitt, J. (1998). Macrobenthos of Spartina foliosa (Pacific cordgrass) salt marshes in southern California: community structure and comparison to a Pacific mudflat and a Spartina alterniflora (Atlantic smooth cordgrass) marsh. Estuaries, 21, 129-144. Lilley, S.A. & Schiel, D.R. (2006). Community effects following the deletion of a habitat-forming alga from rocky marine shores. Oecologia, 148, 672-681. Lohrer, A.M., Chiaroni, L.D., Hewitt, J.E. & Thrush, S.F. (2008). Biogenic disturbance determines invasion success in a subtidal soft-sediment system. Ecology, 89, 1299-1307. Luckenbach, M.W. (1984). Biogenic structure and foraging by five species of shorebirds (Charadrii). Estuarine, Coastal and Shelf Science, 19, 691-696.

186 Markert, A., Wehrmann, A. & Kröncke, I. (2010). Recently established Crassostrea-reefs versus native Mytilus-beds: differences in ecosystem engineering affects the macrofaunal communities (Wadden Sea of Lower Saxony, southern German Bight). Biological Invasions, 12, 15-32. Matsuzaki, S.S., Usio, N., Takamura, N. & Washitani, I. (2007). Effects of common carp on nutrient dynamics and littoral community composition: roles of excretion and bioturbation. Fundamental and Applied Limnology, 168, 27- 38. Matsuzaki, S.S., Usio, N., Takamura, N. & Washitani, I. (2009). Contrasting impacts of invasive engineers on freshwater ecosystems: an experiment and meta-analysis. Oecologia, 158, 673-686. Mattila, J., Chaplin, G., Eilers, M.R., Heck, Jr., K.L., O‘Neal, J.P. & Valentine, J.F. (1999). Spatial and diurnal distribution of invertebrate and fish fauna of a Zostera marina bed and nearby unvegetated sediments in Damariscotta River, Maine (USA). Journal of Sea Research, 41, 321-332. Mayer, C.M., Keats, R.A., Rudstam, L.G. & Mills, E.L. (2002). Scale-dependent effects of zebra mussels on benthic invertebrates in a large eutrophic lake. Journal of the North American Benthological Society, 21, 616-633. McCorry, M.J. & Otte, M.L. (2001). Ecological effects of Spartina anglica on the macro-invertebrate infauna of the mud flats at Bull Islands, Dublin Bay, Ireland. Web Ecology, 2, 71-73. 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. Mercer, J.M., Whitlatch, R.B. & Osman, R.W. (2009). Potential effects of the invasive ascidian (Didemnum vexillum Kott, 2002) on pebble-cobble bottom habitats in Long Island Sound, USA. Aquatic Invasions, 4, 133-142. Mistri, M. (2002). Ecological characteristics of the invasive Asian date mussel, Musculista senhousia, in the Sacca di Goro (Adriatic Sea, Italy). Estuaries, 25, 431-440.

187 Montiero, S.M., Chapman, M.G. & Underwood, A.J. (2002). Patches of the ascidian Pyura stolonifera (Heller, 1978): structure of habitat and associated intertidal assemblages. Journal of Experimental Marine Biology and Ecology, 270, 171-189. Mörtl, M. & Rothhaupt, K.O. (2003). Effects of adult Dreissena polymorpha on settling juveniles and associated macroinvertebrates. International Review of Hydrobiology, 88, 561-569. Mouritsen, K.N. & Poulin, R. (2005). Parasites boost biodiversity and changes animal community structure by trait-mediated indirect effects. Oikos, 108, 344- 350. Munari, C. (2008). Effects of the exotic invader Musculista senhousia on benthic communities of two Mediterranean lagoons. Hydrobiologia, 611, 29-43. 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. Neira, C., Levin, L.A., Grosholz, E.D. & Mendoza, G. (2007). Influence of invasive Spartina growth stages on associated macrofaunal communities. Biological Invasions, 9, 975-993. Nicastro, A., Bishop, M.J., Kelaher, B.P., Benedetti-Cecchi, L. (2009). Export of non-native gastropods shells to a coastal lagoon: alteration of habitat structure has negligible effects on infauna. Journal of Experimental Marine Biology and Ecology, 374, 31-36. Norkko, A., Hewitt, J.E., Thrush, S.F. & Funnell, G.A. (2001). Benthic-pelagic coupling and suspension-feeding bivalves: linking site-specific sediment flux and biodeposition to benthic community structure. Limnology and Oceanography, 46, 2067-2072. Norkko, A., Hewitt, J.E., Thrush, S.F. & Funnell, G.A. (2006). Conditional outcomes of facilitation by a habitat-forming subtidal bivalve. Ecology, 87, 226-234.

188 Norling, P. & Kautsky, N. (2007). Structural and functional effects of Mytilus edulis on diversity of associated species and ecosystem functioning. Marine Ecology Progress Series, 351, 163-175. O‘Brien, A.L., Ross, D.J. & Keough, M.J. (2006). Effects of Sabella spallanzanii physical structure on soft sediment macrofaunal assemblages. Marine and Freshwater Research, 57, 363-371. Olsen, J., Kayikci, O., Reyhe, T.S. & Happel, E.M. (2008). The effects of Marzelleria spp. on the benthic macro fauna community structure in the Isefjord. Thesis Project, Roskilde University, Denmark. Retrieved from http://diggy.ruc.dk/handle/1800/3809 Osman, R.W. & Whitlatch, R.B. (1995). The influence of resident adults on recruitment: a comparison to settlement. Journal of Experimental Marine Biology and Ecology, 190, 169-198. Pagliosa, P.R. & Lana, P.C. (2005). Impact of plant cover removal on macrobenthic community structure of a subtropical salt marsh. Bulletin of Marine Science, 77, 1-17. Piazzi, L. & Balata, D. (2008). The spread of Caulerpa racemosa var. cylindracea in the Mediterranean Sea: an example of how biological invasions can influence beta diversity. Marine Environmental Research, 65, 50-61. Posey, M.H. (1988). Community changes associated with the spread of an introduced seagrass, Zostera japonica. Ecology, 69, 974-983. Posey, M.H., Dumbauld, B.R. & Armstrong, D.A. (1991). Effects of a burrowing mud shrimp, Upogebia pugettensis (Dana), on abundances of macro-infauna. Journal of Experimental Marine Biology and Ecology, 148, 283-294. Posey, M.H., Wigand, C. & Stevenson, J.C. (1993). Effects of an introduced aquatic plant, Hydrilla verticillata, on benthic communities in the upper Chesapeake Bay. Estuarine, Coastal and Shelf Science, 37, 539-555. Pranovi, F., Curiel, D., Rismondo, A., Marzocchi, M. & Scattolin, M. (2000). Variations of the macrobenthic community in a seagrass transplanted area of the Lagoon of Venice. Scientia Marina, 64, 303-310.

189 Rabaut, M., Guilini, K., Van Hoey, G., Vincx, M. & Degraer, S. (2007). A bio- engineered soft-bottom environment: the impact of Lanice concheliga on the benthic species-specific densities and community structure. Estuarine, Coastal and Shelf Science, 75, 525-536. Ragnarsson, S.A. & Raffaelli, D. (1999). Effects of the mussel Mytilus edulis L. on the invertebrate fauna of sediments. Journal of Experimental Marine Biology and Ecology, 241, 31-43. Reed, B.J. & Hovel, K.A. (2006). Seagrass habitat disturbance: how loss and fragmentation of eelgrass Zostera marina influences epifaunal abundance and diversity. Marine Ecology Progress Series, 326, 133-143. Reed, T., Wielgus, S.J., Barnes, A.K., Schiefelbein, J.J., & Fettes, A.L. (2004). Refugia and local controls: Benthic invertebrate dynamics in Lower Green Bay, Lake Michigan following zebra mussel invasion. Journal of Great Lakes Research, 30, 390-396. Ricciardi, A., Whoriskey, F.G. & Rasmussen, J.B. (1997). The role of the zebra mussel (Dreissena polymorpha) in structuring macroinvertebrate communities on hard substrates. Canadian Journal of Fisheries and Aquatics Sciences, 54, 2596-2608. Robinson, T.B. & Griffiths, C.L. (2002). Invasion of Langebaan Lagoon, South Africa, by Mytilus galloprovincialis – effects on natural communities. African Zoology, 37, 151-158. Robinson, T.B., Griffiths, C.L., Branch, G.M. & Govender, A. (2007). The invasion and subsequent die-off of Mytilus galloprovincialis in Langebaan Lagoon, South Africa: effects on natural communities. Marine Biology, 152, 225-232. Ross, D.J., Keough, M.J., Longmore, A.R. & Knott, N.A. (2007). Impacts of two introduced suspension feeders in Port Phillip Bay, Australia. Marine Ecology Progress Series, 340, 41-53. Sahade, R., Tatián, M., Kowalke, J., Kühne, S. & Esnal, G.B. (1998). Benthic fauna associations on soft substrates at Potter Cove, King George Island, Antarctica. Polar Biology, 19, 85-91.

190 Sánchez-Moyano, J.E., Estacio, F.J., García-Adiego, E.M. & García-Gómez, J.C. (2001). Effect of the vegetative cycle of Caulerpa prolifera on the spatio- temporal variation of invertebrate macrofauna. Aquatic Botany, 70, 163-174. Sardiña, P., Cataldo, D.H. & Boltovskoy, D. (2008). The effects of the invasive mussel, Limnoperna fortunei, on associated fauna in South American freshwaters: importance of physical structure and food supply. Fundamental and Applied Limnology, 173, 135-144. Schaffner, L.C. (1990). Small-scale organism distributions and patterns of species diversity: evidence for positive interactions in an estuarine benthic community. Marine Ecology Progress Series, 61, 107-117. Schmidt, A.L. & Scheibling, R.E. (2007). Effects of native and invasive macroalgal canopies on composition and abundance of mobile benthic macrofauna and turf-forming algae. Journal of Experimental Marine Biology and Ecology, 341, 110-130. Schwindt, E. & Iribarne, O.O. (2000). Settlement sites, survival and effects on benthos of an introduced reef-building polychaete in a SW Atlantic coastal lagoon. Bulletin of Marine Science, 67, 73-82. Siebert, T. & Branch, G.M. (2007). Influences of biological interactions on community structure within seagrass beds and sandprawn-dominated sandflats. Journal of Experimental Marine Biology and Ecology, 340, 11-24. Stewart, T.W., Miner, J.G. & Lowe, R.L. (1998). Macroinvertebrate communities on hard substrates in western Lake Erie: structuring effects of Dreissena. Journal of Great Lakes Research, 24, 868-879. Strayer, D.L., Lutz, C., Malcom, H.M., Munger, K. & Shaw, W.H. (2003). Invertebrate communities associated with a native (Vallisneria americana) and an alien (Trapa natans) macrophyte in a large river. Freshwater Biology, 48, 1938-1949. Strong, J.A., Dring, M.J. & Maggs, C.A. (2006). Colonisation and modification of soft substratum habitats by the invasive macroalga Sargassum muticum. Marine Ecology Progress Series, 321, 87-97.

191 Sylvester, F., Boltovskoy, D. & Cataldo, D. (2007). The invasive bivalve Limnoperna fortunei enhances benthic invertebrate densities in South American floodplain rivers. Hydrobiologia, 589, 15-27. Thayer S.A., Haas R.C., Hunter R.D. & Kushler, R.H. (1997). Zebra mussel (Dreissena polymorpha) effects on sediment, other zoobenthos, and the diet and growth of adult yellow perch (Perca flavescens) in pond enclosures. Canadian Journal of Fisheries and Aquatics Sciences, 54, 1903-1915. Theel, H.J., Dibble, E.D. & Madsen, J.D. (2008). Differential influence of a monotypic and diverse native aquatic plant bed on a macroinvertebrate assemblage; an experimental implication of exotic plant induced habitat. Hydrobiologia, 600, 77-87. Tokeshi, M. & Romero, L. (1995). Filling a gap: dynamics of space occupancy on a mussel-dominated subtropical rocky shore. Marine Ecology Progress Series, 119, 167-176. Tsuchiya, M. (2002). Faunal structures associated with patches of mussels on East Asian coasts. Helgoland Marine Research, 56, 31-36. Turner, S.J. & Kendall, M.A. (1999). A comparison of vegetated and unvegetated soft-sediment macrobenthic communities in the River Yealm, south-western Britain. Journal of the Marine Biological Association of the United Kingdom, 79, 741-743. Valdivia, N. & Thiel, M. (2006). Effects of point-source nutrient addition and mussel removal on epibiotic assemblages in Perumytilus purpuratus beds. Journal of Sea Research, 56, 271-283. van Houte-Howes, K.S.S., Turner, S.J. & Pilditch, C.A. (2004). Spatial differences in macroinvertebrate communities in intertidal seagrass habitats and unvegetated sediment in three New Zealand estuaries. Estuaries, 27, 945- 957. Vaughn, C.C. & Spooner, D.E. (2006). Unionid mussels influence macroinvertebrate assemblage structure in streams. Journal of the North American Benthological Society, 25, 691-700.

192 Werner, S. & Rothhaupt, K.O. (2008). Effects of the invasive Asian clam Corbicula fluminea on benthic macroinvertebrate taxa in laboratory experiments. Fundamental and Applied Limnology, 173, 145-152. Whitcraft, C.R. & Levin, L.A. (2007). Regulation of benthic algal and animal communities by salt marsh plants: impact of shading. Ecology, 88, 904-917. Witman, J.D. (1985). Refuges, biological disturbance, and rocky subtidal community structure in New England. Ecological Monographs, 55, 421-445. Young, C.M. (1989). Larval depletion by ascidians has little effect on settlement of epifauna. Marine Biology, 102, 481-489. Zhou, H.X., Liu, J. & Qin, P. (2009). Impacts of an alien species (Spartina alterniflora) on the macrobenthos community of Jiangsu coastal inter-tidal ecosystem. Ecological Engineering, 35, 521-528. Zühlke, R. (2001). Polychaete tubes create ephemeral community patterns: Lanice conchilega (Pallas, 1766) associations studied over six years. Journal of Sea Research, 46, 261-272.

193 Appendix C. Data sources for meta-analysis of the effects of exotic ecosystem engineers (Chapter 3).

Table C.1. List of exotic ecosystem engineering species included in this meta-analysis with data sources. ES = effect size (impact on a native species or genus). Superscripts denote references for engineer status (distinct or shared taxon), if different from effect size data sources.

Ecosystem Engineer Engineer status # ES Location Effect size data sources Species Family Genus Family d lnR

Caulerpa taxifolia Caulerpaceae St. George Basin, Australia1,2 shared shared McKinnon et al. (2009) — 20 Codium fragile ssp. Codiaceae Cranberry Cove, Canada3 distinct distinct Schmidt & Scheibling (2007) 6 6 Tomentosoides Corophium curvispinum Corophiidae Lower Rhine River, Netherlands distinct distinct van der Brink et al. (1993) — 3 Crassostrea gigas Ostreidae Bahia Anegada, Argentina4 distinct shared Escapa et al. (2004) 7 7 Wadden Sea, Germany5,6 distinct shared Kochman et al. (2008) 20 21 Dreissena polymorpha Dreissenidae Lake Champlain, USA distinct distinct Beekey et al. (2004) 9 — Lake Constance, Germany distinct distinct Mörtl & Rothhaupt (2003) 4 4 Lake Erie, USA distinct distinct Stewart et al. (1998) 32 32 Botts et al. (1996) 5 5 Lake Michigan, USA distinct distinct Kuhns & Berg (2003) 6 6 Lake St. Louis, Canada distinct distinct Ricciardi et al. (1997) 14 14 Hydrilla verticilatta Hydrocharitaceae Chesapeake Bay, USA7,8 distinct shared Posey et al. (1993) 10 10 Musculista senhousia Mytilidae Mission Bay, USA9 distinct shared Crooks & Khim (1999) 21 21 Sacco di Goro, Adriatic Sea5 distinct shared Mistri (2002) 16 26 Munari (2008) 7 — Tyrrhenian Sea, Sardinia5 distinct shared Munari (2008) 17 —

194 Mytilus galloprovincialis Mytilidae Langebaan Lagoon, South Africa5 distinct shared Robinson & Griffiths (2002) — 59 Sabella spallanzanii Sabellidae Outer Harbour, Australia10,11 distinct shared Holloway & Keough (2002a) 1 1 Holloway & Keough (2002b) 7 7 Port Phillip Bay, Australia10,11 distinct shared Ross et al. (2007) 5 5 Holloway & Keough (2002a) 5 5 Sargassum muticum Sargassaceae Langestone Harbour, Britain1,12 distinct shared Strong et al. (2006) 5 5 Strangford Lough, Ireland1,12 distinct shared Strong et al. (2006) 9 9 Spartina alterniflora Poaceae Willapa Bay, USA7,8 distinct shared Ferraro & Cole (2007) — 13 Spartina alterniflora x Poaceae San Francisco Bay, USA shared shared Neira et al. (2007) 15 15 foliosa hybrid Levin et al. (2006) 3 3 Neira et al. (2005) 14 14 Spartina anglica Poaceae Arcachon Bay, France13,14 shared shared Cottet et al. (2007) — 27 Dublin Bay, Ireland15,16 distinct shared McCorry & Otte (2001) 5 5 Little Swanport estuary, Tasmania17,18 distinct shared Hedge & Kriwoken (2000) — 17 Westerschelde estuary, Netherlands19 shared shared Bouma et al. (2009) 26 26 Styela clava Styelidae Port Phillip Bay, Australia10,20 distinct shared Ross et al. (2007) 5 5 Trapa natans Trapaceae Hudson River, USA7,8 distinct distinct Strayer et al. (2003) — 22 Zostera japonica Zosteraceae Coos Bay, USA7,21 shared shared Posey (1988) 16 16

Total Species: 16 Total Studies: 30 Total ES: 290 429

195 Effect size data sources

Beekey, M.A., McCabe, D.J. & Marsden, J.E. (2004). Zebra mussel colonisation of soft sediments facilitates invertebrate communities. Freshwater Biology, 49, 535-545. Botts, P.S., Patterson, B.A. & Schloesser, D.W. (1996). Zebra mussel effects on benthic invertebrates: physical or biotic? Journal of the North American Benthological Society, 15, 179-184. Bouma, T.J., Ortells, V. & Ysebaert, T. (2009). Comparing biodiversity effects among ecosystem engineers of contrasting strength: macrofauna diversity in Zostera noltii and Spartina anglica vegetations. Helgoland Marine Research, 63, 3-18. Cottet, M., de Montaudouin, X., Blanchet, H. & Lebleu, P. (2007). Spartina anglica eradication experiment and in situ monitoring assess structuring strength of habitat complexity on marine macrofauna at high tidal level. Estuarine, Coastal and Shelf Science, 71, 629-640. Crooks, J.A. & Khim, H.S. (1999). Architectural vs. biological effects of a habitat-altering, exotic mussel, Musculista senhousia. Journal of Experimental Marine Biology and Ecology, 240, 53-75. Escapa, M., Isacch, J.P., Daleo, P., Alberti, J., Iribarne, O., Borges, M., Santos, E.P.D., Gagliardini, D.A. & Lasta, M. (2004). The distribution and ecological effects of the introduced Pacific Oyster Crassostrea gigas (Thunberg, 1793) in Northern Patagonia. Journal of Shellfish Research, 23, 765-772. Ferraro, S.P. & Cole, F.A. (2007). Benthic macrofauna – habitat associations in Willapa Bay, Washington, USA. Estuarine, Coastal and Shelf Science, 71, 491-507. Hedge, P. & Kriwoken, L.K. (2000). Evidence for effects of Spartina anglica invasion on benthic macrofauna in Little Swanport estuary, Tasmania. Austral Ecology, 25, 150-159. Holloway, M.G. & Keough, M.J. (2002a). An introduced polychaete affects recruitment and larval abundance of sessile invertebrates. Ecological Applications, 12, 1803-1823.

196 Holloway, M.G. & Keough, M.J. (2002b). Effects of an introduced polychaete, Sabella spallanzanii, on the development of epifaunal assemblages. Marine Ecology Progress Series, 236, 137-154. Kochmann, J., Buschbaum, C., Volkenborn, N. & Reise, K. (2008). Shift from native mussels to alien oysters: differential effects of ecosystem engineers. Journal of Experimental Marine Biology and Ecology, 364, 1-10. Kuhns, L.A. & Berg, M.B. (2003). Benthic invertebrate community responses to round goby (Neogobius melanostomus) and zebra mussel (Dreissena polymorpha) invasion in southern Lake Michigan. Journal of Great Lakes Research, 25, 910-917. Levin, L.A., Neira, C. & Grosholz, E.D. (2006). Invasive cordgrass modifies wetland trophic function. Ecology, 87, 419-432. McCorry, M.J. & Otte, M.L. (2001). Ecological effects of Spartina anglica on the macro-invertebrate infauna of the mud flats at Bull Islands, Dublin Bay, Ireland. Web Ecology, 2, 71-73. 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. Mistri, M. (2002). Ecological characteristics of the invasive Asian date mussel, Musculista senhousia, in the Sacca di Goro (Adriatic Sea, Italy). Estuaries, 25, 431-440. Mörtl, M. & Rothhaupt, K.O. (2003). Effects of adult Dreissena polymorpha on settling juveniles and associated macroinvertebrates. International Review of Hydrobiology, 88, 561-569. Munari, C. (2008). Effects of the exotic invader Musculista senhousia on benthic communities of two Mediterranean lagoons. Hydrobiologia, 611, 29-43. 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.

197 Neira, C., Levin, L.A., Grosholz, E.D. & Mendoza, G. (2007). Influence of invasive Spartina growth stages on associated macrofaunal communities. Biological Invasions, 9, 975-993. Posey, M.H. (1988). Community changes associated with the spread of an introduced seagrass, Zostera japonica. Ecology, 69, 974-983. Posey, M.H., Wigand, C. & Stevenson, J.C. (1993). Effects of an introduced aquatic plant, Hydrilla verticillata, on benthic communities in the upper Chesapeake Bay. Estuarine, Coastal and Shelf Science, 37, 539-555. Ricciardi, A., Whoriskey, F.G. & Rasmussen, J.B. (1997). The role of the zebra mussel (Dreissena polymorpha) in structuring macroinvertebrate communities on hard substrates. Canadian Journal of Fisheries and Aquatics Sciences, 54, 2596-2608. Robinson, T.B. & Griffiths, C.L. (2002). Invasion of Langebaan Lagoon, South Africa, by Mytilus galloprovincialis – effects on natural communities. African Zoology, 37, 151-158. Ross, D.J., Keough, M.J., Longmore, A.R. & Knott, N.A. (2007). Impacts of two introduced suspension feeders in Port Phillip Bay, Australia. Marine Ecology Progress Series, 340, 41-53. Schmidt, A.L. & Scheibling, R.E. (2007). Effects of native and invasive macroalgal canopies on composition and abundance of mobile benthic macrofauna and turf-forming algae. Journal of Experimental Marine Biology and Ecology, 341, 110-130. Stewart, T.W., Miner, J.G. & Lowe, R.L. (1998). Macroinvertebrate communities on hard substrates in western Lake Erie: structuring effects of Dreissena. Journal of Great Lakes Research, 24, 868-879. Strayer, D.L., Lutz, C., Malcom, H.M., Munger, K. & Shaw, W.H. (2003). Invertebrate communities associated with a native (Vallisneria americana) and an alien (Trapa natans) macrophyte in a large river. Freshwater Biology, 48, 1938-1949.

198 Strong, J.A., Dring, M.J. & Maggs, C.A. (2006). Colonisation and modification of soft substratum habitats by the invasive macroalga Sargassum muticum. Marine Ecology Progress Series, 321, 87-97. van den Brink, F.W.B., van der Velde, G. & bij de Vaate, A. (1993). Ecological aspects, explosive range extension and impact of a mass invader, Corophium curvispinum Sars, 1895 (Crustacea: Amphipoda), in the Lower Rhine (The Netherlands). Oecologia, 93, 224-232.

Engineer status references

1. Guiry, M.D. & Guiry, G.M. (2009). AlgaeBase. Galway: National University of Ireland. Retrieved from: http://www.algaebase.org. 2. Millar, A.J.K. & Kraft, G.T. (1994). Catalogue of marine benthic green algae (Chlorophyta) of New South Wales, including Lord Howe Island, South- western Pacific. Australian Systematic Botany, 7, 419-453. 3. South, G.R. & Cardinal, A. (1970). A checklist of marine algae of eastern Canada. Canadian Journal of Botany, 48, 2077-2095. 4. Carranza, A., Defeo, O. & Beck, M. (2009). Diversity, conservation status and threats to native oysters (Ostreidae) around the Atlantic and Caribbean coasts of South America. Aquatic Conservation: Marine and Freshwater Ecosystems, 19, 344-353. 5. Appeltans, W., Bouchet, P., Boxshall, G.A., Fauchald, K., Gordon, D.P., Hoeksema, B.W., Poore, G.C.B., van Soest, R.W.M., Stöhr, S., Walter, T.C. & Costello, M.J. (Eds.) (2009). The World Register of Marine Species. Retrieved from: http://www.marinespecies.org. 6. Riesen, W. & Reise, K. (1982). Macrobenthos of the subtidal Wadden Sea: revisited after 55 years. Helgoländer Meeresuntersuchungen, 35, 409-423. 7. USDA, NRCS (2009). The PLANTS Database. Baton Rouge, LA: National Plant Data Center. Retrieved from: http://plants.usda.gov. 8. U.S. Native Plant Information Network. (2009). Lady Bird Johnson Wildflower Center, Austin, TX. Retrieved from: http://www.wildflower.org/plants/.

199 9. Austin, W.C. (1985). An Annotated Checklist of Marine Invertebrates in the Cold Temperate Northeast Pacific. Cowichan Bay, Canada: Khoyatan Marine Laboratory. 10. ABRS (2009). Australian Faunal Directory. Australian Biological Resources Study, Canberra. Retrieved from: http://www.environment.gov.au/biodiversity/abrs/online- resources/fauna/afd/index.html. 11. Knight-Jones, P. & Perkins, T.H. (1998). A revision of Sabella, Bispira and Stylomma (Polychaeta: Sabellidae). Zoological Journal of the Linnean Society, 123, 385-467. 12. National Biodiversity Network (2009). National Biodiversity Network Gateway. Nottingham, UK. Retrieved from: http://data.nbn.org.uk/. 13. Nielsen, L.B., Finster, K., Welsh, D.T., Donelly, A., Herbert, R.A., de Wit, R. & Lomstein, B.A. (2001). Sulphate reduction and nitrogen fixation rates associated with roots, rhizomes and sediments from Zostera noltii and Spartina maritima meadows. Environmental Microbiology, 3, 63-71. 14. Baumel, A., Ainouche, M.L. & Levasseur, J.E. (2001). Molecular investigations in populations of Spartina anglica C.E. Hubbard (Poaceae) invading coastal Brittany (France). Molecular Ecology, 10, 1689-1701. 15. Boyle, P.J. & Kavanagh, J.A. (1961). A Spartinetum at Baldoyle in Ireland. Nature, 192, 81-82. 16. Botanical Society of the British Isles (BSBI). (2009). Distribution Maps Scheme. Retrieved from: http://www.bsbi.org.uk. 17. Morley, B.D. & Toelken, H.R. (Eds.) (1983). Flowering Plants in Australia. Adelaide, Australia: Rigby. 18. Australian National Botanic Gardens (2009). CPBR and ANBG databases. Retrieved from: http://www.anbg.gov.au/. 19. Clapham, A.R., Tutin, T.G. & Moore, D.M. (1987). Flora of the British Isles. New York, NY: Cambridge University Press.

200 20. Cohen, B.F., Currie, D.R. & McArthur, M.A. (2000). Epibenthic community structure in Port Phillip Bay, Victoria, Australia. Marine and Freshwater Research, 51, 689-702. 21. Hansen, G.I. (1997). A revised checklist and preliminary assessment of the macrobenthic marine algae and seagrasses of Oregon. In T.N. Kaye, A. Liston, R.M. Love, D.L. Luoma, R.J. Meinke & M.V. Wilson (Eds.), Conservation and Management of Native Flora and Fungi (pp. 175-200). Corvallis, OR: Native Plant Society of Oregon.

201 Appendix D. Data sources for meta-analysis of Dreissena effects (Chapter 4).

Table D.1. Site locations and data sources for Dreissena meta-analysis: total taxonomic richness (S), taxa density (T), numerical density (N), biomass density (B), diversity indices (D), evenness indices (E). Study site Effect sizes Data sources S T N B D E 1. Appletree Bay, Lake Champlain x x Beekey et al. (2004) 2. Hawkins Bay, Lake Champlain x x Beekey et al. (2004) 3. North Bass Island, Lake Erie (5.5m) x x Bially & MacIssac (2000) 4. North Bass Island, Lake Erie (7.3m) x x Bially & MacIssac (2000) 5. North Bass Island, Lake Erie (5.5m) x x Bially & MacIssac (2000) 6. Presque Isle Bay, Lake Erie x Botts et al. (1996) 7. Long Point, Lake Erie x x x x x Dermott & Kerec (1997) 8. Longbeach, Lake Erie x x Dermott et al. (1993) 9. Station I, Mikolajskie Lake, Poland x Dusoge (1966) 10. Station II, Mikolajskie Lake, Poland x Dusoge (1966) 11. Site 63, Lake St. Clair x x x x x Griffiths (1993) 12. Christiana Creek, USA x x x x x Horvath et al. (1999) 13. Lukomskoe Lake, Belarus x x Karataev & Burlakova (1995); Karatayev et al. (1997) 14. Calumet Harbor, Lake Michigan x x Kuhns & Berg (2003) 15. Low depositional zone, Lake Ontario x Lozano et al. (2001) 16. Lake Lukoml', Belarus x x Lyakhnovich et al. (1988) 17. Mud site, Oneida Lake, New York x Mayer et al. (2002) 18. Curtis Point, Rice Lake x Mercer et al. (1999) 19. Lake Constance, Germany x x x x Mörtl & Rothhaupt (2003) 20. Statn. 11, Saginaw Bay, Lake Huron x x x x x x Nalepa, et al. (2002); Nalepa, et al. (2003) 21. Statn. 13, Saginaw Bay, Lake Huron x x x x x x Nalepa, et al. (2002); Nalepa, et al. (2003) 22. Statn. 14, Saginaw Bay, Lake Huron x x x x x x Nalepa, et al. (2002); Nalepa, et al. (2003) 23. Statn. 16, Saginaw Bay, Lake Huron x x x x x x Nalepa, et al. (2002); Nalepa, et al. (2003) 24. Olcott, Lake Ontario (55m) x Owens & Dittman (2003)

202 25. Olcott, Lake Ontario (75m) x Owens & Dittman (2003) 26. Rochester, Lake Ontario (55m) x Owens & Dittman (2003) 27. Rochester, Lake Ontario (75m) x Owens & Dittman (2003) 28. Oswego, Lake Ontario (55m) x Owens & Dittman (2003) 29. Oswego, Lake Ontario (75m) x Owens & Dittman (2003) 30. Site 24, Lower Green Bay, Lake x Reed et al. (2004) Michigan (1m) 31. Site 24, Lower Green Bay, Lake x Reed et al. (2004) Michigan (3m) 32. Site 17, Lower Green Bay, Lake x Reed et al. (2004) Michigan (1m) 33. Site 17, Lower Green Bay, Lake x Reed et al. (2004) Michigan (3m) 34. Site 1, Fox Inlet, Lower Green Bay, x Reed et al. (2004) Lake Michigan 35. Site 9, Fox Inlet, Lower Green Bay, x Reed et al. (2004) Lake Michigan 36. Site 14, Lower Green Bay, Lake x Reed et al. (2004) Michigan 37. Site 4E, Lower Green Bay, Lake x Reed et al. (2004) Michigan (1m) 38. Site 4E, Lower Green Bay, Lake x Reed et al. (2004) Michigan (3m) 39. Lake St. Louis, Lachine x x x x x Ricciardi et al. (1997) 40. Lake St. Francois, Cornwall x x x x x Ricciardi et al. (1997) 41. Soulanges Canal, Pte-des-Cascades x x x x x Ricciardi et al. (1997) 42. Cobble site, Lake Ontario x x x x Stewart & Haynes (1994); Haynes, et al. (1999); Haynes, et al. (2005) 43. Reef site, Lake Ontario x x x x Stewart & Haynes (1994); Haynes, et al. (1999); Haynes, et al. (2005) 44. Fishery Bay, Lake Erie x x x x x x Stewart, et al. (1998a); Stewart, et al. (1998b); Stewart, et al. (1999) 45. Hudson River (shallow) x x Strayer, et al. (1998); Strayer & Smith (2001) 46. Hudson River (deep) x Strayer, et al. (1998);

203 Strayer & Smith (2001) 47. Pond enclosures, Lake St. Clair x Thayer et al. (1997)

Data sources

Beekey, M.A., McCabe, D.J. & Marsden, J.E. (2004). Zebra mussel colonisation of soft sediments facilitates invertebrate communities. Freshwater Biology, 49, 535-545. Bially, A. & MacIsaac, H.J. (2000). Fouling mussels (Dreissena spp.) colonize soft sediments in Lake Erie and facilitate benthic invertebrates. Freshwater Biology, 43, 85-97. Botts, P.S., Patterson, B.A. & Scholesser, D.W. (1996). Zebra mussel effects on benthic invertebrates: physical or biotic? Journal of the North American Benthological Society, 15, 179-184. Dermott, R. & Kerec, D. (1997). Changes to the deepwater benthos of eastern Lake Erie since the invasion of Dreissena: 1979-1993. Canadian Journal of Fisheries and Aquatic Sciences, 54, 922-930. Dermott, R., Mitchell, J., Murray, I. & Fear, E. (2003). Biomass and production of zebra mussels (Dreissena polymorpha) in shallow waters of northeastern Lake Erie. In T. F. Nalepa & D. W. Schloesser (Eds.), Zebra Mussels: Biology, Impacts and Control (pp. 399-413). Boca Raton, FL: Lewis Publishers. Dusoge, K. (1966). Composition and interrelations between macrofauna living on stones in the littoral of Mikolajskie Lake. Ekologia Polska – Seria A, 39, 1-8. Griffiths, R.W. (1993). Effects of zebra mussels (Dreissena polymorpha) on the benthic fauna of Lake St. Clair. In T. F. Nalepa & D. W. Schloesser (Eds.), Zebra Mussels: Biology, Impacts and Control (pp. 414-437). Boca Raton, FL: Lewis Publishers. Haynes, J.M., Stewart, T.W. & Cook, G.E. (1999). Benthic macroinvertebrate communities in southwestern Lake Ontario following invasion of Dreissena: Continuing change. Journal of Great Lakes Research, 25, 828-838.

204 Haynes, J.M., Tisch, N.A., Mayer, C.M. & Rhyne, R.S. (2005). Benthic macroinvertebrate communities in southwestern Lake Ontario following invasion of Dreissena and Echinogammarus: 1983 to 2000. Journal of the North American Benthological Society, 24, 148-167. Horvath, T.G., Martin, K.M. & Lamberti, G.A. (1999). Effect of zebra mussels, Dreissena polymorpha, on macroinvertebrates in a lake-outlet stream. American Midland Naturalist, 142, 340-347. Karataev, A.Y. & Burlakova, L.E. (1995). The role of Dreissena in lake ecosystems. Russian Journal of Ecology, 26, 207-211. Karatayev, A.Y., Burlakova, L.E. & Padilla, D.K. (1997). The effects of Dreissena polymorpha (Pallas) invasion on aquatic communities in Eastern Europe. Journal of Shellfish Research, 16, 187-203. Kuhns, L.A. & Berg, M.B. (1999). Benthic invertebrate community responses to round goby (Neogobius melanostomus) and zebra mussel (Dreissena polymorpha) invasion in southern Lake Michigan. Journal of Great Lakes Research, 25, 910-917. Lozano, S.J., Scharold, J.V. & Nalepa, T.F. (2001). Recent declines in benthic macroinvertebrate densities in Lake Ontario. Canadian Journal of Fisheries and Aquatic Sciences, 58, 518-529. Lyakhnovich, V.P., Karataev, A.Y., Mitrakhovich, P.A., Gur‘yanova, L.V. & Vezhnovets, G.G. (1988). Productivity and prospects for utilizing the ecosystem of Lake Lukoml‘, Thermoelectric station cooling reservoir. Soviet Journal of Ecology, 18, 255-259. Mayer, C.M., Keats, R.A., Rudstam, L.G. & Mills, E.L. (2002). Scale-dependent effects of zebra mussels on benthic invertebrates in a large eutrophic lake. Journal of the North American Benthological Society, 21, 616-633. Mercer, J.L., Fox, M.G. & Metcalfe, C.D. (1999). Changes in benthos and three littoral zone fishes in a shallow, eutrophic Ontario lake following the invasion of the zebra mussel (Dreissena polymorpha). Journal of Lake and Reservoir Management, 15, 310-323.

205 Mörtel, M. & Rothhaupt, K.-O. (2003). Effects of adult Dreissena polymorpha on settling juveniles and associated macroinvertebrates. International Review of Hydrobiology, 88, 561-569. Nalepa, T.F., Fanslow, D.L., Lansing, M.B. & Lang, G.A. (2003). Trends in the benthic macroinvertebrate community of Saginaw Bay, Lake Huron, 1987 to 1996: Responses to phosphorus abatement and the zebra mussel, Dreissena polymorpha. Journal of Great Lakes Research, 29, 14-33. Nalepa, T.F., Fanslow, D.L., Lansing, M.B., Lang, G.A., Ford, M, Gostenik, G. & Hartson, D.J. (2002). Abundance, biomass, and species composition of benthic macroinvertebrate populations in Saginaw Bay, Lake Huron, 1987-96. NOAA Technical Memorandum GLERL-122. Ann Arbor, MI: Great Lakes Environmental Research Laboratory. Owens, R.W. & Dittman, D.E. (2003). Shifts in the diets of slimy sculpin (Cottus cognatus) and lake whitefish (Coregonus clupeaformis) in Lake Ontario following the collapse of the burrowing amphipod Diporeia. Aquatic Ecosystem Health and Management, 6, 311-323. Reed, T., Wielgus, S.J., Barnes, A.K., Schiefelbein, J.J., & Fettes, A.L. (2004). Refugia and local controls: Benthic invertebrate dynamics in Lower Green Bay, Lake Michigan following zebra mussel invasion. Journal of Great Lakes Research, 30, 390-396. Ricciardi, A., Whoriskey, F.G. & Rasmussen, J.B. (1997). The role of the zebra mussel (Dreissena polymorpha) in structuring macroinvertebrate communities on hard substrata. Canadian Journal of Fisheries and Aquatic Sciences, 54, 2596-2608. Stewart, T.W. & Haynes, J.M. (1994). Benthic macroinvertebrate communities of southwestern Lake Ontario following invasion of Dreissena. Journal of Great Lakes Research, 20, 479-493. Stewart, T.W., Miner, J.G. & Lowe, R.L. (1998a). Macroinvertebrate communities on hard substrates in western Lake Erie: Structuring effects of Dreissena. Journal of Great Lakes Research, 24, 868-879.

206 Stewart, T.W., Miner, J.G. & Lowe, R.L. (1998b). Quantifying mechanisms for zebra mussel effects on benthic macroinvertebrates: organic matter production and shell-generated habitat. Journal of the North American Benthological Society, 17, 81-94. Stewart, T.W., Miner, J.G. & Lowe, R.L. (1999). A field experiment to determine Dreissena and predator effects on zoobenthos in a nearshore, rocky habitat of western Lake Erie. Journal of the North American Benthological Society, 18, 488-498. Strayer, D.L. & Smith, L.C. (2001). The zoobenthos of the freshwater tidal Hudson River and its response to the zebra mussel (Dreissena polymorpha) invasion. Archiv fur Hydrobiologie. Supplementband. Monographic Studies, 139, 1-52. Strayer, D.L., Smith, L.C. & Hunter, D.C. (1998). Effects of the zebra mussel (Dreissena polymorpha) invasion on the macrobenthos of the freshwater tidal Hudson River. Canadian Journal of Zoology, 76, 419-425. Thayer S.A., Haas R.C., Hunter R.D. & Kushler, R.H. (1997). Zebra mussel (Dreissena polymorpha) effects on sediment, other zoobenthos, and the diet and growth of adult yellow perch (Perca flavescens) in pond enclosures. Canadian Journal of Fisheries and Aquatic Sciences, 54, 1903-1915.

207 Appendix E. Macroinvertebrate taxa collected from experimental substrata (Chapter 5).

Table E.1. List of macroinvertebrate taxa collected from experimental substrata in the three experiments. Numbers are mean abundances (± SE) per brick. Dreissena spp. were excluded from calculations of the total number of taxa, total abundance and taxa density.

Experiment 1 Experiment 2 Experiment 3 Control Half Full Control Half Checker Full Control Half Checker Full Mollusca Bivalvia Dreissenidae — — — Dreissena bugensis Andrusov 0.3 0.3 0.2 0.6 0.1 — 0.2 — (0.1) (0.1) (0.1) (0.2) (0.1) (0.1) Dreissena polymorpha (Pallas) 0.4 0.4 0.8 0.5 0.3 0.5 0.1 0.1 (0.1) (0.2) (0.2) (0.2) (0.2) (0.2) (0.1) (0.1) Sphaeriidae — — — Pisidium sp. — — 0.1 — — — — — (0.1) Gastropoda Ancylidae — — — Ferrissia rivularis (Say) 0.1 — — 0.1 0.2 0.1 0.2 0.1 (0.1) (0.1) (0.2) (0.1) (0.1) (0.1) Bithyniidae Bithynia tentaculata (L.) — 0.2 0.1 2.5 2.9 2.7 3.7 0.6 1.2 0.2 0.6 (0.1) (0.1) (0.2) (0.5) (0.6) (0.5) (0.2) (0.5) (0.1) (0.3)

208 Hydrobiidae 1.0 1.0 0.7 (0.4) (0.5) (0.4) Amnicola limosa (Say) 9.9 12.9 8.8 10.5 3.4 6.9 6.4 7.5 (1.0) (2.0) (1.2) (1.5) (0.6) (1.2) (1.5) (1.8) Marstonia decepta (Baker) 0.1 0.1 0.1 — 0.3 0.4 0.1 0.8 (0.1) (0.1) (0.1) (0.2) (0.2) (0.1) (0.4) Probythinella lacustris (Baker) 0.2 — 0.1 — — — — — (0.1) (0.1) Lymnaeidae — — — Stagnicola catascopium (Say) 9.3 9.7 9.6 9.9 8.4 8.9 8.1 9.6 (1.2) (1.0) (1.2) (1.1) (1.4) (1.6) (1.8) (1.3) Physidae — — — Physa gyrina (Say) 2.4 2.8 2.6 2.2 4.0 2.8 2.5 2.8 (0.4) (0.4) (0.5) (0.5) (0.7) (0.6) (0.7) (0.6) Planorbidae 1.8 3.5 1.7 (0.8) (1.2) (0.7) Gyraulus deflectus (Say) 13.1 19.4 17.6 27.2 12.2 12.4 16.4 26.9 (1.6) (3.4) (2.4) (3.1) (2.1) (2.3) (2.7) (7.9) Helisoma anceps (Menke) — — 0.1 — — — — 0.1 (0.1) (0.1) Helisoma campanulatum (Say) — 0.1 — — — — — — (0.1) Planorbella trivolvis (Say) 0.1 — — — — — — — (0.1) Planorbula armigera (Say) — — — 0.1 — — — — (0.1) Pleuroceridae — — — Elimia livescens (Menke) 0.2 0.2 0.1 0.3 0.6 0.6 0.7 1.0

209 (0.2) (0.1) (0.1) (0.1) (0.4) (0.2) (0.3) (0.4) Pleurocera acuta Rafinesque 0.1 0.4 0.4 0.3 1.0 0.2 1.1 0.7 (0.1) (0.2) (0.2) (0.1) (0.3) (0.1) (0.3) (0.3) Valvatidae — 0.1 — (0.1) Valvata tricarinata (Say) 2.5 1.8 2.5 2.9 0.7 1.0 1.3 0.6 (0.6) (0.3) (0.7) (0.5) (0.3) (0.4) (0.4) (0.2) Viviparidae — — — Campeloma decisum (Say) — — — — — 0.1 0.1 — (0.1) (0.1) Crustacea Amphipoda Gammaridae Gammarus fasciatus Say 1.3 42.0 109.8 102.1 94.3 78.7 74.5 7.9 24.0 28.4 49.9 (0.7) (6.6) (12.4) (20.0) (13.5) (8.7) (11.4) (1.8) (6.3) (8.1) (21.7) Echinogammarus ischnus 0.3 0.1 — 25.6 31.6 36.3 39.2 5.4 15.4 10.4 8.7 (Stebbing) (0.2) (0.1) (6.1) (7.4) (6.1) (5.2) (1.3) (3.8) (2.0) (3.3) Hyalellidae — — — Hyalella azteca Saussure — — — — — 0.1 — — (0.1) Isopoda Asellidae Caecidotea 1.4 1.6 2.6 — — — — — — — — (1.0) (0.4) (0.8) Insecta Coleoptera Elmidae — — — Stenelmis 0.2 0.1 — — — — 0.1 0.3

210 (0.1) (0.1) (0.1) (0.3) Diptera Chironomidae 6.0 7.5 6.4 (1.4) (0.9) (0.9) Rheotanytarsus — — — 0.1 0.2 0.4 — 0.2 (0.1) (0.1) (0.2) (0.2) other Chironomidae 37.0 29.3 23.8 38.1 45.5 43.3 37.5 30.2 (6.6) (4.0) (2.9) (6.9) (11.0) (7.8) (7.9) (6.5) Empididae — — — Hemerodromia 0.3 — — 0.1 — — — — (0.1) (0.1) Ephemeroptera Baetidae — — — Cloeon — — — 0.1 0.1 0.1 0.1 0.2 (0.1) (0.1) (0.1) (0.1) (0.1) Caenidae 0.8 3.3 2.7 (0.4) (0.6) (0.9) Caenis 0.1 0.2 0.1 0.1 — — 0.1 0.1 (0.1) (0.1) (0.1) (0.1) (0.1) (0.1) Ephemerellidae — — — Attenella — — — — 0.1 — — — (0.1) Eurylophella 0.1 0.1 0.2 0.2 0.3 0.2 0.1 — (0.1) (0.1) (0.1) (0.1) (0.2) (0.1) (0.1) Heptageniidae 0.1 0.3 — (0.1) (0.2) Stenacron 0.1 0.2 0.5 0.3 3.5 3.6 2.5 1.7 (0.1) (0.1) (0.2) (0.2) (1.8) (1.2) (1.2) (0.7)

211 Lepidoptera Pyralidae — — — Petrophila — — — — 0.1 1.1 0.8 0.4 (0.1) (0.9) (0.5) (0.2) Trichoptera Brachycentridae — 0.1 — (0.1) Brachycentrus incanus Hagen 0.1 0.2 0.3 0.2 — 0.1 — — (0.1) (0.1) (0.1) (0.1) (0.1) Glossosomatidae — — — Protoptila 0.1 0.1 — 0.1 — — — — (0.1) (0.1) (0.1) Helicopsychidae — — — Helicopsyche borealis (Hagen) 1.7 2.9 2.1 1.9 3.3 3.2 2.2 3.6 (0.4) (0.5) (0.5) (0.4) (0.7) (0.6) (0.6) (0.7) Hydropsychidae — 0.2 — (0.1) Cheumatopsyche — 0.2 0.1 0.8 — 0.1 — 0.1 (0.1) (0.1) (0.3) (0.1) (0.1) Hydropsyche 0.2 0.7 1.2 1.1 — 0.1 0.4 0.1 (0.1) (0.2) (0.2) (0.2) (0.1) (0.3) (0.1) Hydroptilidae — 0.3 — (0.2) Agraylea 1.0 0.5 0.2 0.1 — — — — (0.3) (0.2) (0.1) (0.1) Hydroptila 0.2 0.1 0.1 — 0.8 1.1 0.3 0.9 (0.1) (0.1) (0.1) (0.3) (0.3) (0.2) (0.3) Leptoceridae 0.1 0.7 0.7

212 (0.1) (0.3) (0.2) Ceraclea alces (Ross) 0.2 0.1 0.2 0.2 — — — — (0.1) (0.1) (0.1) (0.1) Ceraclea ancylus (Vorhies) 0.2 0.2 0.2 0.1 — — — — (0.1) (0.1) (0.1) (0.1) Ceraclea annulicornis (Stephens) 0.1 0.1 0.2 — — — — — (0.1) (0.1) (0.1) Ceraclea maculata (Banks) 0.1 — — — — — — — (0.1) Ceraclea neffi (Resh) — — — 0.3 — — — — (0.1) Ceraclea resurgens (Walker) 0.1 0.1 0.1 0.2 — — — — (0.1) (0.1) (0.1) (0.1) Ceraclea transversa (Hagen) 0.1 — — — — — — — (0.1) Ceraclea sp. Moira River Resh — — — 0.1 — — — — (0.1) Mystacides 0.2 — 0.1 0.1 — — 0.1 — (0.1) (0.1) (0.1) (0.1) Nectopsyche 0.2 — — 0.1 — — — — (0.1) (0.1) Oecetis 0.3 0.2 0.1 0.2 — — — — (0.1) (0.1) (0.1) (0.1) Limnephilidae — — — 0.1 — — — — — — — (0.1) Polycentropodidae — 0.3 0.2 (0.2) (0.1) Neureclipsis — 0.1 0.1 0.2 — — — — (0.1) (0.1) (0.1)

213 Polycentropus 0.8 0.8 0.9 0.8 0.6 0.2 0.3 1.1 (0.3) (0.2) (0.2) (0.3) (0.4) (0.1) (0.2) (0.3) Psychomyiidae — — — Psychomyia 0.5 0.5 0.4 0.6 — — — 0.1 (0.3) (0.2) (0.2) (0.2) (0.1) immature Trichoptera 2.7 4.1 3.7 4.1 0.7 1.6 1.3 0.6 (0.7) (0.5) (0.7) (1.3) (0.3) (1.0) (0.7) (0.3) Annelida Hirudinea Glossiphoniidae 0.3 0.9 0.8 (0.2) (0.3) (0.3) Abloglossiphonia heteroclita (L.) — — 0.1 — — — — — (0.1) Batracobdella phalera (Graf) 0.1 — — — 0.4 — — — (0.1) (0.3) Glossiphonia complanata — 0.2 0.2 0.3 0.2 — 0.1 0.2 (0.1) (0.1) (0.1) (0.2) (0.1) (0.1) Helobdella fusca (Castle) — — 0.1 — — — — 0.1 (0.1) (0.1) Oligochaeta Naididae 0.1 1.3 0.6 (0.1) (0.5) (0.4) Chaetogaster — 0.2 — — — — — — (0.2) Nais sp. 0.1 0.1 — 0.1 0.1 0.1 0.3 0.1 (0.1) (0.1) (0.1) (0.1) (0.1) (0.3) (0.1) Stylaria lacustris (L.) — 0.1 — — 0.6 — — — (0.1) (0.5)

214 other Naididae 0.7 1.6 1.2 4.0 6.2 3.4 2.7 3.0 (0.4) (0.5) (0.6) (1.2) (3.5) (1.9) (1.8) (1.1) Tubificidae — 0.2 0.1 (0.1) (0.1) Spirosperma nikolskyi — 0.1 0.1 0.1 — — — — (Lastockin & Sokolskaya) (0.1) (0.1) (0.1) Platyhelminthes Proseriata Plagiostomidae — — — Hydrolimax grisea Haldeman — 0.5 0.3 0.8 — — 0.1 0.5 (0.2) (0.1) (0.4) (0.1) (0.3) Tricladida Dugesiidae — — — Cura foremanii Girard — — — 0.1 0.1 — — — (0.1) (0.1) Planariidae 1.9 5.9 4.3 (0.6) (1.1) (1.1) Dugesia tigrina Girard 4.2 11.1 13.1 20.2 4.9 19.2 12.3 24.8 (1.1) (3.1) (2.6) (5.4) (1.2) (6.7) (2.5) (5.7) Dugesia sp. 0.4 0.5 0.7 0.7 0.3 0.4 0.3 — (0.1) (0.2) (0.2) (0.2) (0.2) (0.3) (0.2) Arachnida Hydracarina 3.7 2.9 3.8 4.9 4.2 3.3 4.2 1.8 2.9 1.0 2.1 (1.1) (0.5) (1.4) (1.2) (0.5) (0.4) (0.6) (0.4) (0.8) (0.4) (0.5) Cnidaria Hydrozoa Hydridae — — — Hydra — — — — — — 0.1 —

215 (0.1) Nemertea Hoplonemertea Tetrastemmatidae — — — Prostoma graecense (Bohmig) 0.2 0.9 0.2 1.7 0.4 0.5 0.3 1.0 (0.1) (0.6) (0.1) (0.5) (0.3) (0.2) (0.2) (0.5)

Total No. Taxa 13 20 14 45 43 43 46 32 32 34 34 Total abundance (#/brick 18.8 72.2 134.3 224.9 235.4 212.5 252.0 114.9 155.7 138.9 180.7 ± SE) (2.6) (7.6) (13.2) (24.9) (21.1) (12.4) (25.5) (17.1) (23.0) (21.5) (31.3) Taxa density (#/brick 5.4 10.3 8.4 15.8 17.1 16.5 18.1 16.2 16.3 15.1 17.1 ± SE) (0.8) (0.4) (0.4) (0.5) (0.6) (0.5) (0.9) (1.1) (0.7) (0.9) (0.7)

216