USF Libraries
USF Digital Collections

Depositional dynamics in seagrass systems of tampa bay, fl :

MISSING IMAGE

Material Information

Title:
Depositional dynamics in seagrass systems of tampa bay, fl : influence of hydrodynamic regime and vegetation density on ecosystem function
Physical Description:
Book
Language:
English
Creator:
Meyers, Alison
Publisher:
University of South Florida
Place of Publication:
Tampa, Fla
Publication Date:

Subjects

Subjects / Keywords:
Thalassia testudinum
Water flow
Fauna
Particle accumulation
Artificial seagrass units
Dissertations, Academic -- Biology - Integrative -- Doctoral -- USF   ( lcsh )
Genre:
non-fiction   ( marcgt )

Notes

Abstract:
ABSTRACT: Many coastal ecosystems around the world are dominated by submerged aquatic vegetation (SAV) habitats. These SAV habitats are known to provide many highly valuable ecosystem services such as habitat for commercial important species and increased water clarity. Water flow is an environmental variable which can have measurable effects on the ecosystem services provided by SAV, but is often not considered in studies assessing these services. This dissertation sought to investigate the links between SAV, primarily seagrasses, and hydrodynamics, paying special attention to the effects on sediments and fauna. Three main areas are discussed: (1) the effects of SAV on flow, (2) the effects of SAV and flow on deposition in SAV beds, and (3) the effects of SAV and flow on faunal communities in SAV beds. Seagrasses and other SAV reduce currents, attenuate waves, and dampen turbulence within their vegetative canopies, which in turn can enhance deposition and reduce the resuspension of sediment, organic matter, and passively settling larvae. The ability of SAV to retard flow may be further enhanced by increases in vegetated structure, such as shoot density, biomass, or canopy height, which can promote increased abundance and diversity of in- and epifauna within SAV beds. Ultimately, it is clear that hydrodynamics is an important factor that shapes SAV communities both physically (e.g. deposition, sediment structure, etc.) and biologically (e.g. faunal community composition, predation pressure, food availability, etc.).
Thesis:
Dissertation (Ph.D.)--University of South Florida, 2010.
Bibliography:
Includes bibliographical references.
System Details:
Mode of access: World Wide Web.
System Details:
System requirements: World Wide Web browser and PDF reader.
Statement of Responsibility:
by Alison Meyers.
General Note:
Title from PDF of title page.
General Note:
Document formatted into pages; contains X pages.
General Note:
Includes vita.

Record Information

Source Institution:
University of South Florida Library
Holding Location:
University of South Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
usfldc doi - E14-SFE0003343
usfldc handle - e14.3343
System ID:
SFS0027659:00001


This item is only available as the following downloads:


Full Text

PAGE 1

Depositional Dynamics in Seagrass Systems of Tampa Bay, FL: Influence of Hydrodynamic Regime and Ve getation Density on Ecosystem Function by Alison Cheryl Meyers A dissertation submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy Department of Biology College of Arts and Sciences University of South Florida Co-Major Professor: Susan S. Bell, Ph.D. Co-Major Professor: Florence I.M. Thomas, Ph.D. Thomas Chrisman, Ph.D. Gordon Fox, Ph.D. Date of Approval: March 25, 2010 Keywords: Thalassia testudinum water flow, fauna, particle accumulation, artificial seagrass units Copyright 2010, Alison Cheryl Meyers

PAGE 2

Dedication This doctoral dissertatio n is dedicated to my family. Without their continued support, encouragement, patience, and love th is dissertation would not possible.

PAGE 3

Acknowledgements I would like to express my grat itude to both of my major prof essors, Drs. Flo Thomas and Susan Bell. Dr. Flo Thomas for her years of support, insightful discussions of concepts and data, and friendship throughout my doctoral degree, and Dr. Susan Bell for her graciousness at accepting a student middegree and endless hours on manuscript revisions. I would also like to thank my committee member s, Drs. Tom Chrisman and Gordon Fox, for all of their advice and input re garding my dissertation and future career. Thanks must also be given to Drs. Gary Huxel and Markus Hue ttel for all of their assistance in the early years of my degree. I am also indebted to the numerous students, friends, and post-docs that contri buted considerable time and effort toward my research both in the field and in the lab. Specifical ly, I would like to ex tend thanks to Bill Ryerson, Alicia Fox, Dr. Lisa Whitenack, Neal Halstead, Mike Middlebrooks, Laura Bedinger, Hank Custin, Justin Krebs, Kris R obbins, Justin Bowles, Dr. Alex Tewfik, Dr. Louise Kregting, and last but not least Dr. Louise Firth. I would like also to single out the invaluable help of Stacy Villanueva and Dr. Brian Badgley who helped me more than I am sure they are aware. I would also lik e to give special tha nks to Dr. Kyle AveniDeforge, who in addition to his field he lp spent hours upon hour s helping with my hydrodynamic data.

PAGE 4

i Table of Contents List of Tables ..................................................................................................................... iii List of Figures ............................................................................................................... ..... iv Abstract ...................................................................................................................... ........ ix Chapter 1 Overview of Research ......................................................................................1 Research Goals .........................................................................................................4 Chapter Objectives ...................................................................................................5 Significance of Research ..........................................................................................6 Chapter 2 Depositional Processes in Seagrass Beds: Rethinking Density as a Measure of Ecosystem Function ....................................................................................8 Introduction ..............................................................................................................8 Methods..................................................................................................................10 Experimental Design ..................................................................................10 Particle Accumulation ................................................................................13 Site Characterization ..................................................................................14 Data Analysis .............................................................................................16 Hydrodynamic Characterization ................................................................18 Results ....................................................................................................................20 Hydrodynamic Characterization ................................................................20 Site Characterization ..................................................................................27 Particle Accumulation ................................................................................30 Discussion ..............................................................................................................38 Hydrodynamic Characterization ................................................................38 Particle Accumulation ................................................................................41 Implications................................................................................................43 Chapter 3 Effects of Patchy Habitat Structure on Depositional Processes in Seagrass, Thalassia testudinum Systems of Lower Tampa Bay, FL ..........................47 Introduction ............................................................................................................47 Methods..................................................................................................................50 Experimental Design ..................................................................................50 Particle Accumulation ................................................................................53 Total Suspended Solids ..............................................................................54 Data Analysis .............................................................................................55 Hydrodynamic Characterization ................................................................56 Results ....................................................................................................................58

PAGE 5

ii Hydrodynamic Characterization ................................................................58 Total Suspended Solids ..............................................................................66 Particle Accumulation ................................................................................66 Discussion ..............................................................................................................76 Hydrodynamic Characterization ................................................................76 Particle Accumulation ................................................................................78 Conclusions ................................................................................................81 Chapter 4 Investigation of the Effect s of Submerged Aquatic Vegetation and Hydrodynamic Regime on Select Ecosyste m Processes in Vegetated Systems: A Review .....................................................................................................................83 Introduction ............................................................................................................83 Background ............................................................................................................86 Hydrodynamics and SAV ..........................................................................86 Hydrodynamic Conditions .............................................................86 Canopy Height ...............................................................................89 Vegetation Structure ......................................................................90 Hydrodynamics, SAV, and Sedimentation ................................................92 Hydrodynamics, SAV, and Faunal Communities ......................................96 Review ...................................................................................................................98 Methods......................................................................................................98 Sedimentation ............................................................................................99 Sedimentation in Vegetated and Unvegetated Habitats .................99 Variation in SAV Struct ure and Sedimentation ...........................113 SAV, Sedimentation, and the Effects of Flow .............................115 Faunal Communities ................................................................................117 Faunal Communities in Vegetated and Unvegetated Habitats ........................................................................................117 Variation in SAV Structur e and Faunal Communities ................120 SAV, Faunal Communities, and the Effects of Flow...................122 Conclusions & Synthesis .....................................................................................124 Implications..........................................................................................................125 Literature Cited ................................................................................................................129 Appendix A Additional Tables and Figures .................................................................144 About the Author ................................................................................................... End Page

PAGE 6

iii List of Tables Table 1 Average ( SD) percent dry weight, organic matter, and carbonate content by sediment size fraction ( m) in sediments collected from vegetated (Thalassia testudinum) and unvegetated (bare sand) benthic habitats located within fast (0.078 0.041 m s-1) and slow (0.025 0.01 m s-1) experimental fl ow sites at Emerson Point Park in lower Tampa Bay, FL. ...................................................................32 Table 2 Summary of studi es investigating influence of submerged aquatic vegetation (SAV) structure on sedimentation. .........................................100 Table 3 Summary of studi es investigating influence of submerged aquatic vegetation (SAV) structure on asso ciated faunal communities. ..............104

PAGE 7

List of Figures Figure 1 Map of lower Tampa Bay, FL with locations of the fast (0.078 0.041 m s-1) and slow (0.025 0.01 m s-1) experimental flow sites. ...........................................................................................................12 Figure 2 Average ( SEM, n = 6) bulk flow speeds (cm s-1) measured over artificial seagrass unit (ASU) plot s with either high (1500 shoots m-2) or low (300 shoots m-2) seagrass shoot densities at fast and slow experimental flow sites at Emerson Point Park in lower Tampa Bay, FL. .........................................................................................21 Figure 3 Average ( SEM, n = 6) per cent reduction in flow speeds in the water column above compared to flow speeds within the canopy of artificial seagrass unit (ASU) plot s with either high (1500 shoots m-2) or low (300 shoots m-2) seagrass shoot densities at fast (0.078 0.041 m s-1) and slow (0.025 0.01 m s-1) experimental flow sites at Emerson Point Park in lower Tampa Bay, FL. ......................22 Figure 4 Relationship between reduc tion in overlying flow speed in artificial seagrass unit canopies and overlying flow speed (m s-1) for high (1500 shoot m-2) and low (300 shoots m-2) shoot density ASUs at Emerson Point Park in lower Tampa Bay, FL. ..............23 Figure 5 Speed profiles (mean SEM, n = 6) from field measures of flow within and above canopy of artif icial seagrass unit (ASU) plots with either high (1500 shoots m-2) or low (300 shoots m-2) shoot densities at fast and slow experime ntal flow sites at Emerson Point Park in lower Tampa Bay, FL. ...................................................................24 Figure 6 Relationship between turbulent kinetic energy (m2 s-2) within and above the canopy of artificial seag rass units (ASU) and overlying flow speed (m s-1) for high (1500 shoot m-2) and low (300 shoots m-2) shoot density ASUs at Em erson Point Park in lower Tampa Bay, FL. Best fit lines represented by solid line () for the high and dashed line (----) for the low shoot density treatments. ..............25 iv

PAGE 8

Figure 7 Relationship between Reynolds sh ear stress (Pa) within and above the canopy of artificial seagrass units (ASU) and overlying flow speed (m s-1) for high (1500 shoot m-2) and low (300 shoots m-2) shoot density ASUs at Emerson Point Park in lower Tampa Bay, FL. Best fit lines re presented by solid line () for the high and dashed line (----) for the low shoot density treatments. ......................26 Figure 8 Relationship between Reynolds shear stress (Pa) within the canopy (i.e. 5 cm above bottom or 20% of the canopy height) of artificial seagrass units (ASU) and overlying flow speed (m s-1) for high (1500 shoot m-2) and low (300 shoots m-2) shoot density ASUs at Emerson Point Park in lower Tampa Bay, FL. ......................................28 Figure 9 Average ( SD, n = 3) a) dry weight (g L-1) of total suspended solids (TSS), b) dry weight of organic matter in TSS (g L-1), and c) percent organic matter in the T SS, across 12 replicate weeks at fast (0.078 0.041 m s-1) and slow (0.025 0.01 m s-1) experimental flow sites at Emerson Point Park in lower Tampa Bay, FL. .....................................................................................................29 Figure 10 Average ( SD, n = 4) percen t dry weight by sediment size fraction a) at fast (0.078 0.041 m s-1) and slow (0.025 0.01 m s-1) experimental flow sites and b) in vegetated (Thalassia testudinum ) and unvegetated (bare sand) benthi c habitats within those study sites. ...........................................................................................................31 Figure 11 Average ( SEM, n = 10) dry weight (g m-2 day-1) of particles in 63 m and <63 m particle size fractions accumulated in artificial seagrass units (ASU) with either high (1500 shoot m-2) or low (300 shoots m-2) seagrass shoot densities at fast (0.078 0.041 m s-1) and slow (0.025 0.01 m s-1) experimental flow sites located at Emerson Point Park in lower Tampa Bay, FL. ..........................34 Figure 12 Average ( SEM, n = 10) dry weight (g m-2 day-1) of organic matter in 63 m and <63 m particle size fractions accumulated in artificial seagrass units (ASU) with either high (1500 shoot m-2) or low (300 shoots m-2) seagrass shoot densities at fast (0.078 0.041 m s-1) and slow (0.025 0.01 m s-1) experimental flow sites located at Emerson Point Park in lower Tampa Bay, FL. .........................................................................................35 v

PAGE 9

Figure 13 Average ( SEM, n = 10) dry weight of carbonates (g m-2 day-1) in 63 m particle size fraction accumulate d in artificial seagrass units (ASU) with high (1500 shoot m-2) or low (300 shoots m-2) seagrass shoot densities at fast (0.078 0.041 m s-1) and slow (0.025 0.01 m s-1) experimental flow sites loca ted at Emerson Point Park in lower Tampa Bay, FL. ...............................................................................36 Figure 14 Exponential reductions in pa rticle accumulation efficiency by artificial seagrass unit (ASU) plot s with increasing water column flow speed (m s-1). Artificial seagrass units with either high (1500 shoots m-2) or low (300 shoots m-2) seagrass shoot densities at fast (0.078 0.041 m s-1) and slow (0.025 0.01 cm s1) experimental flow sites at Emerson Point Park in lower Tampa Bay, FL. .....................................................................................................37 Figure 15 Location of study sites, North skyway (27.11N, 82.57W) and East Beach (2738.77N, 82.70W), in lower Tampa Bay, FL. ...............................................52 Figure 16 Average ( SEM, n = 6) bulk flow speeds (m s-1) measured above experimentally thinned patches of seagrass ( Thalassia testudinum ) with shoot densities ranging from full density (Full), half of the original shoot density (50%), 10% of the original density (10%), and complete shoot removal (Bar e) at two study sites ( North Skyway and East Beach) in Tampa Bay, FL. ........................................59 Figure 17 Speed profiles (m s-1) from field measures of flow within and above canopies of experimentally thinned patches of seagrass ( Thalassia testudinum ) with shoot densities ranging from full density (Full), half of the orig inal shoot density (50%), 10% of the original dens ity (10%), and complete shoot removal (Bare) and averaged ( SD, n = 6) across two study sites (North Skyway and East Beach) in Tampa Bay, FL. ............................................61 Figure 18 Normalized speed profiles (U/Umaximum) from field measures of flow within and above the ca nopy of experimentally thinned patches of seagrass ( Thalassia testudinum ) on days with a) maximum ( 0.04 m s-1) and b) minimum ( 0.02 m s-1) flows. .................62 vi

PAGE 10

Figure 19 Relationship between turbulent kinetic energy (m2 s-2) within and above the canopy of experimentally thinned patches of seagrass ( Thalassia testudinum ) and overlying flow speed (m s-1) for shoot densities ranging from full density (Full), half of the original shoot density (50%), 10% of the original density (10%), and complete shoot removal (Bare) at two study sites (North Skyway and East Beach) in Tampa Bay, FL. ..........................................................63 Figure 20 Relationship between turbulent kinetic energy (m2 s-2) within the canopy (i.e. 5 cm above bottom or 20% of the canopy height) of experimentally thinned patches of seagrass ( Thalassia testudinum ) and overlying flow speed (m s-1) for shoot densities ranging from full density (Full), half of th e original shoot density (50%), 10% of the original density (10 %), and complete shoot removal (Bare) at two study sites (North Skyway and East Beach) in Tampa Bay, FL. .....................................................................................................64 Figure 21 Relationship between Reynolds shear stress (Pa) within and above the canopy of experimentally thinned patches of seagrass ( Thalassia testudinum ) and overlying flow speed (m s-1) for shoot densities ranging from full density (Full), half of the original shoot density (50%), 10% of the original density (10%), and complete shoot removal (Bare) at two study sites (North Skyway and East Beach) in Tampa Bay, FL. ..........................................................65 Figure 22 Average ( SD, n = 3) a) dry weight (g L-1) of total suspended solids (TSS), b) dry weight of organic matter in TSS (g L-1), and c) percent organic matter in TSS, acr oss seven replicate weeks at two experimental study sites ( North Skyway and East Beach) in Tampa Bay, FL. .........................................................................................67 Figure 4 Average ( SEM, n = 10) dry weight (g m-2 day-1) of particles accumulated by experimentally thinned patches of seagrass ( Thalassia testudinum ) with shoot densitie s ranging from full density (Full), half of the original shoot density (50%), 10% of the original density (10%), and comple te shoot removal (Bare) at two study sites (North Skyway and East Beach) in Tampa Bay, FL. ...............68 Figure 24 Average ( SEM, n = 5) dry weight (g m-2 day-1) of particles by particle size fraction ( m) accumulated in experimentally thinned patches of seagrass ( Thalassia testudinum ) at a) North Skyway and b) East Beach study sties............................................................................69 vii

PAGE 11

viii Figure 5 Average ( SEM, n = 5) pe rcent dry weight by particle size fraction ( m) of particles accumulated in experimentally thinned patches of seagrass ( Thalassia testudinum ) at a) North Skyway and b) East Beach study sties............................................................................71 Figure 26 Average ( SEM, n = 10) percent organic matter in particles accumulated by experimentally thinned patches of seagrass ( Thalassia testudinum ) with shoot densitie s ranging from full density (Full), half of the original shoot density (50%), 10% of the original density ( 10%), and comple te shoot removal (Bare) created at two study sites (Nor th Skyway and East Beach) in Tampa Bay, FL. ..............................................................................................................72 Figure 27 Average ( SEM, n = 5) percent organic matter composition by particle size fraction ( m) of particles accumulated in experimentally thinned patches of seagrass ( Thalassia testudinum ) at a) North Skyway and b) East Beach study sties. ....................................73 Figure 28 Average ( SEM, n = 10) percent carbonates in the particles accumulated by experimentally thinned patches of seagrass ( Thalassia testudinum ) with shoot densitie s ranging from full density (Full), half of the original shoot density (50%), 10% of the original density (10%), and comple te shoot removal (Bare) created at two study sites (Nor th Skyway and East Beach) in Tampa Bay, FL. ..............................................................................................................74 Figure 6 Average ( SEM, n = 5) pe rcent carbonate composition by particle size fraction of particles accumula ted in experimentally thinned patches of seagrass ( Thalassia testudinum ) at a) North Skyway and b) East Beach study sties............................................................................75 Figure 30 Conceptual models of the in teractions between submerged aquatic vegetation (SAV) and sedime ntation under equivalent hydrodynamic conditions and how those interactions are modified by varying SAV densities. .........................................................................93 Figure 31 Conceptual models of interac tions between equivalent densities of submerged aquatic vegetation (S AV) and sedimentation and how those interactions are modified by hydrodynamic conditions. ..................94

PAGE 12

ix Depositional Dynamics in Seagrass Systems of Tampa Bay, FL: Influence of Hydrodynamic Regime and Ve getation Density on Ecosystem Function Alison Cheryl Meyers ABSTRACT Many coastal ecosystems around the worl d are dominated by submerged aquatic vegetation (SAV) habitats. These SAV ha bitats are known to provide many highly valuable ecosystem services such as hab itat for commercial im portant species and increased water clarity. Water flow is an environmental variable which can have measurable effects on the ecosystem servic es provided by SAV, but is often not considered in studies assessing these services. This dissertation s ought to investigate the links between SAV, primarily seagrasses, a nd hydrodynamics, paying special attention to the effects on sediments and fauna. Three ma in areas are discussed: (1) the effects of SAV on flow, (2) the effects of SAV and fl ow on deposition in SAV beds, and (3) the effects of SAV and flow on faunal communities in SAV beds. Seagrasses and other SAV reduce currents, attenuate wa ves, and dampen turbulence within their vegetative canopies, which in turn can enhance depositi on and reduce the resusp ension of sediment, organic matter, and passively settling larvae. The ability of SAV to retard flow may be further enhanced by increases in vegetated stru cture, such as shoot density, biomass, or canopy height, which can promote increased abu ndance and diversity of inand epifauna within SAV beds. Ultimately, it is clear that hydrodynamics is an important factor that

PAGE 13

x shapes SAV communities both physically (e.g. de position, sediment structure, etc.) and biologically (e.g. faunal community compositi on, predation pressure, food availability, etc.).

PAGE 14

1 Chapter 1 Overview of Research Seagrass communities are both highly producti ve and vital ecosystems within the marine realm. They can act as habitats and nursery grounds fo r many commercially and ecologically important species (Irlandi 1996, Jackson et al 2001 Nagelkerken et al. 2002), sinks for both particles (larvae, sedime nt, detritus) and nutrients (Asmus & Asmus 2000), bioindicators of anthropogenic nutri ent inputs (Yamamuro et al. 2003), and sediment stabilizers (Orth 1977). It has al so been suggested that marine macrophyte communities act as a global carbon sink (Smith 1981). Thus, changes to these communities can have a direct impact on pr ocesses as widely varying as the global carbon cycle and coastal erosion. What is becoming apparent is that the complex set of processes that characterize seagrass systems are modified by the template of environmental variables, such as sedime nt properties, seagrass structure, and hydrodynamic conditions (Touchette & Burkhol der 2000, Eyre & Ferguson 2002, Vizzini & Mazzola 2006, Hasegawa et al. 2008) that compose these systems. Many of the environmental variables th at influence seagrass ecosystem function, such as the extent and complexity of seag rass structure and se diment properties in seagrass systems, have been extensivel y studied (Lynts 1966, Kenworthy et al. 1982, Bos

PAGE 15

2 et al. 2007), but often ignored is hydrodyna mic setting. Numerous studies address the interaction between seagrass beds and prevailing hydrodynamic conditions. Specifically, studies have demonstrated seagrasses reduce currents (Almasi et al. 1987, Ackerman & Okubo 1993, Heiss et al. 2000, Madsen et al. 2001), attenuate waves, and dampen turbulence (Koch & Gust 1999) within their canopies. How hydrodynamic conditions modify measures of ecosystem function su ch as sedimentation, are not addressed by many of the studies that inves tigate seagrass-flow relationships in seagrass systems. Due to the lower energy environmen t present within seagrass canopies, seagrasses should act as highly effective traps for sediments and other suspended particles. The ability of seagrass beds to effectively trap sediments has been demonstrated by studies that found fine, organic rich sediments within seagrass be ds when compared to habitats with bare substrate that lack vegetation (Peterson et al. 1984), but is less often described by direct measures of sediment deposition and resuspen sion. The lack of direct measures of sedimentation makes it difficult to assess the influence of flow on depositional dynamics in seagrass systems and how that influence may be modified by changes in seagrass density. It has been suggested that seagrass density can have a measurable effect on local hydrodynamics. Flow speed has been measured to decrease with increases in seagrass shoot density and increasing distance into the bed (Peterson et al. 2004), but most previous studies have not been able to conclusively establish a negative linear relationship between seagrass density and wate r flow (Gambi et al. 1990). This trend holds true under both high and low flow c onditions (Eckman 1987) and for other species of submerged aquatic vegetation (SAV), such as saltmarsh grasses ( Scirpus americanus

PAGE 16

3 Eckman 1983, Spartina alterniflora Leonard & Croft 2006). The interaction between seagrass shoot density and the overall current regime can also have measureable effects on the trapping and retention of particles su spended in the water column by seagrass beds. Most studies have measured increas es in deposition (Gacia et al. 1999) and reductions in resuspension (Hasegawa et al. 2008) in the low energy environment present under increasing seagrass density conditions. Ther efore, it is expected that seagrass beds with higher shoot densities should be the most effective at tr apping and retaining suspended particles from the water column. Additionally, the ability of both high and low density seagrass beds to trap and retain suspended particles should be enhanced in low flows, where the chance of particle deposit ion is greater and resu spension is reduced. Previous studies have investigated sedi mentation across varying flow conditions (Hendricks et al. 2008) and differing seagrass densities (Fonseca & Fisher 1986, Peralta et al. 2008), but one facet of sedimentation in seagrass systems that warrants further investigation is whether equivalent levels of seagrass density provide equivalent levels of particle accumulation under varying hydrodyna mic conditions. Studies such as these would be helpful when establishing target s eagrass densities for restoration efforts across varying physical settings. Faunal communities that utilize seagrass be ds both as a habitat and food source can also be modified by hydrodynamic setting. As with sedimentation, the both the presence (O'Gower & Wacasey 1967, Stone r 1980, Edgar 1990, Edgar et al. 1994) and density (Orth 1973, Homziak et al. 1982, Edga r & Robertson 1992) of seagrass has a positive effect on faunal communities (i.e. increased measures of faunal diversity, abundance, richness, evenness, biomass, and production) that inhabit seagrass systems.

PAGE 17

4 Hydrodynamics can affect faunal abundances directly by altering larval supply and settlement (Eckman 1987, Grizzle et al. 1996, Bologna & Heck 2002), and/or indirectly by modifying sediment characteristics (M urphey & Fonseca 1995, Gambi et al. 1998). The relationship between seagrass density and faunal communities is arguably well studied, but how varying hydrodynamic conditions modify seagrass-faunal relationships has yet to be sufficiently elucidated. More studies that address how flow alters the presence and abundance of faunal species ma y help define how faunal communities are shaped by underlying physical envi ronmental variables. RESEARCH GOALS The overall goal of my doctoral rese arch is to determine the impact of hydrodynamic setting on measures of ecosystem function (i.e. sedimentation and faunal community characteristic) across varying degrees of seagrass and other submerged aquatic vegetation (SAV) structure (i.e. shoot and blade density and length, biomass, leaf area index). In the followi ng chapters a series of in situ experiments focusing on the manipulation of seagrass density will be combined with detailed hydrodynamic characterization, using an Acoustic Doppler Velocimeter (ADV), and an extensive literature review in order to further the current understandin g of the modification of flow by and the effects of flow on particle entrai nment and faunal assemblages in seagrass and other SAV systems.

PAGE 18

5 CHAPTER OBJECTIVES In Chapter 2, I experimentally investigate in situ whether ecosystem function, measured here as the accumulation of particles in a seagrass ( Thalassia testudinum mimics) bed, maintains its relationship to seagrass shoot density across different hydrodynamic regimes. The objectives of Chap ter 2 are to determine if accumulation of particles in seagrass beds of equal shoot densities differs under fast and slow flow conditions. This chapter also seeks to de termine experimentally to what extent hydrodynamic regime and seagrass shoot dens ity can modify the amount and types of particles accumulated in seagrass systems.. In Chapter 3 I further investigate how seagrass loss might impact ecosystem function, linked to depositional processes (her e measured as particle accumulation) in seagrass systems, a series of field experi ments were conducted in which small scale (1 m2) bare and reduced density seagrass patches were created within larger seagrass ( Thalassia testudinum ) beds and subsequent modificat ions to flow and particle accumulation were directly measured. The objectives of this study were to determine: 1) how presence of small scale bare and redu ced density seagrass patches within larger seagrass beds modified flow patterns; 2) if reduced T. testudinum shoot density altered patterns of particle accumulation within experime ntal patches; and 3) if modifications to flow could be used to predict patterns of particle accumulation within experimental patches. In Chapter 4 I evaluate existing in formation on links between hydrodynamic regime and ecosystem function associated with SAV systems. Specifically, the objectives are to: 1) summarize current knowledge regarding effects of SAV on

PAGE 19

6 hydrodynamic characteristics (e.g. velocity, turb ulence, shear stress etc.) and cascading effects on sedimentation and faunal comm unities in SAV systems; 2) summarize relationships recorded betw een sedimentation and the pr esence and/or amount of structural components (e.g. de nsity, canopy height, biomass, etc.) of SAV, and how hydrodynamic conditions can modify that relatio nship; 3) explore th e link between SAV presence and/or amount of structure a nd faunal community characteristics (e.g. abundance, richness, diversity, biomass, et c.) and effects hydrodynami c regime may have on that link; and 4) discuss the importa nce of considering hydrodynamics when exploring measures of ecosystem function, su ch as sedimentation and faunal community characteristics, across flow regimes. SIGNIFICANCE OF RESEARCH By meeting the above objectives, I aim to further our knowledge of depositional processes in seagrass and other SAV system s, which demonstrates the importance of hydrodynamic regime on measures of ecosystem function (i.e. sedi mentation and faunal community characteristics) in these systems. This information can be important to help us better understand both seagrass habitat lo ss and restoration success. An increase in seagrass habitat loss is being witnessed worl dwide (Orth et al. 2006). As the degree of seagrass habitat loss increases dramatic changes to physic al and ecological processes become more pronounced. Previous studies have shown that reductions in seagrass habitat had significantly negative effects on faunal communities (i.e. reductions measures of faunal abundance, richness, evenness, dive rsity, biomass, and productivity) that inhabit seagrass systems (Hughes et al. 2002, Reed & Hovel 2006). In order to mediate seagrass

PAGE 20

7 loss and related loss of ecosystem function, re searchers seek to determine the underlying causes (i.e. eutrophication, reduced light levels, propeller scar ing) and restoration efforts endeavor to successfully re-est ablish these lost habitats. Gaining a better understanding of what environmental factors can modify ecosystem function in seagrass systems can provide important insights into targets, su ch as seagrass patch si ze, shoot density, and replacement ratios, to set for restoration success. In the following chapters my hope is to further the knowledge of seagrass ecosystem function as they relate to the critical issues of seagrass habitat loss and restoration.

PAGE 21

8 Chapter 2 Depositional Processes in Seagrass Beds: Reth inking Density as a Measure of Ecosystem Function INTRODUCTION Loss of critical seagrass habitat has b een reported widely (Duarte 2002). A number of studies have argued that loss of ecosystem function is a likely result of habitat loss. Therefore, focus has been placed seagrass shoot density as a measure of ecosystem function in seagrass systems. Reliance on this metric as a proxy of ecosystem health can be attributed to a positive relationship be tween shoot densities and faunal abundance, including epifauna (Edgar & Robertson 1992, Fonseca et al. 1996, Lee et al. 2001, Bartholomew 2002, Deegan et al. 2002), infa una (Edgar & Robertson 1992, Gambi et al. 1998, Bartholomew 2002, Deegan et al. 2002), and fish (Stoner 1983, Fonseca et al. 1996, Deegan et al. 2002). Likewise, high seagrass densities have been linked to increased quantities of particle accumulation and retention compared to that recorded in areas of low seagrass density (Gacia & Duarte 2001, Widdows et al. 2008). Within dense stands of seagrass, sediment properties, as well as the probability of deposition and resuspension, are altered compared to more sparsely vegetated canopies. Through either direct contact of particles with seagrass blades (Hendricks et al. 2008) or

PAGE 22

9 a result of the lower energy environment pres ent within the canopy, ca pacity for particles to remain in suspension is reduced, resulti ng in accumulation of fine, organic rich sediments (Lynts 1966, Scoffin 1970, Eckman 1983, Neumeier & Ciavola 2004). In addition, the probability of increased depositi on (Bos et al. 2007) and reduced erosion (Widdows et al. 2008) may accompany the lower energy environment. Although sparsely vegetated canopies can also accumula te more fine sediments than unvegetated habitats (Eckman 1987), their ability to affect reductions in current velocity (Sand-Jensen & Mebus 1996) and turbulent stress (Luhar et al. 2008) may not differ. Effectively, the function of sparse stands of vegetation may be reduced to the point where it cannot be differentiated from unvegetated habitats. Although it is clear that vegetation density is linked to particle accumulation and ecosystem function in seagrass systems, the abili ty of the physical set ting to modify these relationships is often not a ddressed. Seagrass canopies reduce currents (Almasi et al. 1987, Ackerman & Okubo 1993, Heiss et al. 2000, Madsen et al. 2001), attenuate waves, and dampen turbulence w ithin their vegetative canopies (Madsen 1983), but modifications to flow by seagrass canopies may not necessarily be consistent across flow regimes. Specifically, current reductions im posed by seagrass canopies are greater when ambient flow increases (Fonseca et al. 1982, Eckman 1987) or possibly non-existent under slower, ambient flow conditions (Heiss et al. 2000). This suggests that effects of low seagrass densities on current reducti on and particle accumulation may only be significant under fast flow conditions, while under slow flow conditions, these effects may not be indistinguishable from measur es of flow or particle accumulation in

PAGE 23

10 unvegetated habitats. Therefor e, understanding of the imp lications of seagrass loss on ecosystem function may be in need of reconsideration. In this study I experimentally investig ate whether ecosystem function, measured here as accumulation of partic les, maintains its relationship to seagrass shoot density across different hydrodynamic regimes. My objectives were to determine if accumulation of particles in seagrass beds of equal shoot densities differs under fast and slow flow conditions. Based on previous studies, I speculate that under fast water flow conditions, high densities of vegetation will accumulate fewer, specifically fine grained, particles when compared to particles accumulated in sparse canopies subject to more sluggish flow conditions. This study also s eeks to determine experimentally to what extent hydrodynamic regime and seagrass shoo t density can modify the amount and types of particles accumulated in seagrass systems. Under fast flow conditions, the probably of resuspension is generally increased, so it is expected that seagrass beds subject to slow flow conditions, regardless of vegetation density, will exhibit higher levels of particle accumulation. The lower energy environment inside canopies of high density seagrass beds, compared to their sparse counterparts should increase the probability of passive particle accumulation in high density beds. METHODS Experimental Design In order to test how particle accumula tion differs between identical seagrass densities subject to different flow regimes, artificial seagra ss units (ASU) were used to simulate Thalassia testudinum (eelgrass) beds of specifi c densities. Two levels of

PAGE 24

11 seagrass density were te sted, high (1500 shoots m-2) and low (300 shoots m-2). Simulated shoots were composed of two blades, each 20 cm long, constructed of ~1 cm wide green polypropylene ribbon and tied to plastic mesh (opening size ~2.45 cm) attached to black window screening 4 m2 (2 m x 2 m) in size and weighted by small fishing weights sewn to the window screening backi ng. The level of seagrass density, number of blades per shoot, and blade height measurements were based on previous morphometric measurements of T. testudinum from Tampa Bay, FL (Meyers, unpublished data). Size of the ASU was selected due to results of previous studies that indicated canopy induced flow reductions are usually minimal beyond 1 m from the leading edge of the bed (Gambi et al. 1990, Peterson et al. 2004, Folkard 2005), as well as the prohibitory nature of transporting and deploying ASU larger than 4 m2 in size. Two levels of flow, fast (0.078 0.041 m s-1) and slow (0.025 0.01 m s-1), were tested at Emerson Point Park located at the mouth of the Manatee River (27.36 N and 82.02 W) ( Figure 1 ). Fast and slow flow sites were separated by <200 m Four experimental treatments were estab lished: slow flow and low density (SL); fast flow and low density (FL); slow flow and high density (SH); and fast flow and high density (FH). For each experimental treatm ent, one replicate ASU was deployed for a duration of seven days. All of the ASUs, regardless of treatment, were deployed 5 m from each other at similar water depths (~1m ) and secured to the sediment with plastic garden stakes (22.5 cm x 2.5 cm). This was re peated for a total of ten replicate paired experiments over a single summ er season (June August 2007).

PAGE 25

Fast Flow Slow F low N 500 m Emerson Pt. ParkTerra Ceia Bay Manatee River Gulf of Mexico Tam p a Ba y Figure 1 Map of lower Tam pa Bay, FL with locations of the fast (0.078 0.041 m s-1) and slow (0.025 0.01 m s-1) experimental flow sites. 12

PAGE 26

13 Particle Accumulation For each of the ten replicate paired experi ments, five particle collectors composed of blue fiberglass air conditioning filter ma terial approximately 1 cm thick and 144 cm2 (12 cm x 12 cm) in size were attached to each ASU. Following the week long deployment, particle collectors were retr ieved before the ASUs themselves were retrieved and redeployed with new pa rticle collectors the same day. Particle collectors used provided a quanti tative estimate of particle flux (g m-2 day-1) to ASUs, referred to in this study as particle accumulation. Specifically, particles retained on collectors mainly reflect meas ure of deposition with minor amounts of resuspension assumed, but not directly measur ed. Probability of particle resuspension was artificially reduced due to adhesive nature of the material that composes collectors and placement of collectors flat on the botto m in region of reduced shear stress. Measures of deposition coupled with direct measures of in canopy flow have seldom been used in similar in situ studies (Gacia et al. 1999, Granata et al. 2001, Hasegawa et al. 2008). To detect any differences in the com position of particles accumulated among the experimental treatments, particle collectors were returned to th e lab and rinsed over sieves and particles divided into sand (63 m) and siltclay (<63 m) size fractions. Both size fractions were drie d to constant weight at 60oC for 24 to 48 h and combusted for organics at 500oC for 4 h. The sand size fraction was also combusted for carbonates at 950C for 2 h. The pre-combusted (500oC for 4 h) Whatman GF/C filters (47 mm diameter; 1.2 m pore size) onto which silt-clay partic le size fractions we re filtered are

PAGE 27

14 only usable at temperatures up to 500C, which prohibited combusting this size fraction at 950C for carbonates. Site Characterization Total suspended solid (TSS) samples, sediment cores, and flow measurements (described in hydrodynamic charact erization section) were also collected from each site. Three TSS sub-samples (1 L) were collected, and later averaged, from the mid-water column at the start and end of each week long re plicate experiment at each flow site. TSS samples were filtered onto pre-combusted (500oC for 4 hours) Whatman GF/F filters (47 mm diameter; 0.7 m pore size), dried to constant weight at 60oC for 24 to 48 hours, then combusted for organics at 500oC for 4 hours. Similar methods have been employed by other studies (Ward et al. 1984, Irlandi & Peterson 1991, Koch 1999, James et al. 2004). Discriminant sampling, as was employed in this study, can be limiting in the sense that collected samples may not accura tely reflect an integrated variable measurement as is provided by continuous samp les. Specifically, fluctuations in TSS concentrations in the water column over th e experimental time period, as a result of storms and other high wind events or even tid al intensity (i.e. spring v. neap tides), may not be reflected in discrimi nant TSS samples. To gain a better understa nding of how concentrations of TSS in the water column fl uctuate, samples collected over the entire experimental time period (e.g. daily TSS sa mpling) would have been desirable. TSS samples represent a measurem ent of the concentration (g L-1) of suspended solids, both sedimentary and organic in na ture, present in the water column. These

PAGE 28

15 samples provide an indication of the amount of particles suspended in the water column overlying the ASU that theoretically could be accumulated. The samples can also indicate if a significant site bias exists. It is important to note that while concentrations of TSS in the water column may be similar be tween two sites, the flux of particles over a defined area can differ greatly. Even when T SS concentrations are similar, the amount of TSS passing over an area per unit time should be greater for sites with higher flow speeds when compared to sites with slower flows. To account for this discrepancy, I calculated particle accumulation efficiencies, (see Data Analysis) which provides a metric to account for differences in TSS particle flux between the sties. Additionally, at the start and end of each replicate experiment, six sediment cores (2.54 cm diameter x 10 cm deep), three in bare sand and three in situ T. testudinum beds, were collected from areas adjacent to the ASU to compare particle size distributions, organic matter, and carbonate content of the sediments in natural settings within fast and slow flow sites. Consistent with sediment grain size analysis methods, sediment cores were rinsed through a series of sieves ( 500, 250, 125, 63, and <63 m), and each particle size fraction was drie d to constant weight at 105oC for 12 to 24 h, combusted for organics at 500oC for 4 h, and combusted for carbonates at 950oC for 2 hours (Heiri et al 2001). Again, the <63 m particle size fractions were unable to be combusted for carbonates because of the melting point of the pre-combusted (500oC for 4 h) Whatman GF/C filters (47 mm diameter; 1.2 m pore size) onto which th ey were filtered.

PAGE 29

16 Data Analysis Particles accumulated on collectors, as well as TSS and sediment samples, were analyzed for percent organic and carbonate content following the loss on ignition (LOI) method, detailed in Heiri et al (2001). Differences in the amounts and types of particles (dry weight, % dry weight, dry weight of orga nic matter, % organic matter, dry weight of carbonates, and % carbonates for each particle sample and size fraction) accumulated in ASU plots with either high or low seagrass shoot densities deployed at fast or slow experimental flow sites were tested for us ing a two factor MANOVA, when assumptions of normality and equality of variances were met (SPSS Statistics 17.0). When significant differences were detected, subsequent pos t-hoc tests (ANOVA) were performed. Each week long ASU deployment (n = 10) was tr eated as a replicate, and within these replicates, the five particle samples per experimental treatment were averaged for analysis. One factor MANOVA was used to test for differences in the amounts (dry weight, g L-1) and organic matter composition (dry weight of organics, g L-1 and % organics) of TSS measured in the water column at the fast and slow experimental flow sites, when assumptions of normality and equality of va riances were met (SPSS Statistics 17.0). Post-hoc tests (ANOVA) were performed when significant differences were detected. To investigate the relationship between particle accumulation and flow, particle accumulation efficiency values for our expe rimental treatments were derived from particle flux rates and measures of particle accumulation in ASUs. Flux rate of particles in the water column over ASU s (PFlux, g m-2 s-1) were calculated as follows:

PAGE 30

TSS A u PFlux* (1) where u is the averaged overlying velocity (m s-1), A (m2) the area over which fluid is flowing (i.e. canopy height*plot length/plot area), and T SS the concentration (g m-3) of TSS in the water column at the time of the ve locity measurement. The greater the flux of flow through ASUs or the greater the concen tration of TSS in the water column, the greater the resulting particle flux rate. Particle accumulation efficiency values (E) were subsequently calculated using the equation: Flux AccumP P E (2) where PAccum (g m-2) is the amount of particles accumulated in the <63 m size fraction in the ASUs at the time of the velocity and TSS measurements. Particle accumulation efficiency values equal to one suggest all of the TSS particles in the water column that fluxed through ASU were captured, while values less than one suggest more TSS particles in the water column fluxed through ASU than particles accumulated in ASU. Values greater than one, a situation which should not occur, suggests more particle were accumulated in ASU than TSS particles in the water column that fluxed through the ASU. Differences in the sediment composition (d ry weight, % dry wei ght, dry weight of organic matter, % organic matter, dry we ight of carbonates, % carbonates for each sediment sample and size fraction) in vegetated (T. testudinum ) and unvegetated (bare sand) habitats present at the fa st and slow experimental flow sites were examined using a 17

PAGE 31

18 two way MANOVA, when assumptions of normali ty and equality of variances were met (SPSS Statistics 17.0). Significant differences were further investigated with post-hoc tests (ANOVA). Hydrodynamic Characterization Velocity profiles characterizing water flow through and above the ASU canopies were measured within each experimental plot at the end of the week long experiment. Profiles were obtained with an Acoustic Doppl er Velocimeter (Field ADV, YSI/Sontek), which can measure velocity components al ong the X, Y, and Z axes. Velocity measurements along each axis are derived from signals scattered by sm all particles within a specific sampling volume. The ADV probe was attached to a vertically adjustable arm extended ~0.4 m from a vertical pole affixed to a flat bottomed weight base, allowing for velocity measurements to be collected at varying heights above the bottom. Profiles for each experimental plot were measured in the center of the plot and included at least six heights above the sediment, with at least f our of the measurements occurring within and at the top of the ASU canopy. Velocity m easurements were collected with the probe facing down and the X-axis element of the pr obe aligned parallel w ith the direction of dominant flow at a sampling rate of 25 Hz and sampling lengths at least 2 minutes in duration to obtain velocity measurements fr om which turbulence could be quantified. Any seagrass blades in direct contact with the ADV sensors were removed for velocity measurements within the seagrass canopy to pr event interference with data collection. Previous studies have shown no significant e ffect of removing a relatively small number of leaves on flow measurements taken in vegetative communities (Ikeda and Kanazawa

PAGE 32

1996). Much of water flow in Tam pa Bay is tidally driven, although wind waves also contribute to the hydrodynamic setting, so all velocity profiles were measured either during mid-incoming or -outgoing tides to ensure that tidal flow was at its peak, thereby representing maximum flows experienced by the sites and experimental plots. All velocity profile data files generated were low pass filtered to remove the high frequency noise. Files were filtered by discarding any samples within the file with signal to noise ratios (SNR) less than 15 and correla tion coefficients, which provide measure of the reliability of and amount of noise in samp les, less than 90. Additionally, any velocity readings greater than 1.5 standa rd deviations from the local velocity mean were filled in with local velocity means, which essentially fl attens out any regions in the data with large spikes. Files with too many samples exceeding the 1.5 standard deviation threshold value were discarded. Using filtered data, mean velocities (m s-1) in X, Y, and Z directions, speed (m s-1), turbulent kinetic energy (m2 s-2), and Reynolds shear stress (Pa) were calculated (Bouma et al. 2007). Bulk flow (m s-1) values represent sp eed of the overlying flow and were measured at mid-water column depth above the canopy of each experimental treatment. Reduction ratios (i .e. reduction in flow speeds due to the presence of the canopy) were calculated using the equation: MAX MIN MAXU UU Ratio Reduction (3) where UMAX is the maximum water column flow speed (m s-1) above canopy and UMIN the minimum flow speed (m s-1) 5 cm above bottom within the ASU canopy. 19

PAGE 33

20RESULTS Hydrodynamic Characterization As expected, bulk flow speeds were greater at the fast (0.078 0.041 m s-1) compared to the slow flow site (0.025 0.01 m s-1) (ANOVA: F1, 17 = 19.755, P < 0.001; Appendix: Table 4; Figure 2 ). At both flow sites, flow speed was reduced more within the canopy of the high (63.40 9.12 %) vers us low (44.61 14.07 %) seagrass density treatment, but reduction in flow within bot h seagrass density treatments was greater under fast (59.54 13.19 %) compared to slow (49.39 15.45 %) flow conditions ( Figure 3 ). Reduction ratios did not appear depe ndent on flow speed, but generally w ere less in the low when compared to the high density treatment. The exception is for flow speeds above 0.08 m s-1, where reduction ratios in both the high and low density treatment appear similar ( Figure 4 ). Profiles of average flow speeds illus trate flow was generally lower both within and above the canopy at the slow flow site in comparison to flow speeds experienced within and above the canopy at the fast flow site. Flow speeds also appear generally lower within the canopy of the high when compared to th e low seagrass density treatment under both fast and slow flow treatment conditions ( Figure 5 ). Measurem ents of turbulent kinetic energy (m2 s-2) within and a bove the canopy of both seagrass density treatments increased as overlying flow speed (m s-1) increased ( Figure 6 ), but when the high (y = 610.34x1.54, R2 = 0.82) was compared to the low (y = 821.85x 1.67, R2 = 0.82) seagrass density treatment the relationship was not found to be density dependent (t0.05(2),101 = -1.16, P = 0.25). Reynolds sh ear stress (Pa) also increased as overlying flow speed increased within and above the ASU canopies (Figure 7 ), but the

PAGE 34

0 0.02 0.04 0.06 0.08 0.1 0.12 FastSlow Flow regimeBulk flow (m s-1) High Low Figure 2 Average ( SEM, n = 6) bulk flow speeds (cm s-1) measured over artificial seagrass unit (ASU) plots with either high (1500 shoots m-2) or low (300 shoots m-2) seagrass shoot densities at fast and slow expe rimental flow sites at Emerson Point Park in lower Tampa Bay, FL. 21

PAGE 35

0 10 20 30 40 50 60 70 80 FastSlow Flow regime% reduction in flow High Low Figure 3 Average ( SEM, n = 6) percent reducti on in flow speeds in the water column above compared to flow speeds within the ca nopy of artificial seag rass unit (ASU) plots with either high (1500 shoots m-2) or low (300 shoots m-2) seagrass shoot densities at fast (0.078 0.041 m s-1) and slow (0.025 0.01 m s-1) experimental flow sites at Emerson Point Park in lower Tampa Bay, FL. 22

PAGE 36

0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 00.020.040.060.080.10.120.140.160.18 Speed (m s-1)Reduction Ratio High Low Figure 4 Relationship between reduction in overl ying flow speed in artificial seagrass unit canopies and overlying flow speed (m s-1) for high (1500 shoot m-2) and low (300 shoots m-2) shoot density ASUs at Emerson Point Park in lower Tampa Bay, FL. 23

PAGE 37

0 0.5 1 1.5 2 2.5 3 3.5 4 00.020.040.060.080.10.120.14 Speed (m s-1)Z/hd Slow High Slow Low Fast High Fast Low canopy Figure 5 Speed profiles (mean SEM, n = 6) from field measures of flow within and above canopy of artificial s eagrass unit (ASU) plots with either high (1500 shoots m-2) or low (300 shoots m-2) shoot densities at fast and slow experimental flow sites at Emerson Point Park in lower Tampa Bay, FL. Profiles were measured for each flow and density combination for a total of four treatments: slow flow and high density; slow flow and low density; fast flow and high density; and fast flow and low density. Profiles shown for heights above the bottom (Z) normalized to deflected canopy height (hd) with canopy height indicated by horiz ontal dashed line (----). 24

PAGE 38

0 10 20 30 40 50 60 70 00.0250.050.0750.10.1250.150.175 Speed (m s-1)Turbulent Kinetic Energy (m2 s-2 104) High Low Figure 6 Relationship between turbulent kinetic energy (m2 s-2) within and above the canopy of artificial seagrass units (ASU) and overlying flow speed (m s-1) for high (1500 shoot m-2) and low (300 shoots m-2) shoot density ASUs at Emerson Point Park in lower Tampa Bay, FL. Best fit lines represented by solid line ( ) for the high and dashed line (----) for the low shoot density treatments. 25

PAGE 39

0 1000 2000 3000 4000 5000 6000 7000 00.0250.050.0750.10.1250.150.175 Speed (m s-1)Reynolds Shear Stress (Pa 104) High Low Figure 7 Relationship between Reynolds shear st ress (Pa) within and above the canopy of artificial seagrass units (ASU ) and overlying flow speed (m s-1) for high (1500 shoot m-2) and low (300 shoots m-2) shoot density ASUs at Emerson Point Park in lower Tampa Bay, FL. Best fit lines represented by solid line ( ) for the high and dashed line (----) for the low shoot density treatments. 26

PAGE 40

27 relationship was not dependent on treatment density as regr ession coefficients did not significantly differ (t0.05(2),101 = 0.48, P = 0.63) between high (y = 51093x1.61, R2 = 0.47) and low (y = 38089x1.63, R2 = 0.58) seagrass shoot densities. Within the canopy (i.e. 5 cm above the bottom or 20% of the canopy height) of the high (y = 17294x 86.14, R2 = 0.72) seagrass density treatment, Reynolds shear stress increased at a faster rate with increasing overlying flow speed (m s-1) and measurements across overlying flow speeds were greater when compared (t0.05(2),15 = 2.25, P = 0.04) to the low (y = 7736.8x 37.31, R2 = 0.67) density treatment ( Figure 8 ). Site Characterization Characterization of fast and slow experi mental study sites provided an indication of both the amount of particles suspended in the water column (TSS) that theoretically could be accumulated in ASU plots and a way to check for any significant site bias. Significant differences in the dry weight (g L-1), dry weight of organic matter (g L-1), and percent organic matter of TSS in the water column were f ound across the replicate weeks (MANOVA: F36, 39 = 4.191, P < 0.001; Appendix: Table 5), but not between flow sites (MANOVA: F3, 22 = 1.548, P = 0.230; Appendix: Figure 5) For the first four replicate weeks of the study, the amount of particles and organic matter suspended in the water column appear to be elevated at both study sites ( Figure 9 ), which may account for significan t differences detected in TSS ch aracteristics across re plicate weeks. To characterize the two flow sites further, sediment samples were taken from vegetated ( Thalassia testudinum ) and unvegetated habitats within each site. When sediment characteristics (dry weight, % dry weight, dry weight of organic matter, %

PAGE 41

-50 0 50 100 150 200 250 300 350 00.0050.010.0150.020.0250.03 Speed (m s-1)Reynolds Shear Stress (Pa 104) High Low Figure 8 Relationship between Reynolds shear st ress (Pa) within the canopy (i.e. 5 cm above bottom or 20% of the canopy height) of artificial seagrass units (ASU) and overlying flow speed (m s-1) for high (1500 shoot m-2) and low (300 shoots m-2) shoot density ASUs at Emerson Point Park in lower Tampa Bay, FL. Best fit lines represented by solid line ( ) for the high and dashed line (----) for the low shoot density treatments. 28

PAGE 42

0 0.005 0.01 0.015 0.02 0.025 0.03 0.035 0.04 01234567891011121314Dry weight of TSS (g L-1) Fast Slow 0 0.002 0.004 0.006 0.008 0.01 0.012 01234567891011121314Organic matter in TSS (g L-1) 0 10 20 30 40 50 60 70 80 01234567891011121314 Experimental week% organic matter in TSS b c a Figure 9 Average ( SD, n = 3) a) dry weight (g L-1) of total suspended solids (TSS), b) dry weight of organic matter in TSS (g L-1), and c) percent organic matter in the TSS, across 12 replicate weeks at fast (0.078 0.041 m s-1) and slow (0.025 0.01 m s-1) experimental flow sites at Emerson Po int Park in lower Tampa Bay, FL. 29

PAGE 43

30 organic matter, dry weight of carbonates, % carbonates for each sediment sample and size fraction) were compared between flow sites and benthic habitats, significant differences were found between the benthic habitats (MANOVA: F12, 1 = 203.964, P = 0.055; Appendix: Table 6), but no significan t interaction was found between the two factors (MANOVA: F12, 1 = 31.693, P = 0.138; Appendix: Table 6) or significant differences in sediment characteri stics between flow sites (MANOVA: F12, 1 = 134.762, P = 0.067; Appendix: Table 6). Between the two benthic habitats, si gnificant differences were determined in the percent dry weight of the smallest (<63 m; ANOVA: F1, 12 = 54.135, P < 0.001; Appendix: Table 7) and largest ( 500 m; ANOVA: F1, 12 = 7.458, P = 0.018; Appendix: Table 7) sediment size fr actions measured, with a greater percentage of both sediment size fractions pr esent in the vegetated habitat ( Figure 10 ). Significantly greater percentages of organic m atter (ANOVA: F1, 12 = 12.417, P = 0.004; Appendix: Table 7) and carbonates (ANOVA: F1, 12 = 18.735, P = 0.001; Appendix: Table 7) were also found in sediments of vegetate d habitats of both flow sites ( Table 1 ). Particle Accumulation When all characteristics ( 13 variables) of particle accumulation measures were tested, significant differences were detected between flow treatments (MANOVA: F10, 22 = 39.992, P < 0.001; Appendix: Table 9). Ho wever, no significant interaction between the factors (MANOVA: F10, 22 = 0.284, P = 0.978) or differences between density treatments (MANOVA: F10, 22 = 0.328, P = 0.964) were detected (Appendix: Table 9). Post-hoc tests indicated that dry weight (g m-2 day-1) of the particle samples (ANOVA: F1, 31 = 11.080, P = 0.002; Appendix: Table 10 ), as well as dry weight (g m-2 day-1) of the

PAGE 44

0% 20% 40% 60% 80% 100% 120% FastSlow Flow regime% dry weight of sediments 500 m 250 m 125 m 63 m <63 m a 0% 20% 40% 60% 80% 100% 120% Bare sandSeagrass Benthic structure% dry weight of sediments 500 m 250 m 125 m 63 m <63 m b Figure 10 Average ( SD, n = 4) percent dry weight by sediment size fraction a) at fast (0.078 0.041 m s-1) and slow (0.025 0.01 m s-1) experimental flow sites and b) in vegetated ( Thalassia testudinum ) and unvegetated (bare sand) benthic habitats within those study sites. Sediment size fracti ons measured included silt-clays (<63 m), very fine sands (63 m), fine sands (125 m), medium sands (250 m), and very coarse sands (500 m). 31

PAGE 45

32 Table 1 Average ( SD) percent dry weight, or ganic matter, and car bonate content by sediment size fraction ( m) in sediments collected from vegetated ( Thalassia testudinum ) and unvegetated (bare sand) benthic ha bitats located w ithin fast (0.078 0.041 m s-1) and slow (0.025 0.01 m s-1) experimental flow sites at Em erson Point Park in lower Tampa Bay, FL. Benthic structure Sediment size fraction (m) Thalassia testudinum Bare sand 500 2.447 ( 1.365) 1.306 ( 0.8757) 250 8.540 ( 2.564) 8.179 ( 3.599) 125 76.95 ( 5.303) 79.84 ( 6.565) 63 10.51 ( 3.947) 10.14 ( 3.251) % dry weight <63 1.549 ( 0.2583) 0.6334 ( 0.2715) Combined 1.074 ( 0.2268) 0.7669 ( 0.4380) 500 0.1034 ( 0.0434) 0.0905 ( 0.0726) 250 0.0794 ( 0.0226) 0.1015 ( 0.0955) 125 0.3082 ( 0.0776) 0.2763 ( 0.1498) 63 0.1965 ( 0.0969) 0.1221 ( 0.0788) % organic matter <63 0.3868 ( 0.0853) 0.1902 ( 0.0861) Combined 0.6168 ( 0.3939) 0.2652 ( 0.0741) 500 0.3981 ( 0.4140) 0.1323 ( 0.0972) 250 0.0333 ( 0.0087) 0.0293 ( 0.0256) 125 0.1085 ( 0.0114) 0.0862 ( 0.0254) % carbonates 63 0.0768 ( 0.0271) 0.0388 ( 0.0212)

PAGE 46

33 63 m size fraction (ANOVA: F1, 31 = 11.820, P = 0.002; Appendix: Table 10) of accumulated particles in the ASU plots at the slow flow site were significantly greater than fast flow site ( Figure 11 ). Dry weights (g m-2 day-1) of particles in the 63 m size fraction accumulated in ASUs at the slow flow site were over two times greater than what was accumulated at the fast flow site. Likewise, significantly greater (g m-2 day-1) organic matter (ANOVA: F1, 31 = 43.676, P < 0.001; Appendix: Table 10) was accum ulated in ASU plots at the slow flow site when compared to fast flow site ( Figure 12 ). This significant difference was driven by 63 m -sized organic particles accumulating under slow flow conditions (ANOVA: F1,31 = 57.060; P < 0.001). Significantly greater carbonate particles of the 63 m size fraction (ANOVA: F1, 31 = 12.073, P = 0.002; Appendix: Table 10) also accumulated at the slow compared to fast flow site ( Figure 13 ). To investiga te the relationship between particle accumulation and flow further, particle accumulation efficiency values were calculated by dividing measured particle accumulation rates by predicted particle accumu lation rates (i.e. flux rate of TSS in the water column). Plots of particle accumula tion efficiency versus flow speed (cm s-1) in the water column above the ASU canopies de monstrate a negative relationship, best described as exponential, as particle accumula tion efficiency values decreased markedly with increasing flow speeds enc ountered under both high (y = 1.2e-0.41x, R2 = 0.86) and low (y = 1.3e-0.38, R2 = 0.77) seagrass density settings (Figure 14 ). No significant difference was detected between reg ression coefficients (t0.05(2), 14 = 0.431, P = 0.673) for the high and low seagrass density treatments The efficiency of particle capture by seagrasses, regardless of vegetation density, was reduced as flow speeds increased.

PAGE 47

0 10 20 30 40 50 60 70 80 90 100 Slow HighSlow LowFast HighFast Low Treatment (Flow Density)Dry weight of trapped particles (g m-2 day-1) 63 m <63 m A B Figure 11 Average ( SEM, n = 10) dry weight (g m-2 day-1) of particles in 63 m and <63 m particle size fractions accumulated in artificial seagrass units (ASU) with either high (1500 shoot m-2) or low (300 shoots m-2) seagrass shoot densities at fast (0.078 0.041 m s-1) and slow (0.025 0.01 m s-1) experimental flow sites located at Emerson Point Park in lower Tampa Bay, FL. Statistically different groupings indicated by post-hoc analysis (ANOVA) are repres ented by upper case lettering (A or B). 34

PAGE 48

0 1 2 3 4 5 6 7 8 Slow HighSlow LowFast HighFast Low Treatment (Flow Density)Dry weight of organic matter (g m-2 day-1) 63 m <63 m A B Figure 12 Average ( SEM, n = 10) dry weight (g m-2 day-1) of organic matter in 63 m and <63 m particle size fractions accumulated in artificial seagrass units (ASU) with either high (1500 shoot m-2) or low (300 shoots m-2) seagrass shoot densities at fast (0.078 0.041 m s-1) and slow (0.025 0.01 m s-1) experimental flow sites located at Emerson Point Park in lower Tampa Bay, FL. Statistically different groupings indicated by post-hoc analysis (ANOVA) are repres ented by upper case lettering (A or B). 35

PAGE 49

0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 Slow HighSlow LowFast HighFast Low Treatment (Flow Density)Dry weight of 63 m carbonates (g m-2 day-1) A B Figure 13 Average ( SEM, n = 10) dry weight of carbonates (g m-2 day-1) in 63 m particle size fraction accumulate d in artificial seagrass uni ts (ASU) with high (1500 shoot m-2) or low (300 shoots m-2) seagrass shoot densities at fast (0.078 0.041 m s-1) and slow (0.025 0.01 m s-1) experimental flow sites located at Emerson Point Park in lower Tampa Bay, FL. Statistically different groupings indicated by post-hoc analysis (ANOVA) are represented by uppe r case lettering (A or B). 36

PAGE 50

0 0.2 0.4 0.6 0.8 1 1.2 1.4 00.0250.050.0750.10.1250.150.175 Speed (m s-1)Particle accumulation efficiency High Low Figure 14 Exponential reductions in particle accumulation efficiency by artificial seagrass unit (ASU) plots with increa sing water column flow speed (m s-1). Artificial seagrass units with either high (1500 shoots m-2) or low (300 shoots m-2) seagrass shoot densities at fast (0.078 0.041 m s-1) and slow (0.025 0.01 cm s-1) experimental flow sites at Emerson Point Park in lower Ta mpa Bay, FL. Best fit lines represented by solid line ( ) for the high and dashed line (----) for the low shoot density treatments. 37

PAGE 51

38DISCUSSION Ecosystem function, measured here as the a ccumulation of particles and organic matter, displayed no difference as seagrass shoot de nsity was increased in either hydrodynamic regime. Levels of particle accumulation, driv en by differences in dry weight, organics, and carbonates of 63 m sized particles, were always gr eater under slow than fast flow conditions. Even the low density treatment at the slow flow site accumulated more particles, again driven by differences in dry weight, organics, and carbonates of 63 m sized particles, than high density treatment at the fast flow site. Thus, comparison of ecosystem function of a SAV based only on ve getation structure may only be valid if flow conditions are identical. This suggests that hydrodynamic setting must be taken into account when considering measures of ecosy stem function, such as deposition, in seagrass systems, and that equivalent levels of seagrass density may not provide similar levels of ecosystem function. Hydrodynamic Characterization Reduced flow speeds imposed by high and low density seagrass beds across fast and slow flow regimes followed expected trends based on past studies (Eckman 1987, Gambi et al. 1990, Worcester 1995, Peterson et al. 2004, Widdows et al. 2008). In contrast to the present study, previous studies have focused on the impact of seagrass on hydrodynamics, often in laboratory flumes, and then predict impacts on some measure of sedimentation (e.g. suspended sediment concen tration, measure of entrainment). This study sought to investigate the impact of seagrass canopy density on flow in situ and directly measure particle accumulation within canopies with different flows. This

PAGE 52

39 approach thus provides an assessment of the linkages of seagrass density, flow and one measure of sediment dynamics in a natural setting. Regardless of canopy density, reduction in overlying flow speed in the ASUs was greater when the fast (59.5 13.2%) was compared to the slow flow treatment (49.4 15.5%) (Appendix: Figure 32), but when flow reductions were considered across overlying flow speeds, and not just treatment categories, reduction ra tios did not change with increasing flow speed ( Figure 4 ). Widdows et al. (2008) al so found that at current velocitie s less than 0.12 m s-1 the amount of flow reduction due to the presence of a seagrass canopy ( Zostera noltii ) was also substantially reduced In this study, while flow reduction did not appear to differ across the range of flow speeds encountered, reduction ratios were generally lower (i.e. less flow reduction) for the low when compared to the high density seagrass treatment ( Figure 3 Figure 4 ). Flow reduction has been measured to be density dependent under flow and dens ity conditions sim ilar to this study (Eckman 1987, Gambi et al. 1990, Peterson et al. 2004), but above 0.08 m s-1 reduction ratios did not appear dependent on density It is possible that sk imming flow developed over the low seagrass density treatments at higher flows speeds, thus reducing the extent to which flow could penetrate the canopy, consequently increasing reduction ratios for the low density treatment. Gambi et al. (1990) al so found skimming flow to develop at flow speeds as low as 0.05 m s-1, but only when seagrass ( Zostera marina) shoot densities exceeded that found in the low seagrass density treatment (1000 vs. 300 shoots m-2). Measures of both turbulent kinetic energy (m2 s-2) and Reynolds shear stresses (Pa) were found to increase as overlying fl ow speed increased, but these relationships were not found to be density dependent ( Figure 6 Figure 7 ). Presence of a canopy

PAGE 53

40 disrupts flow and fluid momentum is extracte d by the canopy in the form of turbulence (Denny 1988). Gambi et al. (1990) suggested th at, in contrast to laminar flows where turbulence can be generated, turbulence is likely rescaled by the presence of a canopy in the case of turbulent flows. This accounts fo r increases in both tu rbulent kinetic energy and Reynolds shear stress, a shear stress resultin g from turbulent velocity fluctuations, as flow speeds increase. The lack of density dependence for measures of turbulent kinetic energy and Reynolds shear stress es indicates a lack of infl uence on flow by the seagrass density treatments, particularly with regards to flow characteristics that are thought to dictate levels of particle de position and resuspension (Grana ta et al. 2001, Widdows et al. 2008). These measures offer an explanation fo r the lack of density effects on particle accumulation by the seagrass density treatments in this study. Reynolds shear stress within the canopy measured at 5 cm above the bottom (20% of canopy height) was found to be significantly greater across all overlying flow speeds encountered by the high compared to th e low seagrass density treatments ( Figure 8 ). W iddows et al. (2008) also found shear stress (derived from turbulent kinetic energy) to increase with increases in current velocity (m s-1) and seagrass ( Z. noltii ) density and to reach a maximum 0.5 cm above the bottom (5 10% of canopy height) at the highest density tested (12600 leaves m-2). As an indicator of vertical mixing and downward fluxes toward the bottom, higher Reynolds sh ear stress in the ca nopy of the high density seagrass treatment would suggest that a co rresponding difference in particle accumulation (e.g. increased accumulation due to greater dow nward flux of particles) between high and low density seagrass treatments should be pr esent. Lack of differences in particle accumulation between the different seagrass density treatments may be attributed to the

PAGE 54

41 lack of density dependence in Reynolds shear stress (i.e. lack of di fference in vertical mixing) at the top of the canopy and in th e water column above the canopy of the seagrass density treatments (Appendix: Tabl e 12). Lack of density effects on flow suggest that the differences obs erved in particle accumulati on between the fast and slow flow treatments are in response to differences in flow between the experime ntal sites. Particle Accumulation Experimental evidence from the present st udy demonstrates that flow was a more important factor than density of seagrass with respect to depositional processes in seagrass beds. The amount of fine particles (63 and <63 m) accumulated was remarkably similar within a flow regime, even with a five fold difference in seagrass shoot density. However, other studies have f ound levels of sediment erosion in seagrass ( Zostera noltii ) beds to be linked to seagrass density when flows exceeded 0.20 m s-1 (Widdows et al. 2008), a speed over 2X the aver age bulk speed of the fast flow treatment in this study. For the flow speeds tested here, canopy density did not have a significant effect on depositional processes, but measurable differences in the amounts and types of particles retained in low compared to high density beds were detected. In general, greater amounts of particles, as well as organic matte r and carbonates, were retained in the low density ASU plots compared to that found in the high density plots. Under the tide dominated or unidirectional flow conditions experienced at both the study sites, currents often cause blades to bend in a single di rection for hours at a time, only to change direction with the turning tide. The overla pping, bent-over canopy can effectively create a barrier between the environments above and within the bed, which in turn can reduce

PAGE 55

42 the amount of mixing between the overlying water column and the bed (Koch & Gust 1999). Reduced mixing between the water column and the bed potentially can reduce the probability of resuspension, but may also reduce the deposition of sediments from the overlying water column as intrusion into the be d is blocked by the ba rrier that the blades create. As seagrass density increases the canopy can more eff ectively act as a barrier to deposition, likely resulting in reduced leve ls of particle accumulation. The unexpected reduced particle accumulation by the high density treatment challenges the convention that high densities of vegetation equate to high levels of ecosystem function in seagrass systems. In low flow environments, the presen ce of a canopy may have little effect on deposition and sediment properties. Irlandi (1996) found a greater amount of both fine particles (<63 m) and organic matter in sediments exposed to slow (0.07 m s-1) compared to fast flow (0.35 m s-1) conditions, but no difference in the amount of fine particles present in vegetated ( Halodule wrightii and Z. marina) versus unvegetated sediments under those slow flow conditions. In a study similar to this one, the amount of mud (<63 m) accumulated in artificial seagrass units was significantly greater than that accumulated in unvegetated areas, but only seas onally when wave energy, and therefore resuspension, was higher (Almasi et al. 1987). Differences in sediment accumulation due to canopy densities may only be important under fast flow speeds as resuspension is often reduced or absent at highly sheltered site s (Gacia et al. 1999). If resuspension only occurs under fast flow conditions, then struct ures that inhibit resuspension should only become important in habitats that experien ce fast flows. This again suggests that vegetation density, or just the presence ve getation, may only become important to

PAGE 56

43 depositional processes when velocity within the canopy reaches or exceeds a critical resuspension threshold, which may only be ach ieved at faster free stream velocities than was witnessed in this study. Implications This study has shown that equivalent densi ties of vegetation structure subjected to different flow conditions did not produce equivalent levels of ecosystem function as reflected in particle retention. This in turn, has important implicati ons for restoration of seagrass systems subject to both natural and human disturbance (Bos & van Katwijk 2007). Restoration efforts often either l ook to reestablish seag rass shoot densities equivalent to natural levels to restore eco system function, or it is assumed similar densities of seagrass will pr ovide equivalent levels of resilience to disturbance. Although, this line of thinking has not alwa ys been supported by evidence (Fonseca et al. 1996). For example, in a restoration effort, transplant survival of seagrass ( Z. marina ) decreased with increasing hydrodynamic exposur e, classified on the basis of sediment grain size, distance to the shore, and duration of hydrodynamic exposure (Bos & van Katwijk 2007). Under high exposure conditions survival was significan tly greater in the high density planting units, while under low hydrodynamic exposure conditions, low (5 plants m-2) and high (14 plants m-2) density planting units had similar survivorship (Bos & van Katwijk 2007). Planting densities that result in successful restoration efforts under slow flow conditions may not be sufficient for survival and persistence when similar densities are planted in a fast flow regime. Likewise, the ecosystem function of sediment retention may also vary for identical densities of seagrass planted in different physical

PAGE 57

44 settings. Therefore, discussions of restoration efforts should cons ider physical setting when establishing replacement ratios (i.e. am ount of habitat restored for every unit area lost) that currently are based on seagrass dens ity alone. Specifically, replacement ratios or amount of seagrass planted may need to be greater if a restorati on project is designed for a site with high compared to lower flow conditions: 1) to account for possible reduced transplant survival; and 2) as shown in this study, to achieve a similar level of particle accumulation. Altered flow conditions due to both natu ral (hurricanes) and human modifications (sand bar movement and dredging) can also cause unexpected changes in ecosystem function provided by seagrasses. Some studies have noted complete removal of seagrass as the result of disturbances, such as hurri canes (van Tussenbroek et al. 2008), but few have assessed how even short-lived altere d flow conditions that might accompany such changes affect ecosystem function of seagrass systems. In a study by Bell et al. (2008), physical disturbance and transport of sediments and associated seagrass ( Halophilia decipiens) seed bank by hurricane-generated distur bance resulted in patches of seagrass appearing in previously barre n areas and other formerly vegetation occupied patches disappearing. Thus over hundreds of meters the spatial distribution of seagrass can be modified. Other studies have shown hurricanes to alter se diment composition in seagrass beds such that coarser sediment sizes and the complete loss of silt-clays may occur (Kalbfleisch & Jones 1998) w ith substantial erosion resul ting. Interestingly these modifications of sediment and topographic feat ures may recover little by three years post disturbance (Fourqurean & Rutten 2004). Ag ain, under severe disturbance by water flow, net sediment distribution and composition may be highly impacted. Here then the

PAGE 58

45 physical setting and links to sedimentary cha nges may have implications for ecosystem function. If flow can influence sediment retenti on in seagrass beds, then measures of ecosystem function of seagrass systems in th e absence of flow, both in laboratory and field settings, particularly with regards to m easures of particle accumulation, may need to be interpreted cautiously. For example, sediments of seagrass beds are characteristically fine grained and organic rich, attributed to lower energy cond itions inside the beds (Orth 1977, Grady 1981, Kenworthy et al. 1982, Peterson et al. 1984). In Mediterranean seagrass systems sedimentary organic matter has been found to be one of the main primary producers transferred in Posidonia oceanica food webs (Vizzini & Mazzola 2006) and changes in the content and bioavail ability of sedimentary organic matter can dictate temporal changes in meiofaunal daily production (i.e. secondary production) (Danovaro et al. 2002). Consequently, reduc tions in sedimentary organic matter resulting from hydrodynamic alterations (e.g. high wind or storms events) may decrease trophic energy transfer and have cascading effects on production by higher tropic levels. In other systems, such as the rocky intertidal, sites that experience different hydrodynamic forces are not considered equivalent ha bitats (Leonard et al. 1998, R obles et al. 2001), but in the soft sediment literature, physical setting is often overlooked when a ddressing questions of ecosystem function. For example, top-down c ontrol in seagrass syst ems often attributed to epiphyte grazers even in presence of botto m-up influences, such as increased nutrient inputs (Jaschinski & Sommer 2008). This type of study does not consider modifications to ecosystem function as hydrodynamic conditions are altered, such as reduced seagrass presence resulting from the physical removal of epiphyte grazers by increased water flow

PAGE 59

46 (Schanz et al. 2002). Other studies unsuccessfully try to make predictions concerning presence and abundance of faunal species using metrics such as seagrass density (Worthington et al. 1992) without considering influence of flow on larval recruitment (Eckman 1987) or deposition (Bol ogna & Heck 2002). The re sults of this study suggests that those ecosystem services provided by seagrasses (e.g. water quality, costal protection) dependent on their ability to reduce currents, attenuate waves, and decrease turbulence, may be reduced under faster ener gy conditions regardless of the prevailing vegetation density. As flow diminishes th e import of seagrasses to these ecosystem services (e.g. water quality, co stal protection) may also diminish as particle flux (i.e. amount of particles in the wate r column that pass over an area per unit time) decreases (Granata et al. 2001) and resuspension probabilities are reduced (Har lin et al. 1982). Although seagrass density has been designa ted as a major factor underlying the ecosystem function of seagrass systems (Orth 1973, Homziak et al. 1982, Bos et al. 2007, Widdows et al. 2008), the role that vegetation plays must be considered within the context of physical setting. Integrati on of hydrodynamic measurements into studies assessing seagrass ecosystem services provide a more complete view of the complex interactions between underlying environmental variables th at dictate levels of ecosystem processes in these important coastal habitats.

PAGE 60

47 Chapter 3 Effects of Patchy Habitat Structure on Deposition Processes in Seagrass, Thalassia testudinum Systems of Lower Tampa Bay, FL INTRODUCTION The ability to trap and bind suspended pa rticles has long been recognized as one of the important functions of seagrass ecosy stems. Reduction in water column turbidity increases light levels, which in turn enhances both the prim ary and overall productivity of these ecosystems, and the tendency of seagra sses to retain sediments and other trapped particles help to reduce coastal erosion risks (Koch et al. 2009). These important ecosystem services, among others, provided by seagrasses and algal beds have been estimated to be valued at upwar ds of 3.8 trillion dollars per year (Costanza et al. 1997). It is clear that the removal or reduction of seagrass ecosyste ms causes substantial erosion (Wilson 1949), which in turn could be detrimental both ecologically and economically, but the extent of the negative consequences is unclear. Removal, reduction, and fragmentation of seagrass systems can occur via natural disturbances, such as waves, currents, and bioturbation (Townsend & Fonseca 1998), both on large and small, more local scales. Increasingly though these systems are being modified by anthropogenic disturbances rang ing from coastal development, propeller

PAGE 61

48 scarring (Bell et al. 2002), anchor usage (Creed & Amado Filho 1999, Francour et al. 1999), fishing practices (Neckles et al. 2005) trampling (Eckrich & Holmquist 2000), and plant harvesting. Disturbances and fragmentation, regardless of cause, generally result in some form of seagrass loss (i .e. complete removal or reduced density) and modification to cover and spatial configurati on. Most previous st udies have addressed questions of modified cover and spatia l configuration due to disturbance and fragmentation by investigating measures of seagrass ecosystem function at larger scales (e.g. large v. small patches, continuous v. fragmented landscapes) (Frost et al. 1999, Bowden et al. 2001, Hirst & Attrill 2008). Smaller scale disturbances, such as anchor usage, may only affect areas <1 m2 in size within larger seagrass bed and result in reduced shoot densities, but not complete seagrass removal (Francour et al. 1999). Effects of small scale disturba nces within larger seagrass beds can vary greatly from disturbances that cause fragmentation on a la ndscape scale. For example, studies testing effects of small scale reduc tions in seagrass density (E dgar & Robertson 1992) or creation of gaps (Reed & Hovel 2006) within larger seagrass beds on faunal communities found decreased faunal abundances and altered faunal assemblages with greater seagrass removal, while Johnson & Heck (2006) found fragmentation at the scale of 1 to 100 m2 sized patches may have little impact on faunal community assemblages. Although presence of gaps and reduced density patche s in larger seagrass beds have been well documented for faunal community measur es, less is known about how hydrodynamic characteristics and depositional processes are altered within these systems. Flow across and through small scale gaps in seagrass beds have been previously investigated, but no attempts have been made to capture in situ measures of flow in

PAGE 62

49 reduced density patches located within larger, continuous seagrass beds. Past flume studies addressing flow modi fications in small scale gaps and patches found the formation of flow recirculation cells and st agnant regions of flow within the gaps (Maltese et al. 2007) and incr eased turbulence at the canopy water interface downstream of gaps (Folkard 2005). Within continuous be ds of seagrass and other submerged aquatic vegetation (SAV), higher vegetation densities effect greater current reductions (Madsen & Warnke 1983, Leonard & Luther 1995, Widdow s & Brinsley 2002), increased wave attenuation (Koch & Gust 1999, Chen et al. 2007 ), and decreases in turbulence (Leonard & Luther 1995, Koch & Gust 1999, Luhar et al. 2008), when compared to more sparsely vegetated canopies. Specifically, some st udies have found up to a 40 % reduction in near-bed flow in dense SAV canopies compar ed to their sparser counterparts (Widdows et al. 2008). Whether or not similar flow modifications (i.e. greater current speeds, reductions in wave attenuation, and increas ed turbulence) occur within canopies of reduced density patches located within larger, continuous s eagrass beds has yet to be tested. Reduced attenuation of flow within lowe r density seagrass canopies is often linked to decreases in particle deposition and increases in acc umulation of coarse, organic poor sediments relative to higher density canopies (Lynts 1966, Scoffin 1970, Eckman 1983, Eckman 1987, Neumeier & Ciavola 2004), but this has not been tested for smaller patches within larger seagrass beds. Als o, few studies have address the effects of seagrass loss on depositional dynamics with in seagrass systems. Anecdotally Wilson (1949) noted a large scale die off of Zostera marina along the Atlantic coasts of Europe in the early 1930s resulted in substantial sediment erosi on, and Marshall & Lukas (1970)

PAGE 63

50 found no shift in sediment characteristics one month following the experimental removal of seagrass ( Zostera marina) from ~200 m2 plots. Thus, further e xperimental studies that address small scale seagrass reduction and loss are needed to assess implications of local disturbances and fragmentation on larger se agrass ecosystems and services. In order to further investigate how seag rass loss might impact ecosystem function, linked to depositional processes (here measured as particle accumulation) in seagrass systems, a series of field experiments were conducted in which small scale (1 m2) bare and reduced density seagrass patches we re created within larger seagrass ( Thalassia testudinum ) beds and subsequent modifications to flow and particle accumulation were directly measured. The objectives of this study were to determine: 1) how presence of small scale bare and reduced density seagra ss patches within larger seagrass beds modified flow patterns; 2) if reduced T. testudinum shoot density altered patterns of particle accumulation within experimental patche s; and 3) if modifications to flow could be used to predict patterns of particle accu mulation within experimental patches. Given previous studies in continuous seagrass beds, it was predicted that a full density canopy would attenuate flow to a grea ter extent within the canopy than patches of seagrass with reduced or removed shoot densities. Along w ith predicted reductions in flow attenuation, it was predicted levels of particle accumul ation would also be reduced in bare and reduced seagrass density patches in comp arison to a full density seagrass bed.

PAGE 64

51 METHODS Experimental Design In order to test the effect s of bare and reduced density seagrass patches within a larger, continuous seagrass bed on with in canopy hydrodynamics and depositional processes, experimental patche s with shoots densiti es ranging from full density to bare were created in seagrass, Thalassia testudinum beds. Four density tr eatments (full, 50%, 10%, and bare) were create d from naturally existing T. testudinum beds by removing specific shoots percentages. Accordingly, for the full density treatment no shoots were removed, half of the shoots were removed fo r the 50% treatment, 90% of the original shoot density was removed for the 10% treatment, and all of the shoots were removed from the patches of the bare treatment. Shoots and all above ground biomass were removed by hand in all shoot removal treatment s, and full patches with no shoot removal were likewise disturbed by simu lating blade removal. The study was conducted at two site s, North Skyway (27.11N, 82.57W) and East Beach (2738.77N, 82.70W), in lower Tampa Bay, FL near Ft. DeSoto Park ( Figure 1 ). The sites were separated by ~2 km and were characterized by continuous, m onospecific T. testudinum beds in ~1 m water depth that experienced at daily tidal range of ~0.5 m. Average ( SD) shoot densities were 449.1 87.53 shoots m-2 at the North Skyway site and 481.6 93.31 shoots m-2 at the East Beach site. At each study site, two experimental patches 1m2 (1m x 1m) in size were created for each density treatment, for a total of eight experimental patches created per site. The size of the patches was chosen based upon the results of a pilot study, which suggested

PAGE 65

East Beach 1 km N North Skyway Gulf of Mexico Tampa Bay St. Petersburg Ft. DeSoto Gulf of Mexico Tampa Bay Figure 15 Location of study sites, North s kyway (27.11N, 82.57W) and East Beach (27.77N, 8241.70W), in lower Tampa Bay, FL. 52

PAGE 66

53 that increases in within canopy flow speeds due to reductions in seagrass shoot density could be detected within patches of these dimensions (i.e. 1 m x 1 m). Patches were placed 5 m from the leading edge of larger, continuous seagrass beds, 5 m apart from each other within the seagrass beds, and were marked by PVC poles at each of the four corners of the patch. The study was conducted at the North Skyway site from July 9th to August 13th, 2008 and from July 23rd to July 30th and September 18th to October 16th, 2008 at the East Beach site. Timing of experiments was limite d by weather conditions. Over the 9 week study period, week long experiments were c onducted, for a total of five replicate experiments per seagrass density treatment at each of the stud y sites. Results of a pilot study indicated particle accu mulation over a week long period provided a sufficient sample size to detect differences in pa rticle accumulation between seagrass density treatments. Particle Accumulation To provide quantitative measures of sedi ment accumulation in seagrass habitats, particle collectors, 144 cm2 (12 cm x 12 cm) in size and composed of 1 cm thick blue fiberglass air conditioning filter material, were placed in the center of each experimental patch, flush with the sediment surface and secu red by garden stakes. Particle collectors were retrieved at the end of each week long experiment. Particle collectors used provided a quanti tative estimate of particle flux (g m-2 day-1) to experimental patches, referred to in this study as particle accumulation. Specifically, particles retained on collectors mainly reflect measure of deposition with

PAGE 67

54 minor amounts of resuspension assumed, but not directly measured. Probability of particle resuspension was artifici ally reduced due to adhesive nature of the material that composes collectors and placement of collector s flat on the bottom in region of reduced shear stress. Measures of deposition coupled with direct measures of in canopy flow have seldom been used in similar in situ studies (Gacia et al. 1999, Granata et al. 2001, Hasegawa et al. 2008). To detect differences in the amounts and composition of the particles that were accumulated among seagrass density treatments, co llected particles were first rinsed over a series of sieves and divided into large ( 125 m), medium (125 to 63 m), and small ( 63 m) size fractions. All the size fractions were dried to a constant weight at 60C for 24 to 48 h and combusted for organic content at 500C for 4 h. Large and medium size fractions were also combusted fo r carbonate content at 950C for 2 h. The Whatman GF/C filters (47 mm diameter; 1.2 m pore size) onto which sma ll particle size fractions were filtered are only usable at temperatur es up to 500C, which prohibited combusting this size fraction at 950C for carbonates. Total Suspended Solids In addition to the particle samples, at the start and end of each week long experiment, total suspended solid (TSS) sample s were collected from the water column. Three 1 L TSS sub-samples, that were later averaged, were collected each week from mid-water depth. The TSS samples were filtered onto pre-combusted (500C for 4 h) Whatman GF/F filters (47 mm diameter; 0.7 m pore size), dried to a constant weight at 60C for 24 to 48 h and combusted for organics at 500C for 4 h. Similar methods have

PAGE 68

55 been employed by other studies (Ward et al. 1984, Irlandi & Peterson 1991, Koch 1999, James et al. 2004). Discriminant sampling, as was employed in this study, can be limiting in the sense that collected samples may not accura tely reflect an integrated variable measurement as is provided by continuous samp les. Specifically, fluctuations in TSS concentrations in the water column over th e experimental time period, as a result of storms and other high wind events or even tid al intensity (i.e. spring v. neap tides), may not reflected in discriminant TSS samples. To gain a better understanding of how concentrations of TSS in the water column fluctuate samples collected over the entire experimental time period (e.g. daily TSS sa mpling) would have been desirable. TSS samples represent a measurem ent of the concentration (g L-1) of suspended solids, both sedimentary and organic in na ture, present in the water column. These samples provide an indication of the amount of particles suspended in the water column overlying the experimental patches that theoretically could be accumulated and to check for significant site bias. It is important to note that while concentrations of TSS in the water column may be similar between two site s the flux of particles over a defined area can differ greatly. Even when TSS concentra tions are similar, the amount of TSS passing over an area per unit time should be greater for sites with higher flow speeds when compared to sites with slower flows. Data Analysis Particles accumulated on collectors, as well as TSS and sediment samples, were analyzed for percent organic and carbonate content following the loss on ignition (LOI)

PAGE 69

56 method detailed in Heiri et al (2001). Differences in the am ounts and types of particles (dry weight, % dry weight, dry weight of orga nic matter, % organic matter, dry weight of carbonate content, and % carbonate content for each particle sample and size fraction) accumulated in the experimental density treatments, both between sites and among density treatments, were tested for usi ng a two factor MANOVA (21 variables), when assumptions of normality and equality of va riances were met (SPSS Statistics 17.0). When significant differences were detected, subsequent post-hoc tests (ANOVA) were preformed. Each week long experiment (n = 5) was treated as a re plicate, and within these replicates, particle accumulation meas ures for duplicate experimental patches created were averaged for analysis. One factor MANOVA was used to test for differences in the amounts (dry weight, g L-1) and organic matter composition (dry weight of organics, g L-1 and % organics) of TSS measured in the water column at the experimental study sites (North Skyway and East Beach), when assumptions of normality and equality of variances were met (SPSS Statistics 17.0). Post-hoc tests (ANOVA) were performed wh en significant differences were detected. Hydrodynamic Characterization Velocity profiles characterizing water flow through and above the canopies of the experimental density treatments were measured within each experimental patch at the end of the week long experiment. Profiles were obtained with an Acoustic Doppler Velocimeter (ADV; Nortek Field Vector), which can measure velocity components along the X, Y, and Z axes. Velocity measurements along each axis are derived from signals

PAGE 70

57 scattered by small particles within a sp ecific sampling volume. The ADV probe was attached to a vertically adjustable arm exte nded ~0.4 m from a vertical pole affixed to a flat bottomed weight base, allowing for velocity measurements to be collected at varying heights above the bottom. Profiles for each experimental patch were measured in the center of the patch and included at least six he ights above the sediment with at least four of the measurements occurring within and at the top of the canopies of the experimental patches. Velocity measurements were coll ected with the probe facing down and the Xaxis element of the probe aligned parallel w ith the direction of dominant flow at a sampling rate of 32 Hz and sampling lengths at least 2 minutes in duration to obtain velocity measurements from which turbulence could be quantified. Any seagrass blades in direct contact with the ADV sensors were removed for velocity measurements within the seagrass canopy to prevent in terference with data collectio n. Previous studies have shown no significant effect of removing a re latively small number of leaves on flow measurements taken in vegetative communities (Ikeda and Kanazawa 1996). Much of water flow in Tampa Bay is tidally driven, although wind waves also contribute to the hydrodynamic setting, so all veloci ty profiles were measured either during mid-incoming or -outgoing tides to ensure that tidal flow was at its pe ak, thereby representing maximum flows experienced by the sites and experimental patches. All velocity profile data files generated were low pass filtered to remove the high frequency noise. Files were filtered by discarding any samples within the file with signal to noise ratios (SNR) less than 15 and correla tion coefficients, which provide measure of the reliability of and amount of noise in samples, less than 90. Additionally, any velocity readings greater than 1.5 standa rd deviations from the local ve locity mean were filled in

PAGE 71

58 with local velocity means, which essentially fl attens out any regions in the data with large spikes. Files with too many samples exceedi ng the 1.5 standard deviation threshold value were discarded. Using filtered data, mean velocities (m s-1) in X, Y, and Z directions, speed (m s-1), turbulent kinetic energy (m2 s-2), and Reynolds shear stress (Pa) were calculated (Bouma et al. 2007). Bulk flow (m s-1) values represent sp eed of the overlying flow and were measured at mid-water column depth above the canopy of each experimental treatment. RESULTS Hydrodynamic Characterization At both experimental sites, North Skyw ay and East Beach, overall flow was sluggish with maximum flow speeds not exceeding 0.08 m s-1 (Figure 16). Statistical testing indicated no significant interaction be tween study site and e xperimental seagrass density on bulk flow values (ANOVA: F3, 31 = 0.956, P = 0.426, Appendix: Table 13). Additionally, no significant differences in bulk flow values were detected either between the two study sites (ANOVA: F1, 31 = 3.264, P = 0.081; Appendix: Table 13) or among the four experimental seagra ss density treatments (ANOVA: F3, 31 = 0.428, P = 0.735; Appendix: Table 13). No significant reduc tion in flow speed within the canopy was found either among seagrass density treatments (F3,32 = 0.65, P = 0.59) or between study sites (F1,34 = 1.46, P = 0.23). Profiles of average flow speeds illust rate flow both within and above the canopies of the seagrass density treatment patc hes were similar, suggesting that density had little effect on flow (Figure17). When pr ofiles within and above the canopies of full

PAGE 72

0 0.01 0.02 0.03 0.04 0.05 0.06 0.07 Bare10%50% Full Experimental density treatmentSpeed (m s-1) North Skyway East Beach Figure 16 Average ( SEM, n = 6) bulk flow speeds (m s-1) measured above experimentally thinned patches of seagrass ( Thalassia testudinum ) with shoot densities ranging from full density (full), half of the original shoot density (50%), 10% of the original density (10%), and complete shoot removal (bare) at two study sites ( North Skyway and East Beach) in Tampa Bay, FL. 59

PAGE 73

60 shoot density beds and bare patches were co mpared on days with maximum flow speeds, no attenuation of flow was measured either within or above the full density canopy (Figure 18a). In contrast, on days with minimum flow speeds, there is some flow attenuation within the full seagrass density canopies (Figure18b), suggesting that the presence of a canopy had an effect on flow under slow flow conditions (<0.02 m s-1). Measurements of turbulent kinetic energy (m2 s-2) within and a bove the canopy of the seagrass density treatments incr eased as overlying flow speed (m s-1) increased (Figure 19), but comparison of regressi on coefficients for the full (y = 0.19x1.79, R2 = 0.912), 50% (y = 0.17x1.81, R2 = 0.90), 10% (y = 0.13x1.73, R2 = 0.91), and Bare (y = 0.31x1.96, R2 = 0.92) seagrass density treatments f ound the relationship not to be density dependent (ANOVA: F3,189 = 0.51, P = 0.68). Within the canopy 5 cm above the bottom (i.e. 20% of the canopy height) of the fu ll seagrass density treatment (y = 0.20x1.72, R2 = 0.88), turbulent kinetic energy increased at a slower rate with in creasing overlying flow speed (m s-1) when compared (ANOVA: F3,28 = 3.34, P = 0.03) to the 50% (y = 1.58x2.24, R2 = 0.95), 10% (y = 1.12x2.13, R2 = 0.99), and Bare (y = 2.25x2.36, R2 = 0.98) seagrass density treatments (Figure 20). Reynolds shear stress (Pa) also increa sed as overlying flow speed increased within and above the canopies of the seagrass density treatments (Figure 21). The relationship was not dependent on treatment density as regr ession coefficients did not significantly differ (ANOVA: F3,189 = 0.64, P = 0.589) among full (y = 4.49x1.70, R2 = 0.64), 50% (y = 4.77x1.80, R2 = 0.67), 10% (y = 5.46x1.77, R2 = 0.58), and Bare (y = 9.44x1.93, R2 = 0.75) seagrass density treatments.

PAGE 74

0 0.5 1 1.5 2 2.5 00.0150.030.0450.060.075 Speed (m s-1)Z/hd Bare 10% 50% Full Canopy Figure 17 Speed profiles (m s-1) from field measures of fl ow within and above canopies of experimentally thinned patches of seagrass ( Thalassia testudinum ) with shoot densities ranging from full density (Full), half of the original shoot dens ity (50%), 10% of the original density (10%), and complete shoot removal (Bare) and averaged ( SD, n = 6) across two study sites (North Skyway and East Beach) in Tampa Bay, FL. Profiles shown for heights above the bottom (Z) normalized to deflected canopy height (hd) with canopy height indicated by horiz ontal dashed line (----). 61

PAGE 75

62 0 0.5 1 1.5 2 2.5 00.20.40.60.811.2 U/UmaximumZ/hd Bare Full a Canopy 0 0.5 1 1.5 2 2.5 00.20.40.60.811.2 U/UmaximumZ/hd Bare Full b Canopy Figure 18 Norm alized speed profiles (U/Umaximum) from field measures of flow within and above the canopy of experimenta lly thinned patches of seagrass ( Thalassia testudinum ) on days with a) maximum ( 0.04 m s-1) and b) minimum ( 0.02 m s-1) flows. Experimental shoot densities included full shoot density (Full) or complete shoot removal (Bare). Profiles shown for heights above the bottom (Z) nor malized to deflected canopy height (hd) with canopy height indicated by horizontal dashed line (----).

PAGE 76

0 0.0005 0.001 0.0015 0.002 0.0025 00.0150.030.0450.060.0750.090.105 Speed (m s-1)Turbulent kinetic energy (m2 s-2) Full 50% 10% Bare Figure 19 Relationship between tur bulent kinetic energy (m2 s-2) within and above the canopy of experimentally thinned patches of seagrass ( Thalassia testudinum ) and overlying flow speed (m s-1) for shoot densities ranging from full density (Full), half of the original shoot de nsity (50%), 10% of the or iginal density (10%), ad complete shoot removal (Bare) at two study sites (North Skyway and East Beach) in Tampa Bay, FL. n 63

PAGE 77

0 0.0001 0.0002 0.0003 0.0004 0.0005 0.0006 0.0007 0.0008 00.0050.010.0150.020.0250.030.035 Speed (m s-1)Turbulent Kinetic Energy (m2 s-2) Bare 10% 50% Full Figure 20 Relationship between tur bulent kinetic energy (m2 s-2) within the canopy (i.e. 5 cm above bottom or 20% of the canopy height ) of experimentally thinned patches of seagrass ( Thalassia testudinum ) and overlying flow speed (m s-1) for shoot densities ranging from full density (Full), half of the original shoot dens ity (50%), 10% of the original density (10%), and complete shoot removal (Bare) at two study sites (North Skyway and East Beach) in Tampa Bay, FL. 64

PAGE 78

0 0.05 0.1 0.15 0.2 0.25 00.0150.030.0450.060.0750.090.105 Speed (m s-1)Reynolds shear stress (Pa) Full 50% 10% Bare Figure 21 Relationship between Reynolds shear stre ss (Pa) within and above the canopy of experimentally thinned patches of seagrass ( Thalassia testudinum ) and overlying flow speed (m s-1) for shoot densities ranging from full density (Full), half of the original shoot density (50%), 10% of the original density (10%), and complete shoot removal (Bare) at two study sites (North S kyway and East Beach) in Tampa Bay, FL. 65

PAGE 79

66 Total Suspended Solids Significant differences in the dry weight (g L-1), dry weight of organic matter (g L-1), and percent organic matter of TSS in the water column was found between the two study sties (MANOVA: F3, 34 = 10.868, P < 0.001; Appendix: Ta ble 14). Differences were also found across replicate week s at the North Skyway (MANOVA: F15, 36 = 3.101, P = 0.003; Appendix: Table 14) and East Beach (MANOVA: F18, 39 = 5.268, P < 0.001; Appendix: Table 14) sites ( Figure 22; Appendix: Table 14). Partic le Accumulation When all characteristics (21 variables) of accumulated particle measures were considered, significant differences both between the two study sites (MANOVA: F21, 47 = 33.856, P < 0.001; Appendix: Table 18) and among the four seagrass density treatments (MANOVA: F63, 147 = 1.492, P = 0.026; Appendix: Table 18 ) were detected. However, no significant interaction between factors was detected (MANOVA: F63, 147 = 1.272, P = 0.169; Appendix: Table 18). Significant differences were f ound in the dry weight (g m-2 day-1) accumulation of particles among seagrass dens ity treatments (ANOVA: F3,67 = 5.318, P <0.001; Appendix: Table 19). Generally, decreased shoot density was accompanied with increased dry weight particle accumulation (Figure 23). Post-hoc testing (Tukey B) indicated dry weight particle accumulation to be significantly less by the full and half (50%) seagrass density treatments in comparison to the bare treatment (Figure 23). These differences were driven by changes in the dry weight accumulation of 125 m-sized particles (ANOVA: F3, 67 = 4.813, P = 0.004; Appendix: Table 19, Figure 24). This is not

PAGE 80

0 0.005 0.01 0.015 0.02 0.025 0.03 0123456Dry weight of TSS (g L-1) 7 North Skyway East Beach a 0 0.001 0.002 0.003 0.004 0.005 0.006 0.007 0.008 0123456Organic matter in TSS (g L-1) 7 b 67 0 5 10 15 20 25 30 35 40 45 01234567 Replicate weeks% Organic matter in TSS c Figure 22 Average ( SD, n = 3) a) dry weight (g L-1) of total suspended solids (TSS), b) dry weight of organic matter in TSS (g L-1), and c) percent organic matter in TSS, across seven replicate weeks at two experi mental study sites ( North Skyway and East Beach) in Tampa Bay, FL.

PAGE 81

0 50 100 150 200 250 300 350 400 450 Bare 10% 50% Full Experimental density treatmentDry weight of accumulated particles (g m-2 day-1) A AB B B Figure 23 Average ( SEM, n = 10) dry weight (g m-2 day-1) of particles accumulated by experimentally thinned patches of seagrass ( Thalassia testudinum ) with shoot densities ranging from full density (Full), half of the original shoot density (50%), 10% of the original density (10%), and complete shoot removal (Bare) at tw o study sites (North Skyway and East Beach) in Tampa Bay, FL. Values for each experimental treatment were averaged across the study sites. Statistically differe nt groupings indicated by posthoc analysis (Tukey B) are represen ted by upper case lettering (A or B). 68

PAGE 82

0 50 100 150 200 250 300 <6363 125 Particle size fraction ( m)Dry weight of accumulated particles (g m-2 day-1) Bare 10% 50% Full a 0 50 100 150 200 250 300 350 <6363 125 Particle size fraction ( m)Dry weight of accumulated particles (g m-2 day-1) Bare 10% 50% Full b Figure 24 Average ( S EM, n = 5) dry weight (g m-2 day-1) of particles by particle size fraction ( m) accumulated in experimentally thinned patches of seagrass (Thalassia testudinum ) at a) North Skyway and b) East Be ach study sties. Experimental shoot densities ranged from full density (Full), half of the original shoot density (50%), 10% of the original density (10%), and co mplete shoot removal (B are) created at two study sites in Tampa Bay, FL. 69

PAGE 83

70 unexpected, as most of the particles accumu lated (>50%) by the experimental patches, across all seagrass density treatments, were in the largest particle size fraction ( 125 m) (Figure 25). ANOVA revealed significant differences in percent organic matter composition of particles accumulated among the density treatment (ANOVA: F3, 67 = 15.287, P < 0.001; Appendix: Table 19). Particles accumulated in the full and half (50%) seagrass density treatment beds had higher organic matter com position in than the lower seagrass density treatments (Figure 26). Differences in organic matter composition between seagrass density treatments were not driven by change s in a single particle size class. Particles accumulated by patches with reduced in seagrass density generally had lower percent organic matter composition in the <63 m (ANOVA: F3, 67 = 8.973, P <0.001), 63 m (ANOVA: F3, 67 = 12.211, P <0.001), and 125 m (ANOVA: F3, 67 = 13.657, P <0.001) particle size fractions (Appe ndix: Table 19, Figure 27). Percent carbonate composition of part icles accumulated were significantly different between seagrass de nsity treatments (ANOVA: F3, 67 = 3.697, P = 0.016; Appendix: Table 19). Post-hoc testing (Tukey B) indicated pa rticles accumulated in bare treatment patches had significantly lowe r percent carbonate composition than full seagrass density patches (Figure 28). Although there was a general decrease in the percent carbonate composition of accumulate d particles with redu ced seagrass density (Figure 29), significant differences were driven by changes in percent carbonate composition of particles accumulated in the 63 particle size fraction (ANOVA: F3, 67 = 8.576, P < 0.001; Appendix: Table 19).

PAGE 84

0 10 20 30 40 50 60 70 80 <6363 125 Particle size fraction ( m)% dry weight of accumulated particles Bare 10% 50% Full a 0 10 20 30 40 50 60 70 80 90 <6363 125 Particle size fraction ( m)% dry weight of accumulated particles Bare 10% 50% Full b Figure 25 Average ( S EM, n = 5) percent dr y weight by particle size fraction ( m) of particles accumulated in experimentally thinned patches of seagrass (Thalassia testudinum ) at a) North Skyway and b) East Be ach study sties. Experimental shoot densities ranged from full density (Full), half of the original shoot density (50%), 10% of the original density (10%), and co mplete shoot removal (B are) created at two study sites in Tampa Bay, FL. 71

PAGE 85

0 2 4 6 8 10 12 Bare10%50%Full Experimental density treatment% organic matter in accumulated particles B B A A Figure 26 Average ( S EM, n = 10) percent orga nic matter in particles accumulated by experimentally thinned patches of seagrass ( Thalassia testudinum ) with shoot densities ranging from full density (Full), half of the original shoot density (50%), 10% of the original density ( 10%), and complete shoot removal (Bare) crea ted at two study sites (North Skyway and East Beach) in Tampa Bay, FL. Values for each experimental treatment were averaged across the study sites. Statistically different groupings indicated by post-hoc analysis (Tukey B) are represen ted by upper case lettering (A or B). 72

PAGE 86

0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5 <6363 125 Particle size fraction ( m)% organic matter in accumulated particles Bare 10% 50% Full a 0 1 2 3 4 5 6 7 8 <6363 125 Particle size fraction ( m)% organic matter in accumulated particles Bare 10% 50% Full b Figure 27 Average ( S EM, n = 5) percent orga nic matter composition by particle size fraction ( m) of particles accumulated in experimentally thinned patches of seagrass ( Thalassia testudinum ) at a) North Skyway and b) East Beach study sties. Experimental shoot densities ranged from full density (Fu ll), half of the original shoot density (50%), 10% of the original density (10 %), and complete shoot removal (Bare) created at two study sites in Tampa Bay, FL. 73

PAGE 87

0 1 2 3 4 5 6 Bare10%50%Full Experimental density treatment% carbonates in accumulated particles B AB AB A Figure 28 Average ( SEM, n = 10) percent carbona tes in the particles accumulated by experimentally thinned patches of seagrass ( Thalassia testudinum ) with shoot densities ranging from full density (Full), half of the original shoot density (50%), 10% of the original density (10%), and complete shoot removal (Bare) crea ted at two study sites (North Skyway and East Beach) in Tampa Bay, FL. Values for each experimental treatment were averaged across the study sites. Statistically different groupings indicated by post-hoc analysis (Tukey B) are represen ted by upper case lettering (A or B). 74

PAGE 88

0 0.2 0.4 0.6 0.8 1 1.2 1.4 63 125 Particle size fraction ( m)% carbonates in accumulated particles Bare 10% 50% Full a 0 1 2 3 4 5 6 7 63 125 Particle size fraction ( m)% carbonates in accumulated particles Bare 10% 50% Full b Figure 29 Average ( S EM, n = 5) percent car bonate composition by particle size fraction of particles accumulated in experimentally thinned patches of seagrass ( Thalassia testudinum ) at a) North Skyway and b) East Beach study sties. Experimental shoot densities ranged from full density (Fu ll), half of the original shoot density (50%), 10% of the original density (10 %), and complete shoot removal (Bare) created at two study sites in Tampa Bay, FL. 75

PAGE 89

76 DISCUSSION Particle accumulation in bare and reduced density patches within larger seagrass beds differ, most notably in the bare patc hes, but may not be a function of particle deposition from the water column. Lack of c onsistent density depe ndent flow reductions or modifications suggest presence of bare or reduced density pa tches within larger seagrass beds have little effect on flow in these systems. This may account for the low amount of density dependent particle accumu lation by reduced density seagrass patches measured by this study and suggests that larg e scale measure of eco system function (i.e. particle accumulation) within seagrass system s may be modified little by the presence of small scale gaps or patches within these systems. Hydrodynamic Characterization Density dependent flow reductions were not found for any of the seagrass density treatments. Gacia et al. ( 1999) detected reductions in cu rrent velocities proportional to canopy height by altering seagrass ( Posidonia oceanica ) structure (i.e. canopy height) and Peterson et al. (2004) found greater flow reductions within higher de nsity seagrass ( Zostera marina ) canopies, but both did so at the lead ing edge of a bed where flow first encounters the canopy. Patches for this study were placed 5 m in from the leading edge of the beds, so flow patterns through and above the T. testudinum canopy were well established before encountering patches of altered seagrass densities. On the scale tested (1 m2), gaps within a larger, cont inuous seagrass bed did little to disrupt pre-established flow patterns through and above the seagrass canopy. Only on days with minimum flow ( 0.02 m s-1) was some attenuation of flow measured with in the canopy of the dull

PAGE 90

77 density bed in comparison to the bare patc hes, which suggests flow was skimming over the patches on days with maximum flow ( 0.04 m s-1). Gambi et al. (1990) measured skimming flows in a flume at overlying velocities as low as 0.05 m s-1, but at seagrass ( Zostera marina ) shoot densities 1000 shoots m-1. This can have implications for particle accumulation, as skimming flows potentially reduce the amount of mixing between the water column and the canopy (K och & Gust 1999), and may account for the low amount of density dependence in particle ac cumulation by the patches. Measures of both turbulent kinetic energy (m2 s-2) and Reynolds shear stresses (Pa) were found to increase as overlying fl ow speed increased, but these relationships were not found to be density dependent. Flui d momentum is extracted in the form of turbulence as flow is disrupted by canopy pr esence (Denny 1988). Gambi et al. (1990) found turbulence within a seagrass canopy to increase with greater distance from the leading edge of the canopy as fluid momentum was progres sively extracted as more plants were encountered by the flow, but found li ttle effect of density. This suggests that as flow encounters gaps and low density pa tches within larger seagrass beds, smaller sized patches (i.e. 1 m across) may not be of sufficien t size to alter fluid momentum extraction, turbulence, or Reynolds shear stress, a shear stre ss resulting from turbulent velocity fluctuations, regardless of the pres ence, density, or absence of seagrass within these patches. Lack of density dependence fo r flow measures that are through to dictate levels of particle depositi on and resuspension (Granata et al. 2001, Widdows et al. 2008), such as turbulent kinetic energy and Reynolds sh ear stresses, indicate a lack of influence on flow by the seagrass density treatments, whic h may be reflected in the lack of density effects on particle accumulation by the s eagrass density treatments in this study.

PAGE 91

78 Turbulent kinetic energy within the ca nopy 5 cm above the bottom (20% of canopy height) was found to increase at a slower ra te in full seagrass density patches when compared to the reduced seagrass density treatments. Nepf (1999) found increased turbulence with the addition of low densities (200 stems m-2) of rigid emergent vegetation ( Spartina alterniflora and mimics) to previously unvege tated areas due to the production of additional wakes, but decr eased turbulence as increased densities (up to 2000 stems m2) reduced flow speed. In this study reductions in flow speed were generally not found when flow through the full density bed was co mpared to reduced density (50% and 10%) or bare patches, but on minimum flow ( 0.02 m s-1) days some flow attenuation was measured in the canopy of the full density bed. This may account for the lower increase in turbulent kinetic energy with flow with in the canopy of the full density bed, but the lack of differences between the remaining de nsity treatments suggests that the presence of patches within larger seag rass beds have little effect on flow in these systems. Particle Accumulation Despite the lack of consistent density de pendent flow reductions or modifications, differences in particle accumulation in bare and reduced density patches within larger seagrass beds were measured, suggesting pa rticle accumulation in these patches may not be a function of particle depos ition from the water column. Unexpectedly, dry weight (g m-2 day-1) particle accumulation was significantly gr eater in bare treatment patches when compared to the half (50%) and full seagra ss density patches, driven by accumulation differences in 125 m-sized particles. Maltese et al. (2007) found flow recirculation cells and stagnant regions form w ithin 1 m long gaps of seagrass ( Posidonia oceanica )

PAGE 92

79 mimics tested in a flume under both fast (0.055 m s-1) and slow (0.017 m s-1) flow conditions. In the current study presence of fl ow recirculation cells and stagnant regions could help account for increased dry weight particle accu mulation in bare treatment patches. Advection of flow into the ga p by the flow recirculation cell would have increased mixing between the bottom and the water column, which could have increased particle transfer to the collectors either via deposition from the water column or bed load transfer from within the patch. Gacia et al. (1999) and Gacia & Duarte (2001) found amounts of secondary deposition (i.e. deposition of resuspended particles) to be greater than primary deposition, suggesting resuspen sion may increase particle accumulation when the difference between secondary de position and resuspension is greater than primary deposition. Similarly, Granata et al (2001) found a gradient of decreasing deposition following a storm to be inversely proportional to Posidonia oceanica shoot density resulting from reduced resuspension with increased densit y. This suggest the presence of resuspension in bare treatm ent patches could have increased particle accumulation when compared to reduced a nd full density patches, which likely experienced low to no resuspension. Although dry weight (g m-2 day-1) particle accumulation was significantly greater in bare treatment patches, the percent organic matter and carbonate composition of particles accumulated in full seagrass density patches were significantly greater in comparison to bare seagrass patches. This suggests some differences in accumulation among density treatments may be the result of the seagrass beds and the organisms occupying them contributing material (i.e. orga nic and inorganic detritus) to the particles accumulated in the beds. Previous studies have suggested deposition of particles from

PAGE 93

80 the water column can be minimal in seagrass beds (Gacia et al. 1999; Gacia and Duarte 2001; Neumeier and Ciavola 2004; Widdows et al. 2008), indicating some of the particles accumulated must be generated by compone nts (e.g. mollusk shells, calcareous macroalgae, senescent seagrass leaves) within the bed. Duar te et al. (1999) found large biovolumes of seagrass derived detritus, some times five-fold greater than that of the living seagrass, within Posidonia oceanica seagrass beds. Additionally, the presence of seagrass increases faunal abundances (O'Gower & Wacasey 1967, Lewis & Stoner 1983, Edgar 1990, Hirst & Attrill 2008), which could further contribute to the organic and inorganic matter accumulated within seagrass beds. As was previous suggested, formation of a flow recirculation cell w ithin the bare treatment patches may have enhanced deposition of particles from the water column in these bare patches. If within the full density beds the particles accumula ting were bed generated this would account for the significant differences found in pe rcent organic matter and carbonate composition of particles accumulated by the ba re and full density patches. With the exception of a flume set study that utilized seagrass ( Posidonia oceanica ) mimics (Maltese et al. 2007), measur ements of flow in gaps and reduced density patches within larger seagrass canopi es has not been previously reported. Past studies have focused on measures of fl ow either at the upstream edge of in situ seagrass canopies (Gacia et al. 1999, Peterson et al. 2004 ) or isolated patches in flumes (Folkard 2005). In addition, no study has coupled flow m easurements in bare and reduced density patches with measures of part icle accumulation to address pr ocesses of sedimentation in locally patchy or fragmented seagrass system s. Previous studies have focused on the influence of vegetation presence on partic le accumulation (Gacia et al. 2003, Hasegawa

PAGE 94

81 et al. 2008), but not th e effects of seagrass loss. Thos e studies that do address density effects focus on differences in the sediment characteristics (i.e. degree of sorting, sediment size distribution, and organic matter co ntent) as indicators of modified particle accumulation, but do not consider direct measures of particle accumulation (Lynts 1966, Kenworthy et al. 1982, Eckman 1983, Fonseca et al. 1983, Eckman 1987) as was measured in this study. More studies similar to this are needed to fully address the extent to which the presences of small scale bare and reduced density patches in larger, continuous seagrass beds alter ecosystem function (i.e. particle accumulation) and associated services (e.g. sediment stabili zation, water clarity) provided by seagrass systems. Conclusions Although gaps and reduced density patches can be common within larger seagrass beds (Townsend & Fonseca 1988, Creed & Am ado Filho 1999, Bell et al. 2002), the presence of small scale (1 m2) gaps and patches ma y not significantly alter hydrodynamic conditions and depositional proce sses in seagrass systems. Significant reductions in flow generally were not measur ed within the canopy of any of the treatment patches, save for some attenuation of flow with the canopy of the full seagrass density bed on minimal flow days. Lack of altered flow in treatment patches was reflected in the lack of strong density dependent effects on particle accumulation. Bare patches were found to have significantly greater dry weight (g m-2 day-1) particle accumulation when compared to reduced and full seagrass density patches, possibly resulting from the formation of a flow recirculation cell increa sing particle deposition or bed load transfer

PAGE 95

82 within the bare patches. In contrast, pe rcent organic matter and carbonate composition of the particles accumulated in the full seagrass density beds was significantly greater, suggesting that the particles accumulated in these plots are likely generated by the beds themselves. Although this study suggests small scale (1 m2) and widely spread (>5 m apart) bare and reduced density patches within larger seagrass beds may have little effect on seagrass ecosystem processes, as seagrass landscapes become increasingly patchy or fragmented the resulting altered flow condi tions (Folkard 2005, Maltese et al. 2007) and faunal community composition (Edgar & Robertson 1992, Reed & Hovel 2006) may modify irreplaceable functions and servic es provided by this highly productive ecosystem.

PAGE 96

83 Chapter 4 Investigation of the Effects of Subm erged Aquatic Vegetation and Hydrodynamic Regime on Select Ecosystem Processe s in Vegetated Systems: A Review INTRODUCTION Submerged aquatic vegetation (SAV), su ch as seagrasses, macroalgae, and marshgrasses, dominate coastal regions worldw ide. These SAV habitats, in addition to being highly productive communities, provide extensive and highly valuable ecosystem services. Services linked to SAV syst ems include providing habitats for many commercially important fauna and their pr ey (Heck et al. 1995), nutrient cycling (Erftemeijer & Middelburg 1995), food sources for endangered species (Bjorndal 1980, Reich & Worthy 2006), such as dugongs, manatees, and green turtles, and reduced coastal erosion and increased water clarity (Marba & Duarte 1997) in SAV systems and adjacent habitats due to sediment trappi ng and stabilization (Ward et al. 1984, Koch 1999, Terrados & Duarte 2000, Gacia & Duarte 2001). Ecosystem services provided by SAV systems are highly dependent on the overa ll processes and inter actions that operate within these ecosystems. Becoming more appa rent is that the complex set of underlying processes, such as trophic energy transfer denitrification, and sedimentation, that characterize ecosystem function in these syst ems and are modified by the template of

PAGE 97

84 environmental variables, such as sedime nt properties, SAV structure, hydrodynamic conditions (Touchette & Bu rkholder 2000, Eyre & Ferguson 2002, Vizzini & Mazzola 2006, Hasegawa et al. 2008) present in these ecosystems. Many of the environmental and ecological variables that influence the ecosystem function of SAV systems, such as extent a nd complexity of SAV structure, sediment properties, and faunal commun ity composition, have been arguably well studied (e.g. Orth 1973, Fonseca et al. 1983, Eckman 1987, Edgar & Robertson 1992). One often ignored or understudied environmental variab le that can have a measurable effect on SAV ecosystem function is hydrodynamic setti ng, characterized by diverse variables such as flow speed, turbulence intensity, flow origin (i.e. tides v. waves), and shear stress (Koch and Gust 1999). In seagrass and ot her soft bottom systems, changes in hydrodynamic conditions alter habitat char acteristics such as SAV density and morphology (Polte et al. 2005), presence of ep iphytes (Schanz et al. 2002), as well as geochemical sediment properties (Kenworthy et al. 1982), which in turn can result in highly variable ecosystem function, such as am ount of food resources, sediment stability, and nutrient cycling. In addition to hydrodynamic conditions modi fying habitat characteristics, flow can also impact faunal communities residing in these habitats. High seawater flux to seagrass beds is accompanied by enhanced larval recruitment (Eckman 1987), followed by larval deposition, especially in areas where flow is reduced at the canopy edge (Bologna & Heck 2002). Growth rate s in filter feeding bivalves ( Mercenaria mercenaria ) are increased by enhanced flux rates of suspended food particles to seagrass ( Halodule wrightii and Zostera marina) habitats under high flow conditions (Irlandi &

PAGE 98

85 Peterson 1991, Irlandi 1996). These findings suggest that ecosyste m function in these systems may not be equivalent across flow regimes. The purpose of this review is to evalua te existing information on links between hydrodynamic regime and ecosystem function associated with SAV systems. Specifically, the objectives ar e to: 1) summarize current know ledge regarding effects of SAV on hydrodynamic characteristics (e.g. veloci ty, turbulence, shear stress, etc.) and cascading effects on sedimentation and faunal communities in SAV systems; 2) summarize relationships recorded between sedi mentation and the pres ence and/or amount of structural components (e.g. density, canopy height, biomass, etc.) of SAV, and how hydrodynamic conditions can modify that relatio nship; 3) explore th e link between SAV presence and/or amount of structure a nd faunal community characteristics (e.g. abundance, richness, diversity, biomass, et c.) and effects hydrodynami c regime may have on that link; and 4) discuss the importa nce of considering hydrodynamics when exploring measures of ecosystem function, su ch as sedimentation and faunal community characteristics, across flow regimes. This review, focuses mainly on experime ntal studies that considered marine, estuarine, and some freshwater submerge d aquatic vegetation (SAV) inhabiting soft sediments. It is restricted to surveying studies investigating inve rtebrate faunal species with well known affinities for SAV as a primary or exclusive habitat, such as epifauna and species with low mobility, as thes e would most likely be influenced by hydrodynamics and/or sediment characteristics.

PAGE 99

86 BACKGROUND Hydrodynamics and SAV Many studies have demonstrated that s eagrasses and other SAV reduce currents (Almasi et al. 1987, Ackerman & Okubo 1993, Nikora et al. 1998, Heiss et al. 2000, Madsen et al. 2001), attenuate waves, and da mpen turbulence within their vegetative canopies (Madsen 1983, Leonard & Luther 1995 ). The extent to which SAV both reduces currents and attenuates waves is dependent on multiple factors including prevailing hydrodynamic conditions (e.g. tides vs. waves, turbulences, shear stress, etc.), canopy height, and vegetation structure. Hydrodynamic Conditions The relationship between ecosystem function (e.g. sediment capture and resuspension, larval trapping, etc.) and SAV is highly dependent on the hydrodynamic environment (e.g. dominance of tide or wave action, flow speed, etc.), present in the habitat. In tide dominated or unidirectional flow conditi ons, currents cause blades to bend in a single direction, often for hours at a time, only to change direction with the tide. The overlapping, bent canopy can effectively cr eate a barrier between the environments above and within SAV beds. Mixing betw een the overlying water column and the bed should be reduced in this situation (Koch & Gust 1999). Under wave dominated or oscillatory flow conditions, blades tend to flap at high frequencies as flow oscillates back and forth, effectively opening and closing the canopy every few seconds. Under these conditions, there should be increased mixing between the overlying water column and SAV bed (Koch & Gust 1999). The process of mixing between the water column and the

PAGE 100

87 canopy allows for nutrient (e.g. nitrogen, phosphor us, etc.) and particle (e.g. sediments, organic matter, larvae, etc.) exchange between these two environments, thus the degree of mixing can have important consequences for measures of ecosystem function, such as sedimentation and faunal community characteri stics, which will be discussed further in subsequent sections. The extent of mixing between the water column and the canopy is also highly dependent on flow velocity and can transfor m the influences of unidirectional verses oscillatory flows in SAV systems. In hi gh flows, SAV canopies deflect currents above and around the bed, which produces intensificatio n in deflected flow (Gambi et al. 1990, Sand-Jensen & Mebus 1996, Verduin & B ackhaus 2000, Backhaus & Verduin 2008, Widdows et al. 2008). This may potentia lly result in reduced mixing between the canopy and the overlying water column (Koch & Gust 1999). Conversely, any significant reductions in flow due to SAV may be reduced or negligible under low flow conditions (Heiss et al. 2000), so deflection and intensifica tion of flow above and around SAV beds and the possible accompanying re ductions in mixing observed under higher flow conditions are usually not found (Koch & Gust 1999). Certain hydrodynamic conditions (i.e. ambient currents in excess 10 cm s-1) result in monami (i.e. seagrass blades moving in a waving motion) due to hydr oelasticity of plants causing velocity fluctuations both within and above the canopy (Ackerman & Okubo 1993, Grizzle et al. 1996). When monami is present, turbulent vertical transpor t of momentum is enhanced (Ghisalberti & Nepf 2002), and exchange between the water colu mn and within the meadow increased (Koch & Gust 1999, Granata et al. 2001). Thus, the degree of mixing

PAGE 101

88 between the water column and the ca nopy is highly dependent on prevailing hydrodynamic conditions (e.g. tide ve rsus wave dominated water fl ow, flow speed, etc.). Prevailing hydrodynamic conditions modi fy not only the degree of mixing between the water column and SAV canopies, bu t also that within SAV canopies due to turbulence and shear stress. As flow transitions from laminar to turbulent, fluctuations in velocity resulting from vibra tions in flow or imperfectio ns in substrate are no longer dampened by viscosity and eddies are form ed (Denny 1988). The latter have properties of the fluid contained within them, so they are an efficient way of transferring mass and momentum. The size and number of eddies formed provide an indication of the amount of turbulence present in the system. Shear stresses give an indi cation of the velocity gradient, so the greater the difference in fl ow between two heights above the bottom, the greater the shear stress (Denny 1988). At th e canopy-water interface, turbulence intensity and shear stress increase dramatically, whic h enhances exchange between the canopy and overlying waters (Gambi et al. 1990, Ik eda & Kanasawa 1996, Nepf & Vivioni 2000, Hendricks et al. 2008, Widdows et al. 2008). Within SAV canopies, the hydrodynamic environment is usually characterized by suppr essed turbulence and low shear stresses (Anderson & Charters 1982, Gambi et al. 1990, Nepf & Vivioni 2000, Hendricks et al. 2008), and consequently reduced mixing (Ackerman & Okubo 1993), as currents are reduced and waves are attenuated, and momentum is lost due to the friction drag of the SAV canopy. The exception is sparse SAV canopies, which tend to experience higher turbulence intensities than their dense counterparts (Worcester 1995, Nepf 1999). Reduced mixing within SAV canopies can lead to both reduced sediment trapping and

PAGE 102

89 larval capture (Grizzle et al. 1996, Granata et al. 2001) and reduced mass transfer of dissolved nutrients and gasses needed for photosynthesis by SAV. Canopy Height The proportion of the water column occupied by SAV beds can have a measureable effect on flow in these systems. When water depth is greater than SAV canopy depth, waves are not effectively attenuated, but when the SAV canopy extends through the entire water colum n, current velocities are effectively reduced (Ward et al. 1984, Fonseca & Fisher 1986, Neumeier & Ci avola 2004, Moller 2006) and wave energy is attenuated (Koch 1999, Chen et al. 2007). Fonseca & Cahalan (1992) showed that percent wave energy reductions were 40% when canopy height and water depth were near equal. Similarly, Gacia et al. (1999) found that current velocities were reduced proportionally with height of seagrass ( Posidonia oceanica ) canopy. Under wave dominated flow conditions, the ability of SAV to attenuate waves either plateaus or is reduced as wave height increases with re spect to canopy height (Moller 2006, Bradley & Hauser 2009). Consequently, th e ability of emergent SAV, such as saltmarsh grasses, to attenuate waves to a greater extent than subtidal SAV, such as seagrasses and macroalgae, complicates comparisons of ecosystem function (e.g. sediment trapping, larval capture, etc.) across SAV types (Peral ta et al. 2008) and di fferent water depths. Distinct hydrodynamic environments above and within SAV canopies are seen in tide dominated systems when the canopy does not occupy the entire water column. Above the structure of the canopy, as with over bare sand, flow profiles tend to be logarithmically shaped, with flow increasing until the prevailing flow speed is reached

PAGE 103

90 (Nikora et al. 1998, Abdelrhman 2003, Neumei er & Ciavola 2004). Within the canopy, flow profiles follow a sigmodal shape. Flow sharply increases bot h with distance from the bottom and as resistance due to the presence of the canopy is reduced at the canopywater interface, resulting in both horizontal mixing in the canopy and vertical mixing between the canopy and the water column (N epf & Vivoni 2000). In contrast, flow profiles in emergent SAV canopi es or those that occupy the entire water column often resemble profiles over bare sand or above th e structure of an SAV canopy (Nepf et al. 1999), where only horizontal mixing is pres ent between the water column and SAV canopies and small scale turbulence makes di ffusive transport slow within the canopy (Nepf & Vivoni 2000). Thus, the extent S AV canopies occupy the water column can have important implications concerning the extent of diffusion and mixing both within the canopy and between the water column and the canopy. Vegetation Structure While impacts of canopy height on hydrodynamic characteristics have been investigated at length, some disagreement exists concerning the relationship between other measures of SAV structure, such as SAV density, and reduc tion of currents and attenuation of waves in SAV systems. Studi es considering the effects of SAV density on flow patterns using laboratory flumes have reported no consistent effect of density on current reduction or turbulence generation (Fonseca et al. 1982, F onseca & Fisher 1986, Gambi et al. 1990, Fonseca & Cahalan 1992). In contrast, field studies focused on flow measures in high versus low SAV densities have found the opposite (i.e. consistent reductions in currents and turbulence w ith increasing SAV densities) (Eckman 1987,

PAGE 104

91 Leonard & Luther 1995, Peterson et al. 2004). While conditions of studies concerning the interaction between SAV a nd flow conducted using flumes have the benefit of being more controlled than field experiments, they are also usually constr ained by width, depth, and working length of the flume, in addition to that distribution of the SAV within the flume. For example, Fonseca and Koehl ( 2006) found SAV canopies that extended the entire width of a flume working section were less effective at reducing current velocities within the SAV canopy than narrow SAV patches. Thus, discrepancies between field and flume studies may be more a result of expe rimental design than actual differences. Although effects may not be consistent across experimental settings, increases in SAV density generally reduce currents (L eonard & Luther 1 995, Widdows & Brinsley 2002), increase attenuation of waves (Koch & Gust 1999, Chen et al. 2007), and decrease turbulence (Leonard & Luther 1995, Koch & Gust 1999, Luhar et al. 2008). Specifically, a study by Widdows et al. (2008) showed up to a 40% reduc tion in near-bed flow in dense seagrass canopies ( Zostera noltii ) when compared to prevailing flow, and Peterson et al. (2004) showed that flow within the canopy of a seagrass ( Zostera marina ) bed was predicted to vary inversely w ith the square root of the shoot density. In contrast, significant reductions in velocity (Sand-Jensen & Mebus 1996) and turbulent stress (Luhar et al. 2008) inside versus outside SAV beds can be reduced or disappear completely, when SAV density is sparse. Functionally, sparse beds can act very similarly to unvegetated bare areas, and the ecosystem se rvices they provide may be quite different than those provided by dense SAV beds, particularly with respect to sediment trapping and faunal capture. Thus, many of the ecosystem characteri stics often associated with SAV beds, such as high amounts of sedimentary silt-clays and orga nic matter, increased

PAGE 105

92 water clarity, and distinct faunal communities, may no longer be evident in sparse beds when the hydrodynamic environment (i.e. flow and turbulence) experienced inside SAV beds resembles conditions outside the beds. Hydrodynamics, SAV, and Sedimentation Hydrodynamics and SAV characteristics, as discussed above, set up a complex suite of potential effects on sedimentation with in SAV habitats. Perhaps the most often reported link is that reductions in hydr odynamic conditions are of ten assumed to be accompanied by corresponding increases in deposition (Ward et al. 1984), decreases in resuspension (Gacia & Duarte 2001), and modi fication of sediment characteristics toward more fine and organic rich (Peterson et al. 1984), within versus outside SAV beds ( Figure Figure ). As velocity increases, the probability of particle resuspension and size of particle that can be resuspended also incr ease. T hus, coarser sediments coupled with elevated resuspension experienced by unvegeta ted areas, in comparison to SAV beds, are most likely the direct result of more freque nt and/or extended exposure to velocities above critical friction velocities. Resuspension can vary over temporally as a result of predictable (e.g. daily tidal cycles, annual SAV die-bac k, etc.) and unpredictable (e.g. storms, wide-spread SAV dieoffs, etc.) events (Wilson 1949, Granata et al 2001). During ebb tide, suspended solids, in addition to phosphorus and silicate concentrations, are higher in water ebbing from denuded mudflats than from seagrass-covered mudflats

PAGE 106

WATER SURFACE a b SUSPENDED SUSPENDED SOLIDS SOLIDS DEPOSITION R ESUSPENSION FLOW FLOW SAV SAV 93 SEDIMENT SEDIMENT Figure 30 Conceptual models of th e interactions between su bm erged aquatic vegetation (SAV) and sedimentation under equivalent hydrodynamic conditions and how those interactions are modified by varying SAV de nsities. When SAV densities are high (a), flow is attenuated, deposition increases, and c oncentration of suspended solids is reduced. In contrast, when SAV densities are lo w (b), flow attenuation is reduced, and resuspension and concentrations of suspended solids are increased.

PAGE 107

WATER SURFACE a b SUSPENDED SUSPENDED SOLIDS SOLIDS DEPOSITION 94 Figure 31 Conceptual models of interactions be tween equivalent densities of subm erged aquatic vegetation (SAV) and sedimentation a nd how those interactio ns are modified by hydrodynamic conditions. Under low flow conditions (a), flow is attenuated, sediments are deposited, and concentration of suspended so lids is decreased. As flow increases (b), flow attenuation is reduced, and resuspensi on and suspended solids concentrations are increased. FLOW FLOW SAV SAV R ESUSPENSION SEDIMENT

PAGE 108

95 (Bulthuis et al. 1984), but thes e daily tide induced resuspensi on events may only play a small roll in overall sediment dynamics of a system. Unpredictable, intense resuspension events resulting from storms and other elevat ed wind episodes can prove more important, causing significant increases in particle m ovement (i.e. sediments and seagrass seed banks) or resuspension within SAV beds (Ward et al. 1984, Dauby et al. 1995, Asmus & Asmus 2000, Granata et al. 2001, Paling et al. 2003, Bell et al. 2008). Consequently, large resuspension events may eclipse any di fferences in sedimentation resulting from daily depositional fluctuations. Resuspension and particle (i.e. sediment s and associated organic matter) loss in SAV habitats may be a necessary process for ma intaining these habitats. Elevated levels of organic matter are often accumulated in SAV beds in comparison to adjacent unvegetated areas (Kenworthy et al. 1982), but it has been su ggested that accumulation of >5% organic matter in the sediment may lead to loss of SAV due to build up of toxic compounds (Barko & Smart 1983, Barko & Smart 1986) or changes to st ructural growth patterns (Wicks et al. 2009). Additionally, a ccumulation of sediments with high silt-clay and organic matter content increase concentrations of sulfides and the extent of anoxia in the sediments, which may adversely affect fa unal survivorship (Neira et al. 2006). Under these conditions resuspension and removal of some accumulated sediments and organic matter may be necessary to prevent faunal community loss (Neira et al. 2006). The importance of SAV in shallow, near shore environments is not only due to their ability to reduce resuspen sion, but SAV structure can also act as a filter by trapping and stabilizing sediments and increasing water clarity (Chen et al. 2007). Such impacts can be beneficial to growth and produc tion of both SAV and ecosystems farther off

PAGE 109

96 shore, such as coral reefs (Y amamuro et al. 2003). Depositio n tends to be reduced when the flux of particles (i.e. amount of particles in the water co lumn that pass over the SAV canopies per unit time) to SAV canopies is reduc ed, as a function of either low particle concentrations in the water column or slow flow (Granata et al. 2001). Reductions in deposition resulting from reduced mixing be tween the water column and the canopy can be the result of low flow conditions, charac terized by reduced turbulence and low shear stresses (Gambi et al. 1990, Ikeda & Kana sawa 1996, Nepf & Vivioni 2000, Hendricks et al. 2008, Widdows et al. 2008), or high flow c onditions, which cause a barrier between the water column and as vegetation bends ove r in skimming flows (K och & Gust 1999). In moderate flows, increased deposition from increased exchange between the environments above and within SAV canopies have been noted (Peralta et al. 2008). Thus, effects of hydrodynamics on sedimenta tion in the presence of SAV are complex and highly dependent on both delive ry rate of particles to th e SAV bed and the extent of exchange between the water column and SAV canopy. Hydrodynamics, SAV, and Faunal Communities The complex set of interactions that direct sedimentation in SAV systems is closely linked to factors that govern faunal co mmunity characteristic s, especially with regards to hydrodynamic influences. A dominant theme of papers investigating fauna in SAV systems is the presence of SAV increasing faunal community metrics, such as abundance/density, richness, evenness, divers ity, biomass, and production (OGower & Wacasey 1967, Thayer et al. 1975, Stoner 1980, Lewis & Stoner 1983, Virnstein et al. 1983, Pihl 1986, Edgar 1990, Edgar et al. 1994, Hily & Bouteille 1999, Hirst & Attrill

PAGE 110

97 2008). Another major effort has been direct ed at examining fauna across gradients of SAV density (Orth 1973, Homziak et al. 1982, Edgar & Robertson 1992, Connolly and Butler 1996, Webster et al. 1998, Attrill et al. 2000), with th e expectation that greater amounts of SAV structure facilitate the pr esence and abundance of faunal populations. Generally, these types of studies describe faunal communities by measures of abundance and diversity, such as, richness and evenness. Fewer studies use metrics such as biomass, production, community structure, or mort ality, which arguably offer a better understanding of underlying processes (e.g. trop hic energy transfer, succession, etc.) in these faunal communities (Edgar 1999, Hily & Bouteille 1999). Regardless of the metrics used to describe faunal communities, both the presence of SAV and greater amounts of SAV structure increase measures of faunal community ch aracteristics in SAV systems. Studies exploring the link between faunal communities and SAV beds generally focus on biotic factors associated with S AV structure (e.g. increased food availability, reduced predation, increased passive larval se ttlement, and increased habitat complexity) to explain the relationship between these faunal community components and vegetation structure. For example, Edgar (1999) found that faunal community metrics (i.e. abundance, biomass, and productivity) depend ed on food availability in an SAV (i.e. seagrass mimics) system. What is often not addressed by these studies is how many biotic factors under consideration can al so be modified by abiotic factors (e.g. hydrodynamic conditions, sediment stability and characteristics). For example, sediments of SAV beds are generally charac terized by fine grain sizes and are organic rich, which is attributed to lower ener gy conditions found inside the beds (Orth 1977,

PAGE 111

98 Grady 1981, Kenworthy et al. 1982, Peterson et al. 1984). Enhanced pools of organic detritus in sediments and on the sediment surface within SAV beds provide an important food source for fauna that inhabit these beds (Edgar 1999). Passive larval settlement can also be modified by prevailing hydrodynamic c onditions, with reductions in flow at the edge of the SAV beds enhanceing passive larval settlement (B ologna & Heck 2002). Enhanced larval settlement on seagrass (Zostera marina) blades was found when blades displayed large-amplitude waving (mona mi) under high current speeds (>10 cm s-1) (Grizzle et al. 1996). Enhanced vertical mixing between SAV canopies and the water column and horizontal advection into SAV beds under high flow conditions increase delivery rates and flux of food particles to filter feeders in SAV beds (Worcester 1995, Hendricks et al. 2008). Conve rsely, high flow habitats e xperience greater sediment resuspension, which clogs feeding structures and decreases feeding efficiency of filter feeders (Brun 2009). Ultimately, prev ailing hydrodynamic conditions can have cascading effects on biotic components of S AV systems, which can potentia lly alter ecosystem function. REVIEW Methods Literature survey was conducted using ISI Web of Knowledge database to primarily search for experimental studies with combinations of the keywords: sedimentation, fauna, flow, hydrodynamics, density, seagrass, and submerged aquatic vegetation. Studies included in the review fit the criteria of addressing the link between SAV and either sedimentation or faunal commun ity characteristics. The literature cited

PAGE 112

99 in the studies generated by this database search was also searched to discern any appropriate additional studies that may have been disregarded by the initial search. A total of 89 studies were include d in the review and were subdivided into 21 studies that addressed the link between the presence or absence of SAV and sedimentation (Appendix: Table 20), 31 studi es that dealt with SAV-fa una relationships (Appendix: Table 21), 16 studies investigating inte ractions between varying amounts of SAV structure and sedimentation (F igure 30), and 28 studies that examined the SAV structurefauna association (Figure 31). Sedimentation Sedimentation in Vegetated and Unvegetated Habitats From the body of literature surveyed, it is clear that most investigations of links between SAV and sedimentation have focu sed on differences between vegetated and unvegetated habitats (Appendix: Table 20) Regardless of whether the amount of structure or simply the presence of SAV wa s considered, over 62% of the papers found a positive effect of SAV on sedimentation (Tab le 2, Appendix: Table 20). Results range from descriptive studies (Wilson 1949) noti ng sand being washed away following a large scale seagrass (Zostera marina) die-off, to reports of elev ated silt-clays and organic matter content in sediments present in SAV be ds relative to unvege tated, bare habitats (Orth 1977, Grady 1981, Kenworthy et al 1982, Peterson et al. 1984, Eckman 1987, Heiss et al. 2000). Over one third of all pa pers on his topic focused on measures of sediment grain size and organic content (Append ix: Table 20). These mainly field based

PAGE 113

100 Table 2 Summary of studies investigating influen ce of submerged aquatic vege tation (SAV) structure on se dimentation. Categorized by study type (field and/or flume), vegeta tion type (seagrass and/or sa ltmarsh grass), vegetation st ructural descriptor(s) (sho ot density, canopy height, biomass, etc.), SAV-sedimentation relationship (pre sence, positive, negative, or variable), study result(s), and hydrodynamic modification of result(s). Type of study Vegetation Structural descriptor(s) Relationship Result(s) Hydrodynamics Reference field Halodule wrightii & Zostera marina shoot density no seagrass density sedimentary % silt-clay or organic matter not addressed in results Kenworthy et al. 1982 flume H. wrightii, Syringodium filiforme, Thalassia testudinum & Z. marina shoot density no seagrass shoot density sediment entrainment canopy friction = erosion Fonseca & Fisher 1986 field Posidonia oceanica shoot density, canopy height, biomass, & leaf area index (LAI) variable seagrass LAI = particle trapping seagrass shoot density, biomass, or canopy height erosion = particle trapping Gacia et al. 1999 field P. oceanica shoot density & canopy height seagrass density = deposition resuspension = deposition Granata et al. 2001

PAGE 114

101 Table 2 (Continued) Type of study Vegetation Structural descriptor(s) Relationship Result(s) Hydrodynamics Reference field T. testudinum intensity of T. testudinum growth + dense seagrass growth = fine grained sediments not investigated Lynts 1966 field T. testudinum shoot density + seagrass density = particle trapping no relationship Meyers et al. in prep field T. testudinum mimics shoot density no seagrass density particle trapping flow = particle trapping Meyers et al. in prep field Z. marina LAI & biomass + seagrass LAI = % organic matter in low flow shear velocity = sedimentary % silt-clay & organic matter Fonseca et al. 1983 field Z. marina & mimics shoot density + seagrass density = % sedimentary siltclays flow = sedimentary % silt-clays with equal seagrass shoot densities Eckman 1987 field Z. marina shoot density + seagrass density = sediment accretion not investigated Bos et al. 2007

PAGE 115

102 Table 2 (Continued) Type of study Vegetation Structural descriptor(s) Relationship Result(s) Hydrodynamics Reference field Z. marina aboveground biomass & shoot height no seagrass aboveground biomass or shoot height particle trapping rate seagrass biomass or shoot height = flow = resuspension Hasegawa et al. 2008 field & flume Z. marina & Zostera noltii leaf density & biomass + seagrass density = erosion seagrass density = flow, but TKE and bed shear stress Widdows et al. 2008 flume Z. noltii Spartina anglica & mimics shoot density & flexibility variable SAV density sediment accretion in low flow S. anglica = velocity = sediment trapping; Z. noltii = erosion Peralta et al. 2008 field Spartina alterniflora stem density & biomass + high S. alterniflora density = [total suspended solids] in the water column S. alterniflora = velocity, turbulence intensity, & TKE Leonard & Croft 2006 field & flume Spartina mimics mimic density + Spartina mimic density = sedimentation Spartina mimics = velocity, turbulence intensity, & TKE Bouma et al. 2007

PAGE 116

103 Table 2 (Continued) Type of study Vegetation Structural descriptor(s) Relationship Result(s) Hydrodynamics Reference field & flume Scirpus americanus mimics mimic density + medium & high S. americanus densities = fine grained sediments medium & high S. americanus densities = bed shear stress = erosion Eckman 1983

PAGE 117

104 Table 3 Summary of studies investigating influence of submerge d aquatic vegetation (SAV) st ructure on associated faunal communities. Categorized by vegetation type (seagrass, algae, a nd/or saltmarsh grass), vegetation structural descriptor(s) (sh oot density, canopy height, biomass, etc.), fauna type, SAV-fauna relationship (present, positive, negative, or variable), study re sult(s), and hydrodynamic modification of results. Vegetation Structural descriptor(s) Fauna Relationship Result(s) Hydrodynamics Reference Amphibolis antarctica & Amphibolis griffithii shoot & leaf density epifauna + seagrass shoot density = epifaunal density, richness, and abundance; seagrass leaf density = abundance not investigated Edgar & Robertson 1992 Cymodocea nodosa canopy height epifauna + seagrass canopy height = epifaunal abundance & biomass not investigated Connolly & Butler 1996 C. nodosa & Zostera noltii shoot density, leaf area index (LAI), & leaf standing crop (LSC) polychaetes + seagrass shoot density, LAI, & LSC = polychaete density & high patchiness flow = epifaunal diversity & abundance Gambi et al. 1998

PAGE 118

105 Table 3 (Continued) Vegetation Structural descriptor(s) Fauna Relationship Result(s) Hydrodynamics Reference C. nodosa & Z. noltii # shoots, # leaves/shoot, shoot length & width, LAI, & above& belowground biomass macrofaunal bivalve ( Cerastoderma edule ) no seagrass shoot length, width, or density, aboveground biomass, or LAI bivalve food intake rate hydrodynamic conditions in seagrass = bivalve food intake rate Brun 2009 Halodule wrightii biomass epi& infauna variable seagrass biomass faunal densities for all species not investigated Young & Young 1978 H. wrightii & Zostera marina shoot density benthic macrofauna + seagrass density = faunal abundance, richness, & diversity no consistent effects of hydrodynamics Homziak et al. 1982 H. wrightii & Z. marina shoot density & length macrofaunal bivalve ( Mercenaria mercenaria ) no seagrass shoot density or length clam shell growth blade length = velocity = clam shell growth Irlandi & Peterson 1991 H. wrightii & Z. marina % seagrass cover macrofaunal shrimp ( Penaeus duorarum ) + % seagrass cover = shrimp abundance wave energy = shrimp abundance Murphey & Fonseca 1995 H. wrightii, Syringodium filiforme, & Thalassia testudinum shoot density & canopy height macroepibenthic fauna + 1/3 natural seagrass shoot density = natural faunal densities no observable effects Fonseca et al. 1996

PAGE 119

106 Table 3 (Continued) Vegetation Structural descriptor(s) Fauna Relationship Result(s) Hydrodynamics Reference H. wrightii & Z. marina shoot density, blade height, & species of vegetation macrofaunal bivalve (M. mercenaria ) no seagrass shoot density, blade length, or vegetation species clam growth flow = small clam growth in seagrass Irlandi 1996 H. wrightii, Ruppia maritima, Z. marina, & macroalgae shoot density & aboveground biomass macrofauna variable seagrass shoot biomass in 1992 = faunal abundance relative wave exposure in 1991 = faunal abundance & richness Hovel et al. 2002 Heterozostera tasmanica mimics shoot density & length epibenthic harpacticoid copepods + seagrass blade density = copepod abundance not investigated Jenkins et al. 2002 Posidonia australis & Zostera capricorni shoot height & density, & species of vegetation fish & decapods + Z. capricorni shoot height = faunal richness; P. australis shoot height & density = faunal abundance not investigated Bell & Westoby 1986b S. filiforme T. testudinum & macroalgae aboveground biomass motile epibenthic macroinvertebrate + aboveground vegetation biomass = faunal species richness & abundance not investigated Heck & Westone 1977

PAGE 120

107 Table 3 (Continued) Vegetation Structural descriptor(s) Fauna Relationship Result(s) Hydrodynamics Reference T. testudinum & Halimeda opuntia biomass macrofauna + macrophyte biomass = crustacean abundance & richness not investigated Stoner & Lewis 1985 T. testudinum shoot density & plant biomass epifauna & surficial benthic organisms (>500 m) seagrass shoot density & biomass = faunal density flow at shoot density bed edges = larval deposition Bologna & Heck 2002 Z. capricorni shoot density fish & decapods + seagrass shoot density = faunal abundance not investigated Bell & Westoby 1986a Z. capricorni mimics plant density epibenthic harpacticoid copepods seagrass density copepod abundance not investigated Hicks 1989 Z. capricorni shoot density fish and decapods no seagrass shoot density fish & decapod abundance not investigated Worthington et al. 1992 Z. mariana shoot density macroinfauna + seagrass shoot density = faunal richness, diversity, & evenness not investigated Orth 1973

PAGE 121

108 Table 3 (Continued) Vegetation Structural descriptor(s) Fauna Relationship Result(s) Hydrodynamics Reference Z. marina & mimics shoot density macrofaunal bivalves ( Argopecten irradians & Anomia simplex) seagrass shoot density = bivalve abundance seawater flux under shoot densities = larval recruitment Eckman 1987 Z. marina shoot density infaunal macroinvertebrates + seagrass shoot density = faunal diversity not investigated Webster et al. 1998 Z. marina leaf #/shoot, leaf and stem length, & biomass epifaunal macroinvertebrates + seagrass biomass = faunal richness & abundance not investigated Attrill et al. 2000 Z. marina patch size, inpatch location, & shoot density infauna variable seagrass shoot density faunal variables (abundance, diversity, etc.), but = faunal assemblage composition hydrodynamics not directly measured Bowden et al. 2001 Z. marina mimic # blades, cover/area, surface area, space between blades, & blade width & length epifauna (>500 m) + seagrass structure = small & large faunal abundances flow = small faunal abundances Bartholomew 2002

PAGE 122

109 Table 3 (Continued) Vegetation Structural descriptor(s) Fauna Relationship Result(s) Hydrodynamics Reference Z. marina mosaic density epifauna variable seagrass structure = fauna species richness & abundance in large sized plots not investigated Reed & Hovel 2006 Z. marina mimic shoot length, shoot density, & surface area epifauna variable seagrass surface area = faunal density & diversity, but seagrass shoot length & density not directly measured Sirota & Hovel 2006 Scirpus americanus mimics mimic density infaunal meiofauna no mimic density faunal abundance no consistent effects of hydrodynamics Eckman 1983

PAGE 123

110 generally utilized sediment coring, which prov ides insight into sediment accumulation at these locales, and indirect evidence of curre nt reduction and wave attenuation due to SAV. However, use of cores exclusively offers a poor repr esentation of the contribution of deposition and resuspension events to th e sediment budget. Short term rates (e.g. hourly, daily, weekly) of deposition or resu spension cannot be discerned from sediment cores, and the contribution of shorter time scale events, su ch as increased particle concentrations from terrestr ial storm runoff or increased wave energy conditions in high wind events, may go undetected. Studies that employed measures of deposition and resuspension, such as concentration and loss rates of total susp ended solids in the water column, as an indication of the link between SAV and se dimentation, also generally (>65%) found a positive effect (i.e. increased deposition and decreased resuspension) of SAV on sedimentation (Appendix: Table 20). Comparable results (i.e. SAV presence positively enhancing sedimentation) were found in experimental studies under both controlled flume and natural field conditions. In a fl ume study by Hendriks et al. (2008), particle loss rates from the water column were 14 to 25 times greater in the presence of a seagrass ( Posidonia oceanica ) canopy than in an unvegetated flume, indicating increased deposition. Those studies that addressed deposition under fiel d conditions in both natural and artificial SAV beds (Almas i et al. 1987) and in newly rest ored SAV habitats (Bos et al. 2007) have found similarly enhanced deposition in the presence of SAV. Yet, other studies have reported the opposite effect (i .e. SAV presence does not positively enhance sedimentation) as a result of high wave energies (Paling et al. 2003), the presence of epiphytes (Vermaat et al. 2000), or reduced flows at bed edges (Vermaat et al. 2000,

PAGE 124

111 Neumeier & Ciavola 2004). Under high wave energy conditions, seagrass and SAV beds may become ineffective sediment traps as high wave energies are accompanied by increased turbulence intensity and shear stresses, which can induce sediment resuspension. It has been suggested that epiphytes present on SAV can trap sediments before they reach the sediment surface, wh ich reduces the amount of deposition to the sediment surface (Vermaat et al. 2000). Accumulation of sediments at the edges or leading edge (i.e. the edge of the bed that first encounters flow) of SAV beds can be greater than in the center of the beds, as re duction in flow from outsi de to inside the bed is greatest in this transition zone (Fons eca et al. 1983, Vermaat et al. 2000, Neumeier & Ciavola 2004). As the extent of depositi on in SAV can vary greatly under differing hydrodynamic conditions and epiphytic loads, the positive influence of SAV on sedimentation may be mostly dependent on the ability of SAV to reduce resuspension in these systems (Neumeier & Ciavola 2004, Peralta et al. 2008). A number (40%) of the studies surveyed de scribe the reduction of resuspension in SAV beds (e.g. more than 3 fold in certain cases) as most likely influencing significantly greater sedimentation in vegetated compared to unvegetated habitats (Gacia et al. 1999, Gacia & Duarte 2001, Neumeier & Ciavola 200 4). Ward et al. (1984) found that, in unvegetated areas, wave driven resuspension occurred during periods of high winds, but suspended particulate matter concentrati ons remained stable inside seagrass ( Ruppia maritima) beds. These findings were echoed by Koch (1999), who reported that in unvegetated areas resuspension was greater and the result of wave action, while in a nearby seagrass ( Thalassia testudinum ) bed, resuspension was only initiated during flood tide due to increased flow near the bottom. For diminutive seagrass species, such as

PAGE 125

112 Halophila decipiens, higher threshold velocities are required for sediment movement inside versus outside vegetated regions (F onseca 1989). In these studies, resuspension was recorded in unvegetated regions under lo w energy conditions, while resuspension in an SAV bed required higher prevailing fl ow conditions. SAV appears to maintain sediments in contrast to unvege tated regions that more regular ly experience resuspension. The positive relationship between SAV and sedimentation, as discussed above, is demonstrated geographically from temper ate (Kenworthy et al. 1982, Peterson et al. 1984, Ward et al. 1984, Gacia & Duarte 2001, Bos et al. 2007) to tropical (Fonseca 1989, Agawin & Duarte 2002) and from the nor thern (Orth 1977, Grady 1981, Almasi et al. 1987) to southern (Bulthuis et al. 1984, Heiss et al. 2000) he mispheres. Most studies surveyed were field based (90%) and the seag rass species conformed to a strap-bladed morphology (>80 %), with few othe r blade morphologies studied ( Ruppia maritima, Ward et al. 1984, Halophila decipiens, Fonseca 1989, and Amphibolis griffithii Paling et al. 2003). While Ward et al (1984) and Fonseca (1989) showed a positive relationship between seagrass and sedimentation, Paling et al. (2003) found no such link. Species of Amphibolis have the majority of their biomass in leaves concentrated at the top of long stalks, which allows for resuspension due to increased flows at the sediment-water interface. Given that certain morphologi es (e.g. strap-bladed species such as Thalassia testudinum ) are characterized by higher canopy fr iction and lower sediment movement than recorded for cylindrical counterparts (e.g. Syringodium filiforme ) (Fonseca & Fisher 1986), it is likely that SAV and sedimentation relationships will be species specific.

PAGE 126

113 Variation in SAV Stru cture and Sedimentation Over half of the papers examined concerning sedimentation in SAV systems focused solely on differences in sedimentati on due to either presence or absence of vegetation (Appendix: Table 20). The remaini ng studies addressed the role of varying amounts of SAV structure (e.g. density, ca nopy height, aboveground biomass, etc.) on sedimentation in SAV systems (Table 2). In some cases, descriptors were combined to produce secondary descriptors. For example, Gacia et al. (1999) found that particle trapping by a seagrass ( Posidonia oceanica ) bed was not a function of either shoot density, biomass, or canopy height alone, but instead found that total deposition rate was significantly positively correlated with the leaf area index (LAI), a combined measure of vegetation density and leaf su rface area that describes the pr ojected surface area of the seagrass. However, the most common metric s used to reflect seagrass abundance were shoot, leaf, or plant density. Changing SAV structure, specifically density, influenced sediment properties in SAV habitats in over 50% of studies surveyed (Table 2). Dense stands of SAV are usually associated with fine grained sediments (Lynts 1966, Scoffin 1970, Eckman 1983, Eckman 1987). In addition to modification of sediment properties, many studies surveyed (>30%) indicate that increases in SAV structure (i.e. density, canopy height, and aboveground biomass) ha ve a positive, although not necessarily linear, relationship with other measures (e .g. amount of particle trapping, erosion, and suspended solids in the water column) of se dimentation in SAV habitats (Table 2). Studies that investigated di rect measures of sedimentat ion (i.e. amount of sediment accretion), such as Bos et al. (2007) and Bo uma et al. (2007), found sedimentation to be significantly lower in sparse compared to dense patches of SAV. Indirect measures of

PAGE 127

114 sedimentation (i.e. concentration of suspe nded solids in the water column) were also decreased as SAV densities increased, indicating increased deposition (Leonard & Croft 2006, Widdows et al. 2008). The trend of increased sedimentation with increased SAV structure also holds across both differences in species (seag rasses, Gacia et al. 1999, salt marsh grasses, Leonard & Croft 2006) and t ypes of studies (i.e. field versus flume studies) (Bouma et al. 2007, Wi ddows et al. 2008). From the few studies that report the link between SAV structure and sedimentation, it appears that increased SAV structure enhances sedimentation. It is important to note that not all st udies surveyed (>40%) found a positive or consistent relationship betw een increased SAV structure (e.g. shoot density, biomass, canopy height, etc.) and sediment ation (Table 2). Of the studies surveyed, some (25%) found no link between SAV structure and se dimentation, while others (<10%) found sedimentation to be inversely related to S AV structure. For those studies that found no consistent relationship between sedimentation and SAV stru cture (Kenworthy et al. 1982, Fonseca & Fisher 1986, Hasegawa et al. 2008) no single explanation emerges. Kenworthy et al. (1982) measur ed high percentages of silt-c lay and organic matter in the center of seagrass ( Halodule wrightii and Zostera marina) beds where shoot densities were lower than at the edges of the beds. The center of the beds corresponds with the region of maximum flow entrainment, thus modifying the sediment characteristics towards fine, organic rich sediments that are found in low energy environments. Other studies that reported either no relationship or a negative relationship between sedimentation and SAV structure attributed th is relationship to eith er reduced deposition (Fonseca & Fisher 1986, Peralta et al. 2008) or resuspension (Granata et al. 2001,

PAGE 128

115 Hasegawa et al. 2008) in the presence of a dense SAV canopy. As previously discussed, reductions in resuspension may correspond to reduced deposition in SAV systems. Alternatively, skimming flow that reduces mixing by creating a barr ier between the water column and canopy reduces potential for depos ition in dense seagrass beds (Fonseca & Fisher 1986, Peralta et al. 2008). Link between SAV structure and sedimentation, especially as it pertains to resuspensi on, appears dependent on the influence of hydrodynamics and canopy features. SAV, Sedimentation, and the Effects of Flow While prevailing hydrodynamic conditions can modify the relationship between SAV and sedimentation, including both th e sediment characteristics and amounts of resuspension and deposition, less than 80% of papers that examined the relationships between SAV and sedimentation include the infl uences of flow (Table 2). High energy systems tend to have similar sediment charact eristics both within and outside of SAV beds (Turner et al. 1999), compared to low energy systems, where differences in sediment characteristics are of ten more distinct inside vers es outside vegetation (Peterson et al. 1984). Specifically, at high energy sites with vegetation, sediments are generally of larger gain sizes and have lo wer organic matter content, in some cases by a factor of almost three, in comparison to low ener gy vegetated sites (M urphey & Fonseca 1995, Irlandi 1996). Thus, the extent to which S AV can modify sedimentation in SAV systems under different hydrodynamic conditions is in need of scrutiny. A small subset of studies surveyed, all in temperate Zostera seagrass systems, indicate that when SAV is pr esent at high energy sites, modi fications to sedimentation are

PAGE 129

116 often amplified, while at lower energy s ites, these are dampened. Many high energy systems often experience veloci ties well above the critical friction velocity necessary for resuspension of particle sizes present in that system, while low energy systems may only ever experience a few resuspension events as velocities rarely, if ever, reach above critical friction velocities. Widdows et al. (2008) determin ed that mass of sediment eroded at current speeds above critica l erosion velocity for sand (0.2 m s-1) was not only linked to presence of seagrass ( Zostera noltii ), but was inversely re lated to density of a seagrass canopy. Working at both high and low en ergy sites, Harlin et al. (1982) reported that sediment accreted in seagrass (Zostera marina) beds and eroded in nearby unvegetated locations under hi gh energy conditions, but found no difference in accretion and erosion between unvegetated locations and those with seagrass in low energy conditions. Therefore, for studies conducted in high energy systems, the presence of SAV may become especially critical to particle retention and stabilization, but may be of reduced importance when resuspension probabilities are minimal in low energy systems. The physical structure (i.e. flexibility of the canopy) of SAV canopies can also modify the interaction between flow and th e canopy and any resulting sedimentation. Stiff canopies, such as salt marsh grass canopi es, are highly effec tive at reducing flow and turbulence when in high shoot de nsities (Eckman 1983, Leonard & Croft 2006, Bouma et al. 2007), which translates to greater sediment trapping potenti al (Peralta et al. 2008). Conversely, more flexible seagrasses canop ies might be more efficient at reducing resuspension and erosion, especially as shoot densities increase (Per alta et al. 2008), as these canopies effectively create a barri er between the water column and canopy by bending under skimming flow conditions (F onseca & Fisher 1986). As previous

PAGE 130

117 discussed, reductions in resuspension often translate to reductions in sedimentation, which may account for the a large number of the studies surveyed (>40%) that found increases in seagrass to have a no effect (Kenworthy et al. 1982, Fonseca & Fisher 1986, Hasegawa et al. 2008) or a negative (Granata et al. 2001) effect on sedimentation (Table 2). This is in contrast to the mostly positive effects of increased salt marsh grass on sedimentation in some studies surveyed (Eckman 1983, Leonard & Croft 2006, Bouma et al. 2007). It is clear from the studies surv eyed (Table 2) that sedimentation in SAV systems should not be considered without first being placed in the context of hydrodynamic regime and extent of SAV structur e, as both factors, alone or combined, can greatly modify all facets of sediment ation (e.g. sediment properties, deposition, resuspension etc.) in SAV systems. Thes e studies highlight how the already highly complex relationship between SAV and sedimentation in SAV systems can be impacted positively (i.e. increased deposition under low flow conditions) or negatively (i.e. increased resuspension in high flows) depending on prevailing hydrodynamic conditions. Faunal Communities Faunal Communities in Vegetated and Unvegetated Habitats The previous sections focused on studies th at used sedimentation as a measure of ecosystem function in SAV systems and how biotic (SAV presence and structure) and abiotic (hydrodynamics) environmental factors can modify that function. Even more prominent in the literature (see Table 3 a nd Appendix: Table 21) is the use of faunal community characteristics, measured by an array of metrics (i.e. community composition, diversity, abundance, richness, evenness, bioma ss, mortality, growth rates), as a proxy to

PAGE 131

118 gauge levels of SAV ecosystem function as en vironmental factors in SAV systems vary. Of the studies surveyed, most (>70%) found a positive relationship between faunal community characteristics and presence of SAV (Appendix: Table 21). Where a positive link between presence of SAV and faunal co mmunity were discerned, some studies found species compositions of vegeta ted and unvegetated habitats to be quite similar, but the relative abundance and biomass of the species that compose these communities differed (Connolly 1997, Hirst and Attri ll 2008). This suggests that faunal communities may be responding to more than the pres ence of SAV, including other as pects of habitat, such as increased detritus, decreased predation pre ssure, reduced water flow, resulting in the observed differences in faunal community characteristics. Almost 75% of studies recording a po sitive link between presence of SAV and faunal community characteristics, either did not address hydrodynamic regime (>55%), did not directly measure the hydrodynamic regime (>15%), or if hydrodynamics were considered, reported no consistent effects of flow on faunal communities in SAV beds (<10%) (Appendix: Table 21). These studies that did find consistent hydrodynamic effects generally found that high flow conditions had a negative effect on faunal species and communities. High flows can reduce macr ofaunal densities in seagrass beds (Stoner 1980), while sheltered, low flow seagrass hab itats demonstrate increased macrofaunal abundances (Polte et al. 2005) and filter feeder ( Mercenaria mercenaria ) growth rates (Peterson et al. 1984) relative to that in unvegetated habi tats. Moreover, high flow conditions have been linked to increased grow th in filter feeders vi a to high food fluxes (Irlandi & Peterson 1991, Irlandi 1996) and enha nced abundances of fauna as a result of increased faunal delivery rates (Bartholomew 2002) in SAV systems. Thus, while the

PAGE 132

119 positive link between faunal species and communities and the presence of SAV is relatively well documented, the extent to which it is modified by prevailing hydrodynamic regime remains poorly investigated. Negative relationships or the lack or a relationship between SAV presence and faunal communities were also detected in some cases and generally resulted from unnaturally occurring SAV habitats (Appendix: Table 21). Arrival of an invasive marshgrass hybrid ( Spartina alterniflora and S. foliosa hybrid) initiated a chain of events, including reduced fl ow within the hybrid canopy, altered sediment composition to fine, organic rich particles, increased anoxia and sulfide concentration in the sediments, reduced bent hic macrofaunal invert ebrate survivorship, and decreased in faunal community measures (i.e. diversity, density, and recruitment) (Neira et al. 2006). Similarly, a study that added seagrass ( Zostera marina) mimics to normally unvegetated habitats detected a m easureable decrease in the abundance of epibenthic harpacticoid copepods, which was attr ibuted to the mechanical disturbance of sediments by the sweeping action of the seag rass blade mimics (Hicks 1989). Thus, those faunal species and communities that rely on SAV for refuge (i.e. habitat, food, reduced predation) generally have a positive response (e.g. increased diversity, biomass, production, and survival) to th e presence of SAV, but the presence of vegetation in normally unvegetated areas or arrival of novel or invasive vegetation can have severe negative impacts on both indi vidual faunal species, as well as the larger faunal community. Similar to those papers that detected a positive link between the presence of SAV and faunal community characteristics, most (>75%) studies that f ound either a negative, variable, or no effect of SAV presence on fauna either did not measure hydrodynamic

PAGE 133

120 regime or found no consistent e ffect of flow in the study re sults (Appendix: Table 21). Those papers that did address the effects of flow on faunal communities in SAV beds focused on either changes to infaunal comm unity compositions as a result of differing amounts of wave exposure in seagrass ( Zostera novazelandica ) habitats (Turner et al. 1999) or flow reductions as a consequence of the arrival of a novel SAV ( Spartina alterniflora and S. foliosa hybrid) and recorded cascad ing effects on the faunal community resulting in reduced benthic macrof aunal invertebrate surv ival (Neira et al. 2006). Thus, for those studies that found either no relationship or a variable relationship between the presence of SAV and fauna l communities, additional environmental variables (e.g. hydrodynamic regime, sediment properties, etc.) may become critical for interpreting possible implica tions for modified faunal co mmunity characteristics. Variation in SAV Structure and Faunal Communities The complex sets of interactions that govern the link between presence of SAV and effects on faunal communities may be modifi ed when SAV structural differences (i.e. shoot and leaf density, canopy height, leaf area index, leaf standing crop, biomass, percent cover) are considered. Over 70% of studies surveyed detected a positive relationship between some metr ic of SAV structure and one or more faunal community characteristics (Table 3). The response of faunal community characteristics to SAV structure appears not to be linear. A study by Homziak et al. (1982) determined that certain faunal community characteristics (i.e abundance, richness, diversity) began to plateau above specific seagrass (Halodule wrightii and Zostera marina) densities, no longer increasing with increasi ng shoot densities. On the lower end of SAV structure,

PAGE 134

121 Fonseca et al. (1996) measured faunal abundance in transplanted seagrass ( Halodule wrightii Syringodium filiforme and Thalassia testudinum ) beds equivalent to those in natural seagrass beds when tr ansplanted shoot densities equaled at least one third of natural shoot densities. This suggests that threshold le vel of structure may exist on the both the low and high ends, below or above wh ich effects of increased SAV structure on faunal communities are not discernable. Increased faunal abundances with increas ed SAV structure are thought to be related to the indirect effects of increase d detritus and other food sources, increased number of habitat niches, and decreased predation risk (Webster et al. 1998). Most of these studies focused on correlative relationship between SAV and faunal communities, with the underlying causes behind the SAV-faunal relationships often unexplained. It is important to note that half of the studies surveyed that determined either a negative or variable relationship, or no relationship at all, between SAV structure and faunal communities, found that prevailing hydrodynamic conditions explained the relationship between SAV structure and their associated faunal communities (Table 3). For example, Irlandi and Peterson (1991) and Irlandi (1996) found no link between high amounts of SAV ( Halodule wrightii and Zostera marina ) structure (shoot density and blade height) and faunal ( Mercenaria mercenaria ) community characteristics (shell growth), but measured increased clam shell growth as f ood flux rates to the seagrass beds that clams inhabited were increased under high flow conditions. Thus, to understand better the underlying causes behind the relationship between faunal communities and vegetation structure, links to hydrodynamics may be useful.

PAGE 135

122 SAV, Faunal Comminutes, and the Effects of Flow Similar to modifications witnessed in sedimentation across hydrodynamic regimes in SAV beds, links between SAV and faunal communities are often directed by prevailing hydrodynamic conditions. Studies quantifying the effects of hydrodynamics on the positive relationship between SAV stru cture and faunal community characteristics generally have found the SAV-fauna link to be strengthened by low flow conditions and weakened as flow increased (Table 3). Across a range of faunal species and sizes, reductions in flow had a positive effect on faunal communities in SAV systems. Gambi et al. (1998) measured lower polychaete de nsity and a high amount of faunal patchiness with reductions in seagrass (Cymodocea nodosa and Zostera noltii ) structure (shoot density, leaf area index, and leaf standing crop), and high shrimp (Penaeus duorarum ) abundances were present in habitats with greater pe rcent seagrass ( Halodule wrightii and Zostera marina) cover in a study by Murphey and Fonseca (1995). The positive response of faunal species has been attributed to the ab oitc (i.e. high sedimentary silt-clay content) and biotic (i.e. high amounts of organic matter and dense seagrass) habitat characteristics present under the low energy conditions. One study reported that under high flow conditions epifaunal abundance increased as SAV ( Zostera marina mimic) structure (number of blades, amount of cover per ar ea, surface area of blades, space between blades, and blade width and le ngth) increased and speculated, but did not measure, that increases in food fluxes to filter feeders enhanc ed nutrient fluxes that promoted growth of algal food resources and increased larval delivery augmented epifaunal abundance (Bartholomew 2002). This points out an im portant distinction: fauna that rely on processes such as high fluid fluxes for food and larval recruitment appear to respond

PAGE 136

123 positively to high flow conditions, while those that rely on stable sediments with high silt-clay and organic matter content for their habitat seem to respond better to low flow conditions. Fifty percent of the studies surveyed record ed either a negative or variable link, or no link at all, between SAV structure and faunal communities, and the lack of relationship was able to be explained, at least partially, by prevailing hydrodynamic conditions (Table 3). Irlandi and Peterson (1991) and Irlandi (1996) both of which did not discern a relationship between SAV structure and fauna, measured increased filter feeder ( Mercenaria mercenaria ) growth rates under high flow conditions as the flux of food particles to SAV beds was enhanced. Lack of a consistent relationship between the abundance of macrofauna in seag rass beds from one year to the next (Hovel et al. 2002) was attributed reduction of fauna as a result of high relative wave exposure, which may have reduced faunal feeding rates, larval availability and settlement, or faunal locomotion. Those studies that found negativ e relationships between SAV structure and faunal abundances ascribed their findings to high larval recruitment resulting from increased water flux to SAV beds (Eckman 198 7) and subsequent reductions in flow when the canopy was encountered (Bologna & Heck 2002). Thus, to understand better the complex set of interaction that govern faunal communities within SAV beds and to explain the findings of many of these studies, links to hydrodynamic conditions must be considered.

PAGE 137

124 CONCLUSIONS & SYNTHESIS Complex set of interactions dictate relationships among SAV, hydrodynamic regime, and proposed ecosystem functions (i .e. site of faunal accumulation, sediment deposition and deposition). Review of published papers clearl y illustrates that there is only a small percentage of papers addressing these interrelationships. While it is clear that SAV-sediment relationships are impacted by hydrodynamic regime, these impacts are not necessarily consis tent within categories (i.e. high or low) of SAV density or of flow. For example, as S AV densities increase, fl ow is increasingly attenuated. When flow is completely atte nuated sedimentation increases little with further increases in SAV (Peralta et al. 2008) and can negatively aff ect sedimentation by reducing mixing between the water column a nd the canopy. In contrast, under low flows, sedimentary accretion and erosion may be equivalent in vegetated and adjacent unvegetated habitats (Harlin et al. 1982). Consequently, th reshold values of both flow and SAV densities increase the complexity of defining clear relationship between measures of ecosystem function, such as sedimentation vegetation structure, and hydrodynamics in SAV systems. Further studies that measure sedimentation in a more direct manner (i.e. measures of sediment tr apping or concentrations of suspended solids in the water column) both in controlled flume and natural field setti ngs, in a variety of vegetation morphologies, and across a range of flow speeds and vegetation densities are needed to define clearly the relationship betw een SAV structure and flow to infer levels of ecosystem function (i.e. sediment deposition) in SAV systems. Likewise, there is a need to recognize th at variation in SAV-fauna relationships may differ across hydrodynamic regimes. Most (75%) of the studies surveyed in this

PAGE 138

125 review did not consider the effects of hydrodynamics nor acknowledge that hydrodynamics can modify these relationships Those studies that did demonstrate altered faunal responses by flow found the fauna-flow relationship dependened on the extent to which habitat charac teristics (e.g. food availabilit y, sediment properties) that these fauna rely on can be modified by flow. Faunal species that are highly dependent on delivery of food particles or re cruitment of larvae via the wa ter column, such as filter feeders or broadcast spawners, generally re spond positively to increased flow more so than changes in any other habitat charact eristic (Eckman 1987, Irlandi & Peterson 1991, Irlandi 1996). Species that are closely linked to the sediments in SAV systems, such as infauna and epibenthic fauna, generally respond positively to low flow conditions, which promote increased sedimentary silt-clay and organic matter content (Murphey & Fonseca 1995, Gambi et al. 1998). While it is clear that the presence and stru cture of SAV beds have a positive effects on faunal communities, more detailed studies are needed, to quantify the underlying factors (e.g. hydrodynamic regime, sediment characteristics, and amount of SAV structure) that may be controlling faunal co mmunity characteristics in SAV systems. IMPLICATIONS This review highlights the importance of hydrodynamics in shaping SAV communities, both through physical processes (e.g. deposition, sediment structure, etc.) and biological structure and function (e .g. faunal community composition). Many of those studies that do acknowledge hydrodynamics as an important environmental factor or modifier in SAV systems measure hydrodyna mics either indirectly (i.e. sediment

PAGE 139

126 characteristics) or in a relatively simplisti c manner (i.e. clod cards). Conversely, studies that make detailed hydrodynamic measuremen ts commonly only address the effects of SAV on flow without considering the potential cascading effects on other habitat functions or processes (e.g. sedimentation, predation, community structure, etc.). Frequently, these studies are also conducted outside of context of a natural setting (i.e. in a flume in the laboratory). Therefore, comp arisons between field and laboratory flume studies that evaluate the effects of hydrodynamics on SAV communities can prove problematic as measures of hydrodynamics (i.e detailed versus simplistic measurements) or habitat functions or processes may not be analogous. Results from laboratory flume and field st udies that investigate the effects of hydrodynamics on sedimentation and faunal communities often differ and/or may contradict each other. Laboratory studies have the advant age of having more controlled conditions (i.e. specific SAV densities and hydr odynamic conditions), but can be lacking when attempting to replicate natural field conditions (i.e. constrained by length, width, and depth of flume). This makes it problematic to extrapolate from the laboratory to the field if flow conditions are hi ghly modified or if canopy c onditions do not match those of the field. Promisingly, some recent studies have successfully incorporated additional habitat processes, such as sedimentation, in to laboratory flume studies addressing the SAV-flow relationship (Hendriks et al. 2008). Other studies ha ve incorporated the results of complementary laboratory and field experiments in a single study to determine how applicable laboratory flume results are both to natural processes and to better interpret the results of field experiments in the context of what is known from controlled laboratory experiments (Bouma et al. 2007, Widdows et al. 2008). More comprehensive studies

PAGE 140

127 such as these are needed before results from laboratory flume studies should be used to exclusively answer questions concerning the ecological function SAV beds and how this translates into the important eco system services they provide. It is commonly assumed that equiva lent amounts of SAV structure provide equivalent levels of ecosystem function and/or services (Fonseca et al. 1996). This underlies the goals of many restoration studies that seek to restore specific levels of ecosystem function and/or accompanying servic es (Fonseca et al. 2000). However, this assumption does not necessarily hold true under varying hydrodynamic conditions Vegetation planting densities that result in successful restoration efforts under low flow conditions may not be sufficient for vegetation survival and persistence when equivalent vegetation densities are exposed to high flow s. Under high flow conditions, seagrass transplant survival decrease (Bos & van Katwijk 2007). Mo reover, if restoration is conducted in an area that has a different hydr odynamic regime than that of the damaged setting, then plans for compensatory leve ls of recovery, such as that commonly determined through Habitat Equivalency Analys es (e.g. Fonseca et al. 2000) may need to be adjusted as well. The presence and/or structural char acteristics (e.g. ca nopy height, density, morphology, etc.) of SAV beds are clearly im portant to measure of ecosystem function (i.e. sedimentation and faunal community characteristics), but hydrodynamic conditions (e.g. tide v. wave dominated, high v. low ener gy, etc.) encountered in these systems can have quantifiable implications pertaining to the survival and persis tence of both the SAV bed (e.g. mass transfer of nutrients, water cl arity, etc.) and the co mmunity that inhabit and utilize the bed (e.g. faunal recruitment, susp ension feeding, predation, etc.). There is

PAGE 141

128 a need to consider hydrodynamic setting when selecting sites for rest oration and goals to restore levels of ecosystem services, as the loss of ecosystem services reflected in SAV loss, even for same species, could have a greater or lesser im pact depending on the hydrodynamic setting. Results of this survey provide convincing information that the potential influence of hydrodynamics on measur es of ecosystem func tion is relatively understudied but merits evaluation before accu rate assessments of ecosystem services provided by SAV systems can be determined.

PAGE 142

129 Literature Cited Abdelrhman MA (2003) Effect of eelgrass Zostera marina canopies on flow and transport. Marine Ecolog y-Progress Series 248:67-83 Ackerman JD, Okubo A (1993) Reduced mixing in a marine macrophyte canopy. Functional Ecology 7:305-309 Agawin NS, Duarte CM (2002) Evidence of di rect particle trappi ng by a tropical seagrass meadow. Estuaries 25:1205-1209 Allen BJ, Williams SL (2003) Native eelgrass Zostera marina controls growth and reproduction of an invasive mussel through food limitation. Marine EcologyProgress Series 254:57-67 Almasi MN, Hoskin CM, Reed JK, Milo J ( 1987) Effects of natu ral and artificial Thalassia on rates of sedimentation. Journal of Sedimentary Petrology 57:901906 Anderson SM, Charters AC (1982) A fluid dynamics study of seawater flow through Gelidium nudifrons Limnology and Oceanography 27:399-412 Asmus H, Asmus R (2000) Material exchange and food web of seagra ss beds in the SyltRomo Bight: how significant are community changes at the ecosystem level? Helgoland Marine Research 54:137-150 Attrill MJ, Strong JA, Rowden AA (2000) Are macroinvertebrate communities influenced by seagrass structural complexity? Ecography 23:114-121 Backhaus JO, Verduin JJ (2008) Simulating th e interaction of s eagrasses with their ambient flow. Estuarine, Coastal and Shelf Science 80:563-572 Barko JW, Smart RM (1983) Effects of orga nic-matter additions to sediment on the growth of aquatic plants. Journal of Ecology 71:161-175 Barko JW, Smart RM (1986) Sediment-relate d mechanisms of growth limitation in submersed macrophytes. Ecology 67:1328-1340 Bartholomew A (2002) Faunal colonization of ar tificial seagrass plots: The importance of surface area versus space size relative to body size. Estuaries 25:1045-1052

PAGE 143

130 Bell JD, Westoby M (1986a) Abundance of Macrofauna in Dense Seagrass Is Due to Habitat Preference, Not Pr edation. Oecologia 68:205-209 Bell JD, Westoby M (1986b) Importance of local changes in leaf height and density to fish and decapods associated with seagra sses. Journal of Experimental Marine Biology and Ecology 104:249-274 Bell SS, Hall MO, Soffian S, Madley K (2002) Assessing the impact of boat propeller scars on fish and shrimp utilizing seag rass beds. Ecological Applications 12:206217 Bell SS, Fonseca MS, Kenworthy WJ (2008) Dynamics of a subtropical seagrass landscape: links between disturbance a nd mobile seed banks. Landscape Ecology 23:67-74 Bjorndal KA (1980) Nutrition and grazi ng behavior of the green turtle Chelonia Mydas Marine Biology 56:147-154 Bologna PAX, Heck KL (2002) Impact of habitat edges on density and secondary production of seagrass-associat ed fauna. Estuaries 25:1033-1044 Borg JA, Rowden AA, Attrill MJ, Schembri PJ, Jones MB (2006) Wanted dead or alive: high diversity of macroinvertebrates associated with living and 'dead' Posidonia oceanica matte. Marine Biology 149:667-677 Bos AR, van Katwijk MM (2007) Planting density, hydrodynamic exposure and mussel beds affect survival of transplanted in tertidal eelgrass. Marine Ecology-Progress Series 336:121-129 Bos AR, Bouma TJ, de Kort GLJ, van Ka twijk MM (2007) Ecosystem engineering by annual intertidal seagrass beds: sedime nt accretion and modification. Estuarine Coastal and Shelf Science 74:344-348 Bouma TJ, van Duren LA, Temmerman S, Clav erie T, Blanco-Garcia A, Ysebaert T, Herman PMJ (2007) Spatial flow and sedi mentation patterns w ithin patches of epibenthic structures: combining fi eld, flume and modelling experiments. Continental Shelf Research 27:1020-1045 Bowden DA, Rowden AA, Attrill MJ (2001) Eff ect of patch size and in-patch location on the infaunal macroinvertebrate assemblages of Zostera marina seagrass beds. Journal of Experimental Mari ne Biology and Ecology 259:133-154 Bradley K, Houser C (2009) Relative velocity of seagrass blades: implications for wave attenuation in low-energy environments. Journal of Geophysical Research Earth Surface 114

PAGE 144

131 Brun FG, van Zetten E, Cacabelos E, Boum a TJ (2009) Role of two contrasting ecosystem engineers ( Zostera noltii and Cymodocea nodosa) on the food intake rate of Cerastoderma edule. Helgoland Marine Research 63:19-25 Bulthuis DA, Brand GW, Mobley MC (1984) Suspended sediments and nutrients in water ebbing from seagrass-covered and denuded tidal mudflats in a Southern Australian embayment. Aquatic Botany 20:257-266 Castel J, Labourg PJ, Escaravage V, Auby I, Garcia ME (1989) Influence of seagrass beds and oyster parks on the abundance a nd biomass patterns of meiobenthos and macrobenthos in tidal flats. Estuarin e Coastal and Shelf Science 28:71-85 Chen SN, Sanford LP, Koch EW, Shi F, North EW (2007) A nearshore model to investigate the effects of seagrass bed geometry on wave attenuation and suspended sediment transport. Estuaries Coasts 30:296-310 Como S, Magni P, Baroli M, Casu D, De Fa lco G, Floris A (2008) Comparative analysis of macrofaunal species ri chness and composition in Posidonia oceanica Cymodocea nodosa and leaf litter beds. Marine Biology 153:1087-1101 Connolly RM (1995) Effects of removal of seagrass canopy on assemblages of small, motile invertebrates. Marine Ecology-Progress Series 118:129-137 Connolly RM, Butler AJ (1996) The effects of altering seagrass canopy height on small, motile invertebrates of shallow Medi terranean embayments. Marine Ecology 17:637-652 Connolly RM (1997) Differences in com position of small, motile invertebrate assemblages from seagrass and unvegetated habitats in a southern Australian estuary. Hydrobiologia 346:137-148 Costanza R, dArge R, deGroot R, Farber S, Grasso M, Hannon B, Limburg K, Naeem S, ONeill RV, Paruelo J, Raskin RG, Sutton P, vandenBelt M (1997) The value of the world's ecosystem services a nd natural capital. Nature 387:253-260 Creed JC, Amado GM (1999) Disturbance and recovery of the macroflora of a seagrass ( Halodule wrightii Ascherson) meadow in the Abrolhos Marine National Park, Brazil: an experimental evaluation of an chor damage. Journal of Experimental Marine Biology and Ecology 235:285-306 Danovaro R, Gambi C, Mirto S (2002) Me iofaunal production and energy transfer efficiency in a seagrass Posidonia oceanica bed in the western Mediterranean. Marine Ecology-Progress Series 234:95-104

PAGE 145

132 Dauby P, Bale AJ, Bloomer N, Canon C, Li ng RD, Norro A, Robertson JE, Simon A, Theate JM, Watson AJ, Frankignoulle M (1995) Particle fluxes over a Mediterranean seagrass bed: a one year case study. Marine Ecology-Progress Series 126:233-246 Deegan LA, Wright A, Ayvazian SG, Finn JT, Golden H, Merson BR, Harrison J (2002) Nitrogen loading alters seagrass ecosystem structure and support of higher trophic levels. Aquatic Conservation: Marine and Freshwater Ecosystems 12:193-212 Denny MW (1988) Biology and the mechanics of the wave-swept environment, Vol. Princeton University Press, Princeton, N.J. Duarte CM, Benavent E, Sanchez MD (1999) The microcosm of particles within seagrass Posidonia oceanica canopies. Marine Ecology-Progress Series 181:289-295 Duarte CM (2002) The future of seagrass meadows. Environmental Conservation 29:192206 Eckman JE (1983) Hydrodynamic processes affecting benthic recruitment. Limnology and Oceanography 28:241-257 Eckman JE (1987) The role of hydrodynamics in recruitment, growth, and survival of Argopecten irradians (L) and Anomia simplex (Dorbigny) within eelgrass meadows. Journal of Experimental Marine Biology and Ecology 106:165-191 Eckrich CE, Holmquist JG (2000) Trampling in a seagrass assemblage: direct effects, response of associated fauna, and the role of substrate characteristics. Marine Ecology-Progress Series 201:199-209 Edgar GJ (1990) The influence of plant stru cture on the species richness, biomass and secondary production of macrofaunal a ssemblages associated with Western Australian seagrass beds. Journal of Experimental Marine Biology and Ecology 137:215-240 Edgar GJ, Robertson AI (1992) The influence of seagrass structur e on the distribution and abundance of mobile ep ifauna: pattern and process in a Western Australian Amphibolis bed. Journal of Experimental Marine Biology and Ecology 160:13-31 Edgar GJ, Shaw C, Watson GF, Hammond LS (1994) Comparisons of species richness, size-structure and production of benthos in vegetated and unvegetated habitats in Western-Port, Victoria. Journal of E xperimental Marine Biology and Ecology 176:201-226

PAGE 146

133 Edgar GJ (1999) Experimental analysis of structural versus tr ophic importance of seagrass beds: I. effects of macrofauna l and meiofaunal invertebrates. Vie Et Milieu-Life and Environment 49:239-248 Erftemeijer PLA, Middelburg JJ (1995) Mass ba lance constraints on nu trient cycling in a tropical seagrass beds. Aquatic Botany 50:21-36 Eyre BD, Ferguson AJP (2002) Comparison of carbon production and decomposition, benthic nutrient fluxes and denitrification in seagrass, phytoplankton, benthic microalgaeand macroalgae-dominated warm-temperate Australian lagoons. Marine Ecology Progress Series 229:43-59 Folkard AM (2005) Hydrodynamics of model Posidonia oceanica patches in shallow water. Limnology and Oceanography 50:1592-1600 Fonseca MS, Fisher JS, Zieman JC, Thayer GW (1982) Influence of the seagrass, Zostera marina L., on current flow. Estuarine, Coastal and Shelf Science 15:351-364 Fonseca MS, Zieman JC, Thayer GW, Fisher JS (1983) The role of current velocity in structuring eelgrass ( Zostera marina L.) meadows. Estuarine Coastal and Shelf Science 17:367-380 Fonseca MS, Fisher JS (1986) A comparison of canopy friction and sediment movement between four species of seagrass with reference to their ecology and restoration. Marine Ecology Progress Series 29:15-22 Fonseca MS (1989) Sediment stabilization by Halophila decipiens in comparison to other seagrasses. Estuarine Coastal and Shelf Science 29:501-507 Fonseca MS, Cahalan JA (1992) A Preliminary evaluation of wave attenuation by four species of seagrass. Estuarine Coastal and Shelf Science 35:565-576 Fonseca M (1996) The role of seagrasses in ne arshore sedimentary processes: a review, Vol. John Wiley & Sons Ltd, Chichester Fonseca MS, Meyer DL, Hall MO (1996) Development of planted seagrass beds in Tampa Bay, Florida, USA .2. Faunal co mponents. Marine Ecology-Progress Series 132:141-156 Fonseca MS, Julius BE, Kenworthy WJ (2000) Integrating biology and economics in seagrass restoration: how much is enough and why? Ecological Engineering 15:227-237 Fonseca MS, Koehl MAR (2006) Flow in seagrass canopies: The influence of patch width. Estuarine Coastal and Shelf Science 67:1-9

PAGE 147

134 Fourqurean JW, Rutten LM (2004) The impact of Hurricane Georges on soft-bottom, back reef communities: Siteand species-specific effects in south Florida seagrass beds. Bulletin of Marine Science 75:239-257 Francour P, Ganteaume A, Poulain M (1999) Effects of boat anchoring in Posidonia oceanica seagrass buds in the Port-Cros National Park (north-western Mediterranean sea). Aquatic Conservatio n-Marine and Freshwater Ecosystems 9:391-400 Frost MT, Rowden AA, Attr ill MJ (1999) Effect of habitat fragmentation on the macroinvertebrate infaunal communiti es associated with the seagrass Zostera marina L. Aquatic Conservation-Marine and Freshwater Ecosystems 9:255-263 Gacia E, Granata TC, Duarte CM (1999) An a pproach to measurement of particle flux and sediment retention within seagrass ( Posidonia oceanica ) meadows. Aquatic Botany 65:255-268 Gacia E, Duarte CM (2001) Sediment retention by a Mediterranean Posidonia oceanica meadow: The balance between deposition and resuspension. Estuarine Coastal and Shelf Science 52:505-514 Gacia E, Duarte CM, Marba N, Terrados J, Kennedy H, Fortes MD, Tri NH (2003) Sediment deposition and production in SE -Asia seagrass meadows. Estuarine Coastal and Shelf Science 56:909-919 Gambi MC, Nowell ARM, Jumars PA (1990) Flume observations on flow dynamics in Zostera marina (eelgrass) beds. Marine Ecology-Progress Series 61:159-169 Gambi MC, Conti G, Bremec CS (1998) Polychaete distribution, diversity and seasonality related to seagrass cover in shallow soft bottoms of the Tyrrhenian Sea (Italy). Scientia Marina 62:1-17 Ghisalberti M, Nepf HM (2002) Mixing layers and coherent struct ures in vegetated aquatic flows. Journal of Geophysical Research-Oceans 107 Grady JR (1981) Properties of sea grass and sand flat sediments from the intertidal zone of St. Andrew Bay, Florida. Estuaries 4:335-344 Granata TC, Serra T, Colomer J, Casamitjana X, Duatre CM, Gacia E (2001) Flow and particle distributions in a nearshore s eagrass meadow before and after a storm. Marine Ecology-Progress Series 218:95-106

PAGE 148

135 Grizzle RE, Short FT, Newell CR, Hoven H, Kindblom L (1996) Hydrodynamically induced synchronous waving of seagrasses: 'monami' and its possible effects on larval mussel settlement. Journal of Experimental Marine Biology and Ecology 206:165-177 Harlin MM, Thornemiller B, Boothroyd JC (1982) Seagrass-sediment dynamics of a flood-tidal delta in Rhode Island (USA). Aquatic Botany 14:127-138 Hasegawa N, Hori M, Mukai H (2008) Seasonal changes in eelgrass functions: current velocity reduction, prevention of sediment resuspension, and control of sedimentwater column nutrient flux in relation to eelgrass dynamics. Hydrobiologia 596:387-399 Heck KL, Wetstone GS (1977) Habitat Comp lexity and Invertebra te Species Richness and Abundance in Tropical Seagrass Meadows. J ournal of Biogeography 4:135142 Heck KL, Able KW, Roman CT, Fahay MP (1995) Composition, abundance, biomass, and production of macrofauna in a Ne w England estuary: comparisons among eelgrass meadows and other nursery habitats. Estuaries 18:379-389 Heiri O, Lotter AF, Lemcke G (2001) Loss on ignition as a method for estimating organic and carbonate content in sediments: repr oducibility and comparability of results. Journal of Paleolimnology 25:101-110 Heiss WM, Smith AM, Probert PK (2000) Infl uence of the small intertidal seagrass Zostera novazelandica on linear water flow and sediment texture. N Z J Mar Freshw Res 34:689-694 Hendriks IE, Tomas Sintes, Tjeerd J. Bouma, and Carlos M. Duarte (2008) Experimental assessment and modeling evaluation of the effects of the seagrass Posidonia oceanica on flow and particle trapping. Marine Ecology Progress Series 356:163173 Hicks GRF (1989) Does epibenthic structur e negatively affect meiofauna. Journal of Experimental Marine Bi ology and Ecology 133:39-55 Hily C, Bouteille M (1999) Modifications of the specific diversity and feeding guilds in an intertidal sediment col onized by an eelgrass meadow (Zostera marina) (Brittany, France). Comptres Rendus De L Academie Des Sciences Serie IIISciences De La Vie-Life Sciences 322:1121-1131 Hirst JA, Attrill MJ (2008) Small is beautiful: an inverted view of habitat fragmentation in seagrass beds. Estuarine Coastal and Shelf Science 78:811-818

PAGE 149

136 Homziak J, Fonseca MS, Kenworthy WJ (1982) Macrobenthic community structure in a transplanted eelgrass (Zostera marina ) meadow. Marine Ecology-Progress Series 9:211-221 Hovel KA, Fonseca MS, Myer DL, Kenwort hy WJ, Whitfield PE (2002) Effects of seagrass landscape structure, structur al complexity and hydrodynamic regime on macrofaunal densities in North Carolin a seagrass beds. Marine Ecology-Progress Series 243:11-24 Hughes JE, Deegan LA, Wyda JC, Weaver MJ, Wright A ( 2002) The effects of eelgrass habitat loss on estuarine fish communiti es of southern New England. Estuaries 25:235-249 Ikeda S, Kanazawa M (1996) Three-dimensiona l organized vortices above flexible water plants. Journal of Hydraulic Engineering-Asce 122:634-640 Irlandi EA, Peterson CH (1991) Modificati on of animal habitat by large plants: mechanisms by which seagrasses influen ce clam growth. Oecologia 87:307-318 Irlandi EA (1996) The effects of seagrass pa tch size and energy regime on growth of a suspension-feeding bivalve. Journa l of Marine Research 54:161-185 Jackson JBC (2001) What was natural in the coastal oceans? Proceeding of the National Academy of Sciences of the Unite d States of America 98:5411-5418 James WF, Best EP, Barko JW (2004) Sediment resuspensi on and light attenuation in Peoria Lake: can macrophytes improve water quality in this shallow system? Hydrobiologia 515:193-201 Jaschinski S, Sommer U (2008) Top-down and bottom-up control in an eelgrass-epiphyte system. Oikos 117:754-762 Jenkins GP, Walker-Smith GK, Hamer PA (2002) Elements of habitat complexity that influence harpacticoid copepods associated with seagrass beds in a temperate bay. Oecologia 131:598-605 Johnson MW, Jr. KLH (2006) Effects of hab itat fragmentation per se on decapods and fishes inhabiting seagrass meadows in the northern Gulf of Mexico. Marine Ecology Progress Series 306:233-246 Kalbfleisch WBC, Jones B (1998) Sedimentology of shallow, hurricane-affected lagoons: Grand Cayman, British West Indies. Jour nal of Coastal Research 14:140-161

PAGE 150

137 Kenworthy WJ, Zieman JC, Thayer GW (1982) Evidence for the influence of seagrasses on the benthic nitrogen cycle in a coastal plain estuary near Beaufort, North Carolina (USA). Oecologia 54:152-158 Koch EW (1999) Sediment resuspension in a shallow Thalassia testudinum banks ex Konig bed. Aquatic Botany 65:269-280 Koch EW, Gust G (1999) Water flow in tideand wave-dominated beds of the seagrass Thalassia testudinum Marine Ecology-Progr ess Series 184:63-72 Koch EW, Barbier EB, Silliman BR, Reed DJ, Perillo GME, Hacker SD, Granek EF, Primavera JH, Muthiga N, Polasky S, Halpern BS, Kennedy CJ, Kappel CV, Wolanski E (2009) Non-linearity in ecos ystem services: temporal and spatial variability in coastal protect ion. Front Ecol Environ 7:29-37 Lee SY, Fong CW, Wu RSS (2001) The effects of seagrass ( Zostera japonica ) canopy structure on associated fauna: a study usi ng artificial seagrass units and sampling of natural beds. Journal of Experimental Marine Biology and Ecology 259:23-50 Leonard GH, Levine JM, Schmidt PR, Bert ness MD (1998) Flow-driven variation in intertidal community structure in a Maine estuary. Ecology 79:1395-1411 Leonard LA, Luther MA (1995) Flow hydrodynamics in tidal marsh canopies. Limnology and Oceanography 40:1474-1484 Leonard LA, Croft AL (2006) The effect of standing biomass on flow velocity and turbulence in Spartina alterniflora canopies. Estuarine Coastal and Shelf Science 69:325-336 Lewis FG, Stoner AW (1983) Distribution of macrofauna within seagrass beds: an explanation for patterns of abundance. Bulletin of Marine Science 33:296-304 Lewis FG (1984) Distribution of macrobenthic crustaceans associated with Thalassia, Halodule and bare sand substrata. Marine Ecology-Progress Series 19:101-113 Luhar M, Rominger J, Nepf H (2008) Intera ction between flow, tr ansport and vegetation spatial structure. Environ Fluid Mech 8:423-439 Lynts GW (1966) Relationship of sediment-s ize distribution to ecologic factors in Buttonwood Sound, Florida Bay. Journal of Sedimentary Petrology 36:66-74 Madsen JD, Chambers PA, James WF, Koch EW, Westlake DF ( 2001) The interaction between water movement, sediment dynamics and submersed macrophytes. Hydrobiologia 444:71-84

PAGE 151

138 Madsen TV, Warncke E (1983) Velocities of currents around and within submerged aquatic vegetation. Archiv fur Hydrobiologie 97:389-394 Maltese A, Cox E, Folkard AM, Ciraolo G, La Loggia G, Lombardo G (2007) Laboratory measurements of flow and turbulence in discontinuous distributions of ligulate seagrass. Journal of Hydrau lic Engineering-Asce 133:750-760 Marba N, Duarte CM (1997) Inte rannual changes in seagrass ( Posidonia oceanica ) growth and environmental change in th e Spanish Mediterranean littoral zone. Limnology and Oceanography 42:800-810 Marshall N, Lukas K (1970) Preliminary observations on the properties of bottom sediments with and without eelgrass, Zostera marina cover. Proceedings of the National Shellfisheries Association 60:107-111 Mellors J, Marsh H, Carruth ers TJB, Waycott M (2002) Te sting the sediment-trapping paradigm of seagrass: do seagrasses influence nutrient status and sediment structure in tropical intertidal environm ents? Bulletin of Marine Science 71:12151226 Moller I (2006) Quantifying saltmarsh vege tation and its effect on wave height dissipation: results from a UK East coas t saltmarsh. Estuarine Coastal and Shelf Science 69:337-351 Murphey PL, Fonseca MS (1995) Role of high and low energy seagrass beds as nursery areas for Penaeus duorarum in North Carolina. Marine Ecology-Progress Series 121:91-98 Nagelkerken I, Roberts CM, van der Velde G, Dorenbosch M, van Riel MC, de la Morinere EC, Nienhuis PH (2002) How important are mangroves and seagrass beds for coral-reef fish? The nursery hypot hesis tested on an is land scale. Marine Ecology-Progress Series 244:299-305 Nakamua Y, Sano M (2005) Comparison of invertebrate abundance in a seagrass bed and adjacent coral and sand areas at Amitori Bay, Iriomote Island, Japan. Fisheries Science 71:543-550 Neckles HA, Short FT, Barker S, K opp BS (2005) Disturbance of eelgrass Zostera marina by commercial mussel Mytilus edulis harvesting in Maine: dragging impacts and habitat recovery. Marine Ecology-Progress Series 285:57-73 Neira C, Grosholz ED, Levin LA, Blake R ( 2006) Mechanisms generating modification of benthos following tidal flat invasion by a Spartina hybrid. Ecological Applications 16:1391-1404

PAGE 152

139 Nepf HM (1999) Drag, turbul ence, and diffusion in flow through emergent vegetation. Water Resources Research 35:479-489 Nepf HM, Vivoni ER (2000) Flow structure in depth-limited, vegeta ted flow. Journal of Geophysical Research-Oceans 105:28547-28557 Neumeier U, Ciavola P (2004) Flow resistance and associated sedimentary processes in a Spartina maritima salt-marsh. Journal of Coastal Research 20:435-447 Nikora VI, Suren AM, Brow n SLR, Biggs BJF (1998) The effects of the moss Fissidens rigidulus (Fissidentaceae: Musci) on near-bed flow structure in an experimental cobble bed flume. Limnology and Oceanography 42:1321-1331 O'Gower AK, Wacasey JW (1967) Animal communities associated with Thalassia diplanthera and sand beds in Biscayne Bay I. Analysis of communities in relation to water movements. Bulletin of Marine Science 17:175-210 Orth RJ (1973) Benthic in fauna of eelgrass Zostera marina beds. Chesapeake Science 14:258-269 Orth RJ (1977) The importance of sediment stability in seagrass communities, Vol. University of South Carolina Press, Columbia Orth RJ, Carruthers TJB, Dennison WC, Duarte CM, Fourqurean JW, Heck KL, Hughes AR, Kendrick GA, Kenworthy WJ, Olyarnik S, Short FT, Waycott M, Williams SL (2006) A global crisis for seagra ss ecosystems. Bioscience 56:987-996 Paling EI, van Keulen M, Wheeler KD, Ph illips J, Dyhrberg R (2003) Influence of spacing on mechanically transplanted seagrass survival in a high wave energy regime. Restoration Ecology 11:56-61 Palmer MA (1986) Hydrodynamics and struct ure: interactive effects on meiofauna dispersal. Journal of Experimental Marine Biology and Ecology 104:53-68 Peralta G, van Duren LA, Morris EP, Bouma TJ (2008) Consequences of shoot density and stiffness for ecosystem engineer ing by benthic macrophytes in flow dominated areas: a hydrodynamic flume st udy. Marine Ecology-Progress Series 368:103-115 Peterson CH (1982) Clam predation by whelks ( Busycon Spp): experimental tests of the importance of prey size, prey densit y, and seagrass cover. Marine Biology 66:159-170

PAGE 153

140 Peterson CH, Summerson HC, Duncan PB ( 1984) The influence of seagrass cover on population structure and indivi dual growth rate of a suspension-feeding bivalve, Mercenaria mercenaria Journal of Marine Research 42:123-138 Peterson CH, Luettich RA, Micheli F, Sk illeter GA (2004) Attenuation of water flow inside seagrass canopies of differing structure. Marine Ecology-Progress Series 268:81-92 Pihl L (1986) Exposure, Vegetation and se diment as primary factors for mobile epibenthic faunal community structure a nd production in shallow marine soft bottom areas. Neth J Sea Res 20:75-83 Polte P, Schanz A, Asmus H (2005) Effects of current exposure on habitat preference of mobile 0-group epibenthos for intertidal seagrass beds ( Zostera noltii ) in the northern Wadden Sea. Estuarine Coastal and Shelf Science 62:627-635 Reed BJ, Hovel KA (2006) Seagrass habitat di sturbance: how loss a nd fragmentation of eelgrass Zostera marina influences epifaunal abundance and diversity. Marine Ecology-Progress Series 326:133-143 Reich KJ, Worthy GAJ (2006) An isotopic assessment of the feeding habits of freeranging manatees. Marine Ecology-Progress Series 322:303-309 Robles CD, Alvarado MA, Desharnais RA (2001) The shifting balance of littoral predator-prey interaction in regimes of hydrodynamic stress. Oecologia 128:142152 Sand-Jensen K, Mebus JR (1996) Fine-scale pa tterns of water velocity within macrophyte patches in streams. Oikos 76:169-180 Santos SL, Simon JL (1974) Distribution a nd abundance of Polychaetous Annelids in a South Florida estuary. Bulletin of Marine Science 24:669-689 Schanz A, Polte P, Asmus H (2002) Cascadi ng effects of hydrodynamics on an epiphytegrazer system in intertid al seagrass beds of the Wadden Sea. Marine Biology 141:287-297 Sirota L, Hovel KA (2006) Simulated eelgrass Zostera marina structural complexity: effects of shoot length, shoot densit y, and surface area on the epifaunal community of San Diego Bay, California, USA. Marine Ecology-Progress Series 326:115-131 Scoffin TP (1970) Trapping and Binding of Subtidal Carbonate Sediments by Marine Vegetation in Bimini Lagoon, Bahamas. Journal of Sedimentary Petrology 40:249-273

PAGE 154

141 Smith SV (1981) Marine macrophytes as a global carbon sink. Science 211:838-840 Stoner AW (1980) The role of seagrass biomass in the organization of benthic macrofaunal assemblages. Bulletin of Marine Science 30:537-551 Stoner AW (1983) Distribution of Fishes in Seagrass Meadows Role of Macrophyte Biomass and Species Composition. Fish Bull 81:837-846 Strong JA, Dring MJ, Maggs CA (2006) Co lonization and modification of soft substratum habitats by the invasive macroalga Sargassum muticum. Marine Ecology-Progress Series 321:87-97 Terrados J, Duarte CM (2000) Experimental evidence of reduced pa rticle resuspension within a seagrass ( Posidonia oceanica L.) meadow. Journal of Experimental Marine Biology and Ecology 243:45-53 Thayer GW, Adams SM, LaCroix MW (1975) Structural and functional aspects of a recently established Zostera marina community. Estuarine Research 1:518-540 Touchette BW, Burkholder JM (2000) Review of nitrogen and phosphorus metabolism in seagrasses. Journal of Experimental Marine Biology and Ecology 250:133-167 Townsend EC, Fonseca MS (1998) Bioturbation as a potential mechanism influencing spatial heterogeneity of North Carolin a seagrass beds. Marine Ecology-Progress Series 169:123-132 Turner SJ, Hewitt JE, Wilkinson MR, Morrisey DJ, Thrush SF, Cummings VJ, Funnell G (1999) Seagrass patches and landscapes: The influence of wind-wave dynamics and hierarchical arrangements of spa tial structure on macrofaunal seagrass communities. Estuaries 22:1016-1032 van Tussenbroek BI, Santos MGB, van D ijk JK, Alcaraz SNMS, Calderon MLT (2008) Selective elimination of rooted plants fr om a tropical seagrass bed in a back-reef lagoon: a hypothesis tested by hurricane Wilma (2005). Journal of Coastal Research 24:278-281 Verduin JJ, Backhaus JO (2000) Dynamics of plant-flow interactions for the seagrass Amphibolis antarctica : field observations and model simulations. Estuarine Coastal and Shelf Science 50:185-204 Vermaat JE, Santamaria L, Roos PJ (2000) Water flow across and sediment trapping in submerged macrophyte beds of contrasting growth form. Archiv Fur Hydrobiologie 148:549-562

PAGE 155

142 Virnstein RW, Mikkelsen PS, Cairns KD, Capo ne MA (1983) Seagrass beds versus sand bottoms: the trophic importance of their as sociated benthic invertebrates. Florida Scientist 46:363-381 Vizzini S, Mazzola A (2006) Sources and transfer of organic matter in food webs of a Mediterranean coastal environment: evid ence for spatial variability. Estuarine Coastal and Shelf Science 66:459-467 Ward LG, Kemp WM, Boynton WR (1984) The influence of waves and seagrass communities on suspended particulates in an estuarine embayment. Marine Geology 59:85-103 Webster PJ, Rowden AA, Attrill MJ (1998) Effect of shoot density on the infaunal macro-invertebrate community within a Zostera marina seagrass bed. Estuarine Coastal and Shelf Science 47:351-357 Worthington DG, Ferrell DJ, McNeill SE, Bell JD (1992) Effects of the shoot density of seagrass on fish and decapods: are correla tion evident over larger spatial scales. Marine Biology 112:139-146 Wicks EC, Koch EW, O'Neil JM, Elliston K ( 2009) Effects of sediment organic content and hydrodynamic conditions on the growth and distribution of Zostera marina Marine Ecology-Progress Series 378:71-80 Widdows J, Brinsley M (2002) Impact of biotic and abiotic processes on sediment dynamics and the consequences to the structure and functioning of the intertidal zone. Journal of Sea Research 48:143-156 Widdows J, Pope ND, Brinsley MD, Asmus H, Asmus RM (2008) Effects of seagrass beds ( Zostera noltii and Z. marina) on near-bed hydrodynamics and sediment resuspension. Marine Ecol ogy-Progress Series 358:125-136 Wilson DP (1949) The decline of Zostera marina L at Salcombe and its effects on the shore. Journal of the Marine Biologi cal Association of the United Kingdom 28:395-412 Worcester SE (1995) Effects of eelgrass beds on advection and turbulent mixing in low current and low shoot density environm ents. Marine Ecology-Progress Series 126:223-232 Yamamuro M, Kayanne H, Yamano H (2003) Delta N-15 of seagrass leaves for monitoring anthropogenic nutri ent increases in coral reef ecosystems. Marine Pollution Bulletin 46:452-458

PAGE 156

143 Young DK, Young MW (1978) Regulation of Speci es Densities of Seagrass-Associated Macrobenthos Evidence from Field E xperiments in Indian River Estuary, Florida. Journal of Marine Research 36:569-593

PAGE 157

144 Appendix A Additional Tables and Figures

PAGE 158

145 Appendix A: Additional Tables and Figures Table 4 Results of one way (replicate weeks) and two way (flow x density) ANOVAs testing fo r differences in bulk flow speeds (m s1) measured over artificial seagrass units (ASUs) with either high (1500 shoots m-2) or low (300 shoots m-2) seagrass shoot densities at fast (0.078 0.041 m s-1) and slow (0.025 0.01 m s-1) experimental study sites at Emerson Point Park in lower Tampa Bay, FL across the 10 week study period. Signi ficant result(s) are indicated by *. Factor Hypothesis degrees of freedom Error degrees of freedom F statistic p value Replicate weeks 5 15 1.886 0.157 Flow 1 17 19.755 < 0.001* Density 1 17 0.399 0.536 Flow x Density 1 17 1.520 0.234

PAGE 159

146 Appendix A (Continued) Table 5 Results of one way (replicate weeks) MANOVA and subsequent posthoc one way (replicate weeks) ANOVAs testing for differences in dry weight (g L-1), dry weight of organic matter (g L-1) and percent organic matter of total suspended solids (TSS) measured in the water column over artific ial seagrass units (ASUs) at fast (0.078 0.041 m s-1) and slow (0.025 0.01 m s-1) experimental study sites at Emerson Point Park in lower Tampa Bay, FL across the 10 w eek study period. Signi ficant result(s) a re indicated by *. F statistic values were approximated using Pillais trace statistic. Factor Variable Hypothesis degrees of freedom Error degrees of freedom F statistic p value Replicate weeks MANOVA 36 39 4.191 < 0.001* Dry weight (g L-1) 12 13 64.161 < 0.001* Organic matter (g L-1) 12 13 27.446 < 0.001* % organic matter 12 13 6.394 0.001*

PAGE 160

147 Appendix A (Continued) Table 6 Results of a one way (replicate weeks) and a two way (flow x benthic habitat) MANOVA testing for differences in the percent dry weight, organic ma tter, and carbonates of sediments collected from vegetated ( Thalassia testudinum ) and unvegetated (bare sand) benthic ha bitats at fast (0.078 0.041 m s-1) and slow (0.025 0.01 m s-1) experimental study sites at Emerson Point Park in lower Tampa Bay, FL across the 10 week study period. Possible signif icant result(s) ar e indicated by **. F statistic value s were approximated using Pillai s trace statistic. Factor Hypothesis degrees of freedom Error degrees of freedom F statistic p value Replicate weeks 36 9 0.803 0.700 Flow 12 1 134.762 0.067 Benthic habitat 12 1 203.964 0.055** Flow x Benthic habitat 12 1 31.693 0.138

PAGE 161

148 Appendix A (Continued) Table 7 Results of posthoc one way (benthic habitat) ANOVAs testing for differences in th e percent dry weight, organic matter, and carbonates by sediment size fraction ( m) of sediments collected from vegetated ( Thalassia testudinum ) and unvegetated (bare sand) benthic habitats at fast (0.078 0.041 m s-1) and slow (0.025 0.01 m s-1) experimental study sites at Emerson Point Park in lower Tampa Bay, FL. Significant re sult(s) are indicated by *. Factor Variable Hypothesis degrees of freedom Error degrees of freedom F statistic p value Benthic habitat <63 m % dry weight 1 12 54.135 < 0.001* 63 m % dry weight 1 12 0.093 0.766 125 m % dry weight 1 12 3.800 0.075 250 m % dry weight 1 12 0.152 0.704 500 m % dry weight 1 12 7.458 0.018* % organics 1 12 12.417 0.004* <63 m % organics 1 12 35.880 < 0.001* 63 m % organics 1 12 15.318 0.002* 125 m % organics 1 12 1.556 0.236 250 m % organics 1 12 0.103 0.753 500 m % organics 1 12 2.781 0.121 % carbonates 1 12 18.735 0.001* 63 m % carbonates 1 12 34.139 < 0.001* 125 m % carbonates 1 12 6.838 0.023* 250 m % carbonates 1 12 1.627 0.226 500 m % carbonates 1 12 5.659 0.035*

PAGE 162

149 Appendix A (Continued) Table 8 Average ( SD) percent dry weight, organic matter, and carbonate content by se diment size fraction ( m) of sediments collected from vegetated ( Thalassia testudinum ) and unvegetated (bare sand) be nthic habitats at fast (0.078 0.041 m s-1) and slow (0.025 0.01 m s-1) experimental study sites at Emerson Point Park in lower Tampa Bay, FL. Flow regime Sediment size fraction ( m) Fast Slow % dry weight 500 1.946 ( 1.726) 1.807 ( 0.6203) 250 6.455 ( 2.718) 10.26 ( 1.983) 125 83.21 ( 2.849) 73.58 ( 3.847) 63 7.430 ( 1.277) 13.22 ( 2.326) <63 0.9569 ( 0.5469) 1.226 ( 0.5286) % organic matter All 0.6616 ( 0.2768) 1.179 ( 0.2597) 500 0.0836 ( 0.0613) 0.1102 ( 0.0556) 250 0.0517 ( 0.0265) 0.1292 ( 0.0760) 125 0.2065 ( 0.0521) 0.3780 ( 0.0975) 63 0.0861 ( 0.0347) 0.2324 ( 0.0729) <63 0.2337 ( 0.1254) 0.3432 ( 0.1192) % carbonates All 0.5666 ( 0.4395) 0.3153 ( 0.0451) 500 0.4130 ( 0.3993) 0.1173 ( 0.1087) 250 0.0249 ( 0.0161) 0.0377 ( 0.0197) 125 0.0891 ( 0.0213) 0.1056 ( 0.0212) 63 0.0396 ( 0.0217) 0.0760 ( 0.0280)

PAGE 163

150 Appendix A (Continued) Table 9 Results of two way (flow x density) MANOVA testing for differen ces in the characteristics ( 13 variables: dry weight and percentages of particles, organi c matter, and carbonates of particle samples a nd by size class) of the particles trapped by art ificial seagrass units (ASUs) with either high (1500 shoots m-2) or low (300 shoots m-2) seagrass shoot densities at fast (0.078 0.041 m s-1) and slow (0.025 0.01 m s-1) experimental study sites at Emerson Point Park in lo wer Tampa Bay, FL across. Significant result(s) are indicated by *. F statistic values were approximated using Pillais trace statistic. Factor Hypothesis degrees of freedom Error degrees of freedom F statistic p value Flow 10 22 39.992 < 0.001* Density 10 22 0.328 0.964 Flow x Density 10 22 0.284 0.978

PAGE 164

151 Appendix A (Continued) Table 10 Results of post-hoc one way (flow) ANOVAs testing for differen ces in the characteristics (13 variables) of the particles trapped by artificial seag rass units (ASUs) with either high (1500 shoots m-2) or low (300 shoots m-2) seagrass shoot densities at fast (0.078 0.041 m s-1) and slow (0.025 0.01 m s-1) experimental study sites at Emerson Point Park in lower Tampa Bay, FL across. Significant result(s) are indicated by *. Factor Variable Hypothesis degrees of freedom Error degrees of freedom F statistic p value Flow dry weight (g m-2 day-1) 1 31 11.080 0.002* <63 m dry weight (g m-2 day-1) 1 31 0.603 0.443 63 m dry weight (g m-2 day-1) 1 31 11.825 0.002* <63 m % dry weight 1 31 135.553 < 0.001* 63 m % dry weight 1 31 135.553 < 0.001* Organic matter (g m-2 day-1) 1 31 43.676 < 0.001* <63 m organic matter (g m-2 day-1) 1 31 4.086 0.052 63 m organic matter (g m-2 day-1) 1 31 57.060 < 0.001* % organic matter 1 31 13.996 0.001* <63 m % organic matter 1 31 49.045 < 0.001* 63 m % organic matter 1 31 1.720 0.199 63 m carbonates (g m-2 day-1) 1 31 12.073 0.002* 63 m % carbonates 1 31 5.798 0.022*

PAGE 165

152 Appendix A (Continued) Table 11 Regression of turbulent kinetic energy (TKE, m2 s-2) and overlying flow speed (m s-1) in the canopy (5 cm above the bottom), at the top of the canopy (20 cm above the bottom), and in the water column above the can opy (40 cm above the bottom) f or high (1500 shoot m-2) and low (300 shoots m-2) shoot density artificial se agrass units (ASU) at Emerson Point Park in lower Tampa Bay, FL. TKE measurement ASU shoot density Regression equation R2 t test P In the Canopy High y = 132.47x 0.53 0.96 t0.05(2),15 = -0.11 0.91 Low y = 133.86x 0.73 0.98 Top of the Canopy High y = 227.33x 3.04 0.82 t0.05(2),11 = 0.48 0.64 Low y = 193.85x 2.96 0.88 In the Water Column High y = 52.46x + 0.66 0.39 t0.05(2),14 = -1.01 0.33 Low y = 92.22x + 0.06 0.56 Table 12 Regression of Reynolds shear stress (RE, Pa) and overlying flow speed (m s-1) in the canopy (5 cm a bove the bottom), at the top of the canopy (20 cm above the bottom), and in the water co lumn above the canopy (40 cm above the bottom) for high (1500 sh oot m-2) and low (300 shoots m-2) shoot density artificial seagrass units (ASU) at Emerson Point Park in lower Tampa Bay, FL. Significant differences between regre ssion coefficients indicated by *. RE measurement ASU shoot density Regression equation R2 t test P In the Canopy High y = 17294x 86.14 0.72 t0.05(2),15 = 2.25 0.04* Low y = 7736.8x 37.31 0.67 Top of the Canopy High y = 10770x 77.50 0.23 t0.05(2),11 = -0.01 0.99 Low y = 10874x 107.46 0.64 In the Water Column High y = 6250.6x 64.05 0.57 t0.05(2),14 = 0.28 0.78 Low y = 5242.5x + 15.76 0.31

PAGE 166

Appendix A (Continued) 0 10 20 30 40 50 60 70 80 Flow Density Treatment% reduction in flow Figure 32 Average ( SD, n = 6) percent reduc tion in flow speeds from above in comparison to within the canopy of artificial seagrass units (ASUs) with either high (1500 shoots m-2) or low (300 shoots m-2) seagrass shoot densities at fast (0.078 0.041 m s-1) and slow (0.025 0.01 m s-1) experimental study sites at Emerson Point Park in lower Tampa Bay, FL. 153

PAGE 167

Appendix A (Continued) 0 10 20 30 40 50 60 70 80 90 100 Slow HighSlow LowFast HighFast Low Treatment (Flow Density)Dry weight of trapped particles (g m-2 day-1) A B Figure 33 Average ( SEM, n = 10) dry weight (g m-2 day-1) of particles trapped by artificial seagrass units (ASUs) with either high (1500 shoot m-2) or low (300 shoots m-2) seagrass shoot densities at fast (0.078 0.041 m s-1) and slow (0.025 0.01 m s-1) experimental study sites at Emerson Point Park in lower Tampa Bay, FL. Statistically different groupings indicated by post-hoc testing (ANOVA) are indicated by upper case lettering (A and B). 154

PAGE 168

Appendix A (Continued) 0 1 2 3 4 5 6 7 8 Slow HighSlow LowFast HighFast Low Treatment (Flow Density)Dry weight of organic matter (g m-2 day-1) A B Figure 34 Average ( SEM, n = 10) dry weight (g m-2 day-1) of organic matter in particles trapped by artificial seagrass un its (ASUs) with either high (1500 shoot m-2) or low (300 shoots m-2) seagrass shoot densities at fast (0.078 0.041 m s-1) and slow (0.025 0.01 m s-1) experimental study sites at Emers on Point Park in lower Tampa Bay, FL. Statistically different groupings i ndicated by post-hoc testing (ANOVA) are indicated by upper case le ttering (A and B). 155

PAGE 169

Appendix A (Continued) 0% 20% 40% 60% 80% 100% 120% Slow HighSlow LowFast HighFast Low Treatment (Flow Density)% of organic matter trapped 63 m <63 m Figure 35 Average ( SEM, n = 10) percent or ganic matter of particles in 63 m and <63 m size fractions trapped by artificial s eagrass units (ASUs) with either high (1500 shoot m-2) or low (300 shoots m-2) seagrass shoot densities at fast (0.078 0.041 m s-1) and slow (0.025 0.01 m s-1) experimental study sites at Emerson Point Park in lower Tampa Bay, FL. 156

PAGE 170

Appendix A (Continued) 0 5 10 15 20 25 30 Slow HighSlow LowFast HighFast Low Treatment (Flow Density)% organic matter in trapped particles 63 m <63 m Figure 36 Average ( SEM, n = 10) percent or ganic matter of particles in 63 m and <63 m size fractions trapped by artificial s eagrass units (ASUs) with either high (1500 shoot m-2) or low (300 shoots m-2) seagrass shoot densities at fast (0.078 0.041 m s-1) and slow (0.025 0.01 m s-1) experimental flow sites at Emerson Point Park in lower Tampa Bay, FL. 157

PAGE 171

Appendix A (Continued) 0 0.5 1 1.5 2 2.5 3 3.5 Slow HighSlow LowFast HighFast Low Treatment (Flow Density)% 63 m carbonates in trapped particles B A Figure 37 Average ( SEM, n = 10) percen t carbonates of particles in 63 m size fraction trapped by artificial seag rass units (ASUs) with either high (1500 shoot m-2) or low (300 shoots m-2) seagrass shoot dens ities at fast (0.078 0.041 m s-1) and slow (0.025 0.01 m s-1) experimental study sites at Emerson Point Park in lower Tampa Bay, FL. Statistically different groupi ngs indicated by post-hoc tes ting (ANOVA) are indicated by upper case lettering (A and B). 158

PAGE 172

Appendix A (Continued) 0 0.005 0.01 0.015 0.02 0.025 0.03 Slow FastDry Weight of TSS (g L-1) a 0 0.001 0.002 0.003 0.004 0.005 0.006 0.007 0.008 Slow FastOrganic matter in TSS (g L-1) b 0 10 20 30 40 50 60 Slow Fast Flow treatment% organic matter in TSS c Figure 38 Average ( SD, n = 12) a) dry weight (g L-1) of total suspended solids (TSS), b) dry weight of organic matter in the TSS (g L-1), and c) percentage of organic matter in the TSS measured in the water column at (0.078 0.041 m s-1) and slow (0.025 0.01 m s-1) experimental study sites at Emerson Point Park in lower Tampa Bay, FL over the 12 week study period. 159

PAGE 173

Appendix A (Continued) 0% 20% 40% 60% 80% 100% 120% Fast Slow Flow regime% of sedimnetary organic matter 500 m 250 m 125 m 63 m <63 m a 0% 20% 40% 60% 80% 100% 120% Bare Seagrass Benthic structure% of sedimentary organic matter 500 m 250 m 125 m 63 m <63 m b Figure 39 Average ( SD, n = 4) percent sedime ntary organic matter by size fraction a) at fast (0.078 0.041 m s-1) and slow (0.025 0.01 m s-1) experimental study sites and b) in vegetated ( Thalassia testudinum ) and unvegetated (bare sand) benthic habitats within those study sites. Sediment size fractions measured included silt-clays (<63 m), very fine sands (63 m), fine sands (125 m), medium sands (250 m), and very coarse sands (500 m). 160

PAGE 174

Appendix A (Continued) 0 0.2 0.4 0.6 0.8 1 1.2 1.4 Fast Slow Flow regime% organic matter in sediments 500 m 250 m 125 m 63 m <63 m a 0 0.2 0.4 0.6 0.8 1 1.2 Bare Seagrass Benthic structure% organic matter in sediments 500 m 250 m 125 m 63 m <63 m b Figure 40 Average ( SD, n = 4) percent organic matter by sediment size fraction a) at fast (0.078 0.041 m s-1) and slow (0.025 0.01 m s-1) experimental study sites and b) in vegetated ( Thalassia testudinum ) and unvegetated (bare sand) benthic habitats within those study sites. Sediment size fractions measured included silt-clays (<63 m), very fine sands (63 m), fine sands (125 m), medium sands (250 m), and very coarse sands (500 m). 161

PAGE 175

Appendix A (Continued) 0% 20% 40% 60% 80% 100% 120% 140% Fast Slow Flow regime% of sedimentary carbonates 500 m 250 m 125 m 63 m a 0% 20% 40% 60% 80% 100% 120% 140% Bare Seagrass Habitat structure% of sedimentary carbonates 500 m 250 m 125 m 63 m b Figure 41 Average ( SD, n = 4) percent sediment ary carbonates by size fraction a) at fast (0.078 0.041 m s-1) and slow (0.025 0.01 m s-1) experimental study sites and b) in vegetated ( Thalassia testudinum ) and unvegetated (bare sand) benthic habitats within those study sites. Sediment size fractions measured included very fine sands (63 m), fine sands (125 m), medium sands (250 m), and very coarse sands (500 m). 162

PAGE 176

Appendix A (Continued) 0 0.2 0.4 0.6 0.8 1 1.2 Fast Slow Flow regime% carbonates in sediments 500 m 250 m 125 m 63 m a 0 0.2 0.4 0.6 0.8 1 1.2 Bare Seagrass Benthic structure% carbonates in sediments 500 m 250 m 125 m 63 m b Figure 42 Average ( SD, n = 4) percent carbonates by sediment size fr action a) at fast (0.078 0.041 m s-1) and slow (0.025 0.01 m s-1) experimental study sites and b) in vegetated ( Thalassia testudinum ) and unvegetated (bare sand) benthic habitats within those study sites. Sediment size fractions measured included very fine sands (63 m), fine sands (125 m), medium sands (250 m), and very coarse sands (500 m). 163

PAGE 177

164 Appendix A (Continued) Table 13 Results of two way (study site x experimental plot) ANOVA testing for differences in bulk flow speeds (m s-1) measured among experimentally thinned seagrass (Thalassia testudinum ) patches with shoot densities ranging from full density (full), half of the original shoot density (50%), 10% of the or iginal density (10%), and complete shoot re moval (bare) at two study sites (North Sk yway and East Beach) in Tampa Bay, FL. Factor Hypothesis degrees of freedom Error degrees of freedom F statistic p value Study site 1 31 3.264 0.081 Treatment 3 31 0.428 0.735 Study site x Treatment 3 31 0.956 0.426 Table 14 Results of three one way (study site; North Skyway replicat e weeks; East Beach replicat e weeks) MANOVAs testing for differences in the dry weight (g L-1), organic matter dry weight (g L-1) and percent organic matter of total suspended solids (TSS) measured in the water column over experimentally thinned seagrass ( Thalassia testudinum ) patches at two study sites (North Skyway and East Beach) in Tampa Bay, FL across 5 replicate experi mental weeks. Significant result(s) are indicated by *. F statistic values were approximated using Pillais trace statistic. Factor Hypothesis degrees of freedom Error degrees of freedom F statistic p value Study site 3 34 10.868 < 0.001* North Skyway Replicate weeks 15 36 3.101 0.003* East Beach Replicate weeks 18 39 5.268 < 0.001*

PAGE 178

165 Appendix A (Continued) Table 15 Results of post-hoc one way (study site; North Skyway repl icate weeks; East Beach re plicate weeks) ANOVAs testing for differences in the dry weight (g L-1), organic matter dry weight (g L-1) and percent organic matter of total suspended solids (TSS) measured in the water column over experimentally thinned seagrass ( Thalassia testudinum ) patches at two study sites (North Skyway and East Beach) in Tampa Bay, FL across 5 replicate experi mental weeks. Significant result(s) are indicated by *. Treatment Variable Hypothesis degrees of freedom Error degrees of freedom F statistic p value Study site Dry weight (g L-1) 1 36 10.708 0.002* Organic matter dry weight (g L-1) 1 36 20.349 < 0.001* % organic matter 1 36 3.949 0.055 North Skyway Replicate Weeks Dry weight (g L-1) 5 12 23.523 < 0.001* Organic matter dry weight (g L-1) 5 12 22.586 < 0.001* % organic matter 5 12 6.110 0.005* East Beach Replicate Weeks Dry weight (g L-1) 6 13 3.480 0.028* Organic matter dry weight (g L-1) 6 13 4.154 0.015* % organic matter 6 13 12.732 < 0.001*

PAGE 179

Appendix A (Continued) Table 16 Results of correlations between the dry weight (g L-1), dry weight of organic matter (g L-1), and % organic ma tter of the total suspended solids (TSS) measured in the water column and the dry weight (g m-2 day-1), dry weight of organic matter (g m-2 day-1), and % organic matter of the all of the particles or just the <63 m size fraction trapped by experi mentally thinned seagrass (Thalassia testudinum ) patches located at the North Skyway study site in Tampa Bay, FL. Signi ficant result(s) ar e indicated by *. Correlation Experimental treatment Particle characteristic Pearson's correlation coefficient statistic (r) Standard error of r statistic (sr) t statistic ( rs r t ) p value TSS: All trapped particle fractions All Dry weight -0.1979 0. 5659 0.3496 0.7497 Organic matter 0.4549 0. 5141 0.8848 0.4414 % organic matter -0.5726 0.4733 1.2096 0.3131 Bare Dry weight -0.0571 0.57 64 0.0991 0.9273 Organic matter 0.6231 0.45 16 1.3800 0.2615 % organic matter -0.7808 0.3607 2.1648 0.1190 10% Dry weight 0.5479 0.48 30 1.1345 0.3390 Organic matter 0.7125 0.40 51 1.7586 0.1769 % organic matter -0.5032 0.4989 1.0086 0.3874 50% Dry weight -0.5479 0.48 30 1.1345 0.3390 Organic matter -0.3689 0.53 66 0.6874 0.5412 % organic matter 0.0999 0. 5745 0.1738 0.8731 Full Dry weight -0.2686 0.55 61 0.4831 0.6621 Organic matter 0.3028 0.55 02 0.5503 0.6204 % organic matter -0.7920 0.3525 2.2467 0.1103 166

PAGE 180

Appendix A (Continued) Table 16 (Continued) Correlation Experimental treatment Particle characteristic Pearson's correlation coefficient statistic (r) Standard error of r statistic (sr) t statistic ( rs r t ) p value TSS: <63m trapped particle fraction All Dry weight 0.6266 0. 4500 1.3925 0.2580 Organic matter 0.6050 0. 4597 1.3161 0.2797 % organic matter -0.8917 0.2613 3.4124 0.0421* Bare Dry weight 0.6884 0. 4187 1.6441 0.1987 Organic matter 0.5875 0. 4672 1.2576 0.2975 % organic matter -0.9181 0.2288 4.0118 0.0278* 10% Dry weight 0.8607 0. 2940 2.9276 0.0611 Organic matter -0.2746 0. 5552 0.4946 0.6548 % organic matter -0.8620 0.2926 2.9457 0.0602 50% Dry weight -0.0888 0. 5751 0.1544 0.8871 Organic matter 0.8740 0. 2806 3.1150 0.0527 % organic matter -0.7609 0.3746 2.0310 0.1352 Full Dry weight 0.7033 0. 4104 1.7138 0.1851 Organic matter 0.6885 0. 4187 1.6442 0.1987 % organic matter -0.9358 0.2035 4.5995 0.0193* 167

PAGE 181

Appendix A (Continued) Table 17 Results of correlations between the dry weight (g L-1), dry weight of organic matter (g L-1), and % organic ma tter of the total suspended solids (TSS) measured in the water column and the dry weight (g m-2 day-1), dry weight of organic matter (g m-2 day-1), and % organic matter of the all of the particles or just the <63 m size fraction trapped by experi mentally thinned seagrass (Thalassia testudinum) patches located at the East Beach study site in Tampa Bay, FL. Significant result(s) are indicated by *. Correlation Experimental treatment Particle characteristic Pearson's correlation coefficient statistic (r) Standard error of r statistic (sr) t statistic ( rs r t ) p value TSS: All trapped particle fractions All Dry weight 0.0431 0.57 68 0.0747 0.9452 Organic matter 0.1115 0.57 38 0.1943 0.8584 % organic matter -0.0347 0.5770 0.0601 0.9559 Bare Dry weight -0.7299 0.39 46 1.8497 0.1615 Organic matter -0.8596 0.29 51 2.9132 0.0618 % organic matter 0.2345 0. 5613 0.4177 0.7042 10% Dry weight -0.0559 0.57 64 0.0970 0.9288 Organic matter 0.3293 0.54 51 0.6042 0.5884 % organic matter -0.0288 0.5771 0.0499 0.9633 50% Dry weight 0.9064 0. 2439 3.7169 0.0339* Organic matter 0.3420 0.54 25 0.6303 0.5732 % organic matter 0.4840 0. 5052 0.9579 0.4088 Full Dry weight 0.2573 0.55 79 0.4611 0.6761 Organic matter 0.7411 0.38 76 1.9119 0.1518 % organic matter -0.4377 0.5191 0.8432 0.4611 168

PAGE 182

Appendix A (Continued) Table 17 (Continued) Correlation Experimental treatment Particle characteristic Pearson's correlation coefficient statistic (r) Standard error of r statistic (sr) t statistic ( rs r t ) p value TSS: <63m trapped particle fraction All Dry weight -0.0044 0. 5773 0.0076 0.9944 Organic matter 0.0252 0. 5772 0.0436 0.9680 % organic matter -0.1177 0.5733 0.2053 0.8505 Bare Dry weight -0.6125 0. 4564 1.3422 0.2721 Organic matter -0.7288 0. 3953 1.8434 0.1625 % organic matter 0.5690 0. 4748 1.1983 0.3168 10% Dry weight -0.0498 0. 5766 0.0863 0.9367 Organic matter 0.0132 0. 5773 0.0229 0.9831 % organic matter 0.6437 0. 4418 1.4570 0.2412 50% Dry weight 0.0854 0. 5752 0.1484 0.8915 Organic matter 0.0624 0. 5762 0.1083 0.9206 % organic matter -0.1754 0.5684 0.3085 0.7779 Full Dry weight 0.5765 0. 4717 1.2221 0.3089 Organic matter 0.7797 0. 3615 2.1568 0.1199 % organic matter -0.5351 0.4877 1.0970 0.3528 169

PAGE 183

170 Appendix A (Continued) Table 18 Results of two way (study site x treatment) MANOVA testing for differences in th e characteristics (21 variables: dry weight and percentages of particles, or ganic matter, and carbonates of pa rticle samples and by size clas s) of the particles trapped by experimentally thinned seagrass (Thalassia testudinum) patches with shoot densities ranging fr om full density (full), half of the original shoot density (50%), 10% of the or iginal density (10%), and complete shoot re moval (bare) at two study sites (North Sk yway and East Beach) in Tampa Bay, FL. Signifi cant result(s) are indicated by *. F stat istic values were approximated using Pillai s trace statistic. Factor Hypothesis degrees of freedom Error degrees of freedom F statistic p value Study site 21 47 33.856 < 0.001* Treatment 63 147 1.492 0.026* Study site x Treatment 63 147 1.272 0.121

PAGE 184

171 Appendix A (Continued) Table 19 Results of post-hoc one way (study site & treatment) ANOVAs testi ng for differences in the charact eristics (21 va riables) of the particles trapped by experi mentally thinned seagrass (Thalassia testudinum) patches with shoot densities ranging from full density (full), half of the original shoot density (50%), 10% of the original density (10% ), and complete shoot removal (bare) at two s tudy sites (North Skyway and East Beach) in Tampa Bay, FL. Significant result(s) are indicated by *. Factor Variable Hypothesis degrees of freedom Error degrees of freedom F statistic p value Study site DW (g) 1 67 5.318 0.024* DW of <63 m (g) 1 67 4.862 0.031* DW of 63 m (g) 1 67 12.951 0.001* DW of 125 m (g) 1 67 0.113 0.738 % DW of <63 m 1 67 38.041 <0.001* % DW of 63 m 1 67 67.827 < 0.001* % DW of 125 m 1 67 13.940 < 0.001* Organic matter (g) 1 67 0.000 0.988 Organics in <63 m (g) 1 67 2.485 0.120 Organic in 63 m (g) 1 67 5.914 0.018* Organics in 125 m (g) 1 67 6.697 0.012* % organic matter 1 67 17.500 < 0.001* % organics in <63 m 1 67 27.394 < 0.001* % organics in 63 m 1 67 3.559 0.064* % organics in 125 m 1 67 43.932 < 0.001*

PAGE 185

172 Appendix A (Continued) Table 19 (Continued) Factor Variable Hypothesis degrees of freedom Error degrees of freedom F statistic p value Study site Carbonates (g) 1 67 1.424 0.237 Carbonates in 63 m (g) 1 67 6.983 0.010* Carbonates in 125 m (g) 1 67 2.038 0.158 % carbonates 1 67 18.230 < 0.001* % Carbonates in 63 m 1 67 0.468 0.496 % Carbonates in 125 m 1 67 19.721 < 0.001* Treatment DW (g) 3 67 6.965 < 0.001* DW of <63 m (g) 3 67 0.868 0.462 DW of 63 m (g) 3 67 1.197 0.318 DW of 125 m (g) 3 67 4.813 0.004* % DW of <63 m 3 67 6.966 < 0.001* % DW of 63 m 3 67 1.498 0.223 % DW of 125 m 3 67 0.776 0.512 Organic matter (g) 3 67 0.843 0.475 Organics in <63 m (g) 3 67 1.021 0.389 Organic in 63 m (g) 3 67 0.237 0.871 Organics in 125 m (g) 3 67 0.618 0.606

PAGE 186

173 Appendix A (Continued) Table 19 (Continued) Factor Variable Hypothesis degrees of freedom Error degrees of freedom F statistic p value Treatment % organics 3 67 15.287 < 0.001* % organics in <63 m 3 67 8.973 < 0.001* % organics in 63 m 3 67 12.211 < 0.001* % organics in 125 m 3 67 13.657 < 0.001* Carbonates (g) 3 67 0.703 0.554 Carbonates in 63 m (g) 3 67 0.557 0.645 Carbonates in 125 m (g) 3 67 0.693 0.559 % carbonates 3 67 3.697 0.016* % Carbonates in 63 m36 7 8.576 < 0.001* % Carbonates in 125 m 3 67 1.499 0.223

PAGE 187

Appendix A (Continued) 0 1 2 3 4 5 6 7 8 9 10 100 90 50 Control Experimental treatmentDry weight of trapped carbonates (g m-2 day-1) Bare 10% 50% Full Figure 43 Average (n = 5) dry weight (g m-2 day-1) of carbonates by si ze fraction trapped by experimentally thinned seagrass (Thalassia testudinum) patches with shoot densities ranging from full density (full), half of the original shoot density (50%), 10% of the original density (10%), and complete shoot removal (bare) at two study sites in Tampa Bay, FL. Comparisons of the average per centage trapped carbonates were made both among experimental plots and between the North Skyway ( 63 and 125 m) and East Beach ( 63 and 125 m) study sites. 174

PAGE 188

Appendix A (Continued) 0% 20% 40% 60% 80% 100% 100 90 50 Control Experimental treatment% of trapped carbonates Bare 10% 50% Full Figure 44 Average (n = 5) percent carbonates by size fraction trapped by experimentally thinned seagrass (Thalassia testudinum) patches with shoot dens ities ranging from full density (full), half of the original shoot dens ity (50%), 10% of the original density (10%), and complete shoot removal (bare) at two st udy sites in Tampa Bay, FL. Comparisons of the average percentage trapped carbonates we re made both among expe rimental plots and between the North Skyway ( 63 and 125 m) and East Beach ( 63 and 125 m) study sites. 175

PAGE 189

Appendix A (Continued) 0 2 4 6 8 10 12 100 90 50 Control Experimental treatmentDry weight of trapped organic matter (g m-2 day-1) Bare 10% 50% Full Figure 45 Average ( SD, n = 5) dry weight (g m-2 day-1) of organic matter by size fraction trapped by experime ntally thinned seagrass (Thalassia testudinum) patches with shoot densities ranging from full density (full), half of the original shoot density (50%), 10% of the original density (10%), and comp lete shoot removal (bare) at two study sites in Tampa Bay, FL. Comparisons of the average percentage trapped organic matter were made both among experimental plots and between the North Skyway ( 63, 63, and 125 m) and East Beach ( 63, 63, and 125 m) study sites. 176

PAGE 190

Appendix A (Continued) 0% 20% 40% 60% 80% 100% 100 90 50 Control Experimental treatment% of trapped organic matter 177 Figure 46 Average (n = 5) percent organi c matter by size fraction trapped by experimentally thinned seagrass (Thalassia testudinum) patches with shoot densities ranging from full density (full), half of the original shoot density (50%), 10% of the original density (10%), and complete shoot removal (bare) at two study sites in Tampa Bay, FL. Comparisons of the average percen tage trapped organic matter were made both among experimental plots and between the North Skyway ( 63, 63, and 125 m) and East Beach ( 63, 63, and 125 m) study sites. Bare 10% 50% Full

PAGE 191

Appendix A (Continued) 0 5 10 15 20 25 Bare 10% 50% Full Experimental treatment% dry weight of <63 m trapped particles C BC AB A Figure 47 Average ( SEM, n = 10) % dry weight of <63 m sized particles trapped by experimentally thinned seagrass (Thalassia testudinum) patches with shoot densities ranging from full density (full), half of the original shoot density (50%), 10% of the original density (10%), and complete shoot removal (bare) at two study sites (North Skyway or East Beach) in Tampa Bay, FL. Values for each experimental treatment were averaged across the study sites. Statistica lly different groupings indicated by post-hoc analysis (Tukey B) are represented by upper case lettering (A, B, or C). 178

PAGE 192

179 Appendix A (Continued) Table 20 Summary of studies investigating influence of submerged a quatic vegetation (SAV) on sedime ntation. Categorized by study location (southern hemisphere, temperate, and/or tropical), study t ype (field and/or flume), vegetation type (seagrass, algae, and/or saltmarsh grass), SAV-sedimentation relationship (presences, pos itive, negative, or variable), study result(s), and hydrodynami c modification of result(s). Location Type of study Vegetation Relati onship Result(s) Hydrodynamics Reference southern hemisphere field Amphibolis griffithii & Posidonia coriacea no sediment movement = inside & outside seagrass beds under high wave energy sediment movement due to winter storms Paling et al. 2003 tropical field flume Halophila decipiens + threshold velocity for sediment motion in seagrass leaf biomass at sediment surface near sediment flow & sediment erosion Fonseca 1989 southern hemisphere field Halophila ovalis, Halodule uninervis, & Zostera capricorni no sedimentary structure & [nutrient] = in seagrass & bare sediment not investigated Mellors et al. 2002 tropical field Halodule wrightii & Thalassia testudinum + sedimentary particle size & organic matter in seagrass not investigated Grady 1981

PAGE 193

180 Appendix A (Continued) Table 20 (Continued) Location Type of study Vegetation Relati onship Result(s) Hydrodynamics Reference temperate field H. wrightii & Zostera marina + % silt-clay & organic matter in seagrass no addressed in results Kenworthy et al. 1982 southern hemisphere field Heterozostera tasmanica & Zostera muelleri + [suspended solids] over seagrass during ebb tide [suspended solids] only over mud flat during ebb tide Bulthuis et al. 1984 temperate field Posidonia oceanica particle deposition < particle resuspension particle resuspension peaks correspond to increased bottom water currents Dauby et al. 1995 temperate field P. oceanica + erosion in seagrass flow = in seagrass and bare sediments Terrados & Duarte 2000 temperate field P. oceanica + deposition in seagrass resuspension by >3 fold in seagrass Gacia & Duarte 2001 temperate flume P. oceanica + water column particle loss rate order of magnitude in seagrass turbulence & shear stress in seagrass canopy Hendriks et al. 2008

PAGE 194

181 Appendix A (Continued) Table 20 (Continued) Location Type of study Vegetation Relati onship Result(s) Hydrodynamics Reference temperate field Ruppia maritima + [suspended solids] in seagrass resuspension suppressed & deposition enhanced as wave energy attenuated by seagrass Ward et al. 1984 tropical field chambers Thalassia hemprichii + [suspended solids] up to 4 fold in chambers with seagrass not investigated Agawin & Duarte 2002 temperate & tropical field T. testudinum & Z. marina + sedimentary particle size & organic matter in seagrass not investigated Orth 1977 tropical field T. testudinum & mimics + for sedimentary % silt-clay, mimics > seagrass > sand 18.5% flow from sand to seagrass Almasi et al. 1987 tropical field T. testudinum variable [suspended solids] over seagrass [suspended solids] over seagrass during flood tide Koch 1999

PAGE 195

182 Appendix A (Continued) Table 20 (Continued) Location Type of study Vegetation Relati onship Result(s) Hydrodynamics Reference temperate field Z. marina variable high sedimentary % silt-clay & organic matter = Z. marina presence; sediments same 1 month after Z. marina removal not investigated Marshall & Lukas 1970 temperate field Z. marina + % silt-clay & organic matter in seagrass flow reduced 3 to 5 fold in seagrass Peterson et al. 1984 temperate field Z. marina + sedimentary % silt-clay in seagrass not investigated Bos et al. 2007 southern hemisphere field Zostera novazelandica + sedimentary % silt-clay in seagrass flow 3.7 fold from above & 2.5 fold from outside seagrass to inside bed Heiss et al. 2000 temperate field freshwater macrophytes (Alisma gramineum & Chara aspera) sedimentation less in macrophytes than bare sediment near sediment flow in macrophytes Vermaat et al. 2000

PAGE 196

183 Appendix A (Continued) Table 20 (Continued) Location Type of study Vegetation Relati onship Result(s) Hydrodynamics Reference temperate field Spartina maritima sedimentation rates higher in bare sediments than S. maritima in fair weather flow in S. maritima canopy, so possible erosion protection Neumeier & Ciavola 2004

PAGE 197

184 Appendix A (Continued) Table 21 Summary of studies investigating influence of submerged a quatic vegetation (SAV) on fauna. Categorized by study type (field and/or flume), vegetation type (seag rass, algae, and/or saltmarsh grass), fauna type, SAV-fauna relationship (presence, positive, negative, or variable), study result(s), and hydrodynamic modi fication of result(s). Type of study Vegetation Fauna Relati onship Result(s) Hydrodynamics Reference field Amphibolis antarctica, Halophila ovalis, Heterozostera tasmanica, Posidonia australis, Posidonia sinuosa, & mimics epi& infauna + seagrass = macrofaunal abundance, biomass, & production not directly measured Edgar 1990 field Cymodocea nodosa & Posidonia oceanica macrofauna + P. oceanica = faunal richness & diverse not investigated Como et al. 2008 field Diplanthera wrightii & Thalassia testudinum macrofauna + seagrass = faunal richness, abundance, diversity, & evenness no consistent effects O'Gower & Wacasey 1967

PAGE 198

185 Appendix A (Continued) Table 21 (Continued) Type of study Vegetation Fauna Relati onship Result(s) Hydrodynamics Reference field D. wrightii & T. testudinum polychaetous annelids + seagrass = annelid abundance not investigated Santos & Simon 1974 field Enhalus acoroides epi& infauna + seagrass = faunal density & biomass not investigated Nakamua & Sano 2005 field Halodule wrightii macrofaunal bivalve (Mercenaria mercenaria & Chione cancellata) + seagrass = bivalve mortality rate not investigated Peterson 1982 field H. wrightii, Syringodium filiforme, & T. testudinum macrobenthic invertebrates + seagrass = faunal density, abundance, & richness not investigated Virnstein et al. 1983 field H. wrightii & T. testudinum benthic crustaceans + seagrass = faunal abundance & richness not investigated Lewis 1984 field Halophila australis, H. tasmanica, & Zostera muelleri meio& macrofauna + seagrass = faunal abundance, richness, & annual production not investigated Edgar et al. 1994

PAGE 199

186 Appendix A (Continued) Table 21 (Continued) Type of study Vegetation Fauna Relati onship Result(s) Hydrodynamics Reference field Halophila engelmanni, S. filiforme, & T. testudinum macrofauna + seagrass = faunal density, biomass, & abundance not directly measured Stoner 1980 field P. oceanica motile macroinvertebrates no seagrass faunal abundance or richness not investigated Borg et al. 2006 field T. testudinum & Zostera marina infauna + seagrass = faunal abundance & richness not investigated Orth 1977 field T. testudinum epi& infauna + seagrass = faunal abundance & richness not investigated Lewis & Stoner 1983 field Zostera capricorni mimics epibenthic harpacticoid copepods mimics = copepod abundance not investigated Hicks 1989 field Zostera japonica & mimics epi& infauna + seagrass = faunal richness & abundance not investigated Lee et al. 2001

PAGE 200

187 Appendix A (Continued) Table 21 (Continued) Type of study Vegetation Fauna Relati onship Result(s) Hydrodynamics Reference field Z. marina infauna + seagrass loss = faunal richness, common species & disappearance of rare species not directly measured Wilson 1949 field Z. marina epi& infaunal invertebrates & fish + seagrass = faunal density & biomass not investigated Thayer et al. 1975 field Z. marina macrofaunal bivalve (Mercenaria mercenaria) + seagrass = bivalves density, size, & growth rate flow in seagrass = bivalve growth rate Peterson et al. 1984 field Z. marina epibenthic fauna + seagrass = faunal richness, density, biomass, & annual production unvegetated, semi-exposed habitat = faunal biomass & production Pihl 1986

PAGE 201

188 Appendix A (Continued) Table 21 (Continued) Type of study Vegetation Fauna Relationship Result(s) Hydrodynamics Reference field Z. marina epi& endomacrofauna + seagrass = faunal abundance, richness, biomass, evenness, & Shannon's Diversity Index not investigated Hily & Bouteille 1999 laboratory & field Z. marina macrofaunal bivalve (Musculista senhousia) seagrass = bivalve survival & growth not investigated Allen & Williams 2003 field Z. marina infauna + seagrass = faunal richness & abundance not directly measured Hirst & Attrill 2008 field Z. muelleri small, motile invertebrates variable seagrass = faunal abundance & biomass for some, but not all fauna not investigated Connolly 1995

PAGE 202

189 Appendix A (Continued) Table 21 (Continued) Type of study Vegetation Fauna Relationship Result(s) Hydrodynamics Reference field Z. muelleri small, motile epifaunal invertebrates variable seagrass = faunal abundance & biomass, but total production not investigated Connolly 1997 field Zostera noltii meio& macroinfauna + seagrass = faunal abundance not directly measured Castel et al. 1989 field Z. noltii mobile epibenthic macrofauna + seagrass = faunal abundance of common species sheltered habitat = faunal abundance Polte et al. 2005 field Zostera novazelandica infauna no seagrass faunal abundance, richness, diversity, evenness, or community composition wave exposure explained differences in faunal community composition Turner et al. 1999 field seagrass mimics invertebrates no seagrass faunal abundance, biomass, or productivity not investigated Edgar 1999

PAGE 203

190 Appendix A (Continued) Table 21 (Continued) Type of study Vegetation Fauna Relationship Result(s) Hydrodynamics Reference field Sargassum muticum macroinfauna variable seagrass = juvenile bivalves, but polychaetes no consistent effects Strong et al. 2006 flume Spartina alterniflora meiofauna (copepods & nematodes) + S. alterniflora = faunal diversity & abundance flow = faunal dispersal Palmer 1986 field Spartina alterniflora x foliosa hybrid benthic macrofaunal invertebrates hybrid = faunal diversity, density, & recruitment low flow in hybrid = sedimentary % silt-clay & organic matter = [sulfide] & anoxia = faunal survivorship Neira et al. 2006

PAGE 204

About The Author Alison Cheryl Meyers was born in Thousand Oa ks, CA and was raised in the neighboring town of Ventura, CA. In 2001 she completed a bachelors degree in marine science with an emphasis in biology and minors in chemistr y and anthropology from the University of San Diego in San Diego, CA. In 2002 Alison began her Ph.D. program in Biology under the mentorship of Dr. Florence I.M. Thomas at the University of South Florida in Tampa, FL. When Dr. Thomas transferred to the Ha waii Institute of Marine Biology at Coconut Island, Alison became co-mentored by Dr. Susan S. Bell, and under this co-mentorship has completed her doctoral degree.


xml version 1.0 encoding UTF-8 standalone no
record xmlns http:www.loc.govMARC21slim xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.loc.govstandardsmarcxmlschemaMARC21slim.xsd
leader nam 22 Ka 4500
controlfield tag 007 cr-bnu---uuuuu
008 s2010 flu s 000 0 eng d
datafield ind1 8 ind2 024
subfield code a E14-SFE0003343
035
(OCoLC)
040
FHM
c FHM
049
FHMM
090
XX9999 (Online)
1 100
Meyers, Alison.
0 245
Depositional dynamics in seagrass systems of tampa bay, fl :
b influence of hydrodynamic regime and vegetation density on ecosystem function
h [electronic resource] /
by Alison Meyers.
260
[Tampa, Fla] :
University of South Florida,
2010.
500
Title from PDF of title page.
Document formatted into pages; contains X pages.
Includes vita.
502
Dissertation (Ph.D.)--University of South Florida, 2010.
504
Includes bibliographical references.
516
Text (Electronic dissertation) in PDF format.
538
Mode of access: World Wide Web.
System requirements: World Wide Web browser and PDF reader.
3 520
ABSTRACT: Many coastal ecosystems around the world are dominated by submerged aquatic vegetation (SAV) habitats. These SAV habitats are known to provide many highly valuable ecosystem services such as habitat for commercial important species and increased water clarity. Water flow is an environmental variable which can have measurable effects on the ecosystem services provided by SAV, but is often not considered in studies assessing these services. This dissertation sought to investigate the links between SAV, primarily seagrasses, and hydrodynamics, paying special attention to the effects on sediments and fauna. Three main areas are discussed: (1) the effects of SAV on flow, (2) the effects of SAV and flow on deposition in SAV beds, and (3) the effects of SAV and flow on faunal communities in SAV beds. Seagrasses and other SAV reduce currents, attenuate waves, and dampen turbulence within their vegetative canopies, which in turn can enhance deposition and reduce the resuspension of sediment, organic matter, and passively settling larvae. The ability of SAV to retard flow may be further enhanced by increases in vegetated structure, such as shoot density, biomass, or canopy height, which can promote increased abundance and diversity of in- and epifauna within SAV beds. Ultimately, it is clear that hydrodynamics is an important factor that shapes SAV communities both physically (e.g. deposition, sediment structure, etc.) and biologically (e.g. faunal community composition, predation pressure, food availability, etc.).
590
Advisor: Susan S. Bell, Ph.D.
653
Thalassia testudinum
Water flow
Fauna
Particle accumulation
Artificial seagrass units
690
Dissertations, Academic
z USF
x Biology Integrative
Doctoral.
773
t USF Electronic Theses and Dissertations.
4 856
u http://digital.lib.usf.edu/?e14.3343