University of New Hampshire University of New Hampshire Scholars' Repository
Doctoral Dissertations Student Scholarship
Spring 2012
Ecological effects of ditching and ditch-plugging in New England salt marshes
Robert E. Vincent University of New Hampshire, Durham
Follow this and additional works at: https://scholars.unh.edu/dissertation
Recommended Citation Vincent, Robert E., "Ecological effects of ditching and ditch-plugging in New England salt marshes" (2012). Doctoral Dissertations. 661. https://scholars.unh.edu/dissertation/661
This Dissertation is brought to you for free and open access by the Student Scholarship at University of New Hampshire Scholars' Repository. It has been accepted for inclusion in Doctoral Dissertations by an authorized administrator of University of New Hampshire Scholars' Repository. For more information, please contact [email protected]. ECOLOGICAL EFFECTS OF DITCHING AND DITCH-PLUGGING
IN NEW ENGLAND SALT MARSHES
BY
ROBERT E. VINCENT
B.A. University of Rhode Island, 1987
M.S. Antioch University, 1996
DISSERTATIOiN
Submitted to the University of New Hampshire
In Partial Fulfillment of
The Requirements for the Degree of
Doctor of Philosophy
In
Earth and Environmental Science
May, 2012 UMI Number: 3525069
All rights reserved
INFORMATION TO ALL USERS The quality of this reproduction is dependent upon the quality of the copy submitted.
In the unlikely event that the author did not send a complete manuscript and there are missing pages, these will be noted. Also, if material had to be removed, a note will indicate the deletion.
ttswWioft FtoMsh«i
UMI 3525069 Published by ProQuest LLC 2012. Copyright in the Dissertation held by the Author. Microform Edition © ProQuest LLC. All rights reserved. This work is protected against unauthorized copying under Title 17, United States Code.
ProQuest LLC 789 East Eisenhower Parkway P.O. Box 1346 Ann Arbor, Ml 48106-1346 ALL RIGHTS RESERVED
©2012
Robert E. Vincent This dissertation has been examined and approved.
Dissertation Director. Dr. David M Burdick Research Associate Professor. Department of Natural Resources and the Environment
Dr. Michele Dionne. Research Director. We
Dr. Erik A. Hobbie. Research Associate Professor. Institute for the Study of Earth. Oceans, and Space
/
Dr. Tbomas D. Lee. Associate Professor. Department of Natural Resources and the Environment
Department of Natural Resources and the Environment
Date DEDICATION
This dissertation is dedicated to my family; and especially to Beth for her patience and support during this process.
iv ACKNOWLEDGEMENTS
I'd like to thank my dissertation advisor, Dr. David Burdick, for taking me on as a student and for his guidance throughout my Ph.D. work; my committee members, Dr.
Michele Dionne, Dr. Erik Hobbie, Dr. Tom Lee, and Dr. Fred Short for their advice during my time at the University of New Hampshire (UNH); and Dr. Michele Dionne for her guidance during my fellowship at the Wells National Estuarine Research Reserve
(The Wells Reserve). I'd also like to thank students and faculty at the UNH Jackson
Estuarine Laboratory for their help with field work, especially Alyson Eberhardt, Chris
Peter, and Dr. Greg Moore, as well as staff at The Wells Reserve. Additional thanks go to Dr. Erik Hobbie and Andy Ouimette at the UNH Stable Isotope Laboratory for their assistance with stable isotope processing; and Dr. Ray Konisky at The Nature
Conservancy, Matt Collins at the NOAA Restoration Center, and Beth Lambert and staff at the New Hampshire Coastal Program for help with data collection and use of equipment.
Funding for my research at Moody Marsh was provided by the NOAA National
Estuarine Research Reserve Graduate Research Fellowship Program. Equipment used in my research was funded in part by the UNH Marine Program. The U.S. Fish and
Wildlife Service provided special use permits for work at Chauncey Creek, Moody, and
Parker River marshes. Permission to conduct work at Sprague River Marsh was provided by Bates College and The Nature Conservancy.
V TABLE OF CONTENTS
DEDICATION iv
ACKNOWLEDGEMENTS v
LIST OF TABLES ix
LIST OF FIGURES x
ABSTRACT xiv
CHAPTER PAGE
I. INTRODUCTION 1
Background 1
Objectives 15
Organization of the Dissertation 24
II. DITCHING AND DITCH-PLUGGING IN NEW ENGLAND
SALT MARSHES: EFFECTS ON HYDROLOGY, ELEVATION,
AND SOIL CHARACTERISTICS 25
Abstract 25
Introduction 25
Materials and Methods 26
Results 37
Discussion 53
Conclusion 66
vi CHAPTER PAGE
III. DITCHING AND DITCH-PLUGGING IN NEW ENGLAND
SALT MARSHES: EFFECTS ON PLANT COMMUNITIES 68
Abstract 68
Introduction 69
Materials and Methods 70
Results 74
Discussion 86
Conclusion 97
IV. FISH PRODUCTIVITY AND TROPHIC TRANSFER IN CREATED
AND NATURALLY-OCCURRING SALT MARSH HABITAT 100
Abstract 100
Introduction 101
Materials and Methods 105
Results 117
Discussion 136
Conclusion 153
V. CONCLUSION 156
Historic and Current Hydrologic Alterations 156
Ecosystem Response to Ditching 157
Ecosystem Response to Natural Creek Hydrology 163
Ecosystem Response to Ditch-Plugging 164
vii Ecosystem Response to Natural Pool Hydrology 165
LITERATURE CITED 166
APPENDICES PAGE
APPENDIX A MEAN VALUES (± SE) BY MARSH AND HABITAT FOR
HYDROLOGY, MARSH SURFACE ELEVATION, AND SOIL
CHARACTERISTICS 187
APPENDIX B PLANT CHARACTERISTICS BY MARSH 195
APPENDIX C INVERTEBRATE RICHNESS AND ABUNDANCE BY
HABITAT, LOCATION, AND TREATMENT: MOODY MARSH,
WELLS, MAINE 205
APPENDIX D INSTITUTIONAL ANIMAL CARE AND USE COMMITTEE
APPROVAL LETTERS 214
viii LIST OF TABLES
TABLE PAGE
1. ANOVA analysis for hydrology, soils, and marsh surface elevations 37
2. Tidal ranges by habitat 39
3. Correlation matrix for hydrology, soils, and marsh surface elevations 40
4. Marsh surface elevations by habitat 41
5. Hydrology, soils, and elevation discriminant function actual by
predicted habitat classification matrix 52
6. Discriminant function values for hydrology, soils, and marsh surface
elevations 52
7. Discriminant function scoring coefficients for hydrology, soils, and
marsh surface elevations 52
8. Kruskal-Wallis test for plant biomass 76
9. ANOVA results for plant characteristics 77
10. Plant discriminant function actual by predicted habitat classification matrix... 85
11. Discriminant function values for plant characteristics 86
12 Discriminant function scoring coefficients for plant characteristics 87
13. ANOVA for invertebrate community characteristics 124
14. Invertebrate proportional distribution by major taxa and habitat 126
15. Stable isotope values by species 130
16. Regression results for fish g15N and g13C vs. pool water salinity 131
ix LIST OF FIGURES
FIGURE PAGE
1. Salt marsh self-maintenance process 3
2. Creek habitat 5
3. Wagon wheel ruts in salt marsh habitat 7
4. Ditch habitat 8
5. Salt marsh grid ditching 9
6. Hay staddle erosion pool formation 11
7. Natural pool habitat 11
8. Ditch-plug plywood reinforcement 12
9. Ditch-plug habitat 13
10. Four-marsh study site location map 15
11. Creek habitat conceptual model 18
12. Ditch habitat conceptual model 19
13. Natural pool habitat conceptual model 20
14. Ditch-plug habitat conceptual model 21
15. Natural pool trophic interaction conceptual model 22
16. Ditch-plug trophic interaction conceptual model 23
17. Three-marsh study site location map 27
18. Parker River Marsh study site 28
19. Chauncey Creek Marsh study site 29
20. Sprague River Marsh study site 31
X FIGURE PAGE
21. Water levels by habitat 38
22. Tidal stage duration curves 39
23. Marsh surface elevations by habitat 41
24. Marsh surface elevations by habitat and distance 42
25. Pore-water salinity by habitat 43
26. Soil redox potential by habitat 44
27. Soil strength by habitat 45
28. Soil strength with depth by habitat 46
29. Soil bulk density by habitat 47
30. Soil carbon storage 48
31. Soil percent organic content 49
32. Discriminant function analysis for hydrology and soil
characteristics 51
33. Ditch-plug habitat surface water impoundment and vegetation
dieback 59
34. Areas of excess salt accumulation in ditch-plug habitat 61
35. Ditch-plug habitat plant dieback areas colonized by sulfur
bacteria and algae mat 75
36. Plant percent cover by habitat 76
37. Plant biomass by habitat 77
38. Plant species richness and diversity by habitat 79
xi FIGURE PAGE
39. Plant species composition by habitat 80
40. Plant composition by habitat, distance, and elevation 81
41. Discriminant function analysis for plant characteristics 84
42. Hummock-hollow topography patterns in ditch-plug habitat 95
43. Moody Marsh study site location map and sample stations 107
44. Moody Marsh fish enclosures and exclosures 108
45. Moody Marsh study design with soil core locations 109
46. Moody Marsh pool physical characteristics 119
47. Natural pool and ditch-plug soil cores 120
48. Fish growth by habitat 121
49. Invertebrate community characteristics 123
50. Invertebrate biomass by habitat 124
51. Invertebrate species richness by habitat 125
52. Invertebrate density by habitat 125
53. Invertebrate abundance by major taxa and habitat 127
54. Invertebrate caloric values 128
55. Stable isotope values (513C and 815N) for vegetation,
invertebrates, and fish 131
56. Stable isotope values (813C and 8I5N) for plants by habitat 132
57. Natural pool habitat food web 134
58. Ditch-plug habitat food web 135
xii FIGURE PAGE
59. Ditch habitat observed conditions 158
60. Creek habitat observed conditions 159
61. Ditch-plug habitat observed conditions 160
62. Natural pool habitat observed conditions 161
xiii ABSTRACT
ECOLOGICAL EFFECTS OF DITCHING AND DITCH-PLUGGING
IN NEW ENGLAND SALT MARSHES
by
Robert E. Vincent
University of New Hampshire, May, 2012
Anthropogenic activities in New England salt marshes have altered hydrologic flows in various ways, but unintended consequences from habitat modifications have received little attention. Specifically, ditches have existed on salt marshes for decades, but the effects of these hydrologic alterations are only poorly understood. Ditch-plugging is a more recent methodology used for salt marsh habitat enhancement and mosquito control, but the long-term effects from this management practice are also unclear. I used natural tidal creeks and pools as controls to examine the effects resulting from ditching and plugging, respectively, upon hydrology, soil characteristics, marsh surface elevation, plant characteristics, fish productivity, and trophic webs. Results indicated only slight differences in parameters sampled within habitat adjacent to ditches compared with creeks, and I infer minor ecological impact after 70+ years. Significant differences in hydrology, soil characteristics, marsh surface elevation, plant characteristics, fish productivity and trophic webs were observed in habitat adjacent to natural pools compared with ditch-plugs, and the many structural differences appear to result in ecological dissimilarities in function between the two habitats as well. The results of my study are important for natural resource managers to consider when planning salt marsh
xiv restoration and enhancement projects. The long-term legacy of ditch-plugs, especially as they pertain to changes in climate, may increase vulnerability of salt marshes to sea level rise and conversion of ditches to ditch-plugs should not be undertaken without careful consideration.
XV CHAPTER I
INTRODUCTION
Background
Estuaries are highly productive environments that create and cycle large amounts of organic matter annually, with trophic transfer of organic carbon providing connectivity among estuarine and near-shore environments (Kneib 2000). In New England, estuaries contain a variety of habitats including salt marshes that grade into freshwater wetlands, tidal mud and sand flats, rocky intertidal areas, tidal pools, and seagrass, algal, and shellfish beds. Salt marshes are in essence transition zones in estuaries that connect marine and terrestrial environments. The focus of my study was to analyze the structural and functional responses of salt marsh habitat to artificial hydrologic regimes, that influence trophic webs and connectivity among natural and created salt marsh habitats.
Salt marshes provide a variety of functions and values important for maintaining the integrity of estuarine ecosystems and protection of surrounding areas. Salt marshes are major contributors to the estuarine food web and provide valuable nesting and refuge habitat for various mammals, birds, fish, and aquatic invertebrates. The carbon energy of primary producing plants is passed along to wildlife through trophic transfer in support of offshore consumers (Kneib 2000), and salt marshes provide valuable nursery habitat for fish species in support of commercial and recreational fisheries (Gedan et al. 2009).
Additional salt marsh functions and values include wave energy dampening that reduces erosion and protects against flood damage (Short et al. 2000); soil filtration and trapping
l that counters sea level rise and improves water quality (Morris et al. 2002; Roman et al.
2000); and the carbon storage function of salt marshes has implications for climate- related processes (Chmura et al. 2003). Salt marshes support regional biodiversity through the maintenance of plant and animal communities adapted to cyclical tidally-influenced saline conditions. The communities grade into upland and freshwater habitats that extent its influence beyond the upper boundary of the estuary. Estuarine systems also provide educational, recreational, aesthetic, and scientific research values to local communities (Wilson et al. 2005).
Salt marsh development occurs over thousands of years through a dynamic process involving hydrology, soil accretion, vegetation, and marsh surface elevation
(Redfield 1972). Salt marsh self-maintenance is a process regulated by tidal sediment transport and deposition along with plant growth, soil trapping, and organic matter accumulation, which combined provide a negative feedback mechanism that facilitates marsh accretion and development to high marsh (Figure 1). Hydrologic regimes are integral to the self-maintenance process, and are influenced by geologic, topographic, and other physical factors (i.e., slope, ice scouring, downed trees, and/or deposition of rocks, wrack, soil, and other debris). Hydrologic regimes mediate the development and persistence of biological communities, which in turn influence plant community patterns and salt marsh accretion rates. Goodman et al. (2007) determined the mean rate of accretion in Maine salt marshes was 2.8 mm/yr over the 17-year period from 1986 to
2003, and the mean rate of relative sea level rise along the Maine coast was approximately 2.4 mm/yr for the
2 Inorganic Sediment Trapping
Sea Level
Sediment Transport Organic Input TMalFlow Marsh Building
Low Marsh Mid Marsh High Marsh
Landward migration
Figure 1. Salt marsh self-maintenance feedback process
same time period. The salt marsh self-maintenance process breaks down when marsh accretion rates fail to keep pace with the rate of sea level rise (Morris et al. 2002).
Salt marshes in New England have a long history of human use and anthropogenic impacts. Approximately 37% of salt marsh habitat along the New
England coast has been lost due to human encroachment that has occurred since
European settlement (Bromberg and Bertness 2005). Anthropogenic impacts include dredging, filling, draining, fertilizer runoff, habitat fragmentation from roadways and railways, tidal flow restrictions from undersized culverts, toxic contaminant runoff, and invasive species establishment (Roman et al. 2000; Teal 2009; Gedan et al. 2009). All of these impacts threaten the ecological integrity of salt marsh habitat and estuarine ecosystems, and many have resulted in a mosaic of natural and created water features
3 throughout marshes of the region (Daiber 1986; Silliman et al. 2009). Naturally-occurring water features include a variety of pools, both permanent and ephemeral, along with meandering and often multi-ordered creek channels. Created water features tend to consist of isolated pools or pools connected by linear ditches, often constructed for specific water control, mosquito, and wildlife habitat purposes.
Creek channels in New England range widely in width (<1 to >100 m) and depth
(<1 to >3 m), typically with steeply sloping vegetated banks (Figure 2). Various processes influence creek development in salt marshes. In early stages, drainage area, gradient, tidal prism, and shear stress interact to promote downward cutting channels
(Lawrence et al. 2006; D'Alpaos et al. 2007). The erosional process of tidal forcing encounters resistance from consolidated soils, reducing shear stress and scouring, promoting the headward formation of dendritic channel networks and channel widening
(Fagherazzi and Furbish 2001; Fagherazzi and Sun 2004; D'Alpaos et al. 2007;
Stefanon et al. 2010). As the marsh develops with rising sea level, physical and biological factors influence erosion and deposition processes, modifying channel morphology (Lawrence et al. 2004; Minkoff 2006; D'Alpaos et al. 2006, Kirwan and
Murray 2007). Plants promote equilibrium between erosion and soil deposition, and the loss of plants can lead to rapid loss of marsh habitat (Kirwan and Murray 2007). In more mature marshes, pinnate headward erosion of established channels can occur in response to
4 ' * -Mug*?**-.* . •
HHH| gH •HBHHE ^i HHSmm ml «• fSSM
~% a ,iV '.,; > ^ > •>. •; ",'•. yi-^? A " "' &X- A "A •>,.'-V|. '-/ /? > ."'/•"'' - >-'-V"• -5L-\'.-~ ' • ', > •• <• .,,V » ./~n'_ •'/ t,5\H 'V'-A -"-'X(••>> • • ' •{*>"r -\ Av'.^ -/K~"rt v •/-.---• ' •y/r'/-A.;' /v- • 1K' '-. -A" >V-^— ^/•-HV. rr •/ - -k-vfAl-vfIs'is-:v« \ • ',,' »\\ \\-\-.-\-, . —'/-•'/- ".A A^vig A~Ar>.>ri»* •.. ._• •—: a • • ^ x^r ' A^a^A^A'-* - -- •&%. >*' I , •' mT ViV '•• 'A'^A:^ 7 4'' ' \ V' • A'A^Xf, .'««ta>A ••v;^>. ty'--<"•.- V - ,ATY 'A'^AA'A- A^A' . A:xA-^vA. •• - < • "'••':•'i7-:-.' ' i i./•. MsMsAA //at\ ^ •»x ^v vv/A /A \^^;Vi-^c.- V AA Figure 2. Creek habitat, Kittery, Maine (photo by Rob Vincent). disequilibrium between marsh platform development and the rate of sea level rise (Hughes et al. 2009). Creeks provide connectivity between estuarine and marine environments by conveying hydrologic flows that transport soils, nutrients, and organic matter throughout the marsh and near-shore environments, which contributes to the salt marsh self-maintenance process. Creeks also provide connectivity between estuarine and off-shore areas, acting as travel corridors, refuge, and forage habitat for resident and transient nekton (Kneib 2000), as well as forage and protective travel corridors for a variety of avian species (personal observations). 5 Ditches were constructed in New England salt marshes for a variety of purposes. From colonial times through the 1800's ditches were dug to maintain habitat for salt hay production and cattle grazing (Fogg 1983; Daiber 1986). These ditches were shallow and constructed for the purpose of removing surface water from the marsh (Fogg 1983). Shallow, narrow, linear wheel ruts resulting from salt hay and clam harvesting wagons (Figure 3) also occur throughout the marshes of New England (Fogg 1983; Taylor 1998; personal observations). Some of these agricultural ditches can still be found, but are less common today and, therefore, were not addressed during this study. The more prominent and widespread created channel habitat in New England salt marshes are the ditches constructed by the Civilian Conservation Corps (CCC) in the 1930's as part of the Roosevelt Administration's attempt to create jobs and control salt marsh mosquitoes (Daiber 1986; Wolfe 1996). Ditches created for mosquito control tend to be narrow (£ lm), vary in depth (£ 1 m), with vertical banks (Figure 4). The CCC program involved digging parallel and perpendicular ditches up to 150 feet (46 meters) apart that connected to tidal creeks throughout the marsh (Figure 5), with the expectation that these ditches would efficiently drain the marsh surface, primary breeding habitat for mosquitoes. Studies (predominantly in the US mid-Atlantic region) have since suggested that alterations to natural hydrologic regimes caused by these ditches have drained pools, lowered soil pore- water levels, modified salt marsh accretion processes, and altered plant communities (Daiber 1986; Bourn and Cottam 1950; Taylor 1998; LeMay 2007; Mullan Crain et al. 2009). Other studies suggested ditches have had little or no impact on salt marsh hydrologic regimes (Headlee 1939; Travis et al. 1954; Provost 1974). Although ditches 6 Figure 3. Wagon wheel ruts in high marsh habitat, Newburyport, Massachusetts (photo by Rob Vincent). do not typically display the meandering characteristics of creek systems, they do provide connectivity, convey hydrologic flows, and provide nekton travel corridors similar to creek habitat. In addition, like natural creeks, a variety of avian species, including sharp- tailed sparrows {Ammodramus caudacutus', and Ammodramus nelsoni), use ditches as protected travel corridors when threatened (personal observations). 7 Figure 4. High marsh ditch habitat, Newburyport, Massachusetts (photo by Rob Vincent). Furthermore, some degraded ditches contain overhanging vegetation, slumped banks, and pooled areas during low water periods that provide suitable fish habitat, also similar to creeks (Daiber 1986; personal observations). The important question is: how do ditches compare to creek habitat? 8 Figure 5. Extensive grid and parallel ditches that were dug in the 1930s by the Civilian Conservations Corps in Seabrook, NH; darker and wetter areas are seen surrounding ditches in the center and right portions of the photo (source: Google Earth, photo taken in 2010) Natural pools in northern New England salt marshes typically develop over time as secondary features in response to disturbance from algal mats, wrack deposits, soil deposition, ice scouring (Harshberger 1916; Chapman 1960; Argow and Fitzgerald 2006; Dionne 1989; Wilson et al. 2009), and hay staddle erosion (personal observations; Figure 6). Complex environmental gradients involving elevation, hydrology, soil characteristics, salinity, redox potential, sulfides, and plant community patterns can also influence pool development (Yapp et al. 1917; Chapman 1960; Redfield 1972). Natural pools are often round (Figure 7), and range widely in size (1 to >20m diameter), with characteristic vertical or under-cut banks and uniform bathymetry with depths approximately 25 to 40 cm at the center (Smith and Able 1994; Hunter et al. 2009). Natural pools provide habitat heterogeneity, high marsh intertidal/subtidal refuge, breeding, and forage habitat for a variety of nekton, and forage habitat for avian and mammalian tertiary consumers. Created pools can be either isolated or connected to other pools via linear ditches or shallow channels, often constructed for specific mosquito control and wildlife habitat purposes. Ditch-plugging is part of the Open Marsh Water Management (OMWM) mosquito control methodology established in the 1960s (Ferrino and Jobbins 1968; Daiber 1986); however, ditch-plugging is a more recent addition to the OMWM methodology (Taylor 1998). Ditch-plug creation is a habitat alteration method that attempts to control mosquito populations by increasing surface water habitat for larvivorous fish, with the secondary objective of promoting wading bird and water fowl habitat (Meredith et al. 1985; Taylor 1998; Neidowski 2000). Ditch-plug creation involves excavating marsh soils from an up-stream portion of the mosquito ditch and plugging the seaward end of the ditch with the spoils. Plywood boards are typically 10 Figure 6. Pool development associated with hay staddle remnants in Rowley, Massachusetts (Photo by Rob Vincent) Figure 7. High marsh natural pool habitat, Newburyport, Massachusetts (photo by Rob Vincent). 11 inserted vertically into the plug in an effort to stabilize the soils (Figure 8), and a pool forms behind the channel plug (Figure 9). Ditch-plugs are relatively small in size to start, have gradually tapering and undulating bathymetry that descends to a deep center sump > 1 m deep (Meredith et al. 1985; personal observations). This newly formed pool is flushed only during the higher spring tides. Ditch-plug creation became popular in New England in the 1990s; however, the long-term localized effects on hydrology, marsh processes, surrounding habitat, and food webs are unclear. Figure 8. Plywood inserted into Ditch-plug in an attempt to provide support and stabilize soils, Phippsburg, Maine. Note the disturbed marsh surface, some areas bare, some waterlogged (photo by Rob Vincent). 12 Figure 9. Ditch-plug habitat, Newburyport, Massachusetts, note marsh plant dieback (photo by Rob Vincent) Hydrologic alteration from ditches failed to rid salt marshes of mosquitoes and, therefore, is not considered a long-term solution for mosquito control (Daiber 1986). Although more recent attempts at controlling salt marsh mosquitoes through further hydrologic alterations (i.e., ditch-plugging and pool creation) and biological controls (i.e., larvivorous fish) have a logical basis, they are not likely to result in a long-term solution for controlling mosquito populations either. This is because bare moist soils associated with plant dieback and surface water retention associated with micro-topographic relief in 13 areas surrounding ditch-plugs provide good mosquito breeding habitat, often replacing the original well-drained soils. Further, habitat impacts from ditch-plugging and pool creation can alter marsh structure and function, thereby impacting biological communities in unpredictable and unintended ways. To assess the effects of altered hydrologic regimes on the structure and function of salt marsh habitat, I studied the physical and biological parameters associated with created and natural, pools and channels, in three back barrier salt marshes (Parker River Marsh, Chauncey Creek Marsh, and Sprague River Marsh) located along the Gulf of Maine coastline from Massachusetts to central Maine (Figure 10). I then used a similar back barrier marsh, Moody Marsh, as my study site to investigate the effects of ditch-plugging on salt marsh pool habitat value in terms of fish productivity, food web structure, and habitat connectivity. My study took a new approach to assessing anthropogenic impacts to salt marsh habitat by focusing on localized responses to habitat alterations. Previous studies used sampling designs involving transects that extended great distances across the marsh and combined data from impacted areas with data collected from areas of the marsh not directly associated with the altered habitat (Adamowicz and Roman 2002; James-Pirri et al. 2005, James-Pirri et al. 2011). They found ditch-plug creation increased ground and surface water levels; shifted nekton dominance from fish (Fundulus heteroclitus) to shrimp (Palaemonetes pugio)\ increased water fowl abundance during migration periods (primarily black duck, Anas rubripes); but had no effect on salinity, vegetation, or mosquito production. By focusing on localized habitat adjacent to the water features, my study design characterizes the marsh responses directly associated with hydrologic 14 alterations, which have implications for physical and biological processes that enable self-maintenance of salt marsh habitat. ( •'V-J •* Sprague River Marsh Moody Marsh Chauneey ("reck Marsh Parker River Marsh Mas.sar 'ius>etts> Figure 10. Study marsh locations showing a central Gulf of Maine coast distribution (source: Google Earth 2010) Objectives Based on preliminary field observations, I created conceptual models that describe existing conditions and the effects of altered hydrologic regimes on natural and created salt marsh habitat, and these conceptual models area presented in Figures 11 through 15. 15 The goal of my study was to further our understanding of the similarities and differences in ecological structure and function among natural and created, pool and channel habitats. More specific objectives and hypotheses that guided my study of the effects of created hydrologic features on the structure and function of salt marsh habitat in New England are: 1. Objective: To further our understanding of the localized effects of altered salt marsh hydrologic regimes on the salt marsh self-maintenance process involving hydrology, soil characteristics, and marsh surface elevations, and the implications for habitat stability with regard to sea level rise. Hypothesis: Created hydrologic features alter physical parameters, soil processes, and surface elevations of adjacent salt marsh habitat, disrupting the salt marsh self-maintenance feedback process, resulting in distinct differences between natural and created habitats (Figures 11 through 14). 2. Objective: To further our understanding of the localized effects of altered salt marsh hydrologic regimes on physical factors and their influence on adjacent plant community structure. Hypothesis: Created hydrologic features alter physical parameters that influence plant community structure and primary production, resulting in distinct 16 differences among natural and created, pool and channel habitats (Figures 11 through 14). 3. Objective: To investigate the trophic structure of created and natural pool habitat, and assess the value of the habitats in terms of fish productivity and invertebrate prey community structure. Hypothesis: Ditch-plug creation alters physical parameters that influence fish productivity, invertebrate communities, and trophic webs, resulting in distinct differences among natural pools and ditch-plugs (Figures 15 and 16). 17 Natural Creek Habitat Conceptual Model Moderate pore water drainage promum high manh vegetation growth Sedanentaoowrtion beyond banfci during spring tides contributes to buflt density, mrah accretion, and sett maintenance Sloping banks supporting taMwmStwtbaaiBmffora contribute to hibitit heterogeneity and MdNnent rapping Fibriofeapric peat sofls wtti higher mineral aonoent,ation Diverse Kve vegetation contributes to aoi ihmgtli and stability More narrow ground water tabic fluctuation nrfth better drained wis along creek banks Low Low Redox Nob vtioml 3005 Figure 11. Conceptual model describing physical and biological parameters in creek habitat 18 Created Ditch Habitat Conceptual Model Reduoed sediment Low dfrerafty high marsh aocr«fontxj WahTKH CL Wider grcwid Organic ftbridhemic paat soR horizon witti waar table JhjO reduoed mineral content dua to low surtoial fluctuation mneval aocretion and higher betov^jround HjO plartfbaomass contributes to marsh eooretian Low Tide and salf maintenance Bcttv drained toils oontlbut* to Live vegetation and reduoed soil saturation tower ground water at low tide contributes ID soi strength and sUbitity Hyi Sulfides Medium —» Redox Medium Rob VMOflrt 3005 Figure 12. Conceptual model describing physical and biological parameters in ditch habitat 19 Natural Pool Habitat Conceptual Model Dmii wepetatinci mmim ftirlirm nn hydrulogfe Itoii, reducing erosion potential, increasing sediment deposition. and so# organic Diverse high marsh vegetation sonlint, vtwh contributes to nwsh aooraton induding short-form S.a«ernliof» and sail maintenance increases habitat heterogeneity Biomass torn dvanta Area of dacreased toil HaO HjO vegetation along with pore water saturation reduoad pore water HjO concentrations provide a stab)* mixture of fffarioftiernfe organic and I 'QfcyY-rrvr mineral soils limited area of increased soft pore water Aquatic vegetation (Rvppt* mariSma) saturation Diverse 6ve vegetation and growth along pool bottoms reduoad soi saturation contributes to sol stvngth and stability More unowned bathymetry and abrupt Moderate high Sulfides —> stable pool sides tMth overhanging margins Moderate Redox Low MbvnoariSDK Figure 13. Conceptual model describing physical and biological parameters in natural pool habitat 20 Ditch-plug Pool Habitat Conceptual Model less cffvcrte vegetation dominated by pioneering ipaciai (l«.. ffajpomfr eimpeee) reduoes friction on hydrologic flows. increases suspended sedments and erosion, and increases potential tor wind derived wave erosion andloss of sedvnents from fte habitat preventing marah aocretion Expansionof permanent surface water into the sunouncfing high marsh reduoes redox Vegetation die-back ams potential, increases habitat stress, reduoes with increased temperatures, vegetation oover. and decreases habitat high tatinky, and bare SoU Permanently saturated and degradwg hermofeapric organic and mineral soils Permanent surface water high in algae and sutfur bacteria over-top vegetation Depleted above and below ground Pools with unvegetated. undulating dM-back and bare sol vegetation biomass production and bathymetry and deep sump oenters prolonged aoi saturation leads to the promote water oottam stratification and breakdown of soft structure and kiw dissolved oxygen Permanent soil saturation an reduced five subsMenoe, deooupMng fie self- vegetation degrades soil profile integrity, maintenance process decreasing soil strength and stabiflty High High Redox Low nsbvinrtaaos Figure 14. Conceptual model describing physical and biological parameters in ditch-plug habitat 21 Natural Pool Habitat: Trophic Structure Conceptual Model DtvarM tagh marVi Vtgfftibon including Omhanging pool nwgms aid shoMbim & aDamNon inonaxs Omrtanging (ringing vagatafcn vagatationpnwidaMiforag* moum halaroganaity. mduoaa bottom up banafto IMi and imartabralas by and iduge habitat, andhalpi control. and Inomam inwnabrala •hading. ragutaSngtemparaturai. and modarat* volar quafty lUmi and abundanoa pnwidng raftjg* and feraga habitat MnmH (temp, uKnity. DO.) Aim of daomand nil para Above and below giound water oonoaMnHon bkxnass from Figure 15. Conceptual model describing parameters influencing trophic characteristics in natural pool habitat 22 Ditch-plug Pool Habitat: Trophic Structure Conceptual Model Lest djvm vegetation Vegetation dte-fcacfc and Encroachment of permanent surface water dorriruted by piomcrinQ areas of increased on surraurKfing high rnarah provides iptOWi (i.e., SL CUTDpflM) eyosed sedanenls. habitat tor mosquito larvae. tfters alter* resource aveiabSty temperatures. evaporation, invertebrate community oompmiliuii due and hiQh salinity adversely to increased anaerabio conditions, and consumption, •taring impact invertebrate increased ftsh foraging presenoe outside of invertebrate oonvnunity community structure High temperatures. low D.O.. the primary deep pod. increasing top structure from the bottom up and high safirrty provides down pressure on invertebrates outside of unfavorable water quality far the primary pool habitat fish habitat, leading to reduced fish production an Permanency saturated and degredmg soR contributes to depleted above and below giuutid vegetation biomass, leedng to fte breakdown of soil Unvegetated. undulating Permanent surface water high in saucture, increased sol sulfide*, bathymetry reduoes fish algae and suAir bacteria over habitat quality I top vegetation dte-becfc and bar* subsidence, and surface water impoundment, and reduoed invertebrate soB limits primary production and Stratified water habitatqualty alter* invertebrate community aofurm finite structure (richness, evenness, Deep sump in oenter of pool habitat function composition and dwerwty) (remnants of deep ditch and afters habitat Reduced beiow ground vegetation channel) prevents water mudng. and communis biomass, saturated anaerobic soils, and creating hypoxic conditions structure Permanent sol saturation and increased sulfides reduce invertebrate detrimental to fish and species riohness and abundanoe, reduced vegetation abundance altering invertebrate community and biomass degrades soi stucture and composition profile integrity, decreasing soi strength and stabifity. increasing Hp erosion potential, increasing > suspended sofids, decreasing — water quaSty. altering fish and invertebrate ptesenoa Mbvmniaoos Figure 16. Conceptual model describing parameters influencing trophic characteristics in ditch-plug habitat 23 Organization of the Dissertation Chapter II explores the similarities and differences in hydrologic regimes among created and natural, pool and channel habitats, and the effects of altered hydrology on the salt marsh self-maintenance process. I did this by comparing surface and groundwater hydrology, soil characteristics, pore-water chemistry, and marsh surface elevations in three back-barrier marshes. Chapter III examines the effects of physical parameters observed in Chapter II on the plant community structure and primary production for each of the four habitat types (creek, ditch, natural pool, and ditch-plug). I did this by comparing species richness, abundance, diversity, and biomass in the same three back-barrier salt marshes used to collect data for physical parameters analyzed in Chapter II. Chapter IV compares the trophic structure, habitat connectivity, and habitat value in terms of fish productivity and invertebrate community structure in natural pool and ditch-plug habitats. I did this by conducting a controlled experiment to assess fish productivity in the two habitat types at a back-barrier marsh in Wells, Maine. I used the same study site to assess invertebrate community structure within and adjacent to the two pool habitats. The stable isotopes S13C and 815N were used to analyze similarities and differences in the trophic webs and habitat connectivity between the two habitats. This work was done as part of a National Estuarine Research Reserve Graduate Fellowship, and has been submitted as a report to the National Oceanic and Atmospheric Administration. Final conclusions with implications for management and research are provided in Chapter V. 24 CHAPTER II DITCHING AND DITCH-PLUGGING IN NEW ENGLAND SALT MARSHES: EFFECTS ON HYDROLOGY, ELEVATION, AND SOIL CHARACTERISTICS Abstract The interactions involving marsh surface elevation, soil characteristics, and hydrologic regimes result in feedbacks that regulate the salt marsh self-maintenance process, and these interactions vary among habitat types. Using natural tidal creeks and pools as controls, I examined the effects of ditching and plugging, respectively, on hydrology, surface elevations, and soils. Results indicated that the long-term effects of ditching were minor when compared to creek habitat. In contrast, created ditch-plugs did not mimic natural pools and showed significant effects on these components of the salt marsh self-maintenance process. Introduction Hydrologic alteration is a common management tool used in salt marsh restoration and enhancement projects, yet the long-term effects of such modifications on marsh processes with regard to sea level rise and habitat persistence receives little attention. The objective of my study was to characterize the hydrology of natural and created water features and determine the localized effects of altered hydrologic regimes on physical soil parameters in adjacent habitat (i.e., within 20 meters). Thus, my study focuses on localized responses of the habitats to altered hydrologic regimes, rather than whole-marsh habitat comparisons made in previous studies (Adamowicz and Roman 2002; James-Pirri et al. 2011). 25 My study provides some answers to the questions: how does ditching and ditch- plugging in salt marshes affect hydrologic regimes; and how do the altered hydrologic regimes affect soils in adjacent habitat? My approach to answering these questions was to use natural creeks and pools as controls in comparison with created ditches and ditch- plugs. Results of my study provide insights into the effects of ditching and ditch- plugging on hydrologic and soil components of the salt marsh self-maintenance process, and in doing so allows us to infer habitat stability with regard to sea level rise. Materials and Methods Study Areas Sampling took place during the summer of 2005 at three back barrier salt marshes located along the Gulf of Maine coastline from Massachusetts to central Maine (Figure 17). Habitat in each marsh was similar, consisting of typical New England high marsh dominated by Spartina patens mixed with Distichlis spicata, Juncus gerardii, short-form Spartina alterniflora, and a variety of halophytic forbs. Creek and ditch banks were dominated by tall-form S. alterniflora mixed with A triplex patula, Sueda spp., S. patens, and Salicornia europaea, among others. Natural pool edges were dominated by short-form S. alterniflora, S. patens, and D. spicata, while ditch-plug edges were dominated by tall-form S. alterniflora and S. europaea. 26 Spra<>ue River Marsh ' hatiiKi'N Creek Marsh Parker River Marsh WiSr. ;K n.lb Figure 17. Study marsh locations showing a central Gulf of Maine coast distribution (source: Google Earth 2010) Parker River Marsh, located on Plum Island in Newburyport, Massachusetts, is part of the U.S. Fish and Wildlife Service (USFWS) Parker River National Wildlife Refuge (NWR) and is the southern-most marsh in this study. The Parker River Marsh is a 500-hectare back barrier salt marsh (42° 44' 36.06" N; 70° 48' 10.05" W), and is within the Plum Island estuary (Figure 18). Parker River Marsh receives tidal flow from the Gulf of Maine primarily via Plum Island Sound. Several creek channels meander through 27 the marsh, with smaller tributaries extending into the high marsh. Natural and created pools occur throughout the marsh. Ditches connected to the main channels were dug perpendicular to the upland edge throughout the marsh during the 1930's by the CCC. The three northern-most ditch-plugs that were sampled during my study were constructed in 2002, and the southern-most ditch-plug was constructed in 1995. Figure 18. Parker River Marsh - Newburyport, Massachusetts (source: MassGIS 2010), ellipse shows the study area 28 Chauncey Creek Marsh in Kittery, Maine, is a 22-hectare back barrier salt marsh, and is part of the U.S. Fish and Wildlife Service Rachel Carson NWR (Figure 19). It is located approximately 40 km north of Parker River Marsh (43° 05' 13.58" N; 70° 39' 52.46" W). Chauncey Creek Marsh receives tidal flow through Chauncey Creek, which is a 2 km-long tributary to the Piscataqua River at Portsmouth Harbor. A creek channel meanders through the marsh with several smaller tributaries extending into the high marsh. Natural and created pools occur throughout the marsh. Ditches connected to the main channel were dug perpendicular to the upland edge throughout the marsh during the 1930's by the CCC. All four ditch-plugs sampled during my study were constructed in 2002. Figure 19. Chauncey Creek Marsh - Kittery, Maine (source: MaineGIS 2010), ellipse shows the study area 29 Sprague River Marsh in Phippsburg, Maine, is an 86-hectare back barrier salt marsh that formed over time in a narrow glacial valley (Figure 20). Part of the Bates Morse Mountain Conservation Area (43° 44' 00.63" N: 69° 44' 42.33" W), Sprague River Marsh is jointly managed by Bates College and The Nature Conservancy. Sprague River Marsh is located approximately 100 km north of Chauncey Creek Marsh and is the northern-most marsh in this study. The marsh receives tidal flow from the Gulf of Maine via the Sprague River. Although ditching had occurred at Sprague River Marsh for agricultural purposes since the 1700s, the most extensive human impact occurred in 1958 when a local resident dredged a large central channel the length of the marsh and smaller ditches extending perpendicularly from upland borders to the dredged central channel (Sebold 1998; Bohlen 2007). A narrow gravel and dirt causeway built in the 1940s cuts across the marsh at the northern end, and a bridge over the main channel allows tidal flow to pass under the causeway. In 2002 riprap was removed from below the bridge, \ improving flow to the northern portion of the marsh; however, the causeway itself is slightly higher in elevation than the surrounding marsh surface, and impedes sheet flow over the marsh surface to some degree even during spring tides. Remnants of the primary creek meander through the marsh with several smaller tributaries extending into the high marsh. Natural pools occur throughout the marsh. The two northern-most ditch-plugs that were sampled during my study were constructed north of the causeway in 2002, and the two ditch-plugs at the southern end of the marsh were constructed in 2000. 30 Figure 20. Sprague River Marsh - Phippsburg, Maine (source: MaineGIS 2010), ellipse shows the study area Study Design Data collection at all three marshes targeted habitats adjacent to four types of water features (ditch, ditch-plug, creek, and natural pool). To ensure a broad distribution across the marsh, aerial photos of each marsh were divided into three equal sections (north, middle, and south). A grid was placed over each stratified marsh aerial photo, and each grid intersection was numbered. A random numbers table was used to choose grid 31 intersections in each of the marsh strata, and water features for each habitat type closest to the chosen grid intersections were used as sample stations for the study. Each study marsh contained four replicates of each of the four habitat types. Eight transects of 20 m in length were established for each habitat type (two as subsamples per replicate). Transect locations were limited by the availability of specific habitat types (i.e., ditch-plugs). Transects were not randomly selected, but were selected without bias at each replicate. Unbiased placement was accomplished by extending transects outward from the edge of each water feature at a location closest to the center of the water feature that allowed the transect to end at least 10 meters away from any other water feature. Although preliminary data suggested that the hydrologic influence of water features extended outward up to 18 meters from the edge of the water body, 10 meters was the maximum distance from adjacent water features that a 20-meter transect could consistently be located throughout each study marsh. To minimize the effect of isolated marsh characteristics, representative habitats were subsampled on both sides of primary channels. All data were recorded in field books or on data sheets, and transferred to excel databases. Sampling methods were consistent among all study sites. The Global Program of Action Coalition for the Gulf of Maine (GPAC) protocol (Neckles and Dionne 1999) was used as a basis for field data collection. Attributes sampled included ground and surface water levels, marsh surface elevations, and soil characteristics (salinity; redox potential; bulk density; percent organic content; carbon storage, and soil strength). 32 Hydrology Hydrology was sampled at each marsh separately. Water levels were sampled along one transect randomly selected at each habitat replicate during a spring/neap tidal cycle. Sampling each habitat type at the same time allowed for comparison of hydrologic conditions at each habitat during the same tidal cycle, and eliminated sampling bias due to variable tidal conditions. Three automatic capacitance water level data loggers (Odyssey Data Recording 2005) were placed along one transect for each habitat type at 2, 7, and 15 meters from the edge of the water feature (12 water level loggers). Water level data loggers were enclosed in 5.08 cm diameter PVC pipes with 6.4 mm holes drilled on four sides every 10 cm for the length of the pipe. The PVC pipes extended 200 cm above and 100 cm below the marsh surface to record both surface and ground water levels. One additional data logger was placed in the center of the main tidal channel as a downstream tidal reference. Elevations of the water level loggers were determined using a self- leveling laser level (CST/Berger 2005) tied into state highway benchmarks, and water levels were recorded relative to NAVD 88. Data loggers were programmed to record water levels every 10 minutes and data were uploaded to excel files. This procedure was repeated by rotating sampling among all four habitat replicates for a total of four transects sampled per habitat type at each study marsh. Soil Characteristics Soil cores were collected at 2, 7, and 15 meters along eight transects for each habitat type (ditch, ditch-plug, creek, and natural pool) in each sample marsh. A three- centimeter diameter Eijkelkamp gouge auger was used to remove the first 20 cm of soil at each core location. The Eijkelkamp auger has an open-face design that allows the auger 33 to be inserted without compacting soils. Each core was visually examined at the time of collection, and no signs of compaction were observed in any of the core samples. Cores were cut into four sections, each five centimeters in length. Core segments were placed in individual Ziploc bags, stored on ice in coolers in the field, and transported back to the lab for processing. Each core was placed in individual tins, weighed, and dried at 60°C for three days (constant weight). Dried cores were re-weighed to determine bulk density as mg/cm3. Cores were then combusted in a muffle oven at 450°C for three hours, cooled, and reweighed to determine carbon storage as mg/cm3 of carbon. Percent organic matter was calculated as the proportion of total dry core weight. Carbon content was calculated based on the equation presented in Craft et al. (1991): Organic carbon = (0.40)LC>I + (0.0025)LC)I2 Soil strength was measured using a Field Scout SC-900 soil compaction meter (Spectrum Technologies, Inc. 2005). Soil compaction readings were collected at 0, 2, 7, 15, and 20 meters along all eight transects per habitat at each marsh. The soil compaction •j meter measured soil strength in kilograms per square centimeter (Kg/cm) at one-centimeter intervals down to a depth of 50 cm below the soil surface. Soil pore-water salinity was measured in parts per thousand (ppt) in each habitat type using a hand-held temperature-corrected refractometer. Salinity wells were made of 1.27 cm diameter PVC pipe with 4 mm holes drilled on four sides and spaced every 2.5 cm along the length of the below ground portion of the well. The bottom of each well was sealed, and the top of the wells had removable elbows for ventilation that prevented 34 entry of surface water. Salinity wells were inserted 15 cm into the marsh soils at 2, 7, and 15 meters along four transects for each habitat type. The wells were placed 25 cm away from each water level data logger. A syringe with plastic tubing was used to extract all water from the well. When water had re-filled the well, a water sample was collected and the salinity level was measured. Soil redox potential was measured using a hand-held redox meter, double junction reference, and platinum-tip electrodes (Faulkner et al. 1989) inserted 10 cm into the soil at 2, 7, and 15 meters along four transects for each habitat type. Soil redox readings were collected at the same locations used for hydrology and pore-water salinity data collection. Eh was calculated by adding +244 mV to each value to account for the potential of the reference electrode (Burdick et al. 1989). Marsh surface elevations were measured using a self-leveling laser level (CST/Berger 2005) tied into state highway benchmarks. Marsh elevations were surveyed at 0,2, 7,15, and 20 meters along each transect. A nylon meter tape was stretched out on the ground between wooden stakes marking the ends of each transect. A sighting rod was placed on the tape at each survey point. Placing the sighting rod on the tape prevented the stick from sinking into the soil and helped to ensure the survey reading was representative of the marsh surface. Statistical Analysis Data were compiled into Microsoft Excel® spreadsheets and statistically analyzed using JMP® statistical software (SAS Institute, Inc. 2010). The alpha level was set at 0.05 to control for Type I Error during statistical analyses. Pearson r correlation was used to assess relationships among soil characteristics and water levels. Each dependent metric 35 (i.e., water level, soil characteristics, or marsh surface elevation) was analyzed separately using a two-way ANOVA model blocked by marsh, with water feature (pool vs. channel), and habitat type (natural vs. created), as independent variables, along with the interaction term water feature x habitat type. Data are presented as means ± Standard Error (SE). All analyses included Tukey-Kramer HSD post-hoc means comparisons. Data were examined to ensure they satisfied the assumptions of the general linear model (normal distributions, no extreme outliers, and evenness of variance). Where necessary, data transformations were performed to meet these assumptions. Water level and soil compaction data were log(x + 1) transformed, and bulk density data were square root transformed. Transformations of marsh surface elevation data did not satisfy the assumptions of normality, so the Wilcoxon\Kruskal-Wallis non-parametric test was used to compare means for this metric. Discriminant function analysis was used to see if data could predict habitat type (creek, ditch, natural pool and ditch-plug) based on marsh physical characteristics (bulk density, carbon storage, percent organic matter, soil strength, salinity, redox potential, and water level). The number of discriminant functions used in a model is based on either (a) g-1, where g is the number of categories in the grouping variable (habitat, dependent variable), or (b) the number of discriminating (independent) variables, p, whichever is less. This study used g-1, which resulted in three discriminant functions included in analysis of the means of each of the 48 transects. 36 Results The 2-way ANOVAs revealed differences in water level and soil strength among the three study marshes, with lower water levels found for channel habitats at Chauncey Creek Marsh relative to the levels recorded at Parker River and Sprague River marshes, and an inverse relationship between soil strength and marsh latitude. However, hydrologic, soil characteristics, and pore water patterns were consistent for each habitat type among all three study marshes (Appendix A). Hydrology Water levels relative to the soil surface were significantly different between created and natural (drier), and between pool and channel (drier) habitats based on the 2-way ANOVA (Table 1). In addition to main effects of habitat type (created or natural) and morphology of water feature (pool or channel), their interaction was not significant; however, Tukey-Kramer post hoc test showed significantly higher water levels in ditch- plug habitat than all other habitat types (Figure 21). Water level in habitat adjacent to natural pools was significantly higher than in habitat adjacent to creeks and ditches as well. Table 1. Two-way ANOVA results for hydrology and soil characteristics (DF=5, 6; n=12) Feature Type Interaction (channel vs. pool) (created vs. natural) (feature by type) Metric F p-Value* F p-Value* F p-Value* Water Level 52.56 <.0001 12.99 0.0020 1.56 0.2101 Salinity 41.70 <.0001 21.71 <.0001 22.73 <.0001 Redox 19.74 <.0001 30.48 <.0001 12.18 0.0011 Soil Strength 1.12 0.2967 19.51 <.0001 26.92 <.0001 Bulk Density 11.70 0.0014 6.58 0.0141 3.34 0.0751 Carbon Storage 2.53 0.1192 20.54 <.0001 13.76 0.0006 Percent Organic Matter 6.44 0.0151 0.16 0.6938 11.82 0.0014 *Bold values indicate significant effects; Feature = channel or pool; Type = natural or created 37 Created Ditch Natural Creek Natural Pool Figure 21. Water level by habitat relative to marsh surface (0 cm) (values are means from 20 m transects of four habitat replicates from three-marshes, pooled together ± SE; n=12) In support of these water level differences, tidal stage duration curves are presented for each habitat type in Figure 22. Water levels were observed to be higher for a greater percentage of time in habitat adjacent to pools than channels, and higher for longer periods of time in created than natural habitats. Water level recorders also showed tidal ranges were higher and drainage was more extensive in natural habitats compared with created habitats (Table 2). Water level was positively correlated with salinity, and negatively correlated with redox (Table 3). 38 70 Created Ditch 60 Natural Creek 50 Ditch Plug Natural Pool 40 "E 30 o © 20 © u. 10 & 1U 5 0 10 19 29 -10 -20 -30 -40 Percent of time Figure 22. Tidal stage duration curves showing the extent and duration of surface and ground water inundation by habitat. The marsh surface is at zero centimeters (values are means from 20 m transects of four habitat replicates from three marshes, pooled together; n=12). Table 2. Water levels by habitat (values are means from 20 m transects of four habitat replicates from three marshes, pooled together; n=12) %of %of %of Range Mean % of time 5 time time time Habitat (cm) (cm) Marsh Surface < -5cm* S -10cm* £-15cm Ditch 83.00 -2.52 23.69 53.05 29.49 14.43 Creek 93.00 -4.01 22.19 58.32 40.53 16.59 Ditch-plug 63.00 5.51 91.03 2.17 0.07 0.00 Natural pool 74.00 -0.49 46.74 20.21 2.36 0.10 * Root zone 0-10cm below marsh surface 39 Table 3. Pearson's r Correlation results for hydrology and soils characteristics. Values are based on means from paired 20 m transects of four habitat replicates from three marshes, pooled together and presented as Pearson r, (p value*); n=12. Bulk Percent Soil Redox Water Density Organic Strength Salinity Potential Level Carbon 0.28 0.24 0.47 -0.32 0.37 -0.18 Storage (0.0568) (0.9340) (0.0007) (0.0288) (0.0094) (0.2266) Bulk -0.82 -0.01 -0.30 0.29 -0.23 Density (<0.0001) (0.9479) (0.0413) (0.0432) (0.1220) Percent 0.28 0.08 -0.14 0.08 Organic (0.0586) (0.6053) (0.3312) (0.5928) Soil -0.22 0.32 -0.04 Strength (0.1359) (0.0264) (0.7816) -0.30 0.33 Salinity (0.0404) (0.0226) Redox -0.35 Potential (0.0155) •Bold values are significant Marsh Surface Elevations Comparisons of marsh surface elevations by habitat feature and type followed what might be expected from the water level results, but with some exceptions. As expected, marsh surface elevations were lower in ditch-plug habitat than in all other habitats showing subsidence (Figure 23), but not significantly different from creek (Table 4). Marsh surface elevations were highest for natural pool habitat even though water levels were greater than creeks and ditches. 40 140 E B _o_ 120 BC oo O > 100 < Z c o 80 ffl > o 60 Ul a> o € 40 3 CO JE 20 e (0 0 Created Ditch Natural Creek Ditch Plug Natural Pool Figure 23. Marsh surface elevation by habitat (values are means from paired 20 m transects of four habitat replicates from three marshes, pooled together ± SE; n=12) Table 4. Marsh surface elevation Wilcoxon/Kruskal-Wallis non-parametric test (values are based on means from paired 20 m transects of four habitat replicates from three marshes, pooled together; n=12) Habitat Comparisons p-Value* Natural pool Ditch-plug <.0001 Natural pool Ditch 0.0005 Natural pool Creek 0.0026 Ditch-plug Ditch 0.0054 Creek Ditch-plug 0.1646 Creek Ditch 0.8845 *Bold values are significant 41 Since marsh surface elevations were measured at five distances along each transect, we can examine the trends with distance for the four habitat types. Elevations in natural pool habitat remained fairly consistent for the 20 meters sampled, and were consistently higher than the other three habitats (Figure 24). Elevations in creek habitat trended lower than ditch habitat from 0-2 meters, but elevations were similar for the two habitats from 7-20 meters distance. Elevations in the two channel habitats trended higher with distance and converged towards natural pool elevations at around 15-20 meters. Marsh surface elevations in ditch-plug-habitat trended slightly higher with distance, but were considerably lower than elevations in natural pool habitat from 0-20 meters, and were lower than channel habitats from 7-20 meters. E 120 I 110 z 100 90 - * - Ditch 80 — Ditch-plug • - Creek 70 Natural Pool W ra 60 s 0 2 7 15 20 Distance from edge of water feature (m) Figure 24. Marsh surface elevations by habitat (values are means from paired 20 m transects of four habitat replicates from three marshes, pooled together ± SE; n=12) 42 Pore-water Chemistry Pore-water salinity was different among habitats and the interaction of marsh feature and habitat type was significant (Table 1). The Tukey-Kramer post hoc test showed significantly higher salinity levels in ditch-plug habitat (31 ± 1 ppt) relative to all other habitat types (Figure 25). There was no significant difference in pore-water salinity for habitat adjacent to natural pools (27 ± 1 ppt), creeks (26 ± 1 ppt), or ditches (26 ± 1 ppt). Soil pore-water salinity was positively correlated with water level, and negatively correlated with bulk density, redox potential, and carbon storage (Table 3). Created Ditch Natural Creek Ditch Plug Natural Pool Figure 25. Pore-water salinity by habitat (values are means from 20 m transects of four habitat replicates from three marshes, pooled together ± SE; n=12) 43 Pore-water redox potential was different among habitats and the interaction of marsh feature and habitat type was significant (Table 1). The Tukey-Kramer post hoc test showed significantly lower redox potential in ditch-plug habitat (-166 ± 25 mV) than all other habitat types (Figure 26). Differences in pore-water redox potential for habitat adjacent to natural pools (31 ±24 mV), creeks (52 ± 27 mV), or ditches (8 ± 24 mV) were not significant. Pore-water redox potential was positively correlated with soil strength, bulk density, and carbon storage, and negatively correlated with water level and salinity (Table 3). 100.00 A A T A 50.00 1 4 0.00 . T ~ -50.00 E, B g-100.00 -150.00 • Ditch i i • Creek -200.00 • Ditch Plug • Natural Pool -250.00 Figure 26. Soil redox potential by habitat (values are means from 20 m transects of four habitat replicates from three marshes, pooled together ± SE; n=12) 44 Soil Strength Soil strength was different among the habitats and the interaction of marsh feature and habitat type was significant (Table 1). The Tukey-Kramer post hoc test showed significantly higher soil strength in natural pool habitat (5.87 ±0.16 kg/cm ) than all other habitat types (Figure 27). Soil strength was lowest in ditch-plug habitat (3.57 ±0.16 kg/cm2) but not significantly different from creek habitat (4.21 ±0.16 kg/cm2). Created Ditch Natural Creek Ditch Plug Natural Pool Figure 27. Soil strength by habitat (values are means from paired 20 m transects of four habitat replicates from three marshes, pooled together ± SE; n=12) Soil strength with depth below the marsh surface for each habitat type is presented in Figure 28. In each habitat, an increase in soil strength was observed at 3 cm 45 below the marsh surface, reflecting the dense root mass that typically occurs at this depth. Soil strength increased consistently with depth and with strikingly similar patterns, regardless of habitat type. Soil strength in natural pool habitat was consistently higher than all other habitats, followed by ditches. Soil strength was positively correlated with redox potential and carbon storage (Table 3). 8 (N E o o>1 * o<0 c (0 w *35 0) — ^ Ditch K , — Creek +>»c * Ditch Plug E Natural Pool n 1 1 1 1 r CO 0 3 5 8 10 13 15 18 20 23 25 28 30 33 36 38 41 43 46 Depth below marsh surface (cm) Figure 28. Soil strength with depth below marsh surface by habitat (values are means from paired 20 m transects of four habitat replicates from three marshes, pooled together ± SE; n=12) 46 Bulk Density Bulk density was significantly different between created and natural (higher bulk density), and between pool and channel (higher bulk density) habitats based on the 2-way ANOVA (Table 1). The interaction of the main effects was not significant; however, the Tukey-Kramer post hoc test showed significantly higher bulk density in creek habitat (762 ± 46 mg/cm3) than all other habitat types (Figure 29). Bulk density was lower in ditch-plug habitat, however there were no significant differences in bulk density of soils adjacent to ditches (605 ± 28 mg/cm3), natural pool (575 ± 32 mg/cm3), or ditch-plugs (551 ± 30 mg/cm3). Bulk density was positively correlated with redox potential, and negatively correlated with percent organic matter and salinity (Table 3). 900 800 B B Created Ditch Natural Creek Ditch Plug Natural Pool Figure 29. Bulk density by habitat (values are means from paired 20 m transects of four habitat replicates from three marshes, pooled together ± SE; n=12) 47 Carbon Storage Carbon storage was different among habitats and the interaction of marsh feature and habitat type was significant (Table 1). The Tukey-Kramer post hoc test showed significantly lower carbon storage in soils adjacent to ditch-plugs (28.7 ± 1.6 mg/cm3) than all other habitat types (Figure 30). Carbon storage was higher in soils adjacent to natural pools (37.4 ± 1.9 mg/cm ); however, there was no significant difference in carbon storage for soils adjacent to ditches (34.2 ± 1.6 mg/cm3) or creeks (34.1 ± 1.7 mg/cm3). Carbon storage was positively correlated with soil strength and redox potential, and negatively correlated with salinity (Table 3). Created Ditch Natural Creek Ditch Plug Natural Pool Figure 30. Carbon storage by habitat (values are means from paired 20 m transects of four habitat replicates from three marshes, pooled together ± SE; n=12) 48 Percent Organic Matter Percent organic matter was different among habitats and the interaction of marsh feature and habitat type was significant (Table 1). The Tukey-Kramer post hoc test showed percent organic matter was significantly lower in soils adjacent to creeks (14.7 ± 0.9 %), but not different from that of ditch-plugs (18.0 ± 0.8 %) (Figure 31). There was no significant difference in percent organic matter for soils adjacent to ditches (19.0 ± 0.8%), ditch-plugs, or natural pools (21.6 ± 0.8 %). Percent organic matter was negatively correlated with bulk density (Table 3). Created Ditch Natural Creek Natural Pool Figure 31. Percent organic content by habitat (values are means from paired 20 m transects of four habitat replicates from three marshes, pooled together ± SE; n=12) 49 Discriminant Function Analysis (DFA) My rationale for using DFA was to determine if marsh physical characteristics could predict habitat type. Discriminant function analysis identified significant differences in group membership for soil characteristics and hydrology among the four habitat types (Wilks' Lambda = 0.14; p<0.0001), with 69% of the transect memberships correctly classified into each of the four habitats (Figure 32, and Table 5). Most misclassifications occurred between creeks and ditches. The first two canonical variables captured 99% of the variation and will be interpreted. The first discriminant function axis (Canonical 1) explained 77% of the variation in soil and hydrology characteristics among the four habitats (Table 6), and represented edaphic stress associated with the degree of flooding. The highest positive correlation for Canonical 1 was redox potential, and most negative correlation was water level (Table 7). The lower redox potential and higher water level identified in Canonical 1 for ditch-plugs identified ditch-plug habitat as distinctly different from all others. The second discriminant function (Canonical 2) explained 22% of the variation in dependent variables, and represented soil characteristics associated with soil bulk properties. Canonical 2 was positively correlated with soil strength and carbon storage, and most negatively correlated with bulk density. Canonical 2 identified the relationship between soil strength and carbon storage, and distinguished natural pool habitat as different from all others in terms of these variables. Also, Canonical 2 identified higher bulk density as a distinguishing factor for creek habitat. Canonical 3 contributed very 50 little to the analysis (Table 6); calling out Sprague River marsh which exhibited lower salinity. Carbon Storage Water Level .Redox Salinity lulk Density Canonical Figure 32. Discriminant function analysis comparing habitat type with soil characteristics (values are means from paired 20 m transects of four habitat replicates from three marshes, pooled together; n=12; + = mean; circles = 95% mean confidence; CD = Ditch; DP = Ditch-Plug, NC = Creek, and NP = Natural Pool) 51 Table 5. Discriminant function actual by predicted habitat classification matrix Predicted Habitat Actual Habitat Ditch Ditch-plug Creek Natural Pool Ditch 5 1 3 3 Ditch-plug 1 11 0 0 Creek 4 1 7 0 Natural pool 0 0 2 10 Table 6. Results of discriminant function analysis for soil and hydrology values measured in habitat adjacent to natural and created, pools and channels. Eigenvalues and percents are based on the three canonical functions. Canonical correlations are a measure of the association between groups formed by independent variables and a particular discriminant function. Canonical Canonical Function Eigenvalue Percent Correlation 1 2.85 77.42 0.86 2 0.80 21.66 0.67 3 0.03 0.93 0.18 Table 7. Discriminant function standardized scoring coefficients used to indicate the relative importance of each independent variable for each discriminant function. Bold values identify independent variables that contribute the most to a given canonical. Canonical Percent Carbon Bulk Soil Water Function organic Storage Density Strength Salinity Redox Level Canonl 0.24 0.37 0.30 0.03 -0.27 0.73 -0.53 Canon2 0.04 0.47 -0.68 0.49 -0.01 0.02 0.38 Canon3 -0.34 0.15 0.24 0.31 0.73 0.17 0.18 52 Discussion Hydrology My study determined that hydrologic regimes of created habitats differed from natural habitats, but differences were more distinct for ditch-plugs and less so for ditches. Although mean water level in habitat adjacent to ditches was similar to creek habitat, the longer duration at which pore-water levels were maintained within the rooting zone of ditch habitat contributed to minor differences in soil processes (this chapter) and plant community composition (Chapter 3) observed between the two habitats. Habitat surrounding natural pools had greater tidal ranges and more extensive drainage than found for ditch-plugs. Higher tidal range has been associated with salt marsh building processes and the alleviation of stressful edaphic conditions. Working in Bay of Fundy salt marshes, Chmura et. al. (2001) found that soil accumulation increased with tidal range, and changes in marsh hydrology leading to greater tidal range had the greatest impact on high marsh accretion rates. My study found that although hydrologic alterations of high marsh habitat were the focus of ditches and ditch-plugs, only in ditch-plug habitat were reductions in tidal range severe enough to have dramatic impacts on the surrounding habitat. In South Carolina marshes, Morris et al. (2002) determined that flooding had positive effects on primary production and marsh accretion, yet flooding beyond optimal levels crossed a threshold and negatively impacted plants and marsh accretion processes. Results of my study demonstrate that optimal limits of flooding depth had been exceeded in ditch-plug habitat, as evidenced by prolonged surface water retention, reduction in 53 plant biomass (Chapter 3), loss of soil strength, and resulting subsidence of marsh surface elevations (Figure 24). Previous studies conducted in New England salt marshes described an increase in surface water area associated with ditch-plug habitat 1-4 years post construction (Adamowicz and Roman 2002; James Pirri et al. 2011), and a general increase in pore-water level associated with ditch-plug habitat 1-2 years post-construction (Adamowicz and Roman 2002). The significant differences in hydrologic regimes among natural pool and ditch-plug habitats identified in my study support observations from these previous studies. However, my study sampled ditch-plug habitat 4-10 years post-construction, and indicated that ditch-plugging resulted in smaller tidal range and extended flooding compared with natural pools. Taken together with the work by Chmura et al. (2001) and Morris et. al. (2002), can we conjecture what will happen to the habitat surrounding ditch-plugs over the long term? Bourn and Cottam (1950), and Lesser (1982) working in salt marshes of Delaware observed decreased soil pore-water levels associated with ditch habitat. Adamowicz and Roman (2005) suggested that ditches had the same effect in New England salt marshes. However, studies by Headlee (1939) working in New Jersey, and Provost (1974), working in Florida salt marshes, concluded that ditching did not result in lowered ground water levels. Likewise, Travis et al. (1954) studied salt marshes of Volusia County, Florida, and attributed small differences in plant patterns between ditched and unditched marshes to altered tidal circulation patterns rather than reduced depth to ground water. Their results concluded that marshes throughout the County became wetter as a result of ditching. LeMay (2007) studying marshes in nearby Rowley, Massachusetts, used a grid 54 of tide sticks to assess relative marsh elevation and surface water flow patterns, and determined that ditches conveyed tidal flows to interior areas of the marsh more frequently compared with unditched sections of the marsh. General associations between plant species occurrence and hydrology have been documented (Bertness and Ellison 1987). The studies mentioned above used plants or visual observations of surface water circulation patterns to estimate ground water levels. In contrast, I used automatic data loggers to record ground and surface water over time, which provided a more accurate assessment of water levels and differentiated my study from previous assessments that used plants as a proxy for ground water level. The results of my study do not support previous studies that suggested ditches decreased soil pore- water levels; but rather conclude that ditching did not result in lower water levels. Further, my results showed that ground water remained within the rooting zone of ditch habitat for longer periods of time compared with creek habitat, which would support conclusions by Travis et al. (1954) that soils of ditched marshes actually become wetter than unditched marshes. In addition, the effects of subsidence and pore-water retention in ditch habitat may have had even greater impacts in marshes with more extensive grid ditching, as indicated by the darker, wetter areas surrounding ditches in Figure 4. Marsh Surface Elevations Direct comparisons of marsh surface elevations for natural and created, channel and pool habitats have not been well documented. LeMay (2007) determined relative marsh surface elevation along 75-meter transects in habitat adjacent to ditches, and concluded that interior areas of ditched marshes had lower elevations, compared with unditched areas of the marsh. My study used laser levels to survey marsh surface 55 elevations within 20 meters of ditches and creeks and showed a steady increase in elevation, with mean marsh surface elevations slightly higher in ditch habitat than in creek habitat. I attributed lower mean elevation in creek habitat to sloping banks, which were not present in ditch habitat. In contrast to LeMay (2007), my results showed marsh surface elevations were higher in ditch habitat than creek habitat from 0-2 meters along transects, and only slightly lower than creek habitat from 7-20 meters from the edge of water features. However, since ditches were originally created in high marsh habitat presumably at the same elevation as high marsh surrounding natural pools, the lower marsh surface elevations of ditch habitat from 0-15 meters (Figure 24) indicates that some subsidence has occurred since the time of ditch construction. Stumpf (1983), studied sedimentation associated with creeks in Delaware marshes and concluded that periodic storm events contribute more to marsh accretion than sediment delivered by creeks during daily tidal cycles. However, creeks in Stumpf s study had levees of several centimeters in elevation that trapped coarse materials. My study showed that prominent levees are not typical of creek and ditch banks in northern New England back barrier marshes. Like Stumpf (1983), Chmura and Hung (2004) determined that in marshes of eastern Canada, vertical accretion in the middle and high marsh zones decreased with distance from creeks due to less frequent flooding, and marsh accretion was greater in areas associated with channel flow. In addition, LeMay (2007) found that soil transport and deposition from tidal flow associated with ditches was similar to that of creeks. My study adds to findings by Chmura and Hung (2004) and LeMay (2007) by confirming that hydrologic regimes of ditches that extend into the marsh interior have maintained processes associated with marsh development and self- 56 maintenance, and have enabled marsh surface elevations in habitat adjacent to ditches to keep pace with creek habitat for the past 70+ years. In contrast, creation of ditch-plug habitat does not support processes that maintain marsh surface elevations, but rather facilitates processes that lead to subsidence, as evidenced by the consistently lower elevations at all replicates. This argument is supported by the difference in mean elevation (10 cm) between ditch-plug and ditch habitats relative to the mean accretion rate (2.8 mm/yr) determined by Goodman et al. (2007). At this rate, five-year old ditch-plugs have the potential to accrete 1.4 cm of elevation, which would leave 8.6 cm accounted for by subsidence. Low elevations are not likely to be an artifact of site selection since ditch-plug locations are unbiased with respect to marsh surface elevation. Rather, the location within a marsh for construction of ditch-plugs is the result of ditch presence, equipment limitations, and mosquito population densities; marsh surface elevation surveys are not conducted as part of the ditch-plug site selection process (Meredith et al. 1985; Hruby and Montgomery 1988; Northeastern Massachusetts Mosquito Control & Wetlands Management District 2011). Previous studies have measured surface water depth associated with ditch-plugs in New England salt marshes (Adamowicz and Roman 2002; James-Pirri 2010); however, I am not aware of any studies that have documented marsh surface elevations in ditch-plug habitat. Due to their association with ditches, ditch-plugs would start out lower than marsh elevations adjacent to natural pools (Figure 24). Marsh surface elevation in natural pool habitat is fairly consistent from 0-20 meters, and is significantly higher than elevations in ditch-plug habitat. In contrast, elevation at the edge of ditch-plugs is similar to ditch banks, and rises only slightly with distance from 0-20 meters, while elevations in 57 ditch habitat converges more closely to natural pool elevations at 20 meters from the edge of the water feature. Reed and Cahoon (1992), working in marshes of Louisiana, found a positive relationship between plant vigor and marsh elevation. The authors determined that stressful edaphic conditions associated with poorly drained low areas of the marsh led to plant dieback, resuspension of mineral soils, and reduction in marsh building processes. Bauman and Day (1993) determined that resuspension of mineral soil is a mechanism for subsidence and marsh collapse in Louisiana marshes. Burdick et al. (1989) working in Louisiana observed a positive relationship between plant physiological stress and lower areas of the marsh. DeLaune et al. (1994), also working in Louisiana marshes, attributed prolonged surface water retention to plant dieback, reduced soil trapping, loss of turgor in below-ground roots, and subsidence of marsh soils. Extended periods of surface water impoundment, greatly reduced soil conditions (Figure 33), and reduced plant cover (Chapter 3) were observed in ditch-plug habitat for all replicates in my study. Since the other habitats could not have accreted 2:10 cm during the time since the ditch-plugs were created, I attribute the lower ditch-plug elevations to subsidence. My results, therefore, show that subsidence has occurred in marsh soils adjacent * to ditch-plugs as a result of poor drainage, plant loss (reduced organic inputs, collapse of dead below-ground biomass, diminished soil trapping, and resuspension of mineral soils), and increased decomposition of below-ground organic matter (loss of carbon storage). Thus, altering hydrologic regimes through ditch-plug creation inhibits the salt marsh self- maintenance process, and leads to increased open water habitat and loss of salt marsh habitat quite a distance beyond the created ditch-plug pool itself. 58 Figure 33. Surface water impoundment and plant dieback in ditch-plug habitat at (a) Parker River, (b) Chauncey Creek, and (c) Sprague River marshes (Photos by Rob Vincent). 59 Pore-water Chemistry Pore-water chemistry conditions varied among natural and created habitats, but differences were greatest in pool habitat. The significantly higher salinity and lower redox potential observed in soils adjacent to ditch-plugs relative to natural pool habitat demonstrate the dissimilarities that exist in edaphic conditions between those habitats. Working in New England, southern US Atlantic, and Gulf coast marshes, Bertness and Pennings (2000) observed reduced salt accumulation in association with frequent tidal flushing. Although ditch-plug habitat occurred at lower elevations of the marsh platform, pore-water salinity was consistently higher, likely due to less frequent flushing, greater retention of saline water, and evaporation, leading to the concentration of salinity in ditch-plug habitat. Bertness and Ellison (1987) concluded that soil salinity in New England salt marshes reached its highest levels at unvegetated mid-marsh locations due to high surface temperatures and evaporation rates. During hot dry summer periods that coincide with neap tides, large areas of unvegetated soil concentrate salinity in ditch-plug habitat (Figure 34). Mendelssohn and McKee (1988) conducted plant transplant experiments in Louisiana marshes and observed low redox potential and high sulfide levels in association with waterlogged soils at lower elevations of the marsh platform. Burdick et al. (1989), Reed and Cahoon (1992), and Howarth and Teal (1979) came to similar conclusions with their work in marshes of Louisiana and Massachusetts. Lower marsh surface elevations, persistently higher water levels, and poor drainage are characteristics of ditch-plug habitat, which contributed to significantly lower redox potential and provide conditions suitable for elevated sulfide levels for this habitat. Mendelssohn and McKee 60 Figure 34. Soils and plant dieback areas in ditch-plug habitat of (a) Parker River marsh and (b) Sprague River Marsh that were exposed to increased temperatures, evaporation, and concentrated salinity levels during hot dry summer neap tide periods Arrows are pointing to salt deposits (photos by Rob Vincent). 61 (1988) noted that such conditions lead to positive feedbacks that expand the radius of waterlogged degraded soils and gradually increase the footprint of open water habitat. This process was evident in ditch-plug habitat for each replicate in my study, and supports my conceptual model presented in Chapter 1. Pore-water chemistry was similar between creek and ditch habitats. Although soil drainage was slower for ditches (Table 2); pore-water redox potential remained similar in the two habitats. Frequent tidal flushing reduces salt accumulation, especially along channel banks (Bertness and Pennings 2000). Tidal flushing frequency was similar between the two channel habitats and contributed to similarities in salinity levels. Soil Bulk Properties There was little difference in soil strength among ditch and creek, but habitat around natural pools had significantly greater soil strength than the habitat adjacent to ditch-plugs. Soil strength increased steadily with depth in all four habitats; however, the differences in strength among habitat types did not vary with depth. Since soils at 50 cm are approximately 400 years old, this would suggest that the differences in feedbacks associated with elevation and hydrology that regulate soil processes that exist among the habitats have altered the substrates consistently in more recent times. In addition, the spike in soil strength at 3 cm in each habitat is due to the peak in root density present at this depth, indicating soil aeration in the top several centimeters is important for maintaining belowground biomass production. Turner (2010) working in Louisiana marshes used a Hand Vane Tester to analyze maximum soil resistance in terms of torque, and determined that organic content (root and rhizome biomass) contributes greatly to soil strength. The higher carbon storage in 62 natural pool habitat, therefore, contributed to the significantly higher soil strength compared with all other habitats. The significantly lower carbon storage in soils adjacent to ditch-plugs account for the lower soil strength compared with other habitats. Ditch- plugs are excavated out of previously existing ditches; consequently, significantly lower soil strength in ditch-plug habitat compared with ditch habitat indicates habitat degradation over time in response to the altered hydrologic regime. Further, the lower soil strength associated with ditch-plug habitat indicates increased susceptibility to erosion and a higher degree of habitat instability, compared with the other habitats. Soil strength counteracts erosive forces, and is thus a determining factor for channel formation and marsh stability. The relative amount of organic to mineral content (i.e., percent organic content) in soils of ditch habitat was higher than creek habitat, which contributed to higher soil strength in ditch habitat. Computer models developed by Fagherazzi and Sun (2004) showed how channels with greater soil strength were more resistant to erosion and, therefore, maintained more linear and less branching channel development. Hughes et al. (2009) analyzed creek development in North Carolina marshes and determined that tidal range, soil strength, and plant structure on slumping banks contributed to branching channel morphology. Therefore, the higher soil strength, lack of slumping banks, and smaller tidal range are contributing factors in maintaining the vertical channel walls and linear character of ditches over the past 70 + years. Carbon storage differed in soils among created and natural habitats, indicating processes associated with soil development or loss varied among the habitat types. Soils in habitat adjacent to the two channel types (ditch and creek) had similar carbon storage (mg/cm3), while carbon storage in the two pool habitats differed. Soils in habitat adjacent 63 to natural pools had the highest carbon storage among the four habitats, which can be attributed to less extensive pore-water drainage and lower organic matter decomposition under anaerobic conditions compared with the channel habitats. Thus, areas adjacent to natural pools act as carbon sinks in high marsh habitat. In contrast, carbon storage in soils adjacent to ditch-plugs was lower than all other habitats. LeMay (2007) collected surface soil "scrapings" in Rowley, Massachusetts, and noted increased mineral deposition and decreased organic matter content in interior sections of ditched high marshes. In contrast, results of my study using soil cores showed lower mineral content in soils adjacent to ditches, as evidenced by lower bulk density and higher percent organic content compared with soils adjacent to creeks. If carbon storage is equivalent between the two habitats, then the difference in bulk density must come from differences in mineral inputs. The added mineral contributions in creek habitat are attributed to higher mineral accretion rates in creek habitat. Turner et al. (2000) analyzed soil cores from marshes in Louisiana, Texas, and Rhode Island, and stressed the importance of below-ground organic content to marsh building processes, and concluded that plant biomass facilitates mineral accretion in salt marsh habitats. The authors noted an inverse relationship between marsh building processes and hydrologic alterations that result in increased flood duration and periodic drying that leads to reduction in soil organic matter. Turner et al. (2000) also concluded that bulk density was inversely related to organic content, and the loss of soil organic content through decomposition is usually permanent in hydrologically altered salt marsh habitat. My results showed that ditch-plugs occurred at lower elevations of the marsh, with extended periods of surface water impoundment and brief seasonal dry periods, 64 conditions that stress plants and reduce biomass (Figures 24, 33, and 34) leading to organic decomposition in soils that impedes the salt marsh building processes. The two pool habitats showed similar bulk density, but ditch-plug habitat had lower carbon storage and lower percent organic content compared with natural pool habitat. These results indicate that the higher concentration of mineral material in ditch-plug soils is due to a higher rate of organic decomposition in ditch-plug habitat, rather than increased mineral accretion. Thus, using only percent organic content to compare levels of organic matter and mineral accretion in soils of different habitats can be misleading. Images of soil cores collected from each habitat are provided in Appendix A. Morris et al. (2002) studied marsh accretion rates relative to sea level rise in South Carolina and described salt marsh self-maintenance as a process where marsh surface elevations are regulated by dominant plants that contribute organic matter to soils while at the same time trap and maintain inorganic material. The authors determined that at lower elevations habitats become unstable, resulting in decomposition of soil organic material if unable to maintain accretion rates in step with sea level rise. Mudd et al. (2009) working in the same South Carolina marsh developed a numerical model that estimated reduced soil carbon storage when transport of inorganic material into the system was reduced. Therefore, one can expect that the feedback involving lower marsh surface elevations (indicating negative marsh accretion), and altered hydrology interrupts soil transport and increases edaphic stress. In addition, loss of plants leads to reduced soil organic content that is characteristic of ditch-plug habitat. These feedbacks undo the marsh self-maintenance function and will likely result in additional marsh habitat loss at ditch-plug sites as sea levels continue to rise. 65 Habitat Distinctions Canonical functions are useful in determining relationships among groups of variables. Morgan and Short (2002) working in salt marshes of the Great Bay Estuary in New Hampshire and Maine, used principal components analysis (PCA) to associate restored marshes to reference sites based on unique sets of physical variables, such as marsh width, fetch, slope, and marsh length. Once restored and reference sites were paired using PCA associations, functional trajectory analysis was used to further analyze marsh development over time. Discriminant function analysis was the canonical correlation used in my study to explore habitat distinctions based on unique combinations of soil characteristics. Discriminant function analysis depicted the differences among the four habitat types with 69% accuracy, and assignment had the most error between ditches and creeks. Soil structure characteristics were the primary variables that distinguished natural pool habitat from the others, suggesting a dynamic equilibrium in the feedback between elevation and hydrology exists, creating habitat stability and maintenance of marsh building processes. Finally, the influence of water level over soil characteristics was most apparent in ditch-plug habitat, which was clearly distinguished from all other habitats and indicated that the level of edaphic stress and habitat degradation observed in ditch-plug habitat was unique. Conclusion Environmental gradients that occur throughout New England salt marshes result from feedbacks involving hydrology, elevation, and soil characteristics, which together with plants regulate the salt marsh self-maintenance process (Harshberger 1911; 66 Chapman 1960; Redfield 1972; Bertness and Ellison 1987; Burdick et al. 1989; Ewing et al. 1997; Hacker and Bertness 1995; Turner et al. 2000; Morris et al. 2002). These complex gradients influence localized conditions, resources, and biological communities. My conceptual model presented in Chapter 1 highlighted the differences that I expected to encounter among the four habitats. I hypothesized that created hydrologic features alter physical parameters, soil processes, and surface elevations of adjacent salt marsh habitat by altering the salt marsh self-maintenance feedback process, and would result in distinct differences between natural and created habitats. The hypothesis was accepted for pool habitat comparisons, but rejected for channel habitats. Interrupting tidal flow and increasing surface water impoundment diminishes the ability of the marsh to keep pace with sea level rise and contributes to salt marsh habitat loss over time (Mendelssohn and McKee 1988; DeLaune et al. 1994; Cahoon and Reed 1995; Turner et al. 2000). My results show that the long-term effects of ditching on hydrology, soils and marsh surface elevations were minor, and the salt marsh building processes in ditch habitat are comparable to those observed in creek habitat. However, the effects of subsidence and pore-water retention may have had greater impacts in marshes with more extensive grid ditching, which warrants further study. In contrast, ditch-plugging showed significant effects on the salt marsh self-maintenance process that exceeded thresholds and led to creeping habitat loss surrounding the created pools, and will lead to additional habitat loss with increased sea level rise. 67 Chapter III DITCHING AND DITCH-PLUGGING IN NEW ENGLAND SALT MARSHES: EFFECTS ON PLANT COMMUNITIES Abstract Salt marsh plant communities are regulated by feedback processes involving hydrologic regimes and marsh physical characteristics, and as expected differ among habitats. Using the study sites in three barrier beach salt marshes, I examined the effects of ditching and ditch-plugging on plant characteristics by means of comparisons to creek and natural pool habitats. Results indicated that ditch and creek habitats were similar in terms of species richness and diversity, but plant cover and biomass were significantly higher in habitat adjacent to creeks. Plant composition in ditch habitat was distinguished by the higher percentage of forb species, while the proportion of tall-form S. alterniflora was much higher in creek habitat, which I attribute to the absence of sloping banks along ditches. These results are indicative of differences in hydrologic and disturbance regimes that can influence competitive and facilitative interactions, habitat structure, and heterogeneity. Results for pool comparisons indicated that plant characteristics were significantly different between ditch-plug and natural pools. Species richness, diversity, and biomass were significantly lower in ditch-plug habitat compared with all other habitats, and plant cover averaged only 30% in habitat adjacent to ditch-plugs, which was significantly lower than all other habitats. Species composition in ditch-plug habitat was characterized by a higher percentage of Salicornia europaea and S. alterniflora, while the relative contribution of high marsh turf and forb species was greater in natural pool habitat. These differences have ecological implications in terms of habitat structure and 68 function of ditch-plug habitat. In addition, increased stress leading to plant dieback due to ditch-plugging has resulted in subsidence that can decrease the stability of ditch-plug habitat and expedite the loss of salt marsh habitat as sea levels rise. Introduction Salt marshes have a long history of human use and anthropogenic impacts, resulting in a mosaic of natural and created water features throughout the marshes of the region (Daiber 1986; Silliman et al. 2009). Naturally-occurring water features include meandering and often multi-ordered creek channels that conduct tidal waters, along with a variety of pools, both permanent and ephemeral. Created water features include ditches constructed by farmers that were expanded and ramified by the Civilian Conservation Corps more than 70 years ago, and ditch-plugs constructed more recently over the past decade. More detailed characteristics of the natural and created, channel and pool habitats found within my study sites were provided in Chapter 1. Salt marsh plant zonation occurs in response to elevation and hydrologic gradients, and the associated edaphic conditions that extend from the seaward edge of the marsh to the upland border. Stronger competitors under benign edaphic conditions (i.e., Juncus gerardii and Spartina patens) occur at intermittently flooded higher elevations (Bertness and Ellison 1987; Emery et al. 2001); while the stress tolerant S. alterniflora is competitively displaced to frequently flooded lower elevations of the marsh (Bertness and Ellison 1987; Bertness 1991). A stunted form of S. alterniflora commonly occurs in poorly drained and nutrient limited areas of the high marsh (Mendelssohn 1979a; Mendelssohn 1979b; Howes et al. 1986). Pioneer species such as Distichlis spicata and 69 S. europaea frequently invade disturbed areas of the high marsh providing habitat heterogeneity until competitively excluded (Bertness and Ellison 1987; Bertness et al. 1992). Forb pannes consisting of a variety of less competitive species also occur throughout the high marsh in response to disturbance and poor drainage conditions (Ewanchuk and Bertness 2004). Previous studies have suggested that plant communities were altered as a result of ditch construction (Bourn and Cottam 1950). Yet others have suggested that ditch-plugging had no effect on surrounding plant characteristics (Adamowicz and Roman 2002; James Pirri et al. 2005; James Pirri 2011). However, these previous studies involved whole-marsh investigations, while my study focuses on the localized effects of hydrologic alterations on plants within 20 meters of the various water features. In doing so, I provide a more accurate assessment of the direct impacts to plant communities from altered hydraulic regimes. The goal of my study was to characterize plants in habitat adjacent to natural and created water features, and determine how plant communities adjacent to ditches and ditch-plugs compare with plant characteristics in habitat adjacent to naturally occurring creeks and pools, respectively. Materials and Methods Study Areas and Design My study took place during the summer of 2005 at three back barrier salt marshes (Parker River, Chauncey Creek, and Sprague River), using the study design, habitat replicates, and sampling transects as described previously (Chapter 2). The Global Program of Action Coalition for the Gulf of Maine (GPAC) protocol (Neckles and 70 Dionne 1999) was used as a basis for field data collection. Attributes sampled included plant percent cover, biomass, species richness, and diversity. Each of the three marshes contained four replicates per habitat type, and each replicate contained paired transects (subsamples) with five plots per transect. Two, 20-meter long transects were established at each of the four replicates, for a total of eight transects per habitat type (ditch, ditch- plug, creek, and natural pool). Transects were on opposite sides of water features, perpendicular to the edge of the water feature. The data from each plot along the paired transects were averaged, resulting in n=12 per attribute and habitat type. Marshes were evaluated separately and once values and patterns were determined to be similar, data from the three marshes were combined for later analyses. Data are presented as three- marsh averages by habitat. Plant Abundance and Biomass A 1 m quadrat was used to visually estimate aerial cover by species, bare ground, dead vegetation, standing water, sulfur bacteria, and algal mat (summing to 100% cover) at 0, 2, 7, 15, and 20 meters along each transect. Above-ground biomass was collected by standing two meters from each 1 m2 quadrat, and with back turned tossing a 0.0625 m2 sub-quadrat over the shoulder and into the 1 m2 quadrat (Tiner 1999). All above-ground •Y plants within the 0.0625 m sub-quadrat were clipped flush with the ground surface, placed into individually labeled sample bags and transported back to the lab for processing. My objective was to characterize current above-ground plant production in each habitat. Therefore, dead plants unattached to living plants were removed from each sample and only live plants were included in biomass analysis. Live plant samples were rinsed to remove soils and salt, transferred to individually labeled paper bags, and placed 71 in a drying oven at 60°C for three consecutive days to achieve a constant dry weight. Biomass samples were then removed from the oven and weighed to 0.0 lg. Plant Species Richness and Diversity Species richness was determined by averaging the number of species in each 1 m2 plot along paired transects by replicate and habitat type at each marsh as described above, Tiner (1987) was used to determine species identities. Plant diversity was assessed using the Shannon-Wiener Index of diversity, which provided a measure of uncertainty relative to the likelihood of encountering a given species in a random sample (Brower et al. 1990). Pielou's J index was used to determine the degree to which abundances were equally distributed within each sample. These two calculations take into account species richness, abundance, and evenness of distribution as follows: (a) Shannon-Wiener Index: 0) i=i where H' is the Shannon index of diversity and pt is the proportional abundance of species / derived using the following formula: (2) where nt is the abundance of species i, and N is the total abundance of all species. Shannon Index values are expressed in units of uncertainty, so to make interpretation easier the Shannon units were converted to units that represent the number of equally abundant species required in a sample to produce the same level of uncertainty represented by the Shannon H1 value (Hill 1973): (3) 72 (b) Pielou's J: (4) where J is the Pielou index of evenness, and H'max represents the condition that all species are equally abundant within a community and is calculated as follows: (5) where S is species richness for the community. Statistical Analysis Data were compiled into Microsoft Excel® spreadsheets and statistically analyzed using JMP® statistical software (SAS Institute, Inc. 2010). The alpha level was set at 0.05 to control for Type I Error during statistical analyses of main and interactive effects. Each dependent metric (i.e., percent cover, biomass, richness, or diversity) was analyzed separately using a two-way ANOVA model blocked by marsh, with marsh feature (pool vs. channel), and habitat type (natural vs. created), as independent variables, along with the interaction term marsh feature x habitat type. Data are presented as means ± Standard Error (SE). All analyses included Tukey-Kramer HSD, where an experiment-wise error rate of 0.05 was used to compare post-hoc means. Data were examined to ensure they satisfied the assumptions of the general linear model (normal distributions, no extreme outliers, and evenness of variance). Transformations of plant percent cover data did not satisfy the assumptions of normality, so the Wilcoxon\Kruskal-Wallis non-parametric test was used to compare means for this metric. 73 Discriminant function analysis was used to predict habitat type (creek, ditch, natural pool and ditch-plug) based on marsh plant characteristics. The discriminant function model used habitat as the dependent variable, with plant species, algal mat, sulfur bacteria, and unvegetated area as independent variables. Algal mat (Cyanobacteria in association with green sulfur bacteria of the family Chlorobiaceae) and purple sulfur bacteria (Chromatiaceae) (Figure 35) are indicative of stressful edaphic conditions (i.e., anoxic soils and high levels of hydrogen sulfide; Kaplan et al. 1979) that can limit nutrient availability, increase the presence of plant toxins, and adversely affect plant growth (Mitsch and Gosselink 2000). Therefore, algal mat and purple sulfur bacteria were included in this analysis. The number of discriminant functions used in a model is based on either (a) g-1, where g is the number of categories in the grouping variable (dependent variables), or (b) the number of discriminating (independent) variables, p, whichever is less. This study used g-1, which resulted in three discriminant functions. Results Plant characteristics varied among the three study marshes, with percent cover and biomass highest at Parker River Marsh (the southern-most marsh) and lowest at Sprague River Marsh (the northern-most marsh), showing a distinct latitudinal trend. Species richness and diversity were highest at Sprague River Marsh and lowest at Parker River Marsh. However, plant patterns were consistent for each habitat type among all three study marshes (Appendix B) and, therefore, data were combined for subsequent analyses. 74 Figure 35. Ditch-plug habitat at Sprague River Marsh in Phippsburg, Maine showing plant die-back areas occupied by algal mat and purple sulfur bacteria (indicative of anoxic soils with high sulfide concentrations; Kaplan et al. 1979). Plant Abundance Comparisons of vascular emergent plant cover by habitat feature and type followed what might be expected from the differences in hydrology, specifically higher water levels in ditch-plug habitat (Chapter 2), where plant cover was lowest (Figure 36). In addition, plant cover was higher in habitat adjacent to creeks (85 ± 1 %) than in ditch habitat (81 ± 1 %) (Table 8). 75 100 Created Ditch Natural Creek Ditch Plug Natural Pool Figure 36. Plant percent cover by habitat (values are means from paired 20 m transects of four habitat replicates from three marshes, pooled together ± SE; n=12) Table 8. Plant percent cover Wilcoxon/Kruskal-Wallis non-parametric comparisons (n=12) Habitat Comparison p-Value Creek Ditch-plug <0.0001 Natural pool Ditch-plug <0.0001 Creek Ditch 0.0097 Natural pool Ditch 0.3721 Natural pool Creek 0.6859 Ditch-plug Ditch <0.0001 ""Bold values indicate significant differences between habitats Plant biomass was different among habitats and the interaction of marsh feature and habitat type was significant (Table 9). The Tukey-Kramer post hoc test showed plant biomass in creek habitat (51 ± 1 g/0.0625 m2) was significantly higher than all other 76 habitats, and significantly lower in ditch-plug habitat (7 ± 1 g/0.0625 m2; Figure 37), as compared to all other habitats. Created Ditch Natural Creek Natural Pool Figure 37. Plant biomass by habitat (values are means from paired 20 m transects of four habitat replicates from three marshes, pooled together ± SE; n=12) Table 9. Two-way ANOVA results plant characteristics (DF=5, 6; n=12) Water Feature Type Interaction Metric F p-Value* F p-Value* F p-Value" Biomass 108.4388 <0.0001 128.9254 <0.0001 13.1730 0.0008 Richness 14.4663 0.0005 9.7869 0.0032 7.0251 0.0114 Shannon H' 12.4753 0.0010 6.5157 0.0145 8.2294 0.0065 Shannon Exp H' 10.5524 0.0024 4.3532 0.0437 8.0403 0.0071 Pielou's J 6.9987 0.0115 11.1175 0.0018 7.7863 0.0080 * Bold values indicate significant differences between mbitats; Water Feature = channel or pool; Type = natural or created 77 Plant Species Richness and Diversity Species richness was different among habitats and the interaction of marsh feature and habitat type was significant (Table 9). The Tukey-Kramer post hoc test showed species richness in ditch-plug habitat (1.58 ± 0.33 species/m2) was significantly lower than all other habitats (Figure 38). Species richness was similar in creek (4.15 ± 0.33 species/m2), ditch (3.98 ± 0.42 species/m2), and natural pool (3.75 ± 0.28 species/m2) habitats. f The two indices used to assess plant species diversity (H' and ExpH') showed differences among natural and created habitats, as well as for channel and pool habitats (Figure 38). Differences were also observed for Pielou's J, representing the evenness of species distribution in each habitat. However, the interaction of habitat type and water feature was significant (Table 9), and the Tukey-Kramer post hoc test led to the same pattern for all three indices: habitat adjacent to ditch-plugs possessed significantly lower plant diversity and evenness than the other habitats. The Shannon-Wiener H' index of diversity shows the level of uncertainty within a community and was significantly lower in ditch-plug habitat (0.34 ± 0.08). The Shannon-Wiener ExpH' index of diversity shows the number of equally abundant species required in a sample that equates to the level of community uncertainty represented by H', and the Tukey-Kramer post hoc test showed ExpH' in ditch-plug habitat (1.58 ± 0.16) was significantly lower than all other habitats (Figure 38). Similarly, Pielou's J was significantly lower in ditch-plug habitat (0.32 ± 0.04) compared with the other habitats. Plant diversity and evenness were similar among ditch, creek, and natural pool habitats. 78 • Created Ditch • Natural Creek H Ditch Plug • Natural Pool « 3.00 & 2.50 E 2.00 £Ln Richness Shannon H* exp(H') Pielou's J Figure 38. Species richness and diversity indices by habitat (values are means from paired 20 m transects of four habitat replicates from three marshes, pooled together ± SE; n=12) Plant Community Composition Sampling results for emergent plant composition by habitat are presented in Figure 39. Although the proportional contributions to overall plant cover by S. patens, J. gerardii, and D. spicata were similar among ditch, creek, and natural pool habitats, distinct differences were observed, primarily in percent cover by forb species and S. alterniflora. Specifically, ditch habitat was characterized by a higher proportion of forb species and lower S. alterniflora cover. Creek habitat was characterized by a higher proportion of tall-form S. alterniflora and lower cover by short-form S. alterniflora and forb species. Natural pool habitat was characterized by a higher proportion of short-form and medium-height S. alterniflora and S. europaea. Emergent plant cover in ditch-plug 79 habitat was distinctly different from all other habitats. Plant cover averaged only 30% and was characterized by a higher proportion of S. europaea and tall-form S. alterniflora. High marsh species contributed only 10% cover in ditch-plug habitat (Figure 39). • Sueda sp. • Salicomia europaea WAtriplex patula • Plantago maritima • Limonium nashii D Triglochin maritimum • Glaux maritima O 40 DPuccinellia maritima • Short S. alterniflora <20cm • S. alterniflora 20-50cm • Tall S. alterniflora >50cm • Disticlis spicata • Juncus gerardii • Spartina patens Created Ditch Ditch Plug Natural Creek Natural Pool Figure 39. Emergent plant species composition and cover by habitat (values are means from paired 20 m transects of four habitat replicates from three marshes pooled together; n=12) Plant composition with distance from water features as well as elevation gradients for each habitat is presented in Figure 40. The proportional contributions by D. spicata, S. patens, and J. gerardii were similar in ditch, creek, and natural pool habitats. Ditch 80 HSueda sp. • Salicornia europaea O 100 • Atriplex patula • Plantago maritima • Llmonium nashii •Triglochin maritimum BGIaux maritima • Puccinellia maritima i 40 •Short S. altemiflora <20cm J •S. altemiflora 20-50cm •Tall S. altemiflora >50cm •Disticiis spicata • Juncus gerardii Distance from water feature (m) • Spartina patens (a) Ditch habitat 120 HSueda sp. • Salicornia europaea •Atriplex patula •Plantago maritima •Limonium nashii •Triglochin maritimum • Glaux maritima •Puccinellia maritima •Short S. altemiflora <20cm • S. altemiflora 20-50cm • Tall S. altemiflora >50cm • Disticiis spicata •Juncus gerardii Distance from water feature (m) •Spartina patens (b) Ditch-plug habitat Figure 40. Emergent plant composition and cover by habitat and elevation (red line) for (a) ditch, (b) ditch-plug, (c) creek, and (d) natural pool (values are means from paired 20 m transects of four replicates from three marshes pooled together; n=12) 81 BSueda sp. •Salicomia europaea •Atriplex patula •Plantago maritima •Limonium nashii •Triglochin maritimum •Glaux maritima • Puccinellia maritima •Short S. alterniflora <20cm • S. alterniflora 20-50cm •Tall S. alterniflora >50cm •Disticiis spicata • Juncus gerardii Distance from water feature (m) •Spartina patens (c) Creek habitat BSueda sp. •Salicomia europaea •Atriplex patula • Plantago maritima •Limonium nashii •Triglochin maritimum •Glaux maritima • Puccinellia maritima •Short S. alterniflora <20cm • S. alterniflora 20-50cm •Tall S. alterniflora >50cm •Disticiis spicata •Juncus gerardii •Spartina patens Distance from water feature (m) (d) Natural pool habitat Figure 40 (continued). Emergent plant composition and cover by habitat and elevation (red line) for (a) ditch, (b) ditch-plug, (c) creek, and (d) natural pool (values are means from paired 20 m transects of four replicates from three marshes pooled together; n=12) 82 habitat had a higher proportion of Puccinellia maritima at 0-2 m and higher relative contributions by forb species between 7-20 m. The sloping banks of creek habitat promoted the growth of S. alterniflora and excluded J. gerardii at 0-2 m; and S. patens was displaced by J. gerardii between 2-20 m in creek habitat (Figure 40c). Natural pool habitat had the highest marsh surface elevation, which changed little over the 20 m transect. Natural pool habitat was characterized by moderate proportions of forb species, and higher relative contribution of short-form and medium-height S. alterniflora at all distances along transects. Marsh surface elevation beyond 2 m, and plant cover at all distances were lowest in ditch-plug habitat. Spartina alterniflora and S. europaea had the highest proportional contributions from 0-7 m, with high marsh species contributing more with greater distance (Figure 40b). Discriminant function analysis (DFA) is used to predict categorical dependent variables based on continuous independent variables, and in doing so discriminate between the dependent variables and determine group membership for the predictor variables. My rationale for using DFA was to determine if plant cover by species (the continuous independent variables) could predict habitat type. Interpretation of DFA involves several statistics. Wilks' Lambda is used to test the significance of the discriminant function model, with values closer to zero indicating significant differences between group means. The percentage of group membership correctly predicted is provided along with a summary matrix identifying where incorrect predictions occurred. The first canonical function provides the highest discrimination between groups, followed by the second and third functions. Eigenvalues indicate the discriminating power along with the percentage of variance explained by each canonical 83 function. The canonical correlation describes the relationship between dependent groups and the independent variables that make up the discriminant function. Canonical functions with higher correlation values indicate stronger correlations that account for more of the variability in the model. The standardized scoring coefficients indicate which variables contribute the most to group discrimination. The discriminant function plot is provided in Figure 41. ^Jri vegetated .Sulfur Bacteria' Canonical Figure 41. Discriminant function analysis comparing plant cover by habitat type (values are means from paired 20 m transects of four habitat replicates from three marshes pooled together; n=12; + = mean; circles = 95% confidence of means; CD = Ditch (Red); DP = Ditch-Plug (Green), NC = Creek (Blue), and NP= Natural Pool (Orange) 84 Discriminant function analysis identified significant differences in group membership for plant composition and cover type associations among the four habitats (Wilks' Lambda = 0.014; p<0.0001), with 92% of the group memberships correctly classified into each of the four habitats (Table 10). The first discriminant function axis (canonical 1) captured 72% of the variation in species composition and cover type explained by the analysis (Table 11). High positive values represented species composition associated with edaphic stress and unvegetated area (some of which was inhabited by purple sulfur bacteria) commonly found in transects of ditch-plugs, while species with lower values in ditch-plug habitat had low relative abundance. The highest positive correlations for canonical 1 were unvegetated areas, algal mat, and sulfur bacteria, and most negative correlations were S. patens and short-from S. altemiflora (Table 12). Table 10. Discriminant function actual by predicted habitat classification matrix Actual Predicted Habitat Habitat Ditch Ditch-plug Creek Natural Pool Ditch 11 0 0 1 Ditch-plug 0 12 0 0 Creek 1 0 11 0 Natural Pool 2 0 0 10 The second discriminant function (canonical 2) explained 24% of the remaining variation, and positive values represented plants associated with creek habitat (Table 11). Canonical 2 was positively correlated with tall-form S. altemiflora, J. gerardii, and A. patula. Canonical 2 was most negatively correlated with Plantago maritima, which 85 was relatively more abundant in natural pool habitat (Figure 41 and Table 12). The third Canonical Function only represented 4% of the variation in cover data and was not interpreted. Table 11. Results of discriminant function analysis for plant community composition based on percent cover values measured in all habitats. Eigenvalues and percentages are based on the three canonical functions. Canonical correlations are a measure of the association between groups formed by independent variables and a particular discriminant function. Canonical Cumulative Canonical Function Eigenvalue Percent Percent Correlation 1 9.62 72.19 72.19 0.95 2 3.13 23.46 95.65 0.87 3 0.58 4.35 100.00 0.61 Discussion Self-Maintenance in Marshes: Water Features and Vegetation Turner et al. (2000) working in marshes of Louisiana, Texas, and Rhode Island, and Morris et al. (2002) in South Carolina, characterized the importance of plants in the salt marsh self-maintenance process. Plant growth promotes self-maintenance of salt marsh habitat through below-ground biomass production that contributes organic matter to marsh soils, and above-ground growth that traps sediment conveyed in tidal flows. In combination, these processes determine the rate of salt marsh accretion and elevation relative to sea level (Morris et al. 2002). Thus, variations in plant community 86 Table 12. Discriminant function standardized scoring coefficients used to indicate the relative importance of each independent variable for each discriminant function. Bold values identify independent variables that contribute the most to a given canonical function. Function UV SB AM AP PS GM JG LN PM PU SE TSA SSA SP SU TM Canonl 0.971 0.384 0.730 -0.176 -0.266 0.053 -0.109 -0.186 -0.305 -0.375 0.325 0.321 -0.637 -0.917 -0.388 0.243 Canon2 -0.059 -0.133 0.023 0.459 -0.059 0.306 0.481 -0.145 -0.775 0.240 -0.263 0.914 -0.261 0.073 0.232 0.199 Canon3 0.898 0.458 0.499 0.252 0.166 0.066 0.591 -0.345 -0.669 1.065 0.145 -0.246 -0.171 1.099 0.825 0.349 Independent variables: UV=unvegetated, SB=sulfur bacteria, AM=algal mat, AP=A triplex patula, DS=Distichlis spicata, GM=Glaux maritima, JG=Limonium nashii, PM=Plantago maritima, P\J=Puccinellia maritima, SE-Salicornia europaea, TSA=tall-form Spartina altemiflora, SSA=short-form Spartina altemiflora, SP=Spartinapatens, SU= Sueda spp., TM=Triglochin maritimum. characteristics among habitats can influence the rate of marsh development and ability of the habitat to keep pace with sea level rise (Turner et al. 2000). Hydrology, elevation, plant, and soil characteristics work in combination to establish a hydrologic regime, which is the underlying mechanism that influences biochemical processes and plant communities throughout the marsh. In natural habitats, hydrologic features are determined in response to physical factors and disturbance (Williams and Orr 2002; Wilson et al. 2009). In contrast, the creation of salt marsh water features involves human modifications to elevation and hydrology that result in altered hydrologic regimes to which the plants must adapt. Community Attributes (Biomass. Cover, and Composition) Sediment and nutrient transport along with intermittent flushing and aeration of soils are key functions provided by natural and created channel habitats that promote plant growth and soil accretion, making channel habitats critical components of the marsh self-maintenance process. However, differences in plant characteristics for habitat adjacent to creeks and ditches indicate differences in hydrologic regimes and contributions to marsh building process by the two habitats. For instance, less extensive pore-water drainage in habitat surrounding ditches (as shown in Chapter 2) contributed to significantly lower biomass (Figure 37) and cover (Figure 36) than that observed in creek habitat. The lower biomass and cover in ditch habitat, therefore, reduces contributions to the marsh accretion process from soil trapping and organic matter inputs. The effects that different hydrologic regimes have on species composition were apparent in the two channel habitats. Discriminant function analysis (Figure 41) identified tall-form S. alterniflora as characteristic of creek habitat, representative of 88 more frequent tidal flushing and better drained sloping creek banks (Howes et al. 1986; Burdick 1989; Howes and Goehringer 1994). The displacement of S. patens by J.gerardii at distances between 15-20 m in creek habitat was not observed in ditch habitat, while the higher relative proportion of forb species at greater distances in ditch habitat was not found in creek habitat (Figure 40). Limits to flooding stress tolerance appear to prevent down-slope migration of forb species onto the frequently flooded creek banks, further limiting forb presence in creek habitat. Ellison et al. (1986) found that the greatest amount of belowground biomass occurred within the first 10 cm below the soil surface in marshes of Maine and Georgia, and Bertness et al. (1985) working in North Carolina observed a dense root mat at 0-5cm below the soil surface. Aerobic soils increase nutrient availability and plant uptake, and facilitate gas exchange between plant roots and leaves involved in photosynthesis. Thus, soil drainage within the first 5-10 cm is important for plant production and community composition. Pore-water within the first 10 cm (rooting zone) in creek habitat was drained 41% of the time compared to 30% of the time in ditch habitat (Chapter 2). The greater aeration time within the rooting zone of soils adjacent to creeks contributes to the differences observed in species composition and the higher plant production (as measured by biomass and cover) in creek habitat compared with ditch habitat. These results further exemplify the effects of soil drainage on plant communities. The big unanswered question is how will these differences in plants and hydrology affect the ability of ditches to keep pace with sea level rise? Created and natural pool features differ dramatically in their plant characteristics and contributions to self-maintenance. For instance, habitat surrounding natural pools 89 had significantly higher biomass, cover, and diversity than ditch-plug habitat (Figures 36-38). In fact, plant cover and biomass in natural pool habitat were more closely related to channel habitats than ditch-plugs, with the discriminant function analysis (Figure 41) calling out short-form S. alterniflora as an indicator of less extensive pore-water drainage in natural pool habitat compared to the channel habitats. The association of short-from S. alterniflora with less well drained soils in marshes of Massachusetts was confirmed by Howes et al. (1981, 1986). In contrast, discriminant function analysis characterized ditch-plug habitat as supporting greater proportions of algal mat, sulfur bacteria, and unvegetated area, all indicators of permanently saturated conditions that counteract the marsh building process (Koch and Mendelssohn 1989). The benefits of frequent flushing and soil aeration are lacking in pool habitats where tidal exchange is typically limited. In habitat adjacent to ditch-plugs marsh elevations are lower (showing subsidence) with a gradient that slopes towards the pool, promoting surface water retention and reduced soil conditions (Chapter 2). In contrast, habitat surrounding natural pools are 15 cm or greater in elevation than in ditch-plug habitat, do not slope toward the pool beyond 2 meters, and are flooded less frequently. As a result, marsh surfaces adjacent to natural pools are flooded only 46% of the time, with the top 5cm of the soil drained 20% of the time (Chapter 2). Thus, pore-water drainage in natural pool habitat promotes the growth of short-form S. alterniflora and other plants that contribute to marsh self-maintenance. In contrast, marsh surfaces adjacent to ditch-plugs are flooded 91% of the time, with the top 5cm of the soil drained only 2% of the time. Such conditions lead to plant dieback in habitat surrounding ditch- plugs, counteracting the self-maintenance process (Koch and Mendelssohn 1989). 90 Hvdrologic Disturbance Mediates Species Distributions and Community Richness Tidal flow in salt marsh habitat creates disturbance from waterlogging and anaerobic conditions that occur on varying time scales and mediate plant assemblages. Huston (1979) suggested that since fewer species have the ability to adapt to more stressful conditions, higher stress habitats tend to have lower productivity and species richness. Thus, increased stress associated with habitat disturbance from altered hydrologic flows can alter species richness, productivity, and habitat heterogeneity. Species richness of vascular plants in ditch-plug habitat was significantly lower than all other habitats (Figure 38), and can be attributed to extended periods of surface water retention with areas of high edaphic stress that are occupied by only the most stress tolerant species. Although species richness was similar in ditch, creek, and pool habitats, differences in the extent of pore-water drainage appeared to be enough to alter species composition among the three habitats (Figures 39 and 40). Forb panne species in New England include Triglochin maritimum, Plantago maritima, Atriplex patula, Glaux maritima, Limonium nashii, and Sueda spp. (Ewanchuk and Bertness 2004), and had higher relative abundance in ditch habitat than other habitats (Figure 39). Ewanchuk and Bertness (2004) working in marshes of southern Maine manipulated marsh elevations and soil drainage, and concluded that forb pannes develop in waterlogged areas of the marsh. Saturated soils increase plant stress and prevent competitive exclusion of forb species by the less tolerant high marsh turf building species. The authors reasoned that ditching decreased water logging in the high marsh and diminished forb panne habitat (Miller and Egler 1950; Redfield 1965; Nixon 1982; Bertness 1992). My results support observations by Ewanchuk and Bertness (2004) with 91 respect to the association of forb panne species with waterlogged soils. However, my results differ with regard to the effects of ditching on forb habitat. I found the presence of forb species was higher in ditch habitat compared with others, but soils adjacent to ditches were not drained more than natural areas. I found soils in ditch habitat were drained more extensively than habitat near pools where the presence of forb species was lower, suggesting higher stress associated with extended periods of pore-water saturation is less tolerable for forb species. However, ditch soils retained pore-water within the rooting zone for longer periods of time than creek habitat (Chapter 2), indicating intermediate levels of stress tolerance for forb species. Soil drainage intermediate between creek and pool may be needed to maintain a higher presence of the less competitive forb species. Thus, intermediate levels of edaphic stress tolerable for forb species also exclude less stress tolerant turf species, but are not so extreme that they favor more stress tolerant short-form S. alterniflora and S. europaea. Huston (1979) described press disturbance as a persistent physical change that does not return to pre-event conditions. The delayed pore-water drainage in the rooting zone of ditch habitat has resulted in press disturbance, creating edaphic conditions that alter competitive interactions and enable the greater presence of forb species in habitat adjacent to ditches. The importance of press disturbance in maintaining mixed species assemblages was also evident in the relative proportion of D. spicata in all habitats I observed (Figure 40), and is an indication of the dynamic nature of salt marsh habitats. Bertness and Ellison (1987) noted the relationship between D. spicata and recent disturbance from floating plant debris (wrack) that settles on the marsh platform, smothering more 92 competitive species and creating bare patch opportunities for the more stress tolerant D. spicata to colonize. In addition, areas of the high marsh dominated by D. spicata often have soils that are not as well drained as adjacent areas dominated by S. patens or J. gerardii (personal observations). The abundance of D. spicata in all habitat types I observed highlights the importance of disturbance-mediated coexistence, whether burial or hydrologic, in maintaining habitat heterogeneity and community complexity in all habitats studied. Ditch-plugs in my study ranged in age from 5 to 10 years with high relative abundance of S. europaea common in all replicates (Figure 40). Havill et al. (1985) working in British salt marshes observed a positive correlation between S. europaea and soil sulfide concentrations, and Bertness and Shumway (1992) found that S. europaea initially colonizes disturbed bare patches. Therefore, I attribute the high abundance of S. europaea in ditch-plug habitat to the plant dieback and highly reduced soil conditions associated with ditch-plugs (Chapter 2). Reed and Cahoon (1992) working in Louisiana observed extensive plant dieback that they attributed to reduced soil conditions and extended periods of surface water impoundment at lower elevations of the marsh interior. Mendelssohn et al. (1981) observed similar plant response to marsh surface flooding and attributed the loss of plants to soil phytotoxins and altered plant metabolic processes associated with highly reduced soil conditions. Such conditions represent prolonged habitat disturbance. The annual regeneration of S. europaea that I observed at 5+ years after ditch-plug construction demonstrates that edaphic conditions in ditch-plug habitat do not facilitate transition back to high marsh habitat after flooding, stress continues year 93 after year, and ultimately the processes that maintain high marsh have been altered in this habitat. Hummock-hollow topography occurs in marshes throughout New England where pore-water drainage is severely restricted (personal observation). Stribling et al. (2007) described hummock-hollow microtopography in Maryland salt marshes as clumped plants separated by lower elevations of bare soil. Anaerobic conditions, high sulfide concentrations, and reduced plant belowground production result in low areas of bare soil as plants die off and belowground biomass is reduced (Stribling et al. 2006). In addition, Nyman et al. (1995) observed S. alterniflora hummock development in overly flooded marshes of Louisiana. Hummock-hollow plant growth is common in ditch-plug habitat, and large areas of plant dieback with hummock-hollow topography were characteristic of all ditch-plug replicates in my study (Figure 42). I attribute the formation of this hummock-hollow pattern to prolonged surface water retention and anaerobic soils (Chapter 2) that result from ditch-plugging. A microtopographic pattern is produced where hummocks are typically vegetated with S. alterniflora, S. patens, or D. spicata. The base of these hummocks is often covered by an algal mat containing nitrogen-fixing cyanobacteria. Stribling et al. (2006) studying hummock-hollow topography in Maryland marshes found that hummocks supporting S. alterniflora had soil redox and ammonium levels similar to those found in well drained streamside banks, and concluded that the formation of hummocks and more benign soil conditions are controlled by the plants. However, it is unclear at this time if the hummock-hollow pattern in ditch-plug habitat is maintained by the presence of S. alterniflora and perhaps nitrogen fixing cyanobacteria as suggested by 94 Figure 42. Hummock-hollow plant patterns in ditch-plug habitat at (a) Sprague River Marsh and (b) Chauncey Creek Marsh resulting from prolonged water retention and anaerobic soils (Photos by Rob Vincent). 95 Nyman et al. (1995) and Stribling et al. (2006), or if the presence of these species in habitat adjacent to ditch-plugs is an intermediate step in the progression towards open water habitat. Facilitation and Habitat Heterogeneity Patches of species assemblages, such as forbs or short-form S. alterniflora, provide habitat heterogeneity within the larger matrix dominated by high marsh turf species, and this heterogeneity typically results in higher species richness, as seen in ditch, creek, and natural pool habitats (Figure 38). In addition, it appears that facilitative interactions contribute to the maintenance of heterogeneity within these habitats. Facilitative interactions promote the growth and reproduction of one species while having no immediate negative effects on the facilitating species. Bertness and Hacker (1994) and Konisky and Burdick (2004) observed increased growth of transplants paired with J. gerardii, noting the facilitative qualities of J. gerardii. I observed high percent cover for J. gerardii in creek, ditch, and natural pool habitats (Figure 40), which may serve a facilitative function in maintaining high percent cover of the less competitive high marsh turf species (i.e., S. patens and D. spicata). In addition, Bertness (1991b) found that competitively excluded forb species colonize more stressful areas of the marsh and facilitate the expansion of more competitive turf species through amelioration of edaphic stress. Therefore, the presence of weaker competitors (i.e., forb species) I observed in localized stressful areas of natural pool, creek, and ditch habitats can facilitate the expansion of high marsh species over time, thus creating a give-and-take process between stress tolerant and competitive species. This facilitative process complements the effects of periodic hydrologic and other physical disturbances to high marsh habitat and 96 maintains a high percentage of turf building species in these three habitats, while allowing for the occurrence of stress tolerant forb species and short-form S. alterniflora. My results show that the dynamic processes of competitive exclusion, disturbance, and facilitative interactions are similar in both created and natural habitats and maintain nearly the same levels of percent cover for creek, ditch, and natural pool habitats (Figure 36), while providing heterogeneity within each habitat. Percent cover in ditch- plug habitat, however, was distinctly different from all other habitats, indicating the greater extent of disturbance and high edaphic stress had a strong effect on both competitive and facilitative processes in this habitat. Conclusion My conceptual model presented in Chapter 1 highlighted the differences that I expected to encounter among the four habitats. My hypothesis that created hydrologic features alter physical parameters that influence plant community structure, resulting in distinct differences among natural and created habitats was supported. The influence of hydrology, elevation, and soil processes on plant characteristics in different habitats was evident in the plant results of my study, and can be used to infer the likelihood of self- maintenance in each habitat. Soils in habitat adjacent to creeks and natural pools were better drained with higher redox potential than their created counterparts (Chapter 2), promoting plant growth. While creek and ditch marsh surface elevations were similar, creeks had sloping banks that were not present in ditch habitat, and the extent and timing of pore-water drainage was slower in ditch habitat, resulting in differences in plant community composition and biomass. Although plants adjacent to ditches and creeks were similar in terms of high marsh turf building species, the differences in 97 geomorphology and pore-water drainage resulted in environmental gradients that facilitated higher relative abundance of tall-form S. alterniflora along creek banks, and a greater presence of forb panne species in ditch habitat. Significantly greater biomass and percent cover in creek habitat promotes sediment trapping and organic matter contributions in support of marsh accretion. The question is whether or not the plant attributes that contribute to the current rate of marsh accretion in ditch habitat are suitable to sustain the marsh self-maintenance process, or will the rate of sea level rise outpace marsh accretion in ditch habitat? This question is of particular importance for marshes with more extensive grid ditching (Figure 5), in which it appears subsidence and pore- water retention may have had greater stress on plants, which can interfere with marsh accretion processes. Plant attributes (biomass, cover, richness, and diversity) in habitat adjacent to natural pools were similar to those observed for channel habitats; however, species composition in natural pool habitat reflected edaphic conditions that favor more stress tolerant species. Less extensive pore-water drainage in natural pool habitat resulted in higher relative abundance for short-form S. alterniflora, compared with the two channel habitats. However, the proportional contribution of high marsh turf species was comparable to channel habitats, and higher marsh surface elevations demonstrate that natural pool habitat is similar to channel habitats in its ability to facilitate marsh accretion and support self-maintenance. Habitat adjacent to natural pools had significantly higher marsh surface elevation and more extensive pore-water drainage compared with ditch-plug habitat, resulting in significant differences in plant community characteristics between the two habitats. All 98 plant attributes were significantly higher in habitat adjacent to natural pools compared with ditch-plugs. These differences in community structure are in response to the altered hydrology in ditch-plug habitat and its effects on biogeochemical processes (Chapter 2). The resultant structural dissimilarities between natural pools and ditch-plugs have led to dissimilarities in function between the two habitats, and will be discussed further in Chapter 4 (trophic interactions). Maintaining emergent plant habitat is an essential component of the salt marsh self-maintenance mechanism. Interrupting tidal flow and increasing surface water impoundment severely impacts plants and thereby diminishes the ability of the marsh to keep pace with sea level rise, and increases the likelihood and rate of salt marsh habitat loss over time (Mendelssohn and McKee 1988; DeLaune et al. 1994; Cahoon and Reed 1995; Turner et al. 2000). These effects were clearly evident in the soil processes (Chapter 2) and plant communities of ditch-plug habitat. 99 CHAPTER IV FISH PRODUCTIVITY AND TROPHIC TRANSFER IN CREATED AND NATURALLY-OCCURRING SALT MARSH HABITAT Abstract High marsh pools are natural features in New England salt marshes that provide important subtidal refuge within the intertidal high marsh habitat for the dominant resident fish, Fundulus heteroclitus. Fundulus heteroclitus is the most commonly occurring fish in New England marshes and considered an important component in the trophic transfer pathway for its omnivorous diet and role as a prey species providing connectivity to adjacent near-shore and terrestrial habitats. Pool creation is a common component of habitat restoration and enhancement projects throughout the region, and one type is ditch-plugging. My study combined field experiments measuring fish growth and benthic invertebrate communities with 813C and 815N stable isotope analyses to test the hypothesis that ditch-plugs replicate the structure and function of naturally-occurring salt marsh pool habitat in regard to trophic structure and productivity. Marked fish placed in enclosures were measured for length and weight weekly in natural and created pools. Invertebrates were sieved and sorted from soil cores to characterize invertebrate community structure, and stable isotopes were used to determine diet preferences and trophic pathways associated with each pool type. Fish growth trended higher in natural pool habitat (11% for biomass; 25% for length), as did invertebrate species richness, density, biomass, and caloric value. Stable isotope mixing models identified distinct resource utilization and trophic structure for natural and created pools. My study 100 increases our understanding of the ecology of salt marsh pools, and will aid resource managers with future salt marsh restoration planning. Combined, the significant results indicate that pools created using ditch-plugs do not replicate structure and function of natural pools in support of F. heteroclitus and invertebrate communities. Introduction In the northeast, salt marsh ecosystems have long been subjected to manipulation, beginning with salt hay farming by European colonists (Fogg 1983; Williams et al. 2001; Bromberg and Bertness 2005; Buchsbaum et al. 2009). More recently, salt marsh restoration and enhancement projects in New England have pursued management of target species (fish; James-Pirri et al. 2005), target guilds (water fowl; Erwin et al. 1991; Erwin et al. 1994), public health issues (mosquitoes; Wolfe 1996;), or re-establishing tidal flow (Burdick et al. 1997; Dionne et al. 1999). In each case, whether it's habitat manipulation for specific management objectives or attempts to restore lost structure and function, the impetus for habitat change has been hydrologic alteration that either increased or decreased ground and surface water in various ways through ditching and draining, ditch-plugging and surface water impoundment, pool creation, pool connection using shallow surface channels, or alleviating tidal restrictions (Daiber 1986; Fogg 1983; Wolfe 1996; Burdick et al. 1997; Boumans et al. 2002; Silliman et al. 2009). With the exception of tidal flow restoration, these types of projects focus more on target management, and are less likely to address the ecological integrity of the community as a whole (Palmer and Filoso 2009; Silliman et al. 2009). Projects in New England are typically conducted without prior knowledge of secondary effects on marsh ecology. A 101 more conservative and broader reaching approach uses controlled experimentation and analyses on smaller scales or in step-wise progressions to assess the potential for unintended secondary effects prior to large scale habitat manipulations for management or restoration purposes (Palmer and Filoso 2009; Silliman et al. 2009). Further, the long- term effects of climate change and sea level rise on salt marsh habitat are not well understood, yet they are important to consider when designing salt marsh management and restoration projects. Preliminary experimentation can provide a better understanding of ecosystem responses to habitat manipulations, and thus allow for the design of marsh restoration projects that more effectively incorporate future alterations to hydrologic regimes resulting from rising sea levels. My goal, therefore, was to address these issues through the use of controlled experiments to characterize natural pool habitat and identify some of the secondary effects on ecosystem structure and trophic pathways that occur in response to altered hydrologic regimes of created pool habitat. High marsh pools in New England salt marshes occur at higher elevations of the marsh and typically flood only during spring tide events. These pools provide important low tide refuge and production habitat for resident mummichog fish, F. heteroclitus, the dominant fish species found in these marsh systems (Smith and Able 1994; Raposa 2003; Able et al. 2005; MacKenzie and Dionne 2008). The food web of high marsh pools is connected to vegetated habitats by fish that access the flooded marsh surface to feed during spring tides (Weisberg and Lotrich 1982; Dionne et al. 1999; Able et al. 2006; MacKenzie and Dionne 2008). Due to its omnivorous foraging habits throughout the marsh, F. heteroclitus is considered to be an integral link for trophic transfer of primary producer energy in support of avian, mammalian, and near-shore fish populations (Teal 102 and Teal 1969; Kneib and Wagner 1994; Kneib 1997). Therefore, F. heteroclitus is a key component needed to establish and maintain ecological connectivity among salt marsh and adjacent habitats. For these reasons, pool creation is a common component of present-day salt marsh habitat restoration, management, and enhancement projects throughout the region. Natural pools in northern New England salt marshes typically develop over time as secondary features in response to disturbance from algal mats, wrack deposits, soil deposition, ice scouring (Harshberger 1916; Chapman 1960; Argow and Fitzgerald 2006; Dionne 1989; Wilson et al. 2009), and hay staddles associated with erosion (personal observations; Figure 6). Complex environmental gradients involving elevation, hydrology, soil characteristics, salinity, redox potential, sulfides, and plant community patterns are thought to influence natural pool development (Yapp et al. 1917; Chapman 1960; Redfield 1972). Natural pools are often round (Figure 7), and range widely in size (1 to >20 m diameter), with characteristic vertical or under-cut banks and uniform bathymetry with depths approximately 25 to 40 cm at the center (Smith and Able 1994; Hunter et al. 2009). Ditches were constructed for mosquito control by the Civilian Conservation Corps more than 70 years ago, and are typically narrow (£ 1 m) with varying depth (£ 1 m) and vertical banks (Figure 4). Ditch-plugging in New England has occurred since the 1990s, and attempts to control mosquito populations by increasing surface water habitat for larvivorous fish, as well as provide foraging habitat for wading birds and water fowl (Meredith et al. 1985; Taylor 1998; Neidowski 2000). Ditch-plugs typically start out small with a gradually tapering and undulating bathymetry that descends to a center sump 103 2: 1 m deep (Figure 9). However, little is known of the ecological effects of ditches or ditch-plugs. I conducted a controlled experiment that investigated the trophic relationships and resource use of fish (F. heteroclitus) and invertebrate prey species in five ditch-plugs and five naturally-occurring pools in a New England salt marsh. The study combined field experiments measuring fish productivity and invertebrate benthic communities, along with SI3C and 8,5N stable isotope analyses to test the null hypothesis that ditch-plugs are no different in structure and function than naturally-occurring salt marsh pool habitat. Over the years, the use of stable isotope analysis in estuarine ecological studies has become an important tool, focusing on such subjects as food web structure (Lajtha and Michener 1994; Fry 2006; Michener and Lajtha 2007), consumer resource use (Peterson et al. 1985; Deegan and Garritt 1997; McMahon et al. 2005; Fry et al. 2008), system response to nutrient loading and anthropogenic impacts (Griffin and Valiela 2001; Tobias et al. 2003; Bannon and Roman 2008), and habitat response to restoration (Currin et al. 2003; Wozniak et al. 2006). However, the details of trophic transfer of carbon, food web structure, and resource utilization by resident and transient organisms in estuarine systems are not yet well defined (Douglass et al. 2011). This is especially so for high marsh pool habitat in New England. In order to effectively evaluate the ecological impacts from habitat alterations and inform resource managers, it is important to understand the fate of carbon as it passes through the system to F. heteroclitus and beyond. Multiple stable isotope analysis is one useful tool for investigating the trophic interactions (Pasquaud et al. 2010), resource utilization (Weinstein et al. 2000; Haas et al. 2009), and transfer of organic carbon 104 through the estuarine system (Peterson et al. 1986). The ability of stable isotopes to quantify short-term and long-term environmental and biological relationships is invaluable when assessing ecological food webs and time series (Lajtha and Michener 1994; Fry 2006; Michener and Lajtha 2007). Stable isotope analyses in my study provide an understanding of habitat function in terms of resource utilization and trophic transfer, and a framework for comparing the conditions and ecological processes observed in ditch-plug and natural pool habitats. Materials and Methods Study Area and Experimental Design Moody Marsh is a 138-hectare back barrier salt marsh located in Wells, Maine (43° 16'00.52"N; 70° 35'35.69" W). The study portion of Moody Marsh was an 18-hectare parcel at the center of the marsh, which contained both natural pools and ditch-plugs (Figure 43a). Sampling took place during July and August, 2006. The marsh is bound to the south by a paved parking lot and residential development; to the east by the Ogunquit River protected from the Gulf of Maine by a barrier beach; to the north by the Ogunquit River, which drains a 53.8 km watershed; and to the west by steep upland slopes and residential development. Ditches dug in the 1930s were arranged perpendicular to the Ogunquit River, extending westward through the marsh from the river's edge toward the upland border. A large natural tidal creek splits the study marsh into two sections: (1) Ditch-plug Habitat: high marsh dominated by a mixture of Spartina patens, Spartina alterniflora, and Salicornia europaea in the southern portion of the 105 study site where ditches dug in the 1930s were plugged to form pools in 1999; and (2) Natural Pool Habitat: high marsh dominated by a mixture of S. patens and short form S. alterniflora in the northern portion of the study site. Five study pools were identified in each of the two habitat sections (i.e., natural pool and ditch-plug). To ensure a broad distribution across the marsh, an aerial photo of the marsh was divided into three equal-sized bands extending south to north the entire length of the study site (Figure 43b). A grid was placed over the aerial photo, and each grid intersection was numbered. A random numbers table was used to choose grid intersections in each of the three marsh sections, and pools closest to the chosen grid intersections were used as sample stations for the study. The circumference of each pool was numbered 1 through 12 like the face of a clock, with 6 at the southern end and 12 at the northern end of each pool. A random numbers table was used to choose one number representing the location of fish enclosure and exclosure cages for each pool sampled. The number 6 was randomly chosen, so the cages were placed against the pool wall at the southern end of each pool (Figure 44). Cages were placed in the same location in each pool to control for environmental effects (i.e., sunlight, shading, temperature, wind). Fish Enclosure and Exclosure Cages All fish enclosure and exclosure cages were uniformly constructed 0.25 m2 wooden box frames covered with 0.8 mm nylon mesh. The cages were inserted into pool bottom soils and held in place with plastic tent stakes. Fish enclosure cage bottoms were left open to allow fish to forage in the pool soils. Exclosure cages were similarly placed 50 cm west of the enclosure cage in each pool to prevent fish from foraging on the 106 Natural Pool Habitat NP1D Ditch Plug Habitat Figure 43. Moody Marsh study site: (a) location in Wells, Maine showing barrier beach and upland development, and (b) sample stations (green lines are strata delineators; red lines are study area boundaries; NP=Natural Pool; DP=Ditch-plug); Source: Google Earth 2010. 107 benthic community. An area used to control for any cage effects on invertebrate communities was created in each pool by affixing a piece of 0.8 mm nylon mesh between the two pool cages (Figure 44 inset). Fish exclosures were also placed on the marsh surface adjacent to the five ditch-plug and five natural pools. The marsh surface exclosures were located 3 m south of the edge of each pool and in line with the pool enclosure/exclosure placements (Figures 44and 45). Figure 44. Fish enclosures and exclosures in natural pool habitat at Moody Marsh in Wells, Maine 2006. 108 North v$>\ Ditch Plug Pool = Fish Exclosure/Enclosura Sediment Core Natural Pool Figure 45. Invertebrate soil core collection areas, as presented for a single natural pool and ditch-plug replicate. Fish Sampling Minnow traps were used to collect four male F. heteroclitus fish from each sample pool. Fish densities were based on a fish enclosure study conducted at Moody Marsh in 2003 (Mackenzie and Dionne 2008), which sampled the same pools used in my study. The starting total length (TL) and weight of each fish used in the study was approximately 45 mm and 1.10 g, respectively. Fish were marked with subcutaneous non-toxic acrylic paint (Lotrich and Meredith 1974) and placed inside the enclosure within each pool. Enclosed fish were measured for length (mm) and weight (g) weekly 109 for five weeks. At the end of five weeks, fish from each enclosure were sedated with carbonated water, euthanized by decapitation, placed into labeled sample bags, kept on ice in coolers, and transported back to the lab for stable isotope analysis. Fish Growth Instantaneous growth rates and fish condition were used to determine the change in fish biomass, length, and combined growth over the study period. Instantaneous growth is the natural logarithm ratio for comparing the final weight or length of a fish to its initial weight or length over a specific period of time. Instantaneous biomass growth was determined using the exponential model (Chapman 1978): Gins, = (In wt2 - In wtj) /t2 - tj (1) where Gins, is the instantaneous growth rate, wt2 is the wet weight of a fish at time 2 (t2), and wtj is the wet weight of the same fish at time 1 (tj). Instantaneous length growth was calculated using equation 1 and substituting l2 and // for wt2 and wth respectively. Fish condition is a ratio of weight to cubed length. It is considered to be a measure of fish health, production, and by inference habitat value (Murphy and Willis 1996). Fish with higher condition ratios are considered more robust and likely to persist, suggesting that the habitat provides the resources necessary to sustain healthy populations (Murphy and Willis 1996, Ratz and Lloret 2003). Fish condition was calculated using Fulton's Condition Factor: K = (W/L3)x 100,000 (2) where K is the fish condition, W is the fish weight (g), and L3 is the fish length (mm) cubed. 110 Benthic Invertebrate Sampling Soil cores were collected at the beginning and end of the study period by inserting a 7.5 cm diameter PVC corer into the soils and removing the upper 4 cm for analysis. Using a grid pattern representing the interior of the cages and control areas, and a random numbers table, core sampling points were located. There were five replicates per habitat type (i.e., natural pool and ditch-plug). One pool core and one marsh surface core were collected from each replicate in each habitat type prior to the start of the study, and these cores represented natural control conditions. Cores representing experimental conditions were collected at the end of the experiment. Experimental cores consisted of one soil core collected from each fish enclosure, exclosure, and control area within and adjacent to each pool, for a total of three pool experimental cores and two marsh surface experimental cores for each replicate in each habitat type (Figure 45). Soil cores were placed into individually labeled sample bags, kept on ice in coolers, transported back to the lab, and frozen prior to invertebrate identification. Soil cores were rinsed through sieves with distilled water and material separated in two stages using 2.0 mm and 0.5 mm sieves. Once rinsed, the core material was spread evenly throughout the 0.5 mm sieve. A plastic quadrat placed inside the sieve separated the core into four equally-sized numbered sub-samples. Two of the sub-samples were randomly selected for invertebrate identification, and the remaining material was re-frozen for future reference. Sub-samples were placed into lab trays where invertebrates were sorted, placed into Petri dishes with distilled water, and identified using dissecting scopes. Once identified and counted, invertebrates within each sample were grouped by taxon, placed in tins, and dried in an oven at 60°C for ill 24 hours to achieve constant weight. Dried invertebrates were weighed by taxon to determine biomass and abundance, then individually stored in desiccators for later stable isotope analysis. Mean invertebrate density was determined for each type of sample: habitat (natural or created), location (pool or marsh surface), and treatment (enclosure, exclosure, control). Caloric values for each taxon were obtained from the literature (Cummins and Wuycheck 1971). Total caloric content (Kcal/m2) was determined for each habitat, location, and treatment by multiplying invertebrate biomass (g dry weight/m ) by Kcal/g dry weight for each taxon and replicate. Habitat Physical Parameters Pool water physical characteristics were collected at the time of cage installations, and then weekly through the end of the study period. A handheld YSI 85 multimeter (YSI° Corporation 2006) was used to sample dissolved oxygen (DO), salinity, and temperature by suspending the sensor in the middle of the water column 20 cm to the right of each fish enclosure, and 20 cm out from the edge of the pool. Pool water depth was measured at the same location. Stable Isotope Analysis Vegetation. Plant samples were collected from each habitat and replicate as they occurred adjacent to cage placements by clipping above-ground portions of the plants flush with the surface of soils. Plant samples were not combined and remained as separate distinct samples throughout the process. They were rinsed thoroughly in distilled water to remove salt, foreign matter, and epiphytes, placed in labeled paper bags, and dried at 60°C for 48 hours to achieve constant weight. Samples were ground to a homogenous powder using a common household coffee grinder. The coffee grinder was 112 disassembled and thoroughly cleaned with distilled water between samples. Powdered samples weighing between 1.50 mg and 2.00 mg were placed into tin cups for stable isotope processing. Particulate Organic Matter (POM). Sterile 3.75-liter plastic containers were used to collect estuarine water during the incoming tide approximately one-hour prior to high tide. Water was collected from the primary tidal channel of the Ogunquit River that flows north and south between the salt marsh and barrier dune systems. POM samples were not combined and remained as separate distinct samples throughout the process. Glass microfiber filters (GF/F 47 mm) were pre-combusted, and after cooling each filter was individually wrapped in labeled foil and stored inside a desiccator prior to POM processing (Wetzel and Likens 1991). One 3.75 L container of estuarine water was vacuum drawn through five GF/F filters. Filters were placed into individual tin trays and dried in an oven at 60°C for 48 hours. A razor blade and tweezers were used to separate POM material from the GF/F filters. Between 1.05 mg to 2.00 mg of POM material from each filter was placed into individual tin cups for isotope processing. Benthic Microalgae. Benthic microalgae samples (BMA) were collected adjacent to marsh surface enclosures for each replicate per habitat according to Tibias et al. (2003). Nytex screen (210 ^m) was placed on the marsh surface at low tide to capture benthic diatoms migrating up from the underlying soils. Screens were washed with deionized water, sieved through 20 }im mesh, and filtered onto pre-combusted GF/F 47 mm filters. Filters were placed into individual tin trays and dried in an oven at 60°C for 48 hours. A razor blade and tweezers were used to separate BMA material from the 113 GF/F filters. Between 1.05 mg to 2.00 mg of BMA material from each filter was placed into individual tin cups for isotope processing. Invertebrates. For each of the five replicates, dried invertebrate samples were combined by taxon, habitat type (ditch-plug or natural), and location (pool or marsh surface) to achieve adequate mass for isotope analysis. The samples were then ground to a homogenous powder using mortar and pestle. Powdered samples weighing between 0.80mg and 1.20mg were placed into tin cups for isotope processing. Fish. Fish samples were not combined and remained as separate distinct samples for each fish throughout the process. Fish were removed from the freezer, thawed and filleted. White muscle tissue from each fish was rinsed thoroughly with distilled water, placed in a labeled tin, and dried in an oven at 60°C for 48 hours to achieve constant weight. Dried samples were ground to a homogenous powder using mortar and pestle. Powdered samples weighing between 0.80 mg and 1.20 mg were placed into tin cups and arranged in plastic sample trays. Stable Isotope Processing. Stable isotope values of carbon (813C%o) and nitrogen (815N%o) were determined for primary producers (vegetation), primary consumers (herbivorous invertebrates), secondary consumers (predatory invertebrates), and tertiary consumers (fish). Stable isotope ratios were determined at the University of New Hampshire Stable Isotope Laboratory by combusting samples in a Costech ECS4010 Elemental Analyzer coupled to a Delta Plus XP mass spectrometer (Thermo Finnigan). Stable isotope ratios are reported using delta notation where delta (in per mil units (%o)) equals: AX = [(^sample/#standard) - 1 ] X 1000 (3) 114 Where X is 13C or I5N and R is 13C/12C or 15N/14N, respectively. Vienna Pee Dee Belemnite (VPDB) was used as the standard for 13C. Delta 13C values are reported on the VPDB scale using IAEA-CH6 (-10.449 per mil) and IAEA-CH7 (-32.151 per mil). Atmospheric N2 (air) was used as the standard for 15N. Delta 15N values are reported on the VPDB scale using IAEA-N1 (0.4 per mil) and IAEA-N2 (20.3 per mil). Repeated analyses of laboratory standards varied less than 0.15 per mil for both 15N and 13C. Latex rubber gloves were worn at all times during the preparation and handling of all stable isotope samples, equipment, and materials, and chemwipe sheets were used to clean all instruments and trays between each sample preparation. No stains or preservatives were used during this project, and all samples were stored in desiccators prior to stable isotope analysis. The isotope mixing model MixSIR (Semmens and Moore 2008) was used to assess source contributions to consumer diets. Due to the large number of diet source variables (i.e., plants and invertebrate taxa), consumer diets were assessed in two steps: (1) plant source contributions to primary consumer diets; and (2) invertebrate source contributions to secondary and tertiary consumer diets. Three consumer/source models were analyzed: (a) invertebrates as primary consumers with plants as primary producer diet sources; (b) predatory invertebrates as secondary consumers with herbivorous invertebrates as diet sources; and (c) fish as tertiary consumers with invertebrates as primary and secondary consumer diet sources. The relative percent of combined plants versus combined invertebrates as two major groups contributing to the diets of secondary and tertiary consumer diets was not determined due to the large number of sources and 115 group variability. Rather, results are presented as the species-specific percent contribution to primary consumer diets for plants relative to all plant species sampled for that habitat. Similarly, the species-specific percent contribution for invertebrate sources in secondary and tertiary consumers diets represent the contribution of that invertebrate species relative to all invertebrates for that habitat. Consumer 615N and 813C isotope values were adjusted downward 3.4%o and 0.5%o, respectively, to account for trophic fractionation prior to running the models (DeNiro and Epstein 1978; DeNiro and Epstein 1981; Fry and Sherr 1984; Minagawa and Wada 1984; Vander Zanden and Rasmussen 2001; Phillips and Gregg 2001). Statistical Analysis. Data were compiled into Microsoft Excel® spreadsheets and statistically analyzed using JMP® statistical software (SAS Institute, Inc. 2010). Data were examined to ensure they satisfied the assumptions of the general linear model (normal distributions, no extreme outliers, evenness of variance). Where necessary, data transformations were performed prior to analysis to meet these assumptions. The alpha level was set at 0.05 for main effects and interactions to control for Type I Error during statistical analyses. All parametric analyses included Tukey-Kramer HSD post-hoc means comparisons with Type I Error contained to 0.05. Pool physical parameters did not require transformation. Each dependent variable (temperature, salinity, DO, and water depth) was analyzed separately using a one-way ANOVA with habitat as the independent variable. Invertebrate data, including biomass, richness and density, were log (x + 1) transformed prior to analysis. Three-way analysis of variance (ANOVA) was used to determine main effects for the independent variables: habitat, cage treatment, and 116 location (marsh surface or within pool), along with interaction terms (habitat x treatment; and habitat x location). Natural controls and procedural controls were compared using 1-way ANOVA to test for artifacts of treatments. If no significant differences were found (i.e., p>0.05), it was assumed that no treatment artifacts occurred on invertebrate metrics from enclosure and exclosure cages. Fish growth data were log (x + 1) transformed prior to analysis. Dependent variables included biomass, length, and condition. One-way ANOVA with repeated measures on replicates within pools was used to compare fish biomass and condition by habitat. Transformations of fish growth for length did not satisfy the assumptions of normality, so the Wilcoxon rank sums test was used as an ANOVA analog to compare means for this metric. Within group variation for fish biomass and length growth was not significantly different in either ditch-plug or natural pool habitats. Therefore, fish biomass and length data were averaged by replicate in each habitat, and instantaneous growth rates were based on the means for the four fish in each enclosure. One-way ANOVA with repeated measures on replicates within pools was used to compare the dependent variables fish 813C and 815N by the independent variable, habitat. The relationships between fish 513C and 815N with pool water salinity (averaged by replicate) and habitat were analyzed using linear regressions. Results Pool Physical Parameters Mean salinity and water depth of the pools were significantly higher in natural pool habitat compared with ditch-plugs (Figure 8; 1-way ANOVA: p=0.00330, F=7.60, 117 DF= 1, 3, n=5; and p=0.0247, F=12.30; DF= 1, 3, n=5, respectively). Mean water temperature and DO were not significantly different, but trended higher in ditch-plugs (Figure 46). Dissolved oxygen levels in pool soils of the two habitats were substantially different, as evidenced by soil coloration depicted in Figure 43. Soil cores collected from locations near the edges of ditch-plugs were black (1G 2.5/N on Munsell color chart), indicating severely reduced anaerobic soil conditions (Figure 47a). In contrast, soil cores collected from natural pools were grayish-brown (5YR 6/2 on Munsell color chart), indicating more oxygen availability in the soils (Figure 47b). Fish Growth Instantaneous growth for length was significantly higher in natural pool habitat compared with ditch-plug habitat (Wilcoxon rank sums, p=0.0301, Z=2.17, DF=1, 3, n=5). Instantaneous biomass growth in natural pools and ditch-plugs were not significantly different, but trended higher in natural pools. There was no significant difference in fish condition (weight relative to length cubed) between the two habitats (Figure 48). 118 • Ditch Plug • Natural Pool Salinity (ppt) Temperature (°C) Dissolved Oxygen Mean Water Depth (mg/L) (cm; pool edge) Pool Physical Characteristics Figure 46. Pool physical characteristics collected weekly from the five pools in each habitat type (values are means +SE, n=5). Comparisons by habitat for each physical variable were conducted separately using one-way ANOVA, different letters indicate significant differences. 119 Figure 47. Pool soil cores: (a) ditch-plug soil core, black indicating reduced anoxic conditions; (b) light natural pool soil core indicating less reduced conditions. 120 0.045 • Ditch Plug 0.040 • Natural Pool 0.035 0.030 0.025 0.020 0.015 0.010 0.005 0.000 Biomass Length (a) 1.5 X A •o® c 1.0 •o c o * Ifl c 0.5 o 0.0 Ditch Plug Natural Pool (b) Figure 48. (a) Mean fish instantaneous growth rate for biomass and length, and (b) fish condition by habitat; comparisons by habitat for fish biomass growth and condition were conducted separately using one-way ANOVA with repeated measures on replicates within pools; the Wilcoxon rank sums test was used to compare means for fish length; comparisons with different letters are significantly different (Values are means ± standard error; all data were log(x+l) transformed prior to analysis; n=5) 121 Benthic Invertebrates Benthic invertebrate biomass was assessed as a measure of fish prey resource abundance and trophic transfer potential. No significant difference was observed among natural and procedural controls for either habitat, indicating cage presence had no effect on invertebrate communities. The biomass of invertebrates was significantly greater in natural pool habitat than in ditch-plug habitat (Table 13; Figure 49a). There were no significant differences in biomass for location (pool or marsh surface), or treatment (control, enclosure, or exclosure), and no significant interactions (Figure 50). Species richness and density for each habitat and replicate were assessed as a measure of invertebrate community structure. Mean species richness was significantly higher in natural pool habitat than in ditch-plug habitat (Figure 49b), but there were no other significant differences or effects (Table 13; Figure 51). Mean invertebrate density was significantly higher in marsh surface locations than within pools due to the high number of Acarina found in cores from marsh surfaces adjacent in both habitats (Table 13; Figure 52). There were no significant differences in density for habitat (Figure 49c), or treatment, and no significant interactions (Table 13). Benthic invertebrates were grouped by major taxa to determine relative contributions to overall abundance by habitat and location (Table 14). Acarina comprised the greatest proportion of individuals in both habitats and locations (Figure 53). In ditch-plug habitat, only one species of Acarina was found, which comprised 75% and 83% of invertebrates found in pool and marsh surface soils, respectively; followed by Diptera (13% pool soils and 15% marsh surface soils). Further, Diptera contributed more to the overall community composition of marsh 122 A A 1 <0 B B 'I0'" 0 <0I 2 04 " ra 1 ]_ £• 3 ^ • • z 3 • Ditch Plug Natural Pool Ditch Plug Natural Pool (a) (b) 10,000 ^ 9,000 n 8,000 S 7,000 | 6,000 > 5,000 *5 4,000 Ditch Plug Natural Pool Ditch Plug Natural Pool (c) Figure 49. Invertebrate community characteristics by habitat for (a) biomass, (b) richness, (c) density, and (d) caloric value (values presented as mean ± standard error; log transformed prior to analysis; n=5) surface areas in ditch-plug habitat than in natural pool habitat. Ditch-plug and marsh surface soils each supported three additional taxa that contributed to the remaining proportions. Of these species, Amphipods had a higher presence in ditch-plugs than in natural pools, and Polychaeta were found in samples collected from ditch-plugs, but were absent from natural pools. 123 Table 13. Three-way ANOVA results for benthic invertebrate characteristics (DF= 7,42; n=5) Habitat Location Treatment Interaction 1 Interaction 2 P- P- P- P- P- Metric F Value* F Value* F Value* F Value* F Value* Biomass 7.54 0.0095 0.81 0.3779 2.13 0.1092 1.05 0.3100 0.41 0.6659 Richness 5.65 0.0231 2.43 0.1281 1.16 0.3239 0.04 0.8344 0.19 0.8247 Density 0.05 0.8218 11.76 0.0016 0.01 0.9869 0.01 0.9419 0.68 0.5124 Caloric 6.88 0.0114 0.84 0.3480 0.91 0.4151 0.01 0.9957 0.17 0.8401 Value •"Bold values are significant Habitat = Pool vs. ditch-plug Location = pool vs. marsh surface Treatment = control, enclosure, exclosure Interaction 1 = habitat x location Interaction 2 = habitat x treatment • Ditch Plug • Natural Pool Marsh Marsh Marsh Pool Pool Pool Pool Surface Surface Surface Control Procedural Enclosure Exclosure Control Procedural Exclosure Control Control Figure 50. Benthic invertebrate biomass by habitat and treatment (values presented as mean ± standard error; log transformed prior to analysis; n=5) 124 Ditch Plug •Natural Pool Marsh Marsh Marsh Pool Pool Pool Pool Surface Surface Surface Control Procedural Enclosure Exclosure Control Procedural Exclosure Control Control Figure 51. Benthic invertebrate species richness by habitat and treatment (values presented as mean ± standard error; log transformed prior to analysis; n=5) 20,000 • Ditch Plug 18,000 • Natural 16,000 Pool CM 14,000 co 112,000 110,000 c 8,000 ° 6,000 Figure 52. Benthic invertebrate density (per m ) by habitat and treatment (values presented as mean ± standard error; log transformed prior to analysis; n=5) 125 Table 14. Proportional distribution of benthic invertebrates by major taxonomic groups Major Taxa Ditch-plug Marsh Natural Natural Pool Grouping Ditch-plug Surface Pool Marsh Surface Acarina 75% 83% 59% 94% Coleoptera 0% 1% 15% 2% Diptera 13% 15% 10% 1% Arthropoda 0% 0% 10% 0% Amphipoda 8% 0% 0% 1% Hemiptera 2% 0% 6% 1% Polychaeta 2% 0% 0% 0% Gastropoda 0% 0.5% 0% 1% Arachnida 0% 0% 0% 1% Foraminifera 0% 0.5% 0% 0% Total 100% 100% 100% 100% In natural pool habitat, three species of Acarina comprised 59% and 94% of invertebrates found in pool and marsh surface soils, respectively; followed by Coleoptera spp. (15% pool soils and 2% marsh surface soils). Three and five additional taxa were found in natural pool and adjacent marsh surface soils, respectively. Of these species, Coleoptera and Arthropoda were found in high proportions in natural pools, but were absent from ditch-plugs. Marsh surfaces adjacent to natural pools supported greater numbers of taxa than marsh surfaces of ditch-plug habitat. In addition, when Acarina were removed from analysis, the proportional distribution to overall abundance by individual species was much more evenly distributed in natural pool habitat (Figure 53 a) than was observed for ditch-plug habitat, indicating a greater variety of prey availability in natural pool habitat (Figure 53b). Invertebrate relative abundance by habitat and 126 450 • Ditch Rug 400 • Natural Fbol 350 | 300 B3 | 250 i |200 Z 150 100 50 0 ^ ^ ^ J> J? J. Cf $ / S 35 • Ditch Rug • Natural Pool & £/ / & (b) Figure 53. Abundance by major taxa for benthic invertebrates found in cores collected in each habitat (a) with Acarina included, and (b) with Acarina excluded. 127 treatment are presented in Appendix A. Invertebrate caloric values (Kcal/m2) were analyzed as a proxy for fish habitat value and trophic transfer potential. No significant difference was observed among natural and procedural controls for either habitat. Similar to the biomass results, invertebrate caloric value was significantly higher in natural pool habitat than in ditch- plug habitat (Figure 49d). There were no significant differences in caloric value for location (pool or marsh surface), treatment (control, enclosure, or exclosure) (Figure 54), and no significant interactions (Table 13). 6.0 • Ditch Plug • Natural Pool 5.0 4.0 2.0 1.0 0.0 Marsh Marsh Marsh Pool Pool Pool Pool Surface Surface Surface Control Procedural Enclosure Exclosure Control Procedural Exclosure Control Control Figure 54. Benthic invertebrate caloric values (Kcal/m ) by habitat and treatment (values presented as mean ± standard error; log transformed prior to analysis; n=5) 128 Stable Isotopes Stable isotope values of carbon and nitrogen were determined for primary producers, herbivorous invertebrates, predatory invertebrates, and F. heteroclitus. The mean stable isotope value for each species sampled is presented in Table 15 by habitat (ditch-plug or natural pool) and location (within pool or marsh surface). Enclosed ditch- plug fish were significantly higher in 815N and lower in S13C than enclosed fish in natural pool habitat (ANOVA: p<0.0164, F=23.91, DF=1, 3, n=5; and p<0.0221, F=19.14, DF=1, 3, n=5, respectively) (Figure 55). Plants in ditch-plug habitat were higher in 815N, compared with plants in natural pool habitat (Figure 56). The simple regression showed a significant decrease in fish 615N with increased pool salinity averaged by pool replicate, and a significant increase in fish 813C with increased salinity averaged by pool replicate (Table 16). These results indicate a relationship between pool salinity levels and fish isotope values in the two habitats (i.e., lower salinity and higher 815N in ditch-plugs; and higher salinity and higher 813C in natural pools). Multiple regressions were conducted to determine if habitat influenced these trends (Table 16). Results showed that habitat had no effect on the positive relationship between 813C and salinity, and confounding effects of habitat were observed when assessing the relationship between 815N and salinity. 129 Table 15. Mean (± SE) S13C and 815N stable isotope values by trophic level, habitat, and treatment. Values without standard error represent isotope values from a single sample. Ditch-plus Natural Pool Pool Marsh Surface Pool Marsh Surface Species 8,5N%. 8UC%. 8ISN%. 813C%> 8,SN%. 8UC%« 8,5N%. 8,3C%o Primary Producers Quercus rubra 1.2 (0.7) -26.6 (0.6) 1.3 (0.3) -28.0 (0.2) Salicornia europaea 2.4 (0.3) -28.1 (0.1) 2.6 (0.5) -29.0 (0.1) Fucus vessiculosus 4.6 (0.2) -18.5 (0.6) 5.8 (0.2) -16.9(1.0) Ruppia maritima 1.8 (0.4) -11.9(0.3) 0.5 (0.5) -12.0 (0.3) Spartina alterniflora 5.8 (0.4) -13.8 (0.1) 2.5 (0.2) -13.1 (0.1) Spartina patens 3.1 (0.1) -13.9 (0.1) 2.1 (0.2) -14.2(0.1) Benthic Microalgae 3.4 (0.1) -17.2 (0.1) 3.6 (0.1) -18.6 (0.1) Particulate Organic Matter (POM) * 4.9 (0.1) -21.9 (0.1) 4.9 (0.1) -21.9 (0.1) Primary Consumers Acarina spp. 5.6 (0.2) -19.4 (0.1) 4.2(0.1) -19.2 (0.1) 3.2(0.1) -20.1 (0.0) Amphipoda spp. 2.8 (0.3) -18.3 (0.1) -0.3 -20.2 Diptera spp. 3.8 -17.5 1.6 (0.3) -15.9 (0.2) 1.0 -17.1 5.9 (0.1) -18.8(0.1) Canace spp. 3.8 -16.4 Ephydra spp. 1.1 (0.4) -18.1 (1.6) Odontomyia spp. 2.6 (0.1) -17.2 (0.8) Tabanus spp. 4.0 (0.2) -15.3(0.1) Hydrobia spp. 1.9 (0.3) -3.8 (0.2) Melampus bidentatus 2.3(0.1) -11.1 (0.7) Secondary Consumers Arachnida Araneae 6.2 -19.0 Coleoptera spp. 0.7 (0.2) -19.0(0.1) Coccinellini spp. 5.3 (0.1) -19.4(0.2) Geopinus spp. 1.1 -21.9 Neanthes spp. 4.4 (0.3) -12.8 (0.2) Tertiary Consumer Fundulus heteroclitus 5.3 (0.2) -15.6 (0.4) 4.0 (0.2) -13.9 (0.1) * POM collected from main tidal channel of the Ogunquit River. Table 16. Regression results for fish 815N and 813C versus pool water salinity and habitat (isotope and salinity values were averaged by pool). Regression r2/R2 Linear Fit p Value F DF n 815N 0.51 y=: 9.744 - 0.089 (salinity) 0.0197 8.45 1,8 10 y=: 8.454 - 0.511 (habitat) 815N 0.86 0.0010 21.4 2,7 5 - 0.036 (salinity) 813C 0.72 y= -17.456+0.136 (salinity) 0.0019 20.6 1,8 10 y = -16.109. + 0.534 (habitat) 8,3C 0.95 0.0001 62.6 2,7 5 + 0.081(salinity) -42- Ditch-plug Fish -46- i *4 i * C3 Plants t*hif Natural Pool Fish C4 Spartina spp. * Ruppia POM 15N • BMA § I »POM I l C4 (Spartina spp.) * Invertebrates 'ftverteHrates .•"[sagk-ss"!""T BMA 1 a DP Fish I C3 (Salicornia Ruppia INP Fish europaea) maritima -25 -20 -15 1 -10 13C Figure 55. 813C and 815N stable isotopes values (±SE) for fish collected from ditch-plug (DP) and natural pool (NP) habitats, along with various sources that contribute to fish diets (BMA = benthic microalgae; POM = particulate organic matter). The two fish outliers (one per habitat) were not representative of the study populations in each of the habitats and, therefore, were removed from data prior to analysis. 131 Spartina alterniflora Particulate ADP-SA Oranic Matter 1- (POM) A DP-SP • DP-RM A DP-SE .DP BMA Salicornia Benthic Microalgae iftT- INP-SA europaea I*0 Spartina INP-SP I I patens m NP-RM t H • NP-SE 1 Ruppia iNP BMA maritima • -30 -25 -20 -15 -10 513C Figure 56. 813C and 815N stable isotopes values (±SE) for plants collected from ditch- plug (DP, triangles) and natural pool (NP, squares) habitats. SA = Spartina alterniflora-, SP = Spartina patens-, RM = Ruppia maritima-, SE = Salicornia europaea; BMA = benthic microalgae; POM = particulate organic matter. With the exception of SE, 8I5N is consistently higher for plants in ditch-plug habitat, suggesting greater uptake of human- derived nitrogen from upland runoff. Note, in some cases the error bars are smaller than the data point symbols. Natural Pool Food Web Mixing model results identified four trophic levels in natural pool habitat at Moody Marsh: (1) plant primary producers (benthic microalgae (BMA), S. europaea, S. alterniflora, and S. patens)-, (2) invertebrate primary consumers (Acarina, Diptera, 132 Melampus bidentatus, and Amphipoda); (3) invertebrate secondary consumers (Arachnida and Coleoptera); and (4) fish tertiary consumer (F. heteroclitus) (Figure 57). Benthic microalgae contributed the most to the diets of primary consumers in natural pool habitat (range: 76% to 94%), with POM contributing 15% and 10% to the diets of Acarina. and Amphipoda, respectively. The three most abundant primary consumers in natural pool habitat (Acarina, Diptera, and Amphipoda) contributed most to the diets of secondary invertebrate consumers Arachnida and Coleoptera. Two primary consumers (Diptera and M. bidentatus) contributed the most to the diet of F. heteroclitus in natural pool habitat. Ditch-plug Food Web Mixing model results identified four trophic levels in ditch-plug habitat at Moody Marsh: (1) plant primary producers; (2) invertebrate primary consumers; (3) invertebrate secondary consumer (Polychaeta); and (4) fish tertiary consumer. Primary and secondary consumer trophic levels differed slightly from those in natural pools, as indicated (Figure 58). Benthic microalgae contributed the most to the diets of primary consumers in ditch-plug habitat (range: 48% to 84%). In addition, Acarina consumed S. europaea and S. patens. One primary consumer (Diptera) comprised most of the diet for the secondary consumer Polychaeta in ditch-plug habitat. Two primary consumers (Diptera and Amphipoda) contributed to the diet of the tertiary consumer F. heteroclitus in ditch- plug habitat, a portion of which was through the secondary consumer (Polychaeta). 133 Natural Pool Habitat Trophic Transfer Trophic Laval 4. Terttaiy Fundulus heteroclitus Consumer Arachnida Coleoptera 3. •@r Secondary spp. spp. Consumer Acarina Diptera Melampus Amphipoda spp. 2. spp. Primary spp. bidentatus Consumer Benthic Salicornea 1. Spartina Spartina POM Primary Microalgae europaea alterniflora patens Producer Figure 57. Trophic transfer food web based on consumers and diet sources found within natural pool habitat at Moody Marsh in Wells, Maine. Dark lines show sources that provide the highest contribution to the diet of an individual consumer, and gray lines show source contributions that are <10% of the consumer diet. Black circles show the percentage of total plant contribution to the diet of primary consumers; gray circles show the percentage of total invertebrate contribution to the diet of secondary invertebrate consumers; and white circles show the percentage of total invertebrate contribution to the diet of the tertiary consumer F. heteroclitus. Source contributions to consumer diets were determined using 5 13C, 8 15N, and the MixSIR mixing model (Semmens and Moore 2008). 134 Ditch-Plug Habitat Trophic Transfer Trophic Ltvtl 4. Tertiary Fundulus heteroclitus Consumer PolyShaeta 3. spp. Secondary Consumer Acarina Diptera Amphipoda spp. 2. spp. spp. Primary Consumer 1. Benthic Salicomea Spartina Spartina POM Primary Microaigae europaea altemiflora patens Producer Figure 58. Trophic transfer food web based on consumers and diet sources found within ditch-plug habitat at Moody Marsh in Wells, Maine. Dark lines show sources that provide the highest contribution to the diet of an individual consumer, and gray lines show source contributions that are <10% of the consumer diet. Black circles show the percentage of total plant contribution to the diet of primary consumers; gray circles show the percentage of total invertebrate contribution to the diet of secondary invertebrate consumers; and white circles show the percentage of total invertebrate contribution to the diet of the tertiary consumer F. heteroclitus. Source contributions to consumer diets were determined using 8 13C, 8 15N, and the MixSIR mixing model (Semmens and Moore 2008). 135 Discussion Hydrologic features that naturally occur in salt marshes include the tidal creek system that conveys the flood and ebb of tides, and isolated pools that hold water throughout the tidal cycle and flood when spring tides inundate the marsh. Since European settlement, ditches have been dug in marshes to deliver and drain away floodwaters to enhance salt haying and later to reduce habitat for salt marsh mosquitoes (Daiber 1986). Ditch-plugging is a habitat creation methodology that alters the hydrologic regime in localized areas to provide standing water habitat for larvivorous fish and wading birds (Lesser 1982; Meredith et al. 1985; Wolfe 1996; Taylor 1998). However, alteration of habitat in general results in environmental changes that may or may not reflect naturally occurring conditions (Silliman et al. 2009). Furthermore, changes in environmental conditions can lead to modifications in species composition and community structure, altering trophic pathways within the microhabitat and linkages for trophic transfer to adjacent ecosystems (Menge and Olson 1980; Rakocinski et al. 1992; Feyrer and Healey 2003; Speckman 2005; Crain and Bertness 2006). My study identified significant differences in habitat structure and function between ditch-plug and natural pool habitats, indicating an altered ecological role for the created habitat relative to the naturally-occurring reference habitat. Habitat Physical Parameters Ditch-plug pools frequently extend along their original ditch channels to the upland border of the marsh, which results from plugs at the seaward end of the channel preventing drainage. The lower salinity in ditch-plug habitat results from groundwater inputs and freshwater runoff from adjacent upland habitat channeled into ditch-plugs. 136 These ditches can also carry nutrients from upland runoff (Koch and Gobler 2009) to ditch-plugs. Soil nutrients contribute to microbial biomass and can increase respiration and soil decomposition rates (Wigand et al. 2009). During periods when marsh surface soils are drained, greater nutrient and oxygen availability increases microbial decomposition and respiration, leading to loss of organic matter, subsidence, and soil waterlogging (Wigand et al. 2009; DeLaune et al. 1994; Reed and Cahoon 1992). Higher temperatures in ditch-plugs can be an artifact of sampling in shallow water, and higher DO levels in ditch-plugs may be the result of samples collected during daylight hours when oxygen production from photosynthesis by aquatic plant results in higher DO levels in pool water (Smith and Able 2003). Ditch-plugs are typically excavated with pool edges that slope towards a central sump as deep, or deeper, than the original ditch, which can result in stratification of the water column. Hasanudin et al. (2005) investigating the effects of dredging on microbial communities in soils of Tokyo Bay found that extended anoxic conditions that developed in trench bottoms stratified the water column and altered the microbial community structure, compared with adjacent flat unaltered seabed conditions. I found that in ditch-plugs, elevated DO levels exist higher in the water column, while the deeper zones contain anoxic conditions creating the highly reduced soils that I observed. Such conditions contribute to differences in invertebrate communities that I observed between ditch-plug and natural pools. More study is needed to determine the extent of water column stratification in salt marsh pools. The black soil cores collected from ditch-plugs indicate severely reduced anaerobic soil associated with high sulfide concentrations that are toxic to plants and benthic organisms (Mitch and Gosselink 2000). The lighter colored soil cores collected 137 from natural pools indicated higher concentrations of oxidized iron and manganese within natural pool soils (Tiner 1999; Munsell 2000; Mitch and Gosselink 2000). Long and Seitz (2008) working in estuarine channels of Chesapeake Bay concluded that hypoxic conditions increase stress, forcing benthic organisms closer to the soil surface. The subsequent reduction in the volume of suitable habitat increases competition for space and resources, and exposes invertebrates to increased predation pressure (Long and Seitz 2008). Competition for space can also increase emigration by less competitive species (Lenihan and Micheli 2001). Increased competition, emigration, and predation likely contributed to the lower richness and density of invertebrates in ditch-plug habitat. To better understand dissolved oxygen conditions in pools of the two habitat types, pre dawn vertical profiles are recommended. Fish Growth Positive fish growth occurred in both habitats over the study period, yet growth in natural pool habitat was higher for fish biomass and significantly higher for fish length. Instantaneous growth rates in this study (0.018 to 0.035 mm/d; and 0.01 to 0.06 g wet wt/day) were consistent with previous studies by Haas et al. (2009) (0.009 to 0.014 mm/d), and MacKenzie and Dionne (2008) (0.01 to 0.05 g wet wt/d). Larger sample sizes and longer experimental time periods in future studies may show the biomass trend to be significant as well. There was no difference in the condition of fish in pools of either habitat, as indicated by Fulton's K index. However, fish in natural pool habitat were heavier and longer, and thus larger, indicating higher growth rates. This is consistent with the greater availability of food resources in natural pools. Higher growth rates, therefore, show that natural pools provide better habitat quality in terms of fish 138 productivity. Further, larger fish have the potential to consume larger, higher calorie prey as fish mouth gape increases along with body size (Vince et al. 1976; Cunha and Planas 1999; Waggy et al. 2007; Lopez et al. 2010). Previous studies have documented the ability of F. heteroclitus to tolerate stressful physical conditions (Kneib 1986; Smith and Able 2003); however, prolonged exposure to environmental stressors can alter immunocompetence and growth hormones in F. heteroclitus and other fish species (Fries 1986; Deane and Woo 2009). The lower growth rates for fish in ditch-plug habitat may have been, in part, due to lower abundance of prey, increased stress, or a combination of these factors (Smith and Able 2003; Deane and Woo 2009). Invertebrate Community Structure Invertebrate biomass, density, and richness were all higher in both pool and marsh surface soils of natural pool habitat, and similar to values observed by MacKenzie and Dionne (2008). As discussed previously, varying physical and biological conditions can contribute to the differences in invertebrate community structure among the two habitat types. Whitcraft and Levin (2007) working in California marshes noted that altered plant communities can influence resource availability, physical conditions, microhabitats, and herbivore community structure. Further, Bertness (1999) found that increased stress from higher temperatures and salinity associated with unvegetated high marsh habitat can influence invertebrate community composition. Thus, plant dieback associated with altered hydrology and physical parameters in ditch-plug habitat (Chapter 3) contributes to the significantly lower invertebrate density, richness, and biomass in soils adjacent to ditch-plugs. 139 Energy availability influences species composition and ecosystem function (Odum 1975; Thompson and Townsend 2004; Crustinger et al. 2006). The movement of carbon through a system is contingent in part upon the allocation of carbon for such purposes as structure (i.e., chitin) or energy (i.e., protein) at each trophic step (Sterner et al. 1997; Aber and Melillo 2001). Carbon transformations affect the quality and availability of carbon for trophic transfer. The invertebrate biomass available for fish predators that I observed in ditch-plug habitat was of lower caloric value than that available in natural pool habitat, and I infer these differences in energy availability as a biological response to changes in physical conditions associated with altered ditch-plug habitat. The ecological significance of the lower biomass and caloric value was evidenced by reduced energy available for trophic transfer, which contributed to lower fish growth observed in ditch-plug habitat compared with natural pool habitat. Acarina comprised the greatest proportion of overall invertebrate abundance in soil samples collected from both habitats, with one species of Acarina identified in ditch- plug habitat and three species found in samples collected from natural pool habitat. The abundance of Acarina in natural pool habitat was 53% higher than in ditch-plug habitat, reflecting more stressful conditions and lower resource availability in ditch-plug habitat. Soil moisture content and marsh surface water are two variables that affect the presence and abundance of Acarina in salt marsh habitat (Luxton 1967; Foster et al. 1979). Increased soil pore water content and prolonged marsh surface water impoundment in areas adjacent to ditch-plugs results in stressful (i.e., reduced) edaphic conditions that contribute to lower abundance and richness of Acarina in ditch-plug habitat. 140 Disturbance as a catalyst for competitive release has been suggested as a mechanism for influencing species assemblages (Connell 1978; Grime 1979; Huston 1979). In 1986, Soule noted that habitat complexity and resource availability influences community structure; and variations in environmental stressors play an important role in determining species distributions within salt marsh ecosystems (Bertness and Ellison 1987; Baldwin and Mendelssohn 1998; Mullan Crain et al. 2004; Konisky and Burdick 2004). The reduction in resource availability and habitat heterogeneity (i.e., plant dieback and altered invertebrate species composition), along with changes to environmental stress gradients (i.e., prolonged pore-water saturation and surface water inundation) are a few examples of disturbance resulting from ditch-plugging (Chapters 2 and 3). Habitat displacement and reduction in resource availability in response to habitat disturbance can also account for the absence of Arachnida and Coleoptera I observed in invertebrate samples collected from ditch-plug habitat. My results show that Arachnida and Coleoptera preyed upon Acarina in natural pool habitat. The absence of top-down predation on Acarina from these two predators in ditch-plug habitat may have resulted in competitive exclusion of the two additional species of Acarina that were found in natural pool habitat. Alternatively, the single species of Acarina found in ditch-plug habitat may simply have had a higher level of stress tolerance than the two additional species found in natural pool habitat, again resulting in the exclusion of these two species from ditch-plug habitat. Controlled experiments would be necessary to test these hypotheses. I also found a high abundance of Diptera in ditch-plug habitat. Graham and Stoffolano (1983) conducted laboratory experiments and found that 98% of female Diptera deposited eggs 141 on stems of S. alterniflora, while only 2% oviposited on S. patens or D. spicata; and Rader (1984) working in North Carolina marshes found Diptera associated with surface water and tall-form S. alterniflora, which are conditions that I found to be characteristic of ditch-plug habitat (Chapters 2 and 3). Fewer and different primary consumers and mid-level predators in ditch-plug habitat can be attributed to altered habitat conditions. The altered community structure in ditch-plug habitat, as evidenced by the significantly lower invertebrate biomass, richness, and density have an effect on the trophic structure and transfer of energy through the system, as was observed in the results from stable isotopes. However, further study is needed to understand the effects of altered invertebrate communities on all components of the ecosystem. Stable Isotopes Isotope mixing models use mass-balance equations to determine the proportional contribution of diet sources (prey isotope values) to the diet mixture (predator values), and can do so accurately when the number of diet sources equates to the number of isotopes tested +1 (Moore and Semmens 2008). For example, models that use 315N and 1 3 g C isotopes can accurately determine the contribution of three diet sources to the mixture. Models that use greater than n+1 diet sources must address the uncertainty associated the data, which is the challenge for ecological food web studies that commonly have greater than n+1 diet sources. In essence, the accuracy of a mixing model, as with any model, is limited by the values included in the model. Therefore, prior knowledge of predator diets is essential for building adequate food web mixing models. Diet sources included in the mixing models of my study were based on prior knowledge and 142 confirmed by the literature (Fritz 1974; Baker-Dittus 1978; Allen et al. 1994; James-Pirri et al. 2001; McMahon et al. 2005). The MixSIR software accounts for the variability in diet sources by using a Bayesian model that includes estimates of mean, standard deviation, fractionation, and prior knowledge of diet source contributions. A detailed explanation of model parameters and calculations can be found in Moore and Semmens (2008). Carbon and nitrogen stable isotopes in a variety of species collected from ditch-plug and natural pool habitats revealed distinct foraging patterns for resident fish. The patterns observed showed differences in diet sources and habitat utilization by fish placed in the two habitat types. In addition, anthropogenic influence from adjacent upland development can contribute to the differences in stable isotope values that I observed between the two habitats. Diet in ditch-plug habitat was influenced primarily by invertebrates, with benthic microalgae as the basis of the food chain, and Spartina spp. and POM contributing less overall. Variability in plant 815N enrichment may be due to a number of factors influencing mineralization (Kirkpatrick and Foreman 1998; Mitsch and Gosselink 2000) nitrification and denitrification processes (Valiela and Teal 1979; Kaplan et al. 1979), and other physical factors affecting plant nutrient uptake (Hopkinson and Schubauer 1984; Scudlark and Church 1989; Portnoy and Giblin 1997) that occur in wetland soils. Plants in ditch-plug habitat had higher 515N, compared with plants in natural pool habitat, indicating differences in nitrogen sources or physical conditions that affect nitrogen processing in the two habitats (Wigand et al. 2003; Wigand et al. 2007; Fitch et al. 2009; Bauersachs et al. 2009). Koch and Gobler (2009) observed increased levels of nitrogen 143 in their study of salt marsh ditches throughout the estuaries of Long Island, New York, attributed to upland runoff. In New England, Fitch et al. (2009) studied nitrogen levels in soils and plants of the Webhannet Estuary in Wells, Maine (directly north of Moody Marsh, separated by a narrow strip of upland residential development), and found higher levels of nitrate in salt marsh habitat adjacent to upland development. The findings of those two previous studies, combined with the results of isotopic analysis from my study, suggest ditches that extend from ditch-plugs toward the upland border act as conduits for the transport of upland derived nitrogen into the ditch-plugs. The poor drainage and infrequent flushing of ditch-plugs result in the accumulation of the ditch-transported nitrogen, and higher 815N values in plants of ditch-plug habitat. These increased 815N values are then be passed along the food web to higher consumers (Fry 2006). Determining the exact source of nitrogen enrichment and its transfer through the food web was beyond the scope of my study; however, one source that deserves further investigation is nitrogen loading from surrounding uplands directed into ditch-plugs. Salinity and DO have been shown to affect nitrification and denitrification processes in estuarine environments. Vob et al (1997) observed a decrease in NH/ with increase in DO, and an inverse relationship between 815N levels and NH/ availability. Therefore, as DO increases, 815N values from samples taken within the habitat can also increase. Rysgaard et al. (1999) observed an increase in NH/ availability with increased salinity resulting from competition for negative binding sites among NH/, Na+, and 2+ + 15 Mg . Therefore, as salinity decreases, NH4 can also decrease, and 8 N values from samples within the habitat can increase. My regression results showed an inverse relationship between fish 815N and salinity levels, and this relationship was most apparent 144 in the lower salinity and higher 815N values in fish of ditch-plugs. These observations, in addition to terrestrial nitrogen sources, may explain the increased 815N values observed for fish and plants in ditch-plug habitat, where higher pool DO and lower salinity were observed. The results show that levels of 815N are associated with levels of salinity, indicating differences in 815N sources associated with terrestrial, freshwater, and marine inputs; and differences in nitrogen sources at a the base of the food web are reflected in the diet of fish at higher trophic levels in the two habitats. The higher 815N values in ditch-plug fish indicate differences in diet sources and resource use among habitats. Plant detritus can contribute directly to the isotopic signatures of the fish (Svensson et al. 2007), as well as trophic transfer from predation on invertebrates that graze on the plants (Peterson et al. 1986; Currin et al. 1995; Deegan and Garritt 1997; Weinstein and Litvin 2000; Hughes et al. 2000; Wozniak et al. 2006). The higher 815N in plants of ditch-plug habitat (Figure 56) can contribute to the higher values for ditch-plug fish, complicating the interpretation of 815N as an indicator of trophic level. Thus, fish in ditch-plug habitat are not necessarily feeding higher on the food chain as is commonly assumed with higher 815N values. Therefore, an analysis of alternative 815N sources and their impact on habitat is necessary to fully understand source contributions and trophic relationships within salt marsh food webs. The lower 813C signature of fish held in ditch-plug habitat can be attributed in part 11 to the moderate 8 C values of benthic microalgae, which I found to be an important foundation species in salt marsh food webs. In addition, Wozniak et al. (2006) found that F. heteroclitus from areas with a higher percentage of C3 plants, restricted tidal flow, and 11 lower water column salinities tend to have lower 8 C values. Salicornia europaea, a 145 pioneering opportunistic C3 plant with high stress tolerance, regularly invades the plant die back areas adjacent to ditch-plugs (Chapter 3), and the trophic transfer of 813C from S. europaea can be an additional source of lower 5I3C values for F. heteroclitus in ditch- plug habitat. Further, results of my regression analyses show that salinity influences carbon sources at the base of the food web, and these differences are reflected in the diet of fish at higher trophic levels in the two habitats. The diets of consumers in natural pool habitat were more specialized, focusing on specific food sources. The 813C and 815N values for fish in natural pools were more tightly clustered, indicating more consistent and less variable diets. Although invertebrate species richness was higher in natural pool habitat, fish diets in natural pools received larger contributions from fewer invertebrate taxa that were of higher caloric value; with BMA as the basis of the food web, and Spartina spp. and S. europaea contributing less to the overall diet of primary consumers. However, the 8I3C values of fish in natural pool habitat were more in line with Spartina spp., suggesting a tight coupling for the transfer of carbon energy from the primary C4 plant source, through primary invertebrate consumers that specialized in grazing on Spartina spp., and then on to the tertiary consumer F. heteroclitus. In addition, Allen et al. (1994) and James-Pirri et al. (2001) working in New England marshes noted detritus in gut samples, suggesting that F. heteroclitus can also assimilate carbon directly from plant material or the fungi and bacteria growing on the plants, which can influence the 813C value I observed for fish in both habitats. 146 Trophic Webs Connectance is defined as the proportion of interspecific interactions in a community matrix not equal to zero, with higher species richness resulting in higher connectivity and more complex food webs (DeAngelisl975; Martinez 1992). Early theoretical views of food web stability involved static linkage patterns and suggested fewer trophic interactions reduce the stability of the food web since each individual species relies on fewer resources, which affects the ability of the system to withstand stochastic events (Elton 1958; MacArthur 1957; and Hutchinson 1959). In contrast, more recent mathematical models suggest that populations are less stable and species extinction is more likely in complex food webs (May 1972; Gilpin 1975; Pimm 1991). Still others recognize the adaptive nature of predator-prey interactions and the influence of time over stability in altered food webs (Winemiller 1990; Abrams 1992; Kondoh 2003). However, predator prey responses to changes in habitat structure are not well understood, and determining food web structure under existing conditions is necessary for understanding the effects of habitat modifications on trophic interactions and ecosystem function. Common species tend to play an important role in food web dynamics. Due to their relative abundance, habitat disturbance tends to have large impacts on common species (Gaston 2010). Since common species tend to modulate physical and biological interactions through facilitation (Bertness and Hacker 1994), eco-engineering (Mullan Crain and Bertness 2006), and frequent encounters (Gaston 2010), loss of common species in large numbers will have cascading effects on the species that rely on them, altering the trophic structure of the food web (Gaston 2010). Loss of common high 147 marsh plants as a result of ditch-plugging (Chapter 3), therefore, can alter trophic interactions within the habitat. The vegetated marsh surface is connected to the high marsh pool food web by fish that access the flooded marsh surface to feed during spring tides (Weisberg and Lotrich 1982; Fell et al. 1998; MacKenzie and Dionne 2008). Previous studies observed higher gut content and productivity in F. heteroclitus that had accessed the marsh surface, compared with fish confined to channel habitat (Wiseberg and Lotrich 1982; Fell et al. 1998). MacKenzie and Dionne (2008) found that F. heteroclitus had higher productivity when provided with access to the vegetated marsh surface, compared with fish confined to pools. Plants ameliorate edaphic stressors, provide heterogeneity, and produce organic matter, which affects habitat and food resources utilized by invertebrate species that are preyed upon by fish. Furthermore, Whitcraft and Levin (2007) working in southern California marshes observed trophic cascades involving the effects of plant cover on invertebrate richness. The authors observed shifts in microalgal communities in response to changes in plant cover and degree of shading, which in turn altered benthic invertebrate community composition and richness. Thus, marsh plants can directly affect the connectivity and productivity of high marsh pool food webs. Conclusions made by these previous studies support my hypothesis that altered hydrology and plant communities that occur in response to ditch-plugging contribute to lower resource availability and invertebrate species richness, and altered food webs in ditch-plug habitat. Natural Pool Habitat Food Web Habitat structure can influence the richness and abundance of food web components in natural pool habitat. The abundance of emergent plants and benthic 148 microalgae in natural pool habitat (Chapter 3) provides both the initial energy source for the system, as well as physical habitat used for breeding and refuge by higher trophic organisms. Increased heterogeneity of the habitat, including the physical structure of the emergent plant community (Chapter 3), provides a mosaic of habitat patches with unique characteristics that create advantages and disadvantages for various organisms (Kneib 1984). Abjornsson et al. (2004) found that invertebrate prey species increased use of refuge habitat provided by emergent plants in the presence of predatory fish. The greater abundance of emergent plants found in natural pool habitat (Chapter 3) provides structural refuge for the benthic invertebrate community (Kneib 1984). Such refuge causes structural interference, effectively reducing predation for certain species by physically preventing predators from locating the prey (Hughes and Grabowski 2006; Moody and Aronson 2007; Manatunge et al. 2000). This same structural heterogeneity also provides advantageous habitat for ambush predators (Fulton 1985). The structural heterogeneity provided by abundant plants in natural pool habitat (Chapter 3) helps maintain benthic invertebrate prey populations by providing obstacles for bioturbating organisms and reducing disturbance to the benthic environment (Kneib 1984), as well as by increasing soil oxidation through evapotranspiration (Dacey and Howes 1984; Nuttle 1988). Structural interference also results in predator resource partitioning, reducing niche overlap through predator-prey specializations (Hughes and Grabowski 2006). The greater number of secondary predators that occur in natural pool food webs can also influence the benthic invertebrate community. Posey and Hines (1991) demonstrated that interference from competing predators can reduce predation pressure on the prey community, thus influencing the prey community composition 149 through the indirect effects of predator competition. Likewise, predation pressure from tertiary consumers can help maintain secondary consumer species richness through predator mediated coexistence (Paine 1966, Paine 1980), altering predation pressure on the primary consumers and primary producers in a trophic cascade effect that maintains higher species richness and connectivity in the system (Strong 1992; Silliman and Bertness 2002). Ditch-plug Habitat Food Web Hydrologic alterations from ditch-plugging significantly impact the physical and biological characteristics of surrounding high marsh habitat (Chapters 2 and 3). Large numbers of common plant species that typically dominate high marsh habitats die off in the areas immediately surrounding ditch-plugs (Chapter 3). Plants provide structural refuge for prey species (Cody 1986; Boyer and Fong 2005). Reduced structural heterogeneity commonly results in increased predation (Boyer and Fong 2005), which can lead to lower abundance of preferred prey and reduced species richness (Posey and * Hines 1991). Thus, the reduced plant cover characteristic of ditch-plug habitat (Chapter 3) contributes to fewer primary and secondary invertebrate consumers and reduced trophic connectivity in ditch-plug food webs. The altered food web structure in ditch-plug habitat resulting from loss of biological resources was further evidenced by lower invertebrate species richness, biomass, density, and trophic interactions in ditch- plug compared with natural pool habitat. In addition, benthic invertebrate communities exist in three-dimensional space, making vertical heterogeneity and competition for space important factors in species composition (Lilihan and Micheli 2001; Rodil et al. 2008). Soil biogeochemical 150 conditions, therefore, are of importance to benthic invertebrate communities and the primary producing plants upon which they feed. Subsequently, the quality of benthic habitat can have trophic implications. Ground and surface water retained in ditch-plug habitat decreases the depth to reduced soil conditions, limiting vertical heterogeneity and volume of benthic habitat, increasing benthic invertebrate competition for space and resources (Rodil et al. 2008). Such limiting resources and competitive pressures contribute to reduced species richness and, consequently, lower connectivity in ditch-plug habitat (Montoya et al. 2003; Thebault and Loreau 2003). Further studies focusing specifically on species richness and connectivity in ditch-plug habitat are needed to better understand the function of these altered food webs and how they may change through time. The resulting shift in species richness and trophic structure will depend on species composition, degree of niche overlap among remaining species, and the rate of primary production available in the system (Thebault and Loreau 2003). In addition, limiting resources and competitive pressures can also lead to emigration and reduced species richness at multiple trophic levels (Lenihan and Micheli 2001; Steneck and Carlton 2001; Kneib 1984; Strong 1992). Further, the absence of Arachnida and Coleoptera in the secondary consumer trophic level of ditch-plug habitat results from plant die back and habitat loss that occurs in areas surrounding ditch-plugs in response to changes in the hydrologic regime of the surrounding high marsh habitat. A positive relationship exists between below ground biomass and density of the benthic invertebrate community (Seliskar et al. 2002). Under reducing soil conditions, the primary producers decrease in biomass, richness, and composition (Baldwin and Mendelssohn 1998, Koch et al. 1990), limiting food sources for the benthic invertebrate 151 primary consumers (Seliskar et al. 2002). The increase in competition for limited resources in ditch-plug habitat would, therefore, contribute to the reduction in species richness and abundance of the invertebrate primary and secondary consumers, and decrease trophic interactions through competitive exclusion and trophic cascade (Lenihan and Micheli 2001; Steneck and Carlton 2001; Strong 1992; Thebault and Loreau 2003). The more variable diets of primary consumers in ditch-plug habitat indicate limited availability of preferred plant diet sources. The fish (tertiary consumer) diet was more variable in ditch-plug habitat as well, with Amphipoda and Polychaeta as two species of lower caloric value than to the third diet source, Diptera (Cummins and Wuycheck 1971), indicating limited availability of preferred invertebrate diet sources for tertiary consumers in ditch-plug habitat. Thus, the loss of emergent plants from ditch-plug habitat has resulted in negative impacts on trophic transfer and connectivity of the system. Stability of an ecosystem relies heavily on the degree of connectivity and the function of the organisms in the trophic web. The higher the connectivity of a species, the more influence that species has on the trophic structure of the system. Therefore, the loss of highly connected species has greater importance than the number of species lost (Dunne et al. 2002; Williams and Martinez 2000; Montoya and Soule 2002). The functional attributes of the common plant species, S. patens and S. alternijlora, include primary production at the base of the food web, amelioration of edaphic conditions, and above and belowground structural habitat for both predator and prey. The reduction of these dominant plants from areas surrounding ditch-plugs contributes to habitat degradation for benthic invertebrate species and the loss of secondary consumers Arachnida and Coleoptera from the ditch-plug food web. Both of these species 152 contribute to the diet of the tertiary consumer F. heteroclitus in natural pool habitat, as confirmed by isotopic analysis in my study and field observations (Chapter 3; Vincent 2005, personal observations), as well as exert predation pressure on consumers at lower trophic levels. The absence of these mid-level consumers in ditch-plug habitat reduced food web connectivity in this habitat and likely altered the competitive balance of the primary consumers. Lower connectivity tends to increase vulnerability to disturbance and additional species loss (Dunne et al. 2002; Williams et al. 2002; Naeem et al. 1994). Such effects can lead to further instability, loss of system functionality, and habitat degradation (Naeem et al. 1994). Conclusion My study identified distinct differences in species richness, abundance, and food web structure among ditch-plug and natural pool habitats at Moody Marsh in Wells, Maine, and these differences in community structure have resulted in ecological dissimilarities in function between the two habitats. Ditch-plug creation increases the extent and duration of flooding and soil saturation, resulting in highly reduced soils and reduced plant density and biomass in areas immediately surrounding the created pool habitat (Chapters 2 and 3). My conceptual model presented in Chapter 1 highlighted the differences that I expected to encounter among the two habitats. My hypothesis that ditch-plugging alters physical parameters that influence fish productivity and trophic webs, and result in distinct differences among natural pools and ditch-plugs was accepted. Altered hydrology associated with ditch-plugs resulted in anoxic soils and reduced primary production, the two physical and biological variables that contributed to 153 differences in invertebrate community composition, energy availability, and trophic web structure observed between ditch-plug and natural pool habitats. Fish growth was higher in natural pool habitat, as was invertebrate species richness, density, biomass, and caloric value. The invertebrate biomass available for tertiary consumers (fish) in ditch-plug habitat was of lower caloric value than that available in natural pool habitat, and I infer differences in energy availability among the two habitats as a biological response to changes in physical conditions associated with altered ditch-plug habitat. The ecological significance of reduced energy available for trophic transfer was evidenced by lower fish growth in ditch-plugs compared with natural pools. Higher 815N values in plants of ditch-plug habitat are can be associated with nutrient inputs in runoff from surrounding upland development. More work is needed to gain a better understanding of the relationship between anthropogenic nutrient loading and trapping by ditch-plugs. The higher 615N values in ditch-plug fish appears to be related to the level of 815N in plants of that habitat and not necessarily an indicator of feeding at higher trophic levels. The stable isotope mixing model identified distinct resource utilization and trophic structure for natural and created pools. Food web analysis identified ditch-plug fish as having more varied generalist diets, consuming available prey with lower caloric value. In contrast, the diet of fish in natural pools was more specialized, targeting prey with higher caloric values, demonstrating that natural pools provide higher habitat quality in terms of fish productivity. The altered community composition in ditch-plug habitat appeared to have an effect on the trophic structure and transfer of energy through the 154 system. Although four trophic levels were found in both habitats, fewer primary and secondary consumers were observed in ditch-plug habitat, the result of altered habitat conditions and competitive interactions. The loss of plant biomass, structural heterogeneity, and increased physical stress from anoxia in habitat adjacent to ditch-plugs has resulted in lower food web complexity and connectivity, as evidenced by the loss of one primary consumer, the absence of two secondary consumers, the presence of only one Acarina spp., and significantly lower species richness in ditch-plug habitat. The lower trophic connectivity in ditch-plug habitat increases the vulnerability of the food web to disturbance such as drought or sea level rise, which can lead to further reduction of consumer linkages and trophic interactions. My study increases our understanding of the ecology of salt marsh pools, and will aid resource managers with future salt marsh restoration planning. The results show that pools created using ditch-plugs do not replicate the structure and function of natural pools in support of F. heteroclitus and invertebrate communities, and therefore, ditch-plugging is not recommended. 155 CHAPTER V CONCLUSION Historic and Current Hydrologic Alterations My study investigated two forms of hydrologic alteration, ditching and ditch- plugging. Ditches constructed in the 1930s along the coast from Virginia to Maine failed in their attempts to eliminate water presence on the marsh and mosquito recruitment (Daiber 1986). Ditch-plugs constructed from Maryland to Maine beginning in the 1990s achieved their objective of increasing water presence on the marsh (James-Pirri et al. 2011). However, the inherent value of tidal marshes and understanding their response to these types of hydrologic alterations received little forethought, especially with regard to the scale at which these projects were conducted. In fact, little attempt has been made to assess these impacts (but see Adamowicz and Roman 2002; James-Pirri et al. 2011), and knowledge of the primary and secondary effects of these large scale hydrologic manipulations on ecosystem structure and function remains poor decades later. Altering hydrologic flow can influence physical and biological feedbacks involving hydrology, soil accretion, plant growth, and decomposition. Without human interference, these factors maintain the salt marsh self-maintenance process and enable marsh elevations to keep pace with the rate of sea level rise. Hence, a more conservative management approach using small scale experimental units would have been more effective for developing an a-priori understanding of ecosystem response to habitat manipulation, helping to avoid unintended short-term and long-term impacts prior to large scale habitat 156 disturbance. My research identified some of the long-term habitat responses to ditching and ditch-plugging, as summarized here. Ecosystem Response to Ditching The conceptual models presented in Chapter 1 have been modified based on my study results to illustrate the observed differences in physical and biological conditions among the habitats (Figures 59 through 62). The common perceptions that ditching salt marshes lowered water levels (Daiber 1986) and increased high marsh plant cover and biomass (Bourn and Cottam 1950) are not accurate. In fact, the results of my study show that ditching is associated with areas of higher soil water levels and plant communities indicative of wetter habitat conditions. The physical and biological feedbacks that contribute to the marsh accretion process have been altered in ditched habitat, as evidenced by reduced pore water drainage, lower redox potential, lower mineral deposition, and peat subsidence of 23 cm at the edge of the ditch; but perhaps not to the extent of circumventing the self-maintenance process to date. When compared with creek habitat (Figure 60), the most apparent effects of altered hydrology are prolonged pore-water retention within the rooting zone, changes in plant community composition (i.e., lower abundance of tall-from S. alterniflora and higher abundance of forb species in ditched habitat), lower plant biomass adjacent to ditches, and lower mineral content in soils. The important question is whether the combined alterations to physical processes and plant characteristics will hinder the habitat's ability to keep pace with increasing rates of sea level rise, especially in more heavily ditch marshes. 157 Created Ditch Habitat Observed Conditions Mash accretion appears to Moderate levels of carbon have kept pace with sea storage and percent organic level, but subsidence out to matter results in intermediate 15 meters has occuned over levels of sediment strength time, as indicated by Pore water drainage is sufficient comparisons with high enough to support higher levels of marsh surface elevations Moderate bulk density suggests vegetation richness and percent cover, surrounding natural pools lower rates of mineral deposition but only moderate levels of biomass High Tide t Fluctu^in^S^ment Low Tide Pore-water Less extensive pore water drainage through the rooting zone increases edaphic stress and H,0 facilitates a higher percent cover of more stress Moderate redox levels tolerant fort) panne species that provide habitat resulting from higher heterogeneity in areas of the high marsh ground water suggest increased edaphic stress Vertical banks and moderate mineral content results in narrower ground water fluctuations and higher mean water level 2011 Figure 59. Ditch habitat observed condition. Natural Creek Habitat Observed Conditions High bufc density and mineral content, along with moderate carbon storage, suggests greater mineral deposition that More extensive pore-water drainage Lower percent contributes to higher marsh surface through the rooting zone creates less organic content elevations beyond banks in support of the stressful edaphic conditions that contributes to self maintenance process facilitate high vegetation richness, lower sedffnent cover, and biomass strength High Tide • Fluctuating Sediment Sloping banks provide low marsh conditions dominated by iPwTKte Porfr-water Spartina attemiflora that provide habitat heterogeneity High redox potential suggests longer periods of sediment oxidation and more favorable Sloping banks and high mineral conditions for vegetation growth content facilitates wider ground water fluctuations and lower mean water level NaD Vlnccnt 2011 Figure 60. Creek habitat observed conditions. Ditch-plug Pool Habitat Observed Conditions Apparent decoupling of the marsh self maintenance Less diverse vegetation dominated by process with subsidence out Saiicomia europaea, Spartina altemiflroa, to 20 meters, as indicated by sulfur bacteria, and algae mats comparisons with high marsh surface elevations and sediment characteristics Extensive areas of vegetation die-back surrounding all other habitats characterized by high saJtiity, bare soil, impounded surface water, and Lower fish growth in ditch-plug hummock-holow vegetation growth pools is likely the result of altered patterns invertebrate prey community structure, and lower invertebrate bkxnass and caloric value Permanently Permanently saturated soils Permanently H^O saturated result in low redox potential saturated sediments (and higher sulfide levels as sediments indicated by soil coloration and high percent cover of Higher water levels and sulfur bacteria), which Unvegetated undulating bathymetry and deep sumps that result in lower sediment reduced sediment drainage provides less favorable dissolved oxygen and higher sediment result in stressful edaphic conditions for vegetation sulfide levels (as indicated by sediment conditions that likely Imit growth and benthic coloration) are characteristic of ditch-plug resource availability and invertebrates pools, and result in altered invertebrate contribute to a less community structure as indicated by lower complex food web Reduced vegetation bkxnass, low richness, density, and biomass, and carbon storage, low bulk density, and differences in species composition high water levels result in the compared with natural pools breakdown of sol structure, low sediment strength, erosioin, and decreased habitat stability nob Vincent 2011 Figure 61. Ditch-plug habitat observed conditions. Natural Pool Habitat Observed Conditions Hydrotogic flow and pore-water Pore-water drainage through the rooting drainage are enough to facilitate the salt zone is maintains high vegetation cover, marsh self maintenance process, as biomass, and richness, yet species evidenced by moderate bulk density composition is more representative of combined with high carbon storage, moderate edaphic stress as indicated by the percent organic content, and marsh higher percent cover of short-form Spartma Higher fish growth h natural pools patens and ftxto panne species surface elevations is likely the result of higher invertebrate density, btomass, species composition, and caloric Highjide Fluctuating Sediment Fluctuating Sediment Pore-water Pore-water ^ DO • High redox potential (and lower _ n „ n Moderate Moderate water level fluctuations sulfide levels as indicated by soil and sediment drainage provide coloration and lower presence of edaphic conditions that enable sulfur bacteria) provides more primary production and resource favorable conditions for vegetation availability, and likely contribute to growth and bent hie invertebrates Uniform bathymetry at moderate depths vegetated by Ruppia maritima, higher a more complex food web sediment dissolved oxygen, and lower sediment sulfide levels (as indicated by High percent organic content and sediment coloration) are characteristic of carbon storage result in high natural pools, and likely contribute to higher sediment strength aid habitat invertebrate richness, density, and biomass, stability along with community structure that varies in composition from ditch-plug habitat Rob vncent 2011 Figure 62. Natural pool habitat observed conditions. Stevenson and Kearney (2009) stated that 90 percent of salt marsh habitat world wide could be altered by climate-related impacts by 2100. Pore-water drainage and bulk density in ditched habitat lagged behind levels found in creek habitat, which combined with climate forcing and rising sea levels can lead to greater impacts to high marsh habitat in the future. In order to properly manage historic and on-going anthropogenic impacts to salt marsh habitat, we need to develop a full understanding of the sediment transport and marsh accretion processes associated with natural and created channel habitats, especially in response to changing climactic conditions. Nutrient runoff and eutrophication is also a concern as human encroachment on salt marsh habitat continues. Bertness et al. (2009) describe how nutrient loading alters salt marsh plant competition, favoring S. alterniflora, and triggers top-down consumer control by insect herbivores on primary production. Therefore, the function of ditches as conduits for the transport of upland derived nutrients (and their effects on marsh processes including carbon and nitrogen cycling) deserves further attention. Studies that model the interaction of multiple stressors over time are needed to estimate future impacts to coastal habitats and allow for the development of adaptive management plans that account for physical and biological responses to climate variation. For instance, heavily ditched marshes may have experienced greater impacts than those I observed for moderately ditched marshes. My observations of aerial photos suggest that the overlapping effects of adjacent hydrologic regimes that are associated with closely spaced parallel and grid ditching, such as subsidence and reduced pore-water drainage over large areas of the marsh may have altered marsh feedback processes, 162 t resulting in greater pore-water retention. My results show that such conditions can alter the marsh maintenance process, resulting in subsidence and altered high marsh habitat. Therefore, the addition of ditch-plugs (which promotes water retention) in heavily ditched marshes would enhance the positive feedbacks that lead to subsidence and greater loss of vegetated habitat. I recommend that further investigation with small scale experimentation and modeling is needed to determine the effects of concentrated grid ditching on salt marsh physical and biological processes. My water level and plant results indicate that although ditches provide similar functions as creeks in conveying hydrologic flows and maintaining high marsh habitat, the differences in terms of marsh processes (i.e., lower sediment accretion and bulk density, less extensive pore-water drainage, and subsidence) suggest that created linear ditches are not likely substitutes for creeks formed through natural processes. Ecosystem Response to Natural Creek Hydrology My results demonstrated that some physical and biological variables in natural creek habitat were distinctly different from those in created ditch habitat. For instance, I attribute higher soil bulk density to increased sediment transport and deposition associated with greater tidal ranges observed in natural creek habitat (Table 2), a process confirmed by Chmura et al. (2001) in salt marshes of eastern Canada. The wider groundwater fluctuation and more extensive drainage through the rooting zone in natural creek soils promotes higher redox potential and reduced plant stress, resulting in higher plant biomass. Chmura and Hung (2004) determined that more extensive tidal cycling facilitates marsh accretion processes, and the elevation results from my study confirm 163 that these conditions persist in support of salt marsh self-maintenance in natural creek habitat. Ecosystem Response to Ditch-Plugging It is apparent from my research that ditch-plugging has not only halted, but has reversed marsh building processes to a distance of at least 20 meters surrounding ditch- plugged pools. Ditch-plugging results in the decoupling of physical and biological processes that promote salt marsh self-maintenance and the ability of the habitat to keep pace with sea level rise. High surface water levels, permanently saturated soils, low bulk density and carbon storage, low soil strength, very low redox levels, extensive areas of plant dieback, and marsh subsidence associated with hydrologic alterations from ditch- plugging all support this conclusion (Figure 61). In addition, the high edaphic stress observed in ditch-plug habitat has altered plant and invertebrate communities, affecting food web structure and trophic transfer, which contributed to lower fish growth compared with natural pool habitat (Figure 62). The altered physical and biological feedbacks can affect DO, carbon, and nitrogen cycling, as evidenced by lower soil DO, lower carbon storage, and higher 5I5N in the plants and fish of ditch-plug habitat. Further study is needed to clarify the efFects of ditch-plugging on the salt marsh nitrogen cycle and its efFects on ecosystem processes. Studies that investigate climate-related implications of carbon storage loss and the efFects of deep-sumps on dissolved oxygen and water column stratification in ditch-plug habitat are also recommended. Results of my study indicate that ditch-plug pools are not suitable substitutes for naturally occurring salt marsh pools. 164 Ecosystem Response to Natural Pool Hydrology The hydrologic regime of habitat adjacent to natural pools supports the physical and biological feedbacks that facilitate marsh building processes and enable salt marsh self-maintenance, as evidenced by carbon storage, percent organic matter, soil strength, and marsh surface elevations all higher than in the other habitats. Natural hydrological conditions promote habitat heterogeneity in support of complex food webs with higher caloric value and higher F. heteroclitus productivity, the primary link in transferring marsh energy to adjacent habitats. I conclude that natural pool habitat provides greater value than ditch-plug habitat in terms of fish productivity, and thus greater potential for trophic transfer of marsh derived resources to adjacent ecosystems. Future studies, therefore, should focus on clarifying the connectivity and trophic relationships that exist between salt marsh pools and adjacent habitats (i.e., marine, freshwater wetland, riverine, and terrestrial). In doing so, we may gain a better understanding for the importance of pool habitat in estuarine ecosystem processes, and how to better manage surface water features in restoration design and maintenance activities. 165 LITERATURE CITED Abjornsson, K., L. Hansson, and C. Bronmark. 2004. Responses of prey from habitats with different predator regimes: local adaptation and heritability. Ecology. Vol. 85. pp. 1859-1866 Aber, J.D., and J.M. Melillo. 2001. Terrestrial Ecosystems. Harcourt Academic Press. 556 pp Able, K.W., K.J. Smith, and S.M. Hagan. 2005. Fish composition and abundance in New Jersey salt marsh pools: sampling technique effects. Northeastern Naturalist, Vol. 12. No. 4. pp. 485-502 Able, K.W., S.M Hagan, and S.A. Brown. 2006. Habitat use, movement, and growth of young-of-the-year Fundulus spp. in southern New Jersey salt marshes: comparisons based on tag/recapture. Journal of Experimental Marine Biology and Ecology. Vol. 334. pp. 177-187 Adamowicz, S.C., and C.T. Roman. 2002. Initial ecosystem response of salt marshes to ditch-plugging and pool creation: experiments at Rachel Carson National Wildlife Refuge (Maine). Final Report. USGS Patuxent Wildlife Research Center, Costal Research Field Station. University of Rhode Island. Narragansett, RJ. 102 pp Adamowicz, S.C., and C.T. Roman. 2005. New England salt marsh pools: a quantitative analysis of geomorphic and geographic features. Wetlands. Vol. 25. No. 2. pp. 279-288 Allen, E.A., P.E. Fell, M.A. Peck, J.A. Gieg, C.R. Guthke, and M.D. Newkirk. 1994. Gut contents of common mummichogs, Fundulus heteroclitus L., in restored and impounded marsh and in natural reference marshes. Estuaries. Vol. 17, No. 2. pp. 462-471 Anisfeld, S.C. and G Benoit. 1997. Impacts of flow restrictions on salt marshes: an instance of acidification. Environmental Science Technology. Vol. 31. 1650-1657 Argow, B.A. and D.M. Fitzgerald. 2006. Winter processes on northern salt marshes: evaluating the impact of in-situ peat compaction due to ice loading, Wells, ME. Estuarine, Coastal and Shelf Science. Vol. 69. 360-369 Baldwin, A.H. and I.A. Mendelssohn. 1998. Response of two oligohaline marsh communities to lethal and nonlethal disturbance. Oecologia. Vol.116. pp. 543-555 Bannon, R.O. and C.T. Roman. 2008. Using stable isotopes to monitor anthropogenic nitrogen inputs to estuaries. Ecological Applications. Vol. 18. No. 1. pp. 22-30 166 Bastow, T.P., G. Jackson, and J.S. Edmonds. 2002. Elevated salinity and isotopic composition of fish otolith carbonate: stock delineation of pink snapper, Pargrus auratus, in Shark Bay, western Australia. Marine Biology. Vol. 131. pp. 801-806 Bauman, R.H., and J.W. Day. 1993. Precision measurements of sediment elevation in shallow coastal areas using sediment-erosion tables. Estuaries. Vol. 16. pp. 375-380 Bertness, M.D. 1985. Fiddler crab regulation of Spartina alterniflora production on a New England salt marsh. Ecology. Vol. 66. No. 3. pp. 1042-1055 Bertness, M.D. 1991a. Zonation of Spartina patens and Spartina alterniflora in a New England salt marsh. Estuaries. Vol. 72. pp. 138-148 Bertness, M.D. 1991b. Interspecific interactions among high marsh perennials in a New England salt marsh. Ecology. Vol. 27. No. 1. pp. 125-137 Bertness, M.D. 1992. The ecology of a New England salt marsh. American Scientist. Vol. 80. pp. 260-268 Bertness, M.D. 1999. The Ecology of Atlantic Shorelines. Sinauer Associates, Inc. Sunderland, MA. 417 pp Bertness, M.D., and A.M. Ellison. 1987. Determinants of pattern in a New England salt marsh plant community. Ecological Monographs. Vol. 57. pp. 129-147 Bertness, M.D., L. Gough, and S.W. Shumway. 1992. Salt tolerances and the distribution of fugitive salt marsh plants. Ecology. Vol. 73. pp. 1842-1851 Bertness, M.D., and S.D. Hacker. 1994. Physical stress and positive associations among marsh plants. The American Naturalist. Vol. 144. pp. 363-372 Bertness, M.D., and S.C. Pennings. 2000. Spatial variation in process and pattern in salt marsh plant communities in eastern North America. In, Concepts and Controversies in Tidal Marsh Ecology, Michael P. Weinstein and Daniel A. Kreeger editors, pp.39-57 Bertness, M.D., B.R. Silliman, and C. Holdredge. 2009. Shoreline development and the future of New England salt marsh landscapes. In Silliman, B.R., E.D. Groshlz, and M.D. Bertness. 2009. Human impacts on salt marshes, a global perspective. University of California Press, pp. 137 -148 Bohlen, C. 2007. Sixty years of change in a Maine salt marsh landscape: implications for restoration. http://abacus.bates. edu/" cbohlen/research.htm Boumans, R.M.J., D.M. Burdick, and M. Dionne. 2002. Modeling habitat change in salt marshes after tidal restoration. Restoration Ecology. Vol. 10. No. 3. pp. 543-555 167 Bourn, W.S. and C. Cottam. 1950. Some biological effects of ditching tidewater marshes. Research Report 19. U.S. Fish and Wildlife Service, U.S. Department of Interior,Washington, D.C. Boyer, K. E., and P. Fong. 2005. Co-occurrence of habitat-modifying invertebrates: effects on structural and functional properties of a created salt marsh. Oecologia Vol. 143. pp. 619-628 Bromberg, K. and M.D. Bertness. 2005. Reconstructing New England salt marsh losses using historical maps. Estuaries. Vol. 28. pp. 823-832 Brower, J.E., J.H. Zar, and C.N. von Ende. 1990. Field and laboratory methods for general ecology. Wm. C. Brown publishers. 237 pp Buchsbaum, R.N., L.A. Deegan, J. Horowitz, R.H. Garritt, A.E. Giblin, J.P. Ludlam, and D.H. Shull. 2009. Effects of regular salt marsh haying on marsh plants, algae, invertebrates, and birds at Plum Island Sound, Massachusetts. Wetlands Ecology and Management. Vol. 17. pp. 469-487 Burdick, D.M., M. Dionne, R.M. Boumans, and F.T. Short. 1997. Ecological responses to tidal restorations of two northern New England salt marshes. Wetlands Ecology and Management. Vol. 4. No. 2. pp. 129-144 Burdick, D.M. and R.A. Konisky. 2003. Determinants of expansion for Phragmites australis, common reed, in natural and impacted coastal marshes. Estuaries. Vol. 26, No. 2B. pp. 407-416 Burdick, D.M., I.A. Mendelssohn, and K.L. McKee. 1989. Live standing crop and metabolism of the marsh grass Spartina patens as related to edaphic factors in a brackish, mixed marsh community in Louisiana. Estuaries. Vol. 12. No. 3. pp. 195-204. Cahoon, D.R., and D.J. Reed. 1995. Relationships among marsh surface topography, hydroperiod, and soil accretion in deteriorating Louisiana salt marsh. Journal of Coastal Research. Vol. 11. No.2. pp. 357-369 Chapman DW. 1978. Production. In: Begenal T (ed) Methods for assessment of fish production in freshwaters. Blackwell Scientific Publications, Oxford, pp. 202-218 Clarke, A. and K.J. Gatson. 2006. Climate, energy and diversity. Proceedings of the Royal Society B. Vol. 273, pp. 2257-2266 Chapman, V.J. Salt marshes and salt deserts of the world. 1960. Leonard Hill Books Limited, Interscience Publishers, Inc., New York. 392 pp Chmura, G.L., A. Coffey, and R. Crago. 2001. Variation in surface soil deposition on salt marshes in the Bay of Fundy. Journal of Coastal Research. Vol. 17, No. 1. pp. 221-227 168 Chmura GL, Anisfeld SC, Cahoon DR, Lynch JC. 2003. Global carbon sequestration in tidal, saline wetland soils. Global Biogeochemical Cycles. Vol.17. pp.1111-1133 Chmura, G.L., and G.A. Hung. 2004. Controls on salt marsh accretion: a test in salt marshes of eastern Canada. Estuaries. Vol. 27. No. 1. pp. 70-81 Connell, J.H. 1978. Diversity in tropical rain forests and coral reefs. Science. Vol. 199. No. 4335. pp. 1302-1310 Cottenie, K., N. Nuytten, E. Michels, and L. De Meester. 2001. Zooplankton community structure and environmental conditions in a set of interconnected ponds. Hydrobiologia. Vol. 442. pp. 339-350 Crain, C. and M.D. Bertness. 2006. Ecosystem engineering across environmental gradients: implications for conservation and management. Bioscience. Vol. 56. No. 3. pp. 211-218 Crain, C.M., B.R. Silliman, S.L. Bertness, and M.D. Bertness. 2004. Physical and biotic drivers of plant distribution across estuarine salinity gradients. Ecology. Vol. 85. pp. 2539-2549 Craft, C.B., E.D. Seneca, and S.W. Broome. 1991. Loss on ignition and kjeldahl digestion for estimating organic carbon and total nitrogen in estuarine marsh soils: calibration with dry combustion. Estuaries. Vol. 14. No. 2. pp. 175-179 Crustinger, G.M., M.D. Collins, J.A. Fordyce, Z. Gompert, C.C. Nice, and N.J. Sanders. 2006. Plant genotypic diversity predicts community structure and governs an ecosystem process. Science. Vol. 313. pp. 966-968 CST/Berger 2005. Laser Mark CST LMH-CU laser with LD400 universal laser detector and sighting rod. Watseka, IL, USA Cummins, K.W. and J.C. Wuycheck. 1971. Caloric equivalents for investigations in ecological energetics. International association of Theoretical and Applied Limnology. 158 pp Cunha, I. and M. Planas. Optimal prey size for early turbot larvae (Scophthalus maximus L.) based on mouth and ingested prey size. Aquaculture. Vol. 175. Nos. 1-2. pp. 103-110 Currin, C.A., S.Y. Newell, and H.W. Pearl. 1995. The role of standing dead Spartina alternijlora and benthic microalgae in salt marsh food webs: considerations based on multiple stable isotope analysis. Marine Ecology Progress Series. Vol. 121. pp. 99-116 Currin, C.A., S.C. Wainright, K.W. Able, M.P. Weinstein, and C.M. Fuller. 2003. Determination of food web support and trophic position of the mummichog, Fundulus 169 heteroclitus, in New Jersey smooth cordgrass (Spartina alterniflora), common reed (Phragmites australis) and restored salt marshes. 2003. Estuaries. Vol. 26. pp. 495-510 D'Alpaos, A., S. Lanzoni, S.M. Mudd, S. Fagherazzi. 2006. Modeling the influence of hydroperiod and vegetation on the cross-sectional formation of tidal channels. Estuarine, Coastal and Shelf Science. Vol. 69. pp. 311-324 D'Alpaos, A., S. Lanzoni, M. Marani, A. Bonometto, G. Cecconi, and A. Rinaldo. 2007. Spontaneous tidal network formation within a constructed salt marsh: observation and morphodynamic modeling. Geomorphology. Vol. 91. pp. 186-197 Dacey, W.H. and B.L. Howes. 1984. Water uptake by roots controls water table movement and soil oxidation in short Spartina marsh. Science. Vol. 224. pp. 487-489 Daiber, F.C. 1986. Conservation of Tidal Marshes. Van Norstrand Reinhold Company. New York. 341 pp Deane, E.E. and N.Y.S. Woo. 2009. Modulation of fish growth hormone levels by salinity, temperature, pollutants and aquaculture related stress: a review. Review of Fish Biology and Fisheries. Vol. 19. pp. 97-120 Deegan, L.A., and R.H. Garritt. 1997. Evidence for spatial variability in estuarine food webs. Marine Ecology Progress Series. Vol. 147: pp. 31-47 » DeLaune, R.D., J.A. Nyman, and W.H. Patrick, Jr. 1994. Peat collapse, ponding and wetland loss in rapidly submerging coastal marsh. Journal of Coastal Research. Vol. 10. No. 4. pp. 1021-1030 DeNiro, M. J. and S. Epstein. 1981. Influence of diet on the distribution of nitrogen isotopes in animals. Geochimica et Cosmochimica Acta, 45:341-351 DeNiro, M. J. & S. Epstein. 1978. Influence of diet on the distribution of carbon isotopes in animals. Geochimica et Cosmochimica Acta, Vol. 42: pp. 495-506 Dionne, J.C. 1989. An estimate of shore ice action on a Spartina tidal marsh, St. Lawrence Estuary, Quebec, Canada. Journal of Coastal Research. Vol. 5. No. 2. pp. 281- 293 Dionne, M., F.T. Short, and D.M. Burdick. 1999. Fish utilization of restored, created, and reference salt-marsh habitat in the Gulf of Maine. American Fisheries Society Symposium. Vol. 22. pp. 384-404 Douglass, J.G., E.A. Canuel, and J.E. Duffy. 2011 Food web structure in a Chesapeake Bay eelgrass bed as determined through gut contents and 13C and 15N isotope analysis. Estuaries and Coasts. Vol. 34. pp. 701-711 170 Dunne, J.A., R.J. Williams, and N.D. Martinez. 2002. Network structure and biodiversity loss in food webs: robustness increases with connectance. Ecology Letters. Vol. 5. No. 4. pp. 558-567 Ellison, A.M., M.D. Bertness, and T. Miller. 1986. Seasonal patterns in the belowground biomass of Spartina alterniflora (Gramineae) across a tidal gradient. American Journal of Botany. Vol. 73. No. 11. pp. 1548-1554 Emery, N.C., P.J. Ewanchuk, and M.D. Bertness. 2001. Competition and salt-marsh plant zonation: stress tolerators maybe dominant competitors. Ecology. Vol. 82. No.9. pp. 2471-2485 Ewanchuk, P.J., and M.D. Bertness. 2004. Structure and organization of a northern New England salt marsh plant community. Journal of Ecology. 92: 72-85 Erwin, R.M., D.K. Dawson, D.B. Souts, L.S. McAllister, and P.H. Geissler. 1991. Open marsh water management in the mid-Atlantic region: aerial surveys of waterbird use. Wetlands. Vol. 11. No. 2. pp. 209-227 Erwin, R.M., J.S. Hatfield, M.A. Howe, and S.S. Klugman. 1994. Waterbird use of salt marsh ponds created for open marsh water management. Journal of Wildlife Management. Vol. 58. pp. 516-524 Ewing, K., K.L. McKee, and I.A. Mendelssohn. 1997. A field comparison of indicators of sublethal stress in the salt-marsh grass Spartina patens. Estuaries. Vol. 20. pp. 48-65 Fagherazzi, S. and D. J. Furbish. 2001. On the shape and widening of salt marsh creeks. Journal of Geophysical Research. Vol. 106. No. CI. pp. 991-1003 Fagherazzi, S. and T. Sun. 2004. A stochastic model for the formation of channel networks in tidal marshes. Geophysical Research Letters. Vol. 31. L21503. 4 pp Faulkner, S.P., W.H. Patrick, Jr., and R.P. Grambrell. 1989. Field techniques for measuring wetland soil parameters. Soil Science Society of American Journal. Vol. 53. pp. 883-890 Fell, P.E., S.P. Weissbach, D.A. Jones, M.A. Fallon, J.A. Zeppieri, E.K. Faison, K.A. Lennon, K.J. Newberry, and L.K. Reddington. 1998. Does invasion of oligohaline tidal marshes by reed grass, Phragmites australis (Cav.) Trin. Ex Steud., affect the availability of prey resources for the mummichog, Fundulus heteroclitus L.?. Journal of Experimental Marine Biology and Ecology. Vol. 222. pp. 59-77 Ferrigno, F. and D.M. Jobbins. 1968. Open marsh water management. Proceedings of the New Jersey Mosquito Extermination Association. Vol. 55. pp. 104-115 171 Feyrer, F. and M. P. Healey. 2003. Fish community structure and environmental correlates in the highly altered southern Sacramento-San Joaquin Delta. Environmental Biology of Fishes. Vol. 66. pp. 123-132 Fitch, R., T. Theodose, and M. Dionne. 2009. Relationships among upland development, nitrogen, and plant community composition in a Maine salt marsh. Wetlands. Vol. 29, No. 4, pp. 1179-1188 Fogel, B.N., C.M. Crain, and M.d. Bertness. 2004. Community level engineering effects of Triglochin maritima (seaside arrowgrass) in a salt marsh in northern New England, USA. Journal of Ecology. Vol. 92. pp. 589-597 Fogg, J.D. 1983. Recollections of a salt marsh farmer. Historical Society of Seabrook. Seabrook, NH. 89 pp Foster, W.A., J.E. Treherne, P.D. Evans, and C.N.E. Ruscoe. 1979. Short-term changes in activity rhythms in an intertidal arthropod (Acarina: Bdella interrupta Evans). Oecologia. Vol. 38. pp. 291-301 Fries, C.R. 1986. Effects of environmental stressors and immunosuppressants on immunity in Fundulus heteroclitus. American Zoologist. Vol. 26. pp. 271-282 Fry, B. 2006. Stable Isotope Ecology. Springer Science, Business Media, LLC. New York, New York. 308 pp Fry, B., and E. Sherr. 1984. dl3C measurements as indicators of carbon flow in marine and freshwater ecosystems. Contrib. Mar. Sci. Vol. 27. pp. 15-47 Fulton III, R.S. 1985. Predator-prey relationships in an estuarine littoral copepod community. Ecology. Vol. 66. pp. 21-29 Gaston, K.J. 2010. Valuing common species. Science. Vol. 327, Issue 5962. pp. 254-155 Goodman, J.E., M.E. Wood, and W.R. Gehrels. 2007. A 17-yr record of soil accretion in the salt marshes of Maine (USA). Marine Geology. Vol. 242. pp. 109-121 Graham, N.L. and J.G. Stofolano. 1983. Oviposition behavior of the salt marsh greenhead, Tabanus simulans (Diptera: Tabanidae). Annals of the Entomological Society of America. Vol. 76. No. 4. pp. 703-706 Griffin, M.P., and I. Valiela. 2001. 815N isotope studies of life history and trophic position of Fundulus heteroclitus and Menidia menidia. Marine Ecology Progress Series. Vol. 214. pp. 299-305 Grime, J.P. 1979. Plant strategies and plant processes. John Wiley, Chichester, UK. 172 Haas, L.H., C.J. Freeman, J.M. Logan, L. Deegan, and E.F. Gaines. 2009. Examining mummichog growth and movement: are some individuals making intra-season migrations to optimize growth? Journal of Experimental Marine Ecology. Vol. 369. pp. 8-16 Hacker, S.D., and M.D. Bertness. 1995. Morphological and physiological consequences of a positive plant interaction. Ecology. Vol. 76. pp. 2165-2175 Harshberger, J.W. 1911. A hydrometric investigation of the influence of sea water on the distribution of salt marsh and estuarine plants. Proceedings of the American Philosophical Society. Vol. 50. pp. 457-496 Harshberger, J.W. 1916. The origin and vegetation of salt marsh pools. Proceedings of the American Philosophical Society. Vol. 55. pp. 481-484 Hasanudin, U., M. Fujita, Y. Koibuchi, and K. Fujie. 2005. Dynamic changes in environment condition and microbial community structure in trench and flat seabed soils of Tokyo Bay, Japan. Water Science Technology. Vol. 52. No. 9. pp. 107-14 Havill, D.C., A. Ingold, and J. Pearson. 1985. Sulphide tolerance in coastal halophytes. Vegetation. No. 62. pp. 279-285 Headlee, T.J. 1939. Relation of mosquito control to wildlife. New Jersey Mosquito Extermination Association. Vol. 26. pp. 5-12 Hill, M.O. 1973. Diversity and evenness: a unifying notation and its consequences. Ecology. Vol. 54. pp. 427-432 Hopkinson, C.S. and J.P. Schubauer. 1984. Static and dynamic aspects of nitrogen cycling in the salt marsh graminoid Spartina alterniflora. Ecology. Vol. 65, No. 3. pp. 961-969 Hunter, K.L., M.G. Fox, and K.W. Able. 2009. Influence of flood frequency, temperature and population density on migration of Fundulus heteroclitus in semi-isolated marsh pond habitats. Marine Ecology Progress Series. Vol. 391. pp. 85-96 Huston, M. 1979. A general hypothesis of species diversity. American Naturalist. Vol. 113. pp. 81-101 Howarth, R.W., and J.M. Teal. 1979. Sulfate reduction in a New England salt marsh. Limnology and Oceanography. Vol. 24. No. 6.pp. 999-1013 Howes, B.L., J.W. Dacey, and D.D. Goehringer. 1986. Factors controlling the growth form of Spartina alterniflora: feedbacks between above-ground production, soil oxidation, nitrogen and salinity. Journal of Ecology. Vol. 74. pp. 881-898 173 Howes, B.L., R.W. Howarth, J.M. Teal, and I. Valiela. 1981. Oxidation-reduction potentials in a salt marsh: spatial patterns and interactions with primary production. Limnology and Oceanography. Vol. 26. pp. 350-360 Howes, B.L. and D.D. Goehringer. 1994. Porewater drainage and dissolved organic carbon and nutrient losses through the intertidal creek banks of a New England salt marsh. Marine Ecology Progress Series. Vol.114. pp. 289-301 Hruby, T., and W.G. Montgomery. 1988. Open Marsh Water Management for open tidal marshes in the northeast: a manual of methods. Massachusetts Audubon Society Hughes, J.E., L.A. Deegan, B.J. Peterson, R.M. Holmes, and B. Fry. 2000. Nitrogen flow through the food web in the oligohaline zone of a New England estuary. Ecology. Vol. 81. No. 2. pp. 433-452 Hughes, Z.J., D.M. FitsGerald, C.A. Wilson, S.C. Pennings, K.Wieski, and A. Mahadevan. 2009. Rapid headward erosion of marsh creeks in response to relative sea level rise. Geophysical Research Letters. Vol. 36, L03602. 5 pp Hughes, A.R., and J.H. Grabowski. 2006. Habitat context influences predator interference interactions and the strength of resource partitioning. Oecologia. Vol. 149. pp. 256-264 Hunter, K.L., M.G. Fox, and K.W. Able. 2009. Influence of flood frequency, temperature and population density on migration of Fundulus heteroclitus in semi-isolated marsh pond habitats. Marine Ecology Progress Series. Vol. 391. pp. 85-96 Huston, M. 1979. A general hypothesis of species diversity. The American Naturalist. Vol. 113. No. 1. pp. 81-101. James-Pirri, M.J., K.B. Raposa, and J.G. Catena. 2001. Diet composition of mummichogs, Fundulus heteroclitus, from restoring and unrestricted regions of a New England (U.S.A.) salt marsh. Estuarine, Coastal and Shelf Science. Vol. 53. pp. 205-213 James-Pirri, M.J., R.M. Erwin, and D.J. Prosser. 2005. US Fish and Wildlife Service (Region 5) salt marsh study year 4 report (2001 to 2004). US Fish and Wildlife Service, Region 5, Newington, New Hampshire. 03801. 322 pp James-Pirri, M.J., R.M. Erwin, D.J. Prosser, J.D. Taylor. 2011. Responses of salt marsh ecosystems to mosquito control management practices along the Atlantic coast (U.S.A). Restoration Ecology. Society for Ecological Restoration International. DOI: 10.111/j.l 526-100X.2010.00767.x Kaplan, W, I. Valiela, and J.M. Teal. 1979. Denitrification in a salt marsh ecosystem. Limnology and Oceanography. Vol. 24. No. 4. pp. 726-734 174 Kerr, L.A., D.H. Sedor, and R.T. Draus. 2007. Stable isotope (813C and 5180) and Sr/Ca composition of otoliths as proxies for environmental salinity experienced by an estuarine fish. Marine Ecology Progress Series. Vol. 349: pp. 245-253 Kirkpatrick, J. and K. Foreman. 1998. Dissolved inorganic nitrogen flux and mineralization in Waquoit Bay soils as measured by core incubations. Biological Bulletin. Vol. 195. pp. 240-241 Kirwan, M.L., and A. B. Murray. 2007. A coupled geomorphic and ecological model of tidal marsh evolution. Proceedings of the National Academy of Sciences. Vol. 104. No. 15. pp. 6116-6122 Kneib, R.T. 1984. Patterns of invertebrate distribution and abundance in the intertidal salt marsh: causes and questions. Estuaries. Vol. 7. pp. 392-412 Kneib, R.T. 1986. The role of Fundulus heteroclitus in salt marsh trophic dynamics. American Zoologist. Vol. 26. pp. 259-269 Kneib, R.T. 1997. The role of tidal marshes in the ecology of estuarine nekton. Oceanography and Marine Biology: an Annual Review, Vol. 35, pp. 163-220 Kneib, R.T. 2000. Salt marsh ecoscapes and production transfers by estuarine nekton in the southeastern United States. In M.P. Weinstein and D.A. Kreeger, eds. International Symposium: Concepts and Controversies in Tidal Marsh Ecology. Kluwer Academic Publishers, Dordrecht, The Netherlands, pp. 267-291 Kneib, R.T. and R.L. Wagner. 1994. Nekton use of vegetated marsh habitats at different stages of tidal inundation. Marine Ecology Progress Series. Vol. 107, pp. 227-238 Koch, F and C.J. Gobler. 2009. The effects of tidal export from salt marsh ditches on estuarine water quality and plankton communities. Estuaries and Coasts. Vol. 32. pp. 261-275 Koch, M.S., I.A. Mendelssohn, and K.L. McKee. 1990. Mechanism for the hydrogen sulfide-induced growth limitation in wetland macrophytes. Limnology and Oceanography. Vol. 35. No. 2. pp. 399-408 Konisky, R.A., and D.M. Burdick. 2004. Effects of stressors on invasive and halophytic plants of New England salt marshes: a framework for predicting response to tidal restoration. Wetlands. Vol. 24. pp. 434-447 Lajtha, K. and R.H. Michener. 1994. Stable Isotopes in Ecology and Environmental Science. Blackwell Scientific Publications. Cambridge, Massachusetts. 157 pp 175 Lawrence, D.S.L., J.R.L. Allen, and G.M. Havelock. 2004. Salt marsh morphodynamics: an investigation of tidal flows and marsh channel equilibrium. Journal of Coastal Research. Vol. 20. No. 1. pp. 301-316 LeMay, L. 2007. Impact of drainage ditches on salt marsh flow patterns, sedimentation and morphology: Rowley River, Massachusetts. Master's thesis, College of William and Mary, Virginia. 230 pp Lenihan, H.S., and F. Micheli. 2001. Soft-soil communities. In Bertness, M.D., S.D. Gaines, and M.E. Hay (editors). Marine Community Ecology. Sinauer Associates, Inc., Sunderland, Massachusetts, pp. 253-287 Lesser, C. R. 1982 . Standards for Maryland Open Marsh Water Management. (OMWM). Proc. N.J. Mosq. Contr. Assoc. Vol. 70. pp. 29-34 Long, W.C. and R.D. Seitz. 2008. Trophic interactions under stress: hypoxia enhances foraging in an estuarine food web. Marine Ecology Progress Series. Vol. 362. pp. 59-68 Lopez, J.D., M.S. Peterson, E.T. Lang, and A.M. Charbonnet. 2010. Linking habitat and life history for conservation of the rare saltmarsh topminnow Fundulus jenkinsi: morphometries, reproduction, and trophic ecology. Endangered Species Research. Vol. 12. pp. 141-155 Lotrich, V.A. and W.H. Meredith. 1974. Technique and effectiveness of various acrylic colors for subcutaneous marking of fish. Transactions of the American Fisheries Society. Vol. 103. No. l.pp. 140-142 Luxton, M. 1967. The ecology of saltmarsh Acarina. Journal of Animal Ecology. Vol. 36. No. 2. pp. 257-277 MacKenzie, R.A., and M. Dionne. 2008. Habitat heterogeneity: importance of salt marsh pools and high marsh surfaces to fish production in two Gulf of Maine salt marshes. Marine Ecology Progress Series. Vol. 368. pp. 217-230 Magurran 2004. Measuring biological diversity. Princeton University Press. Maricle, B.R., and R.W. Lee. 2006. Effects of environmental salinity on carbon isotope discrimination and stomatal conductance in Spartina grasses. Marine Ecology Progress Series. Vol. 131. pp. 305-310 Martinez, N.D. 1992. Constant connectance in community food webs. The American Naturalist. Vol. 139. No. 6. pp. 1208-1218 Manatunge, J., T. Asaeda, and T. Priyadarshana. 2000. The influence of structural complexity on fish-zooplankton interactions: a study using artificial submerged macrophytes. Environmental Biology of Fishes. Vol. 58. pp. 425-438 176 Mayor, J.R. and C.E. Hicks. 2009. Potential impacts of elevated CO2 on plant interactions, sustained growth, and carbon cycling in salt marsh ecosystems. In Silliman, B.R., E.D. Groshlz, and M.D. Bertness. 2009. Human impacts on salt marshes, a global perspective. University of California Press, pp. 207 - 228 McMahon, K.W., BJ. Johnson, and W.G. Ambrose, Jr. 2005. Diet and movement of the killifish, Fundulus heteroclitus, in a Maine salt marsh assessed using gut contents and stable isotope analyses. Estuaries. Vol. 28. pp. 966-973 Mendelssohn, I. A. 1979a. Nitrogen metabolism in the height forms of Spartina alterniflora in North Carolina. Ecology. Vol. 60. pp. 575-584 Mendelssohn, I. A. 1979b. The influence of nitrogen level, form, and application method on the growth response of Spartina alterniflora in North Carolina. Estuaries. Vol. 2. pp. 106-112 Mendelssohn, I.A., K.L. McKee, and W.H. Patrick.1981. Deficiency in Spartina alterniflora roots: Metabolic adaptation to anoxia. Science. Vol. 214. No. 4519. pp. 439- 441 Mendelssohn, I.A. and K.L. McKee. 1988. Spartina alterniflora die-back in Louisiana: time-course investigation of sediment waterlogging effects. Journal of Ecology. Vol. 76. 509-521 Menge, B.A. and A.M. Olson. 1990. Role of scale and environmental factors in regulation of community structure. Trends in Ecology and Evolution. Vol. 5. No. 2. pp. 52-57 Meredith, W.H., D.E. Saveikis, and C.J. Stachecki. 1985. Guidelines for "open marsh water management" in Delaware's salt marshes - objectives, system designs, and installation procedures. Wetlands. Vol. 5. pp. 119-133 Michener R., and K. Lajtha. 2007. Stable isotopes in ecology and environmental science. Blackwell Publishing LTD. Maiden, Massachusetts. Second Edition. 566 pp Miller, W.B. and F.E. Egler. 1950. Vegetation of the Wequetquock-Pawtucuk tidal marshes, Connecticut. Ecological Monographs. Vol.20, pp. 143-172 Minagawa, M. and E. Wada. 1984. Setpwise enrichment of 15N along food chains: further evidence and the relation between 8I5N and animal age." Geochimica et Cosmochimica Acia. Vol. 48. pp. 1135-1140 Minkoff, D.R., M. Escapa, F.E. Ferramola, S.D. Maraschin, J.O. Pierini, G.M.E. Perillo, and C. Delrieux. 2006. Effects of crab-halophytic interactions on creek growth in a S.W. 177 Atlantic salt marsh: a cellular automata model. Estuarine, Coastal and Shelf Science. Vol. 69. pp. 403-413 Mitsch, W.J. and J.G. Gosselink. 2000. Wetlands. John Wiley and Sons, Inc., New York. 920 pp. Morgan P. A., and F.T. Short. 2002. Using functional trajectories to track constructed salt marsh development in the Great Bay Estuary, Mane/New Hampshire, U.S.A. Restoration Ecology. Vol 10. No. 3. pp. 461-473 Morris, J.T., P.V. Sundareshwar, C.T. Nietch, B. KjerfVe, and D.R. Cahoon. 2002. Responses of coastal wetlands to rising sea level. Ecology. Vol. 83. No. 10. pp. 2869- 2877 Montoya, J.M., and R.V. Soule. 2002. Small world patterns in food webs. Journal of Theoretical Biology. Vol. 214. pp. 405-412 Montoya, J.M., M.A. Rodriguez, and B.A. Hawkins. 2003. Food web complexity and higher-level ecosystem services. Ecology Letters. Vol. 6. pp. 587-593 Moody, R.M., and R.B. Aronson. 2007. Trophic heterogeneity in salt marshes of the northern Gulf of Mexico. Marine Ecology Progress Series. Vol. 331. pp. 49-65 Mudd, S.M., S.M Howell, and J.T. Morris. 2009. Impact of dynamic feedbacks between sedimentation, sea-level rise, and biomass production on near-surface marsh stratigraphy and carbon accumulation. Estuarine, Coastal and Shelf Science Vol. 82. pp. 377-389 Mullan Crain, C., M.D. Bertness. 2006. Ecosystem engineering across environmental gradients: implications for conservation and management. Bioscience. Vol. 56. No. 3. pp. 211-218 Mullan Crain, C, K. Bromberg Gedan, and M. Dionne. 2009. Tidal restrictions and mosquito ditching in New England marshes. In Silliman Et al., 2009. Human impact on salt marshes: a global perspective. University of California Press. Berkeley, pp. 149-169 Mullan Crain, C., B.R. Silliman, S.L. Bertness, and M.D. Bertness. 2004. Physical and biotic drivers of plant distribution across estuarine salinity gradients. Ecology. Vol. 85. No. 9. pp. 2539-2549 Munsell Color. 2000. Munsell Soil Color Charts. Greta Macbeth, New Windsor, NY Murphy, B.R. and D.W. Willis. 1996. Fisheries Techniques, second edition. American Fisheries Society, Bethesda, Maryland. 732pp 178 Naeem, S., L.J. Thompson, S.P. Lawlor, J.H. Lawton, and R.M. Woodfin. 1994. Declining biodiversity can alter the performance of ecosystems. Nature. Vol. 368. pp. 734-737 Neckles H., and M. Dionne. 1999. Regional standards to identify and evaluate tidal wetland restoration in the Gulf of Maine. A GPAC Workshop, June 2-3, 1999. Wells National Estuarine Research Reserve. Wells, Maine. 21 pp. plus appendices Niedowski, N.L. 2000. New York State salt marsh restoration and monitoring guidelines. New York State Department of Environmental Conservation, Division of Fish, Wildlife and Marine Resources, Bureau of Marine Resources. East Setauket, NY. 147 pp Nichols, W.F., K.T. Killngbeck, and P.V. August. 1998. The influence of geomorphological heterogeneity on biodiversity: II. A landscape perspective. Conservation Biology. Vol. 12. No. 2. pp. 371-379 Nixon, S.W. 1982. The ecology of New England high salt marshes: a community profile. U.S. Fish and Wildlife Service, Office of Biological Services, Washington, DC. FWS/OBS-81/55, 70 pp Nixon, S.W., J.W. Ammerman, L.P. Atkinson, V.M. Berounsky, G Billen, W.C. Boicourt, W.R. Boynton, T.M. Church, D.M. Ditoro, R. Elmgren, J.H. Garber, A.E. Giblin, R.A. Jahnke, N.J.P. Owens, M.E.Q. Pilson, and S.P. Seitzinger. 1996. The fate of nitrogen and phosphorus at the land-sea margin of the North Atlantic ocean. Biogeochemistry. Vol. 35. pp. 141-180 Northeastern Massachusetts Mosquito Control and Wetlands Management District. 2011. Standard for Open Marsh Water Management, OMWM. Newburyport, Massachusetts. 12pp Nuttle, W.K. 1988. The extent of lateral water movement in the soils of a New England Salt Marsh. Water Resources Research. Vol. 24. pp. 2077-2085 Nyman, J.A., R.D. DeLaune, S.R. Pezeshki, W.H. Patrick, Jr. 1995. Organic matter fluxes and marsh stability in a rapidly submerging estuarine marsh. Estuaries. Vol. 18. No. IB. pp. 207-218 Odum, E.P. 1975. Ecology: the link between the natural and the social sciences. Second Edition. Hold, Rinehart and Winston, Inc. New York. 244pp Odyssey Data Recording. 2005. Data flow systems PTY limited. Burnside, Christchurch, New Zealand, [email protected] Paine, R.T. 1966. Food web complexity and species diversity. The American Naturalist. Vol. 100. pp. 65-75 179 Paine, R.T. 1980. Food webs: linkage, interaction strength and community infrastructure. Journal of Animal Ecology. Vol. 49. pp. 667-685 Palmer, A. and S. Filoso. 2009. Restoration of ecosystem services for environmental markets. Science. Vol. 325. pp. 575-576 Peterson, B.J, R.W. Howarth, and R.H. Garritt. 1985. Multiple stable isotopes used to trace the flow of organic matter in estuarine food webs. Science, New Series. Vol. 227. No. 4692. pp. 1361-1363 Peterson, B.J., R.W. Howarth, and R.H. Garritt. 1986. Sulfur and carbon isotopes as tracers of salt-marsh organic matter flow. Ecology. Vol. 67. No. 4. pp. 865-874 Peterson, B. J., and B. Fry. 1987. Stable isotopes in ecosystem studies. Annual Review of Ecology and Systematics. Vol.18, pp. 293-320 Phillips, D.L. and J.W. Gregg. 2001. Uncertainty in source partitioning using stable isotopes. Oecologia Vol. 127. pp. 171-179 Portnoy, J.W. and A.E. Giblin. 1997. Biogeochemical effects of seawater restoration to diked salt marshes. Ecological Applications. Vol. 7. No. 3. pp. 1054-1063 Posey, H.M., and A.H. Hines. 1991. Complex predator-prey interactions within an estuarine benthic community. Ecology. Vol. 72. pp. 2155-2169 Provost, M.W. 1974. Salt marsh management in Florida. Tall Timbers Conference on Ecological Animal Control by Habitat Management. Tall Timbers Research Station, Tallahassee, Florida. Vol. 5. pp 5-17 Rader, D.N. 1984. Salt-marsh benthic invertebrates: small-scale patterns of distribution and abundance. Estuaries. Vol. 7. No. 4A. pp. 413-420 Rakocinski, C.F., D.M. Baltz, and J.W. Fleeger. 1992. Correspondence between environmental gradients and the community structure of marsh-edge fishes in a Louisiana estuary. Marine Ecology Progress Series. Vol. 80. pp. 135-148 Raposa, K. 2003. Overwintering habitat selection by the mummichog, Fundulus heteroclitus, in a Cape Cod (USA) salt marsh. Wetlands Ecology and Management. Vol. 11. pp. 175-182 Ratz, H-J. and J. Lloret. 2003. Variation in fish condition between Atlantic cod (Gadus morhaua) stocks, the effect on their productivity and management implications. Fisheries Research. Vol. 60. pp. 369-380 Redfield, A.C. 1965. Ontogeny of a salt marsh estuary. Science. Vol. 147. pp.50-55 180 Redfield, A.C. 1972. Development of a New England salt marsh. Ecological Monographs. Vol. 42. pp. 201-237 Reed, D.J., and D.R. Cahoon. 1992. The relationship between marsh surface topography, hydroperiod, and growth of Spartina alterniflora in a deteriorating Louisiana salt marsh. Journal of Coastal Research. Vol. 8. No. 1. pp. 77-87 Rodil, I.F., S. Cividanes, M. Lastra, and J. Lopez. 2008. Seasonal variability in the vertical distribution of benthic macrofauna and sedimentary organic matter in an estuarine beach (NW Spain). Estuaries and Coasts. Vol. 31 pp. 382-395 Roman, C.T., N.T. Jaworski, F.T. Short, S. Findlay, and R.S. Warren. 2000. Estuaries of the northeastern United States: habitat and land use signatures. Estuaries. Vol. 23. No. 6. pp. 743-764 Roman, C.T., K.B. Raposa, S.C. Adamowicz, M.J. James-Pirri, J.G. Catena. 2002. Quantifying vegetation and nekton response to tidal restoration of a New England salt marsh. Restoration Ecology, Vol.10, No.3, pp. 450-460 Rozema, J., E. Luppes, and R. Broekman. 1985. Differential response of salt-marsh species to variation of iron and manganese. Vegetation. Vol. 62. pp. 293-301 Rysgaard, S. P. Thastum, T. Dalsgaard, P.B. Christensen, and N.P. Sloth. 1999. Effects of salinity on NH/ adsorption capacity, nitrification, and denitrifxcation in Danish estuarine soils. Estuaries. Vol. 22. No. 1. pp. 21-30 SAS Institute .2010. JMP 9 statistical software. Cary, NC Schudlark, J.R. and T.M. Chruch. 1989. The sedimentary flux of nutrients at a Delaware salt marsh site: a geochemical perspective. Biogeochemistry. Vol. 7. pp. 55-75 Sebold, K.R. 1998. The low green prairies of the sea: economic usage and cultural construction of the Gulf of Maine salt marshes. Ph.D. Dissertation. University of Maine, Orono, Maine Seliskar, D.M., J.L. Gallagher, D.M. Burdick, and L.A. Mutz. 2002. The regulation of ecosystem functions by ecotypic variation in the dominant plant: a Spartina alterniflora salt-marsh case study. Journal of Ecology. Vol. 90. pp. 1-11 Semmens, B.X., and J.W. Moore. 2008. MixSIR: A Bayesian stable isotope mixing model, Version 1.0. http://www.ecologybox.org. Date of download September 2009 Short, F.T., D.M. Burdick, C.A. Short, R.C. Davis, and P.A. Morgan. 2000. Developing success criteria for restored eelgrass, salt marsh and mud flat habitats. Ecological Engineering. Vol. 15. Issues 3-4. pp. 239-252 181 Silliman, B.R., and M.D. Bertness. 2002. A trophic cascade regulates salt marsh primary production. Proceedings of the National Academy of Sciences of the United States of America. Vol. 99. pp. 10500-10505 Silliman, B.R., E.D. Grosholz, and M.D. Bertness. 2009. Human impacts on salt marshes: a global perspective. University of California Press. Berkeley. 413 pp Smith, K.J., and K.W. Able. 1994. Salt-marsh tide pools as winter refuges for the mummichog, Fundulus heteroclitus, in New Jersey. Estuaries. Vol. 17. No. IB. pp. 226- 234 Smith, K.J. and K.W. Able. 2003. Dissolved oxygen dynamics in salt marsh pools and its potential impacts on fish assemblages. Marine Ecology Progress Series. Vol. 258. pp. 223-232 Soule, M.E. 1986. Conservation Biology: the science of scarcity and diversity. Sinauer Associates, Inc. Sunderland, MA. 584 pp Speckman, S.G., J.F. Piatt, C.V. Minte-Vera, and J.K. Parrish. 2005. Parallel structure among environmental gradients and three trophic levels in a subarctic estuary. Progress in Oceanography. Vol. 66. pp. 25-65 Spectrum Technologies, Inc. 2005. Field Scout SC-900 soil compaction meter. Plainfield, IL Stefanon, L., L. Carniello, A. D'Alpaos, and S. Lanzoni. 2010. Experimental analysis of tidal network growth and development. Continental Shelf Research. Vol. 30. pp. 950-962 Steneck, R.S. and J.T. Carlton. Human alterations of marine communities. In Bertness, M.D., S.D. Gaines, and M.E. Hay (editors). Marine Community Ecology. Sinauer Associates, Inc., Sunderland, Massachusetts, pp. 445-468 Sterner, R.W., J.J. Elser, E.J. Fee, S.J. Guilford, and T.H. Chrzanowski. 1997. The light: nutrient ratio in lakes: the balance of energy and materials affects ecosystem structure and process. The American Naturalist. Vol. 150. No. 6. pp663-684 Stevenson, J.C. and M.S. Kearney. 2009. Impacts of global climate change and sea-level rise on tidal wetlands. In Silliman, B.R., E.D. Groshlz, and M.D. Bertness. 2009. Human impacts on salt marshes, a global perspective. University of California Press, pp. 171 - 206 Stribling, J.M., O.A. Glahn, X.M. Chen, and J.C. Cornwell. 2006. Microtopographic variability in plant distribution and biogeochemistry in a brackish-marsh system. Marine Ecology Progress Series. Vol. 320. pp. 121-129 182 Stribling, J.M., J.C. Cornwell, and O.A. Glahn. 2007. Microtopography in tidal marshes: ecosystem engineering by vegetation? Estuaries and Coasts. Vol. 30. No. 6. pp. 1007- 1015 Strong, D.R. 1992. Are trophic cascades all wet? Differentiation and donor-control in speciose ecosystems. Ecology. Vol. 73. pp. 747-754 Stumpf, R.P. 1983. The process of accretion on the surface of a salt marsh. Estuarine, Coastal and Shelf Science. Vol. 17. pp. 495-508 Svensson, C.J., G.A. Hyndes, and P.S. Lavery. 2007. Food web analysis in two permanently open temperate estuaries: consequences of saltmarsh loss? Marine Environmental Research. Vol. 64. No. 3. pp. 286-304 Taylor, J. 1998. Guidance for meeting U.S. Fish and Wildlife Service trust resource needs when conducting coastal management for mosquito control on Region 5 national wildlife refuges. U.S. Fish and Wildlife Service, Region 5 Office. Hadley, Massachusetts. 19 pp Teal, J.M. 1962. Energy flow in the salt marsh ecosystem of Georgia. Ecology. Vol. 43. pp. 614-624 Teal, J., and M. Teal. 1969. Life and Death of the Salt Marsh. Ballantine Books. New York, NY. 274 pp Teo, S.L.H. and K.W. Able. 2003. Habitat use and movement of mummichog (Fundulus heteroclitus) in a restored salt marsh. Estuaries. Vol. 26, No. 3. pp 720-730 Thebault, E. and M. Loreau. 2003. Food-web constraints on biodiversity-ecosystem functioning relationships. Proceedings of the National Academy of Science. Vol. 100. No. 25. pp. 14949-14954 Thompson, R.M. and C.R. Townsend. 2005. Energy availability, spatial heterogeneity and ecosystem size predict food-web structure in streams. Oikos. Vol. 108. pp. 137-148 Tiner, R.W. 1987. A field guide to coastal wetland plants of the northeastern United States. University of Massachusetts Press, Amherst, MA. 285 pp Tiner, R.W. 1999. Wetland Indicators: A Guide to Wetland Identification, Delineation, Classification, and Mapping. Lewis Publishers. Washington, D.C. 392 pp Tobias, C.R., M. Cieri, B.J. Peterson, L.A. Deegan, J. Vallino, and J. Hughes. 2003. Processing watershed-derived nitrogen in a well-flushed New England estuary. Limnology and Oceanography. Vol. 48. No. 5. pp. 1766-1778 Town of East Hampton, New York. 2011. http://www.town.east-hampton.ny.us/ 183 Travis, B.V., G.H. Bradley, and W.C. McDuffie. 1954. The effect of ditching on salt marsh vegetation in Florida. New Jersey Mosquito Extermination Association. Vol. 41. pp. 235-244 Turner, R.E, E. M. Swenson and C. S. Milan. 2000. Organic and Inorganic Contributions to Vertical Accretion in Salt Marsh Soils. In, Concepts and Controversies in Tidal Marsh Ecology, Michael P. Weinstein and Daniel A. Kreeger editors, pp. 583-595 Turner, R.E. 2010. Beneath the salt marsh canopy: loss of sediment strength with increasing nutrients. Estuaries and Coasts. DOI 10.1007/s12237-010-9341 -y Valiela, I., and J.M. Teal. 1979. The nitrogen budget of a salt marsh ecosystem. Nature. Vol. 280. pp. 652-656 Valiela, I., J. Costa, K. Foreman, J.M. Teal, B. Howes, D. Aubrey. 1990. Transport of groundwater-borne nutrients from watersheds and their effects on coastal waters. Biogeochemistry. Vol. 10. no. 3. Groundwater inputs to coastal waters (Aug, 1990). pp. 177-197 Vander Zaden, M.J. and J.B. Rasmussen. 2001.Variation in 815N and 813C trophic fractionation: implications for aquatic food web studies. Limnology and Oceanography. Vol. 46. No. 8. pp. 2061-2066 Vince, S. I. Valiela, and N. Backus. 1976. Predation by salt marsh killifish Fundulus heteroclitus (L) in relation to prey size and habitat structure: consequences for prey distribution and abundance. Journal of Experimental Marine Biology and Ecology. Vol. 23. No. 3. pp. 255-266 Vincent, R. 2005. Personal observations of Fundulus heteroclitus feeding directly on Arachnida spp. and Coleoptera spp. that occurred on the surface of New England salt marsh pools Vob , M., G. Nausch, and J.P. Montoya. 1997. Nitrogen stable isotope dynamics in the central Baltic Sea: influence of deep-water renewal on the N-cycle changes. Marine Ecology Progress Series. Vol. 158. pp. 11-21 Waggy, G. L., M. S. Peterson, and B.H. Comyns. 2007. Feeding habits and mouth morphology of young silver perch (Bairiella chrysoara) from the north-central Gulf of Mexico. Southeastern Naturalist. Vol. 6. No. 4. pp. 743-751 Wei, L., C. Yan, B. Ye, and X. Guo. 2008. Effects of salinity on leaf 513C in three dominant mangrove species along salinity gradients in an estuarine wetland, southeast China. Journal of Coastal Research. Vol. 24. No. 1. pp. 267-272 184 Weinstein, M.P. and S.Y. Litvin. 2000. The role of tidal salt marsh as an energy source for marine transient and resident finfishes: a stable isotope approach. Transactions of the American Fisheries Society. Vol. 129. pp. 797-810 Wetzel, R.G., and G.E. Likens. 1991. Limnological Analyses. 2nd. Ed. Springer-Verlag. 391 pp Weisberg, S.B., and V.A. Lotrich. 1982. The importance of an infrequently flooded intertidal marsh surface as an energy source for the mummichog, Fundulus heteroclitus: an experimental approach. Marine Biology. Vol. 66. pp. 307-310 Wigand, C. , R.A. McKinney, M.A. Charpentier, M.M. Chintala, and G.B. Thursby. 2003. Relationships of nitrogen loadings, residential development, and physical characteristics with plant structure in New England salt marshes. Estuaries. Vol. 26. No. 6. pp. 1494-1504 Wigand, W., R.A. McKinney, M.L. Cole, G.B. Thursby, and J. Cummings. 2007. Varying stable nitrogen isotope ratios of different coastal marsh plants and their relationships with wastewater nitrogen and land use in New England, USA. Environmental Monitoring and Assessment. Vol. 131. pp. 71-81 Wigand, C., P. Brennan, M. Stolt, M. Holt, and S. Ryba. 2009. Soil respiration rates in coastal marshes subject to increasing watershed nitrogen loads in southern New England, USA. Wetlands. Vol. 29. No. 3. pp. 952-963 Whitcraft, C.R. and L.A. Levin. 2007. Regulation of benthic algal and animal communities by salt marsh plants: impacts of shading. Ecology. Vol. 88. No. 4. pp. 904- 917 Whittaker, R.H. 1956. Plantof the great Smokey Mountains. Ecological Monographs. Vol. 26. No. 1. pp. 1-80 Williams, L., G.C. Noblitt IV, and R. Buchsbaum. 2001. The effects of salt marsh haying on benthic algal biomass. Biological Bulletin. Vol. 201. pp. 287-288 Williams, P.B. and M.K. Orr. 2002. Physical evolution of restored breached levee salt marshes in the San Francisco Bay estuary. Restoration Ecology. Vol. 10. No. 3. pp. 527- 542 Williams, R.J. and N.D. Martinez. 2000. Simple rules yield complex food webs. Nature. Vol. 404. pp. 180-183 Williams, R.J., E.L. Berlow, J.A. Dunne, S.L. Barabasi, and N.D. Martinez. 2002. Two degrees of separation in complex food webs. Proceedings of the National Academy of Science. Vol. 99. No. 20. pp. 12913-12916 185 Wilson, M.A., R. Costanza, R. Boumans, and S. Liu. 2005. Integrated assessment and valuation of ecosystem goods and services provided by coastal systems. Pp. 1-24. In: Wilson, J.G. (ed.). The Intertidal Ecosystem. Royal Irish Academy Press, Dublin Wilson, K.R., J.T. Kelley, A. Croitoru, M. Dionne, D.F. Belknap, and R. Steneck. 2009. Stratigraphic and ecophysical characterizations of salt marsh pools: dynamic landforms of the Webhannet salt marsh, Wells, ME, USA. Estuaries and Coasts. Vol. 32. pp. 855- 870 Wolfe, R.J. 1996. Effects of open marsh water management on selected tidal marsh resources: a review. Journal of the American Mosquito Control Association. Vol. 12. pp. 701-712 Wozniak, A.S., C.T. Roman, S.C. Wainright, R.A. McKinney, and M.J. James-Pirri. 2006. Monitoring food web changes in tide-restricted salt marshes: a carbon stable isotope approach. Estuaries and Coasts. Vol. 29. No. 4. pp. 568-578 Yapp, R.H., D. John, and O.T. Jones. 1917. The salt marshes of Dovey Estuary. Journal of Ecology. Vol. 5. pp. 65-103 YSI85 Handheld Multimeter. 2005. YSI corporation, Yellow Springs, OH 186 APPENDIX A MEAN VALUES (± SE) BY MARSH AND HABITAT FOR HYDROLOGY, MARSH SURFACE ELEVATION, AND SOIL CHARACTERISTICS Hydrology • Created Ditch • Natural Creek S Ditch Plug • Natural Pool Parker River Chauncey Creek Sprague River Soil Strength Created Ditch • Natural Creek S Ditch Plug • Natural Pool 1 Parker River Chauncey Creek Sprague River 187 Soil Strength (continued) Parker River Marsh - Newburyport, MA 10 9 i 8 7 * -A £o> 6 ' ' • c ^ —2 -i • — s 5 CO 4 —•— Created Ditch 3 -i * * Natural Creek 2 "8 Ditch Plug CO 1 Natural Pool 0 • — r i i —r 1 1 1 r i ~i r 1 1—' 1 1 1 1 1 \ 0 3 5 8 10 13 15 18 20 23 25 28 30 33 36 38 41 43 46 Depth below marsh surface (cm) Chauncey Creek Marsh - Kittery, ME 10 - 9 - 8 - O) 7 - £ 6 - co> $ 5 - CO c 4 - a> 3 - E Created Ditch T30> ? - Ditch Plug CO 1 - Natural Creek 0 - Natural Pool 8 10 13 15 18 20 23 25 28 30 33 36 38 41 43 46 Depth below marsh surface (cm) Sprague River Marsh - Phippsburg, ME Created Ditch Natural Creek Ditch Plug _i 1 1 1 1 1 1 1 1 1 1 1 r 1 Natural1 1 Poolr 3 5 8 10 13 15 18 20 23 25 28 30 33 36 38 41 43 46 Depth below marsh surface (cm) 188 Marsh Surface Elevations Parker River Marsh - Newburyport, MA =§= Created Ditch Ditch Plug Natural Creek Natural Pool 2 7 15 Distance from edge of water feature (m) Chauncey Creek Marsh - Kittery, ME ^ —zjjt J I 120 00 CO <3 110 * -——: 100 1 90 O / 80 Created Ditch x; -/ Ditch Plug to 70 Natural Creek 5 Natural Pool 60 2 7 15 20 Distance from edge of water feature (m) Sprague River Marsh - Phippsburg, ME Created Ditch Ditch Plug Natural Creek •Y- Natural Pool 7 15 20 Distance From Water Feature (m) 189 Soil Redox Potential Parker River Marsh - Newburyport, MA 150 100 1 50 I T 1 0 1 -50 1 T 5 -100 1 -150 • Created Ditch -200 • • Natural Creek -250 - Ditch Plug Natural Pool -300 Chauncey Creek Marsh - Kittery, ME 150 100 50 0 • I > T -50 < t 1 x: HI -100 -150 • Created Ditch -200 * fc • Natural Creek -250 k Ditch Plug Natural Pool -300 Sprague River Marsh - PWppsburg, ME 150 T 100 1 1 50 <• 0 > -50 S "100 -150 • Created Ditch -200 - • Natural Creek 4 s> -250 - a Ditch Plug Natural Pool -300 190 Soil Pore-water Salinity • Created Ditch 1 1 10 Natural Creek I I 1 H Ditch Plug (0 1 1 • Natural Pool w 15 1 1 I 1 1 i 1 1 1 I Parker River Chauncey Creek Sprague River Soil Percent Organic Content • Created Ditch 1 • Natural Creek • Ditch Plug 1 • Natural Pool 1 1 1 1 1 1 1 1 1 1 1 1 I I Parker River Chauncey Creek Sprague River 191 Soil Carbon Storage 1 • Created Ditch l 1 (0 Natural Creek 1 I 1 • Ditch Plug I I I • Natural Pool I 1 1 i 1 I 1 I I ! 1 I I i Parker River Chauncey Creek Sprague River Soil Bulk Density • Created Ditch •=- 500 • Natural Creek i • Ditch Plug 1 1 1 • Natural Pool | 300 1 1 Parker River Chauncey Creek Sprague River 192 Soil Cores Ditch 193 Ditch-plug Natural Pool 194 APPENDIX B PLANT CHARACTERISTICS BY MARSH Emergent Plant Biomass Emergent Vegetation Biomass by Habitat and Marsh Parker • Chauncey Created Ditch Natural Creek Ditch Plug Natural Pool 195 Vegetation Biomass Parker River Marsh - Newburyport, Massachusetts Distance From Water Feature 0 Meters •2 Meters • 7 Meters B15 Meters •20 Meters Created Ditch Natural Creek Ditch Plug Natural Pool Vegetation Biomass Chauncey Creek Marsh - Kittery, Maine Distance From Water Feature • 0 Meters • 2 Meters • 7 Meters • 15 Meters ~ 60 •20 Meters Created Ditch Natural Creek Ditch Plug Natural Pool 196 Vegetation Biomass Sprague River Marsh - Phippsburg, Maine 90 Distance From Water 80 Feature •0 Meters 70 •2 Meters • 7 Meters 60 S15 Meters • 20 Meters 50 40 30 20 10 0 CreatedI Ditch Natural Creek Ditch Plug Natural Pool Emergent Plant Cover and Composition • Parker River • Chauncey Creek E3 Sprague River 55 50 Created Ditch Natural Creek Ditch Plug Natural Pool 197 Parker River Marsh Vegetation Areal Cover and Composition Created Ditch Habitat Parker River Marsh - Newburyport, Massachusetts • Salicornia europaea •Atriplex patula • Plantago maritima • Limonium nashii • Triglochin maritimum 0) 50 • Puccinellia maritima • S. alterniflora 20-50cm • Tall S. alterniflora >50cm • D. spicata • S. patens Distance from Water Feature (m) Vegetation Areal Cover and Composition Natural Creek Habitat Parker River Marsh - Newburyport, Massachusetts • Salicornia europaea •Atriplex patula • Plantago maritima • Limonium nashii •Triglochin maritimum • Glaux maritima • Puccinellia maritima • S. alterniflora 20-50cm •Tall S. alterniflora >50cm • D. spicata • J. gerardii • S. patens Distance from Water Feature (m) 198 Vegetation Areal Cover and Composition Ditch Plug Habitat Parker River Marsh - Newburyport, Massachusetts • Salicornia europaea • Plantago maritima • Triglochin maritimum • Glaux maritima 2 7 15 Distance from Water Feature (m) Vegetation Areal Cover and Composition Natural Pool Habitat Parker River Marsh - Massachusetts • Salicornia europaea •Atriplex patula • Plantago maritima • Limonium nashii •Triglochin maritimum B Puccinellia maritima • S. alterniflora 20-50cm •Tall S. alterniflora >50cm • D. spicata • J. gerardii • S. patens 2 7 15 Distance from Water Feature (m) 199 Chauncev Creek Marsh Vegetation Areal Cover and Species Composition Created Ditch Habitat Chauncey Creek Marsh - Kittery, Maine • Sueda sp. • Saiicomia europaea • Atriplex patuia • Plantago maritimum • Limonium nashii •Triglocbin maritima • Glaux maritima • Short S. altemiflora <20cm • Tall S. altemiflora >50cm • D. spicata • J. gerardii • S. patens Distance From Water Feature (m) Vegetation Areal Cover and Species Composition Natural Creek Habitat Chauncey Creek Marsh - Kittery, Maine • Salicornia europaea • Atriplex patuia • Triglochin maritima @ Glaux maritima • Puccinellia maritima • Short S. altemiflora <20cm • Tall S. altemiflora >50cm • D. spicata PJ. gerardii • S. patens Distance From Water Feature (m) 200 Vegetation Areal Cover and Species Composition Ditch Plug Habitat Chauncey Creek Marsh - Kittery, Maine 100 90 80 • Salicomia europaea 70 QAtriplex patula • Plantago maritimum 60 g •Triglochin maritima I • Puccinellia maritima sT 50 h • Short S. altemiflora <20cm 40 • S. altemiflora 20-50cm •Tall S. altemiflora >50cm 30 • D. spicata • J. gerardii 20 • S. patens 10 0 II2 7 15 20 Distance From Water Feature (m) Vegetatoin Areal Cover and Species Composition Natural Pool Habitat Chauncey Creek Marsh - Kittery, Maine • Salicomia europaea • Atriplex patula • Plantago maritimum •Triglochin maritima • Short S. altemiflora <20cm • S. altemiflora 20-50cm • D. spicata • J. gerardii • S. patens Distance From Water Feature (m) 201 Sprague River Marsh Vegetation Areal Cover and Composition Created Ditch Habitat Sprague River Marsh - Phippsburg, Maine • Sueda sp. • Salicomia europaea •Atriplex patula • Plantago maritima • Limonium nashii •Triglochin maritimum • Glaux maritima OPuccinellia maritima • Tall S. alterniflora >50cm • D. spicata • J. gerardii • S. patens 2 3 4 Distance from Water Feature (m) Vegetation Areal Cover and Composition Natural Creek Habitat Sprague River Marsh - Phippsburg, Maine • Sueda sp. • Salicomia europaea • Atriplex patula • Limonium nashii • Triglochin maritimum • Glaux maritima • S. alterniflora 20-50cm • Tall S. alterniflora >50cm • D. spicata • J. gerardii • S. patens Distance from Water Feature (m) 202 Vegetation Areal Cover and Composition Ditch Plug Habitat Sprague River Marsh - Phippsburg, Maine 100 90 80 70 60 • Tall S. altemiflora >50cm • D. spicata 50 PJ. gerardii MS. patens 40 30 20 10 s • 0 2 7 15 20 Distance from Water Feature (m) Vegetation Areal Cover and Composition Natural Pool Habitat Sprague River Marsh - Phippsburg, Maine B Sueda sp. • Salicornia europaea • Atriplex patula • Limonium nashii • Plantago maritima • Triglochin maritimum • Glaux maritima • Puccinellia maritima • S. altemiflora 20-50cm • D. spicata • J. gerardii • S. patens 2 7 15 Distance from Water Feature (m) 203 Emergent Plant Richness and Diversity • Parker River • Chauncey Creek E3 Sprague River Created Ditch Natural Creek Ditch Plug Natural Pool 1.2 • Parker River • Chauncey Creek B Sprague River 0.8 co JC CO S 0.6 - &w 0.2 Created Ditch Natural Creek Ditch Plug Natural Pool 204 APPENDIX C INVERTEBRATE RICHNESS AND ABUNDANCE BY HABITAT, LOCAGTION, AND TERATMENT: MOODY MARSH, WELLS, MAINE Natural Controls Invertebrate Richness & Abundance Ditch Plug - Marsh Surface Natural Control • Acarina A • Hemiptera Delphacidae a Diptera Ephydridae Ephydra • Diptera Sciomyzidae • Diptera Chironomidae • Hemiptera Salidae 44% • Diptera Ceratopogonidae B Mesogatropoda 11% Hydrobiinae Hydrobia • Diptera Ephydridae • Oligocheate 20% 205 Invertebrate Richness & Abundance Ditch Plug Within-Pool Natural Control • Acarina A • Acarina B • Hemiptera Delphacidae 1Coleaoptera Carabidae Invertebrate Richeness & Abundance Natural Pool - Marsh Surface Natural Control lAcarina A • Acarina B B Coleaoptera Carabidae Geopinus • Amphipoda Orchestra • Coleaoptera Carabidae 13 Diptera Ceratopogonidae 93% 206 Invertebrate Richness & Abundance Natural Pool Within-Pool Natural Control I Acarina A 4%1 4% 4%- l Coleaoptera Carabidae 8%A I Acarina B 8% • Hemiptera Salidae 60% • Coleoptera 12%^B • Coleaoptera Carabidae Geopinus • Hemiptera Salidae Saluda 207 Ditch-plug Habitat Procedural Treatments Invertebrate Richness & Abunance Ditch Plug Marsh Surface Procedural Control • Acarina A 0 Diptera Ephydridae Ephydra • Diptera Tipulidae • Foraminifera 94% Invertebrate Richness & Abundance Ditch Plug Marsh Surface Exclosure • Acarina A • Diptera Ephydridae Ephydra • Diptera Canaceidae 15% • Coleoptera • Diptera Canaceidae Canace • Diptera Ceratopogonidae 76% • Mesogastropoda Hydrobiinae Hydrobia 208 Invertebrate Richness & Abundance Ditch Plug Within-Pool Procedural Control • Acarina A 0 Amphipoda Corophium 17% • Diptera Chironomidae B Hemiptera Delphacidae 67% Invertebrate Richness & Abndance Ditch Plug Within-Pool Exclosure • Acarina A 0 Diptera Chironomidae Dicrotendipes • Amphipoda Corophiidae Corophium 0 Polychaeta Nereiddidae Neanthes 209 Invertebrate Richness & Abundance Ditch Plug Within-Pool Enclosure 5% • Acarina A • Amphipoda Corophiidae Corophium E Diptera E Diptera Chironomidae Dicrotendipes 85% 210 Natural Pool Habitat Procedural Treatments Invertebrate Richness & Abundance Natural Pool - Marsh Surface Procedural Control B Acarina A I Acarina B • Amphipoda • Amphipoda Ochestria B Arachnida Dictynidae • Araneae Lycosidae Arctosa • Diptera Canacidae • Coleoptera EB Coleoptera Carabidae Geopinus • Diptera 88% • Diptera Ceratopogonidae Atricopogon • Diptera Ephydridae • Hemiptera Saldidae 211 Invertebrate Richness & Abundance Natural Pool - Marsh Surface Exclosure • Acarina A • Acarina C • Hemiptera Salidae • Coleoptera 0 Coleoptera Coccinellidae Cocci nellini • Melampus Bidentatus • Amphipoda 88% • Arachnida Araneae 5 Diptera Tabanidae Tabarius Invertebrate Richness & Abundance Natural Pool Within-Pool Procedural Control S Acarina A • Diptera Chironomidae • Coleoptera Carabidae Geopius 14% 65% • Hemiptera Delphacidae • Hemiptera Saldidae 212 Invertebrate Richness & Abundance Natural Pool Within-Pool Exlcosure • Acarina A 1Anthropoda • Coleoptera S Diptera Tabanidae Tabanus 10% • Acarina B • Diptera 55% • Diptera Chironomidae 18% • Hemiptera Cicadellidae Inertebrate Richness & Abundance Natural Pool Within-Pool Enclosure • Acarina A • Coleoptera a Coleoptera Carabidae Geopinus • Coleoptera Circulionidae 57% • Diptera Stratiomyidae Odontomyia • Hemiptera Delphacidae 213 APPENDIX D INSTITUTIONAL ANIMAL CARE AND USE COMMITTEE APPROVAL LETTERS 214 UNIVmsn'Y of Nfw Hampshiri January 20, 2005 Burdick, David M Natural Resources Jackson Lab Durham, NH 03824 IACUC #: 041101 Approval Date: 11/17/2004 Review Level: B Project: Assessment of hydrologic features in saltmarshes The Institutional Animal Care and Use Committee (IACUC) reviewed and approved the protocol submitted for this study under Category B on Page 4 of the Application for Review of Vertebrate Animal Use in Research or Instruction - the study involves either no pain or potentially involves momentary, slight pain, discomfort or stress. Approval is granted for a period of three years from the approval date above. Continued approval throughout the three year period is contingent upon completion of annual reports on the use of animals. At the end of the three year approval period you may submit a new application and request for extension to continue this study. Requests for extension must be filed prior to the expiration of the original approval. Please Note: 1. All cage, pen, or other animal identification records must include your IACUC # listed above. 2. Use of animals in research and instruction is approved contingent upon participation in the UNH Occupational Health Program for persons handling animals. Participation is mandatory for all principal investigators and their affiliated personnel, employees of the University and students alike. A Medical History Questionnaire accompanies this approval; please copy and distribute to all listed project staff who have not completed this form already. Completed questionnaires should be sent to Dr. Gladi Porsche, UNH Health Services. If you have any questions, please contact either Van Gould at 862-4629 or Julie Simpson at 862- 2003. For thP TAG IP R< Chair cc: File Research Conduct and Compliance Services, Office of Sponsored Research, Service Building, 51 College Road, Durham, NH 03824-3585 * Fax: 603-862-3564 215 UNIVERSITY of NEW HAMPSHIRE May 23, 2006 Burdick, David Natural Resources, Jackson Lab Durham, NH 03824 IACUC #: 060305 Approval Date: 04/28/2006 Review Level: B Project: Ecological Responses to Ditching and Ditch-Plugging in New England Salt Marshes The Institutional Animal Care and Use Committee (IACUC) reviewed and approved the protocol submitted for this study under Category B on Page 4 of the Application for Review of Vertebrate Animal Use in Research or Instruction - the study involves either no pain or potentially involves momentary, slight pain, discomfort or stress. Approval is granted for a period of three years from the approval date above. Continued approval throughout the three year period is contingent upon completion of annual reports on the use of animals. At the end of the three year approval period you may submit a new application and request for extension to continue this project. Requests for extension must be filed prior to the expiration of the original approval. Please Note: 1. All cage, pen, or other animal identification records must include your IACUC # listed above. 2. Use of animals in research and instruction is approved contingent upon participation in the UNH Occupational Health Program for persons handling animals. Participation is mandatory for all principal investigators and their affiliated personnel, employees of the University and students alike. A Medical History Questionnaire accompanies this approval; please copy and distribute to all listed project staff who have not completed this form already. Completed questionnaires should be sent to Dr. Gladi Porsche, UNH Health Services. If you have any questions, please contact either Roger Wells at 862-2726 or Julie Simpson at 862- 2003. For the IACUC, fssica A. Bolker, Ph.D. Chair cc: File Robert Vincent' Research Conduct and Compliance Services, Office of Sponsored Research, Service Building, 51 College Road, Durham, NH 03824-3585 * Fax: 603-862-3564 216 THIS PAGE INTENTIONALLY LEFT BLANK 217