This item is only available as the following downloads:
1 THE BIOGEOCHEMICAL EFFECTS OF SEA LEVEL RISE ON COASTAL WETLAND SOIL CARBON By LISA GARDNER CHAMBERS A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS F OR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2012
2 2012 Lisa Gardner Chambers
3 To my h usband and my m om
4 ACKNOWLEDGMENTS To my adviso r, Dr. K. Ramesh Reddy, I would like to extend my true gratitude for your guidance, patience, and unfaltering belief in my abilities. Dr. Todd Osborne, I greatly appreciate your enthusiasm, time, and valuable insights on my research. I would also like to thank my other committee members: Dr. Mark Clark, Dr. Andy Ogram, and Dr. Tom Frazer, and my collaborat ors in south Florida: Dr. Stephe n Davis (Everglades Foundation) and Dr. Tiffany Troxler (Florida International University), all of whom play ed a significant role in shaping this research. Many thanks go to Mr. Gavin Wilson, Ms. Yu Wang, and Dr. Kanika Inglett for assistance in the laboratory, and to all my friends and colleagues in the Wetland Biogeochemistry Laboratory. I especially want to recognize my fellow students, Rupesh Bhomia, Jing Hu, Lucy Nagatia, and Cassandra Medvedeff, all of whom became close friends of mine during our time together in Newell Hall. Most importantly I want to thank my husband John Chambers, and my family, Marcia Gardner and Chris Gardner, who have provided me with endless support, encouragement, and love. Final ly, I would like to acknowledge that my Ph.D. was supported by several funding sources, including a fellowship from the National Science Foundation through the Adaptive Management of Water, Wetlands, and Watersheds Integrative Graduate Education and Resear ch Traineeship, as well as a four year University of Florida Alumni Graduate Award. R esearch support was provided in part by the Florida Sea Grant Nutrient Dynamics Fellowship, with funds from Scotts Miracle Gro, Inc. and by the Everglades Foundation fel lowship program in support of graduate student research on applied Everglades issues. Portions of this work were developed in collaboration with the Florida Coastal Everglades Long Term Ecological Research program ( National
5 Science Foundation Grant DBI 0620409 ), Everglades National Park, and the South Florida Water Management District. The views expressed are those of the author and do not necessarily reflect the view s of these organizations.
6 TABLE OF CONTENTS page AC KNOWLEDGMENTS ................................ ................................ ................................ .. 4 LIST OF TABLES ................................ ................................ ................................ ............ 9 LIST OF FIGURES ................................ ................................ ................................ ........ 10 LIST OF ABBR EVIATIONS ................................ ................................ ........................... 12 ABSTRACT ................................ ................................ ................................ ................... 14 CHAPTER 1 INTRODUCTION ................................ ................................ ................................ .... 16 Sea Level Ris e and Coastal Wetlands ................................ ................................ .... 16 Coastal Wetland Carbon Biogeochemistry ................................ ............................. 18 Dissertation Overview ................................ ................................ ............................. 19 Objectives and Hypotheses ................................ ................................ .............. 20 Dissertation Organization ................................ ................................ ................. 21 2 SEA LEVEL RISE IMPACTS ON BIOGEOCHEMICA L CARBON CYCLING IN COASTAL WETLANDS ................................ ................................ .......................... 26 Background ................................ ................................ ................................ ............. 26 Coastal Wetland Responses to Sea Level Rise ................................ ...................... 27 Carbon Quantity and Quality ................................ ................................ ................... 31 Primary Production ................................ ................................ ........................... 31 Soil Properties ................................ ................................ ................................ .. 34 Organic Matter Quality ................................ ................................ ...................... 36 Electron Acceptor Availability ................................ ................................ .................. 37 Oxygen ................................ ................................ ................................ ............. 38 Nitrate ................................ ................................ ................................ ............... 39 Iron and Manganese ................................ ................................ ........................ 41 Sulfate ................................ ................................ ................................ .............. 42 Carbon Dioxide ................................ ................................ ................................ 43 Microbial Community ................................ ................................ .............................. 45 Soil Bacteria ................................ ................................ ................................ ..... 46 Fungal Decomposers ................................ ................................ ....................... 48 Enzyme Activity ................................ ................................ ................................ 49 Carbon Balance ................................ ................................ ................................ ...... 51 Decomposition ................................ ................................ ................................ .. 52 Greenhouse Gas Emissions ................................ ................................ ............. 53 Carbon Export ................................ ................................ ................................ .. 55 Carbon Accretion and Storage ................................ ................................ ......... 56 Summary and Research Gaps ................................ ................................ ................ 58
7 3 SHORT TERM RESPNSE OF CARBON CYCLING TO SALINIT Y PULSES IN A FRESHWATER WETLAND ................................ ................................ ................. 67 Background ................................ ................................ ................................ ............. 67 Methods ................................ ................................ ................................ .................. 6 9 Exp erimental Design ................................ ................................ ........................ 69 Soil Properties ................................ ................................ ................................ .. 71 Data Analysis ................................ ................................ ................................ ... 72 Results ................................ ................................ ................................ .................... 73 Soil Properties ................................ ................................ ................................ .. 73 Potential Microbial Respiration ................................ ................................ ......... 73 Potenti al Methanogenesis ................................ ................................ ................ 74 Sulfate Versus NaCl ................................ ................................ ......................... 75 Discussion ................................ ................................ ................................ .............. 76 Potent ial Microbial Respiration ................................ ................................ ......... 77 Potential Methanogenesis ................................ ................................ ................ 79 Summary ................................ ................................ ................................ ................ 81 4 EFFECT OF SALINITY PULSING EVENTS ON SOIL CARBON LOSS ALONG AN INTERTIDAL WETLAND GRADIENT: A LABORATORY EXPERIMENT ......... 91 Background ................................ ................................ ................................ ............. 91 Methods ................................ ................................ ................................ .................. 94 Study Area ................................ ................................ ................................ ........ 94 Experimental Design ................................ ................................ ........................ 94 Soil and Water Properties ................................ ................................ ................. 96 Soil Organic Carbon Loss ................................ ................................ ................. 98 Nutrient Release ................................ ................................ ............................... 99 Microbial Indicators ................................ ................................ .......................... 99 Data Analysis ................................ ................................ ................................ 101 Results ................................ ................................ ................................ .................. 102 So il and Water Properties ................................ ................................ ............... 102 Soil Organic Carbon Loss ................................ ................................ ............... 103 Nutrient Release ................................ ................................ ............................. 107 Microbial Indicators ................................ ................................ ........................ 108 Discussion ................................ ................................ ................................ ............ 109 Freshwater Tidal Marsh ................................ ................................ .................. 109 Brackish Marsh ................................ ................................ ............................... 114 Salt Marsh ................................ ................................ ................................ ...... 117 Summary of Salinity Pulsing Effects ................................ ............................... 120 Sea Level Rise Implications ................................ ................................ ........... 121 Summary ................................ ................................ ................................ .............. 122 5 BIOGEOCHEMICAL EFFECTS OF SIMULATED SEA LEVEL RISE ON C ARBON LOSS IN AN EVERGLADES PEAT SOIL ................................ ............. 138 Background ................................ ................................ ................................ ........... 138
8 Methods ................................ ................................ ................................ ................ 142 Study Area ................................ ................................ ................................ ...... 142 Experimental Design ................................ ................................ ...................... 142 Soil and Water Properties ................................ ................................ ............... 145 CO 2 and CH 4 Flux ................................ ................................ ........................... 145 DOC and Porewater Nutrients ................................ ................................ ........ 146 Data Analysis ................................ ................................ ................................ 147 Results ................................ ................................ ................................ .................. 147 Soil and Water Properties ................................ ................................ ............... 147 CO 2 and CH 4 Flux ................................ ................................ .......................... 149 DOC and Porewater Nutrients ................................ ................................ ........ 150 Discussion ................................ ................................ ................................ ............ 151 Soil and Water Properties ................................ ................................ ............... 151 CO 2 and CH 4 Flux ................................ ................................ ........................... 154 DOC and Porewater Nutrients ................................ ................................ ........ 157 Carbon Budget ................................ ................................ ............................... 159 Implications for Everglades Peat Soils ................................ ........................... 159 Summary ................................ ................................ ................................ .............. 160 6 CONCLUSIONS ................................ ................................ ................................ ... 172 Background ................................ ................................ ................................ ........... 172 Objective 1: Review of Literature ................................ ................................ .......... 173 Objective 2: Salinity and Carb on Cycling ................................ .............................. 175 Objective 3: The Importance of Time Scale ................................ .......................... 176 Objective 4: Tidal Cycle Effects ................................ ................................ ............ 177 Objective 5: Inundation and Carbon Cycling ................................ ......................... 178 Objective 6: Interaction of Salinity and Inundation ................................ ................ 179 Objective 7: Importance of Environmental Variables ................................ ............ 179 Synthesis and Future Research ................................ ................................ ............ 180 LIST OF REFERENCES ................................ ................................ ............................. 186 BIOGRAPHICAL SKETCH ................................ ................................ .......................... 202
9 LIST OF TABLES Table page 2 1 Aboveground biomass estimates for coa stal wetlands ................................ ....... 61 3 1 Soil and water properties according to treatment condition ................................ 83 3 2 Correlations between soil parameters and C loss ................................ .............. 83 3 3 Correlations between soil and water parameters ................................ ................ 84 4 1 Intact core experimental design. ................................ ................................ ....... 125 4 2 Intact core experimental timeline ................................ ................................ ...... 125 4 3 Intact core soil properties ................................ ................................ ................. 126 4 4 Inta ct core surface water properties ................................ ................................ 126 4 5 Correlations for the control condition ................................ ................................ 127 4 6 Correlations for the salinity pulsing ev ents ................................ ....................... 127 4 7 Nutrient release for the control condition ................................ .......................... 127 4 8 Microbial indicators for the control condition ................................ ..................... 128 4 9 Summary of salinity pulsing effects ................................ ................................ .. 128 5 1 Soil properties ................................ ................................ ................................ ... 162 5 2 Surfa ce water properties ................................ ................................ .................. 162 5 3 Rising vs. falling tidal affects on CO 2 C flux rate ................................ .............. 162 5 4 Soil porewater nutrients ................................ ................................ .................... 163 5 5 Correlations between measured properties ................................ ...................... 163
10 L IST OF FIGURES Figure page 1 1 Conceptual diag ram of a coastal wetland salinity gradient ................................ 23 1 2 Simplified systems diagram of a wetland soil organic carbon cycle .................... 24 1 3 Flow diagram of research objectives and dissertation organization .................... 25 2 1 Sea level rise affects on coastal ecosystems over time ................................ ...... 62 2 2 The generalized carbon cycle in coastal wetlands ................................ .............. 63 2 3 Coastal wetland responses to sea level rise ................................ ....................... 64 2 4 Soil carbon and bul k density as a function of salinity ................................ .......... 65 2 5 Decomposition as a function of salinity ................................ ............................... 66 3 1 Site location map ................................ ................................ ................................ 85 3 2 Potential soil respiration over time ................................ ................................ ...... 86 3 3 Total soil respiration. ................................ ................................ .......................... 87 3 4 Po tential methanogenesis over time ................................ ................................ ... 88 3 5 Total methanogenesis. ................................ ................................ ....................... 89 3 6 Total carbon loss. ................................ ................................ ............................... 90 4 1 Site map ................................ ................................ ................................ ........... 129 4 2 A schematic of the intact soil core design ................................ ......................... 129 4 3 Soil fiber content. ................................ ................................ .............................. 130 4 4 Tidal cycle effects on soil respiration ................................ ................................ 131 4 5 High tide salinity pulse effects on soil respiration ................................ ............. 132 4 6 Low tide salinity pulse effects on soil respiration ................................ .............. 133 4 7 High tide salinity pulse effects on potential methanogenesis ............................ 134 4 8 Total soil organic carbon loss ................................ ................................ ........... 135 4 9 Soil organic carbon loss by pathway ................................ ................................ 136
11 4 10 Modeled effects of increased inundation on soil carbon loss ............................ 137 5 1 Site location map. ................................ ................................ ............................. 164 5 2 Schematic of the peat monolith d esign ................................ ............................. 165 5 3 Soil oxidation reduction (redox) potential ................................ .......................... 166 5 4 Surface soil (0 5 cm) bulk density ................................ ................................ ..... 167 5 5 Daytime CO 2 C flux by inundation and salinity ................................ ................. 168 5 6 Daytime CO 2 C flux over time ................................ ................................ ........... 169 5 7 Daytime CO 2 C flux over temperature ................................ .............................. 170 5 8 Methanogenesis by treatment condition ................................ ........................... 171 6 1 Tidal effects by wetland type ................................ ................................ ............ 184 6 2 Conceptual diagram with comprehensive soil respiration results. .................... 185
12 LIST OF ABBREVIATION S ANOVA Analysis of variance C Carbon CA Control (water level) ambient (salinity) CE Control (water level) elevated (salinity) DDI Distilled de ionized DI De ionized DO Dissolved oxygen DOC Dissolved organic carbon EC Electrical conductivity FID Flame ionization detector GWP Global warming potential HT High tide IA I nundated (water level) ambient (salinity) IE Inundated (water level) elevated (salinity) IPCC Intergovernmental panel on climate change LOC Labile organic carbon LOI Loss on ignition LOI Loss on ignition LSM Least squares means LT Low tide MBC Microbial bi omass carbon MUF M ethylumbelliferone N Nitrogen NEP Net ecosystem productivity
13 NS Not significant OC Organic carbon OM Organic matter P Phosphorus PLFA Phospholipid fatty acid Ppt Part per thousand PVC Polyvinyl chloride SLR Sea level rise SOC Soil organ ic carbon SRB Sulfate reducing bacteria SRP Soluble reactive phosphorus SRS Shark River Slough TFM Tidal freshwater marsh TKN T otal kjeldahl nitrogen TLOC Total labile organic carbon TOC Total organic carbon TPF 1,3,5 triphenylformazan TTC Triphenyltetraz oliym chloride
14 Abstract of Dissertation Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy THE BIOGEOCHEMICAL EFFECTS OF SEA LEVEL RISE ON COAS TAL WETLAND SOIL CARBON By Lisa Gardner Chambers August 2012 Chair: K. Ramesh Reddy Major: Soil and Water Science Coastal wetlands are among the most productive ecosystems in the world, sequestering and storing a significant portion of global carbon (C ). Sea level rise (SLR) and extreme sea level events ( storm surges, astronomical tides) threaten the sustainability of coastal wetlands and the C stored within them. Three studies were conducted in Florida (USA) with the common goal of improving the mech anistic understanding of how increased salinity and increased inundation affect bigeochemical C cycling in coastal wetland soils. All three studies emphasized the impacts of saltwater intrusion and inundation on soil respiration, methanogenesis, and dissol ved organic C (DOC), but differed in location, scale (i.e., the degree of experimental control), and focus. T he two mechanisms by which sa linity influences C biogeochemistry the sulfate effect (additional electron acceptors) and the osmotic effect (high ionic strength) where separated S ulfate causes a short term stimulation of CO 2 production in proportion to the concentration add ed and a continuous suppression of CH 4 production I onic strength d oes not af fect CO 2 production and causes only minor sup pression of CH 4 C h anges in salinity from fresh to ~14 ppt appear to cause the greatest shift in
15 biogeochemical pathways for C mineralization. However, the availability of other parameters (C lability, nutrient s and microbial community) ultimately deter mines t he persistence of the observed salinity effect s S hort term pulses of higher (or lower) salinity may have a greater impact on C cycling than gradual changes in salinity Freshwater wetlands pulsed with brackish seawater experience significant incr eases in the rate of C loss, as do salt marshes pulsed with freshwater. However, all studies indicate microbial communities adapt quickly (1 2 weeks) to changes in salinity. Increased inundation pose s a more long term threat to coastal wetland sustainabi lity, causing disintegration of soil structure decrease d in soil respiration, and increased in DOC export. Meanwhile, the simultaneous increase of both salinity and inundation cause s a synergistic decline in the rate of C loss. The overall affect of SLR on coastal wetland C storage depends on wetland type, with freshwater tidal wetlands generally having enhanced rates of C loss and brackish and salt marshes demonstrating decreased rates of C loss.
16 CHAPTER 1 INTRODUCTION Sea L evel Rise and C oastal W etl ands Changes in global sea level are a natural part of the geologic history of the earth. Since the end of the last glacial maximum (~21,000 years before present), isostatic adjus tment of the crust and exchange between ocean and land water reservoirs (gla ciers, ice caps, etc.) ha ve increased mean sea level about 120 m. In the 2,000 years prior to 1850, sea level was relatively stable, changing only about 0.0 to +0.2 mm y 1 However, from 1850 to 1950 sea level rose at a rate of ~1.7 mm y 1 and since 199 3, this rate has increased to between 2.8 and 3.1 ( 0.7) mm y 1 This rate of sea level rise (SLR) translates into a ~60 cm increase by 2100 under a business as usual scenario (IPCC 2007). A major concern of SLR is the impact to human populations and infrastructure. Approximately 23 % within 100 m of sea level, and 10 % within 10 m of sea level (IPCC 2007; McGranahan et al. 2007). Threats to livelihoods, public health, and social and economic stability are receiving incr easing attention in the scientific literature, as well as a global debate over coping strategies ( Dasgupta et al. 2011; Nicholls and Cazenave 2010; McGranahan et al. 2007). Nevertheless, a significant portion of coastal land remains sparsely populated or undeveloped, providing important environmental services, such as storm abatement, flood control, wildlife habitat, nutrient cycling, and carbon storage (Barbier et al. 2011). Many of these natural coastal lands are coastal wetlands. Defined by the presen ce of hydrophytic vegetation and waterlogged hydric soils, coastal wetlands also have a tidally influenced hydroperiod. Coastal wetlands occupy the land fringe, tolerating full or even
17 hypersaline conditions ( > 35 ppt), but also extend inland along the en tire salinity gradient. The landward end member, tidal freshwater marshes (TFM) have ocean influenced hydrology but remain fresh (<0.5 ppt; Figure 2 1). Coastal wetlands are considered among the most productive ecosystems in the world provid ing more ec osystem services than any other coastal environment (Gedan et al. 2009). Some of the more tangible ecosystem services include: wildlife habitat, nursery groun ds for shellfish and commercial fisheries, coastline stabilization, storm surge abatement, livest ock fodder, and building material. However, arguably the most important functions of coastal wetlands are less tangible nutrient cycling, pollution removal, and carbon (C) storage (Barbier et al. 2011; Gedan et al. 2009). The storage and burial of C in coastal wetlands is of global importance due to the link between atmospheric CO 2 and climate change (IPCC 2007). Globally, salt marshes are estimated to cover up to 400,000 km 2 and mangroves 138,000 152,000 km 2 (McL eod et al. 2011). Despite their sma ll a real coverage relative to terrestrial and ocean ecosystems, salt marshes and mangroves sequester large amounts of CO 2 approximately 210 g CO 2 m 2 yr 1 (Chmura et al. 2003). The large C sequestration capacity is a function of high primary productivity slow decomposition, and the ability to trap and bury significant amounts of allochthonus C from terrestrial run off and tidal deposition (Armentano and Menges 1986). Coastal wetlands have persisted during centuries of fluctuations in sea level through n atural feedback mechanisms of vertical soil accretion and landward migration (e.g., Kirwan and Temmerman, 2009; Alongi 2008; Donnelly and Bertness 2001). However, there is growing concern over the ability of coastal wetlands to adapt to
18 accelerations in t he rate of SLR, especially in systems already disturbed by adjacent development, dredging and filling, changes in sedimen t loads, invasive species, etc. ( Day et al. 2008; Nicholls et al. 1999). Predictions of global coastal wetland loss associated with SL R are between 20 and 30 % by 2100 (IPCC 2007; Nicholls et al. 1999). The impacts of SLR on coastal wetlands are already evident in many areas of the world. For example, there is a growing area of research that utilizes historical data, remote sensing, and biological indicators to document and assess landscape scale changes in coastal wetlands caused by SLR (e.g., Hussein 2009; Smith 2009; Gaiser et al. 2006 ; William et al. 1999). D espite increasing documentation of ecosystem change s there are still sign ificant gaps in understanding of the mechanisms prompting these landscape scale changes. There is also uncertainty regarding the perpetuation of coastal wetland ecosystem services during these observed transformations. Coastal Wetland C arbon B iogeochemis try From a system s approach, SLR exerts two major d r iving forces on coastal wetlands increases in salinity and inundation. The best way to understand these drivers is to observe the most elementary response. In wetlands, soil microorganisms are usually the first to respon d to environmental changes due to their sensitivity (large surface to volume ratio) and rapid turn over rate. Microbes play a critical role in wetlands by controlling everything from the rate at which nutrients become available to plant s and animals, to the storage and burial of soil C The size, structure, and activity of the soil microbial community is manifested in biogeochemical cycles, the transformation and exchange of energy and ma t ter between biotic and abiotic components of the environment (Reddy and DeLaune 2008). Carbon plays a central role in nearly all biogeochemical cycles as a substrate and electron donor during
19 heterotrophic respiration (Figure 1 2). Primary production (CO 2 fixation) and lateral inflows (overland run of f and deposition) introduce C containing organic matter to a wetland. The organic matter then undergoes a degradation process that includes leaching, enzyme hydrolysis, and microbial respiration. The s e processes break down complex C polymers into monomer s, and then further mineralize the C into CO 2 and CH 4 which are released to the atmosphere (Figure 1 2). The efficiency with which microbes mineralize C is crucial to C fixed and imported into the syste m becomes stored or buried, and how much is mineralized, exported, or released as CO 2 and CH 4 Most coastal wetlands function as net C sinks (Bridgham et al. 2006 ) Dissertation O verview Sea level rise could impact the ability of coastal wetlands to fun ction as global C sinks by altering the rate and nature of biogeochemical C cycling (Morris et al. 2002). A comprehensive review of current literature regarding the impacts of SLR on C cycling reveals significant gaps in the understanding of how the main drivers associated with SLR (salinity and inundation) affect C cycling in coastal wetlands, especially the final stage of the C cycle microbial respiration. The overarching goals of this dissertation are to (1) improve the mechanistic understanding of h ow salinity and inundation a ffect biogeochemical carbon cycling, (2) vary scale (i.e., degree of experimental control) to translate biogeochemical processes into real world observations, and (3) provide insight into the broader question of how coastal wetl and C storage might be affected by SLR. Figure 1 3 provides a conceptual diagram outlining how the system dr ivers (salinity and inundation) a re related to other biogeochemically important processes that I predicted will i mpact the rate of wetland C loss ( e.g., changes in water chemistry, saltwater
20 intrusion, water level, and tidal cycle). This diagram also indicates in which chapter of the dissertation the topic is addressed. O bjectives and H ypotheses The first objective of this dissertation wa s to r eview the current literature and identify knowledge gaps in the understanding of how SLR impacts coastal wetland C cycling which is presented in Chapter 2 of this dissertation Objective 2 wa s to d etermine the process level impact s of salinity on microb ially mediated C cycling in a wetland soil (C hapter 3) I hypothesize d t he high ionic strength of saltwater would decrease the activity of all soil microbes, but the presence of sulfate (SO 4 2 ), an alternative electron acceptor used during anaerobic respi ration will stimulate CO 2 production by promoting sulfate reduction. A ssess ing the importance of time scale (pulsing events vs. gradual change s ) on the biogeochemical response of C cycling to changes in salinity wa s the third objective, and is covered in Chapters 3 5 I expect ed that s hort term, dynamic changes in salinity would have a greater effect on the rate of C cycling than gradual changes a consequence of the high capacity of soil microbial population to adapt to environmental change. Objective 4 addressed in Chapters 4 and 5, wa s to d etermine the importance of tidal cycle on C cycling and how the magnitude of tidal effect differ s along the coastal wetland co ntinuum from TFMs to salt marshes and mangroves The hypothesis for this objective wa s that production of CO 2 would be higher under low tide conditions and will be directly related to the height of the water table regardless of wetland type The fifth objective involve d the q uantif ication of the e ffect of increased inundation (reduced soi l exposure during low tide) on the rate and biogeochemical pathways of C cycling I expect ed i ncreased inundation would reduce the ove rall rate of C cycling, but would increase the contribution of methanogenesis to C
21 loss, relative to respiration. This o bjective is included in Chapters 4 and 5. In o bjective 6 the interactive effects of simultaneous increases in both salinity and inundation on the rate of C cycling were d etermine d (C hapter 5) u nder the hypothesis that i ncreases in salinity would accelera te C loss, wh ile increases in inundation would decrease the rate of C cycling. Both processes occurring simultaneous ly would effectively cancel out the other, resulting in a minimal or no effect on overall C cycling. The final objective wa s to a ssess the importance of other physical and biological variables within the wetlands studied on the rate of C cycling as possible covariants The hypothesis for objective 7 was that i nherent differences in soil properties (organic matter, C, and nitrogen (N) conte nt), nutrient availability (N and phosphorus (P)), a nd microbial community size would co vary with coastal wetland type, salinity, and inundation, and i nfl uencing the response of C cycling to SLR ; this topic is covered in Chapters 4 and 5. Dissertation O rg anization This dissertation is organized into 6 C hapters. Chapter 1 (this Chapter) introduces the concept of s ea level rise, the importance of C storage in coastal wetlands, and provides an overview of the objectives and format of the dissertation. This is followed by Chapter 2, which involves a comprehensive literature review of current knowl edge pertaining to the e ffects of SLR on the three major regulators of C biogeochemistry : C quality, electron acceptor availability, and microbial community. Chapt ers 3 5 focus on individual research studies presented in order of increasing scale and complexity. Chapter 3 focuses on a microcosm study disentangling the impacts of salt (ionic strength) and seawater (containing sulfate) on potential respiration and met hanogenesis in an organic freshwater wetland soil. An intact soil core study investigating the impacts of salinity pulsing events and tidal cycle on C loss along a coastal wetland salinity
22 gradient is the subject of Chapter 4 and Chapter 5 presents a mes ocosm study o n the effects of increased salinity, inundation, and the combination thereof, on C loss and porewater nutrients in an Everglades peat soil. Finally, Chapter 6 provides a synthesis of conclusions drawn from all the dissertation studies and a t heoretical discussion on the fate of coastal wetlands as C sinks during SLR.
23 Figure 1 1 Conceptual diagram of a coastal wetland salinity gradient. Coastal wetlands can have salinities ranging from freshwater to hypersaline, b ut all have a tidally influenced hydroperiod.
24 Figure 1 2 Simplified systems diagram of a wetland soil organic carbon cycle. Carbon enters a wetland through photosynthesis and external inflow. It is then degraded through processes of leaching, hydrolosis, and microbial respiration to be exported or released as carbon dioxide or methane. The remaining carbon is stored in wetland soils.
25 Figure 1 3 Flow diagram of research objectives and dissertation organization. The ccompanying sea level rise: salinity and inundation, and the associated impacts of other ecosystem characteristics.
26 CHAPTER 2 SEA LEVEL RISE IMPAC TS ON BIOGEOCHEMICAL CA RBON CYCLING IN COASTAL WETLANDS Background Biogeochemistry is the scientific study of the interactions between biological, chemical, and physical processes in nature; it includes aspects of biology, soil science, chemistry, and geology Specifically, biogeochemistry focuses on the cycling, or transformation, of elements between various phases (e.g., solid, liquid, gas) and states (e.g., organic or inorganic; oxidized or reduced) with an emphasis on the exchange of energy and matter bet ween biotic and abiotic components of the environment through oxidation reduction reactions ( Reddy and DeLaune 2008 ) Sea level rise can impact C biogeochemistry in coastal wetlands by three main mechanisms: ( 1) the complete loss of wetland area, ( 2) an increase in inundation, and/or, ( 3) an increase in salinity. While the first is fairly straightforward (a decrease in wetland area means a reduction in C cycling and storage), the latter two mechanisms involve more subtle changes in the chemistry, biolog y, and geomorphology of coastal wetlands. This includes multi directional changes in the C balance, varia tions in the end products of C cycling, and numerous feedback loops. The goal of this review is to present the current state of knowledge regarding how C biogeochemistry in coastal wetlands is impacted by SLR, with special emphasis on the role of inundation and salinity on C cycling and storage. The impacts of SLR on coastal wetlands are seen over a range of time scales, which eventually lead to the modification of the chemical, biological, and geological environment of coastal wetlands (Figure 2 1 ). The challenge in discussing SLR impacts on any particular system or process is that ecosystem drivers never occur in isolation. That is, disentangling the
27 response of C biogeochemistry to alterations in inundation and salinity due to SLR, from the response to the many other environmental variables (e.g., temperature, nutrient availability human disturbance, etc.) is nearly impossible. Laboratory and m esocosms studies increase the control researchers have over extraneous variables, but reduce the applicability of results to real world conditions. Both manipulative and observational studies are included in this review, and a discussion of the influence of additional/extraneous variables is included when warranted. On the most basic level, biogeochemical C cycling requires an electron donor (C substrate), an electron acceptor (O 2 NO 3 Fe 3+ Mn 4+ SO 4 2 or CO 2 ), and soil microbes to mediate the energy exchange. As a result of this energy exchange, some C is released to the atmosphere as CO 2 or CH 4 while the remaining C is stored, buried, or exported from the system (Figure 2 2 ). This review is organized according to the 3 major regulators of C bioge ochemistry : ( 1) C quantity and quality (primary production, soil properties, and organic matter quality), ( 2) electron acceptor availability, and ( 3) microbial community (bacteria, fungi, and enzyme activity). The review concludes with a discussion of the overall C balance (decomposition rate s C emissions, C storage and accretion). Brief summaries are included at the end of eac h sub section and predictions of the overall response of the variable in question are presented when there is ample evidence to s upport it. Coastal Wetland Responses to Sea Level Rise The most basic way C bioegeochemistry is disrupted by SLR is through direct loss of wetland area. In recent history, humans have been to blame for the majority of coastal wetland loss; at least 35 % of global mangrove habitat has been lost since 1980 due to activities such as resource harvesting, filling for development, and mariculture
28 ( Valiela et al. 2001) However, the loss of coastal wetland area to SLR is increasing. According to IPCC estimates, SLR of 18 to 59 cm before the end of this century will result in the loss of 30 % ( 2007 ) Coastal wetland loss as a consequence of SLR is a non linear process, but it somewhat predictable with the current knowledge of salt marsh ecomorphological processes ( Fagherazzi et al. 2012 ) Still, the human response to SRL remains a wild card Nicholls et al. ( 1999 ) estimates that a 38 cm increase in sea level will result in a 22 % loss of aerial coastal wetland coverage, but up to 7 0 % of that loss will be the result of human intervention, such as the construction of sea walls. Humans also disrupt natural wetland dynamics with coastal development and the construction of hard shoreline structures (e.g., dikes, walls, armorment) that a lter sediment d elivery patterns. In Louisiana it is estimated that for every 1 m 2 of coastal wetland lost, 30 kg of stored C are also lost ( DeLaune and Whit e 2011 ) Coastal wetland loss not directly attributed to human activity is often traced back to i ncreased salinity. Wetland plant species have different degrees of salinity tolerance, with freshwater species being most sensitive to salinity changes. Salt can negatively impact plant productivity by : ( 1) osmotic stress ( Batzer and Shartiz 2006 ) ( 2) t he accumulation of hydrogen sulfide ( Koch et al. 1990 ) and ( 3) the inhibition of nutrient uptake ( Bradley and Morris 1991 ) If more salt tolerant species are not able to colonize a wetland impacted by increased salinity rapidly, plant mortality can lead to peat collapse and significant elevation loss ( DeLaune et al. 1994 ) ; creating a feedback loop that eventually ends in the conversion of coastal marshes to open wat er ( Nyman et al. 1990 ) The physical mechanisms underpinning the conversion of vegetated m arshes to
29 open water mud flats has been successfully simulated with numeric models, whereby aboveground biomass disturbance reduces soil stability and sediment trapping, local shear stress increases, the elevation of the marsh platform decreases, and the s ystems shifts to a non vegetated equilibrium state ( Fagherazzi et al. 2012) The vulnerability of marshes to submergence during SLR can be estimated with knowledge of the tidal range and sediment availability in the local coastal zone ( Reed 1995 ) Increas ed inundation, on the other hand, appears to be something most coastal wetlands are adept at naturally compensati ng for and is less likely to lead to significant submergence. Coastal wetlands occupy a specific elevation niche relative to sea level and mar sh vegetation thrives when the marsh platform is between mean sea level and mean high tide ( Morris et al. 2002 ) The natural responses of coastal wetlands to increased inundation are : ( 1) keeping pace with sea level through vertical accretion, or ( 2) migr ating landward to maintain optimal elevation relative to sea level Understanding the physical setting of a wetland can often assist in determining the most probable response to SLR ( Figure 2 3 ) Keeping pace with sea level rise occurs through a natural feedback loop that has allowed coastal wetlands to persist since the last ice age: increased inundation depth enhances sedimentation, inorganic sediments accelerate primary production and organic matter accumulation, and the marsh accretes to a new equili brium with sea level ( Fagherazzi et al. 2012 ) Inorganic sediment promotes plant growth by providing nutrients (Ca, K, and P), increasing the availability of cation adsorption sites on the soil, and diminishing nutrient leaching ( Patrick and Khalid 1974 ) Inorganic sediments also contain appreciable amounts of iron (III) that can precipitate S 2 thereby removing this
30 toxin from the system ( Nyman and DeLaune 1999 ) Enhanced plant productivity strengthens the soil matrix with additional root production, p romotes organic matter accumulati on, and contributes to vertical accretion ( Mudd et al. 2009 ) In the coastal wetlands of the Mississippi River delta, organic soil formation accounts for 70 80 % of vertical accretion and sediment deposition the remaining 2 0 30 % ( Day et al. 2000 ) Despite this natural feedback, some researchers question the ability of coastal wetlands to keep pace with the current rate of SLR, citing concerns over the human alteration of the volume of sediment presently delivered to the coa stal zone ( Day et al. 2008 ) Landward marsh migration is a process observed in coastal wetlands with low relief and minimal impediment from human development. The seaward edge of an existing wetland may submerge with SLR, while simultaneously expanding lan dward to maintain the optimal level of tidal inundation ( Gaiser et al. 2006; Adam 2002; Donnelly and Bertness 2001 ) Certain wetland types fare better during the process of marsh migration due to factors such as coastal morphology, tidal range, soil type, and human infrastructure ( Moorhead and Brinson 1995 ) In coastal Georgia, models predict a 20 % decline in salt marsh area, a 24 % decline in tidal freshwater swamp area and a 10 % increase in brackish marshes area; salt marshes are at greater risk of subm ersion and freshwater wetlands at a greater risk of infrastructure impediment ( Craft et al. 2009 ) Landward marsh migration has been extensively documented using historic aerial photographs and biological indicators as proxies of past salinity. In the Ch esapeake Bay region, soil profiles reveal a current landward marsh migration rate s of 3.5 to 6.8 m y 1 up from a rate of 0.2 to 1.3 m y 1 prior to 1850, suggesting a capacity to adjust to changing rates of SLR ( Hussein 2009 ) Marsh migration has also bee n reco rded in
31 Florida ( Gaiser et al. 2006; Ross et al. 2000; Williams et al. 1999) Cape Cod ( Smith 2009) Rhod e Island ( Donnelly and Bertness 2001 ) and North Carolina ( Moorhead and Brinson 1995; Hackney and Cleary 1987). Carbon Quantity and Quality Carbo n functions as an electron donor during aerobic and anaerobic microbial respiration and is subsequently broken down (decomposed) and/or released into the atmosphere ( as CO 2 or CH 4 ) during the process. Importantly, C not utilized by microbes can remain sto red in the soil for days to centuries, effectively reducing wetland greenhouse gas emissions. The source, amount, and quality of C are critical to predicting the balance between C emissions and C storage under SLR. This section is divided into primary pr oduction, soil propert ies, and organic matter quality; 3 regulators of coastal wetland C biogeochemistry that can be impacted by changes in salinity and inundation. Primary P roduction Primary producers (macrophytes, mangroves, and algae) represent the majo r autochthonous input of organic C to coastal wetlands. The diversity, density, and life history of primary producers differ with salinity and region, but two general patterns are evident: ( 1) species richness is negatively correlated with salinity, and ( 2 ) biomass production decreases with increasing distance from the equator. Macrophyte species richness in Georgia declined by half between the tidal freshwater marsh ( TFM ) plant community and the brackish plant community, and > 50 % again, from brackish mar sh to salt marsh plant communit y (Wieski et al. 2010). Even within a specific wetland, primary production varies spatially with generally decreasing biomass as distance from a tidal creek increases (Gonzalez Trilla et al. 2010; Neubauer et al. 2000).
32 T idal freshwater marshes are typically the most spatially and temporally heterogeneous of all coastal wetlands. Displaying no clear pattern of species 60 species of macrophytes that vary seasonally ( Odum 1988 ) In a Virginia TFM, net primary productivity was 94 % macrophytes (536 to 715 g C m 2 y 1 ) and 6 % macroalgae (59 g C m 2 y 1 ; Neubauer et al. 2000 ) Salt marshes exhibit a similar biomass allocation between macrophytes and algae, but often w ith lower overall productivity (Miller et al., 2001) As salinity increases, plant diversity decreases and zonation patterns based on inundation and salinity tolerance become evident (Odum 1988) A n abbreviated summary of aboveground biomass data by locations and salinity is presented in Table 2 1. Belowground biomass also varies with salinity. Brackish marsh macrophytes tend to have shallower roots (0 8 cm deep) than TFM macrophytes (13 20 cm deep), possibly because TFM species translocate more C and nutrients belowground just prior to the dormant season ( Neubauer et al. 2005 ) Nevertheless, salt marshes have the greatest proportio n of biomass stored belowground. A pproximately 60 % of Spartina alterniflora biomass for example, is contained wit hin the soil ( Good et al. 1982 ) A study of a S. alterniflora marsh in Louisiana found an annual belowground production rate of 11,676 g m 2 compared to only 1,821 g m 2 produced aboveground ( Darby and Turner 2008 ) The accumulation of belowground bioma ss is an important energy reserve tightly coupled to overall marsh and estuarine productivity; this energy is transferred during oxidation reduction reactions, leaching, diffusion, bioturbation, and erosion ( Roman and Daiber 1984 )
33 In tropical locations mangroves replace herbaceous vegetation in saline wetlands. Over 70 % of mangrove dominated wetlands occur between 0 and 10 latitude ( Twilley et al. 1992 ) Generally, mangrove biomass production d ecreases from an average of 2.8 Mg km 2 nearest the equat or, to 0.96 Mg km 2 between 30 and 40 ( Twilley et al. 1992 ) where solar radiation and temperatures are l ower In addition to location, mangrove productivity may also be regulated by salinity. A mesocosm study in China where Aegiceras corniculatum was ex posed to wastewater of varying salinities revealed the highest aboveground biomass in the 15 ppt treatment, with decreases of 33 and 44 % in the comparable 0 ppt and 30 ppt treatments, respectively ( Wu et al. 2008 ) Similar to salt marshes, mangroves mai ntain high belowground biomass. Estimates from south Florida range from 2,317 to 4,674 g m 2 for belowground biomass ( Castaeda Moya et al. 2011) The ratio of aboveground:belowground biomass in mangroves is highly sensitive to nutrient availability and inundation frequency. Where resources are high and inundation low, belowground biomass was 17 33 % of aboveground biomass; limitations in resources and high inundation resulted in root production 2 10 x greater than aboveground production ( Castaeda 2010 ) While salinity tolerance is an important determinant of vegetation zonation patterns in coastal wetlands, inundation can also affect C assimilation rates. In a temperate salt marsh, tidal inundation (fully or partially submerging marsh vegetation) de creased the rate of C fixation approximately 46 % compared to the control condition ( Kathilankal et al. 2008 ) A study in Argentina also indicated higher aboveground productivity in the high marsh (less frequently inundated), compared to the low marsh (Tab le 2 1 ; Gonzalez Trilla et al. 2010 )
34 A crude meta analysis (ignoring latitude and other physical factors) of the effects of salinity on aboveground biomass in herbaceous coastal wetlands revealed no significant differences between fresh (1151 455 g C m 2 y 1 ), brackish (1441 766 g C m 2 y 1 ) and salt marshes (1177 370 g C m 2 y 1 ). The high variability confirms that salinity is not the only (or possibly, even the major) factor controlling primary production in coastal wetlands. Several studies indic ate the importance of inundation/elevation ( Gonzalez Trilla et al. 2010; Roman and Daiber 1984 ) nutrient and sediment delivery ( Wieski et al. 2010 ) litter quality ( Craft et al. 200 9) and temperature ( Gonzalez Trilla et al. 2010 ) in regulating coastal we tland aboveground productivity. Soil P roperties Soils can be short term and long term reservoirs of C in coastal wetlands, as well as the habitat for most of the microbes involved in C cycling. Therefore, the chemistry, texture, structure, and organic mat ter content of the soil greatly influences C biogeochemsity. The general properties of soils found in saltwater and freshwater wetlands differ substantially: TFMs have more autochthonous organic C and receive allocthonous C from terrestrial sources (high in clays, silts, and recalcitrant organic matter), whereas saline wetlands have less authothonous C and higher inorganic C inputs from marine sources (including fine sands and phytoplankton ; Wieski et al. 2010; Paludan and Morris 1999; Odum 1988; Hatton et al. 1983 ) Differences in the amount of sediment deposition are the primary reason that soil bulk density increases with salinity and proximity to the sea ( Craft 2007 ) A comparative study of soils in fresh, brackish, and salt marshes of the Mississippi River delta revealed percent pore space (by volume) decreased from 96 % in the TFM to 88 % in the salt marsh, and percent organic C (by weight) decreased from 17 % in the TFM to 12 % in the salt marsh ( Nyman
35 et al. 1990 ) Differences in pore space volume ha s consequences for accretion rate; 1 cm of soil requires 656 g m 2 of organic matter in a TFM, 565 g m 2 in a brackish marsh, and 758 g m 2 in a salt marsh in Louisiana ( Nyman et al. 2006 ) Given that freshwater and brackish soils tend to have lower bulk d ensity and higher pore space volume, these soils are more vulnerability to compaction and structural collapse. Peat collapse can be caused by vegetation death, erosion, and excessive soil bioturbation. Vegetation death may result from salinity stress ( Wil liams et al. 1999 ) drought ( DeLaune et al. 1994 ) herbivory ( Smith 2009 ) or human disturbance ( Kirwan et al. 2008 ) When vegetation dies, the structural support provided by the live roots is lost, and lacunae (gas filled structures encompassing up to 29 % of root space) are also lost ( DeLaune et al. 1994 ) The extremely low bulk density that results means the soil can no longer support itself and the solid material compacts to fill the void. Understanding the risk factors for peat collapse may help preve nt wetland loss and C release. Root death likely occurs prior to any noticeable decline in aboveground biomass and may be a warning of impending peat collapse ( Turner et al. 2004 ) Other factors, such as eutroph ication, reduce soil stability ( Turner 2011 ) In a South Carolina salt marsh, fiddler crab ( Uca Pugnax ) burrows near tidal creek heads reached > 800 m 2 and was negatively correlated with plant stem density and positively correlated with infiltration rates ( Hughes et al. 2009 ) It was hypothesized that the crabs destabilized the soil by consuming and damaging root structures causing peat collapse. When wetlands submerge, the organic C stored in the soil is released into the adjacent estuary or continental shelf where it is either buried or decompo sed by marine microbes ( Li et al. 2011; White et al. 2009 )
36 Increased tidal inundation raises the ratio of inorganic C to organic C. For example, tidally dominated marshes have 50 % less soil organic C and accrete organic C three times slower than river dominated marshes. The higher concentration of organic C in river dominated marshes is due to higher primary productivity, slower decomposition, and lower sedimentation rates ( Craft 2007 ) However, the relationship between C accumulation and distance from the coast is sometimes non linear. The turbidity maximum of an estuary may coincide with the location of a brackish marsh, creating an area of significantly higher soil C at an intermediate location along the salinity gradient ( Wieski et al. 2010 ) Inter mediate marshes (2 to 10 ppt) have been shown to contain 32 % more C than comparable freshwater marsh soils in the Mississippi River delta ( Nyman et al. 1990) In general, freshwater wetlands tend to have a higher organic C content and lower bulk density (F igure 2 4 ). Overall, the physical soil properties of TFMs make them larger reservoirs of organic C, but they are also more susceptible to peat collapse and submergence than saline wetlands. Sea level encroachment in TFMs will simultaneously increase soil stability and decrease the density of soil organic C. Organic M atter Q uality Carbon is contained within the matrix of organic matter (OM) and the ability of microbes to break down OM depends on the quality of the material, including the complexity of the C bonds and the availability of essential nutrients. The quality of autochthonous OM can vary along salinity gradients. Saltwater wetland plants generally produce lower quality litter than freshwater wetland plants and tend to be more rigid in order to wi thstand tides and turbulence. The salt marsh plant, S. alterniflora, is composed of 80 % lignocellulose ( Benner et al. 1986 ) Only a small group of organisms
37 can decompose lignin (actinomycetes, some fungi and eukaryotes) making lignin content a good predi ctor of decomposition rate ( Penton and Newman 2008; Gennari et al. 2007; Enriquez et al. 1993; Benner et al. 1986 ) Other intrinsic characteristics of organic matter that are influenced by salinity include fiber content, plant morphology, and the presence of inhibitors (waxes, cutins, tannins, etc.; Webster and Benfield 1986 ) Concentrations of sulfide, an end product of sulfate reduction, increase with salinity and ha ve been shown to inhibit plant uptake of N ( Bradley and Morris 1991 ) Therefore, increas e s in salinity can increase the C:N ratio of OM ( Mendelssohn and Morris 2000 ) The ratio of C:N in OM is inversely related to rate of microbial respiration and is a good predictor of the balance between C decomposition and C storage. One study in Georgia found biomass quality (leaf C:N) was greater in the TFM (34.5 2.8), compared to the brackish (40.9 1.3) and salt marsh (36.4 1.8; Craft et al. 2009 ) Other studies have found even lower C:N in TFM soils, with values of 11.6 to 12.1 in Virginia ( Neub auer et al. 2000 ) All of this information provides a clear picture of a decline in autochthonous organic matter quality as salinity increases, which reduces the lability of C substrates for microbial decomposition, and possibly increase soil C storage. Electron Acceptor Availability Both salinity and inundation exert strong control over the abundance of different electron acceptors used during oxidation reduction reactions. Redox potential, a measure of the magnitude of electron pressure in the soil (in dicated by Eh), is high when the water table is low, allowing for greater O 2 diffusion and faster more efficient aerobic C cycling. When inundation increases, redox potential drops and anaerobic microbes capable of using alternative electron acceptors ar e favored. Alternative
38 electron acceptors are used in a specific sequence depending on their availability and energy yield; energy yield decreases from O 2 NO 3 Fe 3+ Mn 4+ SO 4 2 and CO 2 (Patrick and DeLaune 1977). In general, increased salinity raises the soil redox potential of anaerobic wetland soils by providing additional SO 4 2 electron acceptors (Mendelssohn et al. 1999). The subsequent section s discuss specific ele ctron acceptor s, in turn, focusing on how SLR (salinity and inundation) may impact a vailability. Oxygen The solubility of oxygen in water decreases ~30 % as salinity increases from freshwater to marine systems (Capone and Kiene 1988) potentially leading to lower dissolved O 2 in the water column of encroaching seawater in coastal wetlands Oxygen availability in the soil is inversely related to the height of the water table and the level of inundation (Yu et al. 2006). While frequency and duration of inundation will certainly increase in many coastal wetlands due to SLR, the turbid, well mixed nature of seawater and daily tidal dynamics may actually enhance soil oxygenation in certain cases by creating a more dynamic hydroperiod (Capone and Kiene 1988). Soil microbes are particularly active during low tide when respiration rates are 50 t o 300 % greater than at high tide due to increased aerobic respiration (Cha pter 4 ; Gribsholt and Kristensen 2003). Soil oxygen availability is also strongly linked to plant productivity since many wetland plants can translocate O 2 from the atmosphere to th eir root zone to alleviate stress from waterlogging. Salt marsh plants generally allocate more biomass belowground than fresh and brackish marsh species and may subsequently increase oxygen availability in the soil rhizosphere. Organisms such as crabs, w hich cause extensive bioturbation and oxygenation of the soil, are also more common in saline
39 wetlands. Recent research indicates soil hydraulic conductivity, a function of both the composition of organic matter and the aggregative/dispersive forces of sa lt, may be critical in determining how quickly O 2 can diffuse into the soil following the ebb tide. Along a gradient of TFM, brackish, and salt marsh habitats in north Florida, saline soils drained an average of 5 times faster than TFM soils, leading to h igher low tide respiration rates in the brackish and salt marsh soils (2226 230 mg CO 2 C m 2 d 1 ) compared to the freshwater soil (1578 229 mg CO 2 C m 2 d 1 ; Cha pter 4 ). Overall, it is conceivable that SLR could increase O 2 availability in coastal wet lands, as long as the marsh is able to adapt to increased inundation by maintain ing an optimal elevation relative to the tidal cycle. Nitrate Several aspects of the nitrogen cycle are affected by increased salinity. One study indicated ammonification (org anic N NH 4 + ) is stimulated by salinities up to 45 ppt (Pathak and Rao 1998). Although attributed to enhanced N mineralization, the increase in NH 4 + observed by Pathak and Rao is more likely a result of the abiotic displacement of NH 4 + from the soil cati on exchange complex by the influx of sea salts, especially Na 2+ and K + Edmonds et al. (2009) observed a pulse of porewater NH 4 + when 10 ppt saltwater was added to a freshwater sediment, but the increase was transient, lasting only12 days following saltw ater addition. Similarly, the reintroduction of seawater to a diked salt marsh in Cape Cod produced a 50 fold increase in porewater NH 4 + within 1 month, but concentrations returned to baseline after 5 months (Portnoy and Giblin 1997). Concentrations of 1 0 ppt saltwater are believed to completely efflux all the exchangeable NH 4 + from the sediment surface, replacing it primarily with Na 2+ (Baldwin
40 et al. 2006; Weston et al. 2006 a ); salinities >10 ppt do not contribute to any additional NH 4 + exchange (Rysgaa rd et al. 1999). Nitrification, the transformation of NH 4 + to NO 2 /NO 3 appears to be stimulated by dilute salinity (10 ppt), but reduced by salinities >17 ppt in soils (Pathak and Rao 1998). The mechanism of this inhibition is not well understood, but it is thought to be a combination of osmotic stress and the inhibition of nitrifying bacteria enzymes by the abundant chloride (Cl ) in saltwater (Seo et al. 2008; Gennari et al. 2007). The inhibition of nitrification by Cl has been demonstrated in agric ultural soils treated with Cl containing fertilizers (Roseberg et al. 1986), but has not received any attention in the coastal wetland literature. Denitrification (NO 3 N 2 /N 2 O) is more consistently suppressed by increased soil salinities (Wu et al. 200 8). Coastal wetland studies indicate that denitrification potential is 2 3 times greater in freshwater and brackish marshes, compared to salt marshes (Craft 2007) with the highest variability in denitrification rates occurring in salt marshes (Nielsen et al. 2003). Similar to nitrification, chloride toxicity has also been suggested as possible a cause of reduced denitrification in coastal wetlands (Seo et al. 2008). Interestingly, denitrifiers may have the capacity to adapt to high Cl concentrations. I n inland watersheds, denitrification enzyme activity is only reduced by Cl if the system is not already acclimated to low levels of Cl exposure (Hale and Groffman 2006). The overall impact of salinity on the N cycle appears to be driven by both abioti c ( the release of NH 4 + from the soil surface) and biotic ( the suppression of nitrification and denitrification) processes. In general, we can expect SLR to cause a short term pulse of NH 4 + which will likely accu mulate in the soil since nitrifying and den itri fying bacteria
41 are physiologically reduced by salinity. According to current knowledge, the availability of NO 3 as an electron acceptor is likely to decrease as a result of SLR, but additional work on the applicability of Cl toxicity to coastal syst ems and the mechanism for Cl inhibition is greatly needed. Iron and Manganese Iron and manganese reduction occurs at redox potentials between 100 and 300 mV. While manganese is usually found at low concentration in soils, Fe is the fourth most abundan t element on earth, comprising ~ 4 % (by weight) of the average soil ( Reddy and DeLaune 2008 ) Ferric iron (Fe 3+ ) can be a significant source of electrons during C metabolism in certain anaerobic soils ( Lovley and Phillips 1986 ) Along a coastal wetland sa linity gradient, iron availability is tightly coupled with S and P cycling. In freshwater, Fe is commonly assoc iated with dissolved organic C ( Paludan and Morris 1999) metal complexes (Reddy and DeLaune 2008 ), and inorganic P ( Harris 1999 ) In fact, Fe has such a high affinity for P that it causes the latter to be immobilized, making P the major limiting nutrient in most freshwater systems ( Blomqvist et al. 2004) Increasing salinity overwhelms the soil with sulfide (S 2 ) produced during sulfate reducti on, triggering the release of P and allowing Fe to precipitate with S 2 to form pyrite and other minerals ( Blomqvist et al. 2004 ) The precipitation of S with Fe is critical to maintaining low porewater sulfide, a known stressor to wetland pl ant producti vity ( King et al. 1982 ) Iron is unique from the other electron acceptors because it reacts abiotically over half of the time in many wetland soils, and thus does no t always function in C cycling (Neubauer et al. 2005; Kostka et al. 2002 ) Nevertheless the biogeochemical reduction of iron can be high in TFMs, especially during peak biomass production when the soils
42 have lower electron pressure ( Neubauer et al. 2005 ) Freshwater sediments with an abundance of iron oxides occluded to the soil surface may experience a rapid release of Fe when saltwater is introduced. A laboratory study showed a spike in soluble Fe 2+ (+148 % ) following seawater addition to a freshwater soil ( Weston et al. 2006 a ) In brackish marshes, Fe has limited availability, making sul fate reduction the dominant form of anaerobic respiration ( Neubauer et al. 2005 ) Generally, the role of Fe 3+ as an electron acceptor during C cycling is expected to decrease with SLR due to the precipitation of Fe with S and the reduction in terrestrial s ediment inputs, which tend to be high in iron. Sulfate Sulfate concentrations in seawater are normally between 20 and 30 mM, as compared to concentrations of 0.1 to 0.2 mM observed in freshwater (Capone and Kiene 1988 ) The high concentration of SO 4 2 in potential of a wetland to a level suitable for sulfate reductio n, between 175 mV and 350 mV ( Jakobsen et al. 1981 ) Salt marsh studies consistently indicate that sulfate reduction accounts for the majority of C mineraliza tion (70 % to 90 % ), and sometimes up to 100 % of resp iration in surface soils (0 5 cm; Kostka et al. 2002; Howarth 1984 ) The availability of SO 4 2 in the soil of coastal wetlands decreases with distance from the sea because it is rapidly reduced prior to d iffusing into the soil profile. In salt marshes, SO 4 2 is abundant throughout the top 50 cm of the soil profile (~20 g SO 4 2 cm 3 ), while only trace amounts are observed in brackish and freshwater wetland soils (< 1 g SO 4 2 cm 3 ; DeLaune et al. 1983 ) A laboratory study manipulating the abundance of SO 4 2 in a freshwater wetland soil indicated an acceleration of anaerobic soil respiration in direct proportion to the amount of SO 4 2 added ( Chambers et al. 2011 ) Similar studies have
43 shown that the additi on of 10 ppt seawater to freshwater sediments will cause sulfate reduction to replace methanogenesis as the dominate pathway of C mineralization in just 12 days, and can be responsible for up to 95 % of C loss after 35 days ( Weston et al. 2006 b ) It is cle ar that SLR will increase the availability of SO 4 2 and sulfate reduction will become the major pathway for C decomposition with increasing salinity. Carbon Dioxide Used as an electron acceptor by hydrogenotrophic methanogens, CO 2 can be an important comp onent of C decomposition in highly reduced wetland soils. Methanogenesis still releases C into the atmosphere, but as methane (CH 4 ), a greenhouse gas with 25 times greater heat trapping capacity than CO 2 over a 100 year time horizon (IPCC 2007 ) Methanoge ns are archea, can only respire specific monomer substrates, and have a lower energy yield th an all other electron acceptor s (Reddy and DeLaune 2008 ) Nonetheless, CH 4 production is considerable in soils and sediments with extreme electron pressure, indica ted by a redox potential < concentrations of NO 3 and SO 4 2 were added to a highly anaerobic soil. This quickly caused CH 4 production to decrease b y 99 % and 78 90 % respectively, in favor of denitri fication and sulfate reduction ( Dodla et al. 2009 ) Initially, sulfate reduction and methanogenesis were believed to be mutually exclusive, and some studies suggested that SO 4 2 or S 2 exerted a dire ct to xic effect on methanogens ( Zehnder et al. 1982 ) It is now understood that both sulfate reducing bacteria (SRB) and methanogens can be active simultaneously, but normally occur at different zones or soil depths. In freshwater lake sediments, SRB are most abundant at 0 2 cm soil depths ( 100 to 150 mV), while methanogens dominate at 3 6 cm soil
44 depths ( 250 to 300 mV; Cappenberg 1974 ) Sulfate reduction may become limited by the rate of SO 4 2 diffusion into the soil profile, causing CH 4 concentrations t o increase with depth while SO 4 2 conc entrations decrease with depth ( Lee et al. 2008 ) In addition to soil depth, methanogens also have an advantage in salt marsh sediments with low flushing rates; CH 4 evolution of up to 8 g CH 4 cm 3 have been reported in s oils with minimal water exchange (King and Wiebe 1980 ) The difference in energy yield between SO 4 2 and CO 2 is a major factor controlling the balan ce between SRB and methanogens ( Jackobsen et al. 1981 ) but the nature and abundance of various substr ates (electron donors) are also important. A substrate may be either non competitive (utilized only by SRB or methanogens) or competitive (utilized by both SRB and methanogens). While the presence of H 2 a competing i t with higher use efficiencies (Winfrey and Zeikus 1977) the existence of certain substrates will support methanogens and not SRB. For example, only methanogens are capable of using methanol and methylated amines as th eir electron donor s Conversely, H 2 acetate, and formate func tion as competitive substrates ( Oremland and Polcin 1982; Oremland et al. 1982) It is also important to note that CH 4 can be produced without using CO 2 as an el ectron acceptor by aceto clastic methanogens (Wang et al. 1996) Some research suggests the prevalence of organic acids, especially acetate, can predict up to 92 % of the variability in CH 4 emissions from wetlands ( Christensen et al. 2003) In TFM soils, acetoclastic methanogens dominat ed over hydrogenotrophic methanogens, a relationship that was strongly depth dependent, but remain unchanged with slight (5 ppt) increases in salinity ( Weston et al. 2011 )
45 Generally, the use of CO 2 as an electron acceptors is expected to decrease greatly due to salinity (sulfate) increases associated with SLR. Microbial Community Soil and shallow sediments contain an estimated 10 8 microbial cells per g of soil (Reddy and DeLaune 2008). Th e microbial community is mainly comp rised of bacteria, but also in cludes fungi, archea, and eukaryotes. The soil microbial community is limiting step in C cycling. Estimates of total ecosystem metabolism in salt marshes indicate 10 20 % of all metabolism is from soil microbial activity (Miller et al. 2001), an impressive statistic considering the scale difference between bacteria and macrophtyes. Organisms capable of survival in saline conditions occur in all major domains of life (bac teria, archea, and eukarya) and rely on two main strategies for osmoregulation the accumulation of KCl in the cytoplasm, or the biosynthesis and/or accumulation of compatible solutes (Oren 2008). The accumulation of salts requires significant cellular ada ptation, therefore, the accumulation of organic osmotic solutes is more common (Welsh 2000). All soil microbes require the appropriate C substrates and electron acceptors in order to respire and decompose C, but the identity, diversity, and activity of th e microbes ultimately determines the rate of C cycling. The impact of inundation on the soil microbial community is fairly straight forward: as water level increases, the population shifts toward anaerobes, favoring bacteria over fungi, as discussed below However, the response to salinity, tidal cycle, and the complex interactions between microbes and other biogeochemical factors affected by SLR requires more in depth attention.
46 Soil B acteria While the density and diversity of bacteria in freshwater a nd marine soils are comparable, the heterotrophic communities themselves are believed to be distinct (Capone and Kiene 1988). In the coastal Everglades (Florida), bacterial species diversity rema ined high (Shannon Weiner index of 3.7 to 4.0) along a salin ity gradient between 0 to 49 ppt, but the bacterial community structure diverged significantly. The divergence was based primarily on soil salinity, and secondarily on the availability of the limiting nutrient, phosphorus (Ikenaga et al. 2010). Conversel y, a laboratory study found a significant increase in bacterial diversity (using RNA) when a Louisiana freshwater swamp soil was exposed to 3.5 ppt salinity (Jackson and Vallaire 2009). Locations where salinity fluctuations are more commonplace may mainta in a larger diversity of bacterial species that can thrive under a broader suite of conditions. A study conducted in a Baltic bay found that estuarine bacterial communities were able to adapt faster to artificial salinity manipulations than riverine bacte rial communities (Langenheder et al. 2003). Local and regional characteristics also have a strong influence on bacterial composition. In ten salt marshes between Maine and Florida, soil bacterial abundance increased with decreasing latitude, which also c orrelated strongly with soil organic matter content (Blum et al. 2004). Similar correlations between soil bacterial abundance and dissolved organic C have been observed in western salt marshes (USA; Cao et al. 2008). Microbial biomass carbon (MBC) is a co mmon index for estimating the size of the soil microbial community by measuring the amount of C contained within cell walls (Vance et al. 1987). Most studies concerning the effect of salinity on MBC have been conducted in agricultural or upland systems wh ere saline soils are produced by the
47 evapoconcentration of salts. These studies have produced dramatically inconsistent results (Gennari et al. 2007; Yuan et al. 2007; Muhammad et al. 2006; Rasul et al. 2006; Pattnaik et al. 2000). Little additional insi ght is provided by the coastal wetland literature because no studies have measure in situ MBC along a coastal salinity gradient. In the laboratory, freshwater wetland soils incubated with varying concentration of artificial seawater (3 to 35 ppt) did not p roduce any difference s in MBC (Chambers et al. 2011). Similarly, a more in depth look at microbial community structure (using DNA and RNA) following the exposure of freshwater river sediments to 10 ppt salinity revealed no difference in bacterial or arche al community composition after 5 weeks (Edmonds et al. 2009). The metabolic quotient (qCO 2 ) is an index of the efficiency of the microbial community (Wong et al. 2008). High salinity typically causes microbial communities to have a n elevated qCO 2 because stressful environments cause organisms to produce more CO 2 per individual (Saviozzi et al. 2011; Tripathi et al. 2006). However, some studies have found an inverse relationship between qCO 2 and salinity; this could be an artifact of shifting microbial po pulations ( Cha pter 4; Wong et al. 2008 ). The limited data on coastal wetland soil bacterial communities appears to suggest that although metabolic pathways may shift following changes in salinity, the microbial communities themselves are highly adaptive to sal t (Chambers et al. 2011; Edmonds et al. 2009) and SLR is not likely to cause a significant change in bacterial community size or diversity. However, additional in situ studies of MBC, phospholipid fatty acid biomarkers, and genetic information in a va riety of coastal areas i s necessary to tease out salinity effects from other environmental variables.
48 Fungal D ecomposers Historically, fungi were thought to play a minor role in decomposition in wetlands because nearly all are strict aerobes. However, a growing number of studies contradict this, showing significant fungal biomass and fungal respiration associated with decaying wetland plants, up to 70 % in some wetlands (Hackney et al. 2000). High concentrations ing decomposers, which have a lower N requirement (C:N of 10:1) than bacter ia (C:N of 4:1 ; Sterner and Elser 2002). Given that salt marsh plants possess all the above mentioned characteristics, it follows that f ungi must be critical to the decomposition process in salt marshes. However, published literature suggests that fungi are highly sensitive to salinity and are found in lower abundance in saline soils, compared to freshwater soils (Yuan et al. 2007; Rasul et al. 2006). This may be because high salinity can disrupt spore germination and sexual reproduction in fungi (Van Ryckegem and Verneken 2005). A significant shift in the fungi:bacteria ratio, from 9:1 to 1:4, was observed when a floodplain soil experie nced only slight increases in electrical conductivity ( EC ; Sardinha et al. 2003). Similar to bacterial communities, fungal communities ordinate along salinity gradients, with some species adapted only to freshwater environments, some only to saltwater env ironments, and some distributed ubiquitously (Van Ryckegem and Verneken 2005). Fungal species are also somewhat plant specific, colonizing only one or more species of vegetation. In salt marshes, standing dead Spartina species have some of the highest fu ngal diversity and density, with fungal cells exceeding bacterial cells 165:1 (Torzilli et al. 2006). Inundation also drives fungal abundance and diversity. In salt marshes, even slight elevation changes between high marsh and low marsh areas yield consi derable
49 differences in fungal abundance (Cordova Kreylos et al. 2006). When the redox potential of a wetland soil was artificially manipulated in the laboratory, fungal activity exceeded bacterial acti vity in the aerobic range (Eh > +250 mV), while the rev erse was true when Eh dropped below 0 mV (Seo and DeLaune 2010). Understanding the role of predicting how the C balance of coastal systems will be impacted during SLR C urrent literature suggests that a decrease in fungal decomposers is likely to result from both increased salinity and increase d inundation in coastal wetlands. Enzyme A ctivity Soil enzymes consist of amino acids synthesized by bacteria and other cells to catalyze reactions. Enzymes are critical to C mineralization because they hydrolyze complex C polymers (cellulose, proteins, lipids, lignin) to form monomers (sugars, amino acids, fatty acids) that bacteria can utilize for respiration (Reddy and DeLau ne 2008). The rapidity with which enzymes respond to environmental conditions and the relative ease of quantifying their activity makes them ideal indicators of soil metabolic activity (Makoi and Ndakidemi 2008). Several enzymes involved in the hydrolys is of C polymers (dehydrogenase, glucosidase, and cellulase) have been studied under different ionic conditions to determine the impact of osmotic stress on synthesis and activity. Dehydrogenase is an intercellular enzyme involved in basic metabolism an d is commonly found in anaerobic microbes (Tripathi et al. 2007). Considered a strong indicator of overall soil biological activity, dehydrogenase activity increased exponentially when seawater concentrations increased from 0 to 30 ppt in a mangrove mesoc osm (Wu et al. 2008; Frankenberger
50 and Dick 1983). This was directly contrary to the findings of Tripathi et al. (2007), who found a linear decrease in in situ dehydrogenase activity when EC was raised from 0 to 30 mS cm 1 in a coastal rice paddy soil. A n additional study investigating the effect of NaCl on dehydrogenase enzymes indicated a slight decrease in activity when EC was increased from 2 to 8 mS cm 1 but the effect was transient and disappeared after 40 days of incubation (Saviozzi et al. 2011). The enzyme responsible for glucose production, glucosidase, decreased slightly in response to the addition of 3.5 ppt artificial seawater to a freshwater wetland soil, though not significantly (Jackson and Vallaire 2009). In a Korean coastal bay, in situ glucosidase activity decreased by 42 % as salinity increased from 1 to 45 mS cm 1 (Siddikee et al. 2011). Cellulase is responsible for the depolymerization of cellulose and was suppressed in 5 species of fungi exposed to EC increases of 0 to 3 mS c m 1 (Malik et al. 1979). The only study to test the impact of salinity on enzymes involved in N and P cycling (important nutrients in C biogeochemistry) was conducted by Jackson and Vallaire (2009). They found both phosphatase and NAGase enzyme activiti es decreased approximately 20 % following the addition of 3.5 ppt seawater. Differences in enzyme activities may be due to osmotic stress, which is known to interfere with enzyme and membrane activities (Van Ryckegem and Verneken 2005). Changes in nutrien t or C availability that accompany SLR may indirectly alter enzyme synthesis and activity (Baldwin et al. 2006). For the most part, enzyme activity appears to be negatively impacted by salinity, which would decrease the rate of microbial C cycling. Studi es targeting the impact of seawater (rather than just EC) on enzymes and in situ
51 measurements of enzymes along coastal wetland gradients would greatly improve our understanding of the relationship between SLR and soil enzymes. Carbon Balance Although a me chanistic understanding of how each regulator of C cycling is individually impacted by increased salinity and inundation i s necessary, the overall purpose of this review is to provide insight on how the balance of C in coastal wetlands will be affected by SLR. In large part, the future of coastal wetlands as a global C sink depends on whether an area of wetland submerges, accretes vertically, or migrates landward in response to SLR. Basic environmental factors can be used to predict which trajectory a wet land ecosystem will take (Figure 2 3 ). In general, submergence will lead to the loss of existing C stores and the loss of the ability to sequester C in the future. Vertical accretion is expected to add to the C sink potential of a wetland because the si ze and distribution of wetland types remains unchanged, while addition al organic and inorganic C is added to the system. Finally, the most difficult scenario for predicting future C balance is landward marsh migration. Models and observations note that w hen coastal wetlands migrate, some types (e.g., TFM, brackish, salt marshes) tend to lose area, while oth ers gain area (Craft et al. 2009 ). It is during marsh migration that the mechanistic analysis of C cycling regulators presented in this review becomes most valuable. For example, if the proportion of brackish marsh to salt marsh increases, as predicted for part of Georgia (Craft et al. 200 9 ), we could expect to see the following responses in the larger wetland ecosystem: (1) a slight increase in produc tivity, diversity, and quality of above ground biomass and OM (2) an increase in the availability of Fe 3+ NO 3 and O 2 as electron acceptors (though SRB will lik ely still dominate respiration), and (3) greater fungal biomass and enzyme activity contribut ing
52 to decomposition. The magnitude of each of these responses will determine the effect on the overall C balance. Th e final section in this chapter provides a review of current research related to the rate of decomposition, greenhouse gas emissions, C e xport, accretion rates, and C storage capacity in coastal wetlands. Decomposition A growing number of studies have investigated the overall rate of decomposition (C loss through mineralization) along a coastal wetland salinity gradient, but results ha ve been contradictory ( Figure 2 5 ). The decomposition of in situ litter material placed in mesh bags and distributed along a coastal salinity gradient resulted in the loss of 41 49 % of the initial mass in a salt marsh, compared to 29 30 % loss in a brackish marsh, and 30 36 % loss in a TFM (Craft 2007). The difference in decomposition rate was well correlated with salinity and showed no relationship to litter quality (Craft 2007). These findings are in contrast to a similar study in a coastal wetland in Den mark using cotton strips. Cotton tensile strength loss was used as a proxy for the r ate of cellulose decomposition and indicated a general decrease in decomposition rate as salinity increased (Mendelssohn et al. 1999). In this study, decomposition rate w as not influenced by salinity, but was positively correlated with soil N and P. Salinity does not seem to impact the rate of decomposition during the initial leaching stage (first 3 days), but after 7 days, decay rates were significantly faster in freshwa ter sites in a litter bag study in a Portuguese estuary (Quintino et al. 2009). Two additional studies, one in arid streams (Reice and Herbst 1982) and one in transitional wetlands (Sangiorgio et al. 2008) corroborate the results of Mendelssohn et al. (199 9) and Quintino et al. (2009), indicating decomposition rates are faster in freshwater than in saltwater systems. A definitive conclusion regarding how salinity impacts the amount and rate of C loss
53 through decomposition cannot be made with the existing li terature, but it appears that inherent site characteristics ( e.g., nutrient availability and litter quality) play a significant role. Greenhouse G as E missions Carbon dioxide is the end product of aerobic and anaerobic microbial respiration and is typicall y the major greenhouse gas emitted from all types coastal wetlands, representing 94 99 % of all C emissions (Cha pter 4 ). Even in situ measurements from a TFM (which have the lowest salinity and, typically, the highest rates of methanogenesis) indicted only 6 % of C loss through CH 4 flux (Neubauer et al. 2000). The rate of CO 2 production in wetland soils is a product of the size, activity, and metabolic efficiency of the microbial community (the topic of the previous section). Since microbial ecology studie s can be time consuming, the rate of CO 2 and CH 4 flux is often used as a general proxy for their activity. The majority of studies investigating CO 2 flux from natural systems of differing salinities have been conducted in upland soils exposed to evapora tive salinity. Work by Gennari et al. (2007), Malik et al. (1979), Muhammad et al. (2006), and Pathak and Rao (1998) in arid and semi arid agricultural soils all found significant declines in CO 2 flux with increasing salt and EC. This research suggests d ifferential salt tolerances among microbes (Malik et al. 1979) and that increasing salinity is not necessarily fatal to the soil microbes, but just reduces metabolic efficiency (Gennari et al. 2007). Unfortunately, the applicability of these studies to co astal systems is limited. Seawater exposes microbes to not only increased EC and ionic stress, but also excess SO 4 2 One study incubating freshwater wetland soils with varying concentrations of artificial seawater or
54 salt (NaCl). Results indicated the ionic effect of NaCl on CO 2 flux was insignificant, while the SO 4 2 containing seawater cause a short term (~2 weeks) acceleration in CO 2 flux in direct proportion t o the amount of SO 4 2 added (Chambers et al. 2011). In a similar study, 10 ppt seawater added to a freshwater wetland soil doubled the rate of C mineralization and was attributed to increased sulfate reduction and nutrient availability (Weston et al. 2006 b ). Two studies have looked at differences in soil respiration along the existing salinity gradient in Barataria Bay, Louisiana. Both found that the TFM had the highest rate of CO 2 emissions (618 g CO 2 C m 2 y 1 ), followed by the salt marsh (418 g CO 2 C m 2 y 1 ), and the brackish marsh (180 g CO 2 C m 2 y 1 ; Nyman and DeLaune 1991; Smith et al. 1983). Differences in organic matter quality and hydrology were cited as important regulators of CO 2 flux in this system (Nyman and DeLaune 1991). There is a ne ed for additional long term studies of in situ CO 2 production along coastal wetland salinity gradients and an investigation of this non linear response of CO 2 flux to salinity. It has long been recognized that CH 4 production is significantly higher in f reshwater systems than saline systems (DeLaune et al. 1983; Whelan 1974). Measurements of methane emissions along a coastal wetland salinity gradient in Louisiana were 4.3, 73, and 160 g CH 4 C m 2 for saltwater, brackish, and freshwater marshes, respectiv ely (DeLaune et al. 1983). Laboratory studies indicate a salinity of ~14 ppt is sufficient for near complete suppression of CH 4 emissions (Chambers et al. 2011). In some cases, low salinit ies (5 ppt) may actually stimulate CH 4 production in TFM soils (We ston et al. 2011). Field studies indicate the relationship between salinity and CH 4 production is complicated by heterogeneity in soil, vegetation, sulfate
55 availability, and methane oxidation (King and Wiebe 1980). High rates of CH 4 emissions (1.0 to 6.7 x 10 3 kg y 1 ) have been documented in some fully saline estuarine and mangrove wetlands (Purvaja and Ramesh 2001). A recent meta analysis of in situ CH 4 flux rates from numerous coastal marshes revealed significantly lower CH 4 flux rates in marshes wit h salinities >18 ppt (1.1 2 g m 2 y 1 ), and generally lower rates with salinities of 5 18 ppt (16.4 11 g m 2 y 1 ), compared to freshwater (<0.5 ppt) marshes (41.9 76 g m 2 y 1 ). However, oligohaline (0.5 to 5 ppt) marshes had the highest emission ra tes (150 221 g m 2 y 1 ; Poffenbarger et al. 2011 ) This log linear relationship between salinity and CH 4 flux has been observed in 2 separate studies (Poffenbarger et al. 2011; Bartlett et al. 1987), but the reasons for it remain unclear. In sum, alt hough our understanding of how inundation and salinity impact CO 2 and CH 4 emissions is growing, there is still a great deal of uncertainty. Specifically, there is evidence of a dip in CO 2 flux and increase in CH 4 flux in brackish wetlands. How wide sprea d this phenomenon is, and what factors are contributing to it, remains unknown. Carbon E xport Of the coastal wetland C that is not mineralized by microbes, some portion is exported. The greater the tidal range, the greater the proportion of C exported fr om the wetland to the marine food web (Twilley et al. 1992). Mangroves are thought to lose more litterfall to tidal export (10 to 50 % ) than TFMs, which are estimated to lose only 5 10 % mainly to leaching (Neubauer et al. 2000). One study estimated that o nly 1 to 5 % of autochthonous C actually remains in a salt marsh, and the remaining 95 99 % is
56 exported to the ocean (Middelburg et al. 1997). In general, C export decreases with distance from the coast and tidal amplitude. Carbon Accretion and S torage A s ignificant amount of both organic and inorganic C is stored long term in coastal wetland soils. It is estimated that globally, mangroves and salt marshes store 44.6 Tg C y 1 (Chmura et al. 2003). In North America alone, estuarine wetlands store 10.2 Tg C y 1 approximately ~31 % of the total wetland C accumulated annually (Bridgham et al. 2006). A recent review of the potential C sequestration capacity of and 226 39 g C km 2 y 1 respectively (Mc L eod et al. 2011). Annual C sequestration can be calculated as the vertical accretion rate, multiplied by the soil bulk density and soil C content (Armentano and Menges 1986). Generally, soil accretion rates are greatest in wetlan ds in warm climates with high moisture and nutrient availability to fuel primary production (Armentano and Menges 1986). Vertical accretion is one way wetlands can cope with SLR and is likely to increases the overall C sequestration rate ( Figure 2 3 ). Ver tical accretion rates, however, do not directly mimic SLR rates. There is a lag period between the onset of increased inundation and the establishment of a new elevation equilibrium. In a step wise sea level increase, marshes can lose 10+ cm of elevation before they equilibrate; sometimes requiring several decades to reach equilibrium, depending on the sediment supply (Kirwan and Temmerman 2009). Under conti nuous SLR, the marsh platform is always moving toward equilibrium, but never quite reach es it (Fag herazzi et al. 2012). Coastal wetlands located in regions with larger tidal amplitudes are expected to be more resilient to SLR because of the constant influx of nutrient and sediment rich
57 tidal water (Nicholls et al. 1999). The high rate of sediment d eposition and higher than average allocation of C to belowground biomass by salt marsh plants can significantly enhance salt marsh accretion rates (Chmura et al. 2003). A review of coastal wetland a ccretion rates across the globe rang ed from 0 to 43 mm y 1 (DeLaune and White 20 11 ). Sediment accumulation has high spatial variability (due to differences in vegetation density and distance from tidal creeks) and temporal variability (based on storm frequency and sea level variations). The temporal variabil ity was exemplified in a study of sedimentation rates in a New England salt marsh where over a 3 year period rates ranged from < 5 mm y 1 to 24 mm y 1 depending on local storm activity (Roman et al. 1997). Accretion rates and C accumulation rates are ge nerally thought to decrease from TFMs to salt marshes, possibly due to higher organic matter export in saline wetlands (Craft 2007). A review of data on soil accretion rates in TFM along the US Atlantic and Gulf coasts found a range of 0.11 to 2.19 cm y 1 (Neubauer 2008). In mangroves, accretion rates differ according to geomorphology. Twilley et al. (1992) estimated that reef mangrove accretion is less than 1 mm y 1 while basin mangroves and riverine mangrove communities accrete 1 2 mm y 1 and > 2 mm y 1 respectively. Spatial variability of accretion rates are also observed in salt marshes, especially between low marsh areas (117 g C m 2 y 1 ) and high marsh areas (65 g C m 2 y 1 ; Choi and Wang 2004). It is clear that there are a great many site speci fic, spatial, and temporal factors to consider when estimating C accumulation in coastal wetlands. Despite the patterns described above, a study along a coastal salinity gradient in Louisiana revealed no
58 significant difference in annual C accumulation bet ween TFM, brackish, and salt marsh sites, with rates ranging from 183 to 296 g C m 2 y 1 (Smith et al. 1983). Summary and Research Gaps Coastal wetlands represent a significant reservoir of global C, sequestering an estimated 210 g CO 2 m 2 yr 1 an ord er of magnitude greater than peatlands, while also providing such important ecosystem services as nutrient cycling, erosion control, storm abatement, and wildlife habitat (Choi and Wang 2004; Chmura et al. 2003). With global sea level rising at a rate of ~3 mm y 1 the fate of intertidal wetlands depends upon their ability to accrete vertically or migrate laterally. If neither of these options is achievable, the wetland will submerge. Close to 5,000 km 2 of coastal marsh in Louisiana has converted to ope n water since 1930 and predictions for global coastal wetland loss due to SLR range from 22 to 30 % by 2100 (DeLaune and White 2011; IPCC 2007; Nicholls et al. 1999). Our ability to understand the fate of the large quantity of soil organic C stored in coas tal wetlands hinges upon our knowledge of how biogeochemical C cycling will be impacted by seawater encroachment into previously fresh, intermediate, and brackish marshes. Saltwater intrusion produces a few predictable changes to the physical wetland en vironment, such as increased mineral sediment deposition, increased soil bulk density, a shift to less diverse salt tolerant vegetation, and a widening and incising of tidal channels. Chemically, the electrical conductivity will increase, causing many of the cations previously held to the soil exchange complex, such as NH 4 + and Fe 3+ to be released into solution. Phosphorus availability increases as Fe precipitates with the plentiful S 2 possibly causing the limiting nutrient to shift from P to N. But t he principal chemical change brought about by seawater is an increase in the abundance of SO 4 2
59 which decreases the electron pressure and causes sulfate reducing bacteria to proliferate at the expense of methanogens and fermenting bacteria. There is also evidence that increasing salinity can cause saprophytic fungi populations to decline, decreasing the ability to breakdown recalcitrant C compounds. Soil microbes appear to be very resilient to ionic stress, with community composition shifting quickly t oward more salt tolerant species without a detectable decrease in overall community size. Available data suggest that after the initial stress response passes and soil microbes adapt to a higher salinity regime, they produce less CO 2 synthesize fewer enz ymes, and decompose organic matter at a slower rate. However, the slower rate of microbial C cycling in saline wetlands does not necessarily translate into greater C storage under SLR. On the contrary, saline wetlands assimilate less C than freshwater we tlands to begin with, have a higher rate of C export, and are the most vulnerable to submergence This review illuminates many aspects of the biogeochemical response of C cycling to SLR that need additional research. A more detailed study of soil bacteri al abundance, microbial biomass C, and enzyme activity along existing salinity gradients is needed to sort out the current contradictions in the literature. Also, the idea that nitrifiers and denitrifiers experience physiological stress in saline conditio ns, thus reducing the rate of N cycling, needs to be validated with field data, and the physiological mechanism needs to be discerned. Finally, much of the current literature suggests an underlying question how important is the time scale to this discussi on? Several of the chemical and biological responses discussed herein are the result of short term studies involving laboratory manipulations, such as the impact of increasing
60 salinity on a freshwater wetland soil. The results of these studies can be dra matic: a 50 fold increase in porewater NH 4 + within 1 month of saltwater re introduction (Portnoy and Giblin 1997); a tripling of soluble Fe 2+ following a 10 ppt seawater addition to a freshwater soil (Weston et al. 2006 b ); and a 20 % increase in microbial r espiration with 3.5 ppt seawater (Chambers et al. 2011). But we do not know how persistent these responses are in nature. Will they lessen as a system becomes adapted to a new salinity regime? In tidal freshwater marshes specifically, where abundant hig h quality organic C tends to accumulate due to a dearth of alternative electron acceptors, what percent of that labile C will be mineralized once an influx of SO 4 2 occurs, and will the mineralization rate slow once a new equilibrium is established ? While there have been recent many advances recently in our understanding of the ecomorphological processes that led to wetland submergence at the saltwater end (Fagherazzi et al. 2012), the biogeochemical controls on C balance during the processes of marsh migr ation and vertical accretion are critical to enhancing our ability to model and predict how SLR will impact the C storage capacity of coastal wetlands.
61 Table 2 1 Aboveground biomass estimates for coastal wetlands. Aboveground bio mass data from published literature according to salinity regime (fresh (<0.5 ppt), brackish (5 15 ppt), and salt (>15 ppt). Salinity Dominant Plant Species Location Biomass (g m 2 y 1 ) Reference Fresh Peltandra virginica Virginia, USA 626 127 Neubauer et al., 2000 Fresh Zizaniopsis miliacea Georgia, USA 1400 200 Craft et al., 2008 Fresh Zizaniopsis miliacea, Spartina cynos uroides Georgia, USA 1424 91 Wieski et al., 2010 Brackish Juncus roemerianus Georgia, USA 1712 112 Craft et al., 2008 Brackish Juncus roemerianus North Carolina, USA 812 145 Christian et. al., 1990 Brackish Juncus roemerianus, Spartina alterniflor a Georgia, USA 1715 95 Wieski et al., 2010 Brackish Spartina densifolia (High marsh) Argentina 2599 705 Gonzalez Trilla et al., 2010 Brackish Spartina densifolia (Low marsh) Argentina 1392 790 Gonzalez Trilla et al., 2010 Brackish Spartina patens, Schoenoplectus americanus Maryland, USA 418 44 Langley et al., 2009 Salt Spartina alterniflora Georgia, USA 996 152 Craft et al., 2008 Salt Spartina alterniflora South Carolina, USA 780 50 Morris et al., 2002 Salt Spartina alterniflora Georgia, U SA 994 93 Wieski et al., 2010 Salt Spartina alterniflora Louisiana, USA 1821 Darby and Turner, 2008 Salt Spartina alterniflora (Short) Delaware, USA 785 251 Roman and Daiber, 1984 Salt Spartina alterniflora (Tall) Delaware, USA 1487 74 Roman and D aiber, 1984 Salt Spartina anglica Netherlands 1435 Middelburg et al., 1997 Salt Spartina patens Delaware, USA 1118 351 Roman and Daiber, 1984
62 Figure 2 1 Sea level rise a ffects on coastal ecos ystems over time. A conceptual flow chart of how sea level rise successively impacts the chemical, biological, and geological environment of a coastal wetland. The cumulative result is a potential change in the capacity of the soil to store soil organic c arbon, a vital ecosystem service.
63 Figure 2 2. The generalized carbon cycle in coastal wetlands. Soil microbes play a key role in regulating how much carbon is sequestered in the soils, versus how much carbon is mineralized and released to the atmos phere as CO 2 or CH 4
64 Figure 2 3 Coastal wetland responses to sea level rise. The three general responses of coastal wetlands to sea level rise, the physical setting that often leads to the response, and the fate of carbon under each scenario.
65 Figu re 2 4 Soil carbon and bulk density as a function of salinity Soil organic C content is represented by circles (left y axis) and soil bulk density is represented by triangles (right y axis). R squared values represent the best fit regression for each parameter. Data compiled from Neubauer ( 2008 ) ; Craft ( 2007 ) ; and Nyman et al. ( 1990 )
66 Figure 2 5 Decomposition as a function of salinity. Data represent the relationship between average water salinity (ppt) and decomposition rate, as determined usin g mesh litter bags placed on the soil surface. R squared values represent best fit regression from each study. Filled triangles ( ) represent data from Craft ( 2007 ) open squares ( ) are from Quintino et al. ( 2009 ) and filled ( 2008 )
67 C HAPTER 3 SHORT TERM RESPNSE OF CARB ON CYCLING TO SALINI TY PULSES IN A FRESHWATER WETLAND Background The soil microbial pool is responsible for many ecosystem processes, including the transfer of C from the organic pool (e.g., biomass) to the inorganic pool (e.g., CO 2 and CH 4 ; Wetzel 2001). The rate at which microbes mineralize C is especially important in wetla nds, where the high level of primary productivity means changes in the C source/sink potential of wetlands could have implications for the global C cycle. Salinity is a prevalent environmental stressor with the potential to alter the rate of C cycling in wetlands (Sangiorgio et al. 2008; Wong et al. 2008; Pattnaik et al. 2000). The impact of soil salinity in arid and semi arid regions on crop productivity and nutrient cycling has been studied extensively. Findings indicate that high salt concentrations i n upland and paddy soils can decrease the size of the soil microbial community (Muhammad et al. 2006; Pattnaik et al. 2000), decrease the rate of microbial respiration (Gennari et al. 2007; Muhammad et al. 2006; Pathak and Rao 1998), and decrease the rate of methanogenesis (Pattnaik et al. 2000). Such an overall decrease in the rate of C cycling in these systems is often attributed to osmotic/ionic stress inflicted on the microbial population by increased conductivity in the soil water environment (Franken berger and Bingham 1982; Gennari et al. 2007). Rising sea level and increasing frequency of saltwater intrusion events also cause stress to microbial populations in freshwater wetlands near the coast. It is estimated that during the 20 th century sea level rose ~1.7 mm yr 1 Since 1993, this rate has increased to 2.8 to 3.1 0.7 mm yr 1 (IPPC 2007). Increasing sea level will amplify the impacts of storm surges and increase the area inundated by extreme tidal
68 events (Michener et al. 1997). In contrast to saline conditions caused by high evaporation rates in arid and semi arid inland soils, seawater contains a relatively uniform mixture of several salts, macro, and micro nutrients (Kester 1967). Of particular importance to wetland soils is the abundance of sulfate (SO 4 2 ) in seawater, which can function as an alternative electron acceptor during anaerobic microbial respiration. Sulfate concentrations in seawater are normally between 20 and 30 mM, as compared to concentrations of 0.1 to 0.2 mM o f SO 4 2 observed in freshwater (Capone and Kiene 1988). Studies investigating the effect of seawater on C cycling have historically been performed along existing salinity gradients in estuaries and coastal zones, or using intact soil cores to measu re long term fluxes. In Louisiana coastal wetlands, microbial respiration rates are highest in freshwater wetlands, followed by salt marshes, and lowest in brackish wetlands (Nyman and DeLaune 1991; Smith et al. 1983). Methane production rates are signif icantly lower in salt marshes, compared to freshwater wetlands, because the abundant SO 4 2 is more energetically favorable for anaerobic respiration than methanogenesis ( Magenheimer et al. 1996; Bartlett et al. 1987; King and Wiebe 1980). It has been foun d that within 12 days of 10 ppt seawater addition to a freshwater tidal marsh soil in Georgia, USA, the dominant microbial pathway switched from methanogenesis to s ulfate reduction (Weston et al. 2006 b ). Despite the extensive knowledge regarding competiti on between methanogens and sulfate reducing bacteria, few studies have directly addressed the mechanistic process by which salinity exposure influences microbial mineralization pathways and rates. Specifically no studies have attempt to distinguish betwe en the importance of ionic stress (increased conductivity),
69 compared to the addition of the sulfate electron acceptor, in altering C mineralization rates following a saltwater pulse. This is important for evaluating the relevance of the trends observed fo r C cycling in agricultural saline soils (i.e., the documented decrease the size, activity, and composition of the microbial population) to coastal soils subjected to saltwater pulses. This study sought to determine the process by which salinity affe cts C cycling in a freshwater wetland soil. Specifically, is the microbial community inhibited by ionic stress and/or the addition of sulfate, and what effect does this have on the overall rate of C cycling? This was done by comparing how microbial respi ration, methanogenesis, and microbial population size respond to different concentrations of seawater (containing SO 4 2 ) and salt (strictly NaCl) additions. I hypothesized that potential respiration rates would be reduced by additions of both seawater an d NaCl due to increased ionic stress to the microbial community, but potential methanogenesis would be reduced to a much greater extent in the seawater treatments due to competition with sulfate reducers. I also anticipated a reduction in the size of the microbial population with increasing concentrations of both seawater and NaCl. Methods Experimental Design A bulk field composite peat soil sample (0 Marsh Conservation Area (27.91833 N, 80.77389 W), a freshwater wetla nd dominated by an even mix of Typha spp and Salix spp (Figure 3 1). Upon return to the lab, the soil was homogenized and approximately 15 g wet weight soil was added to 70 mL glass serum bottles. Seven treatment types were evaluated. They consisted of seawater at concentrations of 35 ppt, 14 ppt, and 3.5 ppt, NaCl at concentrations of 35
70 ppt, 14 ppt, and 3.5 ppt, and a de ionized (DI) water treatment to serve as the freshwater control. The seawater and NaCl treatments functioned as discrete analogues of ionic stress (measured in ppt) while allowing for the isolation of the sulfate reduction effect on C mineralization. All treatments were prepared in triplicate. Fifteen mL of randomly assigned treatment solution was added to each bottle to form a soil slurry. Three concentrations of seawater treatments were made using Neomarine Reef Salt mix (Brightwell Aquatics, Elysburg, PA). Thirty five grams of salt mix was diluted in 1 L of DI water to create 35 ppt seawater. The solution was purged with ambien t air for several hours to establish the CO 2 /HCO 3 equilibrium, and then pH and specific conductivity (mS cm 1 ) were measured. The seawater solution was further diluted with DI water to 14 ppt and 3.5 ppt seawater treatments and pH and specific conductivit y were again noted. The ionic content of the artificial seawater mimicked that of natural seawater without any additional nutrients or C. Sulfate was the only available electron acceptor in the seawater treatments. Three levels of NaCl treatments were m ade by diluting 35 g crystalline NaCl in 1 L DI water. The solution was purged with ambient air for several hours to establish the CO 2 /HCO 3 equilibrium, and then pH and specific conductivity were measured. The NaCl solution was further diluted with DI wat er to 14 ppt and 3.5 ppt NaCl treatments and pH and specific conductivity were noted. Bottles were capped with butyl stoppers and aluminum crimp caps, evacuated to 75 kpa, and flushed with O 2 free N 2 gas for 1 minute to create anaerobic conditions. Inc ubation bottles were then placed in the dark on a circulating shaker at 30C.
71 Headspace was extracted and measured on a gas chromatograph (Shimadzu Scientific Instruments GC 8A, Columbia, MD) fitted with a TCD detector and FID detector to determine the co ncentrations of CO 2 and CH 4 respectively. Headspace samples were measured after 1, 2, 4, and 6 days to produce a rate of production per g soil, per day. All bottles were then purged with O 2 free N 2 gas for 1 minute to prevent CO 2 accumulation in the hea dspace, and the sampling cycle repeated again. The sampling and purging sequence was repeated for three weeks and the rates of potential CO 2 C and CH 4 C production were calculated over time. Soil Properties Bulk density was determined after oven drying a known volume of subsample at 70 C until constant weight. Percent organic matter (% OM) was determined using the loss on ignition (LOI) method (Nelson and Sommers 1996). Three grams of triplicate ground, dried soils were placed in a muffle furnace at 550 C for 4 hours, cooled, and re weighed. Percent weight LOI was calculated as the difference between the soil weight before and after ashing, multiplied by 100. Soil pH and specific conductivity were measured on all samples after completion of the three week incubation period. A 2:1 (water:soil) suspension was created and allowed to equilibrate for 30 min before measurement (Thomas 1996; US EPA 1983). pH was measured using an Accument Research pH meter AR50 (Fisher Scientific, Waltham, MA) and specific conductivity was measured on a Markson EC Meter model 1054. Total extractable organic C (TOC), extractable organic C (OC), and microbial biomass C (MBC) were determined for all samples following the three week incubation period using the fumigation extract ion method after Vance et al. (1987). TOC was
72 defined as the extractable organic carbon extracted from the fumigated samples and OC was defined as the extractable organic C extracted from of the non fumigated samples. Microbial biomass C (MBC) was determ ined by subtracting the extractable C of the non fumigate from the corresponding fumigate sample. Duplicate 5 g wet weight samples were prepared in 25 mL centrifuge tubes. One set was fumigated with chloroform for 24 h and the other set served as the non fumigated control. Following the chloroform treatment, both fumigates and non fumigates were extracted with 25 mL of 0.5 M K 2 SO 4 agitated for 30 min on a circulating shaker, and centrifuged at 5000 rpm for 10 min. The supernatant was vacuum filtered th rough a Whatman # 42 filter paper and stored at 4 C until analyzed for total organic carbon (Shimadzu Scientific Instrument TOC 5050A, Columbia, MD). An extraction efficiency coefficient of k EC = 0.37 was applied to all samples (Sparling et al. 1990). D ata Analysis Statistical analysis was performed using SAS 9.1 software (SAS Institute Inc., Cary, NC). All data sets were first tested to determine if the assumptions of Wilk Test, re spectively. Where these assumptions were not met, the raw data was log transformed and further statistical analysis was conducted using the dataset that fulfilled the assumptions of homogeneity and normality. A two way repeated measures ANOVA 0.05) was used to determine the interaction between CO 2 /CH 4 production, treatment, and time. Significance differences were identified using the Least Squares Means (LSM) post hoc test. One identify significan t differences between pH, conductivity, sulfate concentration, NaCl
73 correlations were performed to determine if correlations exist between CO 2 /CH 4 production and sulfate con centration, NaCl concentration, pH, conductivity, extractable C indicators, and microbial population indicators. Results Soil Properties The soil consisted of a flocculant peat and had a bulk density of 0.097 0.01 g cm 3 and an organic matter content of 55 0.6 % Initial soil pH was 6.82 0.02; at the conclusion of the experiment, soil pH was highest for the 35 ppt seawater treatment (7.15 0.03) and lowest for the 35 ppt NaCl treatment ( 6.41 0.04). The pH of the 3.5 ppt seawater, 3.5 ppt NaCl and freshwater control treatments did not differ from the intial pH (Table 3 1). Specific conductivity ranged from 28.5 0.4 mS cm 1 to below the detection limit (0.3 mS cm 1 ), for the 35 ppt NaCl and the freshwater control, respectively ( Table 3 1 ). Specific conductivity was significantly different for all treatments (p < 0.01) except the 3.5 ppt seawater and 3.5 ppt NaCl treatments. Sulfate and NaCl concentrations were significantly different between all treatments ( Table 3 1 ). Potential Microbial R espiration The rate of potential CO 2 production was significantly higher in the 35 and 14 ppt seawater treatments, compared to the freshwater control and the NaCl treatments, during week 1 (p < 0.01), but did not differ significantly during weeks 2 and 3 (Figure 3 2). There were no significant differences in CO 2 production rate between the NaCl treatments throughout the entire study. Considering all the treatments, time was a significant factor for CO 2 production (p < 0.001), as well as the interaction b etween time and treatment (p < 0.001). Significantly greater rates of microbial respiration occurred during week 1 for all treatments, compared to weeks 2 and 3 (Figure 3 2). However,
74 the contribution of sulfate reduction to respiration decreased over ti me, as seen by the difference in CO 2 production between the seawater treatments and the freshwater control. During week 1, it can be estimated that 44 % of anaerobic respiration was mediated by sulfate reduction, while sulfate reduction accounted for 21 % a nd 15 % of respiration during weeks 2 and 3, respectively. The total amount of CO 2 produced over the three week incubation period was significantly greater (p < 0.01) for all the seawater treatments, compared to the freshwater control (Figure 3 3 ). Total CO 2 production was 32 % higher in the 35 ppt seawater treatment, compared to the freshwater control, 29 % higher in the 14 ppt seawater treatment, and 20 % higher in the 3.5 ppt seawater treatment. Total CO 2 produced by the NaCl treatments did not differ fro m the freshwater control (Figure 3 3 ). The variables that correlated with CO 2 production differed between the seawater and NaCl treatments. Seawater respiration rates were positively correlated (p < 0.01) with indicators of extractable C (TOC and OC; Tab le 3 2). Respiration in NaCl treatments was positively correlated with pH and negatively correlated with conductivity and organic C (p < 0.01; Table 3 2). Potential Methanogenesis The rate of potential CH 4 production was significantly lower (p < 0.001) for the 35 ppt and 14 ppt seawater treatments, compared to the 3.5 ppt seawater and freshwater control for all three weeks (Figure 3 4). The 35 and 14 ppt NaCl treatments also had a significantly lower rate of CH 4 production during weeks 1 and 2, compare d to the 3.5 ppt NaCl and freshwater control, but did not differ significantly during week 3 (Figure 3 4). Time was not a significant factor for CH 4 production rate.
75 The total amount of CH 4 produced over the three week incubation period was significantly less for the 35 and 14 ppt seawater and NaCl treatments, compared to the freshwater control (Figure 3 5). Total methane production was 94 % and 79 % lower in the 35 and 14 ppt seawater treatments, respectively, compared to the freshwater control. Total met hane production was reduced by 55 % and 23 % in the 35 and 14 ppt NaCl treatment, respectively, compared to the freshwater control (Figure 3 5). Neither the seawater nor NaCl 3.5 ppt treatments differed from the freshwater control in methane production. In contrast to respiration, methane production in the seawater treatments was not correlated with indicators of extractable C, but did show a significant negative correlation with pH and conductivity (Table 3 2). NaCl treatments showed a significant positive correlation with pH and negative correlation with conductivity (p < 0.01 ; Table 3 2). The negative correlation between NaCl and organic C suggests methanogens were not C limited (Table 3 2). Sulfate V ersus NaCl Sulfate (i.e., seawater treatments) and Na Cl additions had significantly different effects on potential respiration, potential methanogenesis, and other variables of interest. Sulfate in high concentrations produced a short term increase in microbial respiration (Figure 3 2). This relationship b etween sulfate and CO 2 was also observed as a significant positive correlation between these variables (r = 0.64; p < 0.01 ; Table 3 3 3). In contrast, methanogenesis decreased as sulfate concentration increased (Figure 3 4). A significant negative corre lation existed between CH 4 and sulfate concentration (r = 0.80; p < 0.01; Table 3 3). Sulfate also showed a strong (p < 0.01) positive correlation with the ratio of CO 2 :CH 4 production and pH (Table 3 3 ). The ratio
76 of CO 2 :CH 4 was similar for all NaCl tre atments, the 3.5 ppt seawater treatment, and the freshwater control (2.0 0.5), but significantly higher for the 14 ppt seawater (11.2 1.4) and 35 ppt seawater (43.3 4.0). The NaCl addition had no effect on potential respiration (Figure 3 2), nor was there a correlation between NaCl concentration and respiration (Table 3 3 ). The affect of the NaCl addition on methanogenesis was slightly greater than for respiration, with the 35 and 14 ppt NaCl treatments reducing CH 4 production for 2 weeks (Figure 3 4). Sodium chloride concentrations > 13 mg L 1 may negatively impact methanogenesis, but concentrations below this (as seen in the 3.5 ppt seawater and NaCl treatments) were not correlated with methanogenesis (Table 3 3 ). The main affect of NaCl was a s ignificant increase in conductivity, total extractable organic C, extractable organic C, and pH (Tables 3 1 and 3 3) Discussion High concentrations of seawater (14 and 35 ppt) caused a significant increase in pH, while high concentrations of NaCl (14 and 35 ppt) decreased pH (Table 3 1). The increase in pH caused by seawater is likely a result of the high CaCO 3 content of the seawater mix and a product of sulfate reduction, while strictly NaCl additions may have displaced H + ions from the cation exchange complex and caused the pH to decrease. A similar displacement of ammonium ions from sediments by NaCl has been observed (Baldwin et al. 2006) and is further supported by the strong negative correlation between pH and NaCl concentration in this study (Tabl e 3 3). While it is unlikely that this near neutral range in pH between treatments (6.4 to 7.2) could cause microbial inhibition, it is an interesting side effect of salinization that could select for specific microbial species in the long term. Specifi c conductivity significantly increased in all 14
77 and 35 ppt treatments (Table 3 1), but did not mirror the conductivity of the added solution due to a dilution effect by the soil pore water. Conductivity was most strongly correlated with NaCl concentratio n (Table 3 3 ). Potential Microbial Respiration In general, seawater additions had a stimulatory effect on potential CO 2 production rate (Figure 3 2) and the total amount of CO 2 produced was directly related to the concentration of seawater added (Figure 3 3). Over the three week incubation, total CO 2 production was 32 % higher in the 35 ppt seawater treatment, 29 % higher in the 14 ppt seawater treatment, and 20 % higher in the 3.5 ppt seawater treatment, as compared to the freshwater control. However, the st imulation of respiration rate was short lived; all seawater treatments returned to a CO 2 production rate similar to the freshwater control by week 3 (Figure 3 2). The difference in the magnitude of the increase in CO 2 production by the seawater treatments can be attributed to the increased availability of sulfate to serve as a terminal electron acceptor in anaerobic microbial respiration. Using the theoretical relationship of 1 mole SO 4 2 :2 moles CO 2 (Equation 3 1) indicates only ~20 % of the sulfate added was reduced. Therefore, sulfate depletion was also not responsible for the decline in activity over time in the seawater treatments. C 6 H 12 O 6 + 3SO 4 2 + 6H + = 3H 2 S + 6CO 2 +6H 2 O ( 3 1) All of the treatments produced significantly more CO 2 the first week, compared to weeks 2 and 3 (Figure 3 2). This could be a limitation of the experimental design. Since CO 2 measurements were taken from a closed system with a finite supply of bioavailable C, over time the substrate limitation surpassed t he alternative electron acceptor availability in regulating the rate of microbial activity. The agitation of the soil
78 slurry may have exaggerated the initial pulse of available electron donors by releasing C compounds previously protected with soil aggreg ates. Regardless, since all incubations were composed of replicate soils substrate, the difference in response between treatments can be attributed to the concentration of sulfate electron acceptors. The positive correlation between CO 2 production and indicators of extractable C (p < 0.05) support the hypothesis of a C limitation. A significant decrease in C flux rate over time has also been observed in studies of intact soil cores, with the decline in CO 2 and CH 4 production attributed to progress ive C limitation (Weston et al. 2011). Previous work has indicated that the percent of microbial respiration mediated by sulfate reduction increases as total respiration increases (Howarth 1984). The present study also found a logarithmic increase in the per cent of respiration attributed to sulfate reduction as respiration rate increased. When respiration rates were highest (week 1), 44 % of respiration was mediated by sulfate reducers, and when respiration was low (week 3), only 15 % of respiration was mediat ed by sulfate reducers. This is owed to the assumption that SO 4 2 is the only alternative electron acceptor present in seawater that is not present in the freshwater control or NaCl treatments, and the fact that the soils were maintained under anaerobic c onditions from the time of collection to the conclusion of the study. In contrast to seawater, the NaCl addition had no effect on CO 2 production (Figure 3 2). Since NaCl does not function as an electron carrier as sulfate does, we can conclude that ionic stress alone does not affect microbial community respiration. However, these findings do not address whether microbial community structure or diversity was altered by the NaCl addition. Baldwin et al. (2006) performed a detailed
79 analysis of microbial co mmunity structure using phospholipid fatty acid (PLFA) biomarkers and discovered that NaCl > 5 mS cm 1 decreased microbial diversity, but did not alter microbial biomass. No significant differences in microbial biomass C between NaCl treatments were ident ified in the present study either (Table 3 1). Our findings suggest the tolerance of anaerobic microbes to ionic stress may be higher than initially anticipated, and the microbial community may have the ability to adapt to increased ionic stress within th e period of 1 week (Figure 3 2). The high species richness of freshwater sediments is believed to allow the community to switch biochemical pathways in a matter of days (Edmonds et al. 2009). Potential Methanogenesis Sulfate reduction is thermodynam ically preferred over methanogenesis because of the higher net energy yield for obligate anaerobes (Capone and Kiene 1988). The higher concentrations of sulfate in seawater led to the hypothesis that methane production would be lower in seawater wetlands, as compared to freshwater wetlands. This has been confirmed by numerous studies (e.g., Abril and Iversen 2002; Purvaja and Ramesh 2001; Reeburgh and Heggie 1977). However, some research has found maximum methane emissions occur at intermediate salinitie s ( Sotomayor et al. 1994; Bartlett et al. 1987) and may still be substantial in saltwater wetlands with high carbon inputs (Biswas et al. 2007; Purvaja and Ramesh 2001). By comparing the repression of CH 4 in seawater and NaCl treatments, this study was a ble to differentiate between the effect of sulfate competition and ionic stress on methanogenic microbes. Low concentration (3.5 ppt) of seawater and NaCl did not affect methane production (Figure 3 4). Seawater additions of 14 ppt and above did significa ntly, and persistently, reduce methanogenesis (Figure 3 4). Other work has
80 proposed a salinity of > 13 ppt is required to alter CH 4 flux (Bartlett et al. 1987). Total methane production in this study was 94 % and 79 % lower in the 35 ppt and 14 ppt seawater treatments, respectively, compared to the freshwater control. A strong negative correlation (p < 0.01) between sulfate concentration and CH 4 production (Table 3 3) suggests sulfate reduction replaced methanogenesis as the main form of anaerobic respiratio n. However, the increase in CO 2 cannot be directly calculated from the decrease in CH 4 because of the use of competitive and non competitive substrates between the two groups of anaerobes (Capone and Kiene 1988). An inverse correlation between conducti vity and CH 4 production has been found previously by Magenheimer et al. (1996). NaCl decreased CH 4 production, but to a lesser extent than the seawater did (Figure 3 4). Overall, the 35 ppt NaCl reduced CH 4 production by 55 % and 14 ppt NaCl reduced CH 4 production by 23 % relative to the freshwater control. This repression of methanogenesis by NaCl is slightly less than found by Baldwin et al. (2006) where at little as 1 mS cm 1 decreased CH 4 production by 30 % over 1 month. Although the soil slurry de sign used herein limits the interpretation of the CH 4 flux rates to an estimation of potential methanogenesis, it does provide evidence for a differential sensitivity of methanogens to salt. The fact that the decline in CH 4 production was not directly cor related with the increase in CO 2 production suggests heterotrophic methanogens, rather than autotrophic (CO 2 /H 2 using) methanogens, dominate in this soil and were most strongly affected by salt additions. Other work suggests high concentrations of NaCl wi ll inhibit acetoclastic (heterotrophic)
81 methanogens (Baldwin et al. 2006), which may have been driven the short term decrease in CH 4 production in the 35 and 14 ppt NaCl treatments (Figure 3 4). Summary This study used laboratory soil slurry incubatio ns to assess the short term effects of NaCl and seawater on anaerobic C cycling in a freshwater wetland soil. These idealized conditions (lack of diffusion barriers, constant redox conditions, and the exclusion of alternative electron acceptors) allowed f or the isolation of the two opposing biogeochemical forces that act on coastal wetland soils subjected to a pulse of seawater: ionic stress and sulfate induced respiration. Findings indicate the concentration (ppt) of the seawater being introduced to the freshwater soil is the critical factor in determining the impact on soil C cycling. Oligohaline seawater (3.5 ppt) accelerates overall C mineralization through the combined production of CO 2 C and CH 4 C (Figure 3 6), thus enhancing the rate of organic C d ecomposition. This occurs as a result of the short term acceleration of sulfate reduction without the inhibition of methanogenesis. The overall C mineralization rate was 17 % higher in the 3.5 ppt seawater treatment than the freshwater control (Figure 3 6 ). Mesohaline and haline concentrations of seawater (14 and 35 ppt) also produced a short term stimulation of anaerobic respiration, but the effect was off set by a decrease in methanogenesis (Figure 3 6). Although the effects on the C cycle observed in this study were temporary (1 2 weeks), the increased frequency of storm surges and extreme tidal events in coastal wetlands that are expected to accompany sea level rise, makes these findings significant. Additionally, the fact that the microbial response was temporary indicates coastal wetland soils than gradual sea level rise.
82 The change in the CO 2 :CH 4 ratio following seawater intrusion may have significant implicati ons for global warming. Assuming a CO 2 equivalent radiative forcing of 25 for CH 4 (IPPC 2007), mid salinity wetlands (14 ppt) have a 72 % lower global warming potential (GWP) than freshwater wetlands, and high salinity wetlands (35 ppt) have 86 % lower GWP than freshwater wetlands. Coastal wetlands in the contiguous United States are estimated to sequester 10.2 Tg C y 1 ; equivalent to 31 % of the total C sequestered in all contiguous U.S. wetlands (Bridgham et al. 2006). With sea level rise occurring at ~3 mm y 1 (IPPC 2007), the gently sloping coastal zone of the Atlantic Ocean and Gulf of Mexico, USA, are already experiencing seawater encroachment into previously fresh and low salinity wetlands (e.g., Hussein 2009; Donnelly and Bertness 2001; Williams et al. 1999). The results of this study suggest the biochemical effects of seawater intrusion, especially pulsing events, on organic C mineralization in coastal wetlands may require a re evaluation of the carbon balance of coastal wetlands in light of predi cted sea level rise.
83 Table 3 1 Soil and water properties according to treatment condition. Mean standard deviation of soil pH and specific conductivity measured following destructive sampling after the 3 week incubation. Sulfate and NaCl con centration in each treatment per g dry soil. Different letters indicate significant differences at p < 0.01. Treatment pH Conductivity ( mS cm 1 ) Sulfate ( mg g 1 ) NaCl (mg g 1 ) MBC (g g 1 ) Seawater 35ppt 7.15 0.03 a 25.0 1.2 a 30.4 3.9 a 120.4 15.5 a 371 53 Seawater 14ppt 6.9 0.06 b 11.3 0.4 b 12.2 0.6 b 48.1 2.4 b 533 84 Seawater 3.5ppt 6.78 0.05 c 2.4 0.5 c 3.5 0.3 c 13.0 1.0 c 415 118 NaCl 35ppt 6.41 0.04 d 28.5 0.4 d BD d 421.7 9.4 d 447 97 NaCl 14ppt 6.54 0.01 e 13.0 0. 5 e BD d 165.8 14.2 e 596 171 NaCl 3.5ppt 6.72 0.01 c 3.64 0.1 c BD d 4.2 0.3 f 498 138 Freshwater Control 6.82 0.01 c BD f BD d BD f 467 32 BD = Below Detection; MBC = Microbial Biomass Carbon Table 3 2 Correlations between soil parameters and C loss. correlation coefficients (r values). For all values, n = 9; df = 7; at r = 0.67, p = 0.05 (*); at r = 0.80, p = 0.01 (**); at r < 0.67, NS = Not Significant. Soil Parameter Treatment CO 2 Production CH 4 Production pH Seawater N S 0.75* NaCl 0.84** 0.94** Conductivity Seawater NS 0.81** NaCl 0.90** 0.96** Total organic C Seawater 0.76* NS NaCl NS 0.72* O rganic C Seawater 0.85** NS NaCl 0.82** 0.93** Microbial biomass C Seawater NS NS NaCl NS NS
84 Table 3 3 C orrelations between soil and water parameters correlation coefficients (r values). For all values, n = 21; df = 19; at r = 0.44 p = 0.05 (*); at r = 0.55, p = 0.01 (**); at r < 0.44, NS = Not Significant. Parameter Sulfate NaCl Cond uctivity pH mg g 1 mg g 1 mS cm 1 CO 2 0.64** NS NS 0.71** CH 4 0.80** NS 0.61** NS CO 2 :CH 4 0.97** NS 0.54* 0.76** Conductivity 0.48* 0.83** NS pH 0.82** 0.59** NS Total organic C NS 0.69** 0.54* NS O rganic C NS 0.84** 0.82** NS
85 Figure 3 1 Site location map. The l ocation of freshwater marsh soils collected for this manipulative laboratory experiment was in east central Florida, USA
86 Figure 3 2 Potential soil respiration over time. Anaerobic microbial respiration rate by treatmen t over the 3 week incubation period. Error bars represent standard deviation; n = 3 for all treatments; different letters indicate significant differences at p < 0.01.
87 Figure 3 3 Total soil respiration Mean total anaerobic CO 2 produced over the 3 week incubation period by treatment and concentration. Horizontal lines represent mean (solid line) and standard deviation (dotted line) of total production by the freshwater control. Percentages represent differences in mean total production compared to the freshwater control (p < 0.05). Error bars represent standard deviation; NS = not significant; n = 3 for all treatments.
88 Figure 3 4 Potential methanogenesis over time. Methanogenesis rate by treatment over the 3 week incubation period. Error bars represent standard deviation; n = 3 for all treatments; different letters indicate significant differences at p < 0.01 for week 1, p < 0.05 for weeks 2 and 3.
89 Figure 3 5 Total methanogenesis. Mean total CH 4 produced over the 3 week incubat ion period by treatment and concentration. Horizontal lines represent mean (solid line) and standard deviation (dotted line) of total production by the freshwater control. Percentages represent differences in mean total production compared to the freshwat er control (p < 0.05). Error bars represent standard deviation; NS = not significant; n = 3 for all treatments.
90 Figure 3 6 Total carbon loss. Total C (CH 4 and CO 2 ) produced over the 3 week incubation period by treatment and concentration. Horizontal lines represent mean (solid line) and standard deviation (dotted line) of total production by the freshwater control. Percentages represent differences in mean total production compared to the freshwater control (p < 0.05). Error bars represent standard deviation; NS = not significant; n = 3 for all treatments.
91 CHAPTER 4 EFFECT OF SALINITY P ULSING EVENTS ON SOI L CARBON LOSS ALONG AN INTERTIDAL WETLAND G RADIENT: A LABORATOR Y EXPERIMENT Background Extreme sea level events (astronomical tides, frontal sy stems, tropical storms, and hurricanes) have a greater impact on the environment and society than mean sea level rise (SLR; Michener et al. 1997). With eustatic SLR occurring at a rate of 2.8 to 3.1 mm y 1 storm surges are amplified and the area of land inundated during extreme coastal water levels is enlarged (Mousavi et al. 2011; IPCC 2007; Michener et al. 1997). Approximately 50 % of SLR is ascribed to thermal expansion; global sea surface temperatures are predicted to increase an addition al 1.1 to 6.4 C over than next century (IPCC 2007). Increasing sea surface temperatures are strongly correlated with longer duration and more intense tropical storms (IPCC 2007). In Florida (USA), there is evidence of increasing storm surge frequency ( Park et al. 20 11) It is predicted that the return period of a given coastal surge in Florida will decrease up to ten fold by 2060 (from a 1 in 50 year return int erval to a 1 in 5 year interval) Most studies concerning coastal surges focus on the human and economic costs of flooding and infrastructure loss ( Bloetscher et al. 2011; Park et al. 2011; Frazier et al. 2010 ), but natural coastal ecosystems are also impacted. Storm surges can introduce a pulse of saltwater to the coastal zone, temporarily inundating the ca rbon rich soils of coastal wetlands. Coastal wetlands are a substantial sink for global C, accumulating 30 to 100 kg organic C m 2 in the same time the adjacent upland to accumulates 5 to 10 kg organic C m 2 (Coultas 1996). The rate of soil organic C (SOC ) loss is highly sensitive to salinity due to the ability of sulfate (SO 4 2 ), an abundant constituent of seawater, to function as an alternative electron acceptor during anaerobic respiration by microbes,
92 resulting in the accelerated mineralization of SOC (Chambers et al. 2011; Edmonds et al. 2009). The introduction of seawater to a freshwater wetland can induce a short term increase (20 32 % ) in soil respiration from accelerated sulfate reduction under anaerobic conditions (Chambers et al. 2011). Evidence also suggests the high ionic strength of seawater and the presence of Cl ions can inhibit some soil microbial activity (Seo et al. seawater on soil microbes seems to be neg ligible in comparison to the stimulatory Freshwater pulses result from run off during the heavy precipitation events that often accompany storms, or as point source stormwater discharges into coastal wetlands surro unded by urban development. According to Mulholland et al. (1997), regional climate models for the southeastern USA predict increasing precipitation in Florida, especially in conjunction with large, intense summer storms. The increase in precipitation is expected to exceed evapotranspiration and cause higher run off volumes and more extreme hydrographs in Florida (Mulholland et al. 1997). In addition to climatic changes, extensive coastal urban development in Florida is continually increas ing the coverag e of impervious surfaces, further contributing to episodic freshwater discharges into coastal wetlands. The City of Jacksonville, Florida, has a 1 2 % annual population growth and has lost 36 % of their coastal wetlands since 1943 (US Census Bureau 2009; FD EP 2008). Much of the run surfaces is discharged directly into the adjacent marsh, creating hot spots of freshwater inputs to otherwise brackish or saline intertidal wetlands.
93 This study seeks to characterize and quanti fy the potential loss of ecosystem services (specifically, C storage) resulting from abrupt, short term changes in surface water salinity in the coastal wetlands of Jacksonville, Florida. The goal was to answer three over arching questions : ( 1) h ow do abr upt, short term pulses of higher (or lower) salinity surface water affect the rate of SOC loss in coastal wetland soils, ( 2) d oes the impact of a salinity pulsing event differ between freshwater, brackish, and salt marshes, and ( 3) w hat nutrients, or other soil and water properties that co vary with salinity, influence the response of SOC loss to salinity pulses? I hypothesized that freshwater marsh soils exposed to pulses of low salinities would have an overall higher rate of SOC mineralization due to com bined effects of accelerated sulfate reduction and the maintenance of high in situ rates of methanogenesis, as seen in a previous study (Chambers et al. 2011). I also predicted that salt marsh soils exposed to fresh or low salinity pulses would have moder ately accelerated rates of soil respiration due to the flushing of toxins (e.g., sulfide) and the alleviation of ionic stress for soil microbes. To test these hypotheses, I conducted a laboratory study using intact soil cores collected from 3 intertidal wetlands along a natural salinity gradient and manipulated surface water salinity and tidal cycle to simulate salinity pulsing events. For the purpose of this study, SOC loss was defined as C loss through soil respiration, methanogenesis, and DOC release Since coastal wetlands experience daily tidal fluctuations that alter the balance between aerobic and anaerobic soil respiration, I further specified CO 2 flux as occurring during low tide (LT) or high tide (HT).
94 Methods Study A rea Thirty six intact soi l cores (12 from each of 3 sites) were collected along the natural salinity gradient of intertidal wetlands in the City of Jacksonville, Florida, USA. The sites were chosen based on accessibility and ambient surface water salinity. Soil cores were collec ted from the intertidal marsh platform adjacent to tributaries of the St. 1). The freshwater tidal site was located along Cedar Creek (3026'48.5"N, 8140'17.1"W), the brackish site along Broward River (3026'22.4"N, 8137'33.1"W), and the salt marsh site along Pablo Creek (3018'29.9"N, 8125'9.8"W ; Figure 4 1). Emergent marsh vegetation typical of a sub tropical estuarine wetland dominated each site. The freshwater wetland had the highest species diversity with Sagittaria lancifo lia Zizaniopsis miliacea and Alternanthera philoxeroides being the dominant species. Juncus romerianus and Spartina patens dominated the brackish marsh, and the salt marsh was a monoculture of Spartina alternifloria All three sites were subject to diur nal micro tidal fluctuations averaging 0.7 to 1.0 m in range (NOAA 2011). Experimental D esign The experiment was design ed and analyzed as a 3 x 3 x 3 mixed model treatment design (Table 4 1). Twelve intact soil cores were collected in each of the three w etland types (freshwater, brackish, and salt marsh) in 40 cm long x 10 cm diameter PVC tubes on April 15, 2011. All 12 soil cores from each site were collected within a 6 m 2 plot using care to minimize soil compaction and site heterogeneity. Aboveground vegetation was removed; the cores were capped on top and bottom, and then transported back to the laboratory. At the time of soil collection, 400 L of surface
95 water from the adjacent tidal creek was also collected, field filtered through a 1 micron filte r bag, and transported back to the lab. During sampling, all three sites were near low tide, with the freshwater and salt marsh sites on the ebb tide and the brackish site on the rising tide. Once at the laboratory, the 12 field replicate soil cores were randomly assigned to one of 4 conditions (freshwater, brackish, or saltwater salinity pulses, or immediate analysis ; Table 4 1). Standard window screen mesh was affixed to the bottom of the cores and a 1 cm diameter drain hole was drilled exactly 10 cm ab ove the soil surface (to mai ntain a 10 cm deep water column ; Figure 4 2). The 10 cm deep water column where the soils were collected (NOAA 2011). The bottom of each cor e was plugged and flooded with ambient surface water (i.e., collected from the same site location) and allowed to acclimate for 1 week. Following acclimation, bottom plugs were removed, a leachate collection container was placed under each core, and surfa ce water was allowed to drain through the soil profile and mesh screen for the first 24 hour dry down period (Figure 4 2). After the first dry down, the 9 cores selected for immediate analysis were destructively sampled by sectioning into 3 depth segments (0 5, 5 10, and 10 20 cm). Following acclimation and initial dry down, all remaining 27 cores underwent 3 cycles of 3 to 5 day salinity pulses, punctuated by 2 ambient surface water periods, each lasting ~12 days (Table 4 2). Between each cycle of surfac e water addition, all cores were unplugged, allowed to drain for 24 h, and leachate was collected. Once the
96 53 day manipulation experiment was complete, the remaining 27 cores were destructively sampled. The goal of this design was to investigate both the short term and cumulative impacts of pulsing events in each wetland type. Data from a hurricane storm surge in the Gulf of Mexico indicated that estuarine surface water salinities peak quickly to a maximum of ~25 ppt, and then receded slowly over the cou rse of approximately 4 days (Li et al. 2009). Due to the abrupt, temporary, and dynamic nature of most saltwater (or freshwater) pulses that occur in coastal wetlands, the design included a return to ambient conditions following each salinity pulsing even t (Table 4 2). This allowed for the identification of any possible legacy impacts from the surge on the rate of SOC loss after natural conditions resumed. Hydroperiod was designated as HT during periods of surface water flooding (10 cm water column), and LT during the dry down periods. Soil cores were stored in the dark at 25 C and the top remained open to the atmosphere throughout the laboratory experiment. Soil and W ater P roperties Nine soil cores were destructively sampled on day 10 of the experiment (the baseline/initial soils), and the remaining 27 soil cores were destructively sampled on day 53 (Table 4 2). All soil cores were sectioned into 3 depth segments (0 5, 5 10, and 10 20 cm), stored at 4 C, and analyzed within 30 days. Soil property ana lysis included % moisture, bulk density, % organic matter (OM), total C, total N, a nd C fiber analysis (% fines (< 0.025 mm), cellulose + hemicellulose, and lignin content). Moisture content and bulk density were determined after drying a subsample at 70 C to a constant weight. Percent OM was estimated by mass loss on ignition (LOI) where dry soils were combusted at 550 C for 5 h and final weight was subtracted from initial weight. Total C
97 and N content were determined using a Costech Model 4010 Elemental Analyzer (Costech 121 Analytical Industries, Inc., Valencia, CA). Dissolved OC was measured on a TOC Analyzer (Shimadzu Scientific Instrument TOC 5050A, Columbia, MD) following EPA method 415.1, which included filtering the water sample through a 0.45 m membrane filter and acidifying the sample with H 2 SO 4 until analysis (USEPA 1993). The fiber analysis was performed using a modified Ankom fiber fractionation method after Roberts and Rowland (1998). Fines (<0.025 mm) were defined as soil particles relea sed from a 0.025 mm mesh bag placed in DI water, agitated, and rinsed. Cellulose + hemicellulose was the fraction solubilized in 24 N H 2 SO 4 lignin content was the fraction combusted in a muffle furnace at 550 C and inorganic ash >0.025 mm was the materi al remaining following combustion. During the laboratory experiment the surface water in each core was regularly monitored for salinity, conductivity, pH, dissolved oxygen (DO), and temperature using a hand held YSI model 85 (YSI Inc., Yellow Springs, OH). S urface water (20 mL) was collected from each core 2 3 times during each event/condition to monitor sulfate (SO 4 2 ) concentration and rate of loss. Sulfate samples were un acidified, manually diluted, and analyzed on a Dionex DX 600 Ion Chromatograph (Th ermo Scientific, Sunnyvale, CA) using standard method 4110B (Standard Methods 1997). The surface water collected from each wetland type was analyzed for ammonium N (NH 4 + N) and soluble reactive P (SRP) on an AQ 2 Automated Discrete Analyzer (Seal Analyti cal, Mequon, WI) using EPA Methods 104 A Rev. 3, and 118 A Rev. 2, respectively (USEPA 1993). Depending on salinity, the sample matrix used was either de ionized (DI) water or 13 ppt artificial seawater (Neomarine Reef Salt mix,
98 Brightwell Aquatics, Elysb urg, PA) with the 26 ppt salinity samples diluted by half. For total kjeldahl nitrogen (TKN) quantification, 10 mL of surface water was digested in glass tubes with a TKN salt catalyst and 0.5 mL of concentrated H 2 SO 4 Samples were digested for 2 h at 160 C, and then at 360 C for 30 min. Tubes were cooled, 10 mL DI water added, vortexed, and the concentration was determined calorimetrically using a Technicon Autoanalyzer II (Seal Analytical, Mequon, WI), EPA Method 351.2 (USEPA 1993). Soil Organic C arbon L oss The rate of SOC loss was estimated by measuring the major pathways of organic C loss CO 2 production (soil respiration, during both LT and HT), CH 4 production (methanogenesis), and DOC release. Soil respiration was determined using a portable infrared gas analyzer (Li Cor 8100, Lincoln, NB) equipped with a 10 cm diameter chamber. The cores were plugged/sealed and CO 2 flux was measured (1 min) a total of 16 times during the 53 day study, including a minimum of one sampling during each dry down salinity pulse, and ambient surface water condition. Methane was determined using soil slurry microcosms after Chambers et al. (2011). Approximately 5 g (wet weight) soil was prepared in triplicate and added to four 60 mL glass serum bottles. Bottles were capped with butyl stoppers and aluminum crimp caps, evacuated to 75 kpa, and flushed with O 2 free N 2 gas for one minute to create anaerobic conditions. Eight mL of 0.5 ppt, 13 ppt, or 26 ppt seawater was added to create a slurry with duplicates eve ry third. All incubations were maintained at a slight over pressure and stored in the dark at 30 C. Headspace was extracted and measured on a gas chromatograph (Shimadzu Scientific Instruments GC 8A, Columbia,
99 MD) fitted with a flame ionization detector (FID) on days 3, 7, 12, and 17. Potential methanogenesis was calculated as CH 4 C production per g dry soil per day. DOC release was quantified in the leachate collected during the soil core dry down. Studies indicate that porewater seepage during ebb tide in salt marshes occurs primarily through the face of tidal creek banks (Gardner 2005). Therefore, I attempted to simulate a receding tide when surface water is drawn through the soil profile and then released to tidal creeks. Leachate was collected in 125 mL nalgene bottles, acidified, and stored at 4 C until analyzed. All DOC samples were filtered through a 0.45 m membrane filter and analyzed using a TOC Analyzer (Shimadzu Scientific Instrument TOC 5050A, Columbia, MD) following EPA method 415.1 (USEPA 1993). Nutrient R elease Nutrient release was determined by analyzing the leachate water collected during the dry down periods for a total of 5 samplings during the study (Figure 4 2). Ammonium N, SRP, and TKN were analyzed as described above for surface water properties. Organic N was calculated as the difference between TKN and NH 4 + N. Microbial I ndicators Microbial biomass C (MBC) was determined by fumigation extraction after Vance et al. (1987) and White and Reddy (2001). Duplicate 5 g (wet weight) samples were prepared in 25 mL centrifuge tubes. One set was fumigated with chloroform for 24 h and the other set served as the non fumigated control. Following the chloroform treatment, both fumigates and non fumigates were extracted with 25 mL of 0.5 M K 2 SO 4 agitated for 30 min on a circulating shaker, and centrifuged at 5000 rpm for 10 min. The supernatant was vacuum filtered through a Whatman # 42 filter paper and stored at 4 C until analyzed for total organic carbon (Shimadzu Scientific In strument
100 TOC 5050A, Columbia, MD). An extraction efficiency coefficient of k EC = 0.37 was applied to all samples (Sparling et al. 1990). Total labile organic C (TLOC) was defined as the TOC for the fumigated samples and labile organic C (LOC) was defined as the TOC for the non fumigated samples. The metabolic quotient (qCO 2 ) was calculated as the rate of soil respiration (mg CO 2 C kg soil 1 d 1 ) divided by MBC (mg MBC kg soil 1 ). Beta glucosidase enzyme activity was measured fluorometrically as described by Marx et al. (2001). Soil samples were homogenized, diluted by 100 with autoclaved distilled de ionized (DDI) water, and sonicated for 10 seconds. Replicate soil slurry samples (150 L) were added to each column of a 96 well plate. The top 4 rows were incubated with 100 L (200 M final concentration) of fluorescently labeled substrate (methyl umbelliferone (MUF) glucoside) for 4 hours. After 4 hours, labeled MUF glucoside was added to the bottom 4 rows and 10 L of 0.1 M NaOH was added to all wells. Formation of the fluorescent product MUF was measured at excitation/emission wavelength of 360/460 on a Synergy HT Multi Mode Microplate Reader ( BioTek, Winooski, VT). Quenching curves were prepared for each of the three wetland soils and coefficients wer e applied to the final values, expressed as mg MUF kg dry soil 1 h 1 Dehydrogenase enzymes activity was determined using the TTC (triphenyltetrazoliym chloride) method developed by Thalmann (1968) and modified by Alef (1995). Five g wet weight soil was added to a 60 mL amber glass vial along with 5 mL TTC solution. Slurries were incubated in the dark for 24 hours, 40 mL acetone was added, the solutions were filtered through a Whatmann #2 filter paper, and the optical density of the solution was measure d at 546 nm wavelength on a Spectrophotometer
10 1 (Shimadzu Scientific Instruments UV 160, Columbia, MD). Dehydrogenase activities were expressed as mg TPF kg dry soil 1 h 1 Data A nalysis Statistical analysis was performed using SAS 9.1 software (SAS Instit ute Inc., Cary, NC). All data sets were first tested to determine if the assumptions of homogeneity and normality were met using the Brown and Forsythe's Test and Shapiro Wilk Test, respectively. Where these assumptions were not met, the raw data was log 1 0 transformed and further statistical analysis was conducted using the dataset that fulfilled the assumptions. A three to determine the interaction between soil respiration rate, treatment condition, an d time. A two way repeated measures ANOVA model was used to determine significantly different means for complete data sets collected at common times for each soil core where time (t) was significant (p < 0.05, according to a treatment specific two way ANO VA). This was determined to be true for DOC release and nutrient release. When t was deemed non significant (p > 0.05; two way ANOVA), the multivariate response was reduced to a univariate response and a one way ANOVA model was used to determine signific a ntly different means (Davis 2002 ). This method was determined to be appropriate for CO 2 flux. Non repeated variables (soil and water properties, CH 4 flux, and microbial indicators) were measured using one and two way ANOVA models Product correlations were calculated to determine correlations between SOC loss, soil and water properties, microbial indicators, and nutrient release. A Chi Square test of independence was used to test if the percent of SOC lost to each pathway (respirat ion, methanogenesis, and DOC release) depend ed upon the treatment condition applied, and a Chi Square of Goodness of Fit was used to test if the percent
102 of SOC lost to each pathway deviated significantly from that of the control condition (all lsed with salt (26 ppt) water) was completely removed fro m the analysis as an outlier (> 2.5 times the standard deviation). A simple theoretical model was developed to illustrate the effects of SLR on SOC loss using the results for the partitioning of C lo ss via the 4 pathways (HT CO 2 LT CO 2 CH 4 and DOC release) in each of the 3 wetland types. In this model, SLR is defined as an increase in time of HT and decreased time of LT. The linear model assumes current conditions are 50 % HT and 50 % LT and models an increase to 100 % HT and 0 % LT, with no real time SLR scenario specified. Results Soil and W ater P roperties The brackish marsh soil had a significantly (p < 0.01) higher bulk density than the freshwater and salt ma r sh soils; bulk density did not vary significantly with depth in any of the soils (Table 4 3). Soil organic matter content showed an inverse relationship with bulk density, with consistently higher OM in the freshwater and salt marsh soils (41 52 % ) compared to the brackish marsh soils (29 34 % ; Table 4 3). Similar to % OM, total C and total N were significantly higher in the freshwater and salt marsh soils compared to the brackish marsh soil (p < 0.05) and did not change significantly with depth (Table 4 3). The ratios of soil total C:N were 16.7 1.4, 18.0 1.7, and 17.9 1.6 for the freshwater, brackish, and salt marshes, respectively. The C:N did not differ significantly with wetland type or depth. Carbon fiber analysis revealed the freshwater soil had significantly few er fines (<0.025 mm; p = 0.005) than the bra ckish and salt marsh soils (Figure 4 3). Both the freshwater and salt marsh soil had higher lignin content (p = 0.02) and lower inorganic ash content (>0.025 mm; p = 0.0 3) than the brackish soil
103 (Figure 4 3). Cellulose + hemic elluloses content did not differ significantly with wetland type. Surface water salinities were 0.56 0.07, 13.5 0.8, and 26.5 1.8 ppt for the freshwater, brackish, and salt marsh surface water, respectively. Sulfate concentration and specific cond uctivity increased significantly with salinity, while DOC concentration in the surface water was inversely related with salinity (Table 4 4). Surface water pH was similar for all sites (averaging 7.7), and dissolved oxygen also did not vary significantly between sites (ranging from 6.4 to 6.9 mg L 1 ; data not shown). Ammonium concentration was significantly higher (p < 0.05) in the freshwater, compared to the brackish site water, whereas TKN was not significantly different between wetland types (Table 4 4 ). SRP concentration was significantly greater (p < 0.01) in the freshwater compared to both the brackish and salt marsh surface water (Table 4 4). Soil Organic C arbon L oss Soil respiration was strongly affected by inundation level, with CO 2 flux signif icantly greater at LT (following a 24 h dry down), compared to HT (10 cm water column), in the brackish and salt marsh soils (Fig ure 4 4). Using only the soil cores treated with the control condition, the average rate of CO 2 flux during HT was significant ly greater (p < 0.001) in the freshwater marsh soil (1,033 347 mg CO 2 C m 2 d 1 ) than the salt marsh soil (500 160 mg CO 2 C m 2 d 1 ) and was negatively cor related with salinity (p < 0.05 ; Table 4 5). The rate of HT CO 2 flux was also positively correl ated with DOC release, methanogenesis, qCO 2 SRP, and Organic N (Table 4 5). During LT in the control condition soil cores, average CO 2 fluxes were greatest in the brackish and salt marsh soils (2450 680 and 2052 1031 mg CO 2 C m 2 d 1 respectively), but were not statistically greater than the freshwater marsh soil at LT
104 (1578 649 mg CO 2 C m 2 d 1 ; Figure 4 4). Soil respiration rate at LT in the control condition cores was not correlated with any of the soil, water, or microbial indicators of interes t (Table 4 5). Comparing average HT and LT CO 2 flux in each wetland type revealed significant differences in the magnitude of the tidal effect on soil respiration. Mean soil respiration rates were 53 % greater at LT than at HT in the freshwater marsh soil 230 % greater at LT than HT in the brackish marsh soil, and 310 % greater at LT than HT in the salt marsh soil (Fig ure 4 4). Time was not a significant factor in soil respiration rate over the course of the experiment (p = 0.25; three way repeated meas ures ANOVA). Therefore, CO 2 flux measurements from the three salinity pulses were treated as replicates. Focusing on the effects of the salinity pulsing manipulation, results indicates that all of the freshwater marsh soil cores had significantly higher (p < 0.01) rates of CO 2 flux during HT (1056 420 mg CO 2 C m 2 d 1 ) than all of the other soil cores (Fig ure 4 5). Soil respiration rates in the brackish marsh soils were negatively affected by the pulse of freshwater (0.5 ppt), decreasing the average ra te of CO 2 flux from 426 85 mg CO 2 C m 2 d 1 in the control, to 286 5 mg CO 2 C m 2 d 1 follo wing the pulse (p < 0.05 ; Figure 4 5). Meanwhile, the pulse of salt (26 ppt) water had no significant impact on the brackish marsh soil CO 2 flux during HT (Fig u re 4 5). In the salt marsh soil, respiration rates during HT were not affected by the addition of either fresh (0.5 ppt) or brackish (13 ppt) water (Fig ure 4 5). Overall, the rate of CO 2 flux during the HT salinity pulsing events were positively correlat ed with DOC release, CH 4 flux, and SRP (all p <0.01 ; Table 4 6). Following each salinity pulse, some of the soil treatments experienced a legacy effect from the pulse, resulting in significantly different rates of CO 2 flux during the
105 subsequent LT (Fig ure 4 6). The freshwater marsh soil receiving a pulse of brackish (13 ppt) water had a CO 2 flux of 2587 1230 mg CO 2 C m 2 d 1 more than double the rate in the control treatment during the same period (1221 426 mg CO 2 C m 2 d 1 ; p < 0.001 ; Fig ure 4 6). In contrast, the freshwater marsh soil pulsed with salt (26 ppt) water showed no legacy effect. In the brackish marsh soil, respiration rate during LT was unaffected by either the freshwater or saltwater pulses, with an overall average flux rate of 1645 291 mg CO 2 C m 2 d 1 (Fig ure 4 6). In the salt marsh soil, the pulse of fresh (0.5 ppt) water had a significant legacy effect, increasing CO 2 flux from 1578 313 mg CO 2 C m 2 d 1 in the control treatment, to 2775 710 mg CO 2 C m 2 d 1 following the pu lse (p < 0.05). The pulse of brackish (13 ppt) water had no legacy effect on soil respiration in the salt marsh soil (Fig ure 4 6). The concentrations of NH 4 + and MBC were positively correlated with CO 2 flux during the LT that followed the salinity pulsin g events (p < 0.05; Table 4 6). Between each salinity pulsing event, all soil cores were flooded with their original ambient site water for ~12 days to mimic a return to baseline, or natural field conditions (Table 4 2). During this ambient water phase, respiration rates for each salinity pulsing treatment condition were not significantly different than the control for that wetland type during both HT and LT measurements, suggesting a return to baseline conditions between each salinity pulse (data not sho wn). The rate of methane production was an order of magnitude higher (p < 0.001) in the freshwater marsh control (0.5 ppt) treatment than in all other treatments (114 9 mg CH 4 C m 2 d 1 ; Figure 4 7). The addition of brackish (13 ppt) and salt (26 ppt) water to the freshwater marsh soil decreased methanogenesis by 98 % and 97%, respectively.
106 In the brackish marsh soil, CH 4 flux ranged from below detection to 2.2 mg CH 4 C m 2 d 1 with no significant affect of surface water salinity changes. The salt mars h soil exposed to fresh (0.5 ppt) water had a higher CH 4 flux (8.0 6.4 mg CH 4 C m 2 d 1 ) than the control condition (0.12 0.13 mg CH 4 C m 2 d 1 ), but was not statistically significant (p = 0.08 ; Fig ure 4 7). Overall, CH 4 flux was positively correlated with DOC release, qCO 2 and SRP (Table 4 6). Methane flux had high within treatment variability and can be interpreted as potential methanogenesis since it was measured in a slurry microcosm, rather than the intact soil cores. DOC release from the co ntrol condition soil cores decreased significantly from freshwater, to brackish, to salt marsh wetland types (p < 0.001 ; Table 4 7). DOC release was inversely correlated with salinity and directly correlated with CH 4 flux, qCO 2 SRP, and organic N (Table 4 5). The salinity pulsing events did have a significant impact of the rate of DOC release in both the freshwater and the salt marsh soils. The pulse of brackish (13 ppt) water in the freshwater marsh soil caused a significant decrease (p < 0.01) in DOC release from, from 18.3 2.8 mg L 1 in the control, to 10.8 2.2 mg L 1 following the pulse. Additionally, the pulse of fresh (0.5 ppt) water in the salt marsh soil caused a significant increase (p < 0.01) in DOC release, from 7.6 0.7 mg L 1 in the co ntrol, to 11.4 0.9 mg L 1 following the pulse. With the exception of the pulsing effects, the rate of DOC release was constant over time. Overall, DOC release following salinity pulsing events was positively correlated with CH 4 flux, qCO 2 and SRP, and negatively correlated with NH 4 + and salinity (Table 4 6). Over 98 % of SOC loss in this study occurred through soil respiration, with the exception of the freshwater marsh control, which lost 94 % of SOC to respiration. The
107 average total mass of SOC lost i n the freshwater marsh control (0.5 ppt) treatment was significantly greater than in the brackish marsh salt (26 ppt) water treatment (p = 0.009), the salt marsh control (26 ppt) treatment (p = 0.003), and the salt marsh brackish (13 ppt) water treatment ( p = 0.05 ; Fig ure 4 8). Since th e total mass of SOC loss (Figure 4 8) was affected by the amount of time each conduction was measure d during the experimental manipulation, a weighted average % SOC loss to each pathway was also calculated for the control co ndition (Fig ure 4 9). The freshwater marsh control lost the largest % of total SOC through HT CO 2 flux (55 % ), while the brackish and salt marsh controls lost the largest % SOC through LT CO 2 flux (62 and 70 % respectively). The freshwater marsh soil puls ed with brackish (13 ppt) water was the only treatment to pathways. This treatment more closely resembled the SOC loss partitioning of the brackish marsh control (54 % lo ss to LT CO 2 flux, 46 % to HT CO 2 flux, and < 1 % to methanogenesis) than the freshwater marsh control (Figure 4 9). Nutrient R elease Among the soil cores receiving the control condition, the salt marsh soil released significantly more NH 4 + than the freshwa ter and the brackish marsh soils (p < 0.01 ; Table 4 7). The salinity pulsing events had a significant effect on the freshwater marsh soil, more than doubling the concentration of NH 4 + release when a pulse of brackish (13 ppt) water was added, from 0.02 0.19 mg L 1 in the control, to 0.57 0.79 mg L 1 following the pulse (p = 0.02). TKN release showed no significant difference between wetland type, salinity pulse, or time, averaging 1.2 0.3 mg L 1 Organic N release differed significantly between wet land types and was greater in the freshwater and
108 brackish marsh soils than the salt marsh soil (p < 0.05 ; Table 4 7). Salinity pulsing events and time had no effect of organic N release. SRP release for the control condition was significant greater for the freshwater and brackish marsh soils than from the salt marsh soil, even when controlling for the concentration in the surface water (p < 0.01 ) ( Table 4 7). Adding a pulse of fresh (0.5 ppt) water to both the brackish marsh and salt marsh soils caused a significant increase in SRP release (p < 0.05), but the effect was directly attributable to the increase in SRP from the added freshwater (i.e., the effect was no longer significant when surface water SRP concentration was controlled for). Microbial I ndicators The concentration of MBC in the salt marsh soil control was more than twice that of the freshwater and brackish marsh soil controls (p < 0.001; Table 4 8). The salinity pulse of brackish water to the salt marsh soil caused a significant increas e in MBC (p < 0.01) from 2105 159 mg kg 1 in the control, to 2974 365 mg kg 1 following the 13 ppt pulse. Time was n ot a significant factor for MBC; the amount of MBC was similar in both the baseline cores (sampled on day 10) and the final cores (sam pled on day 53). MBC was positively correlated with soil organic matter content, TLOC, LOC, dehydrogenase activity, and NH 4 + concentration (p < 0.01; data not shown). TLOC among the control condition cores was highest in the salt marsh soils (p < 0.001; Table 4 8). Similar to MBC, TLOC significantly increased in the salt marsh soil following the pulse of brackish (13 ppt) water, from 2345 389 mg kg 1 in the control, to 3236 389 mg kg 1 following the pulse (p < 0.001). LOC did not differ significantl y between wetland types, nor was it affected by the salinity pulsing events. LOC was positively correlated with glucosidase activity (p < 0.05; data not shown). The metabolic quotient (qCO 2 )
109 was significantly lower in the salt marsh soil compared to t he freshwater marsh soils (p < 0.05; Table 4 8) and was unaffected by the salinity pulsing events. The qCO 2 was positively correlated with LT CO 2 flux and DOC release in the control condition soils (Table 4 5), and positively correlated with CH 4 flux and DOC release following the salinity pulsing events (Table 4 6). Beta glucosidase activity was significantly lower in the brackish marsh soil compared to the freshwater marsh soil (p < 0.05; Table 4 8). The salinity pulsing events did not cause a signif icant change in glucosidase activity. glucosidase activity was positively correlated with soil moisture content, dehydrogenase activity (p < 0.05), and organic matter content (p < 0.001). Dehydrogenase activity showed a similar pattern to glucosid ase activity, but differences between wetland types were not significant (Table 4 8). Dehydrogenase activity was positively correlated with soil moisture (p < 0.01), organic matter content, and MBC (p < 0.05). Discussion Freshwater T idal M arsh The fresh water tidal marsh soil had higher soil moisture content (83 2 % ) and a lower bulk density (0.18 0.04 g cm 3 ) than the brackish marsh, but both measurements were similar to the salt marsh soil (Table 4 3). Surface water concentrations of DOC, NH 4 + TKN, and SRP were, on average, higher in the freshwater marsh than other wetland types (Table 4 4). Overall, the ecological characteristics of the freshwater tidal marsh concurred with observations from other US Atlantic coast freshwater tidal marshes, includ ing high plant diversity, high spatial heterogeneity, high organic matter content, and measurable methanogensis (Neubauer et al. 2000; Odum 1988).
110 While the influence of inundation and water table on soil respiration rates in non coastal wetlands and pea tlands is well established (e.g., Wright and Reddy 2001; DeBusk and Reddy 1998; Clymo 1983), very few studies have investigated the impact of tides on coastal wetland soil respiration. I found the rate of CO 2 flux from the freshwater marsh soil during HT was significantly greater than the salt marsh soil, but LT CO 2 flux in the freshwater marsh was the lowest of the 3 intertidal wetland types (Figure 4 4). As a result, the magnitude of the tidal effect for the freshwater marsh was relatively small, avera ging only a 53 % increase in CO 2 flux between LT and HT. This tidal effect is similar to the 50 % increase at LT found by Neubauer et al. (2000) in a temperate tidal freshwater marsh, but is substantially less than the tidal effects observed for the brackis h marsh (230 % increase at LT) and salt marsh (310 % increase at LT) in this study. Differences in the magnitude of the tidal effect along the salinity gradient is likely attributable to difference in the hydraulic conductivity of the soils. The brackish m arsh soil was observed to drain the quickest during the simulated ebb tide (> 1 to 2 h), followed by the salt marsh soil (2 to 6 h), and the freshwater marsh soil (up to 24 h); a pattern that mirrors the LT CO 2 flux rates in these soils (Fig ure 4 4). Soil s with high water retention could allow anaerobic soil conditions to persist during LT, and subsequently decrease the rate and efficiency of microbial respiration (Freeman et al. 1993). F reshwater marsh soils in this study were classified as Maurepas muck and consisted of highly decomposed herbaceous histisols (USDA 1978). This particular peat composition has some of the lowest soil hydraulic conductivity and acts similar to a sponge in its ability to retain water against gravitational pull (Boelter 1965)
111 Methanogenesis was an order of magnitude greater in the freshwater marsh control (0.5 ppt) compared to all other treatments (Fig ure 4 7 ), as would be expected from pre vious literature indicating a higher energy yield for sulfate reducers, compared to 4 flux to overall SOC loss was still minimal, representing only ~3 % of the total SOC loss in the freshwater marsh (Fig ure 4 9). In situ rates of methanogenesis in freshwater tidal mars hes reported in o ther studies represent about 6 % of C loss; my rates may be underestimated because 95 % of CH 4 is normally released through macrophytes, which were not included in the present study (Neubauer et al. 2000). Microbial indicators for the freshw ater marsh showed intermediate concentrations of MBC, TLOC, and LOC, and the highest average value for qCO 2 (Table 4 8). Overall, the freshwater marsh control (0.5 ppt) lost the greatest total average mass of SOC (Fig ure 4 8), the majority of which was lo st via HT soil respiration (55 % ), followed by LT soil respiration (41 % ), and methanogenesis (3 % ; Figure 4 9). When the freshwater marsh soil was pulsed with brackish (13 ppt) water, HT CO 2 flux remained unchanged (Figure 4 5), but a significant legacy effe ct during LT from the bra ckish pulse was observed (Figure 4 6). The respiration rate of freshwater marsh soil pulsed with brackish (13 ppt) water increased 112 % during the subsequent LT. This increase is at least partially explained by an influx of SO 4 2 to support sulfate reduction (Table 4 3), but the data also suggest a strong correlation (Table 4 6) between the acceleration of LT CO 2 flux and a tripling in the concentration of NH 4 + release. Th e latter indicates a short term disruption of N cycling fo llowing pulses of seawater, which could have significant ramification for coupled N and C cycling in coastal wetlands.
112 Chloride is the most abundant ion in seawater by weight, with approximately 7000 mg Cl L 1 in 13 ppt seawater (Kester et al. 1967). C hloride is known to inhibit both denitrification and nitrification in aquatic, wetland, and terrestrial ecosystems (Seo et al. 2008; Roseberg et al. 1986). Research indicates that aquatic systems that do not naturally contain significant Cl (i.e., inland forested watersheds) experience a complete inhibition of denitrification enzyme activity with Cl concentrations as low as 80 mg L 1 (Hale and Groffman 2006). Interestingly, urban steams commonly exposed to low levels of Cl from the runoff of deicing sal ts do not experience any decline in denitrification when Cl concentrations increase, suggesting a capacity for microbial populations to adapt to long term Cl exposure (Hale and Groffman 2006). I found no correlation between CO 2 flux and NH 4 + under ambie nt (control) conditions (Table 4 5), but a significant correlation following the salinity pulsing events (Table 4 6). The inhibition of nitrification by Cl has been demonstrated in agricultural soils treated with Cl containting fertilizers and results in elevated NH 4 + concentrations in the soil solution (Roseberg et al. 1986). Saltwater also disrupts N cycling by replacing NH 4 + ions on the soil cation exchange complex with other salts (e.g., Na 2+ K + ), and thus releases NH 4 + in to solution (Baldwin et al. 2006; Weston et al. 2006; Portnoy and Giblin 1997). I hypothesize that during the HT 13 ppt pulse there was an increase in NH 4 + availability, a precursor to coupled nitrification denitrification, but microbes were not able to utilize the additional inorg anic N until the stress caused by Cl ions was alleviated during the LT dry down. To date, the mechanism of Cl inhibition of N cycling is unknown and the relevance of the previous finding s in inland systems to coastal systems warrants further investigati on (Seo et al. 2008).
113 Methanogenesis decreased 98 % when the freshwater soil was pulsed with brackish (13 ppt) water (Fig ure 4 7). Past laboratory studies have indicated that salinity concentrations of 14 ppt cause the near complete suppression of methanog enesis (Chambers et al. 2011) which agrees with the current study. In situ methanogenesis along a coastal wetland salinity gradient in Louisiana revealed average annual CH 4 fluxes of 4.3, 73, and 160 g CH 4 C m 2 for fresh, brackish, and salt marshes, resp ectively (DeLaune et al. 1983). Higher measurements of CH 4 flux in the field, compared to laboratory, can be expected due to the presence of low redox microzones within the natural soil structure (King and Wiebe 1980). Microbial indicators were not alte red as a result of the brackish pulse, but DOC release decreased significantly compared to the control condition following the pulse. In other laboratory studies, the addition of artificial seawater (no C source) to a freshwater wetland soils caused no ch ange in DOC release (Weston et al. 2011). Since the ambient concentration of DOC in the brackish water is less than the freshwater (Table 4 4), the decrease in DOC release may have been due to DOC absorption to the peat matrix, or the utilization of DOC a s a substrate for microbial respiration (Freeman et al. 1997). Overall, the freshwater marsh soil pulsed with brackish water was the only treatment in the entire study that showed a significant deviation from the control condition in the partitioning of SOC loss via the 4 pathways measured in this study (Fig ure 4 9). Adding a pulse of saltwater (26 ppt) to the freshwater marsh soil had no significant effect on soil respiration during HT (Fig ure 4 5) or the subsequent LT (Fig ure
114 4 6). Although the incre ase in SO 4 2 occurred (Table 4 4), the increase in NH 4 + release (seen with the 13 ppt pulse) was not observed following the saltwater pulse, nor was the legacy effect of accelerated LT respiration. The reason for this is unclear, but it could be that the salinity was so high (26 ppt) the freshwater marsh microbial community was under too much osmotic stress to respond to the increased SO 4 2 abundance. High ionic conductivity is known to inhibit enzyme activity in the soil and cause the buildup of toxins a ssociated with saltwater, such as sulfide (Frankenberger and Bingham 1982). Similar to the brackish pulse, the 26 ppt pulse did significantly suppress methanogenesis (Figure 4 7). No evidence was found of changes in microbial indicators, DOC release, or nutrient release as a result of the saltwater pulse in the freshwater marsh soil. Brackish M arsh The brackish marsh soil had significantly higher bulk density (p < 0.0001), lower organic matter, total C, and total N content (Table 4 3), and higher fines a nd inorganic ash (Fig ure 4 3) than the other wetland types, indicative of a system with greater tidal influence and higher inorganic sediment deposition. Although these characteristics are more commonly found in salt marshes (Craft 2007; DeLaune et al. 20 02; Odum 1988), relation to my sampling sites may have contributed to the brackish marsh site being the most inorganic of the three wetland types (Wieski et al. 2010; N yman et al. 1990). Surface water from the brackish marsh had intermediate levels of DOC and SRP and the low concentrations of NH 4 + and TKN (Table 4 4). Soil respiration rate in the brackish marsh control was moderate during HT, but had the highest averag e CO 2 flux rate at LT (Fig ure 4 4). As a consequence, the magnitude of the tidal effect in the
115 brackish marsh was large ; a 230 % increase in the rate of C loss via soil respiration between HT and LT. The brackish marsh soil was observed to have significa ntly higher hydraulic conductivity than the other wetland types, draining in a few hours when the bottom core plugs were removed, despite having the lowest overall porosity (Table 4 3). Salt can function as both a flocculating agent, causing hydrophilic c olloids to aggregate, and as a dispersing agent, causing hydrophobic colloids to repel one another (Gregory 1989). Therefore, high concentrations of Na 2+ can increase both the size of soil aggregates and the quantity of macropore spaces in the soil profil e, allowing water to drain quickly and the soil to become aerobic faster during LT (Brady and Weil 2004). The acceleration of the soil respiration rate following a drop in the water table is well established in the literature and can create a mineralizati on rate up to 50 times faster than under flooded conditions (Clymo 1983). Methane production in the brackish mars h control was negligible (Figure 4 7). On average, the availability of labile C (MBC, TLOC, and LOC) was low in the brackish marsh (Table 4 8 ). glucosidase, a soil enzyme involved in glucose production, is considered a good indicator of soil quality and general C mineralization rate (Makoi and Ndakidemi 2008). This study found glucosidase to be significantly lower in the brackish marsh so il than the freshwater marsh soil (Table 4 8), but was no relationship was found between glucosidase and salinity, as was also the case in a study by Jackson and Vallaire (2009). Dehydrogenase is produced during microbial respiration and was of interest in this study due to existing contradictions in the literature. Some research has found that dehydrogenase activity is strongly inhibited in saline soils (Saviozzi et al. 2011; Frankenberger and Bingham 1982), while others have found a positive relations hip
116 between dehydrogenase activity and salinity (Wu et al. 2008). The current study found no significant differences in dehydrogenase activity with salinity or wetland type (Table 4 8). Overall, the total mass of C lost from the brackish marsh was inte rmediate in comparison to the other wetlands types (Figure 4 8). Approximately 62 % of SOC loss was to soil respiration during LT, 38 % to soil respiration during HT, and >1 % to DOC release (Figure 4 9). When the brackish marsh soil was pulsed with fresh (0 .5 ppt) water, there was a significant decline in HT soil respiration rate, with an average decrease of 33 % (Fig ure 4 5). Sulfate reduction is the dominate pathway of soil respiration in brackish marshes (Weston et al. 2006 b ) and SO 4 2 tends to have a sho rt residence time in the soil profile, being quickly utilized by microbes (DeLaune et al. 1983). The abrupt decline in SO 4 2 availability when freshwater was added (Table 4 3) would have diminished the rate of sulfate reduction and required non sulfate re ducing microbes to activate quickly to maintain the overall rate of soil respiration. The brackish marsh soil had low microbial biomass, relative to the other wetland types (Table 4 8). High microbial biomass often correlates with greater microbial diver sity (Cordova Kreylos et al. 2006), so the low biomass of the brackish marsh soil may have prevented the microbial community from adapting quickly to the decrease in SO 4 2 availability. Despite the negative influence of the freshwater pulse on HT CO 2 flux no legacy effect of the pulse was seen during the subsequent LT (Figure 4 6). This supports the earlier hypothesis that LT respiration in the brackish marsh soil is dominated by aerobic microbes due to the high hydraulic conductivity of the soil, making SO 4 2 availability superfluous during LT. Methane
117 production, DOC release, microbial indicators, and nutrient release were not significantly affected by the freshwater pulse (Fig ure 4 7; Table s 4 7 and 4 8 ). Adding a pulse of salt (26 ppt) water to th e brackish marsh soil resulted in no change to the soil re spiration rate during HT (Figure 4 5) or LT (Fig ure 4 6). Both brackish and salt marsh soil respiration is dominated by sulfate reduction (Weston et al. 2006 b ; Kostka et al. 2002; Howarth 1984), so the change in electron acceptor availability would be minimal. Similarly, the pulse of 26 ppt did not affect CH 4 flux, DOC release, microbial indicators, or nutrient release from the brackish marsh soil ( Figure 4 7; Tables 4 7 and 4 8 ). Salt M arsh Th e physical properties of the salt marsh soil were similar to that of the freshwater marsh soil, having significantly lower bulk density than the brackish marsh soil and generally greater organic matter, total C, and total N content (Table 4 3). DOC and SR P in the salt marsh surface water were, on average, lower than the other wetland types, and the concentration of NH4+ and TKN were intermediate (Table 4 4). Under control conditions, the rate of soil respiration at HT was significantly lower in the salt m arsh soil than in t he freshwater marsh soil (Figure 4 4). The magnitude of the tidal effect was greatest in the salt marsh soil comp ared to the other wetland types; CO2 flux increased an average of 310 % from HT or LT (Fig ure 4 3). The rate of drainage in salt marsh soil was about 2 6 hours, indicating a high hydraulic conductivity likely associated with the dispersive powers of high ionic strength saltwater (Brady and Weil, 2004). Similar to the brackish marsh soil, the salt marsh soil showed negligible methanogenesis (Figure 4 7). However, MBC and TLOC were highest in the salt marsh soil, and the metabolic quotient the lowest (Table 4 8). Most studies investigating the
118 affect of salinity on MBC have been performed in agricultural and upland systems and results have varied widely, from a positive relationship (Rasul et al. 2006), to a negative relationship (Muhammad et al. 2006), to no relationship at all (Gennari et al. 2007) between salinity and MBC. The qCO 2 is an index of the efficiency of the micro bial community ; stressed populations tend to produce more CO 2 per individual and have a higher qCO 2 (Wong et al. 2008). More saline environments are presumed to be more stressful to microbes and as a result, typically exhibit a higher metabolic quotient (Saviozzi et al. 2011; Tripathi et al. 2006), but previous studies (Wong et al. 2008) have also found a decrease in qCO 2 with increasing salinity, as I did in the present study. Th e different finding could be a consequence of the types of microbial popul ations present in the soil. For example, fungi a re known to produce less CO 2 per individual than bacteria and a commonly associated with the decomposition of Spartina spp. the dominant macrophyte of salt marsh site (Wong et al. 2008; Torzilli et al. 2006). The salt marsh soil had significantly lower DOC, SRP, and organic N release than the other wetland types (Table 4 7), and the highest concentration of NH 4 + release. Overall, the salt marsh control had the lowest average total C loss of all treatments (Figure 4 8) and the largest percent of total C lost to soil respiration during LT (70 % ; Fig ure 4 9). Adding a pulse of fresh (0.5 ppt) water to the salt marsh soil did not affect the rate of soil respiration during HT (Fig ure 4 5), but did cause a significant increase in CO 2 flux during the subsequent LT, an average increase of 76 % in the rate of soil respiration (Figure 4 6). Past research has found that increasing tidal flushing can
119 enhance productivity in salt marshes through decreased osmotic stress and the removal deleterious compounds such as sulfide (King et al. 1982; King and Wiebe 1980). It follows that a flush of freshwater could accelerate microbial respiration in a salt marsh soil, as observed in this study. The addition of freshwater to the salt marsh soil also increased methanogenesis from 0.12 0.13 mg CH 4 C m 2 d 1 in the control, to 8.05 6.14 mg CH 4 C m 2 d 1 following the pulse (Figure 4 7). Although this change was marginally significant (p = 0.08), the implication s are still important considering CH 4 has a radiative forcing 25 times greater than CO 2 (IPCC 2007). This research suggests the discharge of precipitation run off from urban development directly into salt marshes can accelerate both the production of CO 2 during LT and the release of CH 4 to the atmosphere, which may be an importance consideration for urban development adjacent to salt marshes. The fresh (0.5 ppt) water pulse had no effect on soil microbial indicators (Table 4 8) or nutrient release (Tabl e 4 7), but did significantly increase the concentration of DOC release following the freshwater pulse. This could simply be a result of higher DOC concentration in the freshwater marsh surface water (Table 4 4). The brackish (13 ppt) water pulse did not alter the rate of CO 2 flux in the salt marsh soil during either HT (Fig ure 4 5) or the subsequent LT ( Figure 4 6), nor was there any change in CH 4 flux from the salt marsh soil ( Figure 4 7). The brackish water pulse did, however, significantly increase the concentration of bioavailable C (MBC and TLOC) in the salt marsh soil. The activity of soil enzymes, DOC release, and nutrient release was not altered as a result of the 13 ppt pulse (Table 4 7 and 4 8).
120 Summary of S alinity P ulsing E ffects P ulses of brackish (13 ppt) water to freshwater tidal marshes cause the most significant effects on the overall rate of SOC loss (Table 4 9). This type of salinity pulsing event, which is likely a common occurrence in low relief coastal zones experiencing an incre ase in the frequency of storm surges and extreme tidal events, caused a substantial short term increase in CO 2 flux during LT ( Figure 4 6). Although this increase in CO 2 flux is somewhat off set by a decrease in CH 4 flux and DOC release (Table 4 9), soil respiration is the major pathway of SOC loss in all coastal wetlands ( Figure 4 9), and alterations in soil respiration rates are a primary concern for maintaining SOC storage. Pulses of higher salinity water (26 ppt) in freshwater tidal marshes, which ma y be seen in narrower, high relief coastal zones, do not appear to have consequences for SOC cycling that are as severe as the brackish pulses (Table 4 9). This could be because the intensity of the ionic strength increase prevents any significant microbi al response to the pulsing event. Brackish marshes are a dynamic intermediate environment that tend to be dominated by organisms functioning as generalists (Van Ryckegem and Verneken 2005). As a consequence, pulses of either fresh (0.5 ppt) or salt (26 p pt) water elicit minimal changes to the biotic environment of a brackish marsh and the overall rate of SOC loss (Table 4 9; Figure 4 8). A pulse of lower salinity surface water in a salt marsh often occurs as a consequence of heavy rain events, urban ru n off, or stormwater discharge. Fresh (0.5 ppt) water pulses can significantly accelerate SOC loss in salt marshes through both higher LT CO 2 flux ( Figure 4 6) and increased DOC release (Table 4 9). In comparison, brackish (13 ppt) water pulses are not m anifest ed as changes to the overall rate of SOC
121 loss, but do appear to increase the availability of labile C, which could have significant consequences for microbial activity over a longer time span. Since many salt marshes respond to SLR through a natura l feedback loop that promotes vertical marsh accretion (Fagherazzi et al. 2012), the accelerated SOC loss caused by high volumes of freshwater urban run off could diminish the ability of salt marshes adjacent to urban development to keep pace with SLR thro ugh vertical soil accretion. Sea L evel R ise I mplications Aside from salinity pulsing effects, this study also identified noteworthy differences in the influence of tidal cycles (duration of inundation) on coastal wetland SOC loss which could have far re aching repercussions for wetland C storage capacity during eustatic SLR R esults reported here suggest that as coastal marshes become more saline, carbon loss during LT becomes increasingly important ( Figure 4 9). Using a simple linear model to extrapola te these findings to future tidal inundation patterns suggests total SOC loss from freshwater tidal marsh soils may increase with greater tidal inundation ( Figure 4 10). In contrast, increased tidal inundation may decrease SOC loss in brackish and salt ma rsh soils. A decrease in the rate of SOC loss could allow more C to remain stored in the soils of brackish and salt marshes, but it is important to consider that C inputs will likely also be altered by SLR (Kathilankal et al. 2008) For example, one stud y found that tidal inundation significantly reduced C assimilation in a temperate salt marsh. Fully or partially submerging marsh vegetation decreased the rate of C fixation by approximately 46 % compared to the control condition. Predicting the overall e complete C budget to be calculated (including above ground and belowground biomass and sediment deposition rates). Possibly the only study that has attempted to establish
122 a complete C budget in a coastal wet land impacted by saltwater intrusion was performed by Neubauer (2011) in a tidal freshwater marsh. Neubauer (2011) found that a 10 ppt increase in salinity decreased net ecosystem productivity (NEP) by 55 % while greater inundation increased NEP by 75 % resulting in no overall change when these two parameters (salinity and inundation) where combined. Summary Coastal wetlands function as large storage reservoirs for global C in the form of SOC. In the contiguous USA only, 25,000 km 2 of coastal (estuarin e) wetlands sequester over 10 Tg C y 1 (Bridgham et al. 2006). The combined effects of global SLR and coastal urban development are well documented threats to the health and sustainability of coastal wetlands, as well as the SOC stored within them (DeLaun e and White 2012; Craft et al. 2009). However, limited attention has been given to the short term, dynamic fluctuations in salinity that occur in coastal wetlands as a result of storm surges, extreme tidal events, and urban stormwater discharge. This stu dy found that the impact of salinity pulsing events on the rate of SOC loss i n coastal wetlands depends upon: ( 1) changes in SO 4 2 availability relative to the ambient condition, ( 2) the accumulation of NH 4 + following seawater additions, ( 3) the ability of freshwater to flush toxins and reduce ionic stress from saline soils, and ( 4) the size of the microbial community available to respond to the chemical changes in the environment. Freshwater tidal marshes are highly sensitive to pulses of 13 ppt surface water. Brackish water accelerates soil respiration by increasing SO 4 2 reduction and coupled nitrification denitrification of NH 4 + Ammonium accumulates in the soil solution as a result of both abiotic processes (desorption for NH 4 + from the cation exch ange complex) and biotic processes (inhibition of nitrifiers and denitrifiers by Cl ) initiated by the
123 addition of seawater. While some negative impacts of brackish pulses are also observed in freshwater soils (a decrease in methanogenesis and DOC release) the overall effect is an increase in the SOC mineralization rate. Erratic environmental fluctuations are more commonplace in brackish marshes and likely minimize the consequences of salinity pulsing events in this type of intertidal wetland. However, decreases in CO 2 flux with pulses of freshwater were observed in the brackish soil and attributed to decreased SO 4 2 availability. Salt marshes respond to freshwater pulses with accelerated SOC loss due to the flushing of accumulated toxins, a reduction i n osmotic stress, and the stimulation of low levels of methanogenesis. The salt marsh soil used in this study also benefited from a large soil microbial community, which presumably allowed for quicker adaptation to changing environmental conditions. The ability of intertidal wetland soils to drain quickly during ebb tides proved crucial to the overall rate of CO 2 flux in this study. High soil hydraulic conductivity, a function of both organic matter composition and the aggregation/dispersive forces of sa lts, tends to magnify tidal effects in saline wetlands and allow for increased aerobic respiration in brackish in salt marshes during LT. The difference in soil properties along natural salinity gradients could have important implications for eustatic SLR An increase in the duration of inundation due to SLR may decrease the overall rate of SOC loss in brackish and salt marshes, while increasing the rate of SOC loss in freshwater tidal marshes. DOC concentration was inversely correlated with salinity a nd predicted 69 93 % of the difference in HT soil respiration rate and CH 4 flux. In contrast, NH 4 + concentration and MBC were better predictors of CO 2 flux during LT. Low microbial metabolic
124 efficiency (high qCO 2 ) correlated with higher DOC release and pr edicted 92 % of the total rate of SOC loss in this study.
125 Table 4 1. Intact core experimental design. The experimental was a 3 (wetland types) x 3 (salinity levels) x 3 (triplicate intact soil cores) mixed model design. An additional 3 soil cores from e ach of the 3 sites were analyzed immediately to establish initial/baseline conditions. Salinity Pulse (ppt) Total No. Intact Cores Wetland Type 0.5 13 26 Immediate analysis Freshwater (0.5 ppt) 3 3 3 3 12 Brackish (13 ppt) 3 3 3 3 12 S alt Marsh (26 ppt) 3 3 3 3 12 36 Table 4 2 Intact core experimental timeline. The experiment lasted a total of 53 days and included 3 periods of ambient site water conditions (HT) and 3 salinity pulsing events (HT), each punctuated by a 24 hour dr y down period (LT). A total of 36 intact soil cores were collected (12 from each wetland type); 9 cores (3 from each wetland type) were sacrificed after the acclimation period to establish baseline conditions, while the remaining 27 (9 from each wetland t ype) were exposed to 3 salinity pulsing events. Day Event/Condition Tidal Phase 1 -All 36 intact soil cores collected -2 8 Ambient site water HT 9 & 10 Dry down LT 10 -Initial 9 cores destructively sampled -11 14 Salinity pulse HT 15 & 16 Dry dow n LT 17 26 Ambient site water HT 27 & 28 Dry down LT 29 32 Salinity pulse HT 33 & 34 Dry down LT 35 44 Ambient site water HT 45 & 46 Dry down LT 47 50 Salinity pulse HT 51 & 52 Dry down LT 53 -Final 27 cores destructively sampled -HT= high tide ; LT= low tide
126 Table 4 3 I ntact core soil properties. Soil properties (mean standard deviation; n = 3) by wetland type and soil depth for the control condition. Different letters represent significantly different means (p < 0.05) based on a two way ANOVA. Depth Bulk Density Organic Matter Total C Total N (cm) (g cm 3 ) (%) (%) (%) Freshwater 0 5 0.17 0.10 a 40.7 7.9 ab 19.8 1.1 a 1.25 0.03 a 5 10 0.18 0.01 a 43.8 1.1 ad 19.9 1.0 a 1.23 0.03 a 10 20 0.19 0.01 a 42.8 4.7 abd 19.8 2 .8 a 1.10 0.06 b Brackish 0 5 0.26 0.02 b 33.7 1.9 bc 13.5 1.0 b 0.72 0.06 c 5 10 0.28 0.01 b 32.2 1.5 bc 12.5 1.0 b 0.70 0.02 c 10 20 0.28 0.01 b 28.7 0.8 c 11.5 1.3 b 0.66 0.03 c Salt Marsh 0 5 0.18 0.02 a 43.2 1.1 abd 17.9 1.0 a 1.1 0 0.01 b 5 10 0.16 0.02 a 51.6 2.8 d 21.9 1.6 a 1.13 0.03 b 10 20 0.17 0.01 a 46.2 3.0 abd 19.4 1.6 a 1.07 0.12 b Table 4 4 I ntact core s urface water properties Surface water properties (mean standard deviation; n = 12) for the control condition by wetland type. Different letters represent significantly different means (p < 0.05) based on a two way ANOVA. Wetland Type Sulfate Conductivity DOC NH 4 + TKN SRP (mg L 1 ) (mS cm 1 ) (mg L 1 ) (mg L 1 ) (mg L 1 ) (mg L 1 ) Freshwater 43 6 a 1.0 0.1 a 11.5 2.1 0.19 0.01 a 0.91 0.12 0.16 0.01 a Brackish 1485 172 b 20.5 1.2 b 9.6 2.6 0.03 0.04 b 0.69 0.09 0.07 0.02 b Salt Marsh 3199 842 c 39.3 1.8 c 8.3 2.1 0.13 0.05 ab 0.78 0.18 0.04 0.02 b
127 Table 4 5 Correlations for the control condition. coefficients for variables associated with SOC loss for the control condition of each wetland type (Figure 4 3). For each variable, n = 9 and df = 7; at p = 0.05, r = 0.67, and at p = 0.01, r = 0.79. DOC Release CH 4 Flux qCO 2 SRP Org N Salinity CO 2 Flux at HT 0.83 0.69 0.92 0.89 0.85 0.71 CO 2 Flux at LT NS NS NS NS NS NS CH 4 Flux 0.93 NS NS NS NS DOC Release 0.93 0.67 0.81 0.85 0.87 NS = not significant; HT = high tide; LT = low tide ; DOC = dissolved organic carbon; SRP = soluble reactive phosphorus Table 4 6 Correlations for the salinity pulsing events. coefficients for variables associated with SOC loss for the salinity pulsing event treat ments (Figs. 4 and 5). For each variable, n = 26 and df = 24; at p = 0.05, r = 0.39, and at p = 0.01, r = 0.50. DOC Release CH 4 Flux NH 4 + qCO 2 SRP MBC CO 2 Flux at HT 0.73 0.51 NS NS 0.57 NS CO 2 Flux at LT NS NS 0.41 NS NS 0.39 CH 4 Flux 0.69 NS 0.40 0.47 NS DOC Release 0.69 0.57 0.53 0.76 0.39 NS = not significant; HT = high tide; LT = low tide; DOC = dissolved organic carbon; SRP = soluble reactive phosphorus; MBC = microbial biomass carbon Table 4 7 Nutrient release for the control conditio n. DOC and nutrient release (mean standard deviation; n = 3) for the control condition of each wetland type. Significantly different means (p < 0.05) represented by different letters and based on a repeated measures two way ANOVA. Wetland Type DOC NH 4 + TKN Org N SRP (mg L 1 ) (mg L 1 ) (mg L 1 ) (mg L 1 ) (mg L 1 ) Freshwater 18.3 2.8 a 0.18 0.15 a 1.18 0.30 1.00 0.34 a 0.68 0.50 a Brackish 10.6 2.1 b 0.16 0.21 a 1.07 0.24 0.91 0.47 a 0.30 0.28 a Salt Marsh 7.6 0.7 c 0.71 0.44 b 1.33 0. 40 0.63 0.17 b 0.01 0.01 b DOC = dissolved organic carbon; TKN = total kjeldahl nitrogen; SRP = soluble reactive phosphorus
128 Table 4 8 Microbial indicators for the control condition. Microbial indicators (mean standard deviation; n = 3) for the co ntrol condition of each wetland type (0 5 cm). Significantly different means (p < 0.05) represented by different letters and based on a two way ANOVA. Wetland Type MBC TLOC LOC qCO 2 gluco sidase Dehydro genase (mg kg 1 ) (mg kg 1 ) (mg kg 1 ) (mg CO 2 C k g 1 d 1 / mg kg 1 ) (mg MUF kg 1 h 1 ) (mg TPF kg 1 h 1 ) Freshwater 706 157 a 901 214 a 170 30 0.26 0.14 a 318 139 a 148 25 Brackish 606 187 a 755 175 a 163 23 0.21 0.17 ab 90 25 b 31 11 Salt Marsh 2105 159 b 2345 133 b 245 29 0.06 0.01 b 250 27 ab 165 89 MBC= microbial biomass carbon; TLOC= total labile organic carbon; LOC= labile organic carbon Table 4 9 Summary of salinity pulsing effects. Summary of the observed salinity pulsing event treatment affects o n SOC loss, nutrient release, and microbial indicators. Wetland Type Salinity Pulsing Event Observed Effect compared to control condition Freshwater Brackish (13 ppt) water Increase in LT CO 2 flux (p = 0.004) Decrease in CH 4 flux (p < 0.001) Decrease in DOC release (p = 0.006) Increase in NH 4 + release (p = 0.02) Altered partitioning of SOC loss (p = 0.05) Freshwater Salt (26 ppt) water Decrease in CH 4 flux (p < 0.001) Brackish Fresh (0.5 ppt) water Decrease in HT CO 2 flux (p = 0.02) Brackish Salt (26 p pt) water No effect Salt marsh Fresh (0.5 ppt) water Increase in LT CO 2 flux (p = 0.02) Increase in DOC release (p = 0.006) Salt marsh Brackish (13 ppt) water Increase in MBC (p < 0.001) Increase in TLOC (p < 0.001) LT = low tide; HT = high tide; DOC = dissolved organic C; MBC = microbial biomass C; TLOC = total labile organic C
129 Figure 4 1 Site map. Aerial photograph indicating the approximate locations of each of the 3 wetland types in Jacksonville, FL, USA (see text for coordinates). All sites we re Figure 4 2 A schematic of the intact soil core design. The cores were stored in the dark at 25 C and remained open to the atmosphere. The bottom was sealed with a cap during high tide and all CO 2 flux measurements; the cap was removed and water allowed drainage to a leachate collection bucket during the ebb tide.
130 Figure 4 3 Soil fiber content. Percent fiber content in soils (0 10 cm) according to wetland type. Each bar represents mean percent of t he control condition (n = 12); error bars represent standard error. Different letters represent significantly different means (p < 0.05) based on a one way ANOVA.
131 Figure 4 4 Tidal cycle effects on soil respiration. Effect of tidal cycle (low tide = post 24 h dry down; high tide = 10 cm water column) on CO 2 flux rate according to wetland type. Each bar represents mean flux rate of the control condition (n = 9); error bars represent standard error. Different letters represent significantly different means (p < 0.05) based on a one way ANOVA.
132 Figure 4 5. High tide salinity pulse effect s on soil respiration. Effects of salinity pulse (high tide) on CO2 flux rate according to wetland type and surface water salinity. Horizontal bars represent mean (solid line) and standard error (dashed lines) of the control condition. Each bar represents mean flux rate (n = 18); error bars represent standard error; NS = not significantly different from the control condition; significantly different means (p < 0.0 5) represented by percentages and based on a two way ANOVA.
133 Figure 4 6 Low tide salinity pulse effect s on soil respiration. Legacy effects of salinity pulse on subsequent low tide CO 2 C flux rate according to wetland soil type and pulse salinity. Hor izontal bars represent mean (solid line) and standard error (dashed lines) of the control condition. Each bar represents mean flux rate (n = 18); error bars represent standard error; NS = not significantly different from the control condition; significan tly different means (p < 0.05) represented by percentages and based on a two way ANOVA.
134 Figure 4 7 High tide salinity pulse effects on potential methanogenesis. Effects of salinity pulse (high tide) on CH 4 C flux rate according to wetland soil type an d water salinity. Horizontal bars represent mean (solid line) and standard error (dashed lines) of the control condition. Each bar represents mean flux rate (n = 3); error bars represent standard error; NS = not significantly different from the control condition; significantly different means (p < 0.05) represented by percentages and based on a two way ANOVA.
135 Figure 4 8 Total soil organic carbon loss. Mass of carbon los t (CO2 C + CH4 C, + DOC) during the 53 day laboratory experiment. Each bar repre sents mean mass C (n = 3); error bars represent standard error; significantly different means (p < 0.05) represented by different letters based on a two way ANOVA.
136 Figure 4 9 Soil organic carbon loss by pathway. Mean % of soil organic C lost by pat hways (controlling for duration of measurement) according to soil type. Charts represent control condition (n = 3); the freshwater marsh soil treated ol, according to a chi square goodness of fit ( p < 0.05).
137 Figure 4 10 Modeled effects of increased inundation on soil carbon loss. A theoretical model indicating the effects of increased inundation (increased time of high tide (HT) ; decreased time of ) on total soil organic C loss in 3 wetland types. Model projections assume current conditions are 50% HT, and 50% LT. The graph illustrates an increased to 100% of time under HT conditions and 0% under LT condition. Model assumes a linear response o f soil respiration based on Figure 4 8 data.
138 CHAPTER 5 BIOGEOCHEMICAL EFFEC TS OF SIMULATED SEA LEVEL RISE ON CARBON LOSS IN AN EVERGLADE S PEAT SOIL Background Mangroves have higher rates of net primary productivity and carbon (C) sequestration than most forest ecosystems, with global estimates of 882,215 and 102,300 Mg C km 2 y 1 respectively (Donato et al. 2011; Bouillon et al. 2008 a ). Despite the critical role of mangroves in global C cycling and stor age, they are experiencing an a r eal loss of ~2 % per year (Valiela et al. 2001). A substantial portion of this loss is due to human activities (e.g., development, wood harvest, mariculture), but environmental stressors also contribute. Being located on the land fringe means exposure to extreme environment al conditions, such as intense wave action, high winds, storm surges, and sea level rise (Alongi 2008). The Everglades (south Florida) contains the largest mangrove forest in the contiguous USA (>1,400 km 2 ) and overlies 5 to 6 m of peat soils (Welch et al. 1999; Spackman et al. 1966). Net ecosystem productivity in the mangroves of the Everglades is high, averaging about 1170 g C m 2 y 1 but some characteristics of this expansive ecosystem are suggestive of a high vulnerability to submergence from sea level rise (e.g., low relief carbonate geology and micro tidal inundation; Barr et al. 2010 Alongi 2008; Nicholls et al. 1999). It is estimated that approximately half of Everglades National Park is currently within 0.3 and 0.6 m of sea level (Titus and R ichman 2001). According to predictions by the Intergovernmental Panel on Climate Change (IPCC) for 2100 (IPCC 2007).
139 All coastal wetlands, including mangrove forests, o ccupy a specific elevation niche relative to sea level and thrive when the marsh platform is between mean sea level and mean high tide (Morris et al. 2002). As sea level rises, coastal wetlands must maintain this ideal elevation by processes of vertical s oil accretion or landward migration. Vertical accretion occurs through a natural feedback loop: increased tidal inundation enhances sedimentation, inorganic sediments accelerate primary production and organic matter accumulation, and the marsh platform ac cretes vertically to achieve disturbance, adequate nutrient supply, sediment rich coastal waters, and a moderate to large tidal amp litude (Nicholls et al. 1999). Landward migration is also a natural response to SLR observed in coastal wetlands with low relief, exposed to gradual changes in sea level, and with minimal adjacent development. Duri ng landward migration, the seaward edge of a wetland may submerge, but the wetland simultaneously expands landward to maintain the optimal degree of inundation and salinity with minimal loss of overall area (Adam 2002; Donnelly and Bertness 2001). If sea level rises too quickly and/or the local conditions do not support sufficient vertical accretion or landward migration, the wetland may experience decreased productivity, vegetation death, peat collapse, and eventual submergence (Nyman et al. 1990). Under current rates of SLR, it is predicted that between 20 and 30 % coastal wetlands will be lost by the end of the century (IPCC 2007; Nicholls et al. 1999). The ecological importance of the Everglades as a habitat for wildlife, epicenter of gro undwater recharge, recreational haven, and provider of numerous other ecosystem
140 services (e.g., water quality improvement, nutrient cycling, and C storage) has generated a strong interest in the sustainability of the Everglades during SLR. Landward migrat ion in response to SLR and freshwater inputs has been documented in the southeastern portion of the Everglades, near Taylor Slough. Here, vegetation zones have shifted as much as 3.3 km inland since 1940, as indicated by historic aerial photographs and bi ological indicators of past salinity regimes (Gaiser et al. 2006; Ross et al. 2000). Shark River Slough is a slow moving river that drains the majority of the Everglades and tends to have higher nutrient availability, less groundwater influence, greater primary productivity, higher soil organic matter, and a larger tidal range than Taylor Slough (Rivera Monroy et al., 2011; Ewe et al. 2006). This study sought to understand the fate of soil organic C in Shark River Slough under conditions of rising salini ty, increased inundation, and the combination thereof. Two possible fates of wetland soil organic C include: ( 1) decompos ition an d released back to the atmosphere as CO 2 and CH 4 or ( 2) loss via export within the water column as dissolved organic C (DOC) or dissolved inorganic C (DIC) The remaining C can remain stored in the soil for days to centuries. Increasing surface water salinity from SLR introduces higher concentrations of sulfate (SO 4 2 ) to the soil microbes of coastal wetlands. Sulfate is c apable of functioning as an alternative electron acceptor during microbial respiration, resulting in the mineralization of soil organic C and the release of CO 2 to the atmosphere (Patrick and DeLaune 1977). Previous studies have found that increasing SO 4 2 availability in wetland soils can stimulate sulfate reducing bacteria, and consequently accelerate the loss of CO 2 from the soil (Chambers et al. 2011; Weston et al. 2006 b ). However, high
141 SO 4 2 availability is also known to repress the activity of metha nogens, decreasing the loss of CH 4 (DeLaune et al. 1983; Jackobsen et al 1981). Inundation is known to affect microbially mediated C loss by creating anaerobic conditions under which microbial activity and efficiency is greatly reduced, and subsequently t he rate of CO 2 loss also declines ( Wright and Reddy 2001; DeBusk and Reddy 1998). In coastal wetland soils, inundation is inextricably linked to tidal cycle and the rate of CO 2 loss during low tide can be 53 to 310 % greater than the rate observed in the s ame soils during high tide (Cha pters 4 and 5 ). Despite a growing understanding of the effects of salinity (Chambers et al. 2011; Weston et al. 2011; Edmonds et al. 2009; Weston et al. 2006) and tidal cycle (Cha pter 4 ; Neubauer et al. 2000) on coastal wet land C loss, there is still a significant gap in the understanding of the compounding effects of simultaneous increases in both salinity and inundation. The goals of this study were to: ( 1) disentangle the effects of increased salinity, increased inundatio n, and the combination thereof, on soil C loss in an Everglades peat soil ( 2) determine the effect of physical environmental factors (temperature, the rising and falling action of tides, and soil macroinvertebrates) on the rate of C loss and ( 3) determin e the impact of simulated sea level rise on porewater DOC and nutrient availability. Porewater nutrients are strongly coupled with nutrient export in mangroves and can significantly affect C mineralization rates (Bouillon et al. 2007). I hypothesized t hat increasing salinity in this system, which is already mesohaline (~10 20 ppt) would not significantly alter the rate of C loss; increasing inundation by ~8 cm would reduce the overall rate of C loss; and the effect of increased salinity and inundation o n C loss would be additive. I also expected that temperature and porewater
142 phosphorus (P) concentration (a limiting nutrient in this system) would be positively correlated with CO 2 flux. I tested these hypotheses by manipulating salinity and inundation f or 24 field replicate peat monoliths collected from Shark River Slough utilizing a tidal mesocosm facility Methods Study A rea Twenty four peat monoliths were collected in a mature mangrove forest adjacent to lower Shark River Slough (Figure 1; 2521'52 Everglades Long Term Research site SRS 6 (http://fcelter.fiu.edu/research/sites/)) in the SW Everglades (Florida, USA). The collection area is approximately 4 km inland from the Gulf of Mexico and experiences semi diurna l tides (mean range of 1 m) and seasonally driven (June Oct.) freshwater inputs (Rivera Monroy et al. 2007). The forest is comp ri sed of Rhizophora mangle Laguncularia racemosa, and Avicennia germinans of large stature (10 14 m tall; Chen and Twilley 1 999). The peat monoliths (25 cm deep x 28 cm diameter) were removed from the ground intact, carefully placed in heavily perforated buckets (~16 holes of 3.5 cm diameter; Figure 5 2), and transported to an outdoor tidal mesocosm facility located at the Flo rida Bay Interagency Science Center (Key Largo, FL, USA). At the time of collection, ambient salinity at Shark River Slough was ~17 ppt. Experimental D esign The experiment consisted of a randomized split plot design with repeated observations. The two m anipulated factors were salinity (the whole plot factor with two nested blocks) and inundation (the sub plot factor). Meanwhile, time (days) and water temperatures ( C) were recorded as potentially important covariant factors The study
143 ran for 10 weeks including a 3 week acclimation per iod, 1 week salinity ramp, and 6 week experiment. Once the 24 peat monoliths were brought to the mesocsom facility from the field, standard window screen mesh was affixed to the outside of the buckets to allow for water exchange through the perforations while preventing the loss of soil material (Figure 5 2). Pneumatophores, when present, were clipped to the soil surface and any identifiable litter material was removed. The 24 monoliths were randomly assigned to 1 of 4 proof containers, 0.74 m D x 0.80 m W x 2.21 m L) and 1 of plac ed directly on the crypt floor. A 10 cm diameter PVC collar (10 cm long) was inserted near the center of each peat monolith (Figure 5 2) for placement of a portable 10 cm diameter gas chamber used to collect CO 2 and CH 4 flux measurements. The centralized placement of the collars reduced the likelihood of any soil profile disturbance within the sampling area. A porewater sipper was also inserted (through the side bucket perforations) into the center of each peat monolith at a depth of 10 cm from the soil surface (Figure 5 2). The sipper consisted of a 5 cm long air stone (1 cm diameter) attached to a 1 m long nalgene tube. Oxidation reduction probes were also installed in a subset of monoliths at a depth of 10 cm. All four crypts were filled with ambi ent salinity surface water (15 20 ppt) and the monoliths were allowed to acclimate for 3 weeks while exposed to simulated semi diurnal tides. The outdoor Key Largo meso cos m facility was constructed in 1995 and consists of 12 large cement water tight crypts each of which is equipped with a water inflow, a
144 stand pipe with a ball value spigot to manipulate high tide (HT) and low tide (LT) water levels, and an outflow drain. Four ~9,500 L head tanks are located on an adjacent earthen mound and deliver water v ia gravity flow through a system of PVC pipes to a manifold. The manifold allows for the manual mixing of water from multiple head tanks, and the delivery of water to individual crypts. For this experiment, two head tanks were and were filled with surface water drawn directly from the adjacent Florida Bay (~60 m away) using a hydraulic pump and belowground system of were filled with surface water transported via tanker truck from the C 111 canal (~28 km north). Due to the challenge of maintaining constant head pressure in the tanks during the filling of the crypts, a double hartford loop was constructed using two of the unused crypts, two raised water barrels equipped with stand pipe drains, and multiple aquarium pumps to circulate the water between the holding reservoirs. Semi diurnal tides (every 6 h) were run continuously throughout the entire study using a 20 cm tidal range. The control wat er level peat monoliths had a LT at 12.7 cm and a HT at +12.7 cm, relative to the soil surface, whereas the inundated treatment had a LT at 5.1 cm and HT at +20.3 cm. Following the 3 week acclimation period, the 2 crypts assigned to the elevated salinit y treatment were gradually ramped up to between 30 35 ppt for the remainder of the study. The remaining 2 crypts continued to receive ambient salinity (15 20 ppt) surface water. All crypts were partially shaded with three layers of mesh window screen to simulate the diffuse sunlight penetrat ion of a natural mangrove canopy.
145 Soil and W ater P roperties Soil properties were determined for 5 initial soil cores (5 cm diameter x 25 cm deep) collected at the same time as the peat monoliths. At the conclusion of the 10 week study, all 24 peat monoliths were destructively sampled for an analysis of soil properties. During destructive sampling, monoliths were cored at the same location the soil collars were previously installed using a 10 cm diameter, 25 cm long P VC tube. The soils were divided into 3 depth segments (0 5, 5 15, and 15 25 cm) and analyzed for % moisture, bulk density, % organic matter, total C, and total N. Moisture content and bulk density were determined after drying a subsample at 70 C to a co nstant weight. Percent organic matter was estimated by mass loss on ignition (LOI) where dry soils were combusted at 550 C for 5 h and final weight was subtracted from initial weight. Total C and N content were determined using a Carlo Erba elemental an alyzer ( CE Elantech, Inc., Lakewood, NJ). Surface water salinity and temperature in the crypts was recorded twice daily at HT using a handheld YSI (YSI Inc., Yellow Springs, OH). General water quality parameters (NO 3 NH 4 + SRP, and DOC) of the sourc e water w ere gathered from near by long term monitoring locations ( SFWMD 2012 ). CO 2 and CH 4 F lux Soil respiration (CO 2 flux) was measured ~3 times per week at daytime LT for all 24 soils using a portable infrared gas analyzer (Li Cor 8100, Lincoln, NB) eq uipped with a 10 cm diameter chamber. Each flux rate was collected for 75 seconds before manually moving the chamber to another sample collar. Nighttime CO 2 flux was measured once per week on all monoliths at LT using the same procedure as just discribe d Rising and falling tide CO 2 flux measurements were also obtained once per
146 week on a sub set of 12 soils for the 3 hours directly proceeding and following the daytime LT. This involved 3 6 consecutive flux readings on each monolith as the tide was risin g and falling while concurrently measuring and recording the exact height of soil exposed above the water line at the time of each reading. Methanogenesis (CH 4 flux) was measured once a week on a sub set of monoliths using the 10 cm diameter chamber and a trace g as s ampling k it (Li Cor 8100 664 Lincoln, NB) which allowed of in line gas extraction. For each monolith sampled, t he cham ber was sealed for 20 min and a 5 mL gas sample w as extracted at 5 min intervals and transferred to a 5 mL glass vial (previo usly capped with a butyl stopper and aluminum crimp cap and evacuated to 75 kpa). Within 48 h of collection, CH 4 gas samples were run on a gas chromatograph (Shimadzu Scientific Instruments GC 8A, Columbia, MD) fitted with a flame ionization detector (FI D). Methanogenesis was calculated as the slope of CH 4 C concentration over time. DOC and Porewater N utrients S oil porewater (60 mL) was extracted from the porewater sippers ( 10 cm below the soil surface ) once per week during HT from each of the mon o l iths by applying suction with a plastic syringe. Water was field filtered through 0.45 m filter paper, transferred to a 60 mL nalgen e bottle, and stored at 20 C prior to analysis (within 30 d). Porewater was analyzed for NO 3 NO 2 NH 4 + soluble reacti ve P (SRP), and DOC at the Southeastern Environmental Research Center Nutrient Analysis Laboratory (Florida International University, Miami, FL) Nitrogen and phosphate were analyzed on a four channel Alpkem RFA 300 auto analyzer ( OI Analytical, College S tation, TX ) and DOC was analyzed on a Shimadzu 5000 TOC (Shimadzu Scientific Instruments, Columbia, MD).
147 Data A nalysis Statistical analysis was performed using SAS 9.1 software (SAS Institute Inc., Cary, NC). A heteroscedastic split plot repeated measures linear mixed model (Proc GLIMMIX) was developed to test mean differences and interactions between the 4 main factors of interest: salinity, inundation, time, and temperature. The random effects of crypt and replication were also included in the model, alo ng with an auto regression coefficient that improve d model fit. Response variables included : day and night CO 2 flux, DOC, and porewater nutrients. Where time and temperature were not significant factors, such as with methane flux, a two way ANOVA (Proc G LM) was used. calculated to determine correlation coefficients between CO 2 flux, DOC production, porewater nutrients, and environmental variables. All analyses were performed using R esults Soil and W ater P roperties Soil redox potential at 10 cm indicated all soils were under reduced conditions (Figure 5 3). The only treatment that showed a signific antly different redox potential between HT and LT was the control water level elevated salinity treatment ( 72 mV at LT and 166 mV at HT). The inundated ambient salinity treatment had the lowest average redox potential, with an average Eh of 195 mV. The soil bulk density was 0.24 0.4 g cm 3 (mea n standard deviation) at 0 5 cm, which was significantly higher (p < 0.001) than deeper in the soil profile (5 25 cm 0.17 0.2 g cm 3 ). After the 10 week study, soils subjected to the increased inundation treatments had significantly lower surface soi l (0 5 cm) bulk density (0.20 0.3 g cm 3 ) than soils maintained at the control water level ambient salinity treatment (p = 0.009;
148 Figure 5 4). Fiddler crabs ( Uca pugnax ) were inadvertently brought to the mesocosm facility in the peat monoliths. Crab colonization (as indicated by the number of visible burrow holes in the soil surface ) was significantly greater in the inundated ambient salinity treatment (10 4) compared to the inundated elevated salinity treatment (p = 0.04; 4 3) and had a weak nega tive correlation (p = 0.08) with surface soil bulk density. Average organic matter content in the peat soils was 53 7 % total N averaged 1.3 3 % and total C averaged 22 4 % All three properties were significantly lower (p < 0.05) is the surface s oil (0 5 cm) compared to deeper in the soil profile (5 25 cm) and were not affected by the inundation or salinity treatment (Table 5 1). Soil pH was between 7.8 and 8.0. All 4 crypts maintained salinities between 15 and 20 ppt for the first 3 weeks of th e study (the acclimation period). During the 4 th week, the crypts assigned to the elevated salinity treatment were slowly ramped up (~2 ppt d 1 ) to 30 35 ppt, and remained within that salinity range for the remaining 6 week study period. The freshwater ( trucked in from the C 111 canal) had a salinity of < 0.5 ppt throughout the study period, while the saltwater (collected directly from Florida Bay) began at ~42 ppt and decreased to ~34 ppt by the conclusion of the study. This decrease was due to the on set of the wet season causing lower salinities in all of Florida Bay. Nutrient concentrations were significantly higher in the fresh source water (Table 5 2), but were always diluted by mixing with saltwater to achieve the ambient (15 20 ppt) and elevated (3 0 35 ppt) salinities used in the experiment.
149 CO 2 and CH 4 F lux Daytime LT CO 2 flux rate was significantly lower in the inundated treatments (644 442 mg CO 2 C m 2 d 1 ) compared to the control water level treatments (987 618 mg CO 2 C m 2 d 1 ; p < 0.0001) but the main effect of salinity was not significant during the day (p = 0.4; Figure 5 5). The response of CO 2 flux rate to inundation differed according to the salinity treatment (inundation*salinity effect; p < 0.0001), with the highest average daytime CO 2 flux observed in the control water level elevated salinity treatment (1189 699 mg CO 2 C m 2 d 1 ), followed by control water level ambient salinity (770 426 mg CO 2 C m 2 d 1 ), inundated ambient salinity (663 419 mg CO 2 C m 2 d 1 ), and inundated e levated salinity (624 484 mg CO 2 C m 2 d 1 ) treatments Nearly all of the variability in CO 2 flux rate not explained by the main effects of salinity and inundation were explained by time (p < 0.0001; Figure 5 6) and water temperature (p < 0.0001; Figure 5 7). Time produced a significant non linear effect on CO 2 flux and within treatment variance in CO 2 flux decreased with time (Figure 5 6). The effect of water temperature on CO 2 flux was dependent upon inundation level (temperature*inundation effect, p = 0.009), and revealed a polynomial relationship with a steeper slope in th e control water level treatment compared to the inundated treatment (Figure 5 7). Nighttime CO 2 flux showed a similar pattern to daytime flux with significantly lower flux rat es in the inundated treatments (p < 0.0001), an overall decrease in CO 2 flux over time (p < 0.0001), and a significant interaction between inundation and salinity (p = 0.02). However, salinity was a significant main effect for CO 2 flux at night (p = 0.02) with the elevated salinity treatment (1134 731 mg CO 2 C m 2 d 1 ) being higher than the ambient salinity treatment (940 610 mg CO 2 C m 2 d 1 ). On average,
150 nighttime CO 2 flux rate was ~20 % higher than daytime flux rate, indicating the presence of phot osynthetic activity within the soils during the experiment. Rising and falling tides did not result in significant differences in the response of CO 2 flux rate. The slope (change in CO 2 flux rate/ change in cm of soil exposed above the waterline) was high est in the control water level treatment (93 to 121 mg CO 2 C m 2 d 1 ) and lowest in the inundated elevated salinity treatment (42 to 51 mg CO 2 C m 2 d 1 ), but was not significantly different when the tide was rising versus falling (Table 5 3). Methane flux was significantly lower in the control water level ambient salinity treatment than all other treatments (p = 0.001; Figure 5 8). On average, the inundated ambient salinity treatment had the highest flux rate (27 mg CH 4 C m 2 d 1 ) and all treatments had h igh within treatment va r iance DOC and P orewater N utrients Porewater DOC was significantly greater (linear mixed model; p = 0.04) in the inundated treatments (17.4 7.2 mg L 1 ) compared to the control water level treatments (13.3 4.7 mg L 1 ;Table 5 4) The response of DOC not explained by the main factors of salinity and inundation, were explained by a positive relationship with porewater salinity (continuous variable; p = 0.008) and water temperature (p < 0.0001). Porewater nitrite, ammonium, and SRP were all positively correlated with DOC, while CO 2 flux rate was inversely correlated with DOC (Table 5 5). Porewater nitrate concentration showed a significant non linear relationship with time and water temperature ( both p < 0.0001). Neither inundatio n nor salinity caused a significant response in NO 3 concentration, but the control water level ambient salinity treatment had, on average, the highest porewater NO 3 (1.65 0.29 mg L 1 ; Table 5 4). There was no difference between treatments in terms of NO 2 concentration (averaging
151 0.25 0.25 mg NO 2 L 1 ), but nitrite was affected by time and water temperature (both p < 0.0001). Nitrite concentration was positively correlated with the other N species (NO 3 and NH 4 + ), DOC, salinity, and SRP, but inversely related to CO 2 flux (Table 5 5). Ammonia concentration was moderately affected (p = 0.08) by inundation and significantly affected by the interaction between salinity and inundation (p = 0.003), with the inundated elevated salinity treatment having the h ighest average NH 4 + porewater concentration (287.7 78.5 mg L 1 ; Table 5 4). Ammonium concentration was positively correlated with DOC, salinity, NO 2 and SRP, and inversely related to CO 2 flux (Table 5 5). Soluble reactive phosphorus concentrations in the porewater were significantly affected by inundation (p = 0.0002), with the inundated treatment averaging 9.8 5.2 mg SRP L 1 and the control water level treatment averaging 7.0 5.5 mg SRP L 1 A significant interaction between salinity and inundati on was also found (p = 0.04) with the inundated elevated salinity treatment having the highest SRP concentration 12.0 3.0 mg L 1 SRP concentrations increased over time and w ere positively correlated with DOC, salinity, NH 4 + and NO 2 (Table 5 5). Discuss ion Soil and Water P roperties Sulfate redu ction is the dominate pathway for soil respiration in saline wetlands, accounting for 70 to 99 % of C mineralization and typically occurring at redox po tentials between 100 and 150 mV (Kostka et al. 2002; Howarth 1984; Patrick and DeLaune, 1977). The peat soils of this study were poised for sulfate reduction (Figure 5 3), with the inundated ambient salinity treatment (and the control water level elevated salinity
152 treatment during HT) being reduced enou gh to suppo rt methanogenesis (< 150 mV; Patrick and DeLaune 1977). Placed at a depth of 10 cm, the redox probes were above the elevation of the waterline in the control water level treatments during LT, but below the waterline in the inundated treatments during LT Only the control water level elevated salinity treatments demonstrated significantly different redox potentials between high and low tides (Figure 5 3), suggesting a relatively low hydraulic conductivity i n th ese soils. Peat soils consisting of highly decomposed organic matter, such as those used in this study, have the ability to retain water against the pull of gravity due to great capillary tension in the micropore spaces (Boelter 1965) and can decrease the importance of tidal fluctuations in coastal wetlands by continuing to promote anaerobic conditions even during LT conditions ( C hapter 4 ). In this study, the average increase in CO 2 C flux rate between HT and LT was 83 10 % in the control water level treatments and 57 19 % in the inundated treat ment. The magnitude of this tidal effect on soil respiration rate is comparable to that observed in a freshwater tidal marsh soil with a similar tidal range, but significantly lower than that of a brackish and salt marsh soil, also of a similar tidal rang e, which demonstrated a 230 310 % increase in CO 2 flux from HT to LT ( C hapter 4 ). soil material). In coastal wetlands, significant increases in surface soil bulk density are associated with subsidence and peat collapse, whereby pore space is lost when root causing compaction ( Franzen 20 06; Kool et al. 2006; Portnoy and Giblin 1997). This study found a significa nt decrease in surface soil bulk density (0 5 cm) over just a 10 week period in
153 the inundated treatment, compared to the control water level ambient salinity treatment (Figure 5 3). Intuitively, it is possible a decrease in soil bulk density might immedia tely precede peat collapse and compaction. Correlations between decreased soil bulk density, sudden die back of Spartina alterniflora and longer inundation periods have been noted in the submerging salt marshes of coastal Louisiana (Stagg and Mendelssohn 2010). Vegetation was removed in the present study, and although the severing and death of root structures during the collection of the peat monoliths may treatment so ils, in particular, experienced a significant decline in bulk density. One explanation may be abio tic processes, such as increased shear stress or water turbulence when the depth of the water column increase d which has been observed along tidal creeks as water level rises (Fagherazzi et al. 2012). Excessive water logging has also been shown to reduce soil redox potential and accelerate leaching of certain soil elements, such as iron (Stagg and Mendelssohn 2010; Craft et al. 2002), and may have contribute d to a decline in soil material. T he activity of fiddler crabs inadvertently brought from the field to the experimental mesocosms within the peat monoliths are suspected to have played a role in the reduction of the soil bulk density in the inundated tr eatment. Fiddler crabs can directly reduce the soil bulk density through the construction of burrows within the soil profile, but also ingest and excrete sediments when scavenging for microalgae and soil by fiddler crabs likely reduced material on the surface of the monolith that was easily sloughed off with the rising and
154 falling of the tides. Observation in dicated the development of an unconsolidated soil layer that was more prevalent on the inundated monoliths ; the sloughing of soil material was also observed to be extensive. The split plot experimental design prevented the quantification of DOC export acc ording to water level treatment, but a weak (p = 0.08) correlation between surface soil bulk density and the number of fiddler crab burrows supports the hypothesis that this macroinvertebrate played a role in reducing soil bulk density. A study conducted in a South Carolina salt marsh found an inverse relationship between the number of fiddler crab burrows, vegetation health, and soil stability (Hughes et al. 2009). CO 2 and CH 4 F lux Daytime CO 2 flux was, on average, 34 % greater in the control water lev el treatment, compared to the inundated treatments (Figure 5 5). At LT, 13 cm of soil was exposed above the waterline in control water level treatments, compared to 5 cm exposed in the inundated treatments. The alleviation of electron pressure (i.e., inc reased O 2 diffusion) during LT is likely the main factor contributing to the difference in CO 2 flux in the two water level treatments ( Wright and Reddy 2001; DeBusk and Reddy 1998). The increase in salinity from 15 20 ppt (ambient) to 30 35 ppt (elevat ed) alone did not cause a significant change in CO 2 flux during the daytime measurements (p = 0.4), but did significant ly a ffect CO 2 flux at night (p = 0.004). Among all the treatments, average nighttime CO 2 flux was ~20 % greater than daytime rates, but t he control water level ambient salinity treatment, in particular, had the greatest difference between day and night flux rates (28 % ). This suggests photosynthetic activity in the control water level ambient salinity treatment was greater than the others, possibly due to the
155 combination of lower ionic stress (lower salinity), greater flushing (larger tidal range), and higher nutrient availability. Nutrient availability was slightly higher in the ambient salinity treatment because it received a larger propo rtion of water from the freshwater source (during the on site water mixing), and the freshwater source contained higher N and P concentrations than the saltwater source (Table 5 2). Ammonium and P, in particular, are limiting nutrients for algal productio n in coastal waters ; concentrations of these nutrients were 22 times higher in the fresh source water than the salt source water, possibly promoting higher rates of photosynthesis in the ambient salinity treatment (Boyer 2006; Howarth 1988). Therefore, re moving the effects CO 2 fixation by measuring flux rates after dark revealed salinity does exert a significant main effect on soil respiration, but was likely confounded by higher photosynthetic activity during the day. Increasing both salinity and inundati on simultaneously caused the greatest decline in average CO 2 flux rate (Figure 5 5). When considered individually, elevated salinity stimulated CO 2 flux relative to the control, and inundation only slightly suppressed it, so the fact that the inundated el evated salinity treatment had the lowest average CO 2 flux rate was surprising. Higher salinities have been shown to increase the rate of sulfate reduction and enhance soil respiration (Chambers et al. 2011), but the byproducts of sulfate reduction (HS an d S 2 ) can be toxic to plants and microorganisms (Joye and Hollibaugh 1995; Koch et al. 1990). Tidal flushing is critical to the removal of deleterious sulfide compounds (King et al. 1982) and wetland soils with low flushing often develop microzones where SO 4 2 becomes depleted and methanogenesis becomes the dominate pathway for C mineralization (King and Wiebe,
156 1980). In the present study, CH 4 flux was gre ater in the inundated treatment compared to the control water level ambient salinity treatment (Fig ure 5 8), indicati ng that the inundated treatment contained more microzones of SO 4 2 d epletion, lower redox potential (Figure 5 3), and presumably greater sulfide accumulation due to the smaller tidal range and reduced flushing. The synergistic decline in CO 2 flux when salinity and inundation increase simultaneously is an important consideration when predict ing the fate of coastal wetland C storage as sea level rises. For example, SLR responses in some coastal wetland areas, such as Taylor Slough, appear to be driven more so by changes in salinity (Ross et al. 2000), whereas saline marshes already receiving full strength seawater are more likely to experience changes driven by inundation (Kirwan and Temmerman 2009). Coastal wetlands experiencing both sali nity and inundations drivers will be subject to a synergistic decline in the rate of soil C cycling. Another question this study sought to answer was whether the physical process of a tide rising, versus falling, has an impact on CO 2 flux rate. For exampl e, does the upward momentum of a rising tide push additional CO 2 out of the pore space and therefore accelerate the rate of CO 2 flux? Quantifying the relationship between the instantaneous CO 2 flux rate and the exact height of soil above the waterline dur ing rising and falling tides revealed no significant effect of rising and falling tides on CO 2 flux rate (Table 5 3). However, the lack of difference may have been a consequence of the exclusion of vegetation from the study. Wetland plants are known to t ranslocate and release O 2 and CO 2 from their roots, causing a build up of gases in the soil that may be physically expelled from the soil during a rising tide. Therefore, this question of rising vs. falling tidal impacts on CO 2 flux should be revisited in a coupled soil plant system.
157 Methane flux rates were measured during LT and had a strong correlation with LT redox measurements (r = 0.94; Figs. 5 3 and 5 8). As expected, all CH 4 flux rates were low (between 0.2 and 27 mg CH 4 C m 2 d 1 ) because the ab undant SO 4 2 in seawater favors the more energy efficient process of sulfate reduction over methanogenesis (Jackobsen et al. 1981). The average methane flux rate in a coastal Louisiana salt marsh was 11.8 mg CH 4 C m 2 d 1 (DeLaune et al. 1983), which is c omparable to the present study. However, in mangroves the pneumatophores are thought to contribute significantly to CH 4 flux by providing labile C substrates; the removal of pneumatophores from this study may have caused a 12 50 % underestimation of CH 4 fl ux rates (Laanbroek 2010). DOC and P orewater N utrients Porewater DOC ranged from 12 to 18 mg DOC L 1 which is slightly higher than reported for other mangroves (e.g., 5 to 10 mg DOC L 1 for a mangrove forest in Tanzania; Bouillon et al. 2007). Dissolv ed OC was significantly greater in the inundated treatment, compared to th e control water level treatment (p = 0.04). Past studies have found higher DOC concentrations in high marsh (less inundated) areas, than in low marsh areas (Cao et al. 2008) and hig her porewater DOC during LT than HT (Bouillon et al. 2007). However, no studies have looked at differences in porewater DOC (or DOC export) as a function of SLR (Henman and Poulter 2008). It is estimated that wetlands account for ~20 % of all DOC export to the ocean (Lugo et al. 1989) and the greater the tidal range, the greater the proportion of C lost to the ocean (Twilley et al. 1992). Previous research has estimated that ~30 % of porewater DOC is advected and exported during the ebb tide (Bouillon et al. 2007). If this figure is applied to the current data, an increase in relative water level of 8 cm will
158 cause the export of porewater DOC in Shark River Slough to increase from an average of 4 mg L 1 to 5.2 mg L 1 Past estimates of total DOC export f rom Shark River Slough during the summer months range from 8 to 11 mg L 1 (Romigh et al. 2006). The impact of salinity and inundation on N cycling in coastal wetland soils is of interest due to the importance of N in coastal eutrophication and the role of nitrate as an alternative electron acceptor during C cycling. Salinity increases between 0 and ~10 ppt are known to cause the release of significant amounts of NH 4 + from the soil cation exchange complex into the porewater, but once the salinity has pas se d a threshold of ~10 ppt, the influence of this abiotic process is believed to be greatly diminished (Baldwin et al. 2006; Weston et al. 2006 b ; Rysgaard et al. 1999). This is corroborated by the present study, which s howed no significant increase in porew ater NH 4 + with elevated salinity (Table 5 4). However, this study did find evidence of increased NH 4 + concentrations when the affects of inundation and elevated salinity were combined. Under long periods of inundation, NH 4 + accumulates in the porewater b ecause the anaerobic conditions prevent nitrification (the oxidation of NH 4 + to NO 3 ; Reddy and DeLaune 2008). T he role of salinity in the accumulation of porewater NH 4 + at high concentrations of seawater was probably not due to the abiotic release mechani sms mentioned above, but the inhibition of nitrifying and denitrifying bacteria by high concentrations of chloride and sulfide in seawater (Seo et al. 2008; Hale and Groffman 2006; Joye and Hollibaugh 1995). The interaction of both greater anaerobosis and higher concentrations of Cl and S 2 could account for the highest porewater NH 4 + in the inundated elevated salinity treatment (Table 5 4). Similarly, the treatment with the
159 larger tidal range and lower salinity (the control water level ambient salinity treatment) had the highest observed NO 3 concentration (Table 5 4). Soluble reactive P in the porewater was significantly higher in the inundated elevated salinity treatment compared to the control water level ambient salinity treatment (Table 5 4). Prev ious studies have not found a significant relationship between salinity increases and phosphorus availability (Weston et al. 2011) but this effect may be explained by reduced microbial activity in the inundated elevated treatment, as evidence by the lower rate of CO 2 production in that treatment. Carbon B udget Soil C content in this peat soil averaged 2,094 307 g C m 2 in the top 0 5 cm, and 6,332 1,198 g C m 2 from 5 25 cm. Assuming the majority of C cycling is occurring in the upper 5 cm of so il, the 3 pathways measured in this study (CO 2 flux, CH 4 flux, and DOC export (calculated as 30 % of porewater DOC)) represent a mean mass loss of 3.1 1.4 % of the total soil C between 0 5 cm. On average, 95 % of the total C lost in this study was lost via CO 2 flux, 1 % to CH 4 flux, and 4 % to DOC export. The control water level elevated salinity treatment had a significantly higher (p = 0.02) % C mineralization over the 10 week study (4.5 % ), compared to the other 3 treatments (2.6 % ). However, these predict ions of C mineralization may be significantly underestimated because atmospheric CO 2 flux does not include the lateral transport of dissolved inorganic C, which often exceeds DOC export by a factor of 3 to 10 (Bouillon et al. 2008 b ). Implications for Eve rglades P eat S oils A final goal of this study was to enhance the mechanistic understanding of how SLR may impact biogeochemical processes in the peat soils of Shark River Slough.
160 The currently mesohaline salinity and low topography in this region of the Everglades suggests Shark River Slough will probably be subjected to simultaneous increases in both salinity and inundation as sea level rises. The findings of this mesocosm simulation indicate the rate of atmospheric CO 2 flux from the soil will decline w ith SLR. Methanogenesis and DOC production is expected to increase slightly, but represents such a small portion of the C budget (~4%) that the effect on the ecosystem will be negligible. The availability of NH 4 + and SRP in the porewater will increase si gnificantly. This increase in nutrients, in combination with a decrease in anaerobic soil respiration, may cause a n increase in nutrient export to Florida Bay. The increase in inundation could reduce the surface soil bulk density through increased leachin g and bioturbation, but under natural conditions, an increase in tidally deposited sediments could compensate for this decrease. Summary The high rate of primary productivity, high dens ity of soil C, and low release of greenhouse gases make mangroves an ideal ecosystem for global C sequestration and storage (Chmura et al. 2003). Unfortunately, mangroves are also highly susceptible to human disturbance and environmental stressors, which have caused a 35 % decline in global area since 1980 (Valiela et al. 2001). Results indicate that among the 3 pathways measured (CO 2 flux, CH 4 flux and porewater DOC ) 95% of all soil C lost from this mangrove peat soil was lost through CO 2 flux. Increasing tidal inundation from a LT water depth of 13 cm to a depth of 5 cm significantly decreased the rate of CO 2 flux by 14 % Increasing salinity from ambient (15 20 ppt) to 30 35 ppt increased the rate of CO 2 flux by an average of 54 % The combination of these two variables (increased inundation and elevated salinity) re sulted
161 in a synergistic decline in the rate of CO 2 flux, 19 % less than in the control treatment. The rate of CO 2 flux was positively correlated with water temperature and did not differ significantly whether the tide was rising or falling. Methane flux w as low in all treatments (0.2 to 27 mg CH 4 C m 2 d 1 ), but significantly greater in the increased inundation treatment Porewater DOC was generally higher in the inundated treatment and calculated DOC export represented ~4 % of all C lost. Sulfate reducti on was presumed the major pathway of C loss under all conditions. Increasing inundation and salinity resulted in significantly higher porewater NH 4 + and SRP concentrations, probably due to a combination of decreased soil redox potential, a decline in micr obial respiration, and the inhibition of N cycling by high concentration of chloride and sulfide. Significant physical changes in soil structure as a result of increased inundation were also noted. Specifically, there was a 19 % decrease in surface soil bu lk density after just 10 weeks of exposure to increased inundation. Past studies have suggested the role of increasing salinity (saltwater intrusion) in prom p ting peat collapse (Davis et al. 2005), but this study indicates increased inundation may actuall y contribute more to changes in soil bulk density than salinity in mesohaline mangroves.
162 Table 5 1. Soil properties. Percent organic matter (loss on ignition) total C, and total N according to soil depth. Values represent mean standard deviation. Dif ferent letters represent significantly different means based on a one way ANOVA (p < 0.05). Soil Depth Organic Matter Total N Total C (%) (%) (%) 0 5 45.4 5.7 a 1.0 0.2 a 19.9 3.3 a 5 25 56.1 5.3 b 1.5 0.2 b 23.5 2.9 b Table 5 2. Surfac e water properties. Properties of the fresh source water and salt source water mixed to achieve the desired salinities in the mesocosms during the 10 week study. Values represent mean standard deviation. Different letters represent significantly differe nt means based on a one way ANOVA (p < 0.05). Sal twater data from DB Hydro site FLAB24 (SFWMD, 2012) ; freshwater data courtesy of T. Troxler (Florida International University) site JB (2513'57.7"N, 8031'28.4"W). Parameter Freshwater Saltwater Salinity (ppt) <0.5 a 38 4 b NO 3 (mg L 1 ) 5.8 3.5 a 0.02 0.01 b NH 4 + (mg L 1 ) 2.4 0.7 a 0.11 0.08 b SRP (mg L 1 ) 0.02 0.001 a 0.001 .001 b DOC (mg L 1 ) 7.1 0.4 a 6.7 0.3 a SRP = soluble reactive phosphorus; DOC = dissolved organic carbon Ta ble 5 3. Rising vs. falling tidal affects on CO 2 C flux rate. Relationship between CO 2 C flux rate (mg m 2 d 1 ) and cm of soil exposed above the water line during rising and falling tides; data presented according to treatment and tidal condition. Water Level Salinity Tide Slope 2 C m 2 d 1 cm soil exposed) Difference (Rising Falling) Control Ambient Rising 121.1 26.3 Falling 94.8 Control Elevated Rising 93.0 23.4 Falling 116.4 Inundated Ambient Rising 74.4 42.6 Fallin g 117.0 Inundated Elevated Rising 41.8 8.9 Falling 50.7
163 Table 5 4. Soil porewater nutrients. Mean ( standard deviation) porewater concentration of DOC and nutrients presented according to treatment condition. Different letters represent sign ificantly different means based on a one way ANOVA (p < 0.05). Water Level Salinity DOC NO 3 NO 2 NH 4 + SRP (mg L 1 ) (mg L 1 ) (mg L 1 ) (mg L 1 ) (mg L 1 ) Control Ambient 14.8 1.6 ab 1.65 0.29 0.26 0.05 141.6 57.0 a 7.2 2.3 a Control Elevate d 12.0 2.9 a 1.28 0.13 b 0.27 0.06 140.9 67.1 a 8.6 3.7 ab Inundated Ambient 18.4 4.2 b 1.49 0.20 ab 0.27 0.08 211.7 89.9 ab 8.5 3.0 ab Inundated Elevated 17.5 5.0 b 1.16 0.22 b 0.32 0.13 287.7 78.5 b 12.0 3.0 b Table 5 5. Correlations between measured properties. coefficients (r values with p values in parentheses) for porewater parameters (n = 206), nighttime CO 2 flux (n = 163), and water temperature (n = 150). Time Temp DOC Salinity N O 2 NO 3 NH 4 + SRP Temp 0.873 (<.0001) DOC NS NS Salinity 0.378 (<.0001) 0.271 (<.0001) NS NO 2 0.503 (<.0001) 0.340 (<.0001) 0.223 (.010) 0.180 (.010) NO 3 0.248 (.0004) NS NS NS 0.341 (<.0001) NH 4 + 0.377 (<.0001) 0.28 7 (<.0001) 0.545 (<.0001) 0.323 (<.0001) 0.398 (<.0001) NS SRP 0.332 (<.0001) 0.312 (<.0001) 0.413 (<.0001) 0.251 (.0003) 0.305 (<.0001) NS 0.552 (<.0001) CO 2 0.382 (<.0001) 0.394 (<.0001) 0.198 (.005) NS 0.189 (.007) NS 0.431 (<.0001) NS NS = not significant
164 Figure 5 1. Site location map. This study was conducted in south Florida, USA. The aerial photograph indicates where the peat monoliths were collected (Shark p
165 Figure 5 2. Schematic of the peat monolith design. Standard window screen mesh was wrapped around the sides of the monolith to prevent soil loss through the bucket perforation, bu was inserted into the center of the peat monolith ( 10 cm depth) from the side of the bucket and a 10 cm diameter collar was centrally located.
166 Figure 5 3. Soil oxidation reduction (redox) pote ntial R eadings taken at a depth of 10 cm are presented according to treatment condition (C= control water level; I = inundated water level; A = ambient salinity (15 20 ppt); E = elevated salinity (30 35 ppt)) and tidal condition. Bars represent mean (n = 11); error bars represent standard error. Different letters represent significantly different means based on a two way ANOVA (p < 0.05).
167 Figure 5 4. Surface soil (0 5 cm) bulk density Data collected at the conclusion of the 10 week study accordin g to treatment condition (C= control water level; I = inundated water level; A = ambient salinity (15 20 ppt); E = elevated salinity (30 35 ppt)). Bars represent mean (n = 6); error bars represent standard error. Different letters represent significantly different means based on a two way ANOVA (p < 0.05).
168 Figure 5 5. Daytime CO 2 C flux by inundation and salinity. Data presented according to a) salinity treatment (n = 155), b) inundation (n = 155), and c) the combination of salinity and inundation (n = 78). Bars represent means; error bars represent standard error. Different letters represent significantly different means based on a heteroscedastic split plot linear mixed model.
169 Figure 5 6. Daytime CO 2 C flux over time The experimental pha se began 29 days after soil collection and ended 75 days after collection. Carbon dioxide flux presented a ccording to treatment condition (C= control water level; I = inundated water level; A = ambient salinity (15 20 ppt); E = elevated salinity (30 35 pp t)). Points represent means and error bars represent standard error.
170 Figure 5 7. Daytime CO 2 C flux over temperature. Relationship between daytime CO 2 C flux and water temperature (C) according to water level treatment (control water level or inund ated water level). Points represent means; error bars represent standard error. Trend lines are 2nd degree polynomial fits with regression coefficients (r values) indicated.
171 Figure 5 8. Methanogenesis by treatment condition. Methane (CH 4 C) flux a t low tide according to treatment condition (C= control water level; I = inundated water level; A = ambient salinity (15 20 ppt); E = elevated salinity (30 35 ppt)) and tidal conditions. Bars represent mean (n = 12); error bars represent standard error. D ifferent letters represent significantly different means based on a two way ANOVA (p < 0.05).
172 CHAPTER 6 CONCLUSIONS Background Coastal wetlands sequester and store an inordinate amount of global carbon ( C ) relative to their small a r eal coverage. Over the past century, human development and degradation and loss of thousands of hectares of coastal wetlands. Today, sea level rise ( SLR ) is introducing an additional threat to the health and persistence of coastal wetlands. This dissertation evaluates the ability of coastal wetland soils to store C under conditions of increasing sea level. Specifically, this research at tempts to answer the question: h ow do increasing salinity and inundation affect carbon biogeochemistry in coastal wetland soils? The first goal of this dissertation was to improve the mechanistic understanding of how salinity and inundation affect biogeochemical carbon cycling. This was accomplished by disentangli ng the components of salinity that affect soil microbial processes (C hapter 3), quantifying the end products of microbial processes (CO 2 CH 4 and DOC; C hapters 3 5), and measuring changes in the availability of nutrients, electron acceptors, and labile C as potential co variants impacting wetland biogeochemistry (C hapters 3 5). The second goal was to vary scale (i.e., degree of experimental control) to translate biogeochemical processes into real world observations. In each successive study ( C hapters 3, 4, and 5, respectively), the experimental approach progressed from: (1) a highly controlled microcosm slurry study (no diffusion barriers, DI water, constant temperature), to (2) an intact soil core study (intact soil profiles, in situ wa ter, constant temp erature), to (3) a peat monolith
173 mesocosm study (large soil area, in situ water, temperature and tidal fluctuations, and macroinvertebrate inclusion). Using this deliberate progression of increasing environmental complexity, specific biogeochemical mechan isms identified in the earlier studies were observed in the context of additional confounding and co varying factors that are important in natural systems. Finally, the third goal was to provide insight into the broader question of how coastal wetland C s torage might be affected by SLR. A review of literature revealed three landscape scale responses of coastal wetlands to SLR (C hapter 2). Quantifying the release and export of C by the three major microbial pathways (soil respiration, metha nogenesis, and DOC production; C hapters 3 5) suggested there are physical, chemical, and biological characterisitcs that help to predict the response trajectory of a specific coastal wetland to SLR. In the subsequent section, each research objective presented in the int roduction (C hapter 1) is re evaluated in the context of the new knowledge gained from this research and the three over arching goals are further discussed and dissected. Objective 1: Review of L iterature Coastal wetland C cycling is controlled by three m ain regulators, (1) the quantity and quality of C, (2) the availability of electron acceptors, and (3) the size, structure, and activity of the soil microbial community (C hapter 2). The quantity and quality of C in a wetland is a function of C inputs (pri mary production and inflows/deposition), the legacy of organic matter accumulation, and soil development. According to the literature, the rate of primary productivity is mainly driven by system specific characteristics (e.g., temperature and nutrient ava ilability), rather than salinity and inundation. However, a few general trends do orient along salinity gradients: plant diversity, litter quality, and organic soil C are inversely correlated with salinity while bulk density and inorganic soil
174 C are dire ctly correlated with salinity. In terms of electron acceptor availability, SO 4 2 and NO 3 are most affected by changes in salinity. Sulfate is the third most abundant ion in seawater and is responsible for nearly all anaerobic soil respiration at seawate r salinities >10 ppt. The N cycle (and NO 3 available for denitrification) experiences multi faceted consequences from salinity increases. At low salinities, an aboitic release of NH 4 + from the soil commonly occurs. At higher salinities, the accumulatio n of sulfide (a by product of sulfate reduction) and chloride can reduce the activity of nitrifiers and denitrifiers. In coastal wetlands with significant iron content, salinity can also affect Fe dynamics by promoting a shift from Fe P complexes to Fe S complexes, resulting in increased P availability. Current knowledge regarding the soil microbial community response to SLR is incomplete and includes numerous contradictions. In general, it is believed that microbial communities adapt quickly to changes in salinity and inundation, and that more saline wetlands tend to have lower rates of soil respiration and enzyme synthesis. Saline coastal wetlands (>18 ppt) also have significantly lower global warming potential than tidal freshwater wetlands due to l ow rates of CH 4 production and generally high rates of accretion from greater tidal deposition. Some of the most important knowledge gaps in the current literature include: (1) how does time scale (short term vs. long term) influence the observed changes in C biogeochemistry resulting from SLR (2) what is the mech anism for N cycling suppression, (3) how do microbial and fungal biomass vary along salinity gradients and (4) how do microbial scale biogeochemical processes relate to landscape scale responses of coastal wetland C storage.
175 Objective 2: Salinity and C arbon C ycling M ost of the research investigating biogeochemical responses to salinity to date has been performed in agricultural and uplands soils (e.g., Gennari et al. 2007 ; Muhammad et al. 2006; P attnaik et al. 2000) Previous studies indicated significant decreases in the rate of C cycling with increasing salinity but may have limited applicability to coastal wetlands exposed to SLR because of the central importance of anaerobic conditions and s ulfate availability in wetland C cycling. Therefore, the goal of the microcosm study presented in CHAPTER 3 was to determine the mechanism by which salinity alters C mineralization rates in wetland soils. This was done by quantifying the relative importa nce of ionic stress, compared to the addition of sulfate, on the production of CO 2 and CH 4 A batch incubation study measured potential anaerobic respiration and methanogenesis over time in a freshwater wetland soil exposed to varying concentrations (3.5, 14, and 35 ppt) of seawater or salt (NaCl) solutions. Seawater (sulfate) addition induced a short term (2 week) stimulation of CO 2 production (20 32% greater than the freshwater control) and a continuous suppression of CH 4 production (up to 94% less than freshwater). Ion ic stress (NaCl) did not reduce CO 2 production, but did decrease CH 4 production for 2 weeks in both the 14 and 35 ppt NaCl treatments. Results indicate microbial populations rebound quickly from ionic stress. The intrusion of dilute se awater (3.5 ppt) to freshwater wetlands can accelerate organic C mineralization through the short term increase in sulfate induced respiration without inhibiting methanogenesis. Overall, organic C mineralization rate was 17% higher under 3.5 ppt seawater than the freshwater control.
176 Objective 3: The I mportance of T ime S cale The temporary nature of the microbial response in C hapter 3 suggest ed of seawater may have a greater influence on the rate of C cycling in freshwater wetlands than gradual S LR Therefore, C hapter 4 focused on simulating pulses of higher (or lower) salinity water in coastal wetlands, such as would typically occur during storm surges, extreme tides, heavy rains, or urban run off. This study quantified the impact of salinity pu lsing events on soil organic C ( SOC ) loss (CO 2 flux at high and low tide, CH 4 flux, and DOC release) in three intertid al wetlands in Jacksonville, FL ( USA ) Twelve intact soil cores from a freshwater tidal, brackish, and salt marsh were exposed to simulat ed tides and 3 salinity pulsing events during a 53 day laboratory experiment. Results indicated the freshwater marsh was highly sensitive to brackish (13 ppt) pulses, causing a 112 % increase in LT respiration from accelerated sulfate reduction and N cycl ing, a decrease in both CH 4 flux and DOC export, and a significant change in the partitioning of SOC loss through the 3 biogeochemical pathways studied. Low tide soil respiration rate and DOC export both increased significantly in the salt marsh pulsed wit h fresh (0.5 ppt) water suggesting urban run to keep pace with SLR Overall, this study indicated pulses of different salinities had the greatest impact in the tidal freshwater marsh ( TFM ) and salt marsh, and the e ffects on following the recession of the surface water salinity pulse. At the opposite end of the time spectrum, a more measured increase in salinity was simulated in C h apter 5. In the tidal mesocosm study, a 10 15 ppt increase in salinity (from 15 20 to 30 35 ppt) was gradually introduced over a 1 week period, and then the SOC cycling response was monitored for an additional 6 weeks. Based on
177 previous findings that th e major shifts in biogeochemical pathways occur between 0 ppt and 14 ppt, and that the microbial population can adapt quickly to changes in salinity (C hapters 2 4), this increase was not hypothesized to cause a significant change in the rate of SOC loss. However, a significant and persistent increase in CO 2 flux was observed in Chapter 5 as a result of elevated salinity, which was particularly evident during nighttime CO 2 flux sampling. There may be intrinsic characteristics of the mangrove peat soils u sed in C hapter 5 that caused the soil to be stimulated by the sulfate addition for the long term. For example, high microbial activity caused by high temperatures, high DOC, and consistent flushing of the soil by tides could have created a circumstance wh ere electron acceptor availability was the limiting factor for microbes. In which case, increasing sulfate availability could persistently raise the baseline CO 2 flux rate. In contrast, the microcosm soil slurries (C hapter 3) experienced a confounding li mitation in labile C availability, preventing a long term increase in the rate of soil respiration. Objective 4: Tidal C ycle E ffects Despite extensive p r e vious work on the effects of water table on soil respiration in inland wetlands, there had been almos t no investigation of how tidal cycles impact CO 2 flux in coastal wetlands. In inland wetlands, a fairly straightforward relationship exists between increasing inundation, a decrease in soil respiration rate, and an increase in m ethanogenesis ( Wright and Reddy 2001; DeBusk and Reddy 1998; Clymo 1983) In coastal wetlands, tidal inundation also involves an influx of sulfate, which has been shown to poise the redox potential of wetland soils at an Eh that promotes high rates of sulfate reduction (Mendelssoh n et al. 1999). Additionally, significant variations in inorganic sediment deposition along coastal wetland gradients can result in large
178 differences in soil properties, which also influence SOC loss (Craft 2007; Nyman et al. 1990). The effect of tidal c ycle was investigated in both C hapters 4 and 5. On average, low tide ( LT ) CO 2 flux was higher than high tide ( HT ) CO 2 flux, but the magnitude of the difference between LT and HT varied significantly depending on the coastal wetland type (Figure 6 1). For the TFM, brackish marsh, salt marsh, and mangrove, the difference in CO 2 flux between HT to LT was 53, 230, 310, and 83 % r espectively. Soil properties (organic matter content, total C, bulk density) did not explain these variation s in tidal effect, but significant differences in soil drainage rate s were observed between wetland types This suggests soil hydraulic conductivity (based on the size and distribution of pore spaces) may drive the relationship between CO 2 flux and tidal cycle as much as bio geo chemica l processes Objective 5: Inundation and Carbon C ycling While objective 3 addressed differences in C loss between HT and LT, objective 4 ask ed how reduced soil exposure during LT (an overall increase in inundation) affected SOC loss rate. R esult s of the numeric model developed in C hapter 4, more saline wetlands will store more SOC as inundation increases, while TFMs looses more SOC. TFMs have high rates of SOC loss via HT CO 2 flux, CH 4 flux, and DOC export; under longer durations of HT, this can lead to an overall increase in the rate of SOC loss. In comparison, brackish and salt marshes demonstrated low rates of HT CO 2 flux, translating into a decrease in the rate of SOC loss as inundation increases. C hapter 5 directly addressed this objecti ve with a water level manipulation that decreased the elevation of the soil surface by 8 cm relative to the water table This increase in inundation caused a 35 37 % decrease in the average CO 2 flux rate, compared to the control water level. Methane fl ux also increased significantly
179 compared to the control, but remained low, while DOC concentration increased by approximately 31 % Objective 6: Interaction of S alinity and I nundation Based on the geomorphological characteristics of a coastal wetland, ei ther salinity or inundation may exert a stronger influence on the biogeochemical response to SLR. However, other wetlands will be influenced by simultaneous increases in salinity and inundation. C hapters 3 and 5 clearly demonstrated an increase in the ra te of SOC loss with increasing salinity over both short and long time scales. In C hapter 5, an increase in salinity from 15 20 ppt to 30 35 ppt caused a 54 % increase in the rate of CO 2 flux. In the same system, increased inundation caused a 14 % decrease in the rate of CO 2 flux. Interestingly, when these two drivers were combined, rather than the increase and decrease in rates can celing each other out, there was a synergistic decline of 19 % This data and pervious research suggests the synergistic declin e may be caused by the deleterious by products of increased sulfate reduction (sulfide) accumulating in the soil as salinity increase s. This, in combination with the low redox potential and limited water exchange caused by the increase in inundation, prev ent ed these compounds from being flushed from the soil leading to the overall lowest rate of soil respiration in the inundated elevated salinity treatment Objective 7: Importance of E nvironmental V ariables Coastal wetlands exist in a complex environmen t with each individual ecosystem having unique biological, chemical, and geological characteristics tha t play a role in determining it s response to SLR. Most of this dissertation has investigated SLR in a one in which clim ate drivers, megafauna, and especially C inputs (vegetation and sediment deposition) have been excluded. Providing these
180 artificial boundaries to the soil system allowed for the isolation of individual processes, but also reduced the applicability of the results to natural systems. Still, the inclusion of natural variables increased with each successive research study and did allow for some general conclusions regarding important co variants in coastal wetlands. The brackish marsh soil had a higher bulk density and greater inorganic mat erial >0.025 mm, as well as lower % organic matter, total C, and total N than the TFM, salt marsh, and mangrove soils (C hapters 4 and 5). Surface water DOC was inversely correlated with salinity ( C hapters 4 and 5) and was p ositively correlated with HT soil respiration ( C hapter 4). The salt marsh soil had higher labile C availability and MBC than the TFM and brackish marsh soils ( C hapter 4). Surface water soluble reactive P was inversely correlated with salinity and positivel y correlated with HT soil respiration and DOC release ( C hapter 4). Water temperature caused a positive non linear response in CO 2 flux rate ( C hapter 5). Bioturbation by fiddler crabs appeared to be driven partly by soil inundation level and may increase so il porewater DOC concentration ( C hapter 5). Synthesis and F uture R esearch This body of research answers the questions and objectives presented in the introduction, as outlined above. Carbon dioxide flux was identified as the major pathway of SOC loss in a ll the studies, representing over 90 % of total SOC lost through the measured pathways in each coastal wetland type. Using the conceptual diagram presented in C hapter 1, the response of CO 2 flux to each of the factors and processes of interest in this diss ertation is presented in Figure 6 2. This research also fills many of the research gaps identified in the literature review ( C hapter 2). For example, this work found the impacts of salinity can be observed over both short and long time scales. Short term impacts are abiotic, such as ion exchange
181 on the soil surface ( C hapter 4), and biotic, such as quick spikes in sulfate reduction or reduced microbial respiration due to osmotic stress ( C hapter 3). Whether a short term response persists over time depe nds on the availability of other potentially limiting factors. For example, in a system with a limited supply of labile C (a condition artificially created in C hapter 3), excess SO 4 2 becomes inconsequential to soil microbes over time. Whereas, systems n ot limited by substrate availability or microbial population, an increase in SO 4 2 availability can create a long term increase in the rate of CO 2 flux ( C hapter 5). Therefore, the most appropriate question is not what effect salinity exerts on the system, but rather: (1) i s the system currently limited by electron acceptor availability and (2) h ow much of an increase in biogeochemical cycling can result from the alleviation of electron pressure before another factor (electron donor availability, enzyme sy nthesis, nutrient availability, etc.) becomes limiting? It is the availability of other critical components of the biogeochemical C cycle that ultimately dictates the duration of the salinity effect. The mechanism for the suppression of N cycling is s till unknown, but the results presented herein provide ample new evidence for a significant interaction between N and C cycling as salinity and inundation increase in coastal wetland soils. An accumulation of NH 4 + in the soil porewater can be initiated by cation exchange, organic matter mineralization, and increased inundation. Such an increase in NH 4 + was observed following pulses of higher salinity water in TFM soil ( C hapter 4) and with increased inundation in mangrove soil ( C hapter 5). This increase in NH 4 + availability is an indicator of a bottleneck in the process of nitrification which was also shown to be alleviated by increasing the soil redox potential with a LT event ( C hapter 4). Any
182 disruption of nitrification can have important ecological co nsequences for coastal system because NH 4 + availability is positively correlated with coastal eutrophication and algal blooms (Howarth 1988). Whether the ultimate cause of the NH 4 + accumulation was related to physical conditions (redox potential) or biol ogical toxicity (high sulfide and/or chloride concentrations inhibiting nitrifiers) remains an unanswered question that deserves additional attention. The scaling up process initiated with this dissertation needs to continue so that some of the important processes identified herein can be used to assist in predicting landscape scale changes in coastal wetland C storage as sea level rises. Some important inroads to understanding the fate of coastal wetlands were made in C hapter 4. Coastal wetlands adjace nt to urban development are already prohibited from migrating landward in response to SLR and must rely exclusively on vertical accretion to combat submergence. Pulses of fresh surface water (such as urban stormwater discharge) can significantly accelerat e SOC loss from salt marsh soils, further increasing the need for C inputs to compensate for both SLR and increased soil mineralization in coastal wetlands adjacent to human development. Additionally, C hapter 5 demonstrated how increased inundation can im pact soil structure through a significant decline in soil bulk density after only 10 weeks of water level manipulation. Although the mechanism for the loss of soil structure needs to be confirmed, it is plausible the decrease in bulk density is a precurso r to peat collapse, which often leads to wetland submergence. I believe the logical next steps for biogeochemical research in coastal wetlands exposed to SLR is a focus on three main research avenues: (1) coupled
183 biogeochemical cycles affected by sulfate and inundation (especially N and Fe dynamics), and (2) plant soil interactions influencing biogeochemical cycling (such as substrate availability, rhizosphere microzone dynamics, and nutrient and gas translocation in and out of the soil), and (3) disentang ling abiotic soil responses to salinity and inundation (leaching, solubilization, dispersion) from biotic responses (mineralization). Ecological systems are both complex and individually unique. Understanding micro scale processes, mechanisms, and driver s is critical to being able to make accurate predictions about how other systems that have not been directly studied may respond to SLR. Continuing to research the fundamental processes underlying the response of coastal wetland C to SLR in a deliberate, sequential manner of increasing complexity is the most effective way to accrue knowledge that will lead to robust landscape scale predictions regarding the fate of coastal wetlands as a C sink /source during SLR.
184 Figure 6 1 Tidal effects by wetland ty pe. Rate of CO 2 flux during high and low tide (all 20 cm tidal range) for each of the 4 wetland types studied.
185 Figure 6 2 Conceptual diagram with comprehensive soil respiration results.
186 LIST OF REFERENCES Abril G, Iversen N (2002) Methane dynamics in a shallow non tidal estuary (Randers Fjord, Denmark). Marine Ecology Progress Series 230: 171 181 Adam P (2002) Saltmarshes in a time of change. Environmental Conservation 29(1): 39 61 Alef K (1995) Dehydrogenase activity. In: K Alef and P Nannipieri ( eds) Methods in Applied Soil Microbiology and Biochemistry. Academic Press, London, pp 228 230 Alongi DM (2008) Mangrove forests: Resilience, protection from tsunamis, and responses to global climate change. Estuarine Coastal and Shelf Science 76(1): 1 13 Armentano TV, Menges ES (1986) Patterns of change in the carbon balance of organic soil wetlands of the temperate zone. Journal of Ecology 74(3): 755 774 Baldwin DS, Rees GN, Mitchell AM, Watson G, Williams J (2006) The short term effects of salinization on anaerobic nutrient cycling and microbial community structure in sediment from a freshwater wetland. Wetlands 26(2): 455 464 Barbier EB, Hacker SD, Kennedy C, Koch EW, Stier AC, Silliman BR (2011) The value of estuarine and coastal ecosystem services. Ec ological Monographs 81(2): 169 193 Bartlett KB, Bartlett DS, Harriss RC, Sebacher DI (1987) Methane emissions along a salt marsh salinity gradient. Biogeochemistry 4(3): 183 202 Batzer DP, Sharitz RR (Eds) (2006) Ecology of Freshwater and Estuarine Wetland s. University of California Press, Berkeley Benner R, Moran MA, Hodson RE (1986) Biogeochemical cycling of lignocellulosic carbon in marine and freshwater ecosystems: relative contribution of procaryotes and eucaryotes. Limnology and Oceanography 31(1): 89 100 Biswas H, Mukhopadhyay SK, Sen S, Jana TK (2007) Spatial and temporal patterns of methane dynamics in the tropical mangrove dominated estuary, NE coast of Bay of Bengal, India. Journal of Marine Systems 68(1 2): 55 64 Bloetscher F, Heimlich B, Meeroff DE (2011) Development of an adaptation toolbox to protect southeast Florida water supplies from climate change. Environmental Reviews 19: 397 417 Blomqvist S, Gunnars A, Elmgren R (2004) Why the limiting nutrient differs between temperate coastal seas and freshwater lakes: A matter of salt. Limnology and Oceanography 49(6): 2236 2241
187 Boelter DH (1965) Hydraulic conductivity of peats. Soil Sci 100((4)): 227 231 Bouillon S, Borges AV, Castaneda Moya E, Diele K, Dittmar T, Duke NC, Kristensen E, Lee SY, March and C, Middelburg JJ, Rivera Monroy VH, Smith TJ, Twilley RR (2008a) Mangrove production and carbon sinks: A revision of global budget estimates. Global Biogeochemical Cycles 22(2): 12 Bouillon S, Connolly RM, Lee SY (2008b) Organic matter exchange and cyc ling in mangrove ecosystems: Recent insights from stable isotope studies. Journal of Sea Research 59(1 2): 44 58 Bouillon S, Middelburg JJ, Dehairs F, Borges AV, Abril G, Flindt MR, Ulomi S, Kristensen E (2007) Importance of intertidal sediment processes a nd porewater exchange on the water column biogeochemistry in a pristine mangrove creek (Ras Dege, Tanzania). Biogeosciences 4(3): 311 322 Boyer JN (2006) Shifting N and P limitation along a north south gradient of mangrove estuaries in south Florida. Hydro biologia 569: 167 177 Bradley PM, Morris JT (1991) The influence of salinity on the kinetics of NH 4 + uptake in Spartina alterniflora. Oecologia 85(3): 375 380 Brady NC, Weil RR (2004) Elements of the Nature and Properties of Soils. Pearson Prentice Hall, N ew Jersey Bridgham SD, Megonigal JP, Keller JK, Bliss NB, Trettin C (2006) The carbon balance of North American wetlands. Wetlands 26(4): 889 916 Cao YP, Green PG, Holden PA (2008) Microbial Community Composition and Denitrifying Enzyme Activities in Salt Marsh Sediments. Applied and Environmental Microbiology 74(24): 7585 7595 Capone DG, Kiene RP (1988) Comparison of microbial dynamics in marine and fresh water sediments: contrasts in anaerobic carbon catabolism. Limnology and Oceanography 33(4): 725 749 Cappenberg (1974) Interrelaions between sulfate reducing and methane producing bacteria in bottom deposits of a freshwater lake. I. Field observations. Antonie Van Leeuwenhoek Journal of Microbiology 40(2): 285 295 Castaneda E (2010) Landscape patterns of community structure, biomass and net primary productivity of mangrove forests in the Florida Coastal Everglades as a function of resources, regulators, hydroperiod, and hurricane disturbance. In: Oceanography and Coastal Sciences. Louisiana State Universit y. p 171 Castaneda Moya E, Twilley RR, Rivera Monroy VH, Marx BD, Coronado Molina C, Ewe SML (2011) Patterns of Root Dynamics in Mangrove Forests Along Environmental Gradients in the Florida Coastal Everglades, USA. Ecosystems 14(7): 1178 1195
188 Chambers LG Reddy KR, Osborne TZ (2011) Short Term Response of Carbon Cycling to Salinity Pulses in a Freshwater Wetland. Soil Science Society of America Journal 75(5): 2000 2007 Chen RH, Twilley RR (1999) Patterns of mangrove forest structure and soil nutrient dyna mics along the Shark River estuary, Florida. Estuaries 22(4): 955 970 Chmura GL, Anisfeld SC, Cahoon DR, Lynch JC (2003) Global carbon sequestration in tidal, saline wetland soils. Global Biogeochemical Cycles 17(4): 12 Choi YH, Wang Y (2004) Dynamics of c arbon sequestration in a coastal wetland using radiocarbon measurements. Global Biogeochemical Cycles 18(4): 12 Christensen TR, Ekberg A, Strom L, Mastepanov M, Panikov N, Mats O, Svensson BH, Nykanen H, Martikainen PJ, Oskarsson H (2003) Factors controlli ng large scale variations in methane emissions from wetlands. Geophysical Research Letters 30(7): 4 Christian RR, Bryant WL, Brinson MM (1990) Juncus roemerianus production and decomposition along gradients of salinity and hydroperiod. Marine Ecology Progr ess Series 68(1 2): 137 145 Clymo RS (1983) Peat. Mines: swamp, bog, fen and moor. General Studies. Vol.A.: 159 224 Cordova Kreylos AL, Cao YP, Green PG, Hwang HM, Kuivila KM, LaMontagne MG, Van De Werfhorst LC, Holden PA, Scow KM (2006) Diversity, composi tion, and geographical distribution of microbial communities in california salt marsh Sediments. Applied and Environmental Microbiology 72(5): 3357 3366 and Hesieh, Y P Ecology and management of tidal marshes, CRC Press, Boca Raton, pp 61 69 Craft C (2007) Freshwater input structures soil properties, vertical accretion, and nutrient accumulation of Georgia and U.S. tidal marshes. Limnology and Oceanography 52(3): 1220 1230 Craft C, Broome S, Campbell C (2002) Fifteen years of vegetation and soil development after brackish water marsh creation. Restoration Ecology 10(2): 248 258 Craft C, Clough J, Ehman J, Joye S, Park R, Pennings S, Guo HY, Machmuller M (2009) Forecasti ng the effects of accelerated sea level rise on tidal marsh ecosystem services. Frontiers in Ecology and the Environment 7(2): 73 78 D'Angelo EM, Reddy KR (1999) Regulators of heterotrophic microbial potentials in wetland soils. Soil Biology & Biochemistry 31(6): 815 830
189 Darby FA, Turner RE (2008) Below and aboveground Spartina alterniflora production in a Louisiana salt marsh. Estuaries and Coasts 31(1): 223 231 Dasgupta S, Laplante B, Murray S, Wheeler D (2011) Exposure of developing countries to sea lev el rise and storm surges. Climatic Change 106(4): 567 579 Davis CS (2002) Statistical methods for the analysis of repeated measurements. Springer, New York Davis SM, Childers DL, Lorenz JJ, Wanless HR, Hopkins TE (2005) A conceptual model of ecological in teractions in the mangrove estuaries of the Florida Everglades. Wetlands 25(4): 832 842 Day JW, Christian RR, Boesch DM, Yanez Arancibia A, Morris J, Twilley RR, Naylor L, Schaffner L, Stevenson C (2008) Consequences of climate change on the ecogeomorpholo gy of coastal wetlands. Estuaries and Coasts 31(3): 477 491 Day JW, Shaffer GP, Britsch LD, Reed DJ, Hawes SR, Cahoon D (2000) Pattern and process of land loss in the Mississippi Delta: A spatial and temporal analysis of wetland habitat change. Estuaries 2 3(4): 425 438 DeBusk WF, Reddy KR (1998) Turnover of detrital organic carbon in a nutrient impacted Everglades marsh. Soil Science Society of America Journal 62(5): 1460 1468 Delaune RD, Nyman JA, Patrick WH (1994) Peat collapse, pending and wetland loss i n a rapidly submerging coastal marsh. Journal of Coastal Research 10(4): 1021 1030 Delaune RD, Smith CJ, Patrick WH (1983) Methane release from Gulf coast wetlands. Tellus Series B Chemical and Physical Meteorology 35(1): 8 15 DeLaune RD, White JR (2012) W ill coastal wetlands continue to sequester carbon in response to an increase in global sea level?: a case study of the rapidly subsiding Mississippi river deltaic plain. Climatic Change 110(1 2): 297 314 Dodla SK, Wang JJ, Delaune RD, Breitenbeck G (2009) Carbon gas production under different electron acceptors in a freshwater marsh soil. Chemosphere 76(4): 517 522 Donato DC, Kauffman JB, Murdiyarso D, Kurnianto S, Stidham M, Kanninen M (2011) Mangroves among the most carbon rich forests in the tropics. Nat ure Geoscience 4(5): 293 297 Donnelly JP, Bertness MD (2001) Rapid shoreward encroachment of salt marsh cordgrass in response to accelerated sea level rise. Proceedings of the National Academy of Sciences of the United States of America 98(25): 14218 14223
190 Edmonds JW, Weston NB, Joye SB, Mou XZ, Moran MA (2009) Microbial Community Response to Seawater Amendment in Low Salinity Tidal Sediments. Microbial Ecology 58(3): 558 568 Enriquez S, Duarte CM, Sandjensen K (1993) Patterns in decomposition rates among p hotosynthetic organisms the importance of detritus C N P content. Oecologia 94(4): 457 471 Ewe SML, Gaiser EE, Childers DL, Iwaniec D, Rivera Monroy VH, Twilley RR (2006) Spatial and temporal patterns of aboveground net primary productivity (ANPP) along tw o freshwater estuarine transects in the Florida Coastal Everglades. Hydrobiologia 569: 459 474 Fagherazzi S, Kirwan ML, Mudd SM, Guntenspergen GR, Temmerman S, D'Alpaos A, van de Koppel J, Rybczyk JM, Reyes E, Craft C, Clough J (2012) Numerical models of s alt marsh evolution: ecological, geomorphic, and climate factors. Reviews of Geophysics 50: 28 Florida Department of Environmental Protection (2008) Salt Marshes. http://www.dep.s tate.fl.us/coastal/habitats/saltmarshes.htm Accessed 8 February 2011 Frankenberger WT, Dick WA (1983) Relationships between enzyme activities and microbial growth and activity indexes in soils. Soil Science Society of America Journal 47(5): 945 951 Fran kenberger WT, Bingham FT (1982) Influence of salinity on soil enzyme activities. Soil Science Society of America Journal 46(6): 1173 1177 Franzen LG (2006) Increased decomposition of subsurface peat in Swedish raised bogs: are temperate peatlands still net sinks of carbon? Mires and Peat 1: art.3 Frazier TG, Wood N, Yarnal B, Bauer DH (2010) Influence of potential sea level rise on societal vulnerability to hurricane storm surge hazards, Sarasota County, Florida. Applied Geography 30(4): 490 505 Freeman C, Liska G, Ostle NJ, Lock MA, Hughes S, Reynolds B, Hudson J (1997) Enzymes and biogeochemical cycling in wetlands during a simulated drought. Biogeochemistry 39(2): 177 187 Freeman C, Lock MA, Reynolds B (1993) Fluxes of CO 2 CH 4 and N 2 O from a Welsh peatl and following simulation of water table draw down potential feedback to climatic change. Biogeochemistry 19(1): 51 60 Gaiser EE, Zafiris A, Ruiz PL, Tobias FAC, Ross MS (2006) Tracking rates of ecotone migration due to salt water encroachment using fossil mollusks in coastal South Florida. Hydrobiologia 569: 237 257
191 Gardner LR (2005) Role of geomorphic and hydraulic parameters in governing pore water seepage from salt marsh sediments. Water Resources Research 41(7): 11 Gedan KB, Silliman BR, Bertness MD (2 009) Centuries of Human Driven Change in Salt Marsh Ecosystems. Annual Review of Marine Science 1: 117 141 Gennari M, Abbate C, La Porta V, Baglieri A, Cignetti A (2007) Microbial response to Na2SO4 additions in a volcanic soil. Arid Land Research and Mana gement 21(3): 211 227 Gonzalez Trilla G, Marco Sd, Marcovecchio J, Vicari R, Kandus P (2010) Net primary productivity of Spartina densiflora Brong in an SW Atlantic coastal salt marsh. Estuaries and Coasts 33(4): 953 962 Good RE, Good NF, Frasco BR (1982) A review of primary production and decomposition dynamics of the belowground marsh component. Kennedy, V. S. (Ed.). Estuarine Comparisons; Proceedings of the 6th Biennial International Estuarine Research Conference, Gleneden Beach, Oreg., USA, Nov. 1 6, 19 81. Xviii+709p. Academic Press, Inc.: New York, N.Y., USA; London. Illus: P139 158 Gribsholt B, Kristensen E (2003) Benthic metabolism and sulfur cycling along an inundation gradient in a tidal Spartina anglica salt marsh. Limnology and Oceanography 48(6): 2151 2162 Hackney CT, Cleary WJ (1987) Salt marsh loss in southeastern North Carolina lagoons importance of sea level rise and inlet dredging. Journal of Coastal Research 3(1): 93 97 Hackney CT, Padgett DE, Posey MH (2000) Fugal and bacterial contributio ns to the decomposition of Cladium and Typha leaves in nutrient enriched and nutrient poor areas of the Everglades with a note on ergosterol concentrations in Everglades soils. Mycol. Res. 104: 666 670 Hale RL, Groffman PM (2006) Chloride effects on nitrog en dynamics in forested and suburban stream debris dams. Journal of Environmental Quality 35(6): 2425 2432 Harris GP (1999) Comparison of the biogeochemistry of lakes and estuaries: ecosystem processes, functional groups, hysteresis effects and interaction s between macro and microbiology. Marine and Freshwater Research 50(8): 791 811 Hatton RS, Delaune RD, Patrick WH (1983) Sedimentation, accretion, and subsidence in marshes of Barataria Basin, Louisiana. Limnology and Oceanography 28(3): 494 502 Henman J, Poulter B (2008) Inundation of freshwater peatlands by sea level rise: Uncertainty and potential carbon cycle feedbacks. Journal of Geophysical Research Biogeosciences 113(G1): 11
192 Howarth RW (1984) The ecological significance of sulfur in the energy dynam ics of salt marsh and coastal marine sediments. Biogeochemistry 1(1): 5 27 Howarth RW (1988) Nutrient limitation of net primary productivity in marine ecosystems. Annual Review of Ecology and Systematics 19: 89 110 Hughes ZJ, FitzGerald DM, Wilson CA, Penn ings SC, Wieski K, Mahadevan A (2009) Rapid headward erosion of marsh creeks in response to relative sea level rise. Geophysical Research Letters 36(3): L03602 Hussein AH (2009) Modeling of Sea Level Rise and Deforestation in Submerging Coastal Ultisols of Chesapeake Bay. Soil Science Society of America Journal 73(1): 185 196 Ikenaga M, Guevara R, Dean AL, Pisani C, Boyer JN (2010) Changes in Community Structure of Sediment Bacteria Along the Florida Coastal Everglades Marsh Mangrove Seagrass Salinity Gradi ent. Microbial Ecology 59(2): 284 295 IPCC (2007) Climate Change 2007: A Synthesis Report. Valencia, Spain. p 22 Jackson CR, Vallaire SC (2009) Effects of salinity and nutrients on microbial assemblages in Louisiana wetland sediments. Wetlands 29(1): 277 2 87 Jakobsen P, Patrick WH, Williams BG (1981) Sulfide and methane formation in soils and sediments. Soil Science 132(4): 279 287 Joye SB, Hollibaugh JT (1995) Influence of sulfide inhibition of nitrification on nitrogen regeneration in sediments. Science ( Washington DC) 270(5236): 623 627 Kathilankal JC, Mozdzer TJ, Fuentes JD, D'Odorico P, McGlathery KJ, Zieman JC (2008) Tidal influences on carbon assimilation by a salt marsh. Environmental Research Letters 3(4): 6 Kester DR, Duedall IW, Connors DN, Pytkow ic.Rm (1967) Preparation of artificial seawater. Limnology and Oceanography 12(1): 176 179 King GM, Klug MJ, Wiegert RG, Chalmers AG (1982) Relation of soil water movement and sulfide concentration to Spartina alterniflora production in a Georgia salt mar sh. Science 218(4567): 61 63 King GM, Wiebe WJ (1980) Regulation of sulfate concentrations and methanogenesis in salt marsh soils. Estuarine and Coastal Marine Science 10(2): 215 223 Kirwan M, Temmerman S (2009) Coastal marsh response to historical and fut ure sea level acceleration. Quaternary Science Reviews 28(17 18): 1801 1808
193 Kirwan ML, Murray AB, Boyd WS (2008) Temporary vegetation disturbance as an explanation for permanent loss of tidal wetlands. Geophysical Research Letters 35(5): 5 Koch MS, Mendels sohn IA, McKee KL (1990) Mechanism for the hydrogen sulfide induced growth limitation in wetland macrophytes. Limnology and Oceanography 35(2): 399 408 Kool DM, Buurman P, Hoekman DH (2006) Oxidation and compaction of a collapsed peat dome in Central Kalim antan. Geoderma 137(1 2): 217 225 Kostka JE, Roychoudhury A, Van Cappellen P (2002) Rates and controls of anaerobic microbial respiration across spatial and temporal gradients in saltmarsh sediments. Biogeochemistry 60(1): 49 76 Laanbroek HJ (2010) Methan e emission from natural wetlands: interplay between emergent macrophytes and soil microbial processes. A mini review. Annals of Botany 105(1): 141 153 Langenheder S, Kisand V, Wikner J, Tranvik LJ (2003) Salinity as a structuring factor for the compositio n and performance of bacterioplankton degrading riverine DOC. Fems Microbiology Ecology 45(2): 189 202 Langley JA, McKee KL, Cahoon DR, Cherry JA, Megonigal JP (2009) Elevated CO2 stimulates marsh elevation gain, counterbalancing sea level rise. Proceeding s of the National Academy of Sciences of the United States of America 106(15): 6182 6186 Lee RY, Porubsky WP, Feller IC, McKee KL, Joye SB (2008) Porewater biogeochemistry and soil metabolism in dwarf red mangrove habitats (Twin Cays, Belize). Biogeochemis try 87(2): 181 198 Li CY, White JR, Chen CS, Lin HC, Weeks E, Galvan K, Bargu S (2011) Summertime tidal flushing of Barataria Bay: Transports of water and suspended sediments. Journal of Geophysical Research Oceans 116: 15 Lovley DR, Phillips EJP (1986) Or ganic matter mineralization with reduction of ferric iron in anaerobic sediments. Applied and Environmental Microbiology 51(4): 683 689 Lugo A Brown S Brinson M (1989) Concepts in wetland ecology In: Lugo A Brown S & Brinson M (eds) Ecosystems of the W orld 15 Elsevier, Amsterdam. p 53 85 Magenheimer JF, Moore TR, Chmura GL, Daoust RJ (1996) Methane and carbon dioxide flux from a macrotidal salt marsh, Bay of Fundy, New Brunswick. Estuaries 19(1): 139 145
194 Makoi J, Ndakidemi PA (2008) Selected soil enzym es: Examples of their potential roles in the ecosystem. African Journal of Biotechnology 7(3): 181 191 Malik KA, Bhatti NA, Kauser F (1979) Effect of soil salinity on decomposition and humification of organic matter by some celluloltic fungi. Mycologia 71: 811 820 Marx MC, Wood M, Jarvis SC (2001) A microplate fluorimetric assay for the study of enzyme diversity in soils. Soil Biology & Biochemistry 33(12 13): 1633 1640 McGranahan G, Balk D, Anderson B (2007) The rising tide: assessing the risks of climate change and human settlements in low elevation coastal zones. Environment and Urbanization 19(1): 17 37 McLeod E, Chmura GL, Bouillon S, Salm R, Bjork M, Duarte CM, Lovelock CE, Schlesinger WH, Silliman BR (2011) A blueprint for blue carbon: toward an impro ved understanding of the role of vegetated coastal habitats in sequestering CO(2). Frontiers in Ecology and the Environment 9(10): 552 560 Mendelssohn I, Morris J (2000) Eco physiological constraints on the primary productivity of Spartina alterniflora In : Concepts and Controversies of Tidal Marsh Ecology. Elsevier, Amsterdam Mendelssohn IA, Sorrell BK, Brix H, Schierup HH, Lorenzen B, Maltby E (1999) Controls on soil cellulose decomposition along a salinity gradient in a Phragmites australis wetland in D enmark. Aquatic Botany 64(3 4): 381 398 Michener WK, Blood ER, Bildstein KL, Brinson MM, Gardner LR (1997) Climate change, hurricanes and tropical storms, and rising sea level in coastal wetlands. Ecological Applications 7(3): 770 801 Miller WD, Neubauer S C, Anderson IC (2001) Effects of sea level induced disturbances on high salt marsh metabolism. Estuaries 24(3): 357 367 Miller WD, Neubauer SC, Anderson IC (2001) Effects of sea level induced disturbances on high salt marsh metabolism. Estuaries 24(3): 357 367 Moorhead KK, Brinson MM (1995) Response of wetlands to rising sea level in the lower coastal plain of North Carolina. Ecological Applications 5(1): 261 271 Morris JT, Sundareshwar PV, Nietch CT, Kjerfve B, Cahoon DR (2002) Responses of coastal wetland s to rising sea level. Ecology 83(10): 2869 2877 Mousavi ME, Irish JL, Frey AE, Olivera F, Edge BL (2011) Global warming and hurricanes: the potential impact of hurricane intensification and sea level rise on coastal flooding. Climatic Change 104(3 4): 575 597
195 Mudd SM, Howell SM, Morris JT (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 82(3): 377 389 Muhammad S, M uller T, Joergensen RG (2006) Decomposition of pea and maize straw in Pakistani soils along a gradient in salinity. Biology and Fertility of Soils 43(1): 93 101 Mulholland PJ, Best GR, Coutant CC, Hornberger GM, Meyer JL, Robinson PJ, Stenberg JR, Turner R E, VeraHerrera F, Wetzel RG (1997) Effects of climate change on freshwater ecosystems of the south eastern United States and the Gulf Coast of Mexico. Hydrological Processes 11(8): 949 970 Nelson DW, Sommers LE (1996) Total carbon, organic carbon, and orga nic matter. Soil Science Society of America Inc., Madison Neubauer SC (2008) Contributions of mineral and organic components to tidal freshwater marsh accretion. Estuarine Coastal and Shelf Science 78(1): 78 88 Neubauer SC, Givler K, Valentine SK, Megoniga l JP (2005) Seasonal patterns and plant mediated controls of subsurface wetland biogeochemistry. Ecology 86(12): 3334 3344 Neubauer SC, Miller WD, Anderson IC (2000) Carbon cycling in a tidal freshwater marsh ecosystem: a carbon gas flux study. Marine Ecol ogy Progress Series 199: 13 30 Nicholls RJ, Cazenave A (2010) Sea Level Rise and Its Impact on Coastal Zones. Science 328(5985): 1517 1520 Nicholls RJ, Hoozemans FMJ, Marchand M (1999) Increasing flood risk and wetland losses due to global sea level rise: regional and global analyses. Global Environmental Change Human and Policy Dimensions 9: S69 S87 Nielsen DL, Brock MA, Rees GN, Baldwin DS Effects of increasing salinity on freshwater ecosystems in Australia. In. C S I R O Publishing. 2003. C S I R O Publi shing, p 655 665 NOAA (2011) National Weather Service Jacksonville, FL Weather Forecast Office. http://www.srh.noaa.gov/jax/ Accessed 2 February 2012 Nyman J, DeLaune R (1999) Four potential impacts of globa l sea level rise on coastal marsh stability. Current Topics in Wetland Biogeochemistry 3: 112 117 Nyman JA, Delaune RD (1991) CO 2 emission and soil Eh response to difference hydrological conditions in fresh, brackish, and saline marsh soils. Limnology an d Oceanography 36(7): 1406 1414
196 Nyman JA, Delaune RD, Patrick WH (1990) Wetland soil formation in the rapidly subsiding Mississippi River delatic plain mineral and organic matter relationships. Estuarine Coastal and Shelf Science 31(1): 57 69 Odum WE (198 8) Comparative ecology of tidal fresh water and salt marshes. Annual Review of Ecology and Systematics 19: 147 176 Oremland RS, Marsh LM, Polcin S (1982) Methane production and simultaneous sulfate reduction in anoxic, salt marsh sediments. Nature 296(5853 ): 143 145 Oremland RS, Polcin S (1982) Methanogenesis and sulfate reduction: competitive and noncompetitive substrates in estuarine sediments. Applied and Environmental Microbiology 44(6): 1270 1276 Oren A (2008) Microbial life at high salt concentrations : phylogenetic and metabolic diversity. Saline Systems 4: 2 Paludan C, Morris JT (1999) Distribution and speciation of phosphorus along a salinity gradient in intertidal marsh sediments. Biogeochemistry 45(2): 197 221 Park J, Obeysekera J, Irizarry M, Barn es J, Trimble P, Park Said W (2011) Storm surge projections and implications for water management in South Florida. Climatic Change 107(1 2): 109 128 Pathak H, Rao DLN (1998) Carbon and nitrogen mineralization from added organic matter in saline and alkali soils. Soil Biology & Biochemistry 30(6): 695 702 Patrick WH, Khalid RA (1974) Phosphate release and sorption by soils and sediments effect of aerobic and anaerobic conditions. Science 186(4158): 53 55 Patrick WHJ, DeLaune RD (1977) Chemical and biologic al redox systems affecting nutrient availability in the coastal wetlands. Geosci. Man. 18: 131 137 Pattnaik P, Mishra SR, Bharati K, Mohanty SR, Sethunathan N, Adhya TK (2000) Influence of salinity on methanogenesis and associated microflora in tropical ri ce soils. Microbiological Research 155(3): 215 220 Penton CR, Newman S (2008) Enzyme based resource allocated decomposition and landscape heterogeneity in the Florida Everglades. Journal of Environmental Quality 37(3): 972 976 Poffenbarger HJ, Needelman BA Megonigal JP (2011) Salinity Influence on Methane Emissions from Tidal Marshes. Wetlands 31(5): 831 842 Portnoy JW, Giblin AE (1997) Biogeochemical effects of seawater restoration to diked salt marshes. Ecological Applications 7(3): 1054 1063
197 Purvaja R, Ramesh R (2001) Natural and anthropogenic methane emission from coastal wetlands of South India. Environmental Management 27(4): 547 557 Quintino V, Sangiorgio F, Ricardo F, Mamede R, Pires A, Freitas R, Rodrigues AM, Basset A (2009) In situ experimental s tudy of reed leaf decomposition along a full salinity gradient. Estuarine Coastal and Shelf Science 85(3): 497 506 Rasul G, Appuhn A, Muller T, Joergensen RG (2006) Salinity induced changes in the microbial use of sugarcane filter cake added to soil. Appli ed Soil Ecology 31(1 2): 1 10 Reddy KR, DeLaune R (2008) Biogeochemistry of Wetlands: Science and Applications. CRC Press, New York Reeburgh WS, Heggie DT (1977) Microbial methane consumption reactions and their effects on methane distributions in freshwat er and marine environments. Limnology and Oceanography 22(1): 1 9 Reed DJ (1995) The response of coastal marshes to sea level rise survival or submergence. Earth Surface Processes and Landforms 20(1): 39 48 Reice SR, Herbst G (1982) The role of salinity i n decomposition of leaves of Phragmites australis in desert streams. Journal of Arid Environments 5(4): 361 368 Rivera Monroy VH, Twilley RR, Davis SE, Childers DL, Simard M, Chambers R, Jaffe R, Boyer JN, Rudnick DT, Zhang K, Castaneda Moya E, Ewe SML, Pr ice RM, Coronado Molina C, Ross M, Smith TJ, Michot B, Meselhe E, Nuttle W, Troxler TG, Noe GB (2011) The Role of the Everglades Mangrove Ecotone Region (EMER) in Regulating Nutrient Cycling and Wetland Productivity in South Florida. Critical Reviews in En vironmental Science and Technology 41: 633 669 Roberts JD, Rowland AP (1998) Cellulose fractionation in decomposition studies using detergent fiber pretreatment methods. Communications in Soil Science and Plant Analysis 29(11 14): 1379 1380 Roman CT, Daibe r FC (1984) Above ground and below ground primary production dynamics of 2 Delaware Bay tidal marshes. Bulletin of the Torrey Botanical Club 111(1): 34 41 Romigh MM, Davis SE, Rivera Monroy VH, Twilley RR (2006) Flux of organic carbon in a riverine mangrov e wetland in the Florida Coastal Everglades. Hydrobiologia 569: 505 516 Roseberg RJ, Christensen NW, Jackson TL (1986) Chloride, soil solution osmotic potential, and soil PH effects on nitrification. Soil Science Society of America Journal 50(4): 941 945
198 R oss MS, Meeder JF, Sah JP, Ruiz PL, Telesnicki GJ (2000) The Southeast Saline Everglades revisited: 50 years of coastal vegetation change. Journal of Vegetation Science 11(1): 101 112 Rysgaard S, Thastum P, Dalsgaard T, Christensen PB, Sloth NP (1999) Effe cts of salinity on NH4+ adsorption capacity, nitrification, and denitrification in Danish estuarine sediments. Estuaries 22(1): 21 30 Sangiorgio F, Basset A, Pinna M, Sabetta L, Abbiati M, Ponti M, Minocci M, Orfanidis S, Nicolaidouc A, Moncheva S, Trayano va A, Georgescu L, Dragan S, Beqiraj S, Koutsoubas D, Evagelopoulos A, Reizopoulou S (2008) Environmental factors affecting Phragmites australis litter decomposition in Mediterranean and Black Sea transitional waters. Aquatic Conservation Marine and Freshw ater Ecosystems 18: S16 S26 Sardinha M, Muller T, Schmeisky H, Joergensen RG (2003) Microbial performance in soils along a salinity gradient under acidic conditions. Applied Soil Ecology 23(3): 237 244 Saviozzi A, Cardelli R, Di Puccio R (2011) Impact of S alinity on Soil Biological Activities: A Laboratory Experiment. Communications in Soil Science and Plant Analysis 42(3): 358 367 Seo DC, DeLaune RD (2010) Effect of redox conditions on bacterial and fungal biomass and carbon dioxide production in Louisiana coastal swamp forest sediment. Science of the Total Environment 408(17): 3623 3631 Seo DC, Yu K, Delaune RD (2008) Influence of salinity level on sediment denitrification in a Louisiana estuary receiving diverted Mississippi River water. Archives of Agron omy and Soil Science 54(3): 249 257 SFWMD (2012) Water Quality and Other Sample Data Access http://www.sfwmd.gov/dbhydroplsql/water_quality_interface.main_page Accessed 1 5 March 20 12 Siddikee MA, Tipayno SC, Kim K, Chung J, Sa T (2011) Influence of Varying Degree of Salinity Sodicity Stress on Enzyme Activities and Bacterial Populations of Coastal Soils of Yellow Sea, South Korea. Journal of Microbiology and Biotechnology 21(4): 341 346 Smith CJ, Delaune RD, Patrick WH (1983) Carbon dioxide emissions and carbon accumulation in coastal wetlands. Estuarine Coastal and Shelf Science 17(1): 21 29 Smith SM (2009) Multi decadal Changes in Salt Marshes of Cape Cod, MA: Photographi c Analyses of Vegetation Loss, Species Shifts, and Geomorphic Change. Northeastern Naturalist 16(2): 183 208
199 Sotomayor D, Corredor JE, Morell JM (1994) Methane flux from mangrove sediments along the southwestern coast of Puerto Rico. Estuaries 17(1B): 140 147 Spackman W, Dolsen C, Riegel W (1966) Phytogenic sediments and sedimentary environments in the Everglades mangrove complex. Part I Evidence of a transgressing sea and its effects on environments of the Shark River area of southwest Florida. Palaeonto graphica B 17: 135 152 Sparling GP, Feltham CW, Reynolds J, West AW, Singleton P (1990) Estimation of soil microbial C by fumigation extraction method use on soils of high organic matter content, and a reassessment of the kec factor. Soil Biology & Bioche mistry 22(3): 301 307 Stagg CL, Mendelssohn IA (2010) Restoring Ecological Function to a Submerged Salt Marsh. Restoration Ecology 18: 10 17 Standard Methods (1997) Standard Methods for the Examination of Water and Wastewater (20th Edition). 4110 Determina tion of Anions by Ion Chromatography with chemical suppression of eluent conductivity Sterner RW, Elser JJ (2002) Ecological stoichiometry: the biology of elements from molecules to the biosphere. Ecological stoichiometry: the biology of elements from mole cules to the biosphere.: i xxi, 1 439 Thalmann A (1968) Zur Methodik der Bestimmung der Dehydrogenaseaktivitt im Boden mittels Triphenyltetrazoliumchlorid (TTC). Landwirtsch Forsch 21:249 259 Thomas GW (1996) Soil pH and soil acidity. In: Sparks DL, Page AL, Helmke PA, Loeppert RH, Soltanpour PN, Tabatabai MA, Johnston CT & Sumner ME (eds) Methods of soil analysis. Part 3 chemical methods. p 475 490 Titus JG, Richman C (2001) Maps of lands vulnerable to sea level rise: modeled elevations along the US Atl antic and Gulf coasts. Climate Research 18(3): 205 228 Tripathi S, Chakraborty A, Chakrabarti K, Bandyopadhyay BK (2007) Enzyme activities and microbial biomass in coastal soils of India. Soil Biology & Biochemistry 39(11): 2840 2848 Tripathi S, Kumari S, Chakraborty A, Gupta A, Chakrabarti K, Bandyapadhyay BK (2006) Microbial biomass and its activities in salt affected coastal soils. Biology and Fertility of Soils 42(3): 273 277 Turner RE (2011) Beneath the Salt Marsh Canopy: Loss of Soil Strength with Inc reasing Nutrient Loads. Estuaries and Coasts 34(5): 1084 1093 Turner RE, Swenson EM, Milan CS, Lee JM, Oswald TA (2004) Below ground biomass in healthy and impaired salt marshes. Ecological Research 19(1): 29 35
200 Twilley RR, Chen RH, Hargis T (1992) Carbon sinks in mangroves and their implications to carbon budget of tropical coastal ecosystems. Water Air and Soil Pollution 64(1 2): 265 288 US Census Bureau (2009) Duval County, Florida: QuickFacts. http://quickfacts.census.gov/qfd/states/12/12031.html. Acce ssed 22 February 2012 USEPA (1993) Methods for the Determination of Inorganic Substances in Environmental Samples. USEPA 600/R 93/100: Method 365.1 Valiela I, Bowen JL, York JK (2001) Mangrove forests: One of the world's threatened major tropical environm ents. Bioscience 51(10): 807 815 Van Ryckegem G, Verbeken A (2005) Fungal diversity and community structure on Phragmites australis (Poaceae) along a salinity gradient in the Scheldt estuary (Belgium). Nova Hedwigia 80(1 2): 173 197 Vance ED, Brookes PC, J enkinson DS (1987) An extraction method for measuring soil microbial biomass C. Soil Biology & Biochemistry 19(6): 703 707 Wang ZP, Zeng D, Patrick WH (1996) Methane emissions from natural wetlands. Environmental Monitoring and Assessment 42(1 2): 143 161 Webster JR, Benfield EF (1986) Vascular plant breakdown in fresh water ecosystems. Annual Review of Ecology and Systematics 17: 567 594 Welch R, Madden M, Doren RF (1999) Mapping the Everglades. Photogrammetric Engineering and Remote Sensing 65(2): 163 170 Welsh DT (2000) Ecological significance of compatible solute accumulation by micro organisms: from single cells to global climate. Fems Microbiology Reviews 24(3): 263 290 Weston NB, Dixon RE, Joye SB (2006b) Ramifications of increased salinity in tidal f reshwater sediments: Geochemistry and microbial pathways of organic matter mineralization. Journal of Geophysical Research Biogeosciences 111(G1): 14 Weston NB, Porubsky WP, Samarkin VA, Erickson M, Macavoy SE, Joye SB (2006a) Porewater stoichiometry of te rminal metabolic products, sulfate, and dissolved organic carbon and nitrogen in estuarine intertidal creek bank sediments. Biogeochemistry 77(3): 375 408 Weston NB, Vile MA, Neubauer SC, Velinsky DJ (2011) Accelerated microbial organic matter mineralizati on following salt water intrusion into tidal freshwater marsh soils. Biogeochemistry 102(1 3): 135 151 Wetzel RG (2001) Protists: Key ecosystem regulators. Bioscience 51(12): 997 997
201 Whelan T (1974) Methane carbon dioxide and dissolved sulfate from interst itial water of coastal sediments. Estuarine and Coastal Marine Science 2(4): 407 416 White JR, Fulweiler RW, Li CY, Bargu S, Walker ND, Twilley RR, Green SE (2009) Mississippi River Flood of 2008: Observations of a Large Freshwater Diversion on Physical, C hemical, and Biological Characteristics of a Shallow Estuarine Lake. Environmental Science & Technology 43(15): 5599 5604 White JR, Reddy KR (2001) Influence of selected inorganic electron acceptors on organic nitrogen mineralization in everglades soils. S oil Science Society of America Journal 65(3): 941 948 Wieski K, Guo HY, Craft CB, Pennings SC (2010) Ecosystem Functions of Tidal Fresh, Brackish, and Salt Marshes on the Georgia Coast. Estuaries and Coasts 33(1): 161 169 Williams K, Ewel KC, Stumpf RP, Pu tz FE, Workman TW (1999) Sea level rise and coastal forest retreat on the west coast of Florida, USA. Ecology 80(6): 2045 2063 Winfrey MR, Zeikus JG (1977) Effect of sulfate on carbon and electron flow during microbial methanogenesis in freshwater sediment s. Applied and Environmental Microbiology 33(2): 275 281 Wong VNL, Dalal RC, Greene RSB (2008) Salinity and sodicity effects on respiration and microbial biomass of soil. Biology and Fertility of Soils 44(7): 943 953 Wright AL, Reddy KR (2001) Heterotrop hic microbial activity in northern Everglades wetland soils. Soil Science Society of America Journal 65(6): 1856 1864 Wu Y, Tam NFY, Wong MH (2008) Effects of salinity on treatment of municipal wastewater by constructed mangrove wetland microcosms. Marine Pollution Bulletin 57(6 12): 727 734 Yu KW, Faulkner SP, Patrick WH (2006) Redox potential characterization and soil greenhouse gas concentration across a hydrological gradient in a Gulf coast forest. Chemosphere 62(6): 905 914 Yuan BC, Li ZZ, Liu H, Gao M Zhang YY (2007) Microbial biomass and activity in salt affected soils under and conditions. Applied Soil Ecology 35(2): 319 328 Zehnder AJB, Ingvorsen K, Marti T (1982) Microbiology of methane bacteria. In: Hughes DE (ed) Anaerobic digestion. Elsevier. p 45 68
202 BIOGRAPHICAL SKETCH Lisa Gardner Chambers grew up in North Canton, Ohio with her parents, Bruce and Marcia Gardner and older brother, Chris Gardner Her family instille d in her a great appreciation fo r nature through weekend trips to the lake and family vacations at National Parks. Attending Ohio State Uni versity ( Columbus, O hio), Lisa graduate d magna cum laude with a Bachelor of Science in n atural r esources in 2003. At Ohio State, her desire to study wildlife veterinary medicine pr opelled her to enroll in an environmental science course which permanently changed her academic focus toward a career in ecology and conservation Her undergraduate mentor, Dr. Virginie Bouchard, was instrumental in honing this interest to a concentratio n i n wetland s Following her undergraduate degree, Lisa work ed for two year as a Wetland Consultant and one year as a Watershed Director for a an environmental non profit near Dayton, Ohio However, her passion for academia and interest in large scale we tland restoration projects led her to accept a fellowship at Louisiana Stat e University in Baton Rouge, Louisiana, under Dr. John White. Her m aster s research focused on the biogeochemical impacts of a Mississippi River Freshwater Diversion on coastal wet lands in southern Louisiana S he graduated from the Department of Oceanography and Coastal Sciences with the Outstanding Thesis award in 2008. In the summer of 2008, Lisa received both a Nation al Science Foundation Integrative Graduate Traineeship (IGE RT) fellowship and an Alumni Fellowship from the Soil and Water Science Department to begin her Ph.D. at the University of Florida in Gainesville, Florida under Dr. K Ramesh Reddy Her Ph.D. program focused o n interdisciplinary science and included a li fe changing summer studying in southern Africa. Lisa followed her enthusiasm for coastal wetlands in her Ph.D. program,
203 developing an externally funded research initiative focused on the biogeochemical effects of sea level rise on coastal wetland carbon c ycling. During this time, Lisa also met and married her husband, Dr. John Chambers. Lisa has dedicated her career to studying the fundamental science behind globally important e co systems and being a mentor and leader in the field of wetland science.