<%BANNER%>

A Biogeochemical Survey of Wetlands in the Southeastern United States


PAGE 1

A BIOGEOCHEMICAL SURVEY OF WETLANDS IN THE SOUTHEASTERN UNITED STATES By STACIE GRECO A THESIS PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLOR IDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE UNIVERSITY OF FLORIDA 2004

PAGE 2

Copyright 2004 by Stacie Greco

PAGE 3

This document is dedicated to my friends and family whom have allowed me the time and space for intellectual and emotional growth.

PAGE 4

iv ACKNOWLEDGMENTS It is said that it takes a village to rais e a child. Similarly, it takes a community to write a thesis! I am thankful for the guidance and wisdom my committee provided throughout this process. Dr Mark Clark’s contagious enthusiasm has helped me overcome many doubts and fears. Kevin Grace’s perpetual encouragement and patience has greatly improved the quality of this wor k. The editing expertise of Dr. Tom Crisman has been instrumental to this document. I would also like to ac knowledge the hard work of the Wetland Biogeochemistry Laboratory a nd the many helping hands in the field. Finally, this research was made possible by funding from the USEPA’s Office of Water.

PAGE 5

v TABLE OF CONTENTS page ACKNOWLEDGMENTS.................................................................................................iv LIST OF TABLES............................................................................................................vii ABSTRACT....................................................................................................................... xi CHAPTER 1 INTRODUCTION........................................................................................................1 Regulatory Background................................................................................................2 Water Quality Standards........................................................................................2 Numeric Nutrient Criteria......................................................................................3 Types of Wetlands........................................................................................................4 Defining Ecoregions.....................................................................................................7 Limiting Nutrients and Causal Variables.....................................................................9 Biological Indicators of Nutrient Enrichment..............................................11 Biogeochemical Indicators of Nutrient Enrichment ....................................12 Reference Wetlands....................................................................................................14 Research Objectives....................................................................................................15 Hypotheses..................................................................................................................15 2 METHODS.................................................................................................................18 Site Selection..............................................................................................................18 Identifying Minimally Impaired Sites.................................................................18 Identifying Wetland Community Types..............................................................20 Hydrologic Classification.............................................................................21 Site Selection Criteria...................................................................................23 Sampling and Analytical Protocols............................................................................24 Sample Locations................................................................................................24 Sample Collection and Processing......................................................................27 Water............................................................................................................27 Soil...............................................................................................................28 Leaf litter......................................................................................................29 Data Analysis..............................................................................................................30

PAGE 6

vi 3 RESULTS AND DISCUSSION.................................................................................32 Within Wetland Variability........................................................................................35 Water Column ......................................................................................................35 Litter....................................................................................................................37 Soil.......................................................................................................................38 Discussion............................................................................................................38 Variability among Wetland Types..............................................................................39 Vegetative Comparisons: Swamps and Marshes.................................................39 Water column...............................................................................................40 Litter.............................................................................................................41 Soil...............................................................................................................46 Discussion....................................................................................................47 Hydrologic Comparisons: Rive rine and Non-riverine........................................51 Water column ..............................................................................................51 Litter.............................................................................................................53 Soil...............................................................................................................54 Discussion....................................................................................................54 Spatial Variation.........................................................................................................60 Water Column.....................................................................................................62 Litter....................................................................................................................64 Soil.......................................................................................................................67 Discussion............................................................................................................70 4 CONCLUSIONS........................................................................................................75 APPENDIX A WETLAND CHARACTERIZATION FORM...........................................................79 B WETLAND IDENTIFICA TION AND LOCAtION..................................................82 C PHYSICAL SOIL AND WATER COLUMN DATA................................................87 D SOIL, LITTER, AND WATER COLUMN CHEMICAL DATA.............................96 LIST OF REFERENCES.................................................................................................105 BIOGRAPHICAL SKETCH...........................................................................................110

PAGE 7

vii LIST OF TABLES Table page 1-1 Comparison of wetland characteris tics reported in the literature...............................8 2-1 The NWI classification scheme................................................................................22 2-2 Summary of chemical analyses and methods...........................................................29 3-1 Various aggregations of the wetlands surveyed.......................................................33 3-2 Results of pair-wise comparison of core and edge areas.........................................36 3-3 Results of pair-wise comparison of core and edge areas.........................................37 3-4 Results of pair-wise comparison of core and edge areas.........................................39 3-5 Water column properties..........................................................................................40 3-6 Litter phosphorus, nitrogen, and carbon content......................................................43 3-7 Soil P, N, and C content...........................................................................................45 3-8 Values from the current study comp ared to those in the literature..........................47 3-9 Power analysis for non-significant para meters within community comparisons.....49 3-10 Water column properties..........................................................................................52 3-11 Leaf litter properties.................................................................................................55 3-12 Soil properties..........................................................................................................5 7 3-13 Number of surveyed wetlands within the three USEPA Nutrient Ecoregions.........62 3-14 Water column descriptive statistics for surveyed wetlands by ecoregion................63 3-15 Litter descriptive statistics for surveyed by Ecoregion............................................66 3-16 Soil descriptive statistics for survey ed wetlands aggregated by Ecoregion.............69 3-17 Summary of significan t differences among USEPA Nutrient Ecoregions ..............71

PAGE 8

viii 3-18 Soil total phosphorus statistics.................................................................................74 B-1 Wetland identification and location.........................................................................83 D-1 Chemical soil, litter, and water column data for edge (E) and Core (C) sites..........97

PAGE 9

ix LIST OF FIGURES Figure page 1-1 USEPA Level IIII Nu trient Ecoregions.....................................................................9 1-2 Two approaches for estab lishing reference conditions............................................14 2-1 Sampling areas within the thre e USEPA Nutrient Ecoregions................................20 2-2 Number of wetlands surveyed aggregated by community type...............................25 2-3 Sub-sample locations within the core and edge zones.............................................26 3-1 Total area of the four wetland types.........................................................................34 3-2 Percentage distribution of surv eyed wetlands within ecoregions............................34 3-3 Water column TP and TN values by vegetative type...............................................41 3-4 Litter phosphorus, nitrogen, and carbon values by community type.......................42 3-5 Soil %P, %N, and %C values by community type...................................................44 3-6 Water column TP and TN va lues by hydrologic connectivity.................................53 3-7 Litter phosphorus, nitrogen, and carbon content comparisons.................................55 3-8 Soil TP and TN values by hydrologic connectivity.................................................56 3-9 Distribution of wetlands within th e three USEPA Nutrient Ecoregions..................61 3-10 Comparison of ecoregion s aggregated by hydrology...............................................64 3-11 Comparison of ecoregions aggr egated by vegetative type.......................................65 3-12 Comparison of riverine we tlands in the three ecoregions........................................67 3-13 Comparison of litter total phosphorus among the three ecoregionss.......................68

PAGE 10

x 3-14 Distribution of sampling locations wi thin the USEPA Nutrient Ecoregions...........73

PAGE 11

xi Abstract of Dissertation Pres ented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Master of Science A BIOGEOCHEMICAL SURVEY OF W ETLANDS IN THE SOUTHEASTERN UNITED STATES By Stacie Greco August 2004 Chair: Thomas Crisman Cochair: Mark W. Clark Major Department: Environmental Engineering Sciences The USEPA’s National Water Quality Invent ory Reports consistently cite nutrient enrichment as one of the leading causes of wa ter quality impairment. To target problems associated with nutrients, the Clean Wate r Action Plan of 1998 requires the USEPA to establish numeric nutrient cr iteria specific to geographi c region and waterbody type. Developing nutrient criteria for wetlands is difficult due to a lack of historic data, incompatibility of methods employed in prev ious studies, and inhe rent variability among wetland community types. The primary objectives of this study were to conduct a biogeochemical survey of minimally impaired wetlands within the southeastern US and to determine the effect, if any, of regional, hydrologic, a nd vegetative differences on wetland nutrient condition. One hundred and three wetlands were sample d in three USEPA Nutrient Ecoregions covering four states. Sampling was distri buted among wetlands classified by hydrologic

PAGE 12

xii connectivity into riverine and non-riverine and by dominant vegetative cover into swamps and marshes. Soil and litter parameters did not differ si gnificantly between sw amps and marshes, suggesting a distinction between vegetative types is not neces sary for determining soil or litter numeric nutrient criteria in the s outheast. Water colu mn total phosphorus differences between swamps and marshes impl y a need to set numeric nutrient criteria specific to dominant vegetative cover. Hydrologic connectivity appears to be important when characterizing wetland nutrient regimes, as demonstrated by di fferences in water column, litter, and soil characteristics between riverine and non-ri verine wetlands. Ri verine wetlands had greater water column and litte r total phosphorus content and lower soil total nitrogen content compared to non-riverine wetla nds. It is hypothesize d that hydrologic connectivity to adjacent aqua tic ecosystems and larger contributing watersheds of riverine wetlands drives these differences. The USEPA recognized the importance of regional influences on wetland nutrient regimes when the decision was made to de termine numeric nutrien t criteria specific to ecoregions. Results demonstrate that the Sout hern Coastal Plain (XII) is different from the Southern Forested Plain (IX) and the East ern Coastal Plain (XIV), with greater water column total nitrogen, litter total carbon, so il total nitrogen, soil total carbon, and lower litter total phosphorus content. Variability was still large within a given ecoregion; therefore spatial aggregation at a sub-ecore gion level may be necessary for effective nutrient criteria development

PAGE 13

1 CHAPTER 1 INTRODUCTION From the billions of appropriated dollars for restoration of the Florida Everglades to the coining of wetlands as “nature’s kidne ys,” it is evident that wetlands are the ecological buzzword and ecosystem focus of the millennium. It is hard to believe that only a few decades ago wetlands were viewed as wastelands, portrayed by the popular image of the Swamp Thing surrounded by putri d swamp gas. Before their inherent values were recognized, wetlands were dr ained and converted to human-maintained agricultural and sylvicul tural lands at an alarming rate. The conversion of wetlands to “more productive” land uses has recently de creased to a still alarming rate of 23,674 hectares a year (United States Environmen tal Protection Agency 2002). However, such losses only represent complete destructi on of these ecosystems and do not account for numerous additional hectares where wetland functions have been degraded due to changes in hydrology, vegetation, and/or water qu ality. It is this change in ecosystem function, and thereby potential loss of designated use, that led to implementation of the Clean Water Act (CWA) in 1972 and the curren t directive to estab lish numeric nutrient criteria for water bodies within the USA. This thesis addr esses some of the issues for establishing numeric criteria for wetlands a nd presents results of a wetland survey conducted in the southeastern United States.

PAGE 14

2 Regulatory Background Water Quality Standards Section 304(a) of the CWA mandates th e United States Protection Agency (USEPA) to assist states, tribes, and territo ries in developing wate r quality standards. Such standards contain three major components: 1) determination of designated uses, 2) development of numeric or narrative criteria to protect designated uses, and 3) development of antidegradation policy to avoid impacts not addressed by the developed criteria (USEPA 1983). As of the late 1990s, 39 states lacked water quality standards for wetlands (USEPA 2000). States, tribes, and territories are requ ired to determine designated uses of waterbodies within their jurisd iction. These must meet the goa ls of Section 101(a) of the CWA, which include protection and propagati on of fish, shellfish, and wildlife along with providing for recreation opportunities (USEPA 1983). De fining the designated uses for rivers and lakes is a stra ightforward task since the valu es of swimming, fishing, and water sports are easily recognized. This is not the case with wetland s because historically their values have not been recognized, and they are not always obvious. Wetland values can include flood storage, pollution and sedi ment control, food web support, groundwater replenishment, and habitat for various orga nisms including waterf owl (Moore et al. 1999, Morris 1979). Many states simply assign de signated uses based on wetland type or location in the landscape (USEPA 1990), since it is difficult to assign values to each individual wetland. Once states determine the designated us es of a waterbody, criteria must be developed to protect those uses. The criteria of water quality standa rds can be narrative or numeric. Narrative criteria are import ant for impacts that cannot be addressed by

PAGE 15

3 numeric criteria, such as those that do not directly affect water chemistry. For example, discharge of dredge and fill material can be prevented using narrative criteria. Numeric criteria are values or ranges assigned to measurable chem ical, physical, and/or biological parameters. They can be more useful than narrative criteria because they provide a clear distinction between acceptable and unacceptabl e conditions, and hence, reduce ambiguity for management and enforcem ent decisions (USEPA 2000b). Under Section 305(b) of CWA, states, tribes and territories are required biennially to compare monitoring results with their water quality standards. To identify trends in water quality, the USEPA compiles the data and publishes the National Water Quality Inventory Report. These reports consistently identify nutrients as one of the leading causes of water quality impairment and failu re to sustain the designated uses of waterbodies. Excessive nutrients are responsible for almost 50% of impaired lake area and 60% of impaired river reaches in the US (Smith et al. 1999). Numeric Nutrient Criteria To target problems specifically associat ed with nutrient enrichment, President Clinton introduced the Clean Water Action Pl an of 1998, which requires the USEPA to establish numeric nutrient crite ria specific to ecosystem type and geographic region. The agency responded with a document describing its approach titled the National Strategy for the Development of Regional Nutrient Criteria. The document describes the USEPA’s intention to publish technical guida nce manuals for each of the four waterbody types (lakes and reservoirs, rivers and st reams, estuaries, and wetlands) along with criteria recommendations for specific ecoregions. The USEPA intended to recommend target nutrient ranges on a geographic basis using historical nutrient data reference conditions, and expert knowledge (USEPA 1998).

PAGE 16

4 Lakes/reservoirs, rivers/streams, and estuaries are well-monitored ecosystems with sufficient data available to support numeric criteria development. With exception of the Florida Everglades, wetlands lack even a sk eletal survey of nutrient condition. There is a lack of historical wetland data since their value as aquatic ecosystems is a relatively recent phenomenon. The 2000 National Water Quality Inventory Report was unable to make conclusions concerning wetla nd water quality because only 8% of total wetlands in the US were surveyed, in contra st to 42% of US lakes (USEPA 2000a). For those wetlands that have been monitored, num erous parameters have been measured, and a variety of sampling techniques and me thodologies have been utilized making comparisons and regional characterization di fficult. The exception is for the Florida Everglades, which have been st udied sufficiently to provide data for the USEPA to make wetland numeric nutrient recomm endations (USEPA 2000c). Establishing numeric criteria for wetla nds requires the determination of 1) designated use, 2) appropriate regional or type of wetland aggregation scheme to which criteria are sufficiently but not overly protec tive, 3) limiting nutrient/casual variable to determine which nutrients require criteria deve lopment, or in the absence of a clear cause and effect threshold of impairment, the qua ntification of nutrien t concentrations under reference conditions. As di scussed above, determination of designated use requires recognition of wetland values and benefits to local communities. Determining appropriate aggregation of wetlands for de velopment of numeric criteria requires a thorough investigation of potential differences among wetland types and regions. Types of Wetlands Definitions of wetlands include a suite of ecosystems supporting various functions. Common wetland types of North America in clude freshwater marshes, peatlands,

PAGE 17

5 freshwater swamps, riparian systems, tidal sa lt marshes, tidal fres hwater marshes, and mangrove wetlands (Mitsch and Gosselink 2000). These terms generally define the dominant vegetation type and hydrologic regime Marshes are characterized by annual or perennial herbaceous species, and swamps ar e dominated by perennial woody vegetation (Brinson et al.1981). Hydrologica lly, wetlands are broadly categor ized as riverine, tidal, lake fringe, or isolated. Extensive forested floodplains are comm on in the southeastern United States. These riverine wetlands (also called floodplains, bottomlands and riparian wetlands) are connected to nearby rivers or streams, which supply water and nutrients during flood events. Riverine systems also receive consid erable inputs from r unoff of the surrounding landscape (Craft and Casey 2000). Riparian wetlands play a critical role in maintaining water quality, as they efficiently trap sediments and associated contaminants (Hupp 2000). Between 85 to 90% of sediments leaving agricultu ral fields can be captured by wooded riparian wetlands (Gilliam 1994). These wetlands are also impor tant for flood control and provide valuable forest habitat. Although forested floodplains are more comm on in the southeastern United States, herbaceous wetlands can also be found adjacent to rivers and streams. In riverine wetlands there is a narrow opportunity for co lonization between the exposure of alluvial sediments and the return of high water levels and erosional forces (Willby et al. 2001). Large rivers whose extensive floods deposit se diments in adjacent wetlands may have poorly developed riparian marshes.

PAGE 18

6 Riverine marshes are included in this study because there are few studies comparing nutrient cycling between forested and herbaceous systems. Hopkinson (1992) concluded that the growth form of the dom inant vegetation does not influence nutrient retention, although study results showed that a fo rested riverine system retained slightly more nutrients than a riparian marsh (4.3% vs. <1%). Woody vegetation serves as longterm storage of nutrients, while herbaceous vegetation of marshes provides mainly shortterm storage (Reddy and D’Angelo 1994) These differences may lead to biogeochemical differences be tween these wetland types. Depressional wetlands (non-riverine) differ fr om riverine systems because they are not directly influenced by hydrologic fluxes from rivers and streams. Non-riverine wetlands rely on precipitation or groundwater inputs, which tend to have lower nutrient loads than surface waters (Craft and Case y 2000). Hopkinson (1992) determined that relatively closed marshes and swamps of Ok efenokee Swamp retained 90% of inorganic nutrient inputs, whereas small percentages we re retained in riverine systems. He concluded that the openness of a wetland determines nutrient loading, which is strongly correlated with productivity, organic ma tter decomposition, and nutrient cycling. Systems with low nutrient loading are more ef ficient at cycling nutri ents and have lower net primary production (Craft and Casey 2000). Therefore, riverine and non-riverine wetlands within similar surrounding land-uses may naturally display different nutrient concentrations, organic matter conten t, and biogeochemical processes. Differences among riverine verses non-riverine systems and marshes verses swamps hinder generalizations about wetland s (Table 1-1). One exception is that excessive loading of nutrients can alter ecosy stem dynamics. If wetland functions are to

PAGE 19

7 be protected through developmen t of numeric nutrien t criteria, individual wetland types may need to be studied along with thei r regional variation as background for sound environmental regulations. Defining Ecoregions The Clean Water Action Plan of 1998 in cludes a spatial co mponent in its requirement to establish nutrient criteria by geographic regions. The USEPA is addressing spatial variability via geographic regions called ecoregions, which are areas with relatively homogenous ecosystems that differ from adjacent regions (Omernik and Bailey 1997) and are based on geology, physiolo gy, vegetation, climate, soils, wildlife, and hydrology. Omernik (1987) divided the conterminous US into ecoregions based on regional patterns resulting from the combination of compone nt maps including land-use, land-surface forms, potential na tural vegetation, and soils. The USEPA adopted and adapted Omerni k’s ecoregions and stratified them hierarchically. Level I is the coarsest Un ited States ecoregion and is composed of 15 ecological regions, Level II is represente d by 52 regions, and Level III contains 84 ecoregions (Brewer 1999). Level III ecoregi ons with similar characteristics that contribute to nutrient regimes were aggregat ed to create USEPA Nutrient Ecoregions (Figure 1.1). The USEPA recommends that numeric nutrient criteri a be established for lakes/reservoirs, streams/rivers, estuaries, and wetlands within each of the Nutrient Ecoregions. The current study area include s wetlands within the Sout heastern Forested Plain (IX), Southern Coastal Plain (XII), and Ea stern Coastal Plain (XIV) ecoregions. Comparisons were made among the three ecoregions to determine if they are appropriate aggregations for setting numeric nutrient criteria.

PAGE 20

8 Table 1-1. Comparison of wetland characte ristics reported in the literature Parameter Riverine Non-riverine Major source of inputs (Craft and Casey 2000) Runoff Precipitation Connectivity to other systems (Hopkinson 1992) Open Closed Nutrient Cycling (Hopkinson 1992) Less efficient More efficient Soil C:N ratios (Craft and Casey 2000) Similar Similar Parameter Swamps Marshes Nutrient retention (Wilby et al. 2001) Similar Similar Biomass turnover rates (Hopkinson 1992) One magnitude lower One magnitude higher Live tissure N:P ratios (Bedford et al. 1999) Greater Lower Live tissue N:P ratios (Bedford et al. 1999) Suggest P-limitation or co-limitation by N and P Less than 14, suggesting Nlimitation Litter %N (Bedford et al. 1999) Lower Greater Litter %P (Bedford et al. 1999) Similar Similar Soil N:P ratios (Craft and Casey 2000) Low, suggesting Plimitation or colimitation by N and P Greater, Suggesting Plimitation Average water temperatures (Lee and Bukaveckas 2002) Cooler Warmer Algal growth (Battle and Golladay 2001) Low Greater

PAGE 21

9 Figure 1-1.USEPA Level IIII Nutrient Ec oregions for numeri c nutrient criteria recommendations (USEPA 2003) Limiting Nutrients and Causal Variables. Anthropogenically derived nutrients enter aquatic ecosystems from point sources, such as wastewater effluents, and nonpoint sources including agricultural, urban, and construction runoff. Nonpoint sources are ma jor contributors of nut rients to aquatic systems and are most difficult to regulate (Sm ith et al. 1999). Agricultural is the primary source of nonpoint nutrient pollution in the Un ited States due mainly to fertilizer application and accumulation of animal manure (Carpenter et al. 1998). Methodologies in USEPA technical guida nce manuals for establishing numeric criteria are based on limiting nutrients, impl ying that primary production of plants is limited by the nutrient that is the least availa ble relative to the plants requirement for growth. This concept is Liebigs Law of th e Minimum (Smith et al. 1999). Nitrogen (N)

PAGE 22

10 and phosphorus (P) are the nutrients most commonly cited as limiting plant growth (Carpenter et al. 1998, Gusewell et al. 1998, Koerselman and Meuleman 1996, Smith et al. 1999). Therefore, with increased loadi ng of N and/or P to aquatic ecosystems, primary production usually increases and can lead to eutrophication. The agent that causes change in an ecosystem is referred to as the casual variable, while the factor that reacts is called the response variable (USEPA 2000a). For example, when concentrations of a limiting nutrient (casual variable) incr ease, the dominance of fast-growing species (response variable) increases, and they repla ce less competitive species (Gusewell et al. 1998). Eutrophication is the process whereby an aquatic ecosystem shifts from a low nutrient (oligotrophic) to a highly productive, nutrient rich (eutrophic) system (Mitsch and Gosselink 2000). If the shift is the result of human activities, the process is called cultural eutrophication. Eutrophication is characterized by increased growth of algae and/or macrophytes, which can hinder use of water for fishing, recreation, industry, and domestic consumption. Decomposition of excessive algae and macrophytes reduces oxygen supplies, which can lead to fish kill s (Carpenter et al. 1998). Eutrophication can also alter foodwebs, re sulting in a loss of biodiversity (Car penter et al. 1998, Smith et al. 1999). In fact, high species bi odiversity has been correlated with low nutrient regimes (Bedford et al. 1999). Preventing input of nutrients from an thropogenic sources does not necessarily result in decreased plant growth, due to in ternal biogeochemical cycling of nutrients within wetlands. Decomposition of stored organic matter can provide the nutrients required for plant growth (Reddy and D’Ange lo 1994). Nutrient transformations depend

PAGE 23

11 on many factors, including hydrol ogic regime, influent nutrien t concentrations, existing nutrients in the system, vegeta tion, and sediments (Gopal 1999). Predicting the extent of internal nutrient cycling in wetlands is difficult due to inherent differences among wetland ecosyste ms. For example, nutrient cycling of riverine and non-riverine we tlands is influenced by di ssimilar hydrologic regimes. Hopkinson (1992) found that the dominant plan t growth form was the primary factor influencing biomass turnover rates, with mars hes cycling an order of magnitude greater than swamps. Therefore, to determine nutri ent effects in a wetland, it may be necessary to examine several components of various wetland types. If a limiting nutrient was always the factor limiting the system, it would be simple to develop regulations. But aquatic systems are dynamic, and several factors can affect production. For example, plant biomass cha nges seasonally, fluctuates with land-use, and varies regionally (USEPA 2000a). Therefore, to establish numeric nutrient criteria, it is necessary to develop an efficient tool fo r quantifying the nutrient regime of wetlands. An effective nutrient indicator must be sensi tive to varying nutrient regimes, easy to measure and interpret, inexpens ive to apply, and should have as few temporal and spatial constraints as possible. The USEPA is expl oring biological and/or chemical indicators (or indices) to assess ecosystem integrity. Biological indicators of nutrient enrichment Biological assessments of wetlands often l ook at community-level parameters such as abundance, biomass, density, richness, diversity, and community composition as indicators of anthropogenic st ressors (Adamus and Brandt 19 90). Galatowitsch et al. (1999) looked at possible plant, bird, invertebrate, fish, a nd amphibian metrics in eight wetland types in Minnesota. Thei r results indicate that specif ic metrics would have to be

PAGE 24

12 developed for the different wetland types a nd ecoregions. A study in the prairie pothole region of the US looked at the value of macrophyte abundance, species richness, and amounts of litter and standing dead vegetation as indicators of wetla nd health. None of the examined indicators successfully quantif ied ecosystem health (Kantrud and Newton 1996). However, Lane et al. (2004) succe ssfully developed a wetland condition index based on macrophytes, macroinvertebrates, and diatoms for isolated depressional marshes of peninsular Florida. As nutrient levels in a wetland increase, th e chemical structure of the system is altered, leading to biological changes. Microbes are normally first to respond to nutrient pulses with algae following closely behind. Th ere is a time lag betw een casual variables and response variables, particularly in long-lived species (Fennessy et al. 2001). Biological indicators often rely on the response of larger organisms such as plants, invertebrates, and birds (Gal atowitsch et al. 1999, Kantr ud and Newton 1996, Lane et al. 2004). Once organisms respond to a change in the nutrient regime, some of the original structure of the wetland is lost as the new community evolves. One concern with using macrophyte structure as an indi cator is that once a wetland has been dominated by stress tolerant perennials, less aggre ssive species may not be capable of re-coloni zation after the stress is removed (Galatowitsch et al. 1999). The community structure may be a relic of past disturbances. Biogeochemical indicators of nutrient enrichment There is a well-documented correlation between nutrient additions to aquatic ecosystems and proportional increased growth of algae and macrophytes (Carpenter et al. 1998, Morris 1991, Smith et al. 1999). Likewise, elevated P and N levels have been associated with decreased species diversity (Bedford et al. 1999, Carpenter et al. 1998,

PAGE 25

13 Morris 1991, and Smith et al. 1999). Phos phorus and nitrogen chemistry of aquatic ecosystems can be monitored to determine th e degree of enrichment before species are eliminated. This is important since 14% of the 130 plant species in the conterminous US listed as endangered or threatened are f ound primarily in wetlands (Morris 1991). Biogeochemical processes, such as organic matter decomposition and denitrification, can reflect nutrient budgets before responses are evident in higher organisms (Reddy and D’Angelo 1997). Nitrog en to phosphorus ra tios (N:P) in plant tissue (Gusewell and Koerseleman 2002, Gusewell et al. 1998, Koerseleman and Meuleman 1996, Shaver and Melillo 1984, Wil by et al. 2001), soil (Craft and Casey 2000) and litter (Baker et al. 2001, Shaver and Melillo 1984) have been studied to assess nutrient limitation in wetlands. Koerselema n and Meuleman (1996) concluded that when N and P are controlling plant growth in wetla nds; vegetation N:P ra tios > 16 indicate P limitation, while N:P ratios < 14 indicate N limitation. There is disagreement in the literature regarding the limitation of wetland productivity. Morris (1991) reviewed severa l wetland studies and concluded that most wetlands are N limited. The results from nu merous wetlands in Scotland, France, and Ireland agree that most wetlands are N limite d (Wilby et al. 2001). However, Craft and Casey (2000) suggest that fres hwater marshes and forested wetlands of southwestern Georgia are P limited. Bedford et al. (1999) concluded that within temperate freshwater wetlands of North America, marshes are N limited, while evergreen, shrub, and deciduous wetlands are P limited. This confus ion demonstrates a need for additional information regarding nutrient regimes of wetlands. This study includes a

PAGE 26

14 biogeochemical characterization of the su rveyed wetlands to determine background levels of nutrients and biogeochemical pr ocesses in minimally impaired wetlands. Reference Wetlands Establishing numeric criteria for wetla nds requires determination of reference conditions as a standard for comparison. One strategy for determining reference values is to survey wetlands representi ng the broad range of nutrient impairment. The lower 25th percentile of this population would be recommen ded as reference conditions (Figure 1-2). An alternative strategy explored in this study is to set refe rence conditions equal to the upper 25th percentile (or 75th percentile) of wetlands identified as minimally impaired systems (USEPA 2000a). Eventually, indi vidual waterbodies will be sampled and compared to reference conditions to de termine appropriate management methods (USEPA 2000b). Figure 1-2.Two approaches for establishing reference conditions using total phosphorus as the example variable (modified from USEPA 2000a) Minimally Impaired Wetlands Representative of all Wetlands

PAGE 27

15 Ideally, reference conditions should re flect conditions in the absence of anthropogenic influences and pollution. Howe ver, human activities have impacted all ecosystems to some degree; therefore, refe rence conditions realistically represent the least impacted conditions. The USEPA Science Advisory Board endorses use of conditions representing minimal impact as a baseline that should pr otect the beneficial uses (or designated uses) of aquatic resources (USEPA 2000a). The results of this study will help determine appropriate reference conditions for developing numeric nutrient criteria. Research Objectives The survey of this thesis will assess b ackground nutrient concentrations in wetlands to define water quality require d to maintain ecological inte grity. An additional goal of this research is to explore differences in nutrient regimes among various wetland types to determine appropriate wetland aggregation scheme s for setting criteria. Results from this comparison may be instrumental in developi ng nutrient criteria that are sufficiently protective and feasible. Furt hermore, regional aggregates will be explored to gain additional understanding of th e spatial component of wetland nutrient regimes within the southeastern United States. Hypotheses The “openness” of a system to hydrologic and material influxes influences its nutrient loading and pr oductivity (Hopkinson 199 2). Riverine systems are open to such influxes and typically act as sinks for se diment and phosphorus from the contributing watershed (Craft and Casey 2000), whereas non -riverine systems are considerably less open to influxes. It is hypothesized that rive rine wetlands will have higher nutrient levels within soil and water compared to non-riverine systems.

PAGE 28

16 In open systems with high nutrient influx, plan ts have less efficient nutrient cycling and reabsorb fewer nutrients from senesc ing leaves (Hopkinson 1992). Hence, nutrient content of leaf litter is greater in areas w ith increased nutrient availability (Shaver and Melillo 1984). It is hypothesized that ri verine wetlands will have increased nutrient levels in leaf litter compar ed to non-riverine wetlands. The structure of marshes and swamps is quite different, with the former characterized by herbaceous vegetation and th e latter by woody growth forms. In marsh ecosystems, the majority of C, N, and P is stored in the soil, whereas swamps store a great deal of C, N, and P in plant biomass (Hopkinson 1992). Additionally, biomass turnover rates are an order of magnitude greater in ma rshes than swamps (Hopkinson 1992). This continual decomposition of herbaceo us organic matter releases nutrients into the soil; therefore, it is hypot hesized that marshes will have higher nutrient levels in soil than swamps. Battle and Golladay (2001) found that sedge marshes have higher algal growth than cypress swamps. This difference likely refl ects the absence of overstory cover in marshes. Algal populations quickly sequester nutrients from the water column, hence decreasing soluble nutrients available to ot her growth forms (Kadlec and Knight 1996). It is hypothesized that the water column of marshes will have lower N and P content than swamps. There is a spatial component to the nutr ient regimes of wetlands, as recognized by the USEPA’s use of ecoregions in determini ng numeric nutrient criteria. The various geologic formations of the southeastern United States affect hydrology. Hydrology is often cited as the most important defining pa rameter of wetland systems (Ehrenfeld and

PAGE 29

17 Schneider 1991, Fennessy and Mitsch 2001, J ones et al. 2000, Reinelt et al. 1998). Therefore, it is hypothesized that there will be regional differences in the nutrient regimes of wetlands.

PAGE 30

18 CHAPTER 2 METHODS To address the research goal of determin ing background concentrations of nutrients in minimally impaired wetlands, it was necessa ry to locate and surv ey several types of wetlands within areas of nominal anthropoge nic disturbances. Two vegetative and two hydrologic classes were selecte d, resulting in four wetland cla sses. The four wetland types surveyed were riverine marshes, non -riverine marshes, riverine swamps, and nonriverine swamps. To evaluate the spatia l component of wetland nutrient regimes, selection of wetlands was stratified with in three USEPA Nutrient Ecoregions (Southeastern Forested Plains, Southern Coasta l Plains, and Eastern Coastal Plains) in the southeastern United States. Site Selection The site selection process identified mi nimally impaired wetlands within three ecoregions of the southeastern United States that met the criteria for wetland community type (marsh versus swamp), accessibility (pr oximity to forest roads and within public ownership), and hydrologic conne ctivity (riverine verses non-ri verine). The large spatial extent of the study area necessitated a Geogr aphic Information System (GIS) for locating sampling sites and analyzing spatial relations hips. All GIS analysis was done using ArcGIS 8.1. Identifying Minimally Impaired Sites Nutrient enrichment is often a resu lt of fertilizer runoff from surrounding agricultural and urban areas. It was assumed, as supported by Kantrud and Newton

PAGE 31

19 (1996), that wetlands located close to agri cultural areas would ha ve greater nutrient loading than those farther from intensive agri cultural activities. Locating wetlands that were not influenced by agriculture was problem atic due to the scale of the survey. The study area included Florida, Georgia, Alab ama, and South Carolina. Collecting and analyzing detailed land-use da ta for the entire study area wa s not logistically feasible. Furthermore, utilizing data from various s ources (such as four st ate agencies) can be difficult to integrate because of different scal es and varying standards for data quality and collection. The USEPA’s suggestion to us e sites located within the boundaries of public lands as minimally impaired wetlands was adopted (U SEPA 2000a). It is lik ely that these sites are less influenced by cultural nutrient enri chment than wetlands on private lands, as indicated by a Landscape Devel opment Intensity (LDI) Index for assessing the intensity of various land-uses (Brown and Vivas in press). The index u tilizes calculated LDI coefficients ranging from 1.0 (natural system s) to 7.0 (high intensity agricultural). Forestry is a common land-use on public lands. The LDI coefficient for pine plantations is 1.58, which indicates minimal influence of wetlands near silviculture activity. A public lands coverage was obtained and overlayed on the USEPA Nutrient Ecoregion map (Figure 1-1). The largest publi c land tracts in the southeastern United States lie within the boundaries of National Forests; therefor e, efforts were concentrated on identifying National Forests within the three USEPA Nutrient Ecoregions of the Southeastern US (Figure 2-1). Permits were obtained to sample within Apalachicola, Conecuh, Francis Marion, Ocala, Oconee, Os ceola, Sumter, and Talladega (Oakmulgee District) National Forests. Because Geor gia had considerable aerial gaps without

PAGE 32

20 National Forest lands, portions of Fort Benning Military Preserve, Moody Air Force Base, and Banks Lake National Wildlife Refuge were also sampled. Figure 2-1. Sampling areas within the three USEPA Nutrient Ecoregions. Identifying Wetland Community Types Once an area was selected, it was necessary to identify the wetlands present, categorize them into the four target comm unity types, and randomly select sampling 0240480 120KilometersLegend Apalachicola NF Banks Lake NWF Conecuh NF Fort Benning Military Francis Marion NF Moody Air Force Base Ocala NF Oconee NF Osceola NF Sumter NF Talladega NFlSoutheastern Forested Plain Southern Coastal Plain Eastern Coastal Plain

PAGE 33

21 locations. To complete this task, the Unite d States Fish and Wildlife Service (USFWS) National Wetlands Inventory (NWI) was util ized. NWI maps were created through photo-interpretation of aeri al photography supplemented by soil surveys and field verification. (fttp://www.nw i.fws.gov/arcdata/readme.txt, 2002). NWI data were downloaded in 7.5 minute quadrangles from the USFWS website. Classification of wetlands on NWI maps was based on the USFWS Wetland and Deepwater Habitat Classifi cation System (Cowardin et al. 1979), which groups ecologically similar habitats t ogether (Tiner 1999). For th is study, swamps are analogous to Cowardin’s forested wetlands, whic h include wetlands characterized by woody vegetation at least six meters tall. Mars h sites correspond with Cowardin’s emergent wetland class characterized by erect, roote d, herbaceous hydrophytes th at are present for most of the growing season. For this st udy, eleven NWI sub-class level communities were aggregated into two co mmunity types (Table 2-1). NWI data were not available for Talla dega and Conecuh National Forests in Alabama. A hydric soils shapefile was obt ained from United States Forest Service (USFS) personnel and used to identify wetlands at these sites. Community types were determined during the Alabama site visits, sin ce this distinction could not be made with available GIS data. Hydrologic Classification Mitsch and Gosselink. (2000) defined ri parian wetlands as those ecosystems located where streams or rivers at leas t occasionally flood beyond their confined channels. The littoral zone of lakes is often lumped into the riparian wetland classification. To decrease variability among sampled wetlands, those adjacent to rivers and streams were included in this study, while littoral wetl ands of lakes were excluded.

PAGE 34

22 Table 2-1.The NWI classification scheme aggregated into swamp and marsh wetland types NWI Classification System Subsystem Class Sub-class Current Study Classification Palustrine Forested Broad-leaved Deciduous Palustrine Forested Needle-leaved Deciduous Palustrine Forested Broad-leaved Evergreen Palustrine Forested Needle-leaved Evergreen Palustrine Forested Dead Palustrine Forested Indeterminate Deciduous Palustrine Forested Indeterminate Evergreen Swamp Palustrine Emergent Persistent Palustrine Emergent Non-persistent Riverine Tidal Emergent Non-persistent Riverine Lower Perennial Emergent Non-persistent Marsh To identify riverine wetlands, proximity of wetlands to streams and rivers was determined using stream data from various sources. The National Hydrography Dataset (NHD), compiled by USGS at a scale of 1:100,000, was utilized for the three Florida National Forests. Stream data for the re maining locations were obtained from USFS staff. The majority of th e stream data provided was also compiled by USGS. Wetlands located at least partially within 40 meters of a river or stream were classified as riverine. Upstream activities must be considered when classifying these wetlands as minimally impaired. To avoid this compli cation, wetlands along small streams (first and second order) were targeted because their he adwaters were often within the forest boundaries. Larger rivers not originating within the boundaries of National Forests were

PAGE 35

23 not included in the survey due to concerns th at agricultural and urba n activities outside the forest, but within the watershed, may cha nge the desired least impaired status. There are several definitions of isolated wetl ands in use. Tiner et al. (2002) defined a wetland as isolated if it is geographically isolated from other wetlands by uplands. Winter and LaBaugh (2003) suggested that isol ated wetlands are those not connected by streams to other surface-water bodies. Comm on to both definitions is the absence of hydrologic connectivity between the wetland in question a nd surrounding water bodies. Regardless of definition, classifying isolated systems can be difficult, especially during extremely wet years when surface water overflo ws connect “isolated” systems to other aquatic ecosystems. To eliminate confus ion surrounding classification of isolated wetlands, sites were divided in to riverine (as defined above ) and non-riverine, as defined by those wetlands that are at least 40 meters from rivers and streams. Site Selection Criteria After wetlands were categorized by vege tation type and hydr ologic connectivity, proximity to potential nutrient sources and accessibility was determined. A property ownership shapefile was obtained from USFS personnel to identify tracts of land under private ownership within the forest boundari es. Wetlands located on private property were omitted from the survey. Forest Service road coverages were added to the map projects to ensure that the wetlands were accessi ble. All of the forests had extensive road systems; therefore, it was not necessary to omit sites due to a ccessibility concerns. Wetland sampling sites were determined by assigning a number to each of the individual wetland polygons that met co mmunity type and hydrologic connectivity criteria and that were not omitted due to private ownership. A random number generator

PAGE 36

24 was used to select those non -riverine swamps, riverine swamps, non-riverine marshes, and riverine marshes to be sampled. For each public land tract, approximately 30 wetlands meeting the selection criteria were identified; although only 12 (three from each class) were sampled. The additional sites were necessary to compensate for any sites that could not be sampled due to GIS coverage error, misclassification, inaccessi bility, or other unexpected issues. The goal was to sample three wetlands of each community type (riverine marsh, non-riverine marsh, riverine swamp, and non-riverine swamp) within each public land tract. However, with the exception of th e Ocala and Oconee National Forests, marsh communities were scarce. Furthermore, as t opographic relief increased in the northern and western extents of the study area, non-ri verine systems became less prevalent. Therefore, wetland community types were sampled in proportion to their relative abundance (Figure 2-2). More swamps were sa mpled than marshes, and the majority of surveyed wetlands were riverine systems. A total of 103 minimally impaired wetlands were surveyed. Sampling and Analytical Protocols Sample Locations Selected wetlands were physically loca ted using a GPS unit, topographic maps, and the coordinates of the select ed sites. Ground truthing leas t impaired status, vegetative community type, and hydrologic connectivity was always a first step when visiting wetlands. If GIS classification was not verified on the ground, the site was reclassified or not sampled.

PAGE 37

25 58 8 27 10 0 10 20 30 40 50 60 70 Riverine Swamp Riverine Marsh Non-riverine Swamp Non-riverine MarshNumber of Surveyed Wetlands Figure 2-2. Number of wetlands survey ed aggregated by community type A visual survey was conducted upon arrival, and the wetland was divided into two general zones, referred to as the core wetland and the edge wetland. (Figure 2-3). In riverine systems, the core (C) was adjacent to the stream, but landward of any natural levees that have formed. The edge (E) of riverine wetlands was lo cated parallel to the adjacent upland, approximately 25 % of the di stance between the upland and the stream. With small non-riverine wetlands, it was possibl e to walk the entire edge (E) of the wetland and sample the four cardinal points at approximately 25 % of the distance between the upland and the center of the wetl and. The center was sampled as the core (C). In large non-riverine systems, only one si de of the wetland was sampled, as if it was a section of a riverine wetland. The core (C) was located in the deep center of the wetland, and the edge (E) was located para llel to the upland side of the wetland approximately 25 % of the distance between th e upland and the center of the wetland.

PAGE 38

26 Within the edge (E) and the core (C), three sub-sample sites were located approximately 30 paces from each other. Transe cts were typically orientated parallel to the upland boundary. To prevent bias, a PVC ring was tossed into the air after 30 paces had been traversed, and where it landed ma rked the sampling location. At each subsample location, water (if present), soil, and l eaf litter were collecte d. A characterization form (Appendix A) that include d a visual vegetation surve y, hydrologic characteristics, and other descriptive information was completed at each sub-sample location. Figure2-3. Sub-sample locations. A) Within the core and edge zones of riverine wetlands. B) Small non-riverine wetlands. C) Large non-riverine wetlands. C C C E E E Upland Edge Core ( A E Upland River Core Edge Ecotone (not sampled) Upland Core B C A E E C C C C C C C E E E E Edge

PAGE 39

27 A handheld YSI-556 meter (Yellow Springs, CO ) was used to record water column pH, dissolved oxygen saturation, temperature, redox potential (Eh) and conductivity at each sub-sample location with water present. Redox potential was measured as ORP with an Ag/Ag-cl electrode. Values were converted to Eh by adding 234 mV to each reading. Measurements were made with the probe suspended at mid-depth of the water column, but in shallow wetlands (less than 15 cm), the probe was often placed at the sediment-water interface. Sample Collection and Processing Sampling began in April 2003. The survey began with the most southern sites (Ocala National Forest) and then proceeded to the north. Most of the sampling was completed by August 2003. Moody Airforce Base, Fort Benning, and Banks Lake National wildlife Refuge were sampled in September 2003. Water Water was collected, when present, using acid-washed 125-mL HDPE bottles. The bottles were rinsed three times with site water prior to collecting the sample. Care was taken to minimize non-representative partic ulates in the water column; however, the water column often contained particulate matter that was included with the sample. Samples collected at the three sub-sample locations along transect C or E were poured into a pre-acidified (concentrat ed sulfuric acid) 500-m L HDPE bottle to cr eate the zone composite. Water samples were stored on ice for transport to the Wetland Biogeochemistry Laboratory at the Univer sity of Florida (G ainesville, FL). In the laboratory, a sub-sample of the water composite was filtered through 0.45 M filter paper and analyzed for nitrate and nitrite on a rapid-flow analyzer (Table 2-2). An additional (non-filtered) 10 ml sub-sample was digested for Total Kjendal Nitrogen

PAGE 40

28 (TKN) analysis. Results from nitrate/nitrit e and TKN analyses were added together to determine total nitrog en concentrations. Total phosphorus (TP) was determined on a third sub-sample (10 ml) by sulfuric acid and potassium persulfate digestion (EPA method 365.1 1993), followed by colorimetr ic analysis (Technicon AA II). Soil Three soil samples were collected along each of the wetland transects. Prior to sampling, litter and live vegetation were re moved from the sampling area by lightly raking the area by hand. A pre-cl eaned tenite butyrate tube (7 .3 cm. diameter) was driven into the soil at least 10 cm deep. The core tube was then placed on an extruder piston, which was used to push the top of the soil out of the core and into a 10 cm tenite butyrate collar. Any litter remaining on the top of the core was re moved and discarded. The 10 cm core was sliced from the remainder of the core using a stainless steel bread knife and placed in a re-sealable bag. Soils from the three sub-sample sites along transect C or E were combined to create a composite sample. Samples were stored on ice for transport to the laboratory. Coring of soils in densely rooted envir onments was facilitated by using a coring devise with a sharp coring h ead attached to make cutti ng through roots possible and to avoid compacting the sample. An effort was made to avoid large roots, which complicate bulk density calculations. Several swamps, ho wever, contained large root mats, making it impossible to avoid coring through large amounts of root material. In the laboratory, wet weight of the co mposite sample was recorded for bulk density calculations. Roots la rger than 2 mm in diameter were removed from the sample and discarded. The composite sample was hom ogenized. A sub-sample was placed in a shallow 250 mL container, weighed, then dried at 21oC for at least 48 hours.

PAGE 41

29 Table 2-2.Summary of chemical analyses and methods for each stratum sampled Medium Analysis Method TP Sulfuric acid and potassium persulfate digestion followed by colorimetric analysis TKN Sulfuric acid digestion Water NO2-No3 Rapid Flow Analyzer (RFA) Organic matter content Lost on Ignition (LOI) TN Carlos Erba NA 1500 CNS Analyzer (Haak Buchler instruments Saddlebrook, NJ) TC Carlos Erba NA 1500 CNS Analyzer (Haak Buchler instruments Saddlebrook, NJ) Soil TP Ignition Method (Anderson 1976) TP Ignition Method (Anderson 1976) TN Carlos Erba NA 1500 CNS Analyzer (Haak Buchler instruments Saddlebrook, NJ) Litter TC Carlos Erba NA 1500 CNS Analyzer (Haak Buchler instruments Saddlebrook, NJ) The dry sample was re-weighed for percent mo isture calculations. Dry samples were hand-ground using a mortar and pestle, then further ground mechanica lly using a ball mill grinder for at least eight mi nutes. The ground samples were passed through a 1 mm sieve for quality control purposes and placed in scin tillation bottles for analyses. Soil samples were analyzed for organic matter content by lo ss on ignition (LOI), total nitrogen (TN), total carbon (TC), and TP, as summarized in Table 2-2. Leaf litter Leaf litter samples were collected by placi ng a 40 cm diameter PVC ring on the soil surface and hand-collecting all loose mate rial within the ring. Collection was discontinued when the soil surface was reache d, as indicated by the presence of fine, well-decomposed materials. Litter sampli ng was qualitative, not quantitative, since at

PAGE 42

30 times it was necessary to collect multiple samples at a sub-sample location to ensure adequate material for analysis. Litter samp les from the three sub-sample locations were combined to form a composite sample along the core or edge transect. All samples were stored on ice for transport to the laboratory. In the laboratory, litter samples were placed in a paper bag and dried at 21o C for at least 72 hours. The dry samples were coarse ly ground in a Willey mill to pass through a 1 mm screen. The samples were then further ground to pass through a 40-micron followed by an 80-micron screen. To reduce cross-contamination, the mills were vacuumed between each sample. The litter was analyzed for TP, TN, and TC (Table 2-2). Data Analysis All data were analyzed using JMP 4 (1989) software. Shapiro-Wilks normality test was used to describe the di stribution of data. When appropriate, data were log transformed for further analysis. Mahalanobis distance was used to identify and remove extreme outliers. Matched pairs t-tests were used to determine differences between the core and edge sampling locations within wetland s. OBriens test was used to determine if there was equal variance be tween the populations. Popula tions with equal variance were compared using a standard t-test. P opulations with unequal variance, or non-normal distributions, were compared using a Welch ANOVA test for unequal variance. An alpha level of 0.05 was used as a threshold for de termining when differences were significant. When a significant difference did not exis t between treatments, a power test was applied. Power addresses Type II errors, in whic h there is a failure to reject a false null hypothesis (Rotenberry and Wien s 1985). When a significant difference is not found, as indicated by a high p value, it is often assumed that ther e is no difference between the populations compared. However, there may be differences that were not expressed due

PAGE 43

31 to the limited number of samples compared. Power can be used to determine the probability of finding a significant difference. As the probability of significant differences increases, so does the power. Included in the power test is the Least Significant Number (LSN). The LSN is define d as the number of observations needed to decrease the variance enough to achieve a sign ificant result with the given values of significance level, standard de viation of the error, and e ffect size (JMP 4 1989 Help Files).

PAGE 44

32 CHAPTER 3 RESULTS AND DISCUSSION Water column, litter, and soil data from 103 minimally impaired wetlands in the southeastern United States were analyzed. The goals were to characterize nutrient conditions within these wetlands and determ ine whether differences within wetlands, among wetland types, and between USEPA Nutrie nt Ecoregions were present. Results and discussion of findings will be presented in three separate sections: within wetland variability, variability among wetland types, and spatial variability. There are several ways to aggregate the surveyed wetland data based on the question of interest (Table 3-1). Aggreg ating by hydrologic connec tivity allows for a comparison of riverine and non-riverine system s, whereas aggregati ng by vegetative type allows for a comparison between marshes a nd swamps. The most specific aggregation integrates both hydrologic connectivity and ve getative type resulti ng in four separate wetland community types; riverine swamps, non -riverine swamps, riverine marshes, and non-riverine marshes. Aerial coverage of the f our wetland community types wa s not evenly distributed throughout the study area (Figure 3-1). Swamps were more prevalent than marshes, and riverine marshes were practically non-existent in the northern and western extents of the study area. Wetland community types were sampled in proporti on to their relative abundance; therefore, there are unequal samp le sizes for each wetland community type. It is important to keep the unequal distri bution in mind when comparing the various aggregations of wetlands. For example, 83% of the surveyed wetlands are swamps.

PAGE 45

33 Comparisons based on hydrologic connectivity are biased towards riverine and nonriverine swamps, since only 17% of the syst ems compared were marshes. Similarly, 64% of the surveyed wetlands are riverine system s, which may influence distinctions between marshes and swamps. Table 3-1. Various aggregations of the wetlands surveyed in this study Grouping Criteria Aggregation Number Surveyed None All Wetlands Combined 103 Riverine 66 Hydrologic Connectivity Non-riverine 37 Marsh 18 Vegetative Type Swamp 85 Non-riverine Swamp 27 Riverine Swamp 58 Non-riverine Marsh 10 Wetland Community Type Riverine Marsh 8 The surveyed wetlands are not only une qual in abundance, but also in regional distribution (Figure 32). The surveyed wetlands ar e distributed throughout four southeastern states, which in clude three USEPA Nutrient Ec oregions (Figure 2-1). The Southeastern Forested Plain contained 62% of the surveyed swamps, 50% of the marshes, 67% of the riverine, and 47% of the non-rive rine wetlands. The Southern Coastal Plain had 25% of the swamps, 44% of surveyed mars hes, 21% of the riverine, and 42% of the non-riverine systems. The least represented ecoregion was the Eastern Coastal Plain with only 13% of the swamps, 5% of the marshes, 12% of the riverine, and 11% of the nonriverine wetlands. Compar isons among wetland types (aggregated by hydrologic

PAGE 46

34 10,000 20,000 30,000 40,000 50,000 60,000 70,000 80,000 90,000 Non-riverine Swamp Riverine Swamp Non-riverine Marsh Riverine MarshArea (hectares) Oconee Apalachicola Ocala Osceola Oconee Sumter Francis Marion Figure 3-1.Total area of the four wetland t ypes within seven of the surveyed national forests Figure 3-2.Percentage distribution of surveyed wetlands within ecoregions, aggregated by vegetation type (swamps and marshes) and by hydrologic connectivity (nonriverine and riverine). Marsh Riverine Non-riverine Swamp Southeastern Forested Plain Southern Coastal Plain Eastern Coastal Plain Marsh Riverine Non-riverine Swamp Swamp Southeastern Forested Plain Southern Coastal Plain Eastern Coastal Plain Southeastern Forested Plain Southern Coastal Plain Eastern Coastal Plain

PAGE 47

35 connectivity or vegetative type) combined sites in all three ecoregions. Results may be biased by characteristics of the Southeastern Forested Plain since the wetlands are not equally distributed among the three ecoregions. Within Wetland Variability Surveyed wetlands were sampled along two tr ansects. One transect was located in the core area of the wetland while the other was parallel to the upland edge of the wetland. Specific locations of sampling transe cts within each wetland are detailed in Chapter 2 (Figure 2-3). Before comparisons were made among wetland types, possible within wetland variability was investigated. Samples collected within the core area and from the edge area within each wetland were compared using pair-wise analysis. A comparison of physical and chemical attributes between the core a nd edge transects was evaluated for water column, litter, and soil strata. Water column Core areas were significantly deeper (p<0.05) than edge areas for all aggregations of wetlands compared (all wetlands comb ined, swamps, marshes, riverine, and nonriverine). There were no si gnificant differences in water co lumn temperature, dissolved oxygen saturation, pH, or conductivity between core and edge sites (p>0.05). Many of the surveyed wetlands were narrow linear syst ems with short distan ces between the core and edge areas. Therefore, similar water chemistry and physical characteristics within the core and edge areas are not surprisi ng since the water is probably well mixed. Water was not always present within each z one of the wetland (core and edge), and some wetlands had no standing water at th e time of sampling. Only 52 of 103 sampled wetlands had water within both zone s. Of these 52 wetlands, 34 were swamps

PAGE 48

36 and 18 were marshes. There were 26 rive rine and 26 non-riverine systems with water present in each zone. Table 3-2. Results of pair-wise comparison of core and edge areas for various aggregations of surveyed wetlands. “n ” represents the number of wetlands compared, and “p” is the probability value from the pair-wise comparison. Significant differences (p<0.05) are denoted by bold values. Water column TP Water column TN Grouping Criteria Aggregation n p n P None All Wetlands Combined 50 0.017e 97 0.109 Marsh 17 0.024e 15 0.806 Vegetative Type Swamp 33 0.188 82 0.111 Non-riverine 25 0.104 34 0.181 Hydrologic Connectivity Riverine 25 0.09 63 0.211 Non-riverine Swamp 13 0.808 26 0.023c Non-riverine Marsh 12 0.010e 8 0.903 Riverine Swamp 20 0.079 56 0.222 Wetland Community Type Riverine Marsh 5 0.88 7 0.783 e significantly greater values in edge areas c significantly greater values in core areas A nutrient comparison of core and edge sa mples within these wetlands (Table 3-2) indicates that the edge sites had significantl y higher water column total phosphorus (TP) concentrations (0.132 + 0.147 mg/L) than core sites (0.098 + 0.147 mg/L). Total Nitrogen (TN) was also greater at edge th an core sites, but the difference was not significant (p=0.109). Water column TP and TN of core and edge sites were also compared for various aggregations. Edge locations had significantly greater water column TP for three groupi ng strategies: all wetlands combined, marshes, and nonriverine marshes. Elevated water column TP values in edge samples may indicate that

PAGE 49

37 nutrients are being introduced to wetlands fr om adjacent uplands or there is increased mineralization of nutrients at the shallower edge sites. Litter Litter at edge sites had significantly great er total carbon (TC) content and similar TP and TN values compared to litter at core sites. Litter of core and edge sites was compared for the various aggregations of wetlands (Table 3-3). Table 3-3. Results of pair-wise comparison of core and edge areas for various aggregations of surveyed wetlands. “n ” represents the number of wetlands compared, and “p” is the probability value from the pair-wise comparison. Significant differences (p<0.05) are denoted by bold values. Litter TC Litter TN Litter TP Grouping Criteria Aggregation n p n p n p None All Wetlands Combined 90 0.023e 97 0.109 83 0.799 Marsh 15 0.58 15 0.806 14 0.486 Vegetative Type Swamp 82 0.012e 82 0.111 69 0.783 Non-riverine 34 0.771234 0.181 29 0.487 Hydrologic Connectivity Riverine 63 0.017e 63 0.211 54 0.378 Non-riverine Swamp 26 0.526 26 0.023C 20 0.821 Non-riverine Marsh 8 0.575 8 0.903 8 0.753 Riverine Swamp 56 0.014e 56 0.222 48 0.852 Wetland Community Type Riverine Marsh 7 0.883 7 0.783 5 0.038C e significantly greater values in edge areas c significantly greater values in core areas Edge locations had significantly greater litter TC content for all wetlands combined, swamp vegetative type, riverine hydrologic regime, and riverine swamp community type. One possibility for the hi gher carbon content at the edge of these wetlands is that core wetland areas (within ri verine systems) are adjacent to the stream channel. Therefore, the core is more suscep tible to high velocity fl ow, which can transfer

PAGE 50

38 organic matter downstream while depositing i norganic sediments. Inorganic material deposited on leaf litter was of ten integrated into samples from core locations. These deposits may reduce carbon content at core sites. Soil When all wetlands were combined, so il carbon and nitrogen content was significantly greater in core than edge ar eas, whereas phosphorus content was similar within both areas (Table 3-4). Core areas had greater soil TC content for all aggregations and increased TN when comparing all wetla nds combined, swamp vegetative types, nonriverine hydrologic regimes, and riverine marsh wetland communities. Soil TP content was similar between the core and edge ar eas, except within non-riverine swamp communities where core areas had significantl y greater phosphorus content than edge areas. The core areas are significantly deeper than edge sites, which may lead to longer hydroperiods and anaerobic conditions. Unde r anaerobic conditions, decomposition rates are decreased, and levels of N, C, and P can build up in the soil. This may explain the higher levels of these compounds in core ve rsus edge sampling areas in some of the aggregations of surveyed wetlands. Discussion The overall differences within wetlands indicate that sa mples collected at the edge of a wetland will likely have greater water column TP, increased litter TC content, and lower soil TC and TN content than samples collected within the core area of the same wetland. These differences within wetlands suggest potential im plications of inconsistent sampling techniques on biogeochemical character izations of wetlands. To minimize the

PAGE 51

39 effects of within site variability on the findi ngs of this research, only core site values were used in comparisons for the remainder of this study. Table 3-4. Results of pair-wise comparison of core and edge areas for various aggregations of surveyed wetlands. “n ” represents the number of wetlands compared, and “p” is the probability value from the pair-wise comparison. Significant differences (p<0.05) are denoted by bold values. Soil TC Soil TN Soil TP Grouping Criteria Aggregation n p n p n p None All Wetlands Combined 93 0.0001c95 0.023c 94 0.1 Marsh 13 0.0001c141 15 0.6 Vegetative Type Swamp 78 0.0001c79 0.043c 79 0.2 Non-riverine 32 0.0001c32 0.035c 36 0.1 Hydrologic Connectivity Riverine 60 0.0001c610 58 0.7 Non-riverine Swamp 25 0.0001c250 28 0.017cNon-riverine Marsh 6 0.0001c7 1 8 0.8 Riverine Swamp 53 0.0001c540 51 1 Wetland Community Type Riverine Marsh 7 0.0003c7 0.006c 7 0.1 e significantly greater values in edge areas c significantly greater values in core areas Variability among Wetland Types Vegetative Comparisons: Swamps and Marshes Surveyed wetlands can be aggregated by dominant vegetation into swamps and marshes. Swamps are dominated by woody ve getation and include riverine swamps and non-riverine swamps. Swamps were more ubi quitous in the landscape than marshes. Therefore, 85 out of the 103 surveyed wetlands were swamps. Marshes are characterized by herbaceous vegetation and are an aggregate of riverine marshes and non-riverine marshes. Marshes were less common than swamps; therefore, only 18 marshes were surveyed for this study. Differences between swamps

PAGE 52

40 and marshes will be addressed in the following four sections: water column, litter, soil, and discussion. Water column Swamps had significantly greater (p=0.0487) water column TP concentrations compared to marshes (Table 3-5). Swam ps exhibited slightly higher (p=0.5401) TN values, but the trend was not significant (Figur e 3-3). Water TP data partially support the stated hypothesis that marshes would have lo wer water column nutrients compared to swamps. This difference may be correlated with increased presence of algae in marshes compared to swamps. Algae were present in 47% of surveyed ma rshes and only 10% of surveyed swamps. Algae can quickly sequest er water column P, hence lowering water column TP in marshes (Kadlec and Knight 1996). Table 3-5. Water column properties observed in minimally impaired wetlands aggregated by vegetative type Swamp Marsh Parameters mean + SD median 75th n significancemean + SD median 75thn TP (mg/L) 0.108 + 0.12 0.06 0.177 47 0.049 + 0.053 0.03 0.07 17 TN (mg/L) 2.24 + 1.44 1.88 2.79 48 1.82 + 0.64 1.76 2.31 18 Temp (C) 21.9 + 3.0 21.9 24.3 36 ** 25.9 + 4.4 25.3 29.5 14 pH 4.9 + 1.1 4.9 5.9 36 5.3 + 1.1 5.2 6.4 14 DO (%) 28.2 + 21.1 24.3 42.3 36 38.1 + 24.8 39.4 55.7 14 Cond. (uS/cm) 69 + 49 68 82 36 54 + 39 47 89 12 Eh (mv) 412 + 348 397 513 26 369 + 380 318 525 11 Depth (cm) 16.5 + 16.3 14 22.6 42 ** 41.6 + 27.7 47.5 63.5 15 Significant difference (p<0.05) ** Significant difference (p<0.01) *** Significant difference (p<0.001) Water column temperatures were significantly greater in marshes than swamps. This is most likely due to the absence of an overstory of woody vegetation in marshes, which may also contribute to increased mars h algal growth. The core areas of marshes

PAGE 53

41 were significantly deeper than those of swamps. No signi ficant differences were found between swamps and marshes with respect to water column pH, dissolved oxygen saturation, conductivity, or oxi dation reduction potential. TP (mg/L) 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 Marsh Swamp TN (mg/L) 0 1 2 3 4 5 6 7 8 9 Marsh SwampA A B A Figure 3-3. Water column TP and TN values by vegetative type. The dashed line is the mean of each population, and the solid lin e is the overall mean. The bottom of the “box” is the 25th percen tile, and the top is the 75t h percentile. The center line within the boxplot is the median. The whiskers extend + 1.5 the interquantile range. Different letters indicate a signif icant difference (p<0.05) between marshes and swamps. Litter Litter of swamps and marshes had similar phosphorus, nitrogen, and carbon content (Figure 3-4), with no significant differences between vegetative type s evident. The C:P and C:N ratios also did not differ between th e two vegetative types. These findings agree with the literature review of Bedford et al. (1999) th at slightly higher litter N concentrations are found in marshes (1.22%) than in swamps (1.04%). There was a similar trend in this study, with aver age litter TN concentrations of 1.43 + 0.44 % for marshes and 1.25 + 0.29% for swamps.

PAGE 54

42 Figure 3-4 Litter phosphorus, nitrogen, and carbon values by community type. The dash ed line is the mean of each population, an d the solid line is the overall mean. The bottom of the “box” is the 25th percentile, and the top is the 75th percentile. The center line within th e boxplot is the median. The whis kers extend + 1.5 the in terquantile range. Diffe rent letters indicate a significant difference (p<0.05) between marshes and swamps. TN (%) 0.5 1 1.5 2 TC(%) 20 25 30 35 40 45 50 55 TP (%) 0 0.01 0.02 0.03 0.04 0.05 0.06 Marsh Swamp Marsh Swamp Marsh Swamp A A A A A A TN (%) 0.5 1 1.5 2 TN (%) 0.5 1 1.5 2 TC(%) 20 25 30 35 40 45 50 55 TC(%) 20 25 30 35 40 45 50 55 TP (%) 0 0.01 0.02 0.03 0.04 0.05 0.06 Marsh Swamp Marsh Swamp Marsh Swamp A A A A A A

PAGE 55

43Table 3-6. Litter phosphorus, nitrogen, and car bon content observed in minimally impair ed wetlands aggregated by vegetative typ e Swamp Marsh Parameters mean + SD median 75th n significancemean + SD median 75th n P (%) 0.015 + 0.01 0.011 0.02169 0.019 + 0.018 0.012 0.033 15 N (%) 1.25 + 0.29 1.22 1.44 82 1.43 + 0.44 1.33 1.8 15 C (%) 41.0 + 8.5 43.2 48.8 84 41.5 + 4.8 43.1 45.3 15 C/P ratio 4132 + 2958 3258 6838 67 5193 + 5021 3393 8316 14 C/N ratio 32.71 + 9.87 30.78 39.0682 28.90 + 8.50 26.09 35.59 13 Significant difference (p<0.05) ** Significant difference (p<0.01) *** Significant difference (p<0.001)

PAGE 56

44 Figure 3-5 Soil %P, %N, and %C values by community type. The da shed line is the mean of each population, and the solid line is the overall mean. The bottom of the “box” is th e 25th percentile, and the top is the 75th percentile. Th e center line within the boxplot is the median. The whiskers exte nd + 1.5 the interquantile rang e. Different letters indi cate a significant differenc e (p<0.05) between treatments TP (mg/kg) 0 200 400 600 800 1000 1200 Marsh Swamp TN (g/kg) 0 5 10 15 20 25 30 35 Marsh Swamp TC (g/kg) 0 100 200 300 400 500 Marsh SwampA A A A A A TP (mg/kg) 0 200 400 600 800 1000 1200 Marsh Swamp TN (g/kg) 0 5 10 15 20 25 30 35 TN (g/kg) 0 5 10 15 20 25 30 35 Marsh Swamp TC (g/kg) 0 100 200 300 400 500 Marsh SwampA A A A A A

PAGE 57

45Table 3-7.Soil P, N, and C content observed in minima lly impaired wetlands aggregated by vegetative type Swamp Marsh Parameters mean + SD median 75th n significance mean + SD median 75th n TP (mg/kg) 410 + 260 350 550 82 370 + 270 340 420 15 TP (mg/cm3) 0.19 + 0.15 0.14 0.25 82* 0.13 + 0.12 0.093 0.17 15 TN (g/kg) 5.9 + 5.5 3.7 7.4 80 8.5 + 10.4 4.6 10.5 14 TN (mg/cm3) 2.01 + 0.77 1.87 2.39 80 2.04 + 1.16 1.77 3.1 14 TC (g/kg) 123 + 140 64.3 129 78 136 + 156 61.7 194 14 N/P ratio 17.6 + 17.7 14.4 22.8 79 28.2 + 27.9 17.6 42.2 14 C/P ratio 361 + 329 277 514 77 489 + 456 304 970 14 C/N ratio 19.4 + 5.5 18.6 22 78 17.7 + 4.7 17.7 18.7 14 LOI (%) 28.5 + 26.9 16.8 32.9 82 31.7 + 31.3 15.8 69.2 15 Bulk Density (g/cm3) 0.57 + 0.35 0.57 0.83 85 0.45 + 0.32 0.43 0.67 16 Moisture Content (%) 53 + 21 50 73 85 63 + 23 65 86 16 Significant difference (p<0.05) ** Significant difference (p<0.01) *** Significant difference (p<0.001)

PAGE 58

46 Soil Comparisons of soil nutrient content betw een vegetative community types were conducted on a mass per unit mass basis and a mass per unit volume ba sis. Typically it would not be necessary to express nutrient content using these two methods. However, due to the wide range in bulk density f ound among wetland sites, normalizing for bulk density was desirable. Results of findings using both mass and volumetric measures of P and N are presented and dis tinguished based on units. Soils of swamps and marshes were simila r with respect to TP(mg/kg), TN (g/kg), TN (mg/cm3), TC(mg/kg), N:P ratio, C: P ratio, C:N ratio(on a mass basis), loss on ignition (LOI), bulk density, and moisture content, with no significant differences between vegetative types evident (Figure 3-5) However, when soil total phosphorus was normalized by bulk density, swamps had signi ficantly higher total phosphorus (mg/cm3) then marshes. Craft and Casey (2000) found that forested depressions in southwestern Georgia had higher soil nitrogen and phosphorus concen trations, as well as lower N:P ratios, compared to depressional marshes. In th e current survey, soil TP (mg/cm3) was significantly greater in swamps than mars hes, but soil TP (mg/kg), TN (g/kg), TN (mg/cm3), or N:P ratio (Table 3-7) differences were not significant between marshes and swamps. The mean N:P ratio was 18 for swam ps and 28 for marshes. These values are similar to the N:P ratios Craft and Casey ( 2000) reported for swamps that were thought to be p-limited or co-limited by P and N. Thes e results partially suppor t the observations of Hopkinson (1992) that growth form of dominant vegetation does not seem very important in controlling nutrient regimes.

PAGE 59

47 Discussion Phosphorus and nitrogen data for surveyed swamps and marshes were compared to values in the literature (Table 3-8). Phosphor us values were consistently greater in the literature. This could be due to the least im paired status of wetla nds included in this survey. Nitrogen values were fairly c onsistent between the current study and the literature. Table 3-8. Values from the current study co mpared to those in the literature Current Studya Bedford et al. 1999b Nicholson 1995c Whigham and Richardson 1988d Water TP (mg/L) 0.049 + 0.053 0.248 0.520 Litter %P 0.019 + 0.018 0.16 Marsh Soil %P 0.037 + 0.027 0.25 Water TP (mg/L) 0.108 + 0.12 0.221 0.650 Litter %P 0.015 + 0.01 0.16 Swamp Soil %P 0.041 + 0.026 0.09 0.24 Water TN (mg/L) 1.819 + 0.636 2.09 2.67 Litter %N 1.43 + 0.44 1.22 Marsh Soil %N 0.85 + 1.04 1.41 Water TN (mg/L) 2.243 + 1.444 2.17 3.01 Litter %N 1.25 + 0.29 1.04 Swamp Soil %N 0.59 + 0.56 1.28 1.5 a = Mean values with standard deviations for 103 minimally impaired wetlands within the southeastern US b = Mean values from a literature search of North American freshwater temperate wetlands c = Range of values for wetlands with in Elk Island National Park, Alberta d = Mean values from Acer rubrum swamps in Maryland, USA The vegetative structure of marshes and sw amps is quite different, with the former characterized by herbaceous vegetation and th e latter by woody growth forms. It seems logical that wetlands with di fferent vegetation types woul d have soils with varying nutrient contents. According to Hopkinson (199 2), swamps store a great deal of C, N,

PAGE 60

48 and P in plant biomass, while marshes store th e majority of C, N, and P in the soil. Therefore, it was a hypothesis of this thesis th at marshes would have greater soil nitrogen and phosphorus content than swamps. The re sults of this survey do not support this hypothesis. It was found that swamps had significantly greater TP (mg/cm3) than marshes. These results could be influenced by the fact that 64% of the surveyed wetlands are riverine systems which are often associated with higher nutrient concentrations. There is a lot of variability within the da ta for swamps and marshes. The standard deviation values are often as great as the mean values. Th is trend exists not only when wetlands are aggregated by vegetation t ype, but also when all 103 wetlands are combined. This large variab ility in nutrient content among wetlands likely explains why there were minimal significant differences between swamps and marshes for the limited number of sites surveyed. An analysis of statistical power was used to understa nd the limited statistical differences detected between vegetative comm unity types. Power analysis (JMP 1989) was applied to comparisons between swamps and marshes to determ ine if the lack of significant differences was due to an insuffici ent number of wetland s compared (Table 39). Water column TP differences were identified, therefore Least Significant Number (LSN) values were fairly low for water column comparisons. This means that if additional sample sites were included in the survey, and the data re tained their current structure, then additional si gnificant differences between th e water columns of marshes and swamps would likely have been detected. In the case of litter and soil parameters (on a mass per unit mass basis), LSN values were very large. Therefore, true differences in soil (on a mass per unit mass basi s) and litter parameters between swamp and marsh sites

PAGE 61

49 are very small and unrealistic to quantify. Litter and soils from marshes and swamps in this survey of minimally impaired wetlands of the southeastern US were similar with respect to P (mg/kg), N (g/kg), and C (g/kg) content. Regional differences may also be affec ting marsh and soil results since wetlands from all three USEPA Nutrient Ecoregions we re combined for comparisons. To explore this possibility, the 52 swamps and 9 marshes in the Southeastern Forested Plain were compared. The results were fairly consistent with those including all three ecoregions. There were no significant TC, TP, or TN differences between the vegetative communities within litter, soil (on a mass per unit mass ba sis), and water column strata. The only discrepancy was that marshes and swamps ha d significantly different water column TP content when wetlands from all three ecore gions were compared. There may not be enough samples to detect differences at th e community type and ecoregion level. However, fairly consistent results indicat e that regional differences are not skewing results. Table 3-9.Power analysis for non-significant parameters within community comparisons Measured Parameters Number compared in current study (n) Least Significant Number (LSN) Water TP 66 123 Water TN 67 143 Litter % P 82 5,439 Litter % N 97 21,645 Litter % C 97 1,936 Litter C/N 96 6,302 Litter C/P 81 22,449 Soil %P 97 409 Soil %N 94 5,276 Soil %C 92 480,226 A major difference between swamps and ma rshes is the presence of a canopy in swamps. Canopy cover can limit light penetr ation and reduce algal populations. Battle

PAGE 62

50 and Golladay (2001) found that sedge marshe s had higher algal growth than cypress swamps. Algal populations quickly sequester nutrients from the water column, hence decreasing soluble nutrients available to ot her growth forms (Kadlec and Knight 1996). Therefore, it was hypothesized that the wate r column of marshe s would contain less nitrogen and phosphorus than that of swamps. The results of this study partially support this hypothesis. Water column TP was si gnificantly greater in swamps, but TN was similar regardless of dominant vegetation ty pe. Large variation between wetlands and the limited number of samples are likely respon sible for the lack of statistical differences detected between water column TN of swamps and marshes. Power analysis indicated that fewer than one hundred additional samples may be sufficient to detect significantly greater water column TN content in swamps compared to marshes. If swamps have greater wa ter column nutrient content, they may not be removing nutrients from the water column as effec tively as marshes. Wetlands are commonly valued as nutrient sinks; however, it may be necessary to distinguish by vegetative type when assigning this value to wetlands. Differences in nutrient cycling may have implications for aquatic ecosystems downstr eam, in that marshes may retain more phosphorus and nitrogen than swamps. Distin ctions between swamps and marshes may be necessary for determining water column based numeric nutrient criteria for wetlands. Water column total nitrogen and tota l phosphorus and soil total phosphorus (mg/cm3) concentrations appear to be the most sensitive parameters to differences between marshes and swamps. Water colu mn nutrients can be overly sensitive indicators. For example, if water is samp led following a rain event, nutrients may be diluted. When wetlands are sampled on diffe rent days (or even different seasons),

PAGE 63

51 comparisons between them may be confused by parameters (such as rain events) that are not factored into comparisons. Soil based indicators integrate c onditions over a longer period of time and are not easily influenced by sampling conditions. Soil and/or waterbased numeric nutrient criteri a in the southeastern United States may necessitate a distinction between vege tative wetland types. Hydrologic Comparisons: Ri verine and Non-riverine The surveyed wetlands can be aggregated by hydrologic connectiv ity into riverine and non-riverine systems. Riverine wetlands are adjacent to streams, and non-riverine systems are located at least 40-meters from adjacent water bodies. Riverine wetlands were more common; therefore, 64% of the surveyed wetlands were riverine and 36% were non-riverine. Riverine systems includ e riverine marshes and riverine swamps, while non-riverine systems include non-riveri ne swamps and non-riverine marshes. Differences between riverine and non-riverine wetlands will be addressed in the following four sections : water column, litter, soil, and discussion. Water Column Comparisons based on hydrologic connectivity showed that riverine systems (Table 3-10) had significantly greater water column pH and lower oxidation reduction potentials (Eh) than non-riverine systems. Since Eh and pH are the dominant chemical factors influencing nutrient transformations within wetlands (Reddy and D’Angelo 1994), one would expect to see different nutrient si gnatures dependent on hydr ologic connectivity. However, these differences may be minimal since surveyed riverine and non-riverine wetlands were still considered acidic (ave rage pH<7.0) and had mean Eh values (> 300mV) indicating aerobic condi tions (Reddy and D’Angelo 1994) in the water column.

PAGE 64

52 No significant differences were found be tween riverine and non-riverine water column temperature, dissolved oxygen per cent saturation, conductivit y, or water depth. The presence of algae was significantly great er in non-riverine systems than riverine systems. Algae were noted in 13% of su rveyed non-riverine wetlands and only 5% of riverine systems. It is likely that this di fference is the result of the higher occurrence of non-riverine marsh communities than riverine marsh communities and the higher frequency of algae in marshes (47 %) than that of swamps (10%). Table 3-10.Water column properties obser ved in minimally impaired wetlands aggregated by hydrologic connectivity Riverine Non-riverine Parameters mean + SD median 75th n sig mean + SD median 75th n TP (mg/L) 0.119 + 0.13 0.069 0.193 35 0.075 + 0.087 0.039 0.086 31 TN (mg/L) 2.20 + 1.60 1.81 2.79 36 2.18 + 1.11 1.88 2.51 31 Temp (C) 22.6 + 3.2 23.1 25 24 23.4 + 4.39 22.2 27 26 pH 5.5 + 1.0 5.8 6.3 24 ** 4.6 + 1.1 4.3 5.4 26 DO (%) 31.7 + 18.8 32.1 44.9 24 30.3 + 25.6 18.6 52.3 26 Cond. (uS/cm) 64 + 49 59 82 23 67 + 45 58 83.2 25 Eh (mV) 349 + 356 338 430 18 447 + 342 247 534 19 Depth (cm) 21.1 + 19.8 14.5 38.1 27 24.9 + 24.6 15.2 42.4 30 Significant difference (p<0.05) ** Significant difference (p<0.01) *** Significant difference (p<0.001) Further comparisons between hydrologic classe s (Figure 3-6) indica te that riverine systems had significantly greater water column TP but lacked significantly different TN values when compared to non-riverine syst ems. Water column TP data support the hypothesis that riverine syst ems have at least some hi gher water column nutrient conditions. Riverine wetlands are hydrol ogically connected to adjacent aquatic ecosystems and often integrate a more extensive upstream watershed, which may be a source of nutrients. In contrast, non-riveri ne wetlands often have a smaller and more localized watershed resulting in lower nutrient loading (Craft and Casey 2000).

PAGE 65

53 Figure 3-6.Water column TP and TN values by hydrologic connectivity. The dashed line is the mean of each population, and the solid line is the overall mean. The bottom of the “box” is the 25th percentile, and the top is the 75th percentile. The center line within th e boxplot is the median. The whiskers extend + 1.5 the interquantile range. Different lett ers indicate a signi ficant difference (p<0.05) between treatments. Litter Litter of riverine wetlands had si gnificantly higher phosphorus (p<0.0001) and lower carbon content (p<0.0001) than non-rive rine wetlands (Figure 3-7). Nitrogen content of litter was similar between the tw o systems. Ratios of carbon to nitrogen and carbon to phosphorus followed the carbon c ontent trends, with non-riverine systems having significantly higher ratio s than riverine systems. It is probable that lower litter carbon cont ent within riverine systems is due to hydrologic fluxes of these open systems and tran sport of particulate matter. Watersheds of riverine systems often c ontribute inorganic materials th at are deposited in wetlands during flood events. Litter is often coated in organic and inorgani c materials that were not removed before analyses. It may be that litter in riverine a nd non-riverine systems has similar organic carbon content, but that increased inorganic deposition in riverine systems alters the percenta ge of total carbon relative to non-riverine systems. TN (mg/L) 0 1 2 3 4 5 6 7 8 9 TP (mg/L) 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 A A B A Non-riverine Riverine Non-riverine Riverine TN (mg/L) 0 1 2 3 4 5 6 7 8 9 TP (mg/L) 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 A A B A Non-riverine Riverine Non-riverine Riverine

PAGE 66

54 Soil Soil characteristics of riveri ne and non-riverine systems were also compared (Figure 3-8). Non-riverine systems had si gnificantly greater nitrogen (g/kg and mg/cm3), carbon, N:P, C:N, C:P, LOI, and percent mois ture content than riverine systems. Riverine wetlands had significan tly greater phosphorus (mg/cm3) and bulk density values. Craft and Casey (2000) found that non -riverine forested wetlands had elevated soil TP, organic C, and TN content compared to forested riverine wetlands. Soil total phosphorus results from the curr ent study do not coincide with Craft and Casey’s results. Discussion Hydrologic connectivity of riverine wetlands led to the hypothes is that the water column of riverine wetlands would have highe r nutrient concentra tions compared to nonriverine systems. Results of this survey partially support this hypot hesis. Water column phosphorus was greater in riverine wetlands, but there was no difference in nitrogen content regardless of hydrologic connectiv ity. Increased phosphorus conditions in riverine wetlands are likely due to larger contributing watershe ds relative to non-riverine systems. Power analysis was applied to water column to tal nitrogen data to explore the role of sample size in statistical conclusions. A si gnificant difference is more likely to be detected if approximately 800 additional wetlan ds were included in the survey. The large LSN value indicates that there is not much difference between wa ter column TN of riverine and non-riverine wetlands It is possible that wetlan ds cycle nitrogen similarly regardless of hydrologic connectivity or that water column nitrogen is controlled by factors not captured in this comparison.

PAGE 67

55 Figure 3-7.Litter phosphorus, nitrogen, and carbo n content comparisons between riverine and non-riverine systems. The dashed li ne is the mean of each population, and the solid li ne is the overall mean. The bottom of th e “box” is the 25th percentile, and the top is the 75th percenti le. The center line within the boxplot is the medi an. The whiskers extend + 1.5 the interquantile range. Different letters indi cate a significant difference (p <0.05) between treatments. Table 3-11.Leaf litter properties observed in minimally im paired wetlands aggregated by hydrologic connectivity. Riverine Non-riverine Parameters mean + SD median 75th n significancemean + SD median 75th n %P 0.02 + 0.012 0.046 0.027 55 *** 0.008 + 0.004 0.007 0.01 29 %N 1.24 + 0.28 1.22 1.45 62 1.35 + 0.38 1.31 1.68 34 %C 38.21 + 7.64 39.55 44.55 64 *** 47.96 + 3.37 49.3 50.59 31 C/P ratio 2982 + 2418 1868 4450 54 *** 6984 + 3509 7134 8700 27 C/N ratio 29.9 + 8.6 28.2 34.4 62 ** 36.4 + 10.37 36.9 42 34 Significant difference (p<0.05) ** Significant difference (p<0.01) *** Significant difference (p<0.001) TP (%) 0 0.01 0.02 0.03 0.04 0.05 0.06 TN (%) 0.5 1 1.5 2 TC (%) 20 25 30 35 40 45 50 55 A A B A AB Non-riverine Riverine Non-riverine Riverine Non-riverine Riverine TP (%) 0 0.01 0.02 0.03 0.04 0.05 0.06 TN (%) 0.5 1 1.5 2 TC (%) 20 25 30 35 40 45 50 55 A A B A AB Non-riverine Riverine Non-riverine Riverine Non-riverine Riverine

PAGE 68

56 Figure 3-8.Soil TP and TN values by hydrol ogic connectivity. The dashed line is the mean of eac h population, and the solid lin e is the overall mean. The bottom of the “box” is the 25th percentile, and the top is the 75th percentile. The center line within the boxplot is the median. The whiskers extend + 1.5 the interquantil e range. Different letters i ndicate a significant difference (p<0.05) between treatments. TP (mg/kg) 0 200 400 600 800 1000 1200 Non-riverine Riverine TN (g/kg) 0 5 10 15 20 25 30 35 Non-riverine Riverine TC (g/kg) 0 100 200 300 400 500 Non-riverine RiverineB A B A A A TP (mg/kg) 0 200 400 600 800 1000 1200 Non-riverine Riverine TN (g/kg) 0 5 10 15 20 25 30 35 Non-riverine Riverine TC (g/kg) 0 100 200 300 400 500 Non-riverine Riverine TP (mg/kg) 0 200 400 600 800 1000 1200 TP (mg/kg) 0 200 400 600 800 1000 1200 Non-riverine Riverine TN (g/kg) 0 5 10 15 20 25 30 35 TN (g/kg) 0 5 10 15 20 25 30 35 Non-riverine Riverine TC (g/kg) 0 100 200 300 400 500 Non-riverine RiverineB A B A A A

PAGE 69

57Table 3-12.Soil properties observed in minimally impa ired wetlands aggregated by hydrologic connectivity Riverine Non-riverine Parameters mean + SD median 75th n significance mean + SD median 75th n TP (mg/kg) 390 + 240 330 500 61 410+ 290 370 600 36 TP (mg/cm3) 0.22 + 0.16 0.16 0.29 61*** 0.11 + 0.09 0.08 0.13 36 TN (g/kg) 4.3 + 4.2 2.7 4.9 62*** 10.0 + 8.3 5.4 16.5 32 TN (mg/cm3) 1.86 + 0.82 1.69 2.16 62*** 2.31 + 0.78 2.35 2.83 32 TC (g/kg) 75.6 + 85.2 48.7 84.4 60*** 216.3 + 178.8 115.3 422.9 32 N/P ratio 142 + 18 10.3 17.4 61*** 28.8 + 19.3 23.2 32.7 32 C/P ratio 254 + 261 174.6 325.6 59*** 615 + 381 541 887 32 C/N ratio 17.9 + 4.8 17.6 19.3 60** 21.3 + 5.8 21.0 24.0 32 LOI (%) 19.3 + 17.0 13.4 22.5 61*** 45.6 + 33.7 27.8 82.6 36 Bulk Density (g/cm3) 0.63 + 0.35 0.65 0.87 65*** 0.41 + 0.30 0.39 0.63 36 Moisture Content (%) 0.50 + 0.20 0.46 0.64 65*** 0.64 + 0.22 0.61 0.86 36 Significant difference (p<0.05) ** Significant difference (p<0.01) *** Significant difference (p<0.001)

PAGE 70

58 Riverine systems are open to hydrologic influxes and are reported as sinks for sediment and phosphorus from contributi ng watersheds (Craft and Casey 2000). Therefore, it was hypothesized that riverine systems woul d have higher soil phosphorus and nitrogen content than soils in non-riveri ne wetlands. This hypothesis was partially supported by the data collected. Riverine wetlands had greater to tal phosphorus content when the values were normalized by bulk dens ity. However, non-riverine systems had greater soil nitrogen conten t than riverine wetlands. Alternating anaerobic and aerobic conditi ons are ideal for processing nitrogen through wetlands, since nitrogen loss from wetland soil is limited by nitrification in aerobic zones and ammonium diffusion from anaerobic zones to aer obic zones (Reddy and D’Angelo 1994). It appears that surveyed riverine wetlands are storing less nitrogen in the soil than non-riverine we tlands. Riverine wetlands are subject to pulses of flooding when adjacent streams overflow their banks Sudden flooding followed by recession of floodwaters may create ideal conditions for n itrogen processing, hence lowering nitrogen storages in riverine wetlands. The final hypothesis based on hydrologic di fferences was that riverine wetlands would have increased nutrient le vels in leaf litter compared to non-riverine wetlands. It was thought that riverine syst ems have high nutrient influxes that allow plants to cycle nutrients less efficiently a nd reabsorb fewer nutrients fr om senescing leaves (Hopkinson 1992). Hence, areas with increased nutrient availability produce leaf litter with high nutrient content (Shaver and Melillo 1984). This hypothesi s was partially supported by this study. Riverine wetlands did have higher litter TP c ontent, but TN was similar regardless of hydrologic connectivity.

PAGE 71

59 Power analysis was applied to the litter TN results to determine if the lack of significant difference was due to an insuffi cient number of wetlands compared. A difference between riverine and non-riverine litter TN would likely be detected with approximately 50 additional samples. This i ndicates that sample size is affecting the results. Interestingly, additional sample s would not support the hypothesis, but would show litter in non-riverine systems to have grea ter nitrogen content than litter in riverine wetlands. Regional differences may influence hydrol ogic connectivity results since wetlands from all three USEPA Nutrient Ecoregions we re combined for comparisons. To explore this possibility, the 44 riverine and 17 non-riverine wetlands in the Southeastern Forested Plain were compared. The results were fairly consistent with those including all three ecoregions. The only discrepancy was that riverine and non-ri verine systems had significantly different water co lumn TP content when wetlands from all three ecoregions were compared. This difference was not appare nt within the Southeastern Forested Plain comparisons. Additional samples may be need ed to detect differe nces at the ecoregion level. Comparable results indicate that regional differences are not skewing the noted differences based on hydrologic connectivity. As was found when the data were aggregat ed by vegetative type, there is a lot of variability within the data aggregated by hydrologic connec tivity. Standard deviation values were often as great as the mean va lues. This variability makes it difficult to recommend a single numeric nutr ient criterion for protecting all wetlands. Aggregating by hydrology may reduce some of the variability, since differences were identified between riverine and non-rive rine wetlands for numerous pa rameters (water column TP,

PAGE 72

60 pH, Eh; litter TP, TC, C\P, C\N; soil TP TN, TC, N\P, C\P, C\N, LOI, bulk density, and moisture content). Soil and litter strata appe ar to be the most sensitive to differences between riverine and non-riveri ne wetlands. It may be nece ssary to identify wetlands as riverine or non-riverine for reco mmending numeric nutrient criteria. Spatial Variation The surveyed wetlands were stratified w ithin three USEPA Nutrient Ecoregions (Figure 3-9). The vegetative type and hydrol ogic connectivity of wetlands surveyed were not evenly distributed among the ecoregions (Table 3-13). The types of wetlands sampled essentially reflected the distribution of wetland types within the National Forest being surveyed. A greater percentage of marshes and non-riverine systems were represented within the Southern Coastal Plai n (XII). Only 12% of the surveyed wetlands were located in the Eastern Coastal Plains (XIV). Data were aggregated by ecoregion to e xplore the appropriatene ss of this regional classification as an a priori grouping fo r establishing numeric nutrient criteria. Comparisons among ecoregions were made for all wetlands aggregated together and between specific vegetative and hydrologic groupings. Water column, litter, and soil characteristics were compared. One would expect differences among ecoregions if they represent distinct geographic regions with respect to nutrients. Differences between ecoregions will be addressed in the following f our sections: water column, litter, soil, and discussion.

PAGE 73

61 Figure 3-9.Distribution of wetlands within the three USEPA Nutrie nt Ecoregions aggregated by a) hydrologic connectivity and b) vegetative type. Southeastern Forested Plain (IX) Southeastern Forested Plain (IX) Eastern Coastal Plain (XIV Swamp n=85 Marsh n=18 Southeastern Forested Plain (IX) Southeastern Forested Plain (IX) Eastern Coastal Plain (XIV ) Non-riverine n=37 Riverine n=66

PAGE 74

62 Table 3-13. Number of survey ed wetlands within the thr ee USEPA nutrient ecoregions Grouping Criteria Aggregation Southeastern Forested Plains (IX) Southern Coastal Plain (XII) Eastern Coastal Plain (XIV) None Combined 61 29 12 Riverine 44 14 8 Hydrologic Connectivity Non-riverine 17 15 4 Marsh 9 8 1 Vegetative Type Swamp 52 21 11 Water Column When all wetlands were grouped together, water column total phosphorus and total nitrogen did not differ among the three ecore gions (Table 3-14). However, when aggregated by hydrologic connectivity (Figure 3-10), riverine wetlands in the Southern Coastal Plain (XII) had greater wa ter column TN than those of the Southeastern Forested Plains (IX). There were no detectable water column TN and TP differences among nonriverine wetlands in the three ecoregions. When aggregated by vegetative type, swam ps in the Southern Coastal Plain had greater water column TN content than compar able sites of the Southeastern Forested Plain (Figure 3-11). There were no detectab le water column TN and TP differences among marshes in the three ecoregions. Ho wever, fewer marshes were sampled, and there were no water column data for the one marsh in the Eastern Coastal Plain. Water was not present in this wetland when it was sampled. The Southern Coastal Plain appears to have greater water column TN (statistically significant in swamps and riverine wetlands). Water column total nitrogen differences were not detected in vegeta tive type and hydrologic conne ctivity comparisons in the

PAGE 75

63Table 3-14. Water column descriptive statis tics for surveyed wetlands by ecoregion. S uperscript letters following standard devi ations indicate significance for comparisons made across rows. Different letters i ndicate a significant difference (p<0.05). Southeastern Forested Plains (IX) Southern Coastal Plain (XII) Eastern Coastal Plain (XIV) Parameters mean + SD median 75th n mean + SD median 75th n mean + SD median 75th n TP (mg/L) 0.099 + 0.134 a 0.042 0.096 31 0.09 + 0.09a 0.045 0.140 27 0.120 + 0.087a 0.088 0.217 8 TN (mg/L) 2.02 + 1.69a 1.510 2.150 32 2.40 + 1.15a 2.330 2.860 28 2.80 + 1.82a 2.070 3.940 8 Temp (C) 24.1 + 4.0a 13.60 26.80 21 20.8 + 3.3b 20.2 23.0 14 24.6 + 1.5ab 24.2 25.6 7 pH 5.3 + 0.9a 5.50 6.3 21 4.5 + 1.2a 4.0 5.0 14 5.5 + 1.2a 5.9 6.3 7 DO (%) 35.9 + 24.2a 33.2 53.0 21 31.1 + 21.3a 27.7 52.4 14 23.8 + 17.3a 15.7 42.3 7 Cond. (uS/cm) 61.3 + 52.0a 49.0 82.4 22 70.4 + 58.9a 70.5 126.2 13 103.3 + 59.3a 76.0 130.5 7 Eh (mV) 379 + 353a 346 489 21 481 + 347a 507 575 9 365 + 355a 327 495 6 All Wetlands Combined Depth (cm) 11.1 + 7.6a 9.2 17.4 22 13.2 + 10.5a 9.4 21.2 14 6.2 + 8.1a 3.5 7.3 7 RiverineTP 0.118 + 0.154a 0.042 0.194 19 0.126 + 0.111a 0.074 0.197 12 0.106 + 0.060a 0.088 0.169 4 RiverineTN 1.85 + 1.74a 1.32 1.85 20 2.88 + 1.42b 2.78 3.40 12 1.93 + 0.79ab 2.07 2.61 4 Non-riverineTP 0.070 + 0.094a 0.041 0.082 12 0.063 + 0.073a 0.029 0.086 15 0.134 + 0.116a 0.132 0.239 4 Hydrologic ( m g /L ) Non-riverine TN 2.29 + 1.64a 1.84 2.27 12 2.05 + 0.75a 1.86 2.56 16 3.66 +2.27a 3.11 6.02 4 MarshTP 0.117 + 0.128a 0.082 0.227 7 0.033 + 0.022a 0.021 0.045 11 0 Marsh TN 2.50 + 1.99a 1.79 2.71 8 1.82 + 0.70a 1.73 2.51 11 0 Swamp TP 0.094 + 0.138a 0.040 0.093 24 0.131 + 0.106b 0.095 0.206 16 0.120 + 0.087ab 0.088 0.217 8 Vegetative ( m g /L ) Swamp TN 1.86 + 1.59a 1.32 2.15 24 2.78 + 1.23a 2.77 3.35 17 2.80 + 1.82a 2.07 3.94 8

PAGE 76

64 Figure 3-10.Comparison of ecoregions aggreg ated by hydrology. The dashed line is the mean of each population, and the solid line is the overall mean. The bottom of the “box” is the 25th percentile, and th e top is the 75th percentile. The center line within the boxplot is the median. The whiskers extend + 1.5 the interquantile range. Different lett ers indicate a significant difference (p<0.05) between treatments. sections above. Water column nitrogen concentration appears to be affected by geographic region instead of dominant vegetation or hydrologic connectivity. Water column TN and TP concentrations in the Eastern Coastal Plain (XIV) were not significantly different from the other two ecoregions for any of the aggregation schemes. This may be due to insufficient sa mple size, since there were only 12 surveyed wetlands within this ecoregion. Litter Litter nutrient content was compared among the three ecoregions. The Southern Coastal Plain (XII) had significantly lowe r total carbon and greater total phosphorus content compared to the other ecoregions (Tab le 3-15). Litter total nitrogen content was similar among the three ecoregions. TN (mg/L) 0 1 2 3 4 5 6 7 8 SE. Forested S C o a s t a l E. Coastal TN (mg/L) 0 1 2 3 4 5 6 7 8 9 SE. ForestedS C o a s t a lE. CoastalNon-riverineRiverine AAA AB B ATN (mg/L) 0 1 2 3 4 5 6 7 8 SE. Forested S C o a s t a l E. Coastal TN (mg/L) 0 1 2 3 4 5 6 7 8 9 SE. ForestedS C o a s t a lE. Coastal TN (mg/L) 0 1 2 3 4 5 6 7 8 SE. Forested S C o a s t a l E. Coastal TN (mg/L) 0 1 2 3 4 5 6 7 8 9 SE. ForestedS C o a s t a lE. CoastalNon-riverineRiverine AAA AB B A

PAGE 77

65 Figure 3-11.Comparison of ecoregions aggregat ed by vegetative type. The dashed line is the mean of each population, and the solid line is the overall mean. The bottom of the “box” is the 25th perc entile, and the top is the 75th percentile. The center line within the boxplot is the median. The whiskers extend + 1.5 the interquantile ra nge. Different letters indicate a significant difference (p<0.05) between treatments. When aggregated by hydrologic connectivity and ecoregion, riverine wetlands in the Southern Coastal Plain had significantly lower TC, gr eater TP, and similar TN content than the other ecoreg ions (Figure 3-12). There we re no observed TP, TN, or TC differences among litter of non-riverine wetlands in the three ecoregions. It appears that riverine wetlands are driving the differen ces found when all wetlands are combined. Comparisons were made among the ecore gions when aggregating wetlands by vegetative community. Marshes in the Sout hern Coastal Plain had significantly lower litter TP than marshes in the Southeastern Fo rested Plain (Figure 313). Marshes within all three ecoregions had similar litter TN and TC content. Swamps within the Southern Coastal Plain had significantly lower litter TP (Figure 3-13) and greater TC content TN (mg/L) 0 1 2 3 4 5 6 7 8 SE. Forested S. Coastal TN (mg/L) 0 1 2 3 4 5 6 7 8 9 E. Coastal SE. Forested S. CoastalMarshSwampAB B A A ATN (mg/L) 0 1 2 3 4 5 6 7 8 SE. Forested S. Coastal TN (mg/L) 0 1 2 3 4 5 6 7 8 9 E. Coastal SE. Forested S. CoastalMarshSwampAB B A A A

PAGE 78

66Table 3-15.Litter descriptive statistics for surveyed by Ecoregion. Superscript letters following standard deviations indicate significance for comparisons made acr oss rows. Different letters indicat e a significant difference (p<0.05). Southeastern Forested Plains (IX) Southern Coastal Plain (XII) Eastern Coastal Plain (XIV) Parameters mean + SD median 75th n mean + SD median 75th n mean + SD median 75th n TP (%) 0 .018 + 0.012a 0.014 0.024 44 0.008 + 0.004b 0.007 0.009 25 0.024 + 0.012a 0.022 0.033 12 TN (%) 1.31 + 0.39a 1.25 1.59 56 1.24 + 0.33a 1.26 1.39 26 1.31 + 0.14a 1.33 1.40 11 TC (%) 38.70 + 8.02a 39.30 45.08 57 47.74 + 2.75b 47.97 49.86 25 39.97 + 7.92a 40.97 48.12 12 N/P ratio 136.3 + 192.2a 87.00 149.80 42 177.0 + 86.4b 159.70 202.70 26 78.5 + 53.6a 61.90 85.90 12 C/P ratio 3681 + 4000a 2225 5465 42 7056 + 3482b 6618 9374 26 2532 + 2230a 1470 2676 12 All Wetlands Combined C/N ratio 29.9 + 9.3a 27.70 34.10 56 40.0 + 10.5b 38.40 46.10 26 29.7 + 6.3a 27.40 36.80 12 RiverineTP (%) 0.021 + 0.012a 0.02 0.03 33 0.009 + 0.003b 0.008 0.010 12 0.029 + 0.009a 0.037 0.037 8 RiverineTN (%) 1.25 + 0.37a 1.19 1.45 39 1.23 + 0.29a 1.15 1.42 13 1.36 + 0.13a 1.36 1.43 7 Non-riverineTP (%) 0.008 + 0.004a 0.01 0.01 11 0.007 + 0.004a 0.006 0.008 13 0.011 + 0.005a 0.011 0.017 4 Hydrologic Non-riverine TN (%) 1.46 + 0.42a 1.61 1.81 17 1.25 + 0.38a 1.26 1.37 13 1.22 + 0.13a 1.22 1.34 4 MarshTP (%) 0.024 + 0.09a 0.02 0.04 8 0.005 + 0.002b 0.004 0.008 5 0.033 + NA 0.033 0.033 1 Marsh TN (%) 1.45 + 0.41a 1.46 1.81 8 1.38 + 0.55a 1.30 1.89 6 1.57 + NA 1.57 1.57 1 Swamp TP (%) 0.016 + 0.010a 0.01 0.02 36 0.008 + 0.003b 0.007 0.010 20 0.022 + 0.012a 0.021 0.033 11 Vegetative Swamp TN (%) 1.29 + 0.39a 1.22 1.53 48 1.20 + 0.24a 1.20 1.38 20 1.28 + 0.12a 1.33 1.37 10

PAGE 79

67 Figure 3-12.Comparison of riveri ne wetlands in the three ecore gions. The dashed line is the mean of each population, and the so lid line is the overall mean. The bottom of the “box” is the 25th percentile, and the top is the 75th percentile. The center line within th e boxplot is the median. The whiskers extend + 1.5 the interquantile range. Different letters indicate a si gnificant difference (p<0.05) between treatments. compared to the other two ecoregions. Litter TN content was similar in swamps within all three ecoregions. The Southern Coastal Plain had lower li tter phosphorus content when all wetlands were combined and also for the following a ggregations: riverine systems, swamps, and marshes. Litter nitrogen content was also lower in the Southern Coastal Plain, but the difference was not significant for any of the aggregations. The Eastern Coastal Plain had the greatest litter phosphorus content (signi ficant for all wetlands combined, riverine systems, and swamps). Soil Soil characteristics were compared among the three ecoregions (Table 3-16). When compared without subclassification, soil TN (mg/cm3) and TP (mg/cm3) were Riverine TN (%) 0.5 1 1.5 2 2.5 3 SE. Forested S. Coastal E. Coastal TP (%) 0 0.01 0.02 0.03 0.04 0.05 0.06 SE. Forested S. Coastal E. CoastalRiverine AAAA B A Riverine TN (%) 0.5 1 1.5 2 2.5 3 SE. Forested S. Coastal E. Coastal TP (%) 0 0.01 0.02 0.03 0.04 0.05 0.06 SE. Forested S. Coastal E. CoastalRiverine AAAA B A

PAGE 80

68 Figure 3-13.Comparison of litter total phosphorus among the three ecoregions aggregated by vegetative type. The dashed line is the mean of each population, and the solid line is the overall mean. The botto m of the “box” is the 25th percentile, and the top is the 75th percentile. The center line within the boxplot is the median. The whiskers extend + 1.5 the interquantile range. Different letters indicate a signi ficant difference (p<0.05) between treatments. significantly greater in the Sout hern Coastal Plain than in the Southeastern Forested Plain or Eastern Coastal Plain. However, when comparisons were conducted on a mass per unit mass basis there were no significant di fferences in soil total phosphorus (mg/kg) content among the three ecoregions. When aggregated by hydrology there were no significant soil TP (mg/kg), TP (mg/cm3), TN (g/kg), TN (mg/cm3) or TC (g/kg) differences among non-riverine wetlands in the three ecoregions. However, differences among the ecoregions were apparent with riverine wetlands. The Southe rn Coastal Plain had riverine wetlands with lower TP (mg/cm3) than the other two ecoregions. Th e Southeastern Forested Plain had lower TN (mg/cm3) in riverine systems than the other two ecoregions. Marsh Swamp TP (%) 0 0.01 0.02 0.03 0.04 0.05 0.06 SE. Forested S. Coastal E. Coastal TP (%) 0 0.005 0.01 0.015 0.02 0.025 0.03 0.035 0.04 0.045 SE. Forested S. Coastal E. CoastalA A B A AB B Marsh Swamp TP (%) 0 0.01 0.02 0.03 0.04 0.05 0.06 SE. Forested S. Coastal E. Coastal TP (%) 0 0.005 0.01 0.015 0.02 0.025 0.03 0.035 0.04 0.045 SE. Forested S. Coastal E. CoastalA A B A AB B

PAGE 81

69Table 3-16.Soil descriptive statisti cs for surveyed wetlands aggregated by Ecoreg ion. Superscript letter s following standard de viations indicate significance for comparisons made across rows. Different letters i ndicate a significant difference (p<0.05). Southeastern Forested Plains (IX) Southern Coastal Plain (XII) Eastern Coastal Plain (XIV) Parameters mean + SD median 75th n mean + SD median 75th n mean + SD median 75th n TP (mg/cm3) 0.22 0.17a 0.14 0.27 57 0.096 0.067b 0.075 0.13 27 0.24 0.12a 0.2 0.301 11 TN (mg/cm3) 1.75 0.64a 1.62 2.11 55 2.43 1.05b 2.56 3.34 26 2.43 0.67a 2.34 2.44 11 TC (g/kg) 90.4 + 113.1a 48.5 93.9 54 201.2 + 172.7b 104.7 370 26 125.0 + 141.7ab 53.6 165.9 10 N/P ratio 13.3 + 11.4a 9.9 17.9 55 35.0 + 28.8b 25.3 34.6 25 12.3 + 5.4a 12.7 14.5 11 C/P ratio 266.4 + 273.7a 167.8 354.1 54 669.0 + 397.4b 513.4 936.1 25 246.7 + 206.3a 175.8 14.5 11 C/N ratio 18.2 + 3.6a 18.3 21.2 54 21.3 + 7.5a 19.2 24.3 26 17.7 + 6.0a 15.5 20.5 10 LOI (%) 23.2 + 22.8a 15.3 22.7 57 42.1 + 33.6a 27.7 86.1 27 31.3 + 25.8a 19.3 55.8 11 Bulk Density 0.60 + 0.29a 0.67 0.84 59 0.44 + 0.44b 0.43 0.55 28 0.52 + 0.31ab 0.51 0.87 12 All Wetlands Combined % Moisture 0.50 + 0.21a 0.45 0.66 59 0.65 + 0.20b 0.58 0.85 28 0.57 + 0.21ab 0.55 0.80 12 Riverine TP (mg/cm3) 0.25 0.17a 0.20 0.37 41 0.095 0.055b 0.083 0.13 12 0.30 0.10a 0.30 0.39 7 RiverineTN (mg/cm3) 1.57 0.46a 1.49 1.77 41 2.36 1.25b 2.26 3.47 13 2.63 0.76b 2.39 3.34 7 Non-riverineTP (mg/cm3) 0.12 0.11a 0.084 0.13 16 0.097 0.077a 0.061 0.13 15 0.13 0.038a 0.13 0.15 4 Hydrologic Non-riverine TN (mg/cm3) 2.27 0.82a 2.32 2.65 14 2.50 0.85a 2.71 3.19 13 2.07 0.28a 2.12 2.32 4 MarshTP (mg/cm3) 0.19 0.13a 0.16 0.28 8 0.061 0.034b 0.05 0.096 7 0 Marsh TN (mg/cm3) 1.65 1.13a 1.30 1.95 7 2.43 1.12a 2.92 3.31 7 0 Swamp TP (mg/cm3) 0.22 0.17a 0.14 0.27 49 0.11 0.072b 0.099 0.14 20 0.24 0.12a 0.20 0.30 11 Vegetative Swamp TN (mg/cm3) 1.77 0.55a 1.69 2.14 48 2.43 1.06b 2.36 3.43 19 2.43 0.67b 2.34 2.44 11

PAGE 82

70 Soil data were also aggregated by vegeta tion type and compared among ecoregions. Marsh soil phosphorus and nitrog en concentrations did not sh ow significant differences among the ecoregions when compared on a mass per unit mass basis. When nutrient concentrations were normalized by bulk densit y, marshes in the S outhern Coastal Plain had greater total TN (mg/cm3) and lower total phosphorus (mg/cm3) compared to the Southeastern Forested Plain. There were no soil data for the one marsh within the Eastern Coastal Plain. Swamps within the Southern Coastal Plain had significantly greater soil TN (mg/cm3) and TN (g/kg) content compared to swam ps of the Southeastern Forested Plain. There were no significant di fferences among ecoregions fo r TP (mg/kg) content in swamps. However, the Southern Coastal Plain had significantly lower soil TP (mg/cm3) than the other two ecoregions. Discussion It was hypothesized that there would be regional differen ces in the nutrient regimes of wetlands. Differences among the three ecoregions support this hypothesis. The Southern Coastal Plain (XII) is different from the other two ecoregions (Table 3-17), with greater water column TN, litter TC, soil TN, so il TC, and lower litter TP content. These differences suggest that it is a distinct region with its own nutrient characteristics. However, standard deviations were still hi gh even when wetlands were aggregated by ecoregion and hydrologic connectivity or vege tative type. There is still considerable variability among the aggregated wetlands, i ndicating that the eco regions may be too large to aggregate regional differences among wetland nutrient condi tions appropriately.

PAGE 83

71 Table 3-17.Summary of si gnificant differences (p< 0.05) among the three USEPA Nutrient Ecoregions for the various aggregations of surveyed wetlands. All Wetlands Combined Riverine Nonriverine Marsh Swamp Water column TP = = = = = Water column TN = XII > IX = = XII > IX Litter TN = = = = = Litter TC XII >IX and XIV XII >IX and XIV = = XII >IX and XIV Litter TP XII IX XIV > IX = = XII > IX Soil TC (g/kg) XII > IX = = XII > IX XII > IX Soil TP (mg/kg) = XIV > XII and IX = = = There were no soil TP (mg/cm3) differences between the Southeastern Forested Plain and the Eastern Coastal Plain. However, these ecoregi ons had significantly greater soil TP (mg/cm3) content than the Southern Coastal Plain. The Southeastern Forested Plain was the largest ecoregion surveyed a nd included a few northern Florida sites, all Georgia and Alabama sites, and half of the su rveyed wetlands in South Carolina. Several of these areas are known for their clay mine ral soils. Mineral soils retain phosphorus better than organic soils, due to higher iron and aluminum content (Richardson 1985). Therefore, it is not surprising that the Sout heastern Forested Plain had greater soil TP content than the Southern Coastal Plain. The Eastern Coastal Plain included wetlands within coastal South Carolina. High sedimentation rates in alluvial floodplains are common in this ecoregion. Upland inputs to streams may have resulted in higher phos phorus and nitrogen accumulations in the wetland soils of this ecoregion compared to wetlands in the Sout hern Coastal Plain.

PAGE 84

72 When non-riverine water column, litter, a nd soils were compared among the three ecoregions, there were no significant differe nces. Non-riverine wetlands may be less affected by regional differences because they have smaller contributing watersheds than riverine wetlands. Watershed properties, such as soil types and topography may be driving some of the differences noted between riverine and non-rive rine wetlands in the section above. Variability within ecoregions was explor ed by examining regional differences in wetland nutrient regimes at a scale finer th an the USEPA Nutrient Ecoregions. The Southeastern Forested Plain was subdivided into smaller regions by aggregating the surveyed wetlands by National Forest (or military base). This ecoregion was chosen because it has the largest area and contained mo st of the surveyed wetlands (60%). Fort Benning Military Base, Moody Air Force Base Banks Lake National Wildlife Refuge, Conecuh, Oconee, Sumter, Talladega, along with portions of Apalachicola National Forest are located in the Southeastern Fore sted Plain (Figure 314). Moody Air Force Base and Banks Lake National Wildlife Refuge are adjacent to each other; therefore the two surveyed wetlands from Banks Lake Natio nal Wildlife Refuge were combined with the three surveyed wetlands from Moody Airfor ce Base. These five wetlands are referred to as Moody Air Force Base. Soil total phosphorus (mg/cm3) and nitrogen (mg/cm3) content were compared among these National Forests and military bases within the Southeastern Forested Plain. There were no significant diffe rences regarding soil total nitrogen content. There were, however, significant differences among some of the regions with regards to soil total phosphorus content (Table 3-18) Apalachicola National Fo rest had the lowest TP

PAGE 85

73 Figure 3-14. Distribution of sampling locations within the USEPA Nutrient Ecoregions. 0240480 120KilometersLegend Apalachicola NF Banks Lake NWF Conecuh NF Fort Benning Military Francis Marion NF Moody Air Force Base Ocala NF Oconee NF Osceola NF Sumter NF Talladega NFlN Southeastern Forested Plai n Southern Coastal Plain Eastern Coastal Plain

PAGE 86

74 content while Oconee National Forest had the gr eatest TP content. There is almost an order of magnitude difference between the means of these regions. The USEPA has discussed setting nu meric nutrient criteria at the 75th percentile value of least impaired wetla nds within an ecoregion. Resu lts from this study suggest that the 75th percentile of soil TP in the Southeastern Forested Plain is 0.27 mg/cm3. It is unlikely that the USEPA would adopt this recommendation without additional research. However, if this value was adopted as the numeric nutrient criteria for this ecoregion, wetlands in the Apalachicola area would not be sufficiently protected from nutrient enrichment. The mean soil TP content of these wetlands would have to increase by a factor of five before exceeding the numeri c nutrient criteria. Likewise, wetlands in Sumter and Oconee National Forests already exceed the hypothetical nutrient threshold. It is clear that there are significant regional differences in wetland nutrient regimes at a scale finer than the USEPA Nutrient Ecoregions. Table 3-18.Soil total phosphorus statistics for surveyed we tlands in the Southeastern Forested Plain aggregated by National Forest (or military base). Vertical lines connect means that are not si gnificantly different (p<0.05). Soil TP (mg/cm3) mean SD median 75th n Apalachicola, FL 0.056 0.021 0.048 0.077 5 Moody AB, GA 0.11 0.042 0.09 0.15 5 Conecuh, GA 0.13 0.08 0.089 0.14 7 Oakmulgie, AL 0.15 0.10 0.14 0.14 11 Fort Benning, GA 0.15 0.05 0.14 0.19 8 Sumter, SC 0.38 0.20 0.27 0.51 11 Oconee, GA 0.39 0.12 0.44 0.5 9

PAGE 87

75 CHAPTER 4 CONCLUSIONS Establishing nutrient criteria for wetland ecosystems requires understanding variability in nutrient regimes among wetlands. The primary goal of this study was to use consistent sampling methods to understand b ackground nutrient conditions in some of the least impaired watersheds of the southeas tern US. An additional objective was to contrast results based on ve getative community, hydrologic connectivity, and geographic region. It is hoped that these findings will aid the USEPA in deve loping numeric nutrient criteria for wetlands in this region. One finding from this study stresses the importance of consistent sampling locations within wetlands for surveying a nd/or monitoring programs. Water column, litter, and soil characteristics between the core areas and the edge areas of wetlands demonstrated significant differen ces in some parameters. Samples collected at the edge of a wetland had greater wate r column total phosphorus and l itter total carbon content and lower soil total carbon and total nitrogen cont ent than samples from the core area of the same wetland. These differences within we tlands suggest potential implications of inconsistent sampling techniques on biogeoc hemical characterizations of wetlands. Response of wetlands to nutrient change will likely be partially influenced by vegetative characteristics of the wetland. It was hypothesize d that marshes would have higher soil nutrient concentr ations than swamps. Findi ngs from this survey do not support this hypothesis. It was found that swamps had significantly greater TP (mg/cm3) than marshes. These results could be influe nced by the fact that 64% of the surveyed

PAGE 88

76 wetlands are riverine systems which are often associated with higher nutrient concentrations. An additional hypothesis was that mars hes would have lower water column nutrients than swamps. The results of this su rvey partially support th is hypothesis. Total nitrogen was similar regardless of domin ant vegetation type, and total phosphorus concentrations were significan tly greater in swamps. This difference may be correlated with increased presence of algae in marshe s compared to swamps. Algae can quickly sequester water column P, hence lowering water column TP in marshes (Kadlec and Knight 1996). Litter parameters were similar betw een swamps and marshes, suggesting distinguishing between these two ecosystem types is not necessary for determining numeric nutrient criteria. In cont rast, water column and soil (mg/cm3) total phosphorus differences between swamps and marshes dem onstrate the need to set numeric nutrient criteria specific to dominant vegetative c over. A water column based numeric nutrient criteria may not be the best indicator of wetland nutrient regime. Only 52 of 103 sampled wetlands had water present within the core and edge area of the wetland at the same time. Furthermore, water column nutrients can be overly sensitive indicators, since they are influenced by drought, wind, rain events, and other factors. It is likely that hydrologic connectivity will also affect the response of wetlands to nutrient changes. It was hypothesized that riverine wetlands would have higher soil, litter, and water column nutrien t levels than non-riverine sy stems. The results support some of the hypotheses. Riverine wetlands had greater water column total phosphorus than non-riverine systems, but total nitrogen content was similar. Litter total phosphorus

PAGE 89

77 content was also greater in rive rine systems, but again nitrog en content was similar. The results partially support the hypothesis that riverine wetla nds would have greater soil nutrient levels than non-river ine wetlands. Riverine we tlands had greater soil total phosphorus (mg/cm3), but lower soil total nitrogen (g/kg and mg/cm3) content than nonriverine wetlands. Results indicate that it may be necessary to identify wetlands as riverine or nonriverine in order to assign appropriate numeric nutrient criteria. For example, when riverine and non-riverine wetlands are combined, the soil total nitrogen 75th percentile value is 7.39 g/kg. When aggregated by hydrol ogic connectivity, the value is 4.9 g/kg for riverine systems and 16.5 g/kg for non-riverine systems. If 7.39 g/kg was the numeric nutrient criterion for soil total nitrogen, then the non-riverine system s would be identified as threatened by nutrient enrichment. Howe ver, non-riverine systems appear to have approximately three times the soil total nitr ogen content of riverine systems. Numeric nutrient criteria specific to hydrologic c onnectivity will serve as a more effective threshold for indicating the nutrient status of wetlands than a single criterion for all wetlands combined. The USEPA recognized the importance of regional influences on wetland nutrient regimes when the decision was made to de termine numeric nutrien t criteria specific to ecoregions. Results demonstrate that the Sout hern Coastal Plain (XII) is different from the Southern Forested Plain (IX) and the East ern Coastal Plain (XIV), with greater water column total nitrogen, litter total carbon, so il total nitrogen, soil total carbon, and lower litter total phosphorus content. These differences suggest that it is a distinct region with

PAGE 90

78 its own nutrient characteristics, although va riation was great enough to warrant further investigation. The Southeastern Forested Plain wa s subdivided into smaller regions by aggregating the surveyed wetlands by National Forest (or military base). There were no significant soil total nitrogen (mg/cm3) differences among the sub-regions. However, there were significant differences among some of the regions with regards to soil total phosphorus (mg/cm3) content. There was almost an order of magnitude difference between the extreme regions for soil total phosphor us. It is clear that there are significant regional differences in wetland nutrient regime s at a scale finer than the USEPA Nutrient Ecoregions. If the ecoregions are sub-divided for determin ation of numeric nutrient criteria, the assigned values will more accurately reflect background nutrient concentrations. Surveying additional wetlands will assist in determining appropriate water quality criteria. If similar methods are employed, resu lts can be combined to increase statistical robustness and decrease variab ility. Additional studies should concentrate on regional, vegetative, and hydrologic influenc es on wetland nutrient regimes.

PAGE 91

79 APPENDIX A WETLAND CHARACTERIZATION FORM Wetland ID: Date: Observer Name: Picture ID: Weather Condition: Is the wetland adjacent to a body of water? Circle the appropriate choice: River Stream Lake Estuary Ocean None Characterization for the Entire Wetland (Please circle one of the vegetation classes) 1) Is the vegetation composed predominantly non-vascular (mosses and lichens) ...… Moss-Lichen 2) Is the vegetation herbaceous? i) Is the vegetation dominated by rooted emergent vegetation?..................... Emergent Wetland ii) Is the vegetation predominately submergent, floating-leaved, or free-floating?.... Aquatic Bed 3) Is the vegetation mostly trees and/or shrubs? i) Is it dominated by vegetation less than 6 meters tall? ……………… Scrub-Shrub Wetland ii) Are the dominants 6 meters or greater? …………………………………. Forested Wetland Land-Use Characterization 1) Circle the following land-uses that best characterizes the adjacent upland and estimate the percentage of the area that is represented by the circled land uses: a) Commercial ______ g) Rural (scattered homes) ______ b) Industrial ______ h) Unimproved pasture______ c) Golf course ______ i) Forested or wetland ______ d)High density residential (>20 units/acre) ______ j) Pine plantations ______ e) Low density residential ______ k) Row crops ______ f) Feed lots or Dairy operations ______ l) Other ______ 2) Please circle the following fire indicators present within the vegetation zone: a) Charred ground surface e)Burnt dead trees b) Burnt trees with new shoots f) Burnt crowns of trees c) Burn marks on trees and shrubs g) Burned ground with no understory d) No evidence of fire 3) Is trash present in the wetland?: Yes or No (describe) 4) Is there green algae present in the we tland?: Yes or No (describe) 5) Is there evidence of sedimentation in th e wetland? Yes or No (describe) 6) Is there floating vegetation?: Yes or No (describe) 7) Circle any visible indicators of hydrologic disturbances: a) Ditch e) Dam b) Nearby road impeding flow f) Dyke c) Canals g)Piped inflows d) None noticed h) Other (describe) 8) Circle any visible indicators of vegetative disturbances: a) Large stand of vines e) Cutting or grazing in wetland b) Cutting or grazing in adjacent upland f) Insect damage c) Large stand of exotic species g) Large % of dead trees d) None noticed h) Other (describe) 9) Circle any direct indicators of nutrient loading to the wetland a) Presence of cattle in wetland d) Yard waste dumping in/near wetland b) Fertilizer or manure application in watershed e) None noticed c) Other (describe) 10) What is the approximate size of the wetland: ________________ Shape: ___________________

PAGE 92

80 Vegetation Community Characterizati on Form Sub-sample C (Deep Center) Wetland ID: Date: Start Time: Finish Time: Photo ID: Sub-sample C1 Sub-sample C2 Sub-sample C3 Comments Temp C pH DO % Conductivity ORP Water Depth (inches) Depth of Organic layer (inches) Distance from ground to lichen lines (inches) Algal mats (circle one) Present Not present Present Not present Present Not present Aquatic plants (circle one) Present Not present Present Not present Present Not present Morphological adaptations (circle any that apply) Buttressed roots Adventitious roots Hummocks None Present Buttressed roots Adventitious roots Hummocks None Present Buttressed roots Adventitious roots Hummocks None Present Circle the ONE Characterization that best describes the zone being sampled Emerg. Macrophytes Grasses/sedges Floating aquatics Forested Scrub-Shrub Other: Emerg. Macrophytes Grasses/sedges Floating aquatics Forested Scrub-Shrub Other: Emerg. Macrophytes Grasses/sedges Floating aquatics Forested Scrub-Shrub Other: % cover of overstory List the dominant overstory vegetation within a 10-ft radius of sampling and the % cover they represent % cover of understory List the dominant understory story vegetation within a 10-ft radius of sampling and the % cove r they represent

PAGE 93

81 Vegetation Community Characteriza tion Form Sub-sample E (Edge) Wetland ID: Date: Start Time: Finish Time: Photo ID: Sub-sample E1 Sub-sample E2 Sub-sample E3 Comments Temp C pH DO % Conductivity ORP Water Depth (inches) Depth of Organic layer (inches) Distance from ground to lichen lines (inches) Algal mats (circle one) Present Not present Present Not present Present Not present Aquatic plants (circle one) Present Not present Present Not present Present Not present Morphological adaptations (circle any that apply) Buttressed roots Adventitious roots Hummocks None Present Buttressed roots Adventitious roots Hummocks None Present Buttressed roots Adventitious roots Hummocks None Present Circle the ONE Characterization that best describes the zone being sampled Emerg. Macrophytes Grasses/sedges Floating aquatics Forested Scrub-Shrub Other: Emerg. Macrophytes Grasses/sedges Floating aquatics Forested Scrub-Shrub Other: Emerg. Macrophytes Grasses/sedges Floating aquatics Forested Scrub-Shrub Other: % cover of overstory List the dominant overstory vegetation within a 10-ft radius of sampling and the % cover they represent % cover of understory List the dominant understory story vegetation within a 10-ft radius of sampling and the % cove r they represent

PAGE 94

APPENDIX B WETLAND IDENTIFICA TION AND LOCATION

PAGE 95

83Table B-1. Wetland identif ication and location ID Hydrology CommunityEcoregion Location Longitude Latitude AL1 Riverine Swamp SE Forested Plain (IX) Conecuh NF -86.52833 31.27944 AL10 Riverine Swamp SE Forested Plain (IX) Talladaga NF -87.38222 33.02167 AL11 Riverine Marsh SE Forested Plain (IX) Talladaga NF -87.48722 32.87222 AL12 Riverine Swamp SE Forested Plain (IX) Talladaga NF -87.69389 33.09444 AL13 Riverine Swamp SE Forested Plain (IX) Talladaga NF -87.37389 33.17611 AL14 Riverine Marsh SE Forested Plain (IX) Talladaga NF -87.34556 33.18083 AL15 Riverine Swamp SE Forested Plain (IX) Talladaga NF -87.39139 33.10583 AL16 Riverine Marsh SE Forested Plain (IX) Talladaga NF -87.55278 33.03639 AL17 Riverine Swamp SE Forested Plain (IX) Talladaga NF -87.56583 33.05917 AL18 Riverine Swamp SE Forested Plain (IX) Talladaga NF -87.68639 33.08500 AL19 Riverine Swamp SE Forested Plain (IX) Talladaga NF -87.46028 33.05611 AL2 Riverine Swamp SE Forested Plain (IX) Conecuh NF -86.68444 31.22083 AL20 Riverine Swamp SE Forested Plain (IX) Talladaga NF -87.55833 33.08306 AL3 Riverine Swamp SE Forested Plain (IX) Conecuh NF -86.73472 31.24194 AL4 Non-riverine Marsh SE Forest ed Plain (IX) Conecuh NF -86.57417 31.21528 AL5 Non-riverine Swamp SE Forest ed Plain (IX) Conecuh NF -86.84861 31.14194 AL6 Non-riverine Swamp SE Forest ed Plain (IX) Conecuh NF -86.65667 31.22278 AL7 Non-riverine Marsh SE Forest ed Plain (IX) Conecuh NF -86.85611 31.13000 AL8 Riverine Swamp SE Forested Plain (IX) Conecuh NF -86.75611 31.33972 AL9 Riverine Swamp SE Forested Plain (IX) Talladaga NF -87.32667 32.80056 FL10 Non-riverine Swamp S. Costal Plain (XII) Ocala NF -82.14167 29.44056 Table B-1.Continued ID Hydrology CommunityEcoregion Location Longitude Latitude FL11 Non-riverine Swamp S. Costal Plain (XII) Ocala NF -81.98306 29.35611 FL12 Non-riverine Marsh S. Costal Plain (XII) Ocala NF -81.80000 29.23361 FL13 Non-riverine Marsh S. Costal Plain (XII) Ocala NF -81.84778 29.33861 FL14 Riverine Swamp S. Costal Plain (XII) Ocala NF -81.99472 29.34278 FL15 Riverine Swamp S. Costal Plain (XII) Ocala NF -81.99333 29.34528 FL16 Riverine Marsh S. Costal Plain (XII) Ocala NF -81.81750 29.53917

PAGE 96

84FL17 Non-riverine Swamp S. Costal Plain (XII) Ocala NF -81.80528 29.42778 FL18 Riverine Swamp S. Costal Plain (XII) Ocala NF -81.67722 29.09278 FL19 Riverine Swamp S. Costal Plain (XII) Ocala NF -81.68139 29.08500 FL20 Non-riverine Marsh S. Costal Plain (XII) Ocala NF -81.79250 29.25861 FL21 Riverine Swamp S. Costal Plain (XII) Osceola NF -82.66861 30.34944 FL22 Non-riverine Swamp SE Forest ed Plain (IX) Osceola NF -82.75167 30.31722 FL23 Riverine Marsh S. Costal Plain (XII) Ocala NF -81.73083 29.43250 FL24 Non-riverine Marsh S. Costal Plain (XII) Apalachicola NF -84.48167 30.40944 FL25 Non-riverine Marsh SE Forested Plain (IX) Apalachicola NF -84.68611 30.48611 FL26 Non-riverine Swamp S. Costal Plain (XII) Apalachicola NF -84.58944 30.36417 FL27 Riverine Swamp S. Costal Plain (XII) Apalachicola NF -84.59472 30.36333 FL28 Non-riverine Swamp SE Forested Plain (IX) Apalachicola NF -84.63694 30.53778 FL29 Riverine Swamp Apalachicola NF FL30 Non-riverine Swamp Apalachicola NF FL31 Riverine Swamp SE Forested Plain (IX) Apalachicola NF -84.84000 30.35444 FL32 Non-riverine Marsh S. Costal Plain (XII) Apalachicola NF -84.89139 30.23194 FL33 Riverine Swamp S. Costal Plain (XII) Apalachicola NF -85.15889 30.18500 FL34 Non-riverine Swamp SE Forested Plain (IX) Apalachicola NF -84.73111 30.35556 FL35 Non-riverine Swamp SE Forested Plain (IX) Apalachicola NF -85.01222 30.36778 FL36 Non-riverine Swamp S. Costal Plain (XII) Osceola NF -82.48417 30.35389 FL37 Riverine Swamp S. Costal Plain (XII) Osceola NF -82.62333 30.52083 FL38 Non-riverine Marsh S. Costal Plain (XII) Osceola NF -82.64639 30.62028 FL39 Non-riverine Swamp S. Costal Plain (XII) Osceola NF Table B-1.Continued ID Hydrology Community Ecoregion Location Longitude Latitude FL40 Riverine Swamp S. Costal Plain (XII) Osceola NF -82.66806 30.47111 FL41 Non-riverine Swamp S. Costal Plain (XII) Osceola NF -82.59639 30.46472 FL42 Riverine Swamp S. Costal Plain (XII) Osceola NF -82.57417 30.40667 FL43 Non-riverine Swamp S. Costal Plain (XII) Osceola NF -82.60306 30.48167 FL44 Riverine Swamp S. Costal Plain (XII) Osceola NF FL45 Non-riverine Swamp S. Costal Plain (XII) Osceola NF -82.58417 30.31583

PAGE 97

85GA1 Riverine Swamp SE Forested Plain (IX) Oconee NF -83.54250 33.37333 GA10 Riverine Marsh SE Forested Plain (IX) Oconee NF -83.54639 33.80583 GA16 Non-riverine Swamp SE Forested Plain (IX) Grand Bay NWF -83.42917 30.99750 GA17 Non-riverine Swamp SE Forested Plain (IX) Grand Bay NWF -83.29778 31.09556 GA19 Non-riverine Swamp SE Fore sted Plain (IX) Moody AFB -83.36667 31.08250 GA2 Riverine Swamp SE Forested Plain (IX) Oconee NF -83.77556 33.39028 GA20 Non-riverine Swamp SE Fore sted Plain (IX) Moody AFB -83.27889 30.97083 GA21 Non-riverine Swamp SE Fore sted Plain (IX) Moody AFB -83.28861 31.03861 GA25 Riverine Swamp SE Forested Pl ain (IX) Fort Benning MB -84.91056 32.43806 GA26 Riverine Swamp SE Forested Pl ain (IX) Fort Benning MB -84.90056 32.44694 GA27 Riverine Swamp SE Forested Pl ain (IX) Fort Benning MB -84.88167 32.59917 GA28 Riverine Swamp SE Forested Pl ain (IX) Fort Benning MB -84.92556 32.50000 GA29 Riverine Swamp SE Forested Pl ain (IX) Fort Benning MB -84.90500 32.41500 GA3 Non-riverine Swamp SE Forest ed Plain (IX) Oconee NF -83.53111 33.38556 GA30 Riverine Swamp SE Forested Pl ain (IX) Fort Benning MB -84.77333 32.40972 GA31 Non-riverine Swamp SE Forested Plain (IX) Fort Benning MB -84.95111 32.72000 GA32 Riverine Swamp SE Forested Pl ain (IX) Fort Benning MB -85.06278 32.60111 GA4 Riverine Swamp SE Forested Plain (IX) Oconee NF -83.82722 33.32361 GA5 Non-riverine Marsh SE Forest ed Plain (IX) Oconee NF -83.86167 33.24444 GA6 Riverine Swamp SE Forested Plain (IX) Oconee NF -84.05972 33.35528 GA7 Riverine Swamp SE Forested Plain (IX) Oconee NF -83.87694 33.48306 GA8 Riverine Swamp SE Forested Plain (IX) Oconee NF -84.06056 33.47056 GA9 Riverine Swamp SE Forested Plain (IX) Oconee NF -83.89250 33.23167 Table B-1.Continued ID Hydrology Community Ecoregion Location Longitude Latitude SC1 Riverine Swamp E. Coastal Pl ain (XIV) Francis Marion NF -79.85333 33.35917 SC10 Riverine Swamp E. Coastal Pl ain (XIV) Francis Marion NF -80.11722 33.25917 SC11 Riverine Swamp E. Coastal Pl ain (XIV) Francis Marion NF -79.73611 33.39028 SC12 Riverine Swamp E. Coastal Pl ain (XIV) Francis Marion NF -79.46889 33.35361 SC13 Riverine Swamp SE Forested Plain (IX) Sumter NF -81.70639 34.45361 SC14 Riverine Swamp SE Forested Plain (IX) Sumter NF -81.53806 34.56056

PAGE 98

86SC15 Riverine Swamp SE Forested Plain (IX) Sumter NF -81.81222 34.63000 SC16 Riverine Swamp SE Forested Plain (IX) Sumter NF -81.67611 34.75056 SC17 Riverine Swamp SE Forested Plain (IX) Sumter NF -81.88556 34.77583 SC18 Riverine Marsh SE Forested Plain (IX) Sumter NF -81.97722 34.66667 SC19 Riverine Swamp SE Forested Plain (IX) Sumter NF SC2 Riverine Swamp E. Coastal Pl ain (XIV) Francis Marion NF -79.99917 33.55389 SC20 Riverine Swamp SE Forested Plain (IX) Sumter NF -81.78639 34.56167 SC21 Riverine Swamp SE Forested Plain (IX) Sumter NF -81.73833 34.47556 SC22 Riverine Swamp SE Forested Plain (IX) Sumter NF -81.39694 34.47028 SC23 Riverine Swamp SE Forested Plain (IX) Sumter NF -81.58944 34.43167 SC24 Riverine Swamp SE Forested Plain (IX) Sumter NF -81.68750 34.59500 SC3 Non-riverine Swamp E. Coastal Plain (XIV) Francis Marion NF -79.99083 33.43083 SC4 Non-riverine Swamp E. Coastal Plain (XIV) Francis Marion NF -79.95528 33.44806 SC5 Non-riverine Swamp E. Coastal Plain (XIV) Francis Marion NF -79.79750 33.26639 SC6 Riverine Swamp E. Coastal Pl ain (XIV) Francis Marion NF -79.87194 33.17861 SC7 Riverine Marsh E. Coastal Pl ain (XIV) Francis Marion NF -79.91556 33.27750 SC8 Riverine Swamp E. Coastal Pl ain (XIV) Francis Marion NF -79.97083 33.28278 SC9 Non-riverine Swamp E. Coastal Plain (XIV) Francis Marion NF -79.86306 33.21389

PAGE 99

APPENDIX C PHYSICAL SOIL AND WATER COLUMN DATA

PAGE 100

88Table C-1.Physical soil and water column data Water column data is an average of subsample locations within the core (C) or e dge (E) transect. ID Area Soil moisture content (%) Soil bulk density (g /cm3) Soil LOI (%) Water temp (C) Water pH Water DO (%) Water conductivity (uS/cm) Water Eh (mv) Water Depth (cm) AL1 C 39% 0.73 9.7 AL1 E 30% 0.95 6.3 AL10 C 66% 0.39 20.4 22.4 5.31 52.40 18.0 414.0 10.0 AL10 E 80% 0.21 37.9 19.4 5.21 13.70 22.0 342.7 2.3 AL11 C 50% 0.67 10.0 23.6 5.80 14.30 57.7 267.3 15.3 AL11 E 64% 0.41 15.6 23.4 5.57 49.53 46.7 298.5 5.2 AL12 C 59% 0.51 12.9 24.8 5.69 33.17 53.7 5.3 AL12 E 35% 0.99 7.3 22.3 5.54 18.40 38.0 -3.0 AL13 C 46% 0.70 10.3 AL13 E 38% 0.92 7.2 AL14 C 67% 0.41 15.6 23.0 6.45 4.13 133.7 151.7 26.0 AL14 E 67% 0.42 18.2 23.1 6.30 8.00 125.3 175.1 17.3 AL15 C 36% 0.97 23.2 AL15 E 72% 0.31 24.9 AL16 C 73% 0.30 28.4 26.2 5.46 4.77 47.3 276.1 19.0 AL16 E 56% 0.61 10.4 27.7 5.38 33.53 32.7 291.4 10.0 AL17 C 31% 0.72 8.6 AL17 E 33% 0.76 8.5

PAGE 101

89Table C.1.Continued ID Area Soil moisture content (%) Soil bulk density (g /cm3) Soil LOI (%) Water temp (C) Water pH Water DO (%) Water conductivity (uS/cm) Water Eh (mv) Water depth (cm) AL18 C 82% 0.18 34.8 22.6 5.22 45.77 15.0 346.4 5.2 AL18 E 64% 0.40 17.9 -1.0 AL19 E 66% 0.36 19.0 AL2 C 35% 0.97 25.7 5.80 38.07 44.7 8.0 AL2 E 57% 0.52 15.8 25.6 6.38 13.60 197.5 150.8 6.0 AL20 C 24% 0.88 5.8 AL20 E 20% 0.93 5.4 AL3 C 30% 0.72 8.0 AL3 E 40% 0.61 9.8 AL4 C 86% 0.13 77.7 29.2 4.90 29.67 47.0 25.0 AL4 E 77% 0.24 49.3 -3.0 AL5 C 87% 0.12 86.0 21.7 3.68 13.00 46.3 500.7 3.7 AL5 E 61% 0.38 24.4 28.7 4.87 52.00 55.0 6.3 AL6 C 86% 0.13 82.8 27.3 3.70 70.87 83.7 534.2 5.7 AL6 E 61% 0.44 23.0 -3.0 AL7 C 48% 0.74 19.2 30.1 4.64 53.50 17.0 438.0 15.5 AL7 E 45% 0.87 13.1 24.7 5.44 6.70 85.0 257.0 -0.7 AL8 C 71% 0.31 33.6 23.1 4.51 69.40 28.5 476.3 5.8 AL8 E 72% 0.29 40.6 35.0 4.86 97.03 16.0 435.2 3.5 AL9 C 79% 0.20 24.3 25.1 5.70 42.13 28.7 282.8 20.3 AL9 E 77% 0.26 24.0 -2.7 FL10 C 50% 0.63 15.8 16.2 6.49 9.37 139.0 380.7 5.0 FL10 E 42% 0.75 12.5 25.0 5.87 32.37 60.7 237.9 15.0 FL11 C 83% 0.10 89.1 18.5 3.49 12.97 184.0 506.6 4.7 FL11 E 75% 0.21 64.1 16.8 6.27 12.23 133.8 365.9 5.3 FL12 C 88% 0.10 86.1 18.7 4.57 61.40 22.7 461.1 20.5 FL12 E 86% 0.12 79.1 18.8 3.49 12.60 188.0 499.6 0.4

PAGE 102

90Table C.1.Continued ID Area Soil moisture content (%) Soil bulk density (g /cm3) Soil LOI (%) Water temp (C) Water pH Water DO (%) Water conductivity (uS/cm) Water Eh (mv) Water depth (cm) FL13 C 85% 0.13 69.2 19.8 4.57 51.87 27.7 525.0 23.5 FL13 E 76% 0.22 53.4 18.0 4.37 46.60 28.0 578.4 14.6 FL14 C 87% 0.11 87.5 17.8 3.76 6.50 73.0 478.3 8.8 FL14 E 86% 0.11 96.0 19.4 4.74 90.87 22.3 573.2 8.0 FL15 C 55% 0.47 31.4 FL15 E 62% 0.39 33.3 FL16 C 86% 0.12 89.2 FL16 E 80% 0.18 82.4 FL17 C 89% 0.09 98.2 20.3 3.36 27.40 134.5 5.7 FL17 E 83% 0.16 90.8 -1.3 FL18 C 49% 0.46 16.8 FL18 E 70% 0.27 51.8 FL19 C 83% 0.15 63.3 FL19 E 78% 0.21 64.3 FL20 C 44% 0.62 12.5 22.5 4.06 37.37 27500.0 555.6 25.3 FL20 E 32% 0.62 7.2 -2.0 FL21 C 69% 0.17 23.2 17.1 3.82 30.10 593.5 5.0 FL21 E 63% 0.20 11.2 22.8 4.08 50.35 19960.0 529.3 18.3 FL22 C 77% 0.23 35.6 18.8 3.75 79.73 8.3 FL22 E 43% 0.62 16.2 18.6 3.74 38.93 4.2 FL23 C 80% 0.17 15.8 24.5 6.90 53.87 1146.7 4.7 FL23 E 45% 0.43 6.1 26.6 7.22 57.57 1182.7 3.3 FL24 C 54% 0.66 13.4 28.2 3.84 70.83 12.3 237.6 38.0 FL24 E 42% 0.87 9.9 28.0 3.69 63.13 10.3 573.7 20.3 FL25 C 24% 1.10 2.5 32.4 3.96 80.13 19.3 573.0 10.0 FL25 E 68% 0.13 6.8 30.9 3.93 105.20 23.3 559.3 13.5 FL26 C 51% 0.50 27.7 -5.5

PAGE 103

91Table C.1.Continued ID Area Soil moisture content (%) Soil bulk density (g /cm3) Soil LOI (%) Water temp (C) Water pH Water DO (%) Water conductivity (uS/cm) Water Eh (mv) Water depth (cm) FL26 C 58% 0.41 27.9 -5.5 FL27 C 58% 0.46 28.3 FL27 E 63% 0.50 20.9 FL28 C 38% 0.62 18.6 FL28 E 27% 0.81 10.4 FL29 C 18% 0.88 7.1 FL29 E 30% 0.80 10.4 FL30 C 43% 0.75 9.3 24.3 5.64 10.15 30.0 280.6 1.3 FL30 E 55% 0.43 17.1 FL31 C 25% 1.14 5.9 FL31 E 36% 0.85 7.1 FL32 C 11% 0.92 5.3 FL32 E 24% 1.09 4.5 FL33 C 37% 0.81 6.3 FL33 E 23% 1.04 4.1 FL34 C 55% 0.39 22.1 -5.0 FL34 E 16% 0.68 3.6 FL35 C 59% 0.51 16.2 -1.3 FL35 E 48% 0.59 13.4 FL36 C 88% 0.11 94.1 0.7 FL36 E 78% 0.24 FL37 C 84% 2.35 5.0 FL37 E 87% 0.17 7.1 FL38 E 22% 1.32 3.6 20.3 3.58 9.55 119.0 3.8 FL39 C 88% 0.10 91.1 20.1 3.57 27.95 118.0 10.0 FL39 E 77% 0.23 55.9 22.0 3.95 35.70 53.7 9.0 FL40 C 59% 0.41 21.2 3.87 47.40 59.0 9.0

PAGE 104

92Table C.1.Continued ID Area Soil moisture content (%) Soil bulk density (g /cm3) Soil LOI (%) Water temp (C) Water pH Water DO (%) Water conductivity (uS/cm) Water Eh (mv) Water depth (cm) FL40 E 76% 0.22 16.9 -8.0 FL41 C 55% 0.53 22.1 24.4 4.63 5.22 55.0 3.2 FL41 E 37% 0.64 14.4 -4.0 FL42 C 50% 0.55 14.5 -5.0 FL42 E 34% 0.84 8.6 FL43 C 54% 0.47 25.2 FL43 E 62% 0.21 61.3 -5.5 FL44 C 53% 0.55 13.1 21.4 3.89 17.20 70.5 10.5 FL44 E 48% 0.50 22.5 22.1 5.97 12.10 57.0 12.3 FL45 C 82% 0.17 63.6 21.9 6.21 21.45 78.5 20.3 FL45 E 69% 0.31 30.5 21.1 5.94 13.63 69.3 234.9 20.0 GA1 C 45% 0.76 11.6 21.4 6.38 19.27 68.7 236.0 15.0 GA1 E 46% 0.68 12.2 24.8 6.36 48.83 56.0 6.7 GA10 C 63% 0.46 18.3 23.7 6.35 41.35 99.5 2.0 GA10 E 63% 0.46 15.8 21.0 4.66 9.20 32.0 419.6 10.3 GA16 C 84% 0.16 60.1 21.8 5.32 5.10 50.7 323.2 23.0 GA16 E 81% 0.17 65.8 5.3 GA17 C 62% 0.38 14.7 11.3 GA17 E 57% 0.50 12.1 20.7 3.67 5.65 85.5 545.7 3.0 GA19 C 90% 0.11 81.8 20.8 3.67 7.53 82.7 531.5 8.3 GA19 E 83% 0.16 60.5 GA2 C 36% 0.90 11.8 GA2 E 26% 0.98 8.1 GA20 C 87% 0.11 90.7 20.1 3.66 6.90 79.7 575.2 5.7 GA20 E 78% 0.19 52.7 GA21 C 85% 0.15 78.3 22.0 4.67 5.50 50.7 420.3 6.3 GA21 E 85% 0.15 77.5

PAGE 105

93Table C.1.Continued ID Area Soil moisture content (%) Soil bulk density (g /cm3) Soil LOI (%) Water temp (C) Water pH Water DO (%) Water conductivity (uS/cm) Water Eh (mv) Water depth (cm) GA25 C 54% 0.58 9.5 GA25 E 42% 0.78 10.9 GA26 C 34% 0.59 16.7 GA26 E 28% 0.80 10.0 GA27 C 35% 0.67 17.1 GA27 E 16% 1.07 4.5 GA28 C 65% 0.31 21.8 GA28 E 49% 0.57 16.4 GA29 C 63% 0.41 16.3 GA29 E 38% 0.80 7.7 GA3 C 34% 0.77 15.3 GA3 E 34% 0.79 16.1 GA30 C 77% 0.24 33.6 14.6 5.46 65.60 17.7 343.2 -0.3 GA30 E 53% 0.55 16.3 GA31 C 44% 0.57 19.0 GA31 E 43% 0.57 25.5 -6.0 GA32 C 43% 0.69 12.7 GA32 E 24% 1.13 7.2 GA4 C 52% 0.54 20.0 GA4 E 33% 0.84 10.0 GA5 E 29% 1.07 10.0 22.0 5.75 18.60 52.0 GA6 C 33% 0.94 9.7 GA6 E 34% 0.83 12.4 GA7 C 45% 0.76 13.5 21.4 6.32 27.10 79.0 363.5 -0.3 GA7 E 45% 0.70 16.1 GA8 C 28% 0.89 9.0 GA8 E 40% 0.82 12.9

PAGE 106

94Table C.1.Continued ID Area Soil moisture content (%) Soil bulk density (g /cm3) Soil LOI (%) Water temp (C) Water pH Water DO (%) Water conductivity (uS/cm) Water Eh (mv) Water depth (cm) GA9 C 37% 0.89 14.2 22.8 6.25 17.55 82.0 341.2 6.0 GA9 E 36% 0.84 23.4 6.07 9.70 238.0 262.0 -1.0 SC1 C 48% 0.65 18.2 25.6 6.60 31.97 76.0 334.3 23.7 SC1 E 36% 0.59 19.3 0.3 SC10 C 33% 0.88 9.5 SC10 E 25% 1.01 6.7 SC11 C 62% 0.43 32.6 23.5 6.12 7.00 228.0 213.8 SC11 E 55% 0.52 22.3 SC12 C 81% 0.18 55.8 25.5 5.92 42.30 82.3 319.7 2.3 SC12 E 79% 0.23 48.4 SC13 C 43% 0.81 16.6 SC13 E 37% 0.93 12.0 SC14 C 38% 0.64 SC14 E 42% 0.75 13.0 SC15 C 23% 0.95 6.3 SC15 E 36% 0.97 7.5 SC16 C 32% 0.88 12.3 SC16 E 34% 0.95 12.2 SC17 C 32% 0.79 11.8 SC17 E 38% 0.77 14.5 SC18 C 59% 0.51 12.6 27.8 6.28 16.70 101.0 256.6 18.7 SC18 E 55% 0.58 -5.0 SC19 C 26% 0.92 6.3 SC19 E 33% 0.75 15.8 SC2 C 44% 0.58 19.3 SC2 E 42% 0.53 19.4 SC20 C 26% 0.98 9.0

PAGE 107

95Table C.1.Continued ID Area Soil moisture content (%) Soil bulk density (g /cm3) Soil LOI (%) Water temp (C) Water pH Water DO (%) Water conductivity (uS/cm) Water Eh (mv) Water depth (cm) SC20 E 41% 0.76 SC21 C 31% 0.83 11.4 SC21 E 44% 0.71 16.2 SC22 C 26% 0.84 8.5 SC22 E 34% 0.78 9.1 SC23 C 33% 0.84 9.5 SC23 E 39% 0.78 10.0 SC24 C 26% 1.03 8.3 SC24 E 30% 0.83 9.5 SC3 C 39% 0.88 12.2 26.9 6.25 49.33 76.0 302.9 7.3 SC3 E 40% 0.85 14.1 25.4 4.19 11.60 29.0 -0.5 SC4 C 37% 0.87 9.9 24.1 5.79 8.25 130.5 0.0 2.3 SC4 E 28% 0.83 9.0 28.2 5.96 44.00 68.0 336.8 0.3 SC5 C 87% 0.11 88.6 22.5 3.65 12.20 62.5 537.0 5.3 SC5 E 86% 0.12 90.0 24.0 5.93 7.00 101.0 266.9 SC6 C 63% 0.45 28.2 SC6 E 60% 0.51 25.6 SC7 C 88% 0.12 SC7 E 89% 0.10 63.1 SC8 C 30% 0.86 11.3 SC8 E 27% 0.94 7.2 SC9 C 75% 0.22 58.9 24.2 3.80 15.70 68.0 481.4 0.7

PAGE 108

APPENDIX D SOIL, LITTER, AND WATER COLUMN CHEMICAL DATA

PAGE 109

97Table D-1. Chemical soil, litter and water column data for edge (E) and Core (C) sites ID Area Soil TP (mg/kg) Soil TN (g/kg) Soil TC (g/kg) Litter TP (%) Litter TN (%) Litter TC (%) Water Column TP (mg/L) Water Column TN (mg/L) AL1 C 121.6 1.76 39.01 0.009 0.89 21.63 0.012 1.53 AL1 E 107.2 1.05 20.60 0.011 1.08 37.31 AL10 C 363.7 4.55 80.63 0.032 1.11 39.81 0.021 1.24 AL10 E 405.1 9.19 198.60 0.009 1.36 47.18 0.115 2.56 AL11 C 259.0 1.95 33.79 0.047 1.59 30.45 0.092 1.86 AL11 E 256.9 3.49 71.33 0.035 1.91 38.03 0.465 9.13 AL12 C 157.3 2.88 62.10 0.029 1.15 31.41 0.035 1.32 AL12 E 144.4 1.29 29.53 0.024 1.06 40.08 0.022 7.37 AL13 C 204.8 2.20 40.20 0.023 0.88 27.22 0.013 0.94 AL13 E 165.0 1.44 29.65 0.022 1.33 40.55 AL14 C 339.9 3.17 59.35 0.059 1.31 36.65 0.032 1.58 AL14 E 419.0 3.76 70.33 1.48 39.74 0.150 2.50 AL15 C 438.3 3.51 65.50 0.97 27.29 AL15 E 339.2 5.30 123.50 0.008 1.53 45.18 AL16 C 307.0 6.40 141.20 0.023 1.81 37.82 0.082 1.79 AL16 E 141.9 2.36 52.91 0.022 1.35 29.72 0.120 2.21 AL17 C 156.9 1.48 36.63 0.009 1.44 44.58 0.015 0.95 AL17 E 182.2 1.40 31.00 0.009 1.37 43.46 AL18 C 485.5 9.08 158.07 0.030 1.39 35.81 0.626 8.70

PAGE 110

98Table D.1.Continued ID Area Soil TP (mg/kg) Soil TN (g/kg) Soil TC (g/kg) Litter TP (%) Litter TN (%) Litter TC (%) Water Column TP (mg/L) Water Column TN (mg/L) AL18 E 290.3 3.36 72.01 0.010 1.22 45.10 AL19 E 320.7 3.85 96.50 0.040 1.57 39.40 0.345 4.78 AL2 C 0.024 1.13 30.16 0.042 1.31 AL2 E 221.8 3.28 66.44 0.018 1.24 40.03 0.046 1.93 AL20 C 160.8 1.17 21.92 0.009 1.35 43.34 AL20 E 171.5 1.14 22.36 0.010 1.23 39.97 AL3 C 198.2 1.34 28.42 0.006 7.87 39.30 0.018 1.33 AL3 E 177.7 1.71 38.40 0.005 0.98 47.63 AL4 C 673.8 0.015 1.97 45.76 0.353 7.25 AL4 E 437.5 0.019 1.40 46.24 0.240 5.60 AL5 C 602.4 18.39 445.20 0.006 1.80 51.07 0.019 1.38 AL5 E 266.9 6.40 107.40 0.006 1.27 45.29 AL6 C 594.2 19.10 425.20 0.006 1.61 48.66 0.039 2.29 AL6 E 461.7 3.40 67.40 0.040 1.41 42.49 0.745 13.33 AL7 C 422.2 5.36 60.67 0.012 1.80 45.06 0.024 1.79 AL7 E 145.1 3.08 41.76 0.010 1.23 44.27 0.023 1.93 AL8 C 373.2 6.70 160.20 0.008 1.39 46.37 AL8 E 434.2 9.73 219.60 0.007 1.37 48.10 AL9 C 443.6 5.15 0.00 0.038 1.48 31.51 0.028 0.96 AL9 E 471.5 4.87 104.27 0.033 1.46 29.17 0.031 0.82 FL10 C 219.0 3.74 73.79 0.018 1.33 50.09 1.63 FL10 E 218.6 4.10 62.90 0.012 1.11 48.73 0.409 1.03 FL11 C 538.0 15.57 483.10 0.005 0.87 51.35 0.256 2.38 FL11 E 319.9 9.07 353.77 0.008 0.90 51.76 0.194 1.70 FL12 C 375.5 33.39 465.40 0.009 2.24 45.26 0.019 2.24 FL12 E 596.0 30.42 426.20 0.005 1.71 45.54 0.025 1.02 FL13 C 373.7 24.67 351.77 0.004 1.33 45.40 0.016 0.83 FL13 E 640.6 17.96 248.30 0.005 0.84 43.91 0.026 0.83

PAGE 111

99Table D.1.Continued ID Area Soil TP (mg/kg) Soil TN (g/kg) Soil TC (g/kg) Litter TP (%) Litter TN (%) Litter TC (%) Water Column TP (mg/L) Water Column TN (mg/L) FL14 C 545.8 17.58 461.50 0.006 1.34 50.45 0.351 6.64 FL14 E 431.2 11.44 498.60 0.004 0.93 51.70 0.179 3.22 FL15 C 466.6 7.56 149.21 0.016 1.37 46.60 0.149 3.62 FL15 E 501.7 8.96 190.90 0.019 1.85 46.01 FL16 C 827.7 22.68 424.92 0.039 1.77 45.04 FL16 E 891.0 19.70 399.40 0.009 1.56 47.44 FL17 C 405.5 14.15 495.95 0.005 1.30 52.66 0.026 3.17 FL17 E 363.0 13.71 486.57 0.004 1.08 53.13 0.030 3.78 FL18 C 353.3 4.91 84.25 0.010 1.39 44.47 0.213 2.78 FL18 E 935.2 14.49 257.63 0.011 1.17 47.07 FL19 C 804.5 17.87 302.50 0.011 1.66 47.30 FL19 E 696.6 15.66 288.20 0.014 1.72 48.44 FL20 C 66.0 4.88 87.75 0.021 2.01 FL20 E 45.9 1.60 35.00 0.005 1.32 46.96 0.035 2.37 FL21 C 193.2 4.92 87.44 0.009 1.15 49.36 0.024 2.86 FL21 E 98.4 2.25 55.50 0.009 1.08 49.21 0.055 2.96 FL22 C 298.5 9.30 191.20 0.008 0.00 0.00 0.018 2.14 FL22 E 82.7 4.14 0.00 0.007 1.02 50.86 0.024 2.07 FL23 C 333.1 4.76 87.33 0.007 0.99 43.12 0.079 0.91 FL23 E 51.9 0.85 25.15 0.007 1.04 29.70 0.049 1.76 FL24 C 180.8 4.41 62.70 0.003 1.26 46.71 0.014 1.64 FL24 E 109.7 2.35 48.69 0.005 1.44 44.06 0.028 1.80 FL25 C 44.0 0.25 4.55 0.006 0.95 44.40 0.007 1.22 FL25 E 56.0 2.27 31.15 0.007 1.77 39.47 0.007 1.15 FL26 C 239.4 5.46 126.77 0.006 1.09 49.49 FL26 C 261.1 0.006 1.09 49.49 FL27 C 291.4 7.38 137.70 0.005 0.84 47.97 0.030 1.83 FL27 E 133.3 3.55 137.80 0.004 0.72 49.06

PAGE 112

100Table D.1.Continued ID Area Soil TP (mg/kg) Soil TN (g/kg) Soil TC (g/kg) Litter TP (%) Litter TN (%) Litter TC (%) Water Column TP (mg/L) Water Column TN (mg/L) FL28 C 130.2 4.28 97.51 0.002 1.03 52.04 FL28 E 50.2 2.25 58.45 0.005 1.06 50.21 FL29 C 87.2 2.19 48.02 0.007 1.03 45.88 FL29 E 176.4 2.23 44.12 0.008 1.10 38.60 FL30 C 91.3 2.13 48.69 0.011 1.05 37.36 FL30 E 230.4 3.55 81.12 0.004 0.80 47.17 FL31 C 64.7 1.18 22.76 0.004 0.84 43.98 FL31 E 39.5 1.28 31.10 0.008 0.84 41.96 FL32 C 28.8 0.99 30.32 0.004 0.70 42.78 FL32 E 5.5 0.73 25.12 0.004 0.71 39.03 FL33 C 42.3 5.90 32.42 0.008 0.88 26.51 FL33 E 46.9 0.92 16.80 0.002 0.54 44.49 0.004 0.85 FL34 C 75.9 4.77 117.45 0.007 0.88 50.83 FL34 E 21.9 0.57 16.89 0.003 0.62 51.02 FL35 C 95.1 3.80 81.10 0.007 0.92 38.70 0.086 2.92 FL35 E 71.8 34.70 54.50 0.008 0.75 48.16 FL36 C 540.7 16.12 503.39 0.006 0.91 50.59 0.034 2.86 FL36 E 0.011 1.16 50.10 FL37 C 38.8 1.59 64.31 0.008 0.93 46.43 0.069 2.50 FL37 E 42.0 1.01 31.00 0.008 0.97 49.97 FL38 E 33.5 0.30 6.20 0.014 1.24 34.25 FL39 C 506.5 0.007 1.57 51.08 0.141 3.53 FL39 E 272.0 8.60 290.50 0.006 0.96 50.35 0.471 5.51 FL40 C 2.37 53.67 0.009 1.16 45.07 FL40 E 270.0 3.79 85.30 0.009 1.08 47.93 0.059 1.65 FL41 C 630.0 6.63 96.24 0.005 1.03 49.63 0.184 1.57 FL41 E 240.0 3.49 61.03 0.010 1.15 45.57 FL42 C 137.9 3.40 74.00 0.010 1.46 44.78

PAGE 113

101Table D.1.Continued ID Area Soil TP (mg/kg) Soil TN (g/kg) Soil TC (g/kg) Litter TP (%) Litter TN (%) Litter TC (%) Water Column TP (mg/L) Water Column TN (mg/L) FL42 E 95.2 1.65 43.66 0.009 1.40 48.27 FL43 C 278.5 5.27 113.20 0.007 1.25 49.30 FL43 E 385.6 10.45 338.34 0.004 0.81 50.67 0.063 1.34 FL44 C 120.3 2.53 59.90 0.006 1.13 49.03 0.042 2.29 FL44 E 82.5 4.19 118.47 0.008 1.19 49.00 FL45 C 812.0 17.71 321.40 0.009 1.41 49.55 0.086 1.41 FL45 E 214.1 6.88 154.75 0.010 1.36 49.87 GA1 C 700.0 1.73 19.05 0.025 1.64 34.62 0.194 1.25 GA1 E 851.1 2.01 21.16 0.019 1.37 36.44 0.214 1.82 GA10 C 954.5 3.44 49.30 3.00 GA10 E 845.4 2.90 39.50 0.151 1.44 GA16 C 546.9 16.69 310.27 1.87 42.46 0.049 1.31 GA16 E 323.0 12.06 339.60 0.045 1.54 GA17 C 214.0 3.74 77.16 1.82 47.64 0.026 1.32 GA17 E 133.8 2.12 63.04 1.34 45.01 0.086 1.69 GA19 C 870.2 2.02 47.58 0.044 1.89 GA19 E 716.0 14.73 314.78 1.86 49.22 0.133 3.09 GA2 C 314.9 1.80 30.87 0.000 0.98 38.44 GA2 E 230.1 1.27 23.00 0.009 1.25 44.48 GA20 C 1088.2 23.16 467.53 1.76 51.04 0.070 2.24 GA20 E 867.9 14.96 278.02 2.09 50.77 GA21 C 1215.2 19.05 416.06 1.66 48.81 0.096 1.77 GA21 E 1331.4 18.17 406.95 1.72 48.89 0.090 1.62 GA25 C 180.2 3.05 56.67 1.37 45.14 GA25 E 251.3 2.41 41.92 1.02 44.48 GA26 C 399.6 3.70 64.36 0.98 43.03 GA26 E 230.4 2.13 38.29 1.22 46.14 GA27 C 297.8 3.22 59.75 1.21 41.62

PAGE 114

102Table D.1.Continued ID Area Soil TP (mg/kg) Soil TN (g/kg) Soil TC (g/kg) Litter TP (%) Litter TN (%) Litter TC (%) Water Column TP (mg/L) Water Column TN (mg/L) GA27 E 74.6 0.66 19.48 1.10 43.88 GA28 C 334.7 4.81 92.75 1.22 34.85 GA28 E 476.8 1.27 45.31 GA29 C 289.6 4.74 84.39 0.92 37.32 GA29 E 147.5 1.67 30.37 1.01 40.24 GA3 C 604.1 2.69 32.08 0.014 1.32 32.66 GA3 E 638.5 2.80 39.50 0.013 1.33 35.77 GA30 C 619.7 9.09 146.51 93.75 43.19 0.017 0.63 GA30 E 442.2 3.90 65.88 1.44 47.72 GA31 C 236.6 4.57 97.65 1.23 45.10 GA31 E 285.8 4.81 112.51 GA32 C 200.3 2.53 47.80 1.13 45.17 GA32 E 102.9 1.05 26.19 1.12 41.00 GA4 C 467.4 3.62 66.44 0.018 1.45 36.38 GA4 E 380.0 120.82 30.96 0.014 1.74 41.80 GA5 E 203.4 1.20 16.70 0.057 1.36 GA6 C 470.9 1.86 27.66 0.013 1.01 27.14 GA6 E 539.8 2.47 43.88 0.012 1.12 32.76 GA7 C 456.4 2.32 33.68 0.015 1.70 36.09 0.177 3.08 GA7 E 583.3 2.75 36.73 0.016 1.71 36.56 0.204 2.72 GA8 C 263.3 1.60 22.90 0.007 1.01 43.58 GA8 E 538.1 1.80 21.70 0.013 1.64 41.01 0.355 5.60 GA9 C 623.5 2.25 37.07 0.019 1.12 27.71 0.243 2.16 GA9 E 0.015 0.00 0.00 0.184 2.29 SC1 C 732.5 3.69 45.39 0.031 1.40 36.68 0.057 0.86 SC1 E 702.3 4.47 53.18 0.022 1.68 38.39 SC10 C 232.3 2.00 28.70 0.023 1.36 33.02 0.080 2.26 SC10 E 148.8 1.38 25.33 0.014 1.26 30.82

PAGE 115

103Table D.1.Continued ID Area Soil TP (mg/kg) Soil TN (g/kg) Soil TC (g/kg) Litter TP (%) Litter TN (%) Litter TC (%) Water Column TP (mg/L) Water Column TN (mg/L) SC11 C 894.2 9.17 120.89 0.041 1.96 40.95 0.193 2.73 SC11 E 488.5 4.84 74.65 0.032 1.77 43.04 0.142 2.62 SC12 C 972.5 12.99 0.00 0.033 1.33 42.18 0.096 1.89 SC12 E 691.9 10.60 204.30 0.041 1.53 42.53 SC13 C 848.3 2.03 29.20 0.041 1.07 24.05 0.304 1.14 SC13 E 531.9 1.90 27.00 0.020 1.01 34.78 SC14 C 0.016 1.54 41.42 SC14 E 674.1 2.16 30.16 0.021 1.25 38.58 SC15 C 262.4 0.88 22.40 0.013 0.89 21.91 SC15 E 378.0 1.57 19.73 0.014 0.93 19.81 SC16 C 889.8 1.40 22.80 0.027 1.01 27.05 0.060 0.78 SC16 E 645.1 1.68 24.81 0.035 0.77 22.57 SC17 C 314.8 1.96 25.74 0.021 1.45 35.54 SC17 E 328.6 2.60 32.42 0.023 1.34 31.56 SC18 C 407.0 2.30 39.30 0.023 1.33 37.96 0.227 1.49 SC18 E 0.029 1.72 38.68 0.112 1.39 SC19 C 418.0 1.19 20.02 0.012 2.77 44.41 SC19 E 561.1 0.033 1.64 29.80 SC2 C 522.8 4.24 60.04 0.038 1.24 33.09 SC2 E 314.2 3.80 69.19 0.026 1.18 37.37 SC20 C 152.1 1.50 20.90 0.022 1.19 32.40 SC20 E 2.12 27.06 0.025 1.20 32.85 SC21 C 476.2 1.80 21.70 0.024 0.68 16.99 SC21 E 493.0 2.50 32.80 0.031 1.08 29.30 0.321 5.09 SC22 C 609.5 1.55 23.13 0.014 0.81 24.95 SC22 E 856.2 1.70 25.70 0.019 1.43 35.67 SC23 C 317.6 1.60 23.10 0.020 38.67 11.85 SC23 E 424.3 1.85 23.07 0.030 1.40 28.23

PAGE 116

104Table D.1.Continued ID Area Soil TP (mg/kg) Soil TN (g/kg) Soil TC (g/kg) Litter TP (%) Litter TN (%) Litter TC (%) Water Column TP (mg/L) Water Column TN (mg/L) SC24 C 258.3 1.34 19.23 0.013 1.58 42.62 SC24 E 411.4 0.015 1.53 41.88 SC3 C 177.7 2.58 47.22 0.016 1.10 43.23 0.040 1.87 SC3 E 234.8 2.84 53.15 0.021 1.16 43.93 0.058 2.48 SC4 C 178.3 2.27 39.62 0.017 1.12 26.11 0.225 6.58 SC4 E 98.2 1.81 41.63 0.021 1.29 37.78 SC5 C 674.3 15.54 458.63 0.007 1.35 50.52 0.028 1.86 SC5 E 638.8 18.26 463.42 0.006 1.38 51.50 0.019 1.37 SC6 C 572.1 7.43 123.80 0.021 1.44 49.76 SC6 E 510.2 6.38 106.76 0.021 1.36 38.95 0.061 1.93 SC7 C 0.033 1.57 41.00 SC7 E 1453.3 18.46 25.93 0.014 1.31 41.17 SC8 C 346.8 2.48 33.76 0.013 1.16 32.59 SC8 E 248.3 1.54 18.79 0.016 1.64 36.42 0.213 2.86 SC9 C 531.5 10.68 292.21 0.007 1.32 50.58 0.243 4.35 SC9 E 529.9 10.42 282.99 0.008 1.47 48.76 0.086 3.29

PAGE 117

105 LIST OF REFERENCES Adamus, P.R. and K. Brandt. 1990. Impacts on quality of inland wetlands of the United States: A survey of indicators, technique s, and applications of community level biomonitoring data. US Environmental Prot ection Agency, Office of Research and Development, Washington, D.C. Anderson, J.M. 1976. An ignition method for de termination of total phosphorus in lake sediments. Water Research 10:329-331. Battle and Golladay. 2001. Water quality a nd macroinvertebrate assemblages in three types of seasonally inundated limesink wetla nds in southwest Georgia. Journal of Freshwater Ecology 16:189-207. Bedford, B.L.,M.R. Waldridge, and A. Audous. 1999. Patterns in nut rient availability and plant biodiversity of temperate North American wetlands. Ecological Society of America 80:2151-2169. Brewer, I. 1999. The conceptual devel opment and use of ecoregion classifications. Unpublished master’s thesis. Oregon State University. Brinson, M.M., A.E. Lugo, and S. Brow n. 1981. Primary production, decomposition, and consumer activity in freshwater we tlands. Annual Review of Ecology and Systematics 12:123-161. Brown, M.T. and M.B. Vivas. In press. A landscape development intensity index. Environmental Monitoring a nd Assessment. [in press] Carpenter, S.R., N.F. Caraco, D.L. Correll R.W. Howarth, A.N. Sharpley, and V.H. Smith. 1998. Nonpoint pollution of surface waters with phosphorus and nitrogen. Ecological Applications 8:559-568. Cowardin, L.M., V. Carter, F.C. Golet, and E.T. LaRoe. 1979. Classification of wetlands and deepwater habitats of the United States. U.S. Fish and Wildlife Service, Office of Biological Servic es, Washington, DC, USA. FWS/OBS-79/31. Craft, C.B. and W.P. Casey. 2000. Sedime nt and nutrient accumulation in floodplain and depressional freshwater wetlands of Georgia, USA. Wetlands 20:323-332. Ehrenfield, J.G and J.P. Schneider. 1991. Chamaecyparis thyoides wetlands and suburbanization; effects on hydrology, water quality and plant community composition. Journal of Applied Ecology 28: 467-490.

PAGE 118

106 Fennessey, S., M. Gernes, J. Mack, DH Ward rop. 2001. Methods for evaluating wetland condition: using vegetation to assess environmental conditions in wetlands. APA 822-R-01-007j. U.S. Environmental Pr otection Agency; Office of Water; Washington, DC. Fennessy, M.S. and W.J. Mitsch. 2001. Eff ects of hydrology on spatial patterns of soil development in created riparian wetla nds. Wetlands Ecology and Management 9:103-120. Galatowitsch, S.M., D.C. Whited, and J.R. Tester. 1999. Development of community metrics to evaluate recovery of Minnesot a wetlands. Journal of Aquatic Ecosystem Stress and Recovery 6:217-234. Gilliam, J.W. 1994 Riparian wetlands and wa ter quality. Journal of Environmental Quality 23:896-900. Gopal, B. 1999. Natural and constructed wetl ands for wastewater tr eatmentpotentials and problems. Water Science and Technology 40:27-35. Gusewell, S. and W. Koerselman. 2002. Variation in nitrogen and phosphorus concentrations of wetland plants. Persp ectives in Plant Ec ology, Evolution, and Systematics 5:37-61. Gusewell, S, W. Koerselman, and T.A. Ver hoeven. 1998. The N:P ratio and the nutrient limitation of wetland plants. Bulletin of the Geobotanical Institiute ETH 64:77-90. Hopkinson, C.S. 1992. A comparison of ecosyst em dynamics in freshwater wetlands. Estuaries 15:549-562. Hupp, C.R. 2000. Hydrology, geomorphology, and ve getation of coastal plain rivers in the south-eastern USA. H ydrologic Processes 14:2991-3010. Jones, J.A., F.J. Swanson, B.C. Wemple, and K.U. Snyder. 1999. Effects of roads on hydrology, geomorphology, and disturbance patches in stream networks. Conservation Biology 14:76-85. Kadlec, Robert H, and Robert L. Knight 1996. Treatment Wetlands. CRC Press LLC, Boca Raton, Florida. Kantrud, H.A. and W.E. Newton. 1996. A te st of vegetation-related indicators of wetland quality in the prairie pothole region. Journal of Aquatic Ecosystem Health 5:177-191. Koerselman, W., and A.F.M. Me uleman. 1996. The vegetation N:P Ratio: a new tool to detect the nature of nutrient limitatio n. Journal of Applied Ecology, 33:1441-1450. Kushlan, James A. 1990. Freshwater Marshes. Ecosystems of Flor ida. Orlando: Univ. of Central Florida Press. 324-362.

PAGE 119

107 Lane, Charles R., M.T. Brown, M. Murray-Hudson, and M.B. Vivas. 2004. The wetland condition index (WCI): biological indicat ors of wetland condition for isolated Depressional herbaceous wetlands in Florid a. Florida Department of Environmental Protection Contract #WM-683. Lee, A.A. and P.A. Bukaveckas. 2002. Surf ace water nutrient concentrations and litter decomposition rates in wetlands impacted by agricultural and mi ning activities. Aquatic Botany, 74: 273-285. Mitsch, W.J. and J.G. Goselink. 2000. Wetla nds. John Wiley and Sons, Inc., New York, NY, USA. Moore, H.H.,W.A. Niering, L.J. Marsicano, and M. Dowell. 1999. Vegetation changes in created emergent wetlands (1988-1996) in Connecticut (USA). Wetlands Ecology and Management 7:177-191. Morris, J.T. 1991. Effects of nitrogen load ing on wetland ecosystems with particular reference to atmospheric deposition. Annual Review of Ecological Systems 22:257-279. Nicholson. B.J. 1995. The wetlands of Elk Island National Parkvegetation classification, water chemistry, and hydr otopographic relationships. Wetlands 15:119-133. Omernik, J.M. 1987. Map supplement: ecoregions of the conterminous United States. Annals of the Association of American Geographers 77:118-125. Omernik, J.M. and R.G. Bailey. 1997. Di stinguishing between watersheds and ecoregions. Journal of the American Water Resources Association, 33:935-949. Patrick, W.H. 1994. From wastelands to we tlands. Journal of Environmental Quality 23:892-896. Reddy, K.R. and E.M. D’Angelo. 1997. Biogeoc hemical indicators to evaluate pollutant removal efficiency in constructed we tlands. Water Science Technology 35:1-10. Reddy, K.R. and E.M. D’Angelo. 1994. So il processes regulating water quality in wetlands. Global WetlandsOld World and Ne w. Elsevier Pub lishing. New York, NY, US. Reinelt, L., R. Horner, and A. Azous. 1998. Impacts of urbanization on palustrine (depressional freshwater) wetlandresear ch and management in the Puget Sound Region. Urban Ecosystems 1:219-236. Richardson, C.J. 1985. Mechanisms cont rolling phosphorus retention capacity in freshwater wetlands. Science 228:1424-1427.

PAGE 120

108 Rotenberry, J.T. and J.A. Wiens. 1985. Sta tistical power analysis and community wide patterns. The American Naturalist 125:164-168. Shaver, G.R. and J.M. Melillo. 1984. Nutr ient budgets of marsh plants: efficiency concepts and relation to av ailability. Ecology 65:1491-1510. Tiner,W 2003. Geographically isolated wetla nds of the United Stat es. Wetlands 23:494516. Smith, V.H., G.D. Tilman, J.C. Nekola. 1999. Eutrophication: impact s of excess nutrient inputs on freshwater, marine, and terrestrial ecosystems. Environmental Pollution 100:179-196. U. S. Environmental Protection Agency. 2003. Level III Ecoregions of the continental United States Map. U. S. Environmenta l Protection Agency, Washington, D.C. U. S. Environmental Protection Agency. 2002. Water quality conditions in the United States a profile from the 2000 Nationa l Water Quality Inventory. U. S. Environmental Protection Agency, O ffice of Water, Washington, D.C. U. S. Environmental Protection Agency. 2000a. Nutrient criteria technical guidance manual: lakes and reservoirs. U. S. E nvironmental Protection Agency, Office of Water, Office of Science a nd technology, Washington, D.C. U. S. Environmental Protection Agency. 2000b. Nutrient criteri a technical guidance manual: rivers and streams. U. S. E nvironmental Protection Agency, Office of Water, Office of Science a nd technology, Washington, D.C. U. S. Environmental Protection Agency. 2000c. Ambient water quality criteria recommendations: wetlands in nutrient eco region XII. U. S. Environmental Protection Agency, Office of Water, Washington, D.C U. S. Environmental Protection Agency. 1998. National strategy for the development of regional nutrient criteria. U. S. Environmental Protecti on Agency, Office of Water, Washington, D.C. U. S. Environmental Protection Agency. 1993. Methods for the determination of inorganic substances in environmental samples. 365.1. U. S. Environmental Protection Agency. 1990. National guidance: water quality standards for wetlands. U. S. Environmen tal Protection Agency, Office of Water Regulations and Standards, Office of Wetlands Protection, Washington, D.C. U. S. Environmental Protection Agency. 1983. General Program Guidance of the Water Quality Standards Handbook. U. S. Envi ronmental Protection Agency, Office of Water Regulations and Standards, O ffice of Wetlands Protection, Washington, D.C.

PAGE 121

109 Whigham, D.F. and C.J. Richardson. 1988. Soil and plant chemistry of an Atlantic white cedar wetland on the Inner Coastal Plain of Maryland. Canadian Journal of Botany 66:568-576. Willby, N, J., I.D. Pulford, and T.H. Flowers. 2001. Tissue nutrient signatures predict herbaceous-wetland community responses to nu trient availability. New Phytologist 152: 463-481. Winter, T.C. and J.W. LaBaugh. 2003. Hydrol ogic considerations in defining isolated wetlands. Wetlands 23:494-516.

PAGE 122

110 BIOGRAPHICAL SKETCH Stacie Greco’s environmental career bega n when she visited the coastal North Carolina forest of her childhood, to find a park ing lot and mall where the pines and oaks once flourished. Stacie received her undergra duate degree in environmental studies at Warren Wilson College in Asheville, NC, in 1999. Shortly after graduation Stacie moved to the Virgin Islands to work in the ecotourism industry. U pon returning to Asheville in 2000 she worked as a project manager at an environmental consulting firm. Stacie began pursuing her master’s degree in the fall of 2001in the Department of Environmental Engineering Sciences at the University of Flor ida. In the future Stacie hopes to build partnerships with government, industry, and citizens to improve the environmental and social well being of communities.


Permanent Link: http://ufdc.ufl.edu/UFE0006802/00001

Material Information

Title: A Biogeochemical Survey of Wetlands in the Southeastern United States
Physical Description: Mixed Material
Copyright Date: 2008

Record Information

Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
System ID: UFE0006802:00001

Permanent Link: http://ufdc.ufl.edu/UFE0006802/00001

Material Information

Title: A Biogeochemical Survey of Wetlands in the Southeastern United States
Physical Description: Mixed Material
Copyright Date: 2008

Record Information

Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
System ID: UFE0006802:00001


This item has the following downloads:


Full Text











A BIOGEOCHEMICAL SURVEY OF WETLANDS IN THE SOUTHEASTERN
UNITED STATES














By

STACIE GRECO


A THESIS PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
MASTER OF SCIENCE

UNIVERSITY OF FLORIDA


2004

































Copyright 2004

by

Stacie Greco


































This document is dedicated to my friends and family whom have allowed me the time
and space for intellectual and emotional growth.
















ACKNOWLEDGMENTS

It is said that it takes a village to raise a child. Similarly, it takes a community to

write a thesis! I am thankful for the guidance and wisdom my committee provided

throughout this process. Dr. Mark Clark's contagious enthusiasm has helped me

overcome many doubts and fears. Kevin Grace's perpetual encouragement and patience

has greatly improved the quality of this work. The editing expertise of Dr. Tom Crisman

has been instrumental to this document. I would also like to acknowledge the hard work

of the Wetland Biogeochemistry Laboratory and the many helping hands in the field.

Finally, this research was made possible by funding from the USEPA' s Office of Water.





















TABLE OF CONTENTS


page


ACKNOWLEDGMENT S .............. .................... iv


LI ST OF T ABLE S ........._.___..... .___ ............... vii...


AB STRAC T ................ .............. xi


CHAPTER


1 INTTRODUCTION ................. ...............1.......... ......


Regulatory Background ................. ...............2.......... ......
Water Quality Standards............... ...............2
Numeric Nutrient Criteria............... ...............3

Types of Wetlands ................ ...............4............ ....
Defining Ecoregions ................. ........... ... ...............7.....
Limiting Nutrients and Causal Variables. ............. .. ...............9....
Biological Indicators of Nutrient Enrichment ................. .....................1 1
Biogeochemical Indicators of Nutrient Enrichment ................. ................12
Reference Wetlands ................. ...............14.......... .....
Research Objectives............... ...............1
Hypotheses............... ...............1


2 METHODS ................. ...............18.......... .....


Site Selection .............. ....... .................1
Identifying Minimally Impaired Sites .............. ...............18....
Identifying Wetland Community Types ................. .............. ......... .....20
Hydrologic Classification............... .............2
Site Selection Criteria............... ...............23

Sampling and Analytical Protocols .............. ...............24....
Sample Locations .............. ...............24...
Sample Collection and Processing .............. ...............27....
Water ........._.___..... .___ ...............27.....
Soil .............. ...............28....
Leaf litter ........._.___..... .___ ...............29.....
Data Analysis............... ...............30















3 RE SULT S AND DI SCU SSION ............... ..............3


Within Wetland Variability .............. ...............35....
W ater Column .............. ...............3 5....
Litter .............. ...............37....
S oil ................ ...............3.. 8..............
Discussion ................. ........... .. ...............38......

Variability among Wetland Types ................... ........... ...............39. ....

Vegetative Comparisons: Swamps and Marshes............... ...............39
W ater column .............. ...............40....
Litter ................ ...............41.......... ......
Soil .............. ...............46....
Discussion .............. ..... ........... .........4

Hydrologic Comparisons: Riverine and Non-riverine .............. ....................51
W ater column .............. ...............51....
Litter ............. ...... ...............53....
Soil .............. ...............54....
D discussion .............. ...............54....

Spatial Variation ............. ...... __ ...............60....
W ater Column .............. ...............62....
Litter .............. ...............64....
Soil ............. ...... ...............67....
Discussion ............. ...... ._ ...............70....


4 CONCLUSIONS .............. ...............75....


APPENDIX


A WETLAND CHARACTERIZATION FORM ....._____ ...... ..___ .............__..79


B WETLAND IDENTIFICATION AND LOCAtlON................ ...............8


C PHYSICAL SOIL AND WATER COLUMN DATA. ....._____ ...... ....__..........87


D SOIL, LITTER, AND WATER COLUMN CHEMICAL DATA .............................96


LIST OF REFERENCES ................. ...............105................


BIOGRAPHICAL SKETCH ................. ...............110......... ......


















LIST OF TABLES


Table pg

1-1 Comparison of wetland characteristics reported in the literature. .............. ..... ..........8

2-1 The NWI classification scheme............... ...............22.

2-2 Summary of chemical analyses and methods ................. .............................29

3-1 Various aggregations of the wetlands surveyed ................. .......... ...............33

3-2 Results of pair-wise comparison of core and edge areas ................ .............. .....36

3-3 Results of pair-wise comparison of core and edge areas ................ .............. .....37

3-4 Results of pair-wise comparison of core and edge areas ................ .............. .....39

3-5 Water column properties .............. ...............40....

3-6 Litter phosphorus, nitrogen, and carbon content ................. .......... ...............43

3-7 Soil P, N, and C content ................. ...............45........... ..

3-8 Values from the current study compared to those in the literature. .........................47

3-9 Power analysis for non-significant parameters within community comparisons.....49

3-10 Water column properties. ............. ...............52.....

3-11 Leaf litter properties ................. ...............55................

3-12 Soil properties .............. ...............57....

3-13 Number of surveyed wetlands within the three USEPA Nutrient Ecoregions......... 62

3-14 Water column descriptive statistics for surveyed wetlands by ecoregion................63

3-15 Litter descriptive statistics for surveyed by Ecoregion. ............. .....................6

3-16 Soil descriptive statistics for surveyed wetlands aggregated by Ecoregion. ...........69

3-17 Summary of significant differences among USEPA Nutrient Ecoregions ........._....71











3-18 Soil total phosphorus statistics .............. ...............74....

B-1 Wetland identification and location. ............. ...............83.....

D-1 Chemical soil, litter, and water column data for edge (E) and Core (C) sites..........97
























LIST OF FIGURES
Figure pg

1-1 USEPA Level IIII Nutrient Ecoregions. ............. ...............9.....

1-2 Two approaches for establishing reference conditions .............. ....................1

2-1 Sampling areas within the three USEPA Nutrient Ecoregions. ............. ................20

2-2 Number of wetlands surveyed aggregated by community type. ............. ...... ........._25

2-3 Sub-sample locations within the core and edge zones .............. .....................2

3-1 Total area of the four wetland types ................. ...............34......___..

3-2 Percentage distribution of surveyed wetlands within ecoregions ............................34

3-3 Water column TP and TN values by vegetative type ......__ ............ ....... ........41

3-4 Litter phosphorus, nitrogen, and carbon values by community type.. .....................42

3-5 Soil %P, %N, and %C values by community type............... ...............44..

3-6 Water column TP and TN values by hydrologic connectivity. ............... ...............53

3-7 Litter phosphorus, nitrogen, and carbon content comparisons. .........._... ..............55

3-8 Soil TP and TN values by hydrologic connectivity .............. .....................5

3-9 Distribution of wetlands within the three USEPA Nutrient Ecoregions. .................61

3-10 Comparison of ecoregions aggregated by hydrology. ............... ...................6

3-11 Comparison of ecoregions aggregated by vegetative type. ................. ........_.......65

3-12 Comparison of riverine wetlands in the three ecoregions .........._.... ........_.......67

3-13 Comparison of litter total phosphorus among the three ecoregionss. ................... ...68










3-14 Distribution of sampling locations within the USEPA Nutrient Ecoregions. ..........73















Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Master of Science

A BIOGEOCHEMICAL SURVEY OF WETLANDS IN THE SOUTHEASTERN
UNITED STATES

By

Stacie Greco

August 2004

Chair: Thomas Crisman
Cochair: Mark W. Clark
Maj or Department: Environmental Engineering Sciences

The USEPA's National Water Quality Inventory Reports consistently cite nutrient

enrichment as one of the leading causes of water quality impairment. To target problems

associated with nutrients, the Clean Water Action Plan of 1998 requires the USEPA to

establish numeric nutrient criteria specific to geographic region and waterbody type.

Developing nutrient criteria for wetlands is difficult due to a lack of historic data,

incompatibility of methods employed in previous studies, and inherent variability among

wetland community types.

The primary obj ectives of this study were to conduct a biogeochemical survey of

minimally impaired wetlands within the southeastern US and to determine the effect, if

any, of regional, hydrologic, and vegetative differences on wetland nutrient condition.

One hundred and three wetlands were sampled in three USEPA Nutrient Ecoregions

covering four states. Sampling was distributed among wetlands classified by hydrologic










connectivity into riverine and non-riverine and by dominant vegetative cover into

swamps and marshes.

Soil and litter parameters did not differ significantly between swamps and marshes,

suggesting a distinction between vegetative types is not necessary for determining soil or

litter numeric nutrient criteria in the southeast. Water column total phosphorus

differences between swamps and marshes imply a need to set numeric nutrient criteria

specific to dominant vegetative cover.

Hydrologic connectivity appears to be important when characterizing wetland

nutrient regimes, as demonstrated by differences in water column, litter, and soil

characteristics between riverine and non-riverine wetlands. Riverine wetlands had

greater water column and litter total phosphorus content and lower soil total nitrogen

content compared to non-riverine wetlands. It is hypothesized that hydrologic

connectivity to adj acent aquatic ecosystems and larger contributing watersheds of

riverine wetlands drives these differences.

The USEPA recognized the importance of regional influences on wetland nutrient

regimes when the decision was made to determine numeric nutrient criteria specific to

ecoregions. Results demonstrate that the Southemn Coastal Plain (XII) is different from

the Southemn Forested Plain (IX) and the Eastemn Coastal Plain (XIV), with greater water

column total nitrogen, litter total carbon, soil total nitrogen, soil total carbon, and lower

litter total phosphorus content. Variability was still large within a given ecoregion;

therefore spatial aggregation at a sub-ecoregion level may be necessary for effective

nutrient criteria development















CHAPTER 1
INTTRODUCTION

From the billions of appropriated dollars for restoration of the Florida Everglades

to the coining of wetlands as "nature' s kidneys," it is evident that wetlands are the

ecological buzzword and ecosystem focus of the millennium. It is hard to believe that

only a few decades ago wetlands were viewed as wastelands, portrayed by the popular

image of the Swamp Thing surrounded by putrid swamp gas. Before their inherent

values were recognized, wetlands were drained and converted to human-maintained

agricultural and sylvicultural lands at an alarming rate. The conversion of wetlands to

"more productive" land uses has recently decreased to a still alarming rate of 23,674

hectares a year (United States Environmental Protection Agency 2002). However, such

losses only represent complete destruction of these ecosystems and do not account for

numerous additional hectares where wetland functions have been degraded due to

changes in hydrology, vegetation, and/or water quality. It is this change in ecosystem

function, and thereby potential loss of designated use, that led to implementation of the

Clean Water Act (CWA) in 1972 and the current directive to establish numeric nutrient

criteria for water bodies within the USA. This thesis addresses some of the issues for

establishing numeric criteria for wetlands and presents results of a wetland survey

conducted in the southeastern United States.









Regulatory Background

Water Quality Standards

Section 304(a) of the CWA mandates the United States Protection Agency

(USEPA) to assist states, tribes, and territories in developing water quality standards.

Such standards contain three maj or components: 1) determination of designated uses,

2) development of numeric or narrative criteria to protect designated uses, and

3) development of antidegradation policy to avoid impacts not addressed by the developed

criteria (USEPA 1983). As of the late 1990s, 39 states lacked water quality standards

for wetlands (USEPA 2000).

States, tribes, and territories are required to determine designated uses of

waterbodies within their jurisdiction. These must meet the goals of Section 101(a) of the

CWA, which include protection and propagation of fish, shellfish, and wildlife along

with providing for recreation opportunities (USEPA 1983). Defining the designated uses

for rivers and lakes is a straightforward task since the values of swimming, fishing, and

water sports are easily recognized. This is not the case with wetlands because historically

their values have not been recognized, and they are not always obvious. Wetland values

can include flood storage, pollution and sediment control, food web support, groundwater

replenishment, and habitat for various organisms including waterfowl (Moore et al. 1999,

Morris 1979). Many states simply assign designated uses based on wetland type or

location in the landscape (USEPA 1990), since it is difficult to assign values to each

individual wetland.

Once states determine the designated uses of a waterbody, criteria must be

developed to protect those uses. The criteria of water quality standards can be narrative

or numeric. Narrative criteria are important for impacts that cannot be addressed by









numeric criteria, such as those that do not directly affect water chemistry. For example,

discharge of dredge and fill material can be prevented using narrative criteria. Numeric

criteria are values or ranges assigned to measurable chemical, physical, and/or biological

parameters. They can be more useful than narrative criteria because they provide a clear

distinction between acceptable and unacceptable conditions, and hence, reduce ambiguity

for management and enforcement decisions (USEPA 2000b).

Under Section 305(b) of CWA, states, tribes, and territories are required biennially

to compare monitoring results with their water quality standards. To identify trends in

water quality, the USEPA compiles the data and publishes the National Water Quality

Inventory Report. These reports consistently identify nutrients as one of the leading

causes of water quality impairment and failure to sustain the designated uses of

waterbodies. Excessive nutrients are responsible for almost 50% of impaired lake area

and 60% of impaired river reaches in the US (Smith et al. 1999).

Numeric Nutrient Criteria

To target problems specifically associated with nutrient enrichment, President

Clinton introduced the Clean Water Action Plan of 1998, which requires the USEPA to

establish numeric nutrient criteria specific to ecosystem type and geographic region. The

agency responded with a document describing its approach titled the National Strategy

for the Development of Regional Nutrient Criteria. The document describes the

USEPA' s intention to publish technical guidance manuals for each of the four waterbody

types (lakes and reservoirs, rivers and streams, estuaries, and wetlands) along with

criteria recommendations for specific ecoregions.

The USEPA intended to recommend target nutrient ranges on a geographic basis

using historical nutrient data, reference conditions, and expert knowledge (USEPA 1998).









Lakes/reservoirs, rivers/streams, and estuaries are well-monitored ecosystems with

sufficient data available to support numeric criteria development. With exception of the

Florida Everglades, wetlands lack even a skeletal survey of nutrient condition.

There is a lack of historical wetland data since their value as aquatic ecosystems is

a relatively recent phenomenon. The 2000 National Water Quality Inventory Report was

unable to make conclusions concerning wetland water quality because only 8% of total

wetlands in the US were surveyed, in contrast to 42% of US lakes (USEPA 2000a). For

those wetlands that have been monitored, numerous parameters have been measured, and

a variety of sampling techniques and methodologies have been utilized making

comparisons and regional characterization difficult. The exception is for the Florida

Everglades, which have been studied sufficiently to provide data for the USEPA to make

wetland numeric nutrient recommendations (USEPA 2000c).

Establishing numeric criteria for wetlands requires the determination of

1) designated use, 2) appropriate regional or type of wetland aggregation scheme to which

criteria are sufficiently but not overly protective, 3) limiting nutrient/casual variable to

determine which nutrients require criteria development, or in the absence of a clear cause

and effect threshold of impairment, the quantifieation of nutrient concentrations under

reference conditions. As discussed above, determination of designated use requires

recognition of wetland values and benefits to local communities. Determining

appropriate aggregation of wetlands for development of numeric criteria requires a

thorough investigation of potential differences among wetland types and regions.

Types of Wetlands

Definitions of wetlands include a suite of ecosystems supporting various functions.

Common wetland types of North America include freshwater marshes, peatlands,









freshwater swamps, riparian systems, tidal salt marshes, tidal freshwater marshes, and

mangrove wetlands (Mitsch and Gosselink 2000). These terms generally define the

dominant vegetation type and hydrologic regime. Marshes are characterized by annual or

perennial herbaceous species, and swamps are dominated by perennial woody vegetation

(Brinson et al. 1981). Hydrologically, wetlands are broadly categorized as riverine, tidal,

lake fringe, or isolated.

Extensive forested floodplains are common in the southeastern United States.

These riverine wetlands (also called floodplains, bottomlands, and riparian wetlands) are

connected to nearby rivers or streams, which supply water and nutrients during flood

events. Riverine systems also receive considerable inputs from runoff of the surrounding

landscape (Craft and Casey 2000).

Riparian wetlands play a critical role in maintaining water quality, as they

efficiently trap sediments and associated contaminants (Hupp 2000). Between 85 to 90%

of sediments leaving agricultural fields can be captured by wooded riparian wetlands

(Gilliam 1994). These wetlands are also important for flood control and provide valuable

forest habitat.

Although forested floodplains are more common in the southeastern United States,

herbaceous wetlands can also be found adjacent to rivers and streams. In riverine

wetlands there is a narrow opportunity for colonization between the exposure of alluvial

sediments and the return of high water levels and erosional forces (Willby et al. 2001).

Large rivers whose extensive floods deposit sediments in adjacent wetlands may have

poorly developed riparian marshes.









Riverine marshes are included in this study because there are few studies

comparing nutrient cycling between forested and herbaceous systems. Hopkinson (1992)

concluded that the growth form of the dominant vegetation does not influence nutrient

retention, although study results showed that a forested riverine system retained slightly

more nutrients than a riparian marsh (4.3% vs. <1%). Woody vegetation serves as long-

term storage of nutrients, while herbaceous vegetation of marshes provides mainly short-

term storage (Reddy and D'Angelo 1994). These differences may lead to

biogeochemical differences between these wetland types.

Depressional wetlands (non-riverine) differ from riverine systems because they are

not directly influenced by hydrologic fluxes from rivers and streams. Non-riverine

wetlands rely on precipitation or groundwater inputs, which tend to have lower nutrient

loads than surface waters (Craft and Casey 2000). Hopkinson (1992) determined that

relatively closed marshes and swamps of Okefenokee Swamp retained 90% of inorganic

nutrient inputs, whereas small percentages were retained in riverine systems. He

concluded that the openness of a wetland determines nutrient loading, which is strongly

correlated with productivity, organic matter decomposition, and nutrient cycling.

Systems with low nutrient loading are more efficient at cycling nutrients and have lower

net primary production (Craft and Casey 2000). Therefore, riverine and non-riverine

wetlands within similar surrounding land-uses may naturally display different nutrient

concentrations, organic matter content, and biogeochemical processes.

Differences among riverine verses non-riverine systems and marshes verses

swamps hinder generalizations about wetlands (Table 1-1). One exception is that

excessive loading of nutrients can alter ecosystem dynamics. If wetland functions are to










be protected through development of numeric nutrient criteria, individual wetland types

may need to be studied along with their regional variation as background for sound

environmental regulations.

Defining Ecoregions

The Clean Water Action Plan of 1998 includes a spatial component in its

requirement to establish nutrient criteria by geographic regions. The USEPA is

addressing spatial variability via geographic regions called ecoregions, which are areas

with relatively homogenous ecosystems that differ from adj acent regions (Omernik and

Bailey 1997) and are based on geology, physiology, vegetation, climate, soils, wildlife,

and hydrology. Omernik (1987) divided the conterminous US into ecoregions based on

regional patterns resulting from the combination of component maps including land-use,

land-surface forms, potential natural vegetation, and soils.

The USEPA adopted and adapted Omemik's ecoregions and stratified them

hierarchically. Level I is the coarsest United States ecoregion and is composed of 15

ecological regions, Level II is represented by 52 regions, and Level III contains 84

ecoregions (Brewer 1999). Level III ecoregions with similar characteristics that

contribute to nutrient regimes were aggregated to create USEPA Nutrient Ecoregions

(Figure 1.1). The USEPA recommends that numeric nutrient criteria be established for

lakes/reservoirs, streams/rivers, estuaries, and wetlands within each of the Nutrient

Ecoregions.

The current study area includes wetlands within the Southeastemn Forested Plain

(IX), Southern Coastal Plain (XII), and Eastemn Coastal Plain (XIV) ecoregions.

Comparisons were made among the three ecoregions to determine if they are appropriate

aggregations for setting numeric nutrient criteria.










Table 1-1. Comparison of wetland characteristics reported in the literature


Parameter

Maj or source of inputs
(Craft and Casey 2000)

Connectivity to other systems
(Hopkinson 1992)


Riverine

Runoff



Open


Non-riverine

Precipitation



Closed


Nutrient Cycling
(Hopkinson 1992)


Less efficient


More efficient


Soil C:N ratios
(Craft and Casey 2000)

Parameter

Nutrient retention
(Wilby et al. 2001)

Biomass turnover rates
(Hopkinson 1992)

Live tissure N:P ratios
(Bedford et al. 1999)

Live tissue N:P ratios
(Bedford et al. 1999)


Similar


Swamps

Similar


Similar

Marshes

Similar


One magnitude lower


One magnitude higher


Greater


Lower


Suggest P-limitation
or co-limitation by N
and P


Less than 14, suggesting N-
limitation


Lower


Similar


Greater


Similar


Low, suggesting P-
limitation or co-
limitation by N and P


Litter %N (Bedford et al. 1999)


Litter %P (Bedford et al. 1999)


Soil N:P ratios
(Craft and Casey 2000)

Average water temperatures
(Lee and Bukaveckas 2002)

Algal growth
(Battle and Golladay 2001)


Greater, Suggesting P-
limitation


Cooler


Warmer



Greater


Low










D~ral Aggregatons oBf Level lH~ Ecoregion
jibr thr' jtionbalr Nutr~~ient Slrt~egy


recommendations (USEPA 2003)

ILmiin Nutrent andjn Causa Variables.n~a




const.l.rucio ~runoff. ..I Nopint~. 'P soulrces are mrhajr cnrbtrso uretst qai


sourc of~lAUR nonpoint nofutrientpp pollution~ Iin teUie ttsdemil ofriie




Miue11.S ethdlges i USEPA Ntreni cal guidance manualsi fotr establishingnmei


crieri ae bsedonlimiting nutrients, imlyng that priary roucio f latsi






limited by theoi nutrient that to is the lestav iabe relaties tote plainlt's requireentfr






growth. This concept is Liebig's Law of the Minimum (Smith et al. 1999). Nitrogen (N)










and phosphorus (P) are the nutrients most commonly cited as limiting plant growth

(Carpenter et al. 1998, Gusewell et al. 1998, Koerselman and Meuleman 1996, Smith et

al. 1999). Therefore, with increased loading of N and/or P to aquatic ecosystems,

primary production usually increases and can lead to eutrophication. The agent that

causes change in an ecosystem is referred to as the casual variable, while the factor that

reacts is called the response variable (USEPA 2000a). For example, when concentrations

of a limiting nutrient (casual variable) increase, the dominance of fast-growing species

(response variable) increases, and they replace less competitive species (Gusewell et al.

1998).

Eutrophication is the process whereby an aquatic ecosystem shifts from a low

nutrient (oligotrophic) to a highly productive, nutrient rich (eutrophic) system (Mitsch

and Gosselink 2000). If the shift is the result of human activities, the process is called

cultural eutrophication. Eutrophication is characterized by increased growth of algae

and/or macrophytes, which can hinder use of water for fishing, recreation, industry, and

domestic consumption. Decomposition of excessive algae and macrophytes reduces

oxygen supplies, which can lead to fish kills (Carpenter et al. 1998). Eutrophication can

also alter foodwebs, resulting in a loss of biodiversity (Carpenter et al. 1998, Smith et al.

1999). In fact, high species biodiversity has been correlated with low nutrient regimes

(Bedford et al. 1999).

Preventing input of nutrients from anthropogenic sources does not necessarily

result in decreased plant growth, due to internal biogeochemical cycling of nutrients

within wetlands. Decomposition of stored organic matter can provide the nutrients

required for plant growth (Reddy and D'Angelo 1994). Nutrient transformations depend









on many factors, including hydrologic regime, influent nutrient concentrations, existing

nutrients in the system, vegetation, and sediments (Gopal 1999).

Predicting the extent of internal nutrient cycling in wetlands is difficult due to

inherent differences among wetland ecosystems. For example, nutrient cycling of

riverine and non-riverine wetlands is influenced by dissimilar hydrologic regimes.

Hopkinson (1992) found that the dominant plant growth form was the primary factor

influencing biomass turnover rates, with marshes cycling an order of magnitude greater

than swamps. Therefore, to determine nutrient effects in a wetland, it may be necessary

to examine several components of various wetland types.

If a limiting nutrient was always the factor limiting the system, it would be simple

to develop regulations. But aquatic systems are dynamic, and several factors can affect

production. For example, plant biomass changes seasonally, fluctuates with land-use,

and varies regionally (USEPA 2000a). Therefore, to establish numeric nutrient criteria, it

is necessary to develop an efficient tool for quantifying the nutrient regime of wetlands.

An effective nutrient indicator must be sensitive to varying nutrient regimes, easy to

measure and interpret, inexpensive to apply, and should have as few temporal and spatial

constraints as possible. The USEPA is exploring biological and/or chemical indicators

(or indices) to assess ecosystem integrity.

Biological indicators of nutrient enrichment

Biological assessments of wetlands often look at community-level parameters such

as abundance, biomass, density, richness, diversity, and community composition as

indicators of anthropogenic stressors (Adamus and Brandt 1990). Galatowitsch et al.

(1999) looked at possible plant, bird, invertebrate, fish, and amphibian metrics in eight

wetland types in Minnesota. Their results indicate that specific metrics would have to be









developed for the different wetland types and ecoregions. A study in the prairie pothole

region of the US looked at the value of macrophyte abundance, species richness, and

amounts of litter and standing dead vegetation as indicators of wetland health. None of

the examined indicators successfully quantified ecosystem health (Kantrud and Newton

1996). However, Lane et al. (2004) successfully developed a wetland condition index

based on macrophytes, macroinvertebrates, and diatoms for isolated depressional marshes

of peninsular Florida.

As nutrient levels in a wetland increase, the chemical structure of the system is

altered, leading to biological changes. Microbes are normally first to respond to nutrient

pulses with algae following closely behind. There is a time lag between casual variables

and response variables, particularly in long-lived species (Fennessy et al. 2001).

Biological indicators often rely on the response of larger organisms such as plants,

invertebrates, and birds (Galatowitsch et al. 1999, Kantrud and Newton 1996, Lane et al.

2004). Once organisms respond to a change in the nutrient regime, some of the original

structure of the wetland is lost as the new community evolves. One concern with using

macrophyte structure as an indicator is that once a wetland has been dominated by stress

tolerant perennials, less aggressive species may not be capable of re-colonization after the

stress is removed (Galatowitsch et al. 1999). The community structure may be a relic of

past disturbances.

Biogeochemical indicators of nutrient enrichment

There is a well-documented correlation between nutrient additions to aquatic

ecosystems and proportional increased growth of algae and macrophytes (Carpenter et al.

1998, Morris 1991, Smith et al. 1999). Likewise, elevated P and N levels have been

associated with decreased species diversity (Bedford et al. 1999, Carpenter et al. 1998,









Morris 1991, and Smith et al. 1999). Phosphorus and nitrogen chemistry of aquatic

ecosystems can be monitored to determine the degree of enrichment before species are

eliminated. This is important since 14% of the 130 plant species in the conterminous US

listed as endangered or threatened are found primarily in wetlands (Morris 1991).

Biogeochemical processes, such as organic matter decomposition and

denitrification, can reflect nutrient budgets before responses are evident in higher

organisms (Reddy and D'Angelo 1997). Nitrogen to phosphorus ratios (N:P) in plant

tissue (Gusewell and Koerseleman 2002, Gusewell et al. 1998, Koerseleman and

Meuleman 1996, Shaver and Melillo 1984, Wilby et al. 2001), soil (Craft and Casey

2000) and litter (Baker et al. 2001, Shaver and Melillo 1984) have been studied to assess

nutrient limitation in wetlands. Koerseleman and Meuleman (1996) concluded that when

N and P are controlling plant growth in wetlands; vegetation N:P ratios > 16 indicate P

limitation, while N:P ratios < 14 indicate N limitation.

There is disagreement in the literature regarding the limitation of wetland

productivity. Morris (1991) reviewed several wetland studies and concluded that most

wetlands are N limited. The results from numerous wetlands in Scotland, France, and

Ireland agree that most wetlands are N limited (Wilby et al. 2001). However, Craft and

Casey (2000) suggest that freshwater marshes and forested wetlands of southwestern

Georgia are P limited. Bedford et al. (1999) concluded that within temperate freshwater

wetlands of North America, marshes are N limited, while evergreen, shrub, and

deciduous wetlands are P limited. This confusion demonstrates a need for additional

information regarding nutrient regimes of wetlands. This study includes a










biogeochemical characterization of the surveyed wetlands to determine background

levels of nutrients and biogeochemical processes in minimally impaired wetlands.

Reference Wetlands

Establishing numeric criteria for wetlands requires determination of reference

conditions as a standard for comparison. One strategy for determining reference values is

to survey wetlands representing the broad range of nutrient impairment. The lower 25th

percentile of this population would be recommended as reference conditions (Figure 1-2).

An alternative strategy explored in this study is to set reference conditions equal to the

upper 25th percentile (or 75th percentile) of wetlands identified as minimally impaired

systems (USEPA 2000a). Eventually, individual waterbodies will be sampled and

compared to reference conditions to determine appropriate management methods

(USEPA 2000b).



Minimally Representative
Impaired of all Wetlands
Wetlands






1igher warar qulry g Lower muserquantyy
0 5 10 15 20 25 30 SG 4g0 45 50 tE 00 65 70 75 100
Total phoorphoruo (ug.1..)



Figure 1-2.Two approaches for establishing reference conditions using total phosphorus
as the example variable (modified from USEPA 2000a)










Ideally, reference conditions should reflect conditions in the absence of

anthropogenic influences and pollution. However, human activities have impacted all

ecosystems to some degree; therefore, reference conditions realistically represent the

least impacted conditions. The USEPA Science Advisory Board endorses use of

conditions representing minimal impact as a baseline that should protect the beneficial

uses (or designated uses) of aquatic resources (USEPA 2000a). The results of this study

will help determine appropriate reference conditions for developing numeric nutrient

criteria.

Research Objectives

The survey of this thesis will assess background nutrient concentrations in wetlands

to define water quality required to maintain ecological integrity. An additional goal of

this research is to explore differences in nutrient regimes among various wetland types to

determine appropriate wetland aggregation schemes for setting criteria. Results from this

comparison may be instrumental in developing nutrient criteria that are sufficiently

protective and feasible. Furthermore, regional aggregates will be explored to gain

additional understanding of the spatial component of wetland nutrient regimes within the

southeastern United States.

Hypotheses

The "openness" of a system to hydrologic and material influxes influences its

nutrient loading and productivity (Hopkinson 1992). Riverine systems are open to such

influxes and typically act as sinks for sediment and phosphorus from the contributing

watershed (Craft and Casey 2000), whereas non-riverine systems are considerably less

open to influxes. It is hypothesized that riverine wetlands will have higher nutrient levels

within soil and water compared to non-riverine systems.










In open systems with high nutrient influx, plants have less efficient nutrient cycling

and reabsorb fewer nutrients from senescing leaves (Hopkinson 1992). Hence, nutrient

content of leaf litter is greater in areas with increased nutrient availability (Shaver and

Melillo 1984). It is hypothesized that riverine wetlands will have increased nutrient

levels in leaf litter compared to non-riverine wetlands.

The structure of marshes and swamps is quite different, with the former

characterized by herbaceous vegetation and the latter by woody growth forms. In marsh

ecosystems, the maj ority of C, N, and P is stored in the soil, whereas swamps store a

great deal of C, N, and P in plant biomass (Hopkinson 1992). Additionally, biomass

turnover rates are an order of magnitude greater in marshes than swamps (Hopkinson

1992). This continual decomposition of herbaceous organic matter releases nutrients into

the soil; therefore, it is hypothesized that marshes will have higher nutrient levels in soil

than swamps.

Battle and Golladay (2001) found that sedge marshes have higher algal growth than

cypress swamps. This difference likely reflects the absence of overstory cover in

marshes. Algal populations quickly sequester nutrients from the water column, hence

decreasing soluble nutrients available to other growth forms (Kadlec and Knight 1996).

It is hypothesized that the water column of marshes will have lower N and P content than

swamps.

There is a spatial component to the nutrient regimes of wetlands, as recognized by

the USEPA' s use of ecoregions in determining numeric nutrient criteria. The various

geologic formations of the southeastern United States affect hydrology. Hydrology is

often cited as the most important defining parameter of wetland systems (Ehrenfeld and






17


Schneider 1991, Fennessy and Mitsch 2001, Jones et al. 2000, Reinelt et al. 1998).

Therefore, it is hypothesized that there will be regional differences in the nutrient regimes

of wetlands.















CHAPTER 2
IVETHOD S

To address the research goal of determining background concentrations of nutrients

in minimally impaired wetlands, it was necessary to locate and survey several types of

wetlands within areas of nominal anthropogenic disturbances. Two vegetative and two

hydrologic classes were selected, resulting in four wetland classes. The four wetland

types surveyed were riverine marshes, non-riverine marshes, riverine swamps, and non-

riverine swamps. To evaluate the spatial component of wetland nutrient regimes,

selection of wetlands was stratified within three USEPA Nutrient Ecoregions

(Southeastern Forested Plains, Southern Coastal Plains, and Eastern Coastal Plains) in the

southeastern United States.

Site Selection

The site selection process identified minimally impaired wetlands within three

ecoregions of the southeastern United States that met the criteria for wetland community

type (marsh versus swamp), accessibility (proximity to forest roads and within public

ownership), and hydrologic connectivity riverinee verses non-riverine). The large spatial

extent of the study area necessitated a Geographic Information System (GIS) for locating

sampling sites and analyzing spatial relationships. All GIS analysis was done using

ArcGIS 8.1.

Identifying Minimally Impaired Sites

Nutrient enrichment is often a result of fertilizer runoff from surrounding

agricultural and urban areas. It was assumed, as supported by Kantrud and Newton










(1996), that wetlands located close to agricultural areas would have greater nutrient

loading than those farther from intensive agricultural activities. Locating wetlands that

were not influenced by agriculture was problematic due to the scale of the survey. The

study area included Florida, Georgia, Alabama, and South Carolina. Collecting and

analyzing detailed land-use data for the entire study area was not logistically feasible.

Furthermore, utilizing data from various sources (such as four state agencies) can be

difficult to integrate because of different scales and varying standards for data quality and

collection.

The USEPA' s suggestion to use sites located within the boundaries of public lands

as minimally impaired wetlands was adopted (USEPA 2000a). It is likely that these sites

are less influenced by cultural nutrient enrichment than wetlands on private lands, as

indicated by a Landscape Development Intensity (LDI) Index for assessing the intensity

of various land-uses (Brown and Vivas in press). The index utilizes calculated LDI

coefficients ranging from 1.0 (natural systems) to 7.0 (high intensity agricultural).

Forestry is a common land-use on public lands. The LDI coefficient for pine plantations

is 1.58, which indicates minimal influence of wetlands near silviculture activity.

A public lands coverage was obtained and overlayed on the USEPA Nutrient

Ecoregion map (Figure 1-1). The largest public land tracts in the southeastern United

States lie within the boundaries of National Forests; therefore, efforts were concentrated

on identifying National Forests within the three USEPA Nutrient Ecoregions of the

Southeastern US (Figure 2-1). Permits were obtained to sample within Apalachicola,

Conecuh, Francis Marion, Ocala, Oconee, Osceola, Sumter, and Talladega (Oakmulgee

District) National Forests. Because Georgia had considerable aerial gaps without



























































Figure 2-1. Sampling areas within the three USEPA Nutrient Ecoregions.

Identifying Wetland Community Types

Once an area was selected, it was necessary to identify the wetlands present,

categorize them into the four target community types, and randomly select sampling


National Forest lands, portions of Fort Benning Military Preserve, Moody Air Force

Base, and Banks Lake National Wildlife Refuge were also sampled.


SoutyernCosaPli










120 240


Legend










SOconee NF
SOsceola NF
SSumter NF
STalladega NF


480 Kilometers









locations. To complete this task, the United States Fish and Wildlife Service (USFWS)

National Wetlands Inventory (NWI) was utilized. NWI maps were created through

photo-interpretation of aerial photography supplemented by soil surveys and Hield

verification. (fttp ://www.nwi.fws.gov/arcdata/readme.txt, 2002). NWI data were

downloaded in 7.5 minute quadrangles from the USFWS website.

Classifieation of wetlands on NWI maps was based on the USFWS Wetland and

Deepwater Habitat Classification System (Cowardin et al. 1979), which groups

ecologically similar habitats together (Tiner 1999). For this study, swamps are analogous

to Cowardin's forested wetlands, which include wetlands characterized by woody

vegetation at least six meters tall. Marsh sites correspond with Cowardin's emergent

wetland class characterized by erect, rooted, herbaceous hydrophytes that are present for

most of the growing season. For this study, eleven NWI sub-class level communities

were aggregated into two community types (Table 2-1).

NWI data were not available for Talladega and Conecuh National Forests in

Alabama. A hydric soils shapefi1e was obtained from United States Forest Service

(USFS) personnel and used to identify wetlands at these sites. Community types were

determined during the Alabama site visits, since this distinction could not be made with

available GIS data.

Hydrologic Classification

Mitsch and Gosselink. (2000) defined riparian wetlands as those ecosystems

located where streams or rivers at least occasionally flood beyond their confined

channels. The littoral zone of lakes is often lumped into the riparian wetland

classification. To decrease variability among sampled wetlands, those adjacent to rivers

and streams were included in this study, while littoral wetlands of lakes were excluded.










Table 2-1.The NWI classification scheme aggregated into swamp and marsh wetland
types


NWI Classification


Current Study
Classification








Swamp


System
Palustrine

Palustrine

Palustrine

Palustrine

Palustrine

Palustrine

Palustrine

Palustrine

Palustrine

Riverine

Riverine


Subsystem


Class
Forested

Forested

Forested

Forested

Forested

Forested

Forested

Emergent

Emergent

Emergent

Emergent


Sub-class
Broad-leaved
Deciduous
Needle-leaved
Deciduous
Broad-leaved
Evergreen
Needle-leaved
Evergreen
Dead

Indeterminate
Deciduous
Indeterminate
Evergreen
Persi stent

Non-persi stent

Non-persi stent

Non-persi stent


Marsh


Tidal

Lower
Perennial


To identify riverine wetlands, proximity of wetlands to streams and rivers was

determined using stream data from various sources. The National Hydrography Dataset

(NHD), compiled by USGS at a scale of 1:100,000, was utilized for the three Florida

National Forests. Stream data for the remaining locations were obtained from USFS

staff. The majority of the stream data provided was also compiled by USGS. Wetlands

located at least partially within 40 meters of a river or stream were classified as riverine.

Upstream activities must be considered when classifying these wetlands as

minimally impaired. To avoid this complication, wetlands along small streams (first and

second order) were targeted because their headwaters were often within the forest

boundaries. Larger rivers not originating within the boundaries of National Forests were









not included in the survey due to concerns that agricultural and urban activities outside

the forest, but within the watershed, may change the desired least impaired status.

There are several definitions of isolated wetlands in use. Tiner et al. (2002) defined

a wetland as isolated if it is geographically isolated from other wetlands by uplands.

Winter and LaBaugh (2003) suggested that isolated wetlands are those not connected by

streams to other surface-water bodies. Common to both definitions is the absence of

hydrologic connectivity between the wetland in question and surrounding water bodies.

Regardless of definition, classifying isolated systems can be difficult, especially during

extremely wet years when surface water overflows connect "isolated" systems to other

aquatic ecosystems. To eliminate confusion surrounding classification of isolated

wetlands, sites were divided into riverine (as defined above) and non-riverine, as defined

by those wetlands that are at least 40 meters from rivers and streams.

Site Selection Criteria

After wetlands were categorized by vegetation type and hydrologic connectivity,

proximity to potential nutrient sources and accessibility was determined. A property

ownership shapefile was obtained from USFS personnel to identify tracts of land under

private ownership within the forest boundaries. Wetlands located on private property

were omitted from the survey. Forest Service road coverages were added to the map

proj ects to ensure that the wetlands were accessible. All of the forests had extensive road

systems; therefore, it was not necessary to omit sites due to accessibility concerns.

Wetland sampling sites were determined by assigning a number to each of the

individual wetland polygons that met community type and hydrologic connectivity

criteria and that were not omitted due to private ownership. A random number generator










was used to select those non-riverine swamps, riverine swamps, non-riverine marshes,

and riverine marshes to be sampled.

For each public land tract, approximately 30 wetlands meeting the selection criteria

were identified; although only 12 (three from each class) were sampled. The additional

sites were necessary to compensate for any sites that could not be sampled due to GIS

coverage error, misclassification, inaccessibility, or other unexpected issues.

The goal was to sample three wetlands of each community type riverinee marsh,

non-riverine marsh, riverine swamp, and non-riverine swamp) within each public land

tract. However, with the exception of the Ocala and Oconee National Forests, marsh

communities were scarce. Furthermore, as topographic relief increased in the northern

and western extents of the study area, non-riverine systems became less prevalent.

Therefore, wetland community types were sampled in proportion to their relative

abundance (Figure 2-2). More swamps were sampled than marshes, and the majority of

surveyed wetlands were riverine systems. A total of 103 minimally impaired wetlands

were surveyed.

Sampling and Analytical Protocols

Sample Locations

Selected wetlands were physically located using a GPS unit, topographic maps, and

the coordinates of the selected sites. Ground truthing least impaired status, vegetative

community type, and hydrologic connectivity was always a first step when visiting

wetlands. If GIS classification was not verified on the ground, the site was reclassified or

not sampled.










70
-0 58
5 60


e> 4 0
30 -2

20 -0
2 ~8 1
E 10 -
E 0

Riveri ne Riverine No n-ri veri ne Non-riverine
Swamp Marsh Swamp Marsh



Figure 2-2. Number of wetlands surveyed aggregated by community type

A visual survey was conducted upon arrival, and the wetland was divided into two

general zones, referred to as the core wetland and the edge wetland. (Figure 2-3). In

riverine systems, the core (C) was adj acent to the stream, but landward of any natural

levees that have formed. The edge (E) of riverine wetlands was located parallel to the

adj acent upland, approximately 25 % of the distance between the upland and the stream.

With small non-riverine wetlands, it was possible to walk the entire edge (E) of the

wetland and sample the four cardinal points at approximately 25 % of the distance

between the upland and the center of the wetland. The center was sampled as the core

(C).

In large non-riverine systems, only one side of the wetland was sampled, as if it

was a section of a riverine wetland. The core (C) was located in the deep center of the

wetland, and the edge (E) was located parallel to the upland side of the wetland

approximately 25 % of the distance between the upland and the center of the wetland.









Within the edge (E) and the core (C), three sub-sample sites were located

approximately 30 paces from each other. Transects were typically orientated parallel to

the upland boundary. To prevent bias, a PVC ring was tossed into the air after 30 paces

had been traversed, and where it landed marked the sampling location. At each sub-

sample location, water (if present), soil, and leaf litter were collected. A characterization

form (Appendix A) that included a visual vegetation survey, hydrologic characteristics,

and other descriptive information was completed at each sub-sample location.

A B



Upland





Ecotone
Upland
(not sampled)
Edge Core River
Upland
















Figure2-3. Sub-sample locations. A) Within the core and edge zones of riverine wetlands.
B) Small non-riverine wetlands. C) Large non-riverine wetlands.









A handheld YSI-556 meter (Yellow Springs, CO) was used to record water column

pH, dissolved oxygen saturation, temperature, redox potential (Eh), and conductivity at

each sub-sample location with water present. Redox potential was measured as ORP

with an Ag/Ag-cl electrode. Values were converted to Eh by adding 234 mV to each

reading. Measurements were made with the probe suspended at mid-depth of the water

column, but in shallow wetlands (less than 15 cm), the probe was often placed at the

sediment-water interface.

Sample Collection and Processing

Sampling began in April 2003. The survey began with the most southern sites

(Ocala National Forest) and then proceeded to the north. Most of the sampling was

completed by August 2003. Moody Airforce Base, Fort Benning, and Banks Lake

National wildlife Refuge were sampled in September 2003.

Water

Water was collected, when present, using acid-washed 125-mL HDPE bottles. The

bottles were rinsed three times with site water prior to collecting the sample. Care was

taken to minimize non-representative particulates in the water column; however, the

water column often contained particulate matter that was included with the sample.

Samples collected at the three sub-sample locations along transect C or E were poured

into a pre-acidified (concentrated sulfuric acid) 500-mL HDPE bottle to create the zone

composite. Water samples were stored on ice for transport to the Wetland

Biogeochemistry Laboratory at the University of Florida (Gainesville, FL).

In the laboratory, a sub-sample of the water composite was filtered through 0.45

CIM filter paper and analyzed for nitrate and nitrite on a rapid-flow analyzer (Table 2-2).

An additional (non-filtered) 10 ml sub-sample was digested for Total Kjendal Nitrogen










(TKN) analysis. Results from nitrate/nitrite and TKN analyses were added together to

determine total nitrogen concentrations. Total phosphorus (TP) was determined on a

third sub-sample (10 ml) by sulfuric acid and potassium persulfate digestion (EPA

method 365.1 1993), followed by colorimetric analysis (Technicon AA II).

Soil

Three soil samples were collected along each of the wetland transects. Prior to

sampling, litter and live vegetation were removed from the sampling area by lightly

raking the area by hand. A pre-cleaned tenite butyrate tube (7.3 cm. diameter) was driven

into the soil at least 10 cm deep. The core tube was then placed on an extruder piston,

which was used to push the top of the soil out of the core and into a 10 cm tenite butyrate

collar. Any litter remaining on the top of the core was removed and discarded. The 10 cm

core was sliced from the remainder of the core using a stainless steel bread knife and

placed in a re-sealable bag. Soils from the three sub-sample sites along transect C or E

were combined to create a composite sample. Samples were stored on ice for transport to

the laboratory.

Coring of soils in densely rooted environments was facilitated by using a coring

devise with a sharp coring head attached to make cutting through roots possible and to

avoid compacting the sample. An effort was made to avoid large roots, which complicate

bulk density calculations. Several swamps, however, contained large root mats, making

it impossible to avoid coring through large amounts of root material.

In the laboratory, wet weight of the composite sample was recorded for bulk

density calculations. Roots larger than 2 mm in diameter were removed from the sample

and discarded. The composite sample was homogenized. A sub-sample was placed in a

shallow 250 mL container, weighed, then dried at 21oC for at least 48 hours.









Table 2-2.Summary of chemical analyses and methods for each stratum sampled

Medium Analysis Method
TP Sulfuric acid and potassium persulfate digestion
followed by colorimetric analysis
Water TKN Sulfuric acid digestion

NO2-No3 Rapid Flow Analyzer (RFA)

Organic matter content Lost on Ignition (LOI)
Carlos Erba NA 1500 CNS Analyzer (Haak
TN
Soil Buchler instruments Saddlebrook<, NJ)
Carlos Erba NA 1500 CNS Analyzer (Haak
TC
Buchler instruments Saddlebrook, NJ)
TP Ignition Method (Anderson 1976)

TP Ignition Method (Anderson 1976)
Carlos Erba NA 1500 CNS Analyzer (Haak
Litter TN
Buchler instruments Saddlebrook, NJ)
Carlos Erba NA 1500 CNS Analyzer (Haak
TC
Buchler instruments Saddlebrook, NJ)

The dry sample was re-weighed for percent moisture calculations. Dry samples were

hand-ground using a mortar and pestle, then further ground mechanically using a ball mill

grinder for at least eight minutes. The ground samples were passed through a 1 mm sieve

for quality control purposes and placed in scintillation bottles for analyses. Soil samples

were analyzed for organic matter content by loss on ignition (LOI), total nitrogen (TN),

total carbon (TC), and TP, as summarized in Table 2-2.

Leaf litter

Leaf litter samples were collected by placing a 40 cm diameter PVC ring on the soil

surface and hand-collecting all loose material within the ring. Collection was

discontinued when the soil surface was reached, as indicated by the presence of fine,

well-decomposed materials. Litter sampling was qualitative, not quantitative, since at









times it was necessary to collect multiple samples at a sub-sample location to ensure

adequate material for analysis. Litter samples from the three sub-sample locations were

combined to form a composite sample along the core or edge transect. All samples were

stored on ice for transport to the laboratory.

In the laboratory, litter samples were placed in a paper bag and dried at 21o C for at

least 72 hours. The dry samples were coarsely ground in a Willey mill to pass through a

1 mm screen. The samples were then further ground to pass through a 40-micron

followed by an 80-micron screen. To reduce cross-contamination, the mills were

vacuumed between each sample. The litter was analyzed for TP, TN, and TC (Table 2-2).

Data Analysis

All data were analyzed using JMP 4 (1989) software. Shapiro-Wilks normality test

was used to describe the distribution of data. When appropriate, data were log

transformed for further analysis. Mahalanobis distance was used to identify and remove

extreme outliers. Matched pairs t-tests were used to determine differences between the

core and edge sampling locations within wetlands. O'Brien's test was used to determine

if there was equal variance between the populations. Populations with equal variance

were compared using a standard t-test. Populations with unequal variance, or non-normal

distributions, were compared using a Welch ANOVA test for unequal variance. An alpha

level of 0.05 was used as a threshold for determining when differences were significant.

When a significant difference did not exist between treatments, a power test was

applied. Power addresses Type II errors, in which there is a failure to rej ect a false null

hypothesis (Rotenberry and Wiens 1985). When a significant difference is not found, as

indicated by a high p value, it is often assumed that there is no difference between the

populations compared. However, there may be differences that were not expressed due









to the limited number of samples compared. Power can be used to determine the

probability of finding a significant difference. As the probability of significant

differences increases, so does the power. Included in the power test is the Least

Significant Number (LSN). The LSN is defined as the number of observations needed to

decrease the variance enough to achieve a significant result with the given values of

significance level, standard deviation of the error, and effect size (JMP 4 1989 Help

Files).















CHAPTER 3
RESULTS AND DISCUSSION

Water column, litter, and soil data from 103 minimally impaired wetlands in the

southeastern United States were analyzed. The goals were to characterize nutrient

conditions within these wetlands and determine whether differences within wetlands,

among wetland types, and between USEPA Nutrient Ecoregions were present. Results

and discussion of findings will be presented in three separate sections: within wetland

variability, variability among wetland types, and spatial variability.

There are several ways to aggregate the surveyed wetland data based on the

question of interest (Table 3-1). Aggregating by hydrologic connectivity allows for a

comparison of riverine and non-riverine systems, whereas aggregating by vegetative type

allows for a comparison between marshes and swamps. The most specific aggregation

integrates both hydrologic connectivity and vegetative type resulting in four separate

wetland community types; riverine swamps, non-riverine swamps, riverine marshes, and

non-riverine marshes.

Aerial coverage of the four wetland community types was not evenly distributed

throughout the study area (Figure 3-1). Swamps were more prevalent than marshes, and

riverine marshes were practically non-existent in the northern and western extents of the

study area. Wetland community types were sampled in proportion to their relative

abundance; therefore, there are unequal sample sizes for each wetland community type.

It is important to keep the unequal distribution in mind when comparing the various

aggregations of wetlands. For example, 83% of the surveyed wetlands are swamps.










Comparisons based on hydrologic connectivity are biased towards riverine and non-

riverine swamps, since only 17% of the systems compared were marshes. Similarly, 64%

of the surveyed wetlands are riverine systems, which may influence distinctions between

marshes and swamps.

Table 3-1. Various aggregations of the wetlands surveyed in this study

Number
Grouping Criteria Aggregation
Surveyed
None All Wetlands Combined 103


Hydrologic Riverine 66
Connectivity Non-riverine 37


Marsh 18
Vegetative Type
Swamp 85


Non-riverine Swamp 27
Wetland Community Riverine Swamp 58
Type Non-riverine Marsh 10
Riverine Marsh 8

The surveyed wetlands are not only unequal in abundance, but also in regional

distribution (Figure 3-2). The surveyed wetlands are distributed throughout four

southeastern states, which include three USEPA Nutrient Ecoregions (Figure 2-1). The

Southeastern Forested Plain contained 62% of the surveyed swamps, 50% of the marshes,

67% of the riverine, and 47% of the non-riverine wetlands. The Southern Coastal Plain

had 25% of the swamps, 44% of surveyed marshes, 21% of the riverine, and 42% of the

non-riverine systems. The least represented ecoregion was the Eastern Coastal Plain with

only 13% of the swamps, 5% of the marshes, 12% of the riverine, and 1 1% of the non-

riverine wetlands. Comparisons among wetland types (aggregated by hydrologic












H Oconee
90,000 Apalachicola
I Ocala
go,000 O1 O sceola
I Oconee
I Sumter
70,000
O Francis Marion

60,000


S50,000


40,000,


30,000


20,000


10,000



Non-riverine Riverine Non-riverine Riverine
Swamp Swamp Marsh Marsh


Figure 3-1.Total area of the four wetland types within seven of the surveyed national
forests




I Southeastern Forested Plain

I Southern Coastal Plain

O Eastern Coastal Plain


Swamp


Marsh


Non-riverine


Riverine


Figure 3-2.Percentage distribution of surveyed wetlands within ecoregions, aggregated by
vegetation type (swamps and marshes) and by hydrologic connectivity (non-
riverine and riverine).









connectivity or vegetative type) combined sites in all three ecoregions. Results may be

biased by characteristics of the Southeastern Forested Plain since the wetlands are not

equally distributed among the three ecoregions.

Within Wetland Variability

Surveyed wetlands were sampled along two transects. One transect was located in

the core area of the wetland while the other was parallel to the upland edge of the

wetland. Specific locations of sampling transects within each wetland are detailed in

Chapter 2 (Figure 2-3). Before comparisons were made among wetland types, possible

within wetland variability was investigated. Samples collected within the core area and

from the edge area within each wetland were compared using pair-wise analysis. A

comparison of physical and chemical attributes between the core and edge transects was

evaluated for water column, litter, and soil strata.



Water column

Core areas were significantly deeper (p<0.05) than edge areas for all aggregations

of wetlands compared (all wetlands combined, swamps, marshes, riverine, and non-

riverine). There were no significant differences in water column temperature, dissolved

oxygen saturation, pH, or conductivity between core and edge sites (p>0.05). Many of

the surveyed wetlands were narrow linear systems with short distances between the core

and edge areas. Therefore, similar water chemistry and physical characteristics within

the core and edge areas are not surprising since the water is probably well mixed.

Water was not always present within each zone of the wetland (core and edge), and

some wetlands had no standing water at the time of sampling. Only 52 of 103

sampled wetlands had water within both zones. Of these 52 wetlands, 34 were swamps









and 18 were marshes. There were 26 riverine and 26 non-riverine systems with water

present in each zone.

Table 3-2. Results of pair-wise comparison of core and edge areas for various
aggregations of surveyed wetlands. "n" represents the number of wetlands
compared, and "p-" is the probability value from the pair-wise comparison.
Significant differences (p<0.05) are denoted by bold values.
Water column TP Water column TN
Grouping Criteria Aggregation
n p n P
None All Wetlands Combined 50 0. 01 7e 97 0.109

Marsh 17 0. 024e 15 0.806
Vegetative Type
Swamp 33 0.188 82 0.111


Hydrologic Non-riverine 25 0.104 34 0.181
Connectivity Riverine 25 0.09 63 0.211


Non-riverine Swamp 13 0.808 26 0.023c
Wetland Non-riverine Marsh 12 0. 01 0e 8 0.903
Community Type Riverine Swamp 20 0.079 56 0.222
Riverine Marsh 5 0.88 7 0.783
e significantly greater values in edge areas
a significantly greater values in core areas

A nutrient comparison of core and edge samples within these wetlands (Table 3-2)

indicates that the edge sites had significantly higher water column total phosphorus (TP)

concentrations (0.132 + 0.147 mg/L) than core sites (0.098 + 0. 147 mg/L). Total

Nitrogen (TN) was also greater at edge than core sites, but the difference was not

significant (p=0. 109). Water column TP and TN of core and edge sites were also

compared for various aggregations. Edge locations had significantly greater water

column TP for three grouping strategies: all wetlands combined, marshes, and non-

riverine marshes. Elevated water column TP values in edge samples may indicate that









nutrients are being introduced to wetlands from adj acent uplands or there is increased

mineralization of nutrients at the shallower edge sites.

Litter

Litter at edge sites had significantly greater total carbon (TC) content and similar

TP and TN values compared to litter at core sites. Litter of core and edge sites was

compared for the various aggregations of wetlands (Table 3-3).

Table 3-3. Results of pair-wise comparison of core and edge areas for various
aggregations of surveyed wetlands. "n" represents the number of wetlands
compared, and "p-" is the probability value from the pair-wise comparison.
Significant differences (p<0.05) are denoted by bold values.
Grouping Litter TC Litter TN Litter TP
Aggregation
Criteria n p n p n p
All Wetlands
None .90 0. 023e 97 0.109 83 0.799
Combined


Vegetative Marsh 15 0.58 15 0.806 14 0.486
Type Swamp 82 0. 012e 82 0.111 69 0.783


Hydrologic Non-riverine 34 0.7712 34 0.181 29 0.487
Connectivity Riverine 63 0. 01 7e 63 0.211 54 0.378

Non-riverine Swamp 26 0.526 26 0.023c 20 0.821
Wetland Non-riverine Marsh 8 0.575 8 0.903 8 0.753
Community
Type Riverine Swamp 56 0.014e 56 0.222 48 0.852
Riverine Marsh 7 0.883 7 0.783 5 0.038C
e significantly greater values in edge areas
a significantly greater values in core areas

Edge locations had significantly greater litter TC content for all wetlands

combined, swamp vegetative type, riverine hydrologic regime, and riverine swamp

community type. One possibility for the higher carbon content at the edge of these

wetlands is that core wetland areas (within riverine systems) are adj acent to the stream

channel. Therefore, the core is more susceptible to high velocity flow, which can transfer










organic matter downstream while depositing inorganic sediments. Inorganic material

deposited on leaf litter was often integrated into samples from core locations. These

deposits may reduce carbon content at core sites.

Soil

When all wetlands were combined, soil carbon and nitrogen content was

significantly greater in core than edge areas, whereas phosphorus content was similar

within both areas (Table 3-4). Core areas had greater soil TC content for all aggregations

and increased TN when comparing all wetlands combined, swamp vegetative types, non-

riverine hydrologic regimes, and riverine marsh wetland communities. Soil TP content

was similar between the core and edge areas, except within non-riverine swamp

communities where core areas had significantly greater phosphorus content than edge

areas.

The core areas are significantly deeper than edge sites, which may lead to longer

hydroperiods and anaerobic conditions. Under anaerobic conditions, decomposition rates

are decreased, and levels of N, C, and P can build up in the soil. This may explain the

higher levels of these compounds in core versus edge sampling areas in some of the

aggregations of surveyed wetlands.

Discussion

The overall differences within wetlands indicate that samples collected at the edge

of a wetland will likely have greater water column TP, increased litter TC content, and

lower soil TC and TN content than samples collected within the core area of the same

wetland. These differences within wetlands suggest potential implications of inconsistent

sampling techniques on biogeochemical characterizations of wetlands. To minimize the









effects of within site variability on the findings of this research, only core site values

were used in comparisons for the remainder of this study.

Table 3-4. Results of pair-wise comparison of core and edge areas for various
aggregations of surveyed wetlands. "n" represents the number of wetlands
compared, and "p-" is the probability value from the pair-wise comparison.
Significant differences (p<0.05) are denoted by bold values.
Grouping Soil TC Soil TN Soil TP
Aggregation
Criteria n p n p n p

None All Wetlands Combined 93 0.0001c 95 0.023c 94 0.1


Vegetative Marsh 13 0.0001c 14 1 15 0.6
Type Swamp 78 0.0001c 79 0.043c 79 0.2


Hydrologic Non-riverine 32 0.0001c 32 0.035e 36 0.1
Connectivity Riverine 60 0.0001c 61 0 58 0.7

Non-riverine Swamp 25 0.0001c 25 0 28 0.017"
Wetland Non-riverine Marsh 6 0.0001c 7 1 8 0.8
Community
Type Riverine Swamp 53 0.0001c 54 0 51 1
Riverine Marsh 7 0.0003c 7 0.006" 7 0.1
e significantly greater values in edge areas
a significantly greater values in core areas

Variability among Wetland Types

Vegetative Comparisons: Swamps and Marshes

Surveyed wetlands can be aggregated by dominant vegetation into swamps and

marshes. Swamps are dominated by woody vegetation and include riverine swamps and

non-riverine swamps. Swamps were more ubiquitous in the landscape than marshes.

Therefore, 85 out of the 103 surveyed wetlands were swamps.

Marshes are characterized by herbaceous vegetation and are an aggregate of

riverine marshes and non-riverine marshes. Marshes were less common than swamps;

therefore, only 18 marshes were surveyed for this study. Differences between swamps










and marshes will be addressed in the following four sections: water column, litter, soil,

and discussion.

Water column

Swamps had significantly greater (p=0.0487) water column TP concentrations

compared to marshes (Table 3-5). Swamps exhibited slightly higher (p=0.5401) TN

values, but the trend was not significant (Figure 3-3). Water TP data partially support the

stated hypothesis that marshes would have lower water column nutrients compared to

swamps. This difference may be correlated with increased presence of algae in marshes

compared to swamps. Algae were present in 47% of surveyed marshes and only 10% of

surveyed swamps. Algae can quickly sequester water column P, hence lowering water

column TP in marshes (Kadlec and Knight 1996).

Table 3-5. Water column properties observed in minimally impaired wetlands aggregated
by vegetative type
Swamp Marsh
Parameters mean + SD median 75th n Significance mean + SD median 75th n
0.049 +
TP (mg/L) 0.108 + 0. 12 0.06 0.177 47 0.053 0.03 0.07 17
TN (mg/L) 2.24 + 1.44 1.88 2.79 48 1.82 + 0.64 1.76 2.31 18
Temp (oC) 21.9 + 3.0 21.9 24.3 36 ** 25.9 + 4.4 25.3 29.5 14
pH 4.9 + 1.1 4.9 5.9 36 5.3 + 1.1 5.2 6.4 14
DO (%) 28.2 + 21.1 24.3 42.3 36 38.1 + 24.8 39.4 55.7 14
Cond. (uS/cm) 69 + 49 68 82 36 54 + 39 47 89 12
Eh (my) 412 + 348 397 513 26 369 + 380 318 525 11
Depth (cm) 16.5 + 16.3 14 22.6 42 ** 41.6 + 27.7 47.5 63.5 15
Significant difference (p<0.05)
** Significant difference (p<0.01)
*** Significant difference (p<0.001)


Water column temperatures were significantly greater in marshes than swamps.

This is most likely due to the absence of an overstory of woody vegetation in marshes,

which may also contribute to increased marsh algal growth. The core areas of marshes











were significantly deeper than those of swamps. No significant differences were found


between swamps and marshes with respect to water column pH, dissolved oxygen


saturation, conductivity, or oxidation reduction potential.

0.7 9

0.6- 8-




0.5

0.4-5
E *
MarshSwampMarshSw4-
0. z

Fgr -. 3-te 1ounT n Nvle yvgttietp.Tedse iei





Marsh Swamp Marsh Swampps





(Figure 3-3., Water columgn TPan d TN values bytee vegetative type. Te idased line is th

and C: meanio lof each popuaionr aeend the swoli line aiste ovperalma. These bottomg ofr

withthe "box"tr eistew 25t pefrc entiale, and the to sis the 75h percentieThcnter

concntrion ae withnd the boxplots the2% medan. The whiskes extend) +r 1.5 th

siilr redin terunile range. Diff eraen lettersT iondcante a signfcnt difenc (p<0.05)4

~~~btenmarshes and 12 .9 o swamps.
















1.5- 40- .
0.03 o


0.02 .J

0.01 -1-

0 -10.5 ,,2
Marsh Swamp Marsh Swamp Marsh Swamp

A A A A A A

Figure 3-4 Litter phosphorus, nitrogen, and carbon values by community type. The dashed line is the mean of each population, and
the solid line is the overall mean. The bottom of the "box" is the 25th percentile, and the top is the 75th percentile. The
center line within the boxplot is the median. The whiskers extend + 1.5 the interquantile range. Different letters indicate
a significant difference (p<0.05) between marshes and swamps.





Table 3-6. Litter phosphorus, nitrogen, and carbon content observed in minimally impaired wetlands aggregated by vegetative type
Swamp Marsh
Parameters mean + SD median 75th n si nificance mean + SD median 75th n
P ( 0.015 + 0.01 0.011 0.021 69 0.019 + 0.018 0.012 0.033 15
N ( 1.25 + 0.29 1.22 1.44 82 1.43 + 0.44 1.33 1.8 15
C (%) 41.0 + 8.5 43.2 48.8 84 41.5 + 4.8 43.1 45.3 15
C/P ratio 4132 + 2958 3258 6838 67 5193 + 5021 3393 8316 14
C/N ratio 32.71 + 9.87 30.78 39.06 82 28.90 + 8.50 26.09 35.59 13


* Significant difference (p<0.05)
** Significant difference (p<0.01)
*** Significant difference (p<0.001)















Y 25, 40



40 10-I




Marsh Swamp Marsh Swamp Marsh Swamp

A A A A A A
Figure 3-5 Soil %P, %/N, and %C values by community type. The dashed line is the mean of each population, and the solid line is the
overall mean. The bottom of the "box" is the 25th percentile, and the top is the 75th percentile. The center line within the
boxplot is the median. The whiskers extend + 1.5 the interquantile range. Different letters indicate a significant difference
(p<0.05) between treatments
















75th
420
0.17
10.5
3.1
194
42.2
970
18.7
69.2


Parameters
TP (mg/kg)
TP (m/cm3
TN (g/kg)
TN (m/cm3
TC (/g
N/P ratio
C/P ratio
C/N ratio
LOI (%)


Bulk Density (/cm3
Moisture Content (%)


0.83 85
73 85


0.43 0.67 16
65 86 16


* Significant difference (p<0.05)
** Significant difference (p<0.01)
*** Signifieant difference (p<0.001)


n si nificance


Table 3-7.Soil P, N, and C content observed in minimally impaired wetlands aggregated by vegetative type
Swamp Marsh


mean + SD
410 + 260
0.19 + 0.15
5.9 + 5.5
2.01 + 0.77
123 + 140
17.6 + 17.7
361 + 329
19.4 + 5.5
28.5 + 26.9
0.57 + 0.35
53 + 21


median
350
0.14
3.7
1.87
64.3
14.4
277
18.6
16.8
0.57
50


75th
550
0.25
7.4
2.39
129
22.8
514
22
32.9


mean + SD
370 + 270
0.13 + 0.12
8.5 + 10.4
2.04 + 1.16
136 + 156
28.2 + 27.9
489 + 456
17.7 + 4.7
31.7 + 31.3
0.45 + 0.32
63 + 23


median
340
0.093
4.6
1.77
61.7
17.6
304
17.7
15.8









Soil

Comparisons of soil nutrient content between vegetative community types were

conducted on a mass per unit mass basis and a mass per unit volume basis. Typically it

would not be necessary to express nutrient content using these two methods. However,

due to the wide range in bulk density found among wetland sites, normalizing for bulk

density was desirable. Results of Eindings using both mass and volumetric measures of P

and N are presented and distinguished based on units.

Soils of swamps and marshes were similar with respect to TP(mg/kg), TN (g/kg),

TN (mg/cm3), TC(mg/kg), N:P ratio, C:P ratio, C:N ratio(on a mass basis), loss on

ignition (LOI), bulk density, and moisture content, with no significant differences

between vegetative types evident (Figure 3-5). However, when soil total phosphorus was

normalized by bulk density, swamps had significantly higher total phosphorus (mg/cm3)

then marshes.

Craft and Casey (2000) found that forested depressions in southwestern Georgia

had higher soil nitrogen and phosphorus concentrations, as well as lower N:P ratios,

compared to depressional marshes. In the current survey, soil TP (mg/cm3) was

significantly greater in swamps than marshes, but soil TP (mg/kg), TN (g/kg), TN

(mg/cm3), Or N:P ratio (Table 3-7) differences were not significant between marshes and

swamps. The mean N:P ratio was 18 for swamps and 28 for marshes. These values are

similar to the N:P ratios Craft and Casey (2000) reported for swamps that were thought to

be p-limited or co-limited by P and N. These results partially support the observations of

Hopkinson (1992) that growth form of dominant vegetation does not seem very important

in controlling nutrient regimes.










Discussion

Phosphorus and nitrogen data for surveyed swamps and marshes were compared to

values in the literature (Table 3-8). Phosphorus values were consistently greater in the

literature. This could be due to the least impaired status of wetlands included in this

survey. Nitrogen values were fairly consistent between the current study and the

literature.

Table 3-8. Values from the current study compared to those in the literature
Curet Bedford Ncos Whigham and
a et al. c Richardson
Stda 1999" 95 1988d
Water TP (mg/L) 0.049 + 0.053 0.248 0.520
Marsh Litter %P 0.019 + 0.018 0.16
Soil %P 0.037 + 0.027 0.25

Water TP (mg/L) 0. 108 + 0. 12 0.221 0.650
Swamp Litter %P 0.015 + 0.01 0.16
Soil %P 0.041 + 0.026 0.09 0.24

Water TN (mg/L) 1.819 + 0.636 2.09 2.67
Marsh
Litter %/N 1.43 + 0.44 1.22
Soil %N 0.85 + 1.04 1.41

Water TN (mg/L) 2.243 + 1.444 2.17 3.01
Swamp Litter %/N 1.25 + 0.29 1.04
Soil %N 0.59 + 0.56 1.28 1.5
a = Mean values with standard deviations for 103 minimally impaired wetlands within
the southeastern US
b = Mean values from a literature search of North American freshwater temperate
wetlands
c = Range of values for wetlands within Elk Island National Park, Alberta
d = Mean values from Acer rubrum swamps in Maryland, USA

The vegetative structure of marshes and swamps is quite different, with the former

characterized by herbaceous vegetation and the latter by woody growth forms. It seems

logical that wetlands with different vegetation types would have soils with varying

nutrient contents. According to Hopkinson (1992), swamps store a great deal of C, N,









and P in plant biomass, while marshes store the maj ority of C, N, and P in the soil.

Therefore, it was a hypothesis of this thesis that marshes would have greater soil nitrogen

and phosphorus content than swamps. The results of this survey do not support this

hypothesis. It was found that swamps had significantly greater TP (mg/cm3) than

marshes. These results could be influenced by the fact that 64% of the surveyed wetlands

are riverine systems which are often associated with higher nutrient concentrations.

There is a lot of variability within the data for swamps and marshes. The standard

deviation values are often as great as the mean values. This trend exists not only when

wetlands are aggregated by vegetation type, but also when all 103 wetlands are

combined. This large variability in nutrient content among wetlands likely explains why

there were minimal significant differences between swamps and marshes for the limited

number of sites surveyed.

An analysis of statistical power was used to understand the limited statistical

differences detected between vegetative community types. Power analysis (JMP 1989)

was applied to comparisons between swamps and marshes to determine if the lack of

significant differences was due to an insufficient number of wetlands compared (Table 3-

9). Water column TP differences were identified, therefore Least Significant Number

(LSN) values were fairly low for water column comparisons. This means that if

additional sample sites were included in the survey, and the data retained their current

structure, then additional significant differences between the water columns of marshes

and swamps would likely have been detected. In the case of litter and soil parameters (on

a mass per unit mass basis), LSN values were very large. Therefore, true differences in

soil (on a mass per unit mass basis) and litter parameters between swamp and marsh sites









are very small and unrealistic to quantify. Litter and soils from marshes and swamps in

this survey of minimally impaired wetlands of the southeastern US were similar with

respect to P (mg/kg), N (g/kg), and C (g/kg) content.

Regional differences may also be affecting marsh and soil results since wetlands

from all three USEPA Nutrient Ecoregions were combined for comparisons. To explore

this possibility, the 52 swamps and 9 marshes in the Southeastern Forested Plain were

compared. The results were fairly consistent with those including all three ecoregions.

There were no significant TC, TP, or TN differences between the vegetative communities

within litter, soil (on a mass per unit mass basis), and water column strata. The only

discrepancy was that marshes and swamps had significantly different water column TP

content when wetlands from all three ecoregions were compared. There may not be

enough samples to detect differences at the community type and ecoregion level.

However, fairly consistent results indicate that regional differences are not skewing

results.

Table 3-9.Power analysis for non-significant parameters within community comparisons
Number compared Least Significant
Measred aramters in current study (n) Number (LSN)
Water TP 66 123
Water TN 67 143
Litter % P 82 5,439
Litter % N 97 21,645
Litter % C 97 1,936
Litter C/N 96 6,302
Litter C/P 81 22,449
Soil %P 97 409
Soil %/N 94 5,276
Soil %C 92 480,226

A major difference between swamps and marshes is the presence of a canopy in

swamps. Canopy cover can limit light penetration and reduce algal populations. Battle









and Golladay (2001) found that sedge marshes had higher algal growth than cypress

swamps. Algal populations quickly sequester nutrients from the water column, hence

decreasing soluble nutrients available to other growth forms (Kadlec and Knight 1996).

Therefore, it was hypothesized that the water column of marshes would contain less

nitrogen and phosphorus than that of swamps. The results of this study partially support

this hypothesis. Water column TP was significantly greater in swamps, but TN was

similar regardless of dominant vegetation type. Large variation between wetlands and

the limited number of samples are likely responsible for the lack of statistical differences

detected between water column TN of swamps and marshes. Power analysis indicated

that fewer than one hundred additional samples may be sufficient to detect significantly

greater water column TN content in swamps compared to marshes.

If swamps have greater water column nutrient content, they may not be removing

nutrients from the water column as effectively as marshes. Wetlands are commonly

valued as nutrient sinks; however, it may be necessary to distinguish by vegetative type

when assigning this value to wetlands. Differences in nutrient cycling may have

implications for aquatic ecosystems downstream, in that marshes may retain more

phosphorus and nitrogen than swamps. Distinctions between swamps and marshes may

be necessary for determining water column based numeric nutrient criteria for wetlands.

Water column total nitrogen and total phosphorus and soil total phosphorus

(mg/cm3) COncentrations appear to be the most sensitive parameters to differences

between marshes and swamps. Water column nutrients can be overly sensitive

indicators. For example, if water is sampled following a rain event, nutrients may be

diluted. When wetlands are sampled on different days (or even different seasons),










comparisons between them may be confused by parameters (such as rain events) that are

not factored into comparisons. Soil based indicators integrate conditions over a longer

period of time and are not easily influenced by sampling conditions. Soil and/or water-

based numeric nutrient criteria in the southeastern United States may necessitate a

distinction between vegetative wetland types.

Hydrologic Comparisons: Riverine and Non-riverine

The surveyed wetlands can be aggregated by hydrologic connectivity into riverine

and non-riverine systems. Riverine wetlands are adjacent to streams, and non-riverine

systems are located at least 40-meters from adj acent water bodies. Riverine wetlands

were more common; therefore, 64% of the surveyed wetlands were riverine and 36%

were non-riverine. Riverine systems include riverine marshes and riverine swamps,

while non-riverine systems include non-riverine swamps and non-riverine marshes.

Differences between riverine and non-riverine wetlands will be addressed in the

following four sections: water column, litter, soil, and discussion.

Water Column

Comparisons based on hydrologic connectivity showed that riverine systems (Table

3-10) had significantly greater water column pH and lower oxidation reduction potentials

(Eh) than non-riverine systems. Since Eh and pH are the dominant chemical factors

influencing nutrient transformations within wetlands (Reddy and D'Angelo 1994), one

would expect to see different nutrient signatures dependent on hydrologic connectivity.

However, these differences may be minimal since surveyed riverine and non-riverine

wetlands were still considered acidic (average pH<7.0) and had mean Eh values (>

300mV) indicating aerobic conditions (Reddy and D'Angelo 1994) in the water column.










No significant differences were found between riverine and non-riverine water

column temperature, dissolved oxygen percent saturation, conductivity, or water depth.

The presence of algae was significantly greater in non-riverine systems than riverine

systems. Algae were noted in 13% of surveyed non-riverine wetlands and only 5% of

riverine systems. It is likely that this difference is the result of the higher occurrence of

non-riverine marsh communities than riverine marsh communities and the higher

frequency of algae in marshes (47%) than that of swamps (10%).

Table 3-10.Water column properties observed in minimally impaired wetlands
aggregated by hydrologic connectivity
Paverine Non-riverine
Parameters mean + SD median 75th n sig mean + SD median 75th n
TP (mg/L) 0. 119 + 0. 13 0.069 0.193 35 0.075 + 0.087 0.039 0.086 31
TN (mg/L) 2.20 + 1.60 1.81 2.79 36 2.18 +1.11 1.88 2.51 31
Temp (oC) 22.6 + 3.2 23.1 25 24 23.4 + 4.39 22.2 27 26
pH 5.5 + 1.0 5.8 6.3 24 ** 4.6 + 1.1 4.3 5.4 26
DO (%) 31.7 + 18.8 32.1 44.9 24 30.3 + 25.6 18.6 52.3 26
Cond. (uS/cm) 64 + 49 59 82 23 67 + 45 58 83.2 25
Eh (mV) 349 + 356 338 430 18 447 + 342 247 534 19
Depth (cm) 21.1 + 19.8 14.5 38.1 27 24.9 + 24.6 15.2 42.4 30
Significant difference (p<0.05)
** Significant difference (p<0.01)
*** Significant difference (p<0.001)

Further comparisons between hydrologic classes (Figure 3-6) indicate that riverine

systems had significantly greater water column TP but lacked significantly different TN

values when compared to non-riverine systems. Water column TP data support the

hypothesis that riverine systems have at least some higher water column nutrient

conditions. Riverine wetlands are hydrologically connected to adjacent aquatic

ecosystems and often integrate a more extensive upstream watershed, which may be a

source of nutrients. In contrast, non-riverine wetlands often have a smaller and more

localized watershed resulting in lower nutrient loading (Craft and Casey 2000).







53



0.7 0
0.6-8


E 0.4-1 .E~
0- 0.3_. 24-


0.1- i 1-

Non-riverine Riverine Non-riverine Riverine
A B A A

Figure 3-6.Water column TP and TN values by hydrologic connectivity. The dashed line
is the mean of each population, and the solid line is the overall mean. The
bottom of the "box" is the 25th percentile, and the top is the 75th percentile.
The center line within the box lot is the median. The whiskers extend + 1.5
the interquantile range. Different letters indicate a significant difference
(p<0.05) between treatments.

Litter

Litter of riverine wetlands had significantly higher phosphorus (p<0.0001) and

lower carbon content (p<0.0001) than non-riverine wetlands (Figure 3-7). Nitrogen

content of litter was similar between the two systems. Ratios of carbon to nitrogen and

carbon to phosphorus followed the carbon content trends, with non-riverine systems

having significantly higher ratios than riverine systems.

It is probable that lower litter carbon content within riverine systems is due to

hydrologic fluxes of these open systems and transport of particulate matter. Watersheds

of riverine systems often contribute inorganic materials that are deposited in wetlands

during flood events. Litter is often coated in organic and inorganic materials that were

not removed before analyses. It may be that litter in riverine and non-riverine systems

has similar organic carbon content, but that increased inorganic deposition in riverine

systems alters the percentage of total carbon relative to non-riverine systems.









Soil

Soil characteristics of riverine and non-riverine systems were also compared

(Figure 3-8). Non-riverine systems had significantly greater nitrogen (g/kg and mg/cm3)

carbon, N:P, C:N, C:P, LOI, and percent moisture content than riverine systems.

Riverine wetlands had significantly greater phosphorus (mg/cm3) and bulk density

values. Craft and Casey (2000) found that non-riverine forested wetlands had elevated

soil TP, organic C, and TN content compared to forested riverine wetlands. Soil total

phosphorus results from the current study do not coincide with Craft and Casey's results.

Discussion

Hydrologic connectivity of riverine wetlands led to the hypothesis that the water

column of riverine wetlands would have higher nutrient concentrations compared to non-

riverine systems. Results of this survey partially support this hypothesis. Water column

phosphorus was greater in riverine wetlands, but there was no difference in nitrogen

content regardless of hydrologic connectivity. Increased phosphorus conditions in

riverine wetlands are likely due to larger contributing watersheds relative to non-riverine

sy stem s.

Power analysis was applied to water column total nitrogen data to explore the role of

sample size in statistical conclusions. A significant difference is more likely to be

detected if approximately 800 additional wetlands were included in the survey. The large

LSN value indicates that there is not much difference between water column TN of

riverine and non-riverine wetlands. It is possible that wetlands cycle nitrogen similarly

regardless of hydrologic connectivity or that water column nitrogen is controlled by

factors not captured in this comparison.


















S0.3 1.5 40
|-- 0.3 I
0.02-

0.0125


Non-riverine Riverine Non-riverine Riverine Non-riverine Riverine
A B A A A B

Figure 3-7.Litter phosphorus, nitrogen, and carbon content comparisons between riverine and non-riverine systems. The dashed line is
the mean of each population, and the solid line is the overall mean. The bottom of the "box" is the 25th percentile, and the
top is the 75th percentile. The center line within the boxplot is the median. The whiskers extend + 1.5 the interquantile
range. Different letters indicate a significant difference (p<0.05) between treatments.


Table 3-11.Leaf litter properties observed in minimally impaired wetlands aggregated by hydrologic connectivity.
Riverine Non-riverine
Parameters mean + SD median 75th n significance mean + SD median 75th n
%P 0.02 + 0.012 0.046 0.027 55 *** 0.008 + 0.004 0.007 0.01 29
%/N 1.24 + 0.28 1.22 1.45 62 1.35 0.38 1.31 1.68 34
%C 38.21 + 7.64 39.55 44.55 64 *** 47.96 + 3.37 49.3 50.59 31
C/P ratio 2982 + 2418 1868 4450 54 *** 6984 + 3509 7134 8700 27
C/N ratio 29.9 + 8.6 28.2 34.4 62 ** 36.4 + 10.37 36.9 42 34
* Significant difference (p<0.05)
** Significant difference (p<0.01)
*** Signifieant difference (p<0.001)

















800 20y
m 300
a600- 15 -

Ii 10-



Non-riverine Rive rin e Non-riverine Riverine Non-riverine Riverine

A A A B A B
Figure 3-8.Soil TP and TN values by hydrologic connectivity. The dashed line is the mean of each population, and the solid line is the
overall mean. The bottom of the "box" is the 25th percentile, and the top is the 75th percentile. The center line within the
box lot is the median. The whiskers extend + 1.5 the inter uantile ran e. Different letters indicate a si nificant difference
(p<0.05) between treatments.
















mean + SD
410+ 290
0.11 + 0.09
10.0 + 8.3
2.31 + 0.78
216.3 + 178.8
28.8 + 19.3
615 + 381
21.3 + 5.8
45.6 + 33.7
0.41 + 0.30
0.64 + 0.22


median
370
0.08
5.4
2.35
115.3
23.2
541
21.0
27.8
0.39
0.61


75th
600
0.13
16.5
2.83
422.9
32.7
887
24.0
82.6
0.63
0.86


Parameters
TP (mg/kg)
TP (m/cm3
TN (g/kg)
TN (m/cm3
TC (/g
N/P ratio
C/P ratio
C/N ratio
LOI (%)


***
***
***
***
**
***


Bulk Densit (/cm3) 0.63 + 0.35
Moisture Content ( 0.50 + 0.20
* Significant difference (p<0.05)
** Significant difference (p<0.01)
*** Signifieant difference (p<0.001)


n si nificance


Table 3-12.Soil properties observed in minimally impaired wetlands aggregated by hydrologic connectivity


Riverine


Non-riverine


mean + SD
390 + 240
0.22 + 0.16
4.3 + 4.2
1.86 + 0.82
75.6 + 85.2
142 + 18
254 + 261
17.9 + 4.8
19.3 + 17.0


median
330
0.16
2.7
1.69
48.7
10.3
174.6
17.6
13.4
0.65
0.46


75th
500
0.29
4.9
2.16
84.4
17.4
325.6
19.3
22.5
0.87
0.64









Riverine systems are open to hydrologic influxes and are reported as sinks for

sediment and phosphorus from contributing watersheds (Craft and Casey 2000).

Therefore, it was hypothesized that riverine systems would have higher soil phosphorus

and nitrogen content than soils in non-riverine wetlands. This hypothesis was partially

supported by the data collected. Riverine wetlands had greater total phosphorus content

when the values were normalized by bulk density. However, non-riverine systems had

greater soil nitrogen content than riverine wetlands.

Alternating anaerobic and aerobic conditions are ideal for processing nitrogen

through wetlands, since nitrogen loss from wetland soil is limited by nitrification in

aerobic zones and ammonium diffusion from anaerobic zones to aerobic zones (Reddy

and D'Angelo 1994). It appears that surveyed riverine wetlands are storing less nitrogen

in the soil than non-riverine wetlands. Riverine wetlands are subj ect to pulses of flooding

when adj acent streams overflow their banks. Sudden flooding followed by recession of

floodwaters may create ideal conditions for nitrogen processing, hence lowering nitrogen

storage in riverine wetlands.

The final hypothesis based on hydrologic differences was that riverine wetlands

would have increased nutrient levels in leaf litter compared to non-riverine wetlands. It

was thought that riverine systems have high nutrient influxes that allow plants to cycle

nutrients less efficiently and reabsorb fewer nutrients from senescing leaves (Hopkinson

1992). Hence, areas with increased nutrient availability produce leaf litter with high

nutrient content (Shaver and Melillo 1984). This hypothesis was partially supported by

this study. Riverine wetlands did have higher litter TP content, but TN was similar

regardless of hydrologic connectivity.









Power analysis was applied to the litter TN results to determine if the lack of

significant difference was due to an insufficient number of wetlands compared. A

difference between riverine and non-riverine litter TN would likely be detected with

approximately 50 additional samples. This indicates that sample size is affecting the

results. Interestingly, additional samples would not support the hypothesis, but would

show litter in non-riverine systems to have greater nitrogen content than litter in riverine

wetlands.

Regional differences may influence hydrologic connectivity results since wetlands

from all three USEPA Nutrient Ecoregions were combined for comparisons. To explore

this possibility, the 44 riverine and 17 non-riverine wetlands in the Southeastern Forested

Plain were compared. The results were fairly consistent with those including all three

ecoregions. The only discrepancy was that riverine and non-riverine systems had

significantly different water column TP content when wetlands from all three ecoregions

were compared. This difference was not apparent within the Southeastern Forested Plain

comparisons. Additional samples may be needed to detect differences at the ecoregion

level. Comparable results indicate that regional differences are not skewing the noted

differences based on hydrologic connectivity.

As was found when the data were aggregated by vegetative type, there is a lot of

variability within the data aggregated by hydrologic connectivity. Standard deviation

values were often as great as the mean values. This variability makes it difficult to

recommend a single numeric nutrient criterion for protecting all wetlands. Aggregating

by hydrology may reduce some of the variability, since differences were identified

between riverine and non-riverine wetlands for numerous parameters (water column TP,










pH, Eh; litter TP, TC, C\P, C\N; soil TP TN, TC, N\P, C\P, C\N, LOI, bulk density, and

moisture content). Soil and litter strata appear to be the most sensitive to differences

between riverine and non-riverine wetlands. It may be necessary to identify wetlands as

riverine or non-riverine for recommending numeric nutrient criteria.

Spatial Variation

The surveyed wetlands were stratified within three USEPA Nutrient Ecoregions

(Figure 3-9). The vegetative type and hydrologic connectivity of wetlands surveyed were

not evenly distributed among the ecoregions (Table 3-13). The types of wetlands

sampled essentially reflected the distribution of wetland types within the National Forest

being surveyed. A greater percentage of marshes and non-riverine systems were

represented within the Southern Coastal Plain (XII). Only 12% of the surveyed wetlands

were located in the Eastern Coastal Plains (XIV).

Data were aggregated by ecoregion to explore the appropriateness of this regional

classification as an a priori grouping for establishing numeric nutrient criteria.

Comparisons among ecoregions were made for all wetlands aggregated together and

between specific vegetative and hydrologic groupings. Water column, litter, and soil

characteristics were compared. One would expect differences among ecoregions if they

represent distinct geographic regions with respect to nutrients. Differences between

ecoregions will be addressed in the following four sections: water column, litter, soil, and

discussion.












































Figure 3-9.Distribution of wetlands within the three USEPA Nutrient Ecoregions aggregated by a) hydrologic connectivity and b)
vegetative type.









Table 3-13. Number of surveyed wetlands within the three USEPA nutrient ecoregions
Southeastern Southern Eastern
GroupingAggregation Forested Plains Coastal Plain Coastal Plain
Criteria
(IX) (XII) (XIV)
None Combined 61 29 12

Hydrologic Riverine 44 14 8
Connectivity Non-riverine 17 15 4

Vegetative Marsh 9 8 1
Type Swamp 52 21 11

Water Column

When all wetlands were grouped together, water column total phosphorus and total

nitrogen did not differ among the three ecoregions (Table 3-14). However, when

aggregated by hydrologic connectivity (Figure 3-10), riverine wetlands in the Southern

Coastal Plain (XII) had greater water column TN than those of the Southeastern Forested

Plains (IX). There were no detectable water column TN and TP differences among non-

riverine wetlands in the three ecoregions.

When aggregated by vegetative type, swamps in the Southern Coastal Plain had

greater water column TN content than comparable sites of the Southeastern Forested

Plain (Figure 3-11). There were no detectable water column TN and TP differences

among marshes in the three ecoregions. However, fewer marshes were sampled, and

there were no water column data for the one marsh in the Eastern Coastal Plain. Water

was not present in this wetland when it was sampled.

The Southern Coastal Plain appears to have greater water column TN (statistically

significant in swamps and riverine wetlands). Water column total nitrogen differences

were not detected in vegetative type and hydrologic connectivity comparisons in the















Table 3-14. Water column descriptive statistics for surveyed wetlands by ecoregion. Superscript letters following standard deviations
indicate significance for comparisons made across rows. Different letters indicate a significant difference (p<0.05).
Southeastern Forested Plains (IX) Southern Coastal Plain (XII) Eastern Coastal Plain (XIV)
Parameters mean + SD median 75t n mean + SD median 75th n mean + SD median 75t n
r ~~a a~~n a


0.042 0.096 31 0.09 + 0.09
1.510 2.150 32 2.40 + 1.15a

13.60 26.80 21 20.8 + 3.3b
5.50 6.3 21 4.5 + 1.2a

33.2 53.0 21 31.1 +21.3a
49.0 82.4 22 70.4 + 58.9a
346 489 21 481 + 347a

9.2 17.4 22 13.2 + 10.5a


0.045
2.330

20.2
4.0

27.7
70.5
507

9.4


0.074
2.78
0.029


0.140 27 0.120 + 0.087l
2.860 28 2.80 + 1.82a

23.0 14 24.6 + 1.5ab
5.0 14 5.5 + 1.2a

52.4 14 23.8 + 17.3a
126.2 13 103.3 + 59.3a
575 9 365 + 355a

21.2 14 6.2 + 8.1a


0.197 12 0.106 + 0.060a
3.40 12 1.93 + 0.79ab
0.086 15 0.134 + 0.116a


0.088~ 0.217
2.070 3.940

24.2 25.6
5.9 6.3

15.7 42.3
76.0 130.5
327 495

3.5 7.3


0.088 0.169
2.07 2.61
0.132 0.239

3.11 6.02


TP (mg/L)
TN (mg/L)

Temp (oC)
pH
DO (%)
Cond. (uS/cm)

Eh (mV)

Depth (cm)


0.099 + 0.134
2.02 + 1.69a

24.1 + 4.0a
5.3 + 0.9a

35.9 + 24.2a
61.3 + 52.0a
379 + 353a

11.1 + 7.6a


0.118 + 0.154a
1.85 + 1.74a
0.070 + 0.094a


RiverineTP
o3 RiverineTN

Ef Non-riverineTP
-E Non-riverine
TN


SMarshTP

SMarsh TN

ESwamp TP

Swamp TN


0.042
1.32

0.041


0.194 19
1.85 20

0.082 12


0.126 + 0.111a

2.88 + 1.42b
0.063 + 0.073a


2.29 + 1.64a 1.84 2.27 12 2.05 + 0.75a 1.86 2.56 16 3.66 +2.27a


0.117 + 0.128a

2.50 + 1.99a

0.094 + 0.138a

1.86 + 1.59a


0.082

1.79

0.040

1.32


0.227 7

2.71 8

0.093 24

2.15 24


0.033 + 0.022a

1.82 + 0.70a

0.131 + 0.106b

2.78 + 1.23a


0.021

1.73

0.095

2.77


0.045 11

2.51 11

0.206 16 0.120 + 0.087ab 0.088 0.217

3.35 17 2.80 + 1.82a 2.07 3.94







64


Non-riverine R ive rin e

7-" 8-

6_ 7-

E E 5-
Z Z 4-






-- I

A A B A

Figure ~ ~ ~ ~ ~ ~v 3-0Cmaio feoeiosageae yhdoog.Tedse iei h
men feahpouato adtholi ieih overlen The botom
of th "bx iste2t ecnie n h o ste7t ecnie h







Fgueographic regisoninta of dcomeion antvegeation or hydrologicTh connetivity. sth


not sgnifiantl different fromlthe othe twe oi ecoregios foran ofeal thnTe aggegtion

schemes. This maybe due toe insuffcient smle sz, sn hepinc there w5h ereontly 1 sureye

weladswthin thise ecoreion.tebxlti h ein h hsesetn .




Lecitt oer nutrent content was compaed cntamong thpetree ecoregos aThed Sothr

Ceogastal Plain (II) had significantly loegtio r tota caronadgrete tonetal posporu



contesint icmartydfeedt to the other w ecoregions (Tabl 3-15) Lite tota nitrogencontnwa



similard amohng the tre ecoregions.











e Marsh Swamp
7-1 8

6_ 7-
5- 6
E E 5-
Z Z 4-
2- 3-1- 3 -


0' 0

So o oo
o a 0 0 0
LI.. .I LI..


A A A B AB


Figure 3-11.Comparison of ecoregions aggregated by vegetative type. The dashed line
is the mean of each population, and the solid line is the overall mean. The
bottom of the "box" is the 25th percentile, and the top is the 75th
percentile. The center line within the boxplot is the median. The
whiskers extend + 1.5 the interquantile range. Different letters indicate a
significant difference (p<0.05) between treatments.




When aggregated by hydrologic connectivity and ecoregion, riverine wetlands in

the Southemn Coastal Plain had significantly lower TC, greater TP, and similar TN

content than the other ecoregions (Figure 3-12). There were no observed TP, TN, or TC

differences among litter of non-riverine wetlands in the three ecoregions. It appears that

riverine wetlands are driving the differences found when all wetlands are combined.

Comparisons were made among the ecoregions when aggregating wetlands by

vegetative community. Marshes in the Southern Coastal Plain had significantly lower

litter TP than marshes in the Southeastemn Forested Plain (Figure 3-13). Marshes within

all three ecoregions had similar litter TN and TC content. Swamps within the Southemn

Coastal Plain had significantly lower litter TP (Figure 3-13) and greater TC content















Table 3-15.Litter descriptive statistics for surveyed by Ecoregion. Superscript letters following standard deviations indicate
significance for comparisons made across rows. Different letters indicate a significant difference (p<0.05).


Southeastern Forested Plains (IX)
mean + SD median 75 h


Southern Coastal Plain (XII)


Eastern Coastal Plain (XIV)


median 75


median 75" n


n mean + SD

44 0.008 + 0.004b
56 1.24 + 0.33a
57 47.74 + 2.75b
42 177.0 + 86.4b
42 7056 + 3482b

56 40.0 + 10.5b


33 0.009 + 0.003b
39 1.23 + 0.29a
11 0.007 + 0.004a
17 1.25 + 0.38a


8 0.005 + 0.002b
8 1.38 + 0.55a

36 0.008 + 0.003b
48 1.20 + 0.24a


n mean + SD

25 0.024 + 0.012a
26 1.31 + 0.14a
25 39.97 + 7.92a
26 78.5 + 53.6a
26 2532 + 22308

26 29.7 + 6.3a


12 0.029 + 0.009a
13 1.36 + 0.13a
13 0.011 + 0.005a
13 1.22 + 0.13a


5 0.033 + NA
6 1.57 + NA

20 0.022 + 0.012a
20 1.28 + 0.12a


Parameters

TP (%)
TNr (%)
TC (%/)
N/P ratio
C/P ratio

C/N ratio


0.018 + 0.012a 0.014
1.31 + 0.39a 1.25
38.70 + 8.02a 39.30
136.3 + 192.2a 87.00
3681 + 4000a 2225

29.9 + 9.3a 27.70


0.021 +0.012a 0.02
1.25 + 0.37a 1.19
0.008 + 0.004a 0.01
1.46 + 0.42a 1.61


0.024 + 0.09a 0.02
1.45 + 0.41a 1.46

0.016 + 0.010a 0.01
1.29 + 0.39a 1.22


0.024
1.59
45.08
149.80
5465

34.10


0.007
1.26
47.97
159.70
6618

38.40


0.008
1.15
0.006
1.26


0.004
1.30

0.007
1.20


0.009
1.39
49.86
202.70
9374

46.10


0.010
1.42
0.008
1.37


0.008
1.89

0.010
1.38


0.022
1.33
40.97
61.90
1470

27.40


0.037
1.36
0.011
1.22


0.033
1.57

0.021
1.33


0.033 12
1.40 11
48.12 12
85.90 12
2676 12

36.80 12


0.037 8
1.43 7
0.017 4
1.34 4


0.033 1
1.57 1

0.033 11
1.37 10


RiverineTP (%)
RiverineTN (%)
Non-riverineTP (%)
Non-riverine TN (%)


MarshTP (%)
Marsh TN (%/)

Swamp TP (%)
Swamp TN (%)


0.03
1.45
0.01
1.81


0.04
1.81

0.02
1.53







67


Riverine Riverine
3 0.06

2.5- 0.05-

b.04-
2-a
~0.03-
1.5-j
I ~0.02-



0.5' 0

0, 0
0 O O o a



A A A A B A


Figure 3-12.Comparison of riverine wetlands in the three ecoregions. The dashed line is
the mean of each population, and the solid line is the overall mean. The
bottom of the "box" is the 25th percentile, and the top is the 75th percentile.
The center line within the boxplot is the median. The whiskers extend +
1.5 the interquantile range. Different letters indicate a significant difference
(p<0.05) between treatments.

compared to the other two ecoregions. Litter TN content was similar in swamps within

all three ecoregions.

The Southern Coastal Plain had lower litter phosphorus content when all wetlands

were combined and also for the following aggregations: riverine systems, swamps, and

marshes. Litter nitrogen content was also lower in the Southern Coastal Plain, but the

difference was not significant for any of the aggregations. The Eastern Coastal Plain had

the greatest litter phosphorus content (significant for all wetlands combined, riverine

systems, and swamps).

Soil

Soil characteristics were compared among the three ecoregions (Table 3-16).

When compared without sub-classification, soil TN (mg/cm3) and TP (mg/cm3 wr







68




Marsh Swamp
0.06 0.045
0.04-
0.05-
0.035-
0.04- 0.03-
a -a0.025-
S0.03-
0.02-
0.021 0.015
0.01 0.01 -i

O 0


0 0 o o
o o o o o o



A B AB A B A

Figure 3-13.Comparison of litter total phosphorus among the three ecoregions aggregated
by vegetative type. The dashed line is the mean of each population, and the
solid line is the overall mean. The bottom of the "box" is the 25th percentile,
and the top is the 75th percentile. The center line within the boxplot is the
median. The whiskers extend + 1.5 the interquantile range. Different
letters indicate a significant difference (p<0.05) between treatments.




significantly greater in the Southemn Coastal Plain than in the Southeastemn Forested Plain

or Eastern Coastal Plain. However, when comparisons were conducted on a mass per

unit mass basis there were no significant differences in soil total phosphorus (mg/kg)

content among the three ecoregions.

When aggregated by hydrology there were no significant soil TP (mg/kg), TP


(mg/cm3), TN (g/kg), TN (mg/cm3) Or TC (g/kg) differences among non-riverine

wetlands in the three ecoregions. However, differences among the ecoregions were


apparent with riverine wetlands. The Southemn Coastal Plain had riverine wetlands with

lower TP (mg/cm3) than the other two ecoregions. The Southeastern Forested Plain had

lower TN (mg/cm3) in Tiverine Systems than the other two ecoregions.















Table 3-16.Soil descriptive statistics for surveyed wetlands aggregated by Ecoregion. Superscript letters following standard deviations
indicate significance for comparisons made across rows. Different letters indicate a significant difference (p<0.05).
Southeastern Forested Plains (IX) Southemn Coastal Plain (XII) Eastemn Coastal Plain (XIV)
Parameters mean + SD median 75th n mean + SD median 75th n mean + SD median 75th n


TP (mg/cm3)
TN (mg/cm3)
TC (g/kg)
N/P ratio
C/P ratio
C/N ratio

LOI (%)
Bulk Density
% Moisture


0.22 & 0.17a
1.75 & 0.64a
90.4 + 113.1a
13.3 + 11.4a
266.4 + 273.7a
18.2 + 3.6a
23.2 + 22.8a
0.60 + 0.29a
0.50 + 0.21a


0.14 0.27 57 0.096 & 0.067b
1.62 2.11 55 2.43 A 1.05b
48.5 93.9 54 201.2 + 172.7b
9.9 17.9 55 35.0 + 28.8b
167.8 354.1 54 669.0 + 397.4b
18.3 21.2 54 21.3 + 7.5a
15.3 22.7 57 42.1 + 33.6a
0.67 0.84 59 0.44 + 0.44b
0.45 0.66 59 0.65 + 0.20b


0.075 0.13 27 0.24 & 0.12a
2.56 3.34 26 2.43 & 0.67a
104.7 370 26 125.0 + 141.7ab
25.3 34.6 25 12.3 + 5.4a
513.4 936.1 25 246.7 + 206.3a
19.2 24.3 26 17.7 + 6.0a
27.7 86.1 27 31.3 + 25.8a
0.43 0.55 28 0.52 + 0.31ab
0.58 0.85 28 0.57 + 0.21ab


0.083 0.13 12 0.30 + 0.10a

2.26 3.47 13 2.63 & 0.76b

0.061 0.13 15 0.13 & 0.038a

2.71 3.19 13 2.07 & 0.28a


0.2
2.34
53.6
12.7
175.8
15.5
19.3
0.51
0.55


0.301 11
2.44 11
165.9 10
14.5 11
14.5 11
20.5 10
55.8 11
0.87 12
0.80 12


Riverine TP
(mg/cm3)
- RiverineTN
-O (mg/cm3)
Non-riverineTP
x"(mg/cm3)
Non-riverine TN
(mg/cm3)

MarshTP
(mg/cm3)
Marsh TN
(mg/cnr)
~pSwamp TP
S(mg/cm3)
Swamp TN
(mg/cm3)


0.25 & 0.17a 0.20 0.37 41 0.095 & 0.055b

1.57 & 0.46a 1.49 1.77 41 2.36 & 1.25b

0.12 & 0.11a 0.084 0.13 16 0.097 & 0.077a


0.30 0.39 7

2.39 3.34 7

0.13 0.15 4

2.12 2.32 4


2.27 & 0.82a 2.32 2.65 14


2.50 + 0.85a


0.19 & 0.13a 0.16 0.28 8 0.061 & 0.034b

1.65 A 1.13a 1.30 1.95 7 2.43 A 1.12a

0.22 & 0.17a 0.14 0.27 49 0.11 & 0.072b


0.05 0.096 7

2.92 3.31 7


0.099 0.14 20 0.24 & 0.12a


0.20 0.30 11

2.34 2.44 11


2.43 A 1.06b 2.36 3.43 19 2.43 & 0.67b


1.77 & 0.55a 1.69 2.14 48









Soil data were also aggregated by vegetation type and compared among ecoregions.

Marsh soil phosphorus and nitrogen concentrations did not show significant differences

among the ecoregions when compared on a mass per unit mass basis. When nutrient

concentrations were normalized by bulk density, marshes in the Southemn Coastal Plain

had greater total TN (mg/cm3) and lower total phosphorus (mg/cm3) COmpared to the

Southeastern Forested Plain. There were no soil data for the one marsh within the

Eastern Coastal Plain.

Swamps within the Southemn Coastal Plain had significantly greater soil TN

(mg/cm3) and TN (g/kg) content compared to swamps of the Southeastern Forested Plain.

There were no significant differences among ecoregions for TP (mg/kg) content in

swamps. However, the Southemn Coastal Plain had significantly lower soil TP (mg/cm3)

than the other two ecoregions.

Discussion

It was hypothesized that there would be regional differences in the nutrient regimes

of wetlands. Differences among the three ecoregions support this hypothesis. The

Southern Coastal Plain (XII) is different from the other two ecoregions (Table 3-17), with

greater water column TN, litter TC, soil TN, soil TC, and lower litter TP content. These

differences suggest that it is a distinct region with its own nutrient characteristics.

However, standard deviations were still high even when wetlands were aggregated by

ecoregion and hydrologic connectivity or vegetative type. There is still considerable

variability among the aggregated wetlands, indicating that the ecoregions may be too

large to aggregate regional differences among wetland nutrient conditions appropriately.










Table 3-17.Summary of significant differences (p<0.05) among the three USEPA
Nutrient Ecoregions for the various aggregations of surveyed wetlands.
All Wetlands Non-
Riverine Marsh Swamp
Combmned nivenine
Water
column TP
Water
column TN = XII > IX = = XII > IX
Litter TN
Litter TC XII >IX and XIV XII >IX and XIV = = XII >IX and XIV
Litter TP XII Soil TN
(g/kg) XII > IX XIV > IX = = XII > IX
Soil TC
(g/kg) XII > IX = = XII > IX XII > IX
Soil TP
(mg/kg) = XIV > XII and IX

There were no soil TP (mg/cm3) differences between the Southeastemn Forested

Plain and the Eastern Coastal Plain. However, these ecoregions had significantly greater

soil TP (mg/cm3) COntent than the Southemn Coastal Plain. The Southeastemn Forested

Plain was the largest ecoregion surveyed and included a few northern Florida sites, all

Georgia and Alabama sites, and half of the surveyed wetlands in South Carolina. Several

of these areas are known for their clay mineral soils. Mineral soils retain phosphorus

better than organic soils, due to higher iron and aluminum content (Richardson 1985).

Therefore, it is not surprising that the Southeastern Forested Plain had greater soil TP

content than the Southern Coastal Plain.

The Eastemn Coastal Plain included wetlands within coastal South Carolina. High

sedimentation rates in alluvial floodplains are common in this ecoregion. Upland inputs

to streams may have resulted in higher phosphorus and nitrogen accumulations in the

wetland soils of this ecoregion compared to wetlands in the Southemn Coastal Plain.









When non-riverine water column, litter, and soils were compared among the three

ecoregions, there were no significant differences. Non-riverine wetlands may be less

affected by regional differences because they have smaller contributing watersheds than

riverine wetlands. Watershed properties, such as soil types and topography may be

driving some of the differences noted between riverine and non-riverine wetlands in the

section above.

Variability within ecoregions was explored by examining regional differences in

wetland nutrient regimes at a scale finer than the USEPA Nutrient Ecoregions. The

Southeastern Forested Plain was subdivided into smaller regions by aggregating the

surveyed wetlands by National Forest (or military base). This ecoregion was chosen

because it has the largest area and contained most of the surveyed wetlands (60%). Fort

Benning Military Base, Moody Air Force Base, Banks Lake National Wildlife Refuge,

Conecuh, Oconee, Sumter, Talladega, along with portions of Apalachicola National

Forest are located in the Southeastern Forested Plain (Figure 3-14). Moody Air Force

Base and Banks Lake National Wildlife Refuge are adj acent to each other; therefore the

two surveyed wetlands from Banks Lake National Wildlife Refuge were combined with

the three surveyed wetlands from Moody Airforce Base. These Hyve wetlands are referred

to as Moody Air Force Base.

Soil total phosphorus (mg/cm3) and nitrogen (mg/cm3) COntent were compared

among these National Forests and military bases within the Southeastern Forested Plain.

There were no significant differences regarding soil total nitrogen content. There were,

however, significant differences among some of the regions with regards to soil total

phosphorus content (Table 3-18). Apalachicola National Forest had the lowest TP



















































Figure 3-14. Distribution of sampling locations within the USEPA Nutrient Ecoregions.


Legend ~ I L
Apalachicola NF
SBanks Lake NWF
SConecuh NF
Fort Benning ~isliliary
Francis islarion NF
luloody Air Force Base
SOcala NF
SOconee NF
Osceoa NFSoutheastern Forested Plain
Sme Southern Coastal Plain
I I Tlladea NFEastern Coastal Plain


480 Kilometers









content while Oconee National Forest had the greatest TP content. There is almost an

order of magnitude difference between the means of these regions.

The USEPA has discussed setting numeric nutrient criteria at the 75th percentile

value of least impaired wetlands within an ecoregion. Results from this study suggest

that the 75th percentile of soil TP in the Southeastern Forested Plain is 0.27 mg/cm3. It is

unlikely that the USEPA would adopt this recommendation without additional research.

However, if this value was adopted as the numeric nutrient criteria for this ecoregion,

wetlands in the Apalachicola area would not be sufficiently protected from nutrient

enrichment. The mean soil TP content of these wetlands would have to increase by a

factor of Hyve before exceeding the numeric nutrient criteria. Likewise, wetlands in

Sumter and Oconee National Forests already exceed the hypothetical nutrient threshold.

It is clear that there are significant regional differences in wetland nutrient regimes at a

scale Einer than the USEPA Nutrient Ecoregions.


Table 3-18.Soil total phosphorus statistics for surveyed wetlands in the Southeastern
Forested Plain aggregated by National Forest (or military base). Vertical lines
connect means that are not significantly different (p<0.05).
Soil TP (mg/cm3) mean & SD median 75th n
Apalachicola, FL 0.056 & 0.021 0.048 0.077 5
Moody AB, GA 0.11 & 0.042 0.09 0.15 5
Conecuh, GA 0.13 & 0.08 0.089 0.14 7
Oakmulgie, AL 0.15 & 0.10 0.14 0.14 11
Fort Benning, GA 0.15 & 0.05 0.14 0.19 8
Sumter, SC 0.38 & 0.20 0.27 0.51 11
Oconee, GA 0.39 & 0.12 0.44 0.5 9















CHAPTER 4
CONCLUSIONS

Establishing nutrient criteria for wetland ecosystems requires understanding

variability in nutrient regimes among wetlands. The primary goal of this study was to use

consistent sampling methods to understand background nutrient conditions in some of the

least impaired watersheds of the southeastern US. An additional objective was to

contrast results based on vegetative community, hydrologic connectivity, and geographic

region. It is hoped that these findings will aid the USEPA in developing numeric nutrient

criteria for wetlands in this region.

One finding from this study stresses the importance of consistent sampling

locations within wetlands for surveying and/or monitoring programs. Water column,

litter, and soil characteristics between the core areas and the edge areas of wetlands

demonstrated significant differences in some parameters. Samples collected at the edge

of a wetland had greater water column total phosphorus and litter total carbon content and

lower soil total carbon and total nitrogen content than samples from the core area of the

same wetland. These differences within wetlands suggest potential implications of

inconsistent sampling techniques on biogeochemical characterizations of wetlands.

Response of wetlands to nutrient change will likely be partially influenced by

vegetative characteristics of the wetland. It was hypothesized that marshes would have

higher soil nutrient concentrations than swamps. Findings from this survey do not

support this hypothesis. It was found that swamps had significantly greater TP (mg/cm3)

than marshes. These results could be influenced by the fact that 64% of the surveyed









wetlands are riverine systems which are often associated with higher nutrient

concentrations.

An additional hypothesis was that marshes would have lower water column

nutrients than swamps. The results of this survey partially support this hypothesis. Total

nitrogen was similar regardless of dominant vegetation type, and total phosphorus

concentrations were significantly greater in swamps. This difference may be correlated

with increased presence of algae in marshes compared to swamps. Algae can quickly

sequester water column P, hence lowering water column TP in marshes (Kadlec and

Knight 1996).

Litter parameters were similar between swamps and marshes, suggesting

distinguishing between these two ecosystem types is not necessary for determining

numeric nutrient criteria. In contrast, water column and soil (mg/cm3) total phosphorus

differences between swamps and marshes demonstrate the need to set numeric nutrient

criteria specific to dominant vegetative cover. A water column based numeric nutrient

criteria may not be the best indicator of wetland nutrient regime. Only 52 of 103 sampled

wetlands had water present within the core and edge area of the wetland at the same time.

Furthermore, water column nutrients can be overly sensitive indicators, since they are

influenced by drought, wind, rain events, and other factors.

It is likely that hydrologic connectivity will also affect the response of wetlands to

nutrient changes. It was hypothesized that riverine wetlands would have higher soil,

litter, and water column nutrient levels than non-riverine systems. The results support

some of the hypotheses. Riverine wetlands had greater water column total phosphorus

than non-riverine systems, but total nitrogen content was similar. Litter total phosphorus









content was also greater in riverine systems, but again nitrogen content was similar. The

results partially support the hypothesis that riverine wetlands would have greater soil

nutrient levels than non-riverine wetlands. Riverine wetlands had greater soil total

phosphorus (mg/cm3), but lower soil total nitrogen (g/kg and mg/cm3) COntent than non-

riverine wetlands.

Results indicate that it may be necessary to identify wetlands as riverine or non-

riverine in order to assign appropriate numeric nutrient criteria. For example, when

riverine and non-riverine wetlands are combined, the soil total nitrogen 75th percentile

value is 7.39 g/kg. When aggregated by hydrologic connectivity, the value is 4.9 g/kg for

riverine systems and 16.5 g/kg for non-riverine systems. If 7.39 g/kg was the numeric

nutrient criterion for soil total nitrogen, then the non-riverine systems would be identified

as threatened by nutrient enrichment. However, non-riverine systems appear to have

approximately three times the soil total nitrogen content of riverine systems. Numeric

nutrient criteria specific to hydrologic connectivity will serve as a more effective

threshold for indicating the nutrient status of wetlands than a single criterion for all

wetlands combined.

The USEPA recognized the importance of regional influences on wetland nutrient

regimes when the decision was made to determine numeric nutrient criteria specific to

ecoregions. Results demonstrate that the Southemn Coastal Plain (XII) is different from

the Southemn Forested Plain (IX) and the Eastemn Coastal Plain (XIV), with greater water

column total nitrogen, litter total carbon, soil total nitrogen, soil total carbon, and lower

litter total phosphorus content. These differences suggest that it is a distinct region with









its own nutrient characteristics, although variation was great enough to warrant further

investigation.

The Southeastern Forested Plain was subdivided into smaller regions by

aggregating the surveyed wetlands by National Forest (or military base). There were no

significant soil total nitrogen (mg/cm3) differences among the sub-regions. However,

there were significant differences among some of the regions with regards to soil total

phosphorus (mg/cm3) COntent. There was almost an order of magnitude difference

between the extreme regions for soil total phosphorus. It is clear that there are significant

regional differences in wetland nutrient regimes at a scale finer than the USEPA Nutrient

Ecoregions. If the ecoregions are sub-divided for determination of numeric nutrient

criteria, the assigned values will more accurately reflect background nutrient

concentrations.

Surveying additional wetlands will assist in determining appropriate water quality

criteria. If similar methods are employed, results can be combined to increase statistical

robustness and decrease variability. Additional studies should concentrate on regional,

vegetative, and hydrologic influences on wetland nutrient regimes.

















APPENDIX A
WETLAND CHARACTERIZATION FORM

Wetland ID: Date:
Observer Name: Picture ID:
Weather Condition:
Is the wetland adjacent to a body of water? Circle the appropriate choice:
River Stream Lake Estuary Ocean None
Characterization for the Entire Wetland (Please circle one of the 1..>;. rate.. classes)
1) Is the vegetation composed predominantly non-vascular (mosses and lichens) ......Moss-Lichen
2) Is the vegetation herbaceous?
i) Is the vegetation dominated by rooted emergent vegetation?.....................Emergent Wetland
ii) Is the vegetation predominately submergent, floating-leaved, or free-floating?.. ..Aquatic Bed
3) Is the vegetation mostly trees and/or shrubs?
i) Is it dominated by vegetation less than 6 meters tall? ........._........Scrub-Shrub Wetland
ii) Are the dominants 6 meters or greater? ........._.._._ ....._........ Forested Wetland
Land-Use Characterization
1) Circle the following land-uses that best characterizes the adjacent upland and estimate the
percentage of the area that is represented by the circled land uses:
a) Commercial g) Rural (scattered homes)
b) Industrial h) Unimproved pasture
c) Golf course i) Forested or wetland
d)High density residential (>20 units/acre) j) Pine plantations
e) Low density residential k) Row crops
f) Feed lots or Dairy operations 1) Other
2) Please circle the following fire indicators present within the vegetation zone:
a) Charred ground surface e)Bumnt dead trees
b) Bumnt trees with new shoots f) Bumnt crowns of trees
c) Bumn marks on trees and shrubs g) Bumned ground with no understory
d) No evidence of fire
3) Is trash present in the wetland?: Yes or No (describe)
4) Is there green algae present in the wetland?: Yes or No (describe)
5) Is there evidence of sedimentation in the wetland? Yes or No (describe)
6) Is there floating vegetation?: Yes or No (describe)
7) Circle any visible indicators of hydrologic disturbances:
a) Ditch e) Dam
b) Nearby road impeding flow f) Dyke
c) Canals g)Piped inflows
d) None noticed h) Other (describe)
8) Circle any visible indicators of vegetative disturbances:
a) Large stand of vines e) Cutting or grazing in wetland
b) Cutting or grazing in adjacent upland f) Insect damage
c) Large stand of exotic species g) Large % of dead trees
d) None noticed h) Other (describe)
9) Circle any direct indicators of nutrient loading to the wetland
a) Presence of cattle in wetland d) Yard waste dumping in/near wetland
b) Fertilizer or manure application in watershed e) None noticed
c) Other (describe)
10) What is the approximate size of the wetland: Shape:











Vegetation Community Characterization Form Sub-sample C (Deep Center)
Wetland ID:
Date:
Start Time: Finish Time:
Photo ID:
Sub-sample C1 Sub-sample C2 Sub-sample C3 Comments
Temp "C

pH
DO %

Conductivity
ORP

Water Depth (inches)
Depth of Organic
laer (inches)
Distance from
ground to lichen lines
(inches)
Algal mats (circle Present Present Present
one) Not present Not present Not present

Aquatic plants Present Present Present
(circle one) Not present Not present Not present
Buttressed roots Buttressed roots Buttressed roots
Morphological
adaptations Adventitious roots Adventitious roots Adventitious roots
(circle any that Hummocks Hummocks Hummocks
apply) None Present None Present None Present
Emerg. Emerg. Emerg.
Macrophytes Macrophytes Macrophytes
Circle the ONE Grasses/sedges Grasses/sedges Grasses/sedges
Chaactriatin tat Floating aquatics Floating aquatics Floating aquatics
best describes the
zoe eigsapld Forested Forested Forested
Scrub-Shrub Scrub-Shrub Scrub-Shrub
Other: Other: Other:

% cover of overstory
List the dominant
overstory vegetation
within a 10-ft radius
of sampling and the
% cover they
represent
% cover of
understory
List the dominant
understory story
vegetation within a
10-ft radius of
sampling and the %
cover they represent











Vegetation Community Characterization Form Sub-sample E (Edge)
Wetland ID:
Date:
Start Time: Finish Time:
Photo ID:
Sub-sample El Sub-sample E2 Sub-sample E3 Comments
Temp "C

pH
DO %

Conductivity
ORP

Water Depth (inches)
Depth of Organic
laer (inches)
Distance from
ground to lichen lines
(inches)
Algal mats (circle Present Present Present
one) Not present Not present Not present

Aquatic plants Present Present Present
(circle one) Not present Not present Not present
Buttressed roots Buttressed roots Buttressed roots
Morphological
adaptations Adventitious roots Adventitious roots Adventitious roots
(circle any that Hummocks Hummocks Hummocks
apply) None Present None Present None Present
Emerg. Emerg. Emerg.
Macrophytes Macrophytes Macrophytes
Circle the ONE Grasses/sedges Grasses/sedges Grasses/sedges
Chaactriatin tat Floating aquatics Floating aquatics Floating aquatics
best describes the
zoe eigsapld Forested Forested Forested
Scrub-Shrub Scrub-Shrub Scrub-Shrub
Other: Other: Other:

% cover of overstory
List the dominant
overstory vegetation
within a 10-ft radius
of sampling and the
% cover they
represent
% cover of
understory
List the dominant
understory story
vegetation within a
10-ft radius of
sampling and the %
cover they represent















APPENDIX B
WETLAND IDENTIFICATION AND LOCATION













Table B-1. Wetland identification and location


ID Hydrology
AL1 Riverine
AL10 Riverine
AL11 Riverine
AL12 Riverine
AL13 Riverine
AL14 Riverine
AL15 Riverine
AL16 Riverine
AL17 Riverine
AL18 Riverine
AL19 Riverine
AL2 Riverine
AL20 Riverine
AL3 Riverine
AL4 Non-riverine
AL5 Non-riverine
AL6 Non-riverine
AL7 Non-riverine
AL8 Riverine
AL9 Riverine
FL10 Non-riverine
Table B-1.Continued
ID Hydrology
FL11 Non-riverine
FL12 Non-riverine
FL13 Non-riverine
FL14 Riverine
FL15 Riverine
FL16 Riverine


Community
Swamp
Swamp
Marsh
Swamp
Swamp
Marsh
Swamp
Marsh
Swamp
Swamp
Swamp
Swamp
Swamp
Swamp
Marsh
Swamp
Swamp
Marsh
Swamp
Swamp
Swamp

Community
Swamp
Marsh
Marsh
Swamp
Swamp
Marsh


Ecoregion
SE Forested Plain (IX)
SE Forested Plain (IX)
SE Forested Plain (IX)
SE Forested Plain (IX)
SE Forested Plain (IX)
SE Forested Plain (IX)
SE Forested Plain (IX)
SE Forested Plain (IX)
SE Forested Plain (IX)
SE Forested Plain (IX)
SE Forested Plain (IX)
SE Forested Plain (IX)
SE Forested Plain (IX)
SE Forested Plain (IX)
SE Forested Plain (IX)
SE Forested Plain (IX)
SE Forested Plain (IX)
SE Forested Plain (IX)
SE Forested Plain (IX)
SE Forested Plain (IX)
S. Costal Plain (XII)

Ecoregion
S. Costal Plain (XII)
S. Costal Plain (XII)
S. Costal Plain (XII)
S. Costal Plain (XII)
S. Costal Plain (XII)
S. Costal Plain (XII)


Location
Conecuh NF
Talladaga NF
Talladaga NF
Talladaga NF
Talladaga NF
Talladaga NF
Talladaga NF
Talladaga NF
Talladaga NF
Talladaga NF
Talladaga NF
Conecuh NF
Talladaga NF
Conecuh NF
Conecuh NF
Conecuh NF
Conecuh NF
Conecuh NF
Conecuh NF
Talladaga NF
Ocala NF


Longitude
-86.52833
-87.38222
-87.48722
-87.69389
-87.37389
-87.34556
-87.39139
-87.55278
-87.56583
-87.68639
-87.46028
-86.68444
-87.55833
-86.73472
-86.57417
-86.84861
-86.65667
-86.85611
-86.75611
-87.32667
-82.14167

Longitude
-81.98306
-81.80000
-81.84778
-81.99472
-81.99333
-81.81750


Latitude
31.27944
33.02167
32.87222
33.09444
33.17611
33.18083
33.10583
33.03639
33.05917
33.08500
33.05611
31.22083
33.08306
31.24194
31.21528
31.14194
31.22278
31.13000
31.33972
32.80056
29.44056

Latitude
29.35611
29.23361
29.33861
29.34278
29.34528
29.53917


Location
Ocala NF
Ocala NF
Ocala NF
Ocala NF
Ocala NF
Ocala NF













FL17 Non-riverine
FL18 Riverine
FL19 Riverine
FL20 Non-riverine
FL21 Riverine
FL22 Non-riverine
FL23 Riverine
FL24 Non-riverine
FL25 Non-riverine
FL26 Non-riverine
FL27 Riverine
FL28 Non-riverine
FL29 Riverine
FL30 Non-riverine
FL31 Riverine
FL32 Non-riverine
FL33 Riverine
FL34 Non-riverine
FL35 Non-riverine
FL36 Non-riverine
FL37 Riverine
FL38 Non-riverine
FL39 Non-riverine
Table B-1.Continued
ID Hydrology
FL40 Riverine
FL41 Non-riverine
FL42 Riverine
FL43 Non-riverine
FL44 Riverine
FL45 Non-riverine


Swamp
Swamp
Swamp
Marsh
Swamp
Swamp
Marsh
Marsh
Marsh
Swamp
Swamp
Swamp
Swamp
Swamp
Swamp
Marsh
Swamp
Swamp
Swamp
Swamp
Swamp
Marsh
Swamp


S. Costal Plain (XII)
S. Costal Plain (XII)
S. Costal Plain (XII)
S. Costal Plain (XII)
S. Costal Plain (XII)
SE Forested Plain (IX)
S. Costal Plain (XII)
S. Costal Plain (XII)
SE Forested Plain (IX)
S. Costal Plain (XII)
S. Costal Plain (XII)
SE Forested Plain (IX)


SE Forested Plain (IX)
S. Costal Plain (XII)
S. Costal Plain (XII)
SE Forested Plain (IX)
SE Forested Plain (IX)
S. Costal Plain (XII)
S. Costal Plain (XII)
S. Costal Plain (XII)
S. Costal Plain (XII)

Ecoregion
S. Costal Plain (XII)
S. Costal Plain (XII)
S. Costal Plain (XII)
S. Costal Plain (XII)
S. Costal Plain (XII)
S. Costal Plain (XII)


Ocala NF
Ocala NF
Ocala NF
Ocala NF
Osceola NF
Osceola NF
Ocala NF
Apalachicola NF
Apalachicola NF
Apalachicola NF
Apalachicola NF
Apalachicola NF
Apalachicola NF
Apalachicola NF
Apalachicola NF
Apalachicola NF
Apalachicola NF
Apalachicola NF
Apalachicola NF
Osceola NF
Osceola NF
Osceola NF
Osceola NF


-81.80528
-81.67722
-81.68139
-81.79250
-82.66861
-82.75167
-81.73083
-84.48167
-84.68611
-84.58944
-84.59472
-84.63694


-84.84000
-84.89139
-85.15889
-84.73111
-85.01222
-82.48417
-82.62333
-82.64639


29.42778
29.09278
29.08500
29.25861
30.34944
30.31722
29.43250
30.40944
30.48611
30.36417
30.36333
30.53778


30.35444
30.23194
30.18500
30.35556
30.36778
30.35389
30.52083
30.62028


Community
Swamp
Swamp
Swamp
Swamp
Swamp
Swamp


Location
Osceola NF
Osceola NF
Osceola NF
Osceola NF
Osceola NF
Osceola NF


Longitude
-82.66806
-82.59639
-82.57417
-82.60306

-82.58417


Latitude
30.47111
30.46472
30.40667
30.48167

30.31583













GAl Riverine
GA10 Riverine
GAl6 Non-riverine
GAl7 Non-riverine
GAl9 Non-riverine
GA2 Riverine
GA20 Non-riverine
GA21 Non-riverine
GA25 Riverine
GA26 Riverine
GA27 Riverine
GA28 Riverine
GA29 Riverine
GA3 Non-riverine
GA30 Riverine
GA31 Non-riverine
GA32 Riverine
GA4 Riverine
GAS Non-riverine
GA6 Riverine
GA7 Riverine
GA8 Riverine
GA9 Riverine
Table B-1.Continued
ID Hydrology
SC1 Riverine
SC10 Riverine
SC11 Riverine
SC12 Riverine
SC13 Riverine
SC14 Riverine


Swamp
Marsh
Swamp
Swamp
Swamp
Swamp
Swamp
Swamp
Swamp
Swamp
Swamp
Swamp
Swamp
Swamp
Swamp
Swamp
Swamp
Swamp
Marsh
Swamp
Swamp
Swamp
Swamp


SE Forested Plain (IX)
SE Forested Plain (IX)
SE Forested Plain (IX)
SE Forested Plain (IX)
SE Forested Plain (IX)
SE Forested Plain (IX)
SE Forested Plain (IX)
SE Forested Plain (IX)
SE Forested Plain (IX)
SE Forested Plain (IX)
SE Forested Plain (IX)
SE Forested Plain (IX)
SE Forested Plain (IX)
SE Forested Plain (IX)
SE Forested Plain (IX)
SE Forested Plain (IX)
SE Forested Plain (IX)
SE Forested Plain (IX)
SE Forested Plain (IX)
SE Forested Plain (IX)
SE Forested Plain (IX)
SE Forested Plain (IX)
SE Forested Plain (IX)

Ecoregi;on
E. Coastal Plain (XIV)
E. Coastal Plain (XIV)
E. Coastal Plain (XIV)
E. Coastal Plain (XIV)
SE Forested Plain (IX)
SE Forested Plain (IX)


Oconee NF
Oconee NF
Grand Bay NWF
Grand Bay NWF
Moody AFB
Oconee NF
Moody AFB
Moody AFB
Fort Benning MB
Fort Benning MB
Fort Benning MB
Fort Benning MB
Fort Benning MB
Oconee NF
Fort Benning MB
Fort Benning MB
Fort Benning MB
Oconee NF
Oconee NF
Oconee NF
Oconee NF
Oconee NF
Oconee NF

Location
Francis Marion NF
Francis Marion NF
Francis Marion NF
Francis Marion NF
Sumter NF
Sumter NF


-83.54250
-83.54639
-83.42917
-83.29778
-83.36667
-83.77556
-83.27889
-83.28861
-84.91056
-84.90056
-84.88167
-84.92556
-84.90500
-83.53111
-84.77333
-84.95111
-85.06278
-83.82722
-83.86167
-84.05972
-83.87694
-84.06056
-83.89250

Longitude
-79.85333
-80.11722
-79.73611
-79.46889
-81.70639
-81.53806


33.37333
33.80583
30.99750
31.09556
31.08250
33.39028
30.97083
31.03861
32.43806
32.44694
32.59917
32.50000
32.41500
33.38556
32.40972
32.72000
32.60111
33.32361
33.24444
33.35528
33.48306
33.47056
33.23167

Latitude
33.35917
33.25917
33.39028
33.35361
34.45361
34.56056


Community
Swamp
Swamp
Swamp
Swamp
Swamp
Swamp













SC15
SC16
SC17
SC18
SC19
SC2
SC20
SC21
SC22
SC23
SC24
SC3
SC4
SC5
SC6
SC7
SC8
SC9


Riverine
Riverine
Riverine
Riverine
Riverine
Riverine
Riverine
Riverine
Riverine
Riverine
Riverine
Non-riverine
Non-riverine
Non-riverine
Riverine
Riverine
Riverine
Non-riverine


Swamp
Swamp
Swamp
Marsh
Swamp
Swamp
Swamp
Swamp
Swamp
Swamp
Swamp
Swamp
Swamp
Swamp
Swamp
Marsh
Swamp
Swamp


SE Forested Plain (IX)
SE Forested Plain (IX)
SE Forested Plain (IX)
SE Forested Plain (IX)
SE Forested Plain (IX)
E. Coastal Plain (XIV)
SE Forested Plain (IX)
SE Forested Plain (IX)
SE Forested Plain (IX)
SE Forested Plain (IX)
SE Forested Plain (IX)
E. Coastal Plain (XIV)
E. Coastal Plain (XIV)
E. Coastal Plain (XIV)
E. Coastal Plain (XIV)
E. Coastal Plain (XIV)
E. Coastal Plain (XIV)
E. Coastal Plain (XIV)


Sumter NF
Sumter NF
Sumter NF
Sumter NF
Sumter NF
Francis Marion NF
Sumter NF
Sumter NF
Sumter NF
Sumter NF
Sumter NF
Francis Marion NF
Francis Marion NF
Francis Marion NF
Francis Marion NF
Francis Marion NF
Francis Marion NF
Francis Marion NF


-81.81222
-81.67611
-81.88556
-81.97722

-79.99917
-81.78639
-81.73833
-81.39694
-81.58944
-81.68750
-79.99083
-79.95528
-79.79750
-79.87194
-79.91556
-79.97083
-79.86306


34.63000
34.75056
34.77583
34.66667

33.55389
34.56167
34.47556
34.47028
34.43167
34.59500
33.43083
33.44806
33.26639
33.17861
33.27750
33.28278
33.21389














APPENDIX C
PHYSICAL SOIL AND WATER COLUMN DATA













Table C-1.Physical soil and water column data. Water column data is an average of sub-sample locations within the core (C) or edge


(E) transect.

ID Area


Soil moisture
content
(%)
39%
30%
66%
80%
50%
64%
59%
35%
46%
38%
67%
67%
36%
72%
73%
56%
31%
33%


Soil bulk
density
(g /cm3
0.73
0.95
0.39
0.21
0.67
0.41
0.51
0.99
0.70
0.92
0.41
0.42
0.97
0.31
0.30
0.61
0.72
0.76


Water
temp
O"C)


Water
DO




52.40
13.70
14.30
49.53
33.17
18.40


Water
Eh




414.0
342.7
267.3
298.5


Water
Depth
(CM)


Soil LOI
(%)
9.7
6.3
20.4
37.9
10.0
15.6
12.9
7.3
10.3
7.2
15.6
18.2
23.2
24.9
28.4
10.4
8.6
8.5


Water conductivity
(uS/cm)


Water
pH


AL 1
AL 1
AL 10
AL 10
AL 11
AL 11
AL 12
AL 12
AL 13
AL 13
00AL14
AL 14
AL 15
AL 15
AL 16
AL 16
AL 17
AL 17


22.4
19.4
23.6
23.4
24.8
22.3


5.31
5.21
5.80
5.57
5.69
5.54


18.0
22.0
57.7
46.7
53.7
38.0


133.7
125.3


47.3
32.7


10.0
2.3
15.3
5.2
5.3
-3.0


23.0 6.45 4.13
23.1 6.30 8.00


151.7 26.0
175.1 17.3


26.2
27.7


5.46
5.38


4.77
33.53


276.1
291.4


19.0
10.0