Citation
Developing Biological Indicators for Isolated Forested Wetlands in Florida

Material Information

Title:
Developing Biological Indicators for Isolated Forested Wetlands in Florida
Creator:
REISS, KELLY CHINNERS
Copyright Date:
2008

Subjects

Subjects / Keywords:
Constructed wetlands ( jstor )
Diatoms ( jstor )
Distance functions ( jstor )
Ecoregions ( jstor )
Indicator species ( jstor )
Land use ( jstor )
Macroinvertebrates ( jstor )
Macrophytes ( jstor )
Species ( jstor )
Wetlands ( jstor )
City of Gainesville ( local )

Record Information

Source Institution:
University of Florida
Holding Location:
University of Florida
Rights Management:
Copyright Kelly Chinners Reiss. Permission granted to the University of Florida to digitize, archive and distribute this item for non-profit research and educational purposes. Any reuse of this item in excess of fair use or other copyright exemptions requires permission of the copyright holder.
Embargo Date:
4/30/2004
Resource Identifier:
55802506 ( OCLC )

Downloads

This item is only available as the following downloads:


Full Text

PAGE 1

DEVELOPING BIOLOGICAL INDICATORS FOR ISOLATED FORESTED WETLANDS IN FLORIDA By KELLY CHINNERS REISS A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLOR IDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2004

PAGE 2

Copyright 2004 by Kelly Chinners Reiss

PAGE 3

This document is dedicated to my family and friends: Lena, Lucky, Casey, Nana, Ben, Krista, Wendy, and Mike.

PAGE 4

iv ACKNOWLEDGMENTS I would like to thank my advisor, Dr . Mark Brown, and committee members Dr. Mary Duryea, Dr. Wiley Kitchens, and Dr . Clay Montague for their thoughtful discussion, time, and energy. Biological indi cator research was s upported by a grant to Dr. Mark Brown from the Florida Department of Environmental Protection (FDEP). The FDEP staff provided support fo r this research (particula rly Russ Frydenborg, Ashley OÂ’Neal, Ellen McCarron, Julie Espy, Tom Frick, Joy Jackson, Liz Miller, Johnny Richardson, and Lori Wolf). Additionally, I would like to thank the systems ecol ogy research group at the H.T. Odum Center for Wetlands for stimulating di scussion and valuable input along the way. In particular, Chuck Lane (who worked on bi ological indicators for Florida isolated marshes) provided groundwork for this analys is. Assistance in field-data collection, laboratory analysis, data entry, and/or feedb ack on statistical analyses through the years from Eliana Bardi, Susan Carstenn, Matt Cohen, Tony Davanzo, Melissa Friedman, Kristina Jackson, Althea Moore, Mike Murray-Hudson, Joanna Reilly-Brown, Vanessa Rumancik, Lisa Spurrier, Jim Surdick, Ben Vi vas, and Melissa Yonteck, was particularly valuable. An important statement of gratitude to the land owners and managers who allowed access to the 118 wetlands throughout Florida is particularly important. Without their cooperation, this study never would have been possible.

PAGE 5

v TABLE OF CONTENTS page ACKNOWLEDGMENTS.................................................................................................iv LIST OF TABLES...........................................................................................................viii LIST OF FIGURES...........................................................................................................xi ABSTRACT.....................................................................................................................xi v CHAPTER 1 INTRODUCTION........................................................................................................1 Statement of the Problem..............................................................................................1 Defining Ecosystem Integrity.......................................................................................3 Historical Perspective............................................................................................3 Water Quality Criteria...........................................................................................3 Biological Indicators of Ecosystem Integrity........................................................5 Diatoms as biologi cal indicators....................................................................6 Macrophytes as biological indicators.............................................................9 Macroinvertebrates as biological indicators.................................................12 Review of Isolated Freshwater Forested Wetlands....................................................18 Changes in Hydrology.........................................................................................21 Increased Inflows of Nutrients and/or Toxins.....................................................25 Physical Disturbance...........................................................................................27 Quantifying Anthropogenic Influence........................................................................29 Landscape Development Intensity Index............................................................31 Wetland Rapid Assessment Procedure................................................................32 Minnesota Disturbance Index..............................................................................32 Plan of Study...............................................................................................................33 2 METHODS.................................................................................................................34 Site Selection..............................................................................................................34 Gradients of Landscape De velopment Intensity.........................................................38 Field-data Collection..................................................................................................41 Sampling Design.................................................................................................42 Water Samples.....................................................................................................42 Soil Samples........................................................................................................43

PAGE 6

vi Diatoms................................................................................................................43 Macrophytes........................................................................................................45 Supplementary data......................................................................................45 Floristic Quality Index.................................................................................46 Macroinvertebrates..............................................................................................47 Data Analysis..............................................................................................................49 Water and Soil Parameters..................................................................................49 Summary Statistics..............................................................................................50 Regional Compositional Analysis.......................................................................53 Community Composition....................................................................................53 Metric Development............................................................................................55 Indicator Species Analysis...........................................................................57 Diatom metrics.............................................................................................58 Macrophyte metrics......................................................................................59 Macroinvertebrate metrics............................................................................60 Wetland Condition Index....................................................................................62 Cluster Analysis...................................................................................................62 Comparisons among Wetland C ondition Index Metrics.....................................63 3 RESULTS...................................................................................................................64 Water and Soil Parameters..........................................................................................64 Diatoms.......................................................................................................................6 5 Summary Statistics..............................................................................................66 Compositional Analysis.......................................................................................67 Community Composition....................................................................................68 Metric Selection...................................................................................................72 Tolerance metrics.........................................................................................73 Autecological metrics...................................................................................77 Diatom Wetland Condition Index.......................................................................82 Cluster Analysis...................................................................................................82 Macrophytes...............................................................................................................83 Summary Statistics..............................................................................................85 Compositional Analysis.......................................................................................86 Community Composition....................................................................................87 Metric Selection...................................................................................................89 Tolerance metrics.........................................................................................91 Modified Floristic Quality Index metric....................................................104 Exotic species metric..................................................................................106 Native perennial species metric..................................................................106 Wetland status metric.................................................................................108 Macrophyte Wetland Condition Index..............................................................109 Cluster Analysis.................................................................................................115 Macroinvertebrates...................................................................................................116 Summary Statistics............................................................................................117 Compositional Analysis.....................................................................................118 Community Composition..................................................................................122

PAGE 7

vii Metric Selection.................................................................................................123 Tolerance metrics.......................................................................................124 Community balance metrics.......................................................................130 Functional group metrics............................................................................133 Macroinvertebrate Wetland Condition Index....................................................134 Cluster Analysis.................................................................................................135 Wetland Condition Index..........................................................................................135 4 DISCUSSION...........................................................................................................145 Richness, Evenness, and Diversity...........................................................................146 Describing Biological Integrity................................................................................147 Merits of a Multi-Metric Multi-Assemblage WCI...................................................149 A Case for Regionalization.......................................................................................151 WCI Independent of Wetland Type..........................................................................153 Wetland Value..........................................................................................................155 Limitations and Further Research.............................................................................157 Conclusions...............................................................................................................158 APPENDIX A ENERGY CIRCUIT LANGUAGE..........................................................................160 B QUANTIFYING ANTHROPOGENIC INFLUENCE.............................................161 C STANDARD OPERAT ING PROCEDURES..........................................................163 D COEFFICIENT OF CONSERVATISM SCORES..................................................173 E CANDIDATE METRICS.........................................................................................180 F SUMMARY STATISTICS......................................................................................189 LIST OF REFERENCES.................................................................................................194 BIOGRAPHICAL SKETCH...........................................................................................210

PAGE 8

viii LIST OF TABLES Table page 2-1 Surrounding land use, land ownership, and samp le date for 118 study wetlands in Florida...................................................................................................37 2-2 Field-data collected at 118 sample wetlands............................................................39 2-3 Landscape Development Coefficients used in the calculation of the Landscape Development Intensity index ................................................................41 3-1 Water and soil parameters among 3 a priori land use categories............................65 3-2 Water and soil parameters among LDI groups.........................................................66 3-3 Diatom richness, evenness, and diversity among a priori land use categories........67 3-4 Mean diatom summary stat istics between LDI groups............................................68 3-5 Similarity of diatom community composition using MRPP....................................69 3-6 Pearson correlations between environmen tal parameters and NMS ordination axes based on diatom community composition........................................................71 3-7 SpearmanÂ’s correlations for 7 diatom metrics and LDI...........................................72 3-8 Comparisons among diatom metrics and the diatom WCI for LDI groups.............73 3-9 Spearman correlations of diatom indicat or species over a range of LDI values......74 3-10 Diatom tolerant indicator species.............................................................................75 3-11 Diatom sensitive indicator species...........................................................................76 3-12 Diatom WCI and LDI values for wetland clus ters based on diatom community composition...........................................................................................85 3-13 Mean macrophyte richness, evenness, and diversity among a priori land use categories...........................................................................................................87 3-14 Mean macrophyte richness, evenness , and diversity between LDI groups..............87 3-15 Macrophyte community composition sim ilarity among ecoregions with MRPP....88

PAGE 9

ix 3-16 Pearson correlations between environmental variables and NMS axes based on macrophyte community composition at 118 wetlands.............................................91 3-17 Pearson correlations between environmental variables and NMS axes based on macrophyte community composition at 75 wetlands...............................................93 3-18 Spearman correlations between six macrophyte metrics and LDI...........................94 3-19 Comparisons among 6 macr ophyte metrics for LDI groups....................................94 3-20 Macrophyte ISA calculations were co nducted over a range of LDI values.............95 3-21 Statewide and regional macrophyt e tolerant indicator species................................96 3-22 Statewide and regional macrophyt e sensitive indicator species.............................100 3-23 Statewide and regional m acrophyte indicator species were significantly correlated with LDI................................................................................................103 3-24 Exotic macrophyte species identified at 118 study wetlands.................................107 3-25 Macrophyte WCI and metrics sc ored statewide and regionally for study wetlands in the low LDI group...............................................................................112 3-26 Macrophyte WCI and metrics scored statew ide and regionally for study wetlands in the high LDI group..............................................................................113 3-27 Spearman correlations between the macrophyte WCI, metrics, and LDI..............114 3-28 Macrophyte WCI scores and LDI values for wetland clusters based on macrophyte community composition.....................................................................116 3-29 Macroinvertebrate richness, evenness, and diversity among a priori land use categories.........................................................................................................118 3-30 Macroinvertebrate richness, eve nness, and diversity for LDI groups....................119 3-31 Macroinvertebrate community composition similarity among a priori land use categories and ecoregions.......................................................................................120 3-32 PearsonÂ’s r-squared correlations be tween environmental variables and NMS ordination axes based on macroinve rtebrate community composition.................122 3-33 Spearman correlations between macroinvertebrate metrics and the macroinvertebrate WCI with LDI, pH, dissolved oxygen (DO), and total phosphorus (TP).....................................................................................................123 3-34 Macroinvertebrate metric and WCI scores between LDI groups...........................124

PAGE 10

x 3-35 Macroinvertebrate ISA calculatio ns over a range of LDI values...........................125 3-36 Macroinvertebrate tole rant indicator genera..........................................................126 3-37 Sensitive macroinverte brate indicator genera........................................................128 3-38 Macroinvertebrate WCI scores and LDI values fo r wetland clusters based on macroinvertebrate co mmunity composition......................................................137 3-39 WCI scores for 118 wetlands based on three a ssemblages including diatoms, macrophytes, and macroinvertebrates.....................................................138 3-40 Pearson correlations among 19 metrics..................................................................144 A-1 Symbols used in ener gy circuit diagramming........................................................160 B-1 LDI, WRAP, and Minnesota distur bance index scores for 118 wetlands..............161 D-1 Coefficient of Conservatism (CC) scores for 561 macrophytes identified in isolated depressional freshwater forested wetlands in Florida...........................173 E-1 Candidate metrics based on the diatom assemblage..............................................180 E-2 Candidate metrics based on the macrophyte assemblage.......................................182 E-3 Candidate metrics based on th e macroinvertebrate assemblage............................185 F-1 Summary statistics of richness (R ), evenness (E), Shannon diversity (H), and SimpsonÂ’sindex (S) for the diat om assemblage (genus level).........................189 F-2 Summary statistics of ric hness (R), jackknife estimators of species richness (Jack1, Jack2), evenness (E), Shannon diversity (H), and Whitta kerÂ’s beta diversity ( W) for the macrophyte assemblage (species level).......................190 F-3 Summary statistics of richness (R), evenness (E), Shannon diversity (H), and SimpsonÂ’s index (S) for the macroinvert ebrate assemblage (genus level)............193

PAGE 11

xi LIST OF FIGURES Figure page 1-1 Systems diagram showing major sources, st orages, and flows of a cypress dome.....................................................................................................19 1-2 Aggregate systems diagram of a cypress dome embedded within a developed landscape..................................................................................................................21 1-3 Mechanism of altered hydrology of a wetland in a developed landscape................22 1-4 Increased nutrients and/ or toxin inflows into a wetland from the surrounding developed landscape.................................................................................................26 1-5 Potential physical alterations to a pondcypress wetland..........................................28 2-1 Study site location of 118 isolat ed forested wetlands in Florida..............................35 2-2 Belted transect layout for macrophyte sampling and the location of the water and soil samples..............................................................................................42 2-3 Benthic diatom samples were collect ed at 50 isolated forested wetlands................44 2-4. Macroinvertebrates were sample d at 79 isolated forested wetlands..........................48 3-1 NMS ordination bi-plot of 50 wetlands in diatom species space with an overlay of environmental parameters..................................................................70 3-2 Percent diatom tolerant indicator species increased with increasing development intensity..............................................................................................75 3-3 Percent diatom sensitive indicator species decreased with increasing development intensity..............................................................................................77 3-4 Pollution tolerance class 1 diatoms increased w ith increasing development intensity..............................................................................................78 3-5 Nitrogen uptake metabolism class 3 diatoms increased with increasing development intensity.............................................................................79 3-6 Saprobity class 4 diatoms increa sed with increasing development intensity....................................................................................................................80

PAGE 12

xii 3-7 The pH class 3 diatoms increased with increasing development intensity....................................................................................................................81 3-8 Dissolved oxygen class 1 diatom s decreased with increasing development intensity..............................................................................................81 3-9 D-WCI scores decrease with increasing development intensity..............................83 3-10 Diatom WCI scores for wetland clusters based on diatom community composition .............................................................................................................84 3-11 NMS ordination bi-plot of 118 sa mple wetlands in macrophyte species space with an overlay of environmental parameters................................................90 3-12 NMS ordination bi-plot of 75 sa mple wetlands in macrophyte species space with an overlay of environmental parameters................................................92 3-13 Tolerant macrophyte indicator species increased with increasing development intensity..............................................................................................98 3-14 Macrophyte sensitive indicator species decrea sed with increasing development intensity............................................................................................102 3-15 Modified FQI scores decreased w ith increasing development intensity................104 3-16 Exotic species increased with increasing development intensity...........................105 3-17 Native perennial species decreased w ith increasing development intensity..........109 3-18. The percent wetland status species decreased with increasing development intensity..................................................................................................................110 3-19 Macrophyte WCI scores decreased with increasing development intensity..........111 3-20 Regional macrophyte WCI scores for 5 wetland clusters based on macrophyte community composition.........................................................................................115 3-21 NMS ordination bi-plot for 79 wetlands in macr oinvertebrate genus space with an overlay of environmental parameters........................................................121 3-22 Tolerant macroinvertebrate indicato r genera increased with increasing development intensity............................................................................................127 3-23 Sensitive macroinvertebrate indicator genera decreased with increasing development intensity............................................................................................129 3-24 Florida Index scores decreased wi th increasing development intensity.................130

PAGE 13

xiii 3-25 Macroinvertebrates in the phylum Mollusca increa sed with increasing development intensity............................................................................................131 3-26 Macroinvertebrates in the family No teridae decreased with increasing development intensity............................................................................................132 3-27 Macroinvertebrates that belong to the scraper functiona l feeding group increased with increasing development intensity...................................................133 3-28 Macroinvertebrate WCI scores decrease d with increasing landscape development intensity index...................................................................................134 3-29. Macroinvertebrate WCI scores for 5 wetland clusters based on macroinvertebrate community composition...........................................................136 3-30 Three dimensional scatter plot of the WC I based on three assemblages, including diatoms, macrophytes , and macroinvertebrates.....................................142 3-31 Scatterplots of WCI scores for wetlands based on diatom, macrophyte, and macroinvertebrate assemblages..............................................................................143

PAGE 14

xiv Abstract of Dissertation Pres ented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy DEVELOPING BIOLOGICAL INDICATORS FOR ISOLATED FORESTED WETLANDS IN FLORIDA By Kelly Chinners Reiss May 2004 Chair: Mark T. Brown Major Department: Environmental Engineering Sciences The Wetland Condition Index (WCI) provi ded a quantitative measure of the biological integrity of isolated forested wetlands in Florida. Environmental parameters and community composition of the diat om, macrophyte, and macroinvertebrate assemblages were sampled in 118 isolated forested wetlands throughout Florida to answer the overall question: can change s in the biotic components of pondcypress wetlands (such as the community com position of the diatom, macrophyte, and macroinvertebrate assemblages) be related to changes in development intensity in the landscape immediately adjacent to and surroundi ng them. While richness, evenness, and diversity measures were not sensitive to changes in landscape development intensity, biological indicators along with physical and chemical parameters were useful in defining biological integrity. Differences in diatom, macrophyte, and macroinvertebrate community composition were explored in nonmetric multidimensional scaling (NMS) ordinations. Water-column

PAGE 15

xv pH was correlated with the community co mposition of all 3 assemblages. Each assemblage was used to construct the WCI fo r isolated forested wetlands in Florida, which included 19 total metrics (7 diatom metrics, 6 macrophyte metrics, and 6 macroinvertebrate metrics). All metrics were significantly correlated (SpearmanÂ’s correlation coefficient, p < 0.05) with the La ndscape Development Intensity (LDI) index, a measure of the use of nonrenewable energy in the su rrounding landscape. While the WCI suggested low biological integrity of both agricultural and urban wetlands, these wetlands provide services and do work in th e environment. Therefore, the quantitative score of biol ogical integrity esta blished through the WCI should not be used as a surrogate for wetland value, but ra ther as an objective, quantitative means of comparing changes in community composition along gradients of landscape development intensity. In the future, an integrative multi-metric multi-assemblage WCI could be constructed for wetlands throughout the state, w ith lists of indicator species and metric scores dependent on Florida ecoregions and specific to wetland type.

PAGE 16

1 CHAPTER 1 INTRODUCTION With even casual observation, it is apparent that ecosystems change with increasing levels of human development. The extent of change is observably related to the magnitude of human activity. While prev ious research has identified ecosystem responses to human induced change s such as increased nutrients ( Nessel et al. 1982 ; Lemlich and Ewel 1984 ; Devall 1998 ) or altered hydrology ( Marois and Ewel 1983 ; Lugo and Brown 1986 ; Young et al. 1995 ) few have studied the amalgamated response of ecosystems resultant from the combined effect s of anthropogenic development. This is especially true for wetlands located within urban settings. Our study aimed at understanding the effects of landscape development (from anthropogenic activities), and focused on one specific ecosystem t ype, the isolated pondcypress dome. Statement of the Problem Pondcypress domes are isolated depressional forested wetlands. These historically nutrient-poor ecosystems occur throughout Florida and the s outheastern United States coastal plain. Primary driving energies include inputs from rainfall a nd localized run-off. Because of their position in the landscape, pondcypress domes have specific hydrologic and nutrient regimes that regulate species co mposition. Conversion of lands adjacent to and surrounding isolated wetlands to more in tensively managed land uses may alter the driving energies of the wetland; and since driving energies are fundamental to ecosystem organization, changes in inflows may influen ce the rates and direction of processes and ultimately system organization. The changes resultant from modification of the driving

PAGE 17

2 energies may in turn be manifest in detect able differences in the biotic components. Change is further defined as a detectable difference between current conditions from the reference condition. Reference condition is defined as the condition of wetlands surrounded by undeveloped landscapes and without apparent human induced alterations. While it may be difficult to discern the exact causal agent of such changes, it may be possible to detect differences in various biot ic assemblages of ecosystems and relate them to surrounding land use intensity. Our study focuses on understanding the cha nges in the biotic assemblages that occur in isolated pondcypress wetlands resu lting from differen t land uses in the landscapes surrounding them. The major questio n addressed in this dissertation is: can the changes in biotic co mponents of pondcypress wetlands (such as the community composition of the diatom, macrophyte, and macr oinvertebrate assemblages) be related to changes in development intensity in the landscape immediately adjacent to and surrounding the wetlands. From this main que stion several secondary questions arise. Are there differences among pondcypress domes surrounded by differe nt land uses? If change does occur, what describes the ch ange in pondcypress communities? Biological signals may be apparent in the diatom, macrophyte, and macroinvertebrate community composition and in the abiotic components. Can the presence of particular diatoms, macrophytes, or macroinvertebrates be used as an indication of cha nge? Are differences in physical and chemical water and soil cr iteria detectable in wetlands surrounded by different land uses? The extent of change of wetland biota may be an indicator of change in community structure and thus indicative of what has been termed ecosystem integrity ( Karr and Dudley 1981 ). By analyzing changes in multiple assemblages and relating

PAGE 18

3 them to the intensity of land use surr ounding each wetland, new insights concerning the effects of land use on wetland structur e and integrity may be generated. Defining Ecosystem Integrity Determining ecosystem integrity through th e use of biological indicators requires an accepted definition of integrity. Karr and Dudley ( page 55, 1981 ) defined integrity as “the ability of an aquatic ecosystem to support and mainta in a balanced, integrated, adaptive community of organisms having a species composition, diversity, and functional organization comparable to that of the natura l habitats of the regi on.” This definition requires two things: a definiti on of the natural habitat (o r reference condition) and appropriate regionalization. Gerristen et al. ( 2000 ) concur that biological assessment relies on a characterization of the reference condition. Historical Perspective Over 30 years ago, the passage of the Wa ter Pollution and Control Act (later referred to as the Clean Water Act, 1972) requi red states to “restore and maintain the chemical, physical, and biological in tegrity of the Nation’s waters” ( USEPA 1990 ). This legislation included establishi ng water-quality standards fo r all waters within state boundaries, including wetlands. Criteria for de fining water-quality c ould be narrative or numeric; and it could be addr essed through chemical, physical, or biological standards. Initially, states used chemical and physic al criteria (testing waters for chemical concentrations or physical conditions that exceeded criteria) and assuming losses in ecosystem integrity if the criteria were exceeded ( Danielson 1998a ). Water Quality Criteria There are several shortcomings in deri ving ecosystem integrity based on exceeding established limits for chemical and physical parameters. Such criteria have been

PAGE 19

4 considered rudimentary in their ability to re flect more than the temporal concentration within a water body ( Karr 1993 ). For instance, the use of toxicity parameters for determining ecosystem integrity may falsely indicate high ecosystem integrity simply because a single toxicity parameter went unde tected. This same water body could have undesirable levels of other nontar get toxics or metals; or be physically altered so that it no longer resembles a fully functioning water body ( Karr and Chu 1997 ). Furthermore, chemical and physical sampling may not occu r during specific loading events and may therefore incompletely describe the biologi cal and ecological condition of the system. Adams ( 2002 ) points out that other environmenta l factors (such as sedimentation, alterations to habitat, varying temperature and oxygen levels, and changes in ecological aspects like food availa bility and predator-prey relati onships) are not reflected with chemical criteria alone. James and Kleinow ( 1994 ) add that different organisms respond in different ways to the amount, persiste nce, and exposure of xenobiotics (chemical compounds otherwise foreign to an organism); and single-valued chemical and physical criteria of water quality may overl ook important biological implications. Alternatively, biological indica tors integrate the spatial a nd temporal effects of the environment on resident organisms, and are su itable for assessing the possible effects of multifaceted changes in aquatic ecosystems ( Adams 2002 ). Adams ( 2002 ) and Karr and Chu ( 1997 ) note that biological indicators signal changes in the environment that might otherwise be overlooked or underestimated by methods that depend on chemical criteria alone. The underlying support for using biological indicators is that organisms have an intricate relationship with their environment, which reflects current and cumulative ecosystem conditions ( Karr 1981 ). Biological indicators re flect chemical exposure and

PAGE 20

5 also integrate change s in the community composition of the ecosystem (from physical, chemical, and biological changes) ( Adams 2002 ). The United States Environmental Prot ection Agency (USEPA) recognized the potential of biological criteri a to assess water-quality standards and in the late 1980s required states to use biological indicators to accomplish the goals of the Clean Water Act ( USEPA 1990 ). In effect, biological assessment has evolved into one of the standard monitoring tools of water resource-protecti on agencies over the past 2 decades ( Gerristen et al. 2000 ). Biological criteria and monitori ng programs through the USEPA have been created for lakes and streams throughout the United States ( Barbour et al. 1996a ; Karr and Chu 1999 ; Gerristen et al. 2000 ), and more recently efforts to assess wetland condition have been initiated ( USEPA 2002a ). Currently wetland bioassessment programs are in place or being developed in 14 states ( USEPA 2003 ). Biological Indicators of Ecosystem Integrity Biological monitoring to assess ecosystem condition has been applied widely in ecological research. One trend in biologica l monitoring has led to the development of indices of biological integrity (often referred to as IBIs), for different species assemblages including diatoms ( Fore and Grafe 2002 ); macrophytes ( Galatowitsch et al. 1999a ; Gernes and Helgen 1999 ; Mack 2001 ; Lane 2003 ); macroinvertebrates ( Kerans and Karr 1994 ; Barbour et al. 1996b ); fish ( Schulz et al. 1999 ); and birds ( OÂ’Connell et al. 1998 ). Such indices have been applied to ecosystems throughout the world including in Europe ( Kelly and Whitton 1998 ); Japan ( Mack 2001 ); widely throughout the United States ( Karr 1981 ; Lenat 1993 ; Fore and Grafe 2002 ; Lane et al. 2002 ); and is beginning in Australia by J.E. Ling of the Royal Botani cal Gardens, University of Western Sydney. The primary aim of biological monitoring is to detect changes in abundance, structure,

PAGE 21

6 and diversity of target species assemblages. Danielson ( 1998a ) notes that biological signals are effective mainly because biologi cal monitoring incorporates changes from various collective consta nt or pulsing sources. Many studies have created multimetr ic indices of bi ological condition, incorporating individual metr ics into a quantitative value of community condition or ecosystem integrity. Karr and Chu ( 1997 ) defined metrics as biol ogical attributes that have a consistent and pred ictable response to anthropoge nic activities. Anthropogenic activities can alter the integrity of wetland ecosystems by causing one or more of the following conditions: eutrophica tion, contaminant toxicity, acidification, salinization, sedimentation, burial, thermal alteration, vegetation removal, turbidity, shading, dehydration or inundation, and/or habitat fragmentation ( Danielson 1998a ). Diatoms as biological indicators Diatoms are unicellular or colonial alg ae with siliceous bodies. They are an important basis of wetland food webs; and because they drive many wetland functions through their primary production, they are considered valuable in wetland biological assessment ( Cronk and Fennessy 2001 ; Stevenson 2001 ). The USEPA ( 2002b ) described six fundamental ecosystem functions of algae within water bodies: Providing a food source for organi sms at higher trophic levels Contributing to nutrient a nd biogeochemical cycling Oxygenating the water column Regulating water chemistry Creating habitat for other organisms Acting as physical barriers to erosion Because of their rapid turnover times, algae have a short response time to perturbations including nutri ent and toxic contaminant i nputs; and algae continue production throughout the winter, taking advantage of availa ble nutrients when higher

PAGE 22

7 plants are dormant ( Cronk and Fennessy 2001 ). While the standing stock of algae is typically lower than that of the macrophyte assemblage, al gae can constitute a higher proportion of primary productivity w ithin an aquatic community ( Cronk and Fennessy 2001 ). These factors and others contribute to the utility of the algal assemblage for biological assessment. Among the main a dvantages of using algae for biological assessment include the high diversity within th e algal community (particularly of diatom species) in aquatic environments ( Stevenson 2001 ). There is a depth of knowledge as to the sensitivity of many speci es to different environmental conditions based on their autecological characteristic s, including two published tables of autecological relationships by van Dam et al. ( 1994 ) and Bahls ( 1993 ). Additionally, the rapidresponse time of the algal community to cha nging environmental conditions is a major advantage to their use as biological indicators ( Cronk and Fennessy 2001 ), as well as an overlap in the species present among different aquatic environments ( van Dam et al. 1994 ; Fore and Grafe 2002 ). Diatoms in particular are c onsidered easy to identify based on well-established taxonomic keys of their decay resist ant siliceous structures ( Stevenson et al. 1999 ), and there are well-tested protoc ols for sampling aquatic habitats ( Goldsborough 2001 ). Few significant disadvantages of us ing algae in biological assessment methodologies have been described. Among th em is the necessity of a high-powered microscope for identification ( Doherty et al. 2000 ); although identification is relatively easy, and good taxonomic keys have been established ( Stevenson et al. 1999 ). Additionally, while most algae are not readily motile, wind and current translocation can complicate assessments based on scales of anthropogenic activity in the surrounding

PAGE 23

8 landscape. The third noted disadvantage includes natural seasonal variations in abundance and morphology ( Vymazal and Richardson 1995 ). Overall, algae are considered a valuable assemblage for assessing the biological condition of wetlands. In particular, diat oms are noted as a useful assemblage ( Stevenson 2001 ; Doherty et al. 2000 ). Previous research has correla ted the response of diatoms in streams, lakes, and wetlands, to changes in surrounding land use and to changes in watercolumn characteristics including nutrient loading ( van Dam et al. 1994 ); pH ( Pan and Stevenson 1996 ); heavy metal loading ( Charles et al. 1996 ); and saprobity levels ( LangeBertalot 1979 ). The USEPA ( 2002b ) reported that diatoms are one of the most commonly used assemblages in aquatic ecosys tems for assessing biological, physical, and chemical conditions. Research correlating changes in the diat om community composition to changes in their aquatic environment has been undertaken for isolated freshwater marshes in Florida ( Lane 2003 ); large rivers in Idaho ( Fore and Grafe 2002 ); streams ( Barbour et al. 1999 ; Winter and Duthie 2000 ; Munn et al. 2002 ); depressional wetlands in Michigan ( Pan and Stevenson 1996 ; Stevenson et al. 1999 ); prairie potholes ( Adamus 1996 ); Mid-Atlantic streams ( Pan et al. 1996 ); the Florida Everglades ( Raschke 1993 ); and Florida lakes ( Whitmore 1989 ). Most of the quantitative bi ological indices based on diatom community composition have been constructed for rivers and streams ( Bahls 1993 ; Stevenson and Wang 2001 ). In a study of isolated freshwater marshes in peninsular Florida, Lane ( 2003 ) incorporated fourteen metrics into the Diatom Index of Wetland Condition (DIWC). These included tolerant indicator species, se nsitive indicator species, diatoms requiring

PAGE 24

9 low pH, requiring low salinity, tolerant of hi gh salinity, tolerant of high pH, sensitive to high nitrogen, tolerant of high nitrogen, requiri ng elevated dissolve d oxygen, tolerant of low dissolved oxygen, mesoand polysaprobous diatoms, characteristic of oligotrophic environments, characteristic of eutrophic e nvironments, and pollution-tolerant diatoms. Environmental parameters correlating w ith diatom community composition included specific conductivity, water-column pH, wa ter ammonia-nitrogen concentration, water total Kjeldahl nitrogen (TKN) concentrati on, water total phosphorus concentration (TP), soil pH, and soil TP ( Lane 2003 ). In a study of lotic (relating to moving wa ter) systems in the Mid-Atlantic States, Pan et al. ( 1996 ) found the strongest corre lation with diatom community composition and changes in water-column pH. Additional water-column parameters correlating with diatom community composition included turb idity, aluminum concentration, chlorine concentration, TP, total suspended solids, a nd dissolved organic-carbon concentration. Similarly, in a study of emergent permanen tly flooded floodplain wetlands in western Kentucky, Pan and Stevenson ( 1996 ) found significant correla tions between diatom community composition and 8 water variab les, including alkalinity, conductivity, ammonia-nitrogen concentration, pH, s ilicon concentration, nitrate-nitrogen concentration, chlorine concentration, and TP . Another study of streams in Michigan also correlated the response of diatom co mmunity composition to different land use and water physical and chemical parameters ( Stewart et al. 1999 ). These findings found that the algal assemblage was useful in refl ecting changes in the water environment. Macrophytes as biological indicators Wetland macrophytes are defined as aquati c emergent, submergent, or floating plants growing in or near water ( USEPA 1998 ); and are described as distinguishing

PAGE 25

10 landscape features. The spat ial distribution of macrophytes in the landscape occurs according to a multitude of factors, including hydroperiod, water chemistry, and substrate type, as well as other broader factors such as available seed source and climate. Fennessy et al. ( 2001 ) state that the community compositi on of wetland macrophytes typifies the physical, chemical, and biological wetland dyn amics in time and space. Macrophytes play a vital role in suppor ting the structure and functi on of wetlands by providing food and habitat for other assemblages including al gae, macroinvertebrates, fish, amphibians, reptiles, birds, and mammals. Macrophyte populations can be used as dia gnostic tools to asse ss other aspects of the wetland environment. Crowder and Painter ( 1991 ) state that a lack of macrophytes where they are otherwise expected to grow suggests reduced wildlife populations from lack of food or cover; and/or water quality concerns such as toxic chemical constituents, increased turbidity, or increased salinity. In contrast, an overgr owth of particular macrophytes may signify incr eased nutrient loading ( USEPA 1998 ). Many advantages of studying macrophytes as indicators of wetla nd condition have been noted, including their larg e, obvious size; ease of iden tification, to at least some useful taxonomic level; known response to toxi city tests; and general lack of ability to move to avoid unfavorable conditions ( Danielson 1998a ; Cronk and Fennessy 2001 ). Additionally, macrophytes read ily respond to changes in nutrient, light, toxic contaminant, metal, herbicide, turbidity, water, and salt levels. They can also be sampled in the field with transects, or from the office with aerial photography; and wellestablished field methods of sampling macrophytes exist ( USEPA 2003 ). Furthermore, the USEPA ( 2003 ) states that macrophytes do not requ ire laboratory analysis, can easily

PAGE 26

11 be used for calculating simple abundance me trics, and are superb integrators of environmental condition. In general, macr ophytes represent a useful assemblage for describing wetland condition ( Mack 2001 ). Schindler ( 1987 ) alleged that macrophytes can provide a more integrated picture of wetland function than static measures such as nutrient cycling, productivity, decomposition, or chemical and physical composition. There are however some not ed shortcomings of using macrophytes as biological indicators. These include th e potential delay in response tim e for perennial vegetation, difficulty identifying taxa to the species leve l in certain seasons and for some genera, different herbivory patterns, and varied pest-management practices ( Cronk and Fennessy 2001 ). Despite these limitations, macrophyt es have provided strong signals of anthropogenic influence ( USEPA 2003 ). In fact, many states have begun using macrophytes in their wetland biological a ssessment programs, including Florida ( Lane 2003 ), Minnesota ( Galatowitsch et al. 1999a ; Gernes and Helgen 1999 ), Montana ( Apfelbeck 2000 ), North Dakota ( Mushet et al. 2002 ), and Ohio ( Mack 2001 ). Previous biological assessment studies have included unique and varied macrophyte metrics dependent on we tland type and bioregion. Lane ( 2003 ) calculated 5 macrophyte metrics for inclusion in the marsh Vegetative Index of Wetland Condition (VIWC). The 5 core metrics of the VIWC in cluded tolerant indicator species, sensitive indicator species, exotic species, annual to perennial ratio, and average Coefficient of Conservatism score. In Minne sota, Vegetative Indices of Bi otic Integrity (V-IBIs) have been created for 8 wetland types ( Galatowitsch et al. 1999a ). Macrophyte metrics varied depending on wetland type, and included 15 metrics for high-order river floodplain wetlands, 12 for low-order river floodplain wetlands, 8 for midorder river floodplain

PAGE 27

12 wetlands, 7 for calcareous littoral wetlands, 6 for noncalcareous littoral wetlands, 7 for wet prairie-sedge meadows, 4 for forest glaci al marshes, and 1 single metric for prairie glacial marshes. Another comprehensive biological assessment used to construct multimetric indices of biotic integrit y for Ohio wetlands was designed by Mack et al. ( 2000 ). Separate biological multimetric indices were developed for emergent, forested, and shrub wetlands. Twelve metr ics were incorporated, including Carex species, dicot species, shrub species, hydrophyte species , Rosaceae species, Floristic Quality Assessment Index, tolerant species, intole rant species, invasive graminoids, shrub density, small-tree density, and maximum importance value. The Floristic Quality Assessment Index (FQI) has been included in many of the multimetric indices created for the macrophyt e assemblage. The concept of FQI was developed by Wilhelm and Ladd ( 1988 ) for vegetation around Chicago, Illinois. This method of scoring plant species based on e xpert botanist opinion has been used in Michigan ( Herman et al. 1997 ), Ohio ( Andreas and Lichvar 1995 ; Fennessy et al. 1998 ; Mack 2001 ), Ontario ( Francis et al. 2000 ), North Dakota ( Mushet et al. 2002 ), and Florida ( Lane 2003 ; Cohen et al. 2004 ). The FQI provides a quantitative means of assessing the fidelity of a pl ant to a particular environmen t through the Delphi technique ( Kent 2000 ), where individual botanists assign coefficients to each species, and then reevaluate their scores based on the group mean scores. This technique assumes that the collective decision by a group of expert botanists is more accurate than the professional judgment of one individual ( Kent 2000 ). Macroinvertebrates as biological indicators Biological assessment based on the macroinve rtebrate assemblage has been widely applied for indications of environmental qua lity, and often more specifically water

PAGE 28

13 quality ( Lenat 1993 ; Cummins and Merritt 2001 ). Invertebrates are participants in many fundamental ecological processes, includi ng the breakdown of or ganic matter and recycling of nutrients ; and invertebrates are a vital component of the food web, making up a large portion of the diets of other orga nisms (such as fish, amphibians, and birds) ( Cummins and Merritt 2001 ; Helgen 2001 ). As such, Voshell ( 2002 ) recognizes that freshwater invertebrates have been used more often than any other group of organisms for assessing freshwater ecosystems. A great deal is known about the specific eco logy of lotic (relating to moving waters such as streams) macroinvertebrates; less is known about those found primarily in lentic (relating to still waters such as wetlands) environments. Williams and Feltmate ( 1992 ) noted that, while not well studied, the commun ities of aquatic insects in wetlands include species from most of the major aquatic groups. The community composition of wetland macroinvertebrates differs from that of fl owing waters, because of differences in substrate, dissolved-oxygen level in the water column, hydroperiod, and annual water fluctuations. Macroinvertebrat es have been useful indicato rs of environmental condition in streams; and Karr and Chu ( 1997 ) speculate macroinvertebrates also may be appropriate indicators of envir onmental integrity in wetlands. Since 1997, the use of the macroinvertebrate assemblage for biological assessments has been initiated in 48 states for lakes and streams ( Karr and Chu 1999 ). Macroinvertebrate-based we tland biological assessment methodologies have been initiated in many states, including Florida ( Lane 2003 ), Minnesota, Montana, North Dakota, and Ohio ( Danielson 1998b ). Within the state of Florida and throughout the southeastern Coastal Plain, ecological resear ch on the macroinvertebrate community has

PAGE 29

14 included many ecosystem types from isolat ed marshes of peninsular Florida ( Kushlan 1990 ; Lane 2003 ); isolated wetlands in south Florida ( Stansly et al. 1997 ); amphipods in southeastern wetlands ( Pickard and Benke 1996 ); nontidal wetlands ( Batzer and Wissinger 1996 ); sloughs of the northern Everglades ( Rader and Richardson 1992 ; Rader and Richardson 1994 ); floating islands in Orange Lake in north central Florida ( Haag et al. 1987 ); and bottomland hardwood swamps ( Wharton et al. 1982 ). Doherty et al. ( 2000 ) conclude that the stru cture and function of the macroinvertebrate community accurately refl ects the biological condition of a wetland, and that the macroinvertebrate community co mposition changes in pr edictable ways with increased human influence. Because wetland m acroinvertebrates complete part or all of their lives in the wetland, they are directly exposed to condi tions in the wetland water and soils ( Merritt and Cummins 1996 ; Helgen 2001 ). Also, because of the short length of their life cycles (compared to most macrophytes and vertebrates), Stansly et al. ( 1997 ) noted that macroinvertebrates respond quickly to changes in the physical, chemical, or biological parameters of their host environm ent. Their quick response time, reliance on water (both for the water quality and durati on of inundation), and ease of collection make macroinvertebrates a favorable assemblage for use as biological indicators. Noted disadvantages to using macroinvertebrate s include the amount of time and knowledge necessary for identification to lower taxonomic levels ( Cummins and Merritt 2001 ). The Florida Department of Environmenta l Protection (FDEP) has initiated the development of biological indices based on the macroinvertebrate assemblage for freshwater bodies in Florida. Macroinvert ebrate-based biological indices have been created for isolated marshes through the Ma croinvertebrate Index of Wetland Condition

PAGE 30

15 (MIWC; Lane 2003 ), for Florida streams through th e Stream Condition Index (SCI; Barbour et al. 1996a ; Fore 2003 ), for surface waters in south Florida canals using SCI protocol ( Snyder et al. 1998 ), for the evaluation of restor ation in the Kissimmee River Basin ( Merritt et al. 1996 ), and in freshwater lakes th rough the Lake Condition Index (LCI; Gerristen and White 1997 ). Different core metrics comprise each multimetric biological index. Lane ( 2003 ) incorporated 5 core metrics as biological indica tors of wetland condition for isolated marshes in the MIWC , including sensitive taxa, tolerant taxa, predators, Odonata, and Orthocladiinae. Th e SCI was developed with 7 core metrics, including taxa richness, EPT richness (Epheme roptera, Plecoptera, Trichoptera), Florida Index, percent dominant, Chironomid taxa richness, suspension-filter feeders, and Diptera ( Barbour et al. 1996b ). Similarly, the LCI incorporated 7 metrics, including taxa richness, Shannon diversity, Hulbert index, ETO taxa (Ephemeroptera, Trichoptera, and Odonata), percent dominance, filt er feeders, and gatherers ( Gerristen and White 1997 ). Numerous studies have documented the re sponse of the benthic macroinvertebrate community to anthropogenic activities. Two primary areas of research include changes in trophic state, and additions of stormwater and wastewater. Gerristen and White ( 1997 ) and Cairns and Pratt ( 1993 ) found that the benthic macroinvertebrate community composition responded to changes in trophic status. In the northern Everglades, Rader and Richardson ( 1994 ) found that macroinvertebrates re sponded to nutrient enrichment with a greater number of Coleopteran species present (especially those in the families Hydrophilidae and Dystcidae) in nutrient-enri ched and intermediate areas than in nonenriched areas. With shifts in trophic stat us, the structure of other assemblages also

PAGE 31

16 changed, affecting the benthic macroinverteb rate community composition. For example, Adamus and Brandt ( 1990 ) found that shading from dense stands of emergent vegetation altered the distribution among functional f eeding groups by limiting the production of benthic algae (thus favoring detritovores over grazers). De Szalay and Resh ( 1996 ) similarly found that increased sh ading caused fine particulate organic matter to settle out, making rich detritus accessible to support a large population of benthic macroinvertebrate detritovores. While adding stormwater and wastewater a lters the natural hydr ology of an isolated wetland, it also increases the inflow of nutrients, sediments, and toxic metals. Harris and Vickers ( 1984 ) found that adding wastewater to cypress domes shifted the macroinvertebrate community toward a less-com plex trophic structure. Similarly, when wastewater was directed in to Florida cypress domes, Lemna spp. (duckweed) mats covered the water surface, bl ocking sunlight from the water column, and creating anoxic conditions ( Dierberg and Brezonik 1984 ). This reduced the di versity and biomass of benthic invertebrates, leaving only a few pollution-tolerant organisms ( Brightman 1984 ). In Florida streams, Barbour et al. ( 1996b ) found that the occu rrence of tubificid oligochaetes increased with organic enrichment. Other studies have focused on the effects of adding stormwater to freshwater wetlands. Freshwater marshes (in Savannas Preserve State Park, Florida) receiving stormwater additions showed increase d phosphorus levels, lo wered oxygen levels, increased water-column pH and hardness, and a change in the macroinvertebrate community toward pollution-tolerant species and those intolerant of the typical acidic and oligotrophic environment ( Graves et al. 1998 ). Barbour et al. ( 1996b ) reported that some

PAGE 32

17 chironomids of the family Orthocladi inae, including those in the genus Cricotopus , were found to be tolerant of me tal pollution; while other Orthocladiinae, including Rheocricotopus spp. and Corynoneura spp., were thought to be sensitive to metal pollution in Florida streams. Adding wastewater or stormwater also alters the water level and hydroperiod of cypress domes. Toth ( 1993 ) reported a response of the m acroinvertebrate community to water-level manipulation in a Kissimmee River demonstration project. Hydroperiod had a distinct influence on community compositi on; as some macroinvertebrates either temporarily relocated or cope d with behavioral and biologi cal adaptations to changing water conditions. Macroinvertebrates w ith adaptations to wetland hydroperiods demonstrate both behavioral and physiological adaptations for dr aw-down conditions. For example, in south Florida hydric flatwoods, Gore et al. ( 1998 ) found that Crangonyx spp. and several other aquatic insects burrowed into moist se diments to avoid desiccation. Some macroinvertebrates are thought to be indicative of water level and seasonality, with Caenis spp., Anaz spp., Libellula spp., and Pantala spp. indicative of persistent water; some Chironomus, some Tanytarsus, Beardius spp., and Zavreliella marmorata, indicative of permanent standing water; and Ablabesmyia rhamphe grp., Krenopelopia spp., and Tanytarsus sp. g. indicative of ephemeral wetlands ( Doherty et al. 2000 ). Stansly et al. ( 1997 ) concluded that in isolated wetlands of south Florida, the presence of macroinvertebrates with long life cycles or predatory be havior may indicate hydroperiod stability. Snyder et al. ( 1998 ) found that macroinvertebrates with comparatively short life cycles that are capab le of rapid colonization were typical of

PAGE 33

18 canals surrounded by urban land uses, whereas th e occurrence of macr oinvertebrates with longer life cycles were more common in can als surrounded by more natural landscapes. Review of Isolated Freshwater Forested Wetlands Throughout the world, wetlands have been categorized in many different ways ( Keddy 2000 ; Kent 2000 ; Mitsch and Gosselink 1993 ). Probably the most widely applied classification system in North America is that by Cowardin et al. ( 1979 ). Our study focused on what Cowardin et al. ( 1979 ) categorized forested palu strine wetlands. More specifically, the wetlands targeted in our study are called pon dcypress domes with reference to Vernon ( 1947 ), who was the first to name these systems after their characteristic silhouette in the landscape. Cypress domes range in size from less than 1 hectare to more than 10 hectares ( Wharton et al. 1977 ). There is however much application of ecological theories and resear ch from all types of wetlands worldwide, as wetland species share mechanisms to deal with a fluctuating environment by adapting to periodically inundated and often anaerobic conditions. Figure 1-1 is a systems diagram of the pr imary components, sources, and flows of a typical cypress dome. Symbols and terminology are from Odum ( 1994 ). Appendix A provides an overview of the energy circuit language and symbols used in Figure 1-1. Inflows into the cypress dome are limited to sunlight, wi nd, water (rain, surface run-off, groundwater), and recruitment of plant and anim al species. Water inflow comes almost entirely from rainwater, both as a direct i nput and as “run-off” from a relatively small watershed, as such these wetlands are often termed “isolated” due to their somewhat limited hydrologic connections. Standing water is present in most cypress domes much of the year ( Odum 1978 ; Mitsch and Gosselink 1993 ), and some cypress domes ha ve deep central pools staying

PAGE 34

19 DN OM Soil Animal Migration Wind Sunlight ET Seed SourcePondcypress dome Wetland Rain N Ground Water Runoff N DN DOM Water Overland Flow FireI Biomass Biomass Biomass Species Species Species Woody Vegetation Herbaceous Vegetation Struc ture Biomass Biomass Species Species Other Consumers Macroinvertebrates Algae Figure 1-1. Systems diagram showing major sources, storages, and flows of a cypress dome. wet year round, while others go dry annuall y. There is also variation in maximum flooding depth and length of sta nding water between years, whic h reflects the larger scale climatic and physiogeographic influences to cypress domes. Typically, the wettest period is summer and the driest spring and fall ( Mitsch and Gosselink 1993 ). Taxodium ascendens (pondcypress) is the principal tree species in cypress domes ( Devall 1998 ). Other tree species associated with Taxodium ascendens include: Nyssa biflora (black gum), Pinus spp. (many southern pines), Acer rubrum (red maple), and Magnolia virginiana (sweetbay) ( Wilhite and Toliver 1990 ; Devall 1998 ). Pomdcypress trees characteristically dominate the center, with the pomdcypre ss along the edge in competition with other species that are less tolerant of flooded conditions. There is a greater likelihood of fire and a larger nu mber of seedlings in the drier edges ( Odum 1978 ).

PAGE 35

20 The isolated cypress dome is characterized by a tolerance of lo w nutrient levels and intermittent fire ( Brandt and Ewel 1989 ), with major system i nputs limited to rainfall and surface inflows ( Mitsch and Gosselink 1993 ). These ecosystems are considered successionally stable, but may be replaced by other ecosystems with changing environmental conditions, such as decreased water levels ( Devall 1998 ). In drained cypress domes in northern Florida, Marois and Ewel ( 1983 ) found an increase in the densities of hardwood and shrub species. Pondcypress and the typical co-dominate black gum are deciduous, shedding their leaves from October to December. Ewel ( 1984 ) and Wharton et al. ( 1977 ) found that the perpetuation of cypress domes depends on fl uctuating water levels, with a dry period without standing water necessary for cypre ss generation and higher water levels some time during the year necessary to prevent germ ination of more terrestrial faster-growing pines and hardwoods that are not tolerant of standing wa ter. Altering the typical hydroperiod of a cypress dome would eff ect species composit ion, resulting in encroachment of terrestrial species in a dr ained cypress dome and a lack of regeneration in an artificially flooded dome. Anthropogenic activities in the surroundi ng up-slope landscape can create a wide array of changes to the inflows of these otherwise isolated sy stems. Figure 1-2 presents a systems diagram of the primary component s, sources, and flows of a cypress dome embedded in a developed landscape with va riable land uses. The systems boundary reflects a 100 m buffer zone around the isol ated wetland. Some of the potential alterations to pondcypress wetlands located with in developed land uses include changes

PAGE 36

21 Upland Wetland Goods Sunlight FuelsWetland in 100 m Buffer Biomass Land Area Land Area Land Area Assets Wastes Water Inflows Drainage Imputs Biomass T o xi n sNutrients Figure 1-2. Aggregate systems diagram of a cypress dome embedded within a developed landscape. The systems boundary is a 100 meter buffer around the delineated wetland edge, which coul d be used for the Landscape Development Intensity (LDI) index calculation. in the seasonality and depth of flooding, increas ed nutrient inputs, increased toxin inputs, and physical impacts (canals, retaining walls, stormwater box culverts, etc.). Changes in Hydrology Figure 1-3 is a systems diagram represen ting potential hydrologi c alterations to a pondcypress ecosystem surrounded by developed land uses. Two important mechanisms of the developed landscape are highlighted. Firs t, increased run-off is considered a factor of the amount of increased impervious surf ace in the watershed supporting the isolated pondcypress wetland and the amount of rainfall. This would be particularly apparent in an urban landscape, where previously vegeta ted lands are paved creating increased water flow during rain events, which might otherwis e have been intercep ted by the vegetation.

PAGE 37

22 Herbs Sun Wind ET Water Soil DOM Algae Rain Surface Runoff Ground water Impervious Surface Macroinvertebrates Wetland with Hydrologic AlterationsDrainage Woody Plants Figure 1-3. Mechanism of altered hydrology of a wetland in a developed landscape. Bold lines represent examples of potential hydrological alterations. A second mechanism of the deve loped landscape is an increased outflow of water storage from the wetland resulting from drainage. In Florida, mature cypress trees are consid ered the most flood tolerant of all tree species ( Harms et al. 1980 ; Ewel 1990 ). Past research shows that cypress trees can survive sustained deep flooding ( Lugo and Brown 1986 ; Young et al. 1995 ), but they also found decreased growth rates and no evidence of regeneration, suggesting that while mature cypress ecosystems may be able to wi thstand some threshol d level of long-term flooding, regeneration may be impeded which is otherwise necessary to ensure the longterm survival of the ecosystem. Ultimately, removing the structure of the wetland will predictably alter other ecos ystem components. For example, by removing the tree

PAGE 38

23 canopy, algae may flourish at the water surf ace where it would have otherwise been shaded out. This would in term alter th e food available for the fauna and decrease available sunlight at the soil surface. Ewel ( 1990 ) reported that increasing the length of flooding would also affect soil aeration and the ability of other plants to survive and reproduce. Lugo and Brown ( 1986 ) looked at the response of floodplain tree species to sustained increases in water depth after the da mming of the Ocklawaha River in Florida. They found that while the larg er trees survived some dept hs of flooding, in the deepest areas, where mean water depth was 1 m, ther e was 100% tree mortalit y within 5 years of flooding. In the control system, tree mortalit y was less than 1% pe r year. Additionally, flooded trees responded with a dieback of term inal branches, loss of leaves, reduction in leaf size, and loss of color brightness in leaves ( Lugo and Brown 1986 ). Among the trees in the Ocklawaha River floodplain, 25-53 cm was the threshold flooding depth, beyond which tree mortality sharply in creased. In a different study, Young et al. ( 1995 ) found that the annual radial growth of Taxodium distichum (baldcypress) significantly increased for 4 years after flooding, followed by declining growth in the subsequent 16 years. The researchers offered two potential explanations for the initial growth increases: decreased competition due to the death of less flood tole rant species, or increased nutrient levels immediately following flooding. Marois and Ewel ( 1983 ) studied the effects of d itches and berms on 15 cypress domes situated within an in tensively managed slash pine plantation. In the cypress domes not ditched and bermed, the lengths of flooding and mean water depth were generally greater. Alternatively in the dr ier ditched and bermed cypress domes the

PAGE 39

24 density of hardwoods, shrubs, and vines incr eased. They conclude d that while cypress tree growth increased in the years directly following the drying of the domes, cypress regeneration might be inhibited due to ch anges in vegetative species composition, soil chemistry, and hydrology. Cypress seeds require soaking in water in order to germinate ( Demaree 1932 ), so altering the seasonality and decreasing the depth of flooding in a cypress dome may inhibit or seriously diminish potential germ ination. The higher density of hardwoods, shrubs, and vines may also inhibit cypre ss regeneration by blocking sunlight from reaching the forest floor. Marois and Ewel ( 1983 ) found the highest percentage of light transmittance in unaltered domes. The cypress domes with greater light also had an abundance of grasses and sedges. Ewel ( 1990 ) noted that drainage allows species with low flood tolerance to become established, resul ting in an increased density of shrubs and hardwoods, poor cypress regeneration, increase d fire potential, and a dramatic shift to arboreal species from a quatic and wading fauna ( Marois and Ewel 1983 ; Harris and Vickers 1984 ). More specifically, Marois and Ewel ( 1983 ) found broadleaved predominantly evergreen mid-story plants (such as Ilex cassine , Lyonia lucida , Magnolia virginiana , and Persea palustris ), became more common in swamps when water levels were lowered. Harris and Vickers ( 1984 ) speculated that shifting species in the vegetation layer equates to alte red structure and habitat for fauna which affects organisms in all other trophic levels. Decreasing the mean water level could cause changes in the community composition of many species that rely on c ypress domes for regeneration. Benthic invertebrates may not withstand increased dr y periods, and reproduction may be difficult

PAGE 40

25 or unattainable. Often forming the base of swamp food chains ( Ewel 1990 ), eliminating or decreasing the population of benthic macr oinvertebrates could have repercussions throughout the food chain. Fish, amphibians, and reptiles may be eliminated from cypress domes with decreased hydroperiods due to reproduction diffi culties or altered food availability ( Means et al. 1998 ; Ewel 1990 ). Increased Inflows of Nutrients and/or Toxins In undeveloped landscapes cypress domes receive limited nutrient inputs from rainwater and surface water run-off ( Wharton et al. 1977 ). Figure 1-4 is a systems diagram of a pondcypress dome receiving increased nutrients and/or toxins. The inflow of nutrients and toxins may come from both point (stormwa ter and wastewater additions) and non-point (run-off) sources. An increase in run-off from impervious surfaces in the surrounding landscape may carry an increased loading of nutrients and toxins ( Harper 1994 ). Surface run-off carrying fertilizer used on agricultural crops or home lawns are examples of non-point source cont ributions. Figure 1-4 shows th at as nutrients flow into the wetland the growth of living biomass incr eases, and that nutrients accumulate in the water and soil organic matter storages. Conversely, as toxins flow into and accumulate in the wetland, there is a dele terious effect on biomass. Two nutrients of primary importance in pondcypress domes are phosphorus and nitrogen. These are repres ented in the grouped “N” nutri ent pool sources and storage tanks in Figures 1-4. Phosphorus, an elemen t critical to plant growth, is mostly bound into forms unavailable to plan ts at pH levels below 5.7 ( Brady and Weil 2004 ), higher than the average pH of cypress domes embedded in undeveloped landscapes ( Coultas and Duever 1984 ) Phosphorus is known to accumulate in the clay layers found beneath cypress domes, which makes cypress dome eco systems dependent on a constant input of

PAGE 41

26 Herbs Sun Wind ET Water Soil DOM Algae Rain Surface Runoff Ground water Macroinvertebrates Wetland with Nutrient and/or Toxin Loading Woody Plants N N N StormWater N WasteWater N Other Dispersed Sources N T T T T T T NT Figure 1-4. Increased nutrient s and/or toxin inflows into a wetland from the surrounding developed landscape. The character “N ” represents a pooled nutrient tank, and “T” represents a pools toxins tank. available phosphorus from rainfall. Raisi ng the pH of the wetla nd should increase the concentration of available phosphorus. In cont rast, nitrogen does not accumulate in the clay layer or organic sediments at the botto m of cypress domes due to denitrification processes, and the rate of the nitrogen cycle seems dependent on the cycling of decomposition of organic matter ( Wharton et al. 1977 ). Previous studies in cypress domes show that cypress trees respond to increased nutrient loading with incr eased tree growth rates ( Nessel et al.1982 ; Lemlich and Ewel 1984 ). Nessel et al. ( 1982 ) measured phosphorus concentrat ions in live cy press needles at a cypress dome embedded in silvicultura l land use and a cypress strand receiving

PAGE 42

27 sewage for more than 40 years. Cypress n eedles in the silvicultural wetland had a lower average phosphorus concentr ation than cypress needles in the wetland receiving wastewater. Additionally, th e top 20 cm of sediments in the wetland receiving wastewater had nearly 5.5 times as much phosphorus per m2 as in the silvicultural wetland. They concluded that the trees in th e cypress strand were in fact responding to the increased nutrient inputs (from the se wage) with increased growth rates. Historically cypress domes were oligotrophic systems ( Mitsch and Gosselink 1993 ). Ewel ( 1984 ) and Harris and Vickers ( 1984 ) found that an increase in dissolved nutrients led to the development of thick mats of Lemna spp., Spirodela spp., and/or Azolla spp. on the water surface. Ewel ( 1984 ) also noted that a nutrient enriched cypress dome had similar understory species com position compared to a cypress dome not receiving wastewater, howeve r the leaf area was significan tly higher in the nutrient enriched cypress dome. Changes in the unde rstory vegetation (fro m increased leaf area or covering of the water surface with a layer of vegetation) can have an effect on other trophic levels within the ecosystem. In c ypress domes receiving wastewater additions, Harris and Vickers ( 1984 ) reported an increase in num bers of invertebrates and amphibians, however they also noted a shift in the invertebrate taxa and a high larval mortality of amphibians suggesting the fauna in the nutrient enriched dome were different from the control wetland. Physical Disturbance Figure 1-5 is a systems diagram of th e physical influences to a pondcypress wetland in a developed landscape. Examples of physical changes include the trampling and grazing of domestic cattle, rooting of feral pigs, barriers of roads and retention walls,

PAGE 43

28 Herbs Sun Wind ET Water Soil DOM Algae Rain Surface Runoff Ground water Impervious Surface Macroinvertebrates Wetland with Physical AlterationsDrainage Woody Plants Trampling Grazing Rooting Culvert Conduit Barrier Figure 1-5. Potential physical alterations to a pondcypress wetland. Bold lines highlight physical alterations such as imperv ious surface, culverts, drainage mechanisms, barriers, and trampli ng, grazing, and rooting by animals. and conduits such as stormwater culverts . The trampling, grazing, and rooting by animals is depicted as a drain on the bi omass of herbaceous and woody plants and macroinvertebrates. Barriers to water fl ow and impoundments are represented through a switch operation, showing that wh en a barrier is constructed it acts as a control over the level of water in the storage tank. Additio nally, conduits act in two opposing ways by either increasing the flow of water into the system or by helping to drain water from the system, depending on the construction design. Another important component is the increased flow of surface run-off from impe rvious surface outside of the wetland system boundary.

PAGE 44

29 Little research has been ab le to quantify the effects of physical modifications to wetlands. Findlay and Houlahan ( 1997 ) used existing biological surveys for 30 Ontario, Canada, wetlands, comparing the species ric hness of plants, amphibians, birds, and reptiles with wetland areas, density of paved ro ads, and percent forest cover. They found a negative correlation between wetland species richness and density of paved roads on lands within 2 km of the we tland. They concluded that in creasing the density of paved road surface or decreasing the forest cover by 20% with in 2 km surrounding a wetland would pose significant risks to the biodiversity of the wetland and be as detrimental as losing 50% of the wetland itself, in terms of loss of species richness. Another physical modificati on to cypress domes is removing a portion of the canopy layer. Florida has a long hi story of timber harvesting, and Ewel ( 1990 ) suggested that nearly all of the cypress domes in nor th Florida have been logged since the late 1800s. Studies showed that logged cypr ess domes maintained their defining characteristics after regeneration ( Terwilliger and Ewel 1986 ; Ewel et al. 1989 ); however, during regeneration, there were shifts in the fl ora and fauna of logged wetlands. Physical modifications such as roads, canals, and stor mwater culverts also act as direct conduits for the introduction of exotic species ( Frappier and Eckert 2003 ). Quantifying Anthropogenic Influence Wetlands occupy a large portion of the Flor ida landscape. An estimate from the 1780s reported 8,225,000 ha of wetlands in Florida ( Dahl 2000 ). By the mid-1980s, the National Wetlands Inventory estimated Fl orida had 4,467,000 ha of wetlands remaining, translating into a loss in Florida of 46 % of the pre-1780s wetland area ( Dahl 2000 ; Mitsch and Gosselink 1993 ). Throughout the continental United States, similar trends were apparent, with a drastic declin e in the surface area of wetlands.

PAGE 45

30 Dahl ( 2000 ) reported that 98 % of all wetland losses throughout the continental United States from 1986 to 1997 were losses to freshwater wetlands. Of the remaining freshwater wetlands, 40% of those wetlands sa mpled were adjacent to agricultural lands and therefore potentially affected by land use practices such as herbicide and pesticide application, irrigation, livestock watering a nd wastes, soil erosion, and deposition. An additional 17% of the remaining wetlands were adjacent to urban or rural development. Freshwater non-tidal wetlands experienced the greatest development pressure just inland from coastlines as the demand for housing, tr ansportation infrastructure, and commercial and recreational facilities increased ( Dahl 2000 ). These changes in land use are proportionally more widespread in Florida than much of the continental United States due to the remarkable length of coastline along bot h the Atlantic Ocean and Gulf of Mexico coasts of Florida. Spanning the populated coasts from Jacksonville to Miami on the east coast and from Naples to Tampa along the west coast, most coastal counties are reported to have high wetland loss of non-tidal freshwater wetlands from 1986 to 1997 ( Dahl 2000 ). Dahl ( 2000 ) suggested that many of these wetla nds were harveste d and returned as shrub wetlands. Anthropogenic activities can influence an array of changes in surrounding ecosystems. There have been numerous atte mpts at quantifying anthropogenic influence based on varying scales. Three primary in dices of anthropogenic influences were incorporated throughout our study to co mpare wetland condition, including the Landscape Development Intensity (LDI) index ( Brown and Vivas 2004 ), the Wetland Rapid Assessment Procedure (WRAP; Miller and Boyd 1999 ), and the Minnesota disturbance index ( Gernes and Helgen 1999 ).

PAGE 46

31 Landscape Development Intensity Index The Landscape Development Intensity (LDI) index can be used as an index of human activity based on a development intens ity measure derived from nonrenewable energy use in the surrounding landscape. Th e underlying concept behind calculating the LDI (quantifying the nonrenewable energy use per unit area in the surrounding landscape expressed in emergy terms) stems from earlier works by Odum ( 1995 ), who pioneered emergy analysis for environmental accounting. [Emergy is an established environmental accounting term referring to expressi ng energy use in solar equivalents ( Odum 1995 ).] Brown and Vivas ( 2004 ) suggest that landscape conditi on is strongly related to the surrounding intensity of human activity, and th at ecological commun ities are affected by the direct, secondary, and cumulative impacts of activities in the surrounding landscape. The LDI scale encompasses a gradient from completely natural to highly developed land use intensity. More intense activities such as highways and multi-family residential land uses receive higher LDI scores. Natural lands capes such as wetlands, lakes, and upland forests receive a 1.0, the lowest possible LD I score, based on no use of nonrenewable energy in these ecosystems. The LDI is calculated based on the percent of the area in a particular land use times the Landscape Development Coefficient (L DC), which is defined by the amount of nonrenewable energy use. The LDC coeffici ent does not account for any individual causal agent directly, but instead, may repres ent the combined actions of air and water pollutants, physical damage, changes in the suite of envi ronmental conditions (groundwater levels, increased flooding) or a co mbination of such factors, all of which enter the natural ecological system from the surrounding developed landscape.

PAGE 47

32 Wetland Rapid Assessment Procedure The Wetland Rapid Assessment Procedure (WRAP) attempts to provide accurate and consistent evaluations of wetland sites, and relied on an evaluator with an adequate understanding of the functions of and sp ecies found throughout Florida ecosystems ( Miller and Gunsalus 1997 ). WRAP consists of a qualitative score describing the functional capacity of a wetland. Scores rang ed from 0.0 to 3.0, in 0.5 increments. The 6 scoring categories for WRAP include: 1) Wild life utilization; 2) Overstory/shrub canopy; 3) Vegetative ground cover; 4) Adjacent upland support/buffer; 5) Field indicators of wetland hydrology; and 6) Water quality input and treatment. A score of 3.0 indicates an “intact” wetland, whereas a score of 0.0 indi cates a wetland with a reduced functional capacity ( Miller and Boyd 1999 ). Minnesota Disturbance Index The Minnesota disturbance index is consider ed a gradient of hu man disturbance, or a measure of land use disturbance based on investigator knowle dge, observations, and best professional judgment about the de gree of influence to the ecosystem. Gernes and Helgen ( 1999 ) used the Minnesota disturbance inde x as a baseline for creating an index of vegetative biotic in tegrity for depressional wetlands. There are 2 primary categories and 3 secondary categories used to calculate the Minnesota disturbanc e index score. The primary categories include stormwater and agri cultural influence, and are weighted twice as high as the secondary categories. Wetla nds receive scores assigned according to significantly affected (S = 8), m oderately affected (M = 4), le ast affected (L = 2), and not applicable (NA = 0) depending on the scorers opinion as to the degree of influence. Wetlands only receive a score in 1 of the primary categories, and reference wetlands receive a score of 0 in both primary categor ies. The 3 secondary categories include

PAGE 48

33 hydrologic/miscellaneous influence, historical influence, and buffer, receiving scores of significantly affected (S = 3), m oderately affected (M = 2), le ast affected (L = 1), and not applicable (NA = 0). Wetlands can receive scores in all of the secondary categories, with possible scores ranging from 0 to 17. Plan of Study Physical and chemical environmental pa rameters and the community composition of diatoms, macrophytes, and macroinvertebra tes were sampled in isolated forested wetlands throughout Florida to answer the over all question: can the changes in biotic components of pondcypress wetlands (such as the community composition of the diatom, macrophyte, and macroinvertebrate assemblages) be related to changes in development intensity in the landscape immediately adja cent to and surrounding them. Wetland study sites were sought in various landscape settings that include d natural, agricultural, and urban land uses. Three independent measures of anthropogenic influence were calculated for each wetland including LDI, WRAP, and the Minnesota disturbance index. Compositional differences among the diat om, macrophyte, and macroinvertebrate assemblages were identified and related to th e 3 measures of anthropogenic influence. Each assemblage was used to construct th e Wetland Condition Index (WCI) for isolated forested wetlands in Florida.

PAGE 49

34 CHAPTER 2 METHODS Biological, physical, and chemical para meters were sampled in 118 forested wetlands less than 2 ha in size. This chap ter describes site sel ection, calculations of landscape development intensity, field-data collection, and laboratory analyses. Statistical analyses are described for each a ssemblage and for the creation of the Wetland Condition Index (WCI). Site Selection Field research spanned two growing s easons with 72 wetlands sampled between May-September in 2001 and an additional 46 wetlands sampled between May-October in 2002. Figure 2-1 shows the location of the 118 sample wetlands indicated by generalized a priori land use categories (reference, agri cultural, urban). Hereafter wetlands embedded in primarily undeveloped landscap es were called reference; wetlands embedded in primarily agricultural land uses were called agricult ural wetlands; and wetlands embedded in primarily urban la nd uses were called urban wetlands. Random site selection was not feasible gi ven the necessity of obtaining permission to access private lands and the non-random pattern of land development in Florida. Site selection for agricultural wetlands was acco mplished with the aide of the Natural Resources Conservation Service under the Unit ed States Department of Agriculture and University of Florida Institute of Food and Agricultural Sciences extension agents. Sample wetlands were targeted spatially th roughout Florida, so that a nearly equal distribution of wetlands was sampled within e ach of the 4 Florida ecoregion (panhandle

PAGE 50

35 # S # S % %Ê Ú Ê Ú Ê Ú# S # SÊ Ú%Ê Ú Ê Ú Ê Ú Ê Ú Ê Ú# S # SÊ Ú% % # S # S %Ê Ú% %Ê Ú# S # S # SÊ Ú Ê Ú Ê Ú# S % % %Ê Ú% # S % # SÊ Ú% # SÊ Ú# S %Ê Ú Ê Ú# S % # S # S # SÊ Ú Ê Ú Ê Ú# S % # S %Ê Ú% % # SÊ Ú# S # S % # S % # S %Ê Ú% %Ê Ú% %Ê Ú Ê Ú%Ê Ú# S # SÊ Ú Ê Ú Ê Ú% % # S # S # S % # S # S % % %Ê Ú% % % % # S # SÊ Ú# S # SÊ Ú Ê Ú Ê Ú# SÊ Ú Ê Ú Ê Ú 0200400KilometersFlorida Ecoregions Panhandle North Central South Sample Wetlands (n=118)# SReference%AgriculturalÊ ÚUrban N Figure 2-1. Study site location of 118 isolated forested wetlands in Florida. The state of Florida was separated in to four ecoregions ( Lane 2000 ). Sample wetlands were designated by a priori surrounding land use categ ories: o reference, agricultural, or urban. n=28; north n=31; central n=31; south n=28). Boundaries of the Florida ecoregions were determined with a hydrologic model by Lane ( 2000 ). Florida freshwater palustrine wetlands were classified using a hierarch ical classification t echnique, and physical (surficial geology, soils, digi tal elevation model, slope) and climatic (precipitation, potential evapotranspiration, runoff, annual da ys of freezing) variables were tested for correlation with wetland clusters. Final ecoregion boundaries were based on a spatial water balance model. The number of wetlands sampled per a priori land use category per region varied, with 28 wetlands in the south (n = 9 referen ce, n = 9 agricultural, n = 10 urban), 31 in the

PAGE 51

36 central (n = 11 reference, n = 9 agricultura l, n = 11 urban), 31 in the north (n = 9 reference, n = 12 agricultural, n = 10 urban) , and 28 in the panhandle (n = 8 reference, n = 10 agricultural, n = 10 urban) ecoregions . All 72 wetlands sampled in 2001 and 39 (of the 46) wetlands sampled in 2002 were dominated by Taxodium ascendens , and were considered pond-cypress domes. The re maining 7 wetlands sampled in 2002 had a canopy layer comprised of a mixture of species that was not dominated by Taxodium ascendens alone. These wetlands exhibited the characteristic depressional shape of cypress domes in the landscape and from GIS based aerial photography and were included in the sampling pool as potential variants of cypress domes. Wetlands surrounded by natural landscapes were generally located on c onservation lands including state and national parks and forests, count y and city lands, and private conservation tracts. Wetlands currently surrounded by cattle pasture, row crops, citrus, and silvicultural land uses were included in the agricultural a priori land use category. Urban wetlands located in an urban land use matrix for the longest period of time were given priority for sampling. However, due to th e widespread historic loss of wetlands throughout Florida ( FDNR 1988 ) and early incentives to drain swamplands, few pondcypress domes were found in the oldest urban areas. Many of the urban wetlands sampled were suspected to previously b een embedded in agricultural land uses. Table 2-1 provides some general informati on about each sample wetland, including sample date, surrounding land use, and land ow nership. The sample date provided is the earliest sample date, and correlates to m acrophyte sampling. A minimum water level of 10 cm was standardized to ensure sampling did not occur immediately following a small rain event, or too soon afte r initial hydration for the gr owing season, which would not

PAGE 52

37 Table 2-1. Surrounding land use, land ow nership, and sample date for 118 study wetlands in Florida. Site Code* Sample Date Surrounding Land Use^ Land Owner Site Code* Sample Date Surrounding Land Use^ Land Owner SA1 6/5/01 Cattle & Crops Public CR3 6/20/01 State Park Public SA2 6/6/01 Citrus Private CR4 8/10/01 WMD Public SA3 6/27/01 Cattle Public CR5 8/13/01 State Park Public SA4 7/30/01 Crops Private CR6 8/15/01 State Forest Public SA5 7/31/01 Cattle & Crops Public CR7 5/30/02 City Owned Public SA6 9/5/01 Cattle Private CR8 7/2/02 State Forest Public SA7 7/31/02 Woodland Public CR9 7/11/02 State Preserve Public SA8 7/31/02 County Park Public CR10 10/9/02 State Park Public SA9 8/1/02 Cattle Private CR11 10/9/02 State Park Public SR1 6/28/01 County Park Public CU1 5/31/01 Univ. Campus Public SR2 7/3/01 State Park Public CU2 6/15/01 Residential Private SR3 7/24/01 State Reserve Public CU3 7/16/01 Commercial Private SR4 8/1/01 National Park Public CU4 8/14/01 Road Side Private SR5 8/21/01 State Preserve Public CU5 9/11/01 Road Side Private SR6 9/18/01 NWR Public CU6 9/12/01 Golf Course Private SR7 7/15/02 County Park Public CU7 5/30/02 City Owned Public SR8 7/17/02 County Airport Public CU8 7/1/02 Industrial Private SR9 7/24/02 County Park Public CU9 7/8/02 Commercial Private SU1 6/6/01 Resid. & Golf Private CU10 8/7/02 Park Public SU2 6/29/01 School Campus Public CU11 8/8/02 Park Public SU3 7/4/01 Residential Public NA1 5/21/01 Cattle Public SU4 8/22/01 Residential Private NA2 6/4/01 Cattle Private SU5 8/23/01 Industrial Private NA3 6/19/01 Silviculture Public SU6 9/30/01 Industrial Private NA4 7/20/01 Crops Private SU7 7/16/02 Commercial Private NA5 7/27/01 Cattle Private SU8 7/16/02 Comm. & Resid. Private NA6 7/31/01 Silv., Cat.,Crops Private SU9 7/23/02 Residential Private NA7 5/22/02 Crops Public SU10 7/30/02 Roads & Canals Public NA8 5/21/02 Silviculture Private CA1 5/23/01 Crops Private NA9 6/10/02 Silviculture Ease. CA2 5/30/01 Cattle Private NA10 7/12/02 Silviculture Ease. CA3 6/7/01 Pullet Farm Private NA11 7/24/02 Cattle Public CA4 6/21/01 Cattle Public NA12 7/26/02 Cattle & Crops Public CA5 7/10/01 Cattle Private NR1 5/26/01 University Land Public CA6 7/23/01 Citrus Private NR2 6/18/01 City Park Public CA7 7/3/02 Silv. & Cattle Public NR3 7/10/01 State Forest Public CA8 7/19/02 Dairy Farm Public NR4 7/11/01 WMD Public CA9 7/24/02 Citrus Private NR5 8/6/01 Military Private CR1 5/30/01 Conserv. Tract Private NR6 8/21/01 State Park Public CR2 6/14/01 Conserv. Tract Private NR7 5/28/02 State Park Public

PAGE 53

38 Table 2-1. Continued Site Code* Sample Date Surrounding Land Use^ Land Owner Site Code* Sample Date Surrounding Land Use^ Land Owner NR8 8/5/02 State Park Public PA9 8/13/02 Row Crops Public NR9 8/29/02 State Forest Public PA10 8/14/02 Silviculture Public NU1 5/22/01 Road Side Private PR 1 6/15/01 National Forest Public NU2 6/11/01 Resid. & Golf Pr ivate PR2 7/3/01 WMD Public NU3 6/26/01 Residential Private PR3 7/4/01 Military Public NU4 6/27/01 Residential Private PR 4 8/9/01 State Forest Public NU5 6/28/01 Residential Private PR 5 8/10/01 State Forest Public NU6 8/1/01 Resid. & Instit. Privat e PR6 8/18/01 National Forest Public NU7 5/15/02 Comm. & Residential Private PR7 6/4/02 Conservation Tract Private NU8 6/3/02 Residentia l & Golf Private PR8 8/7/02 NWR Public NU9 6/12/02 Industrial Private PU 1 6/14/01 Residential Private NU10 7/29/02 Resid. & Instit. Priv ate PU2 7/5/01 Residential Private PA1 5/24/01 Cattle Private PU3 8/ 17/01 Resid. & Comm. Private PA2 5/29/01 Cattle Private PU4 8/17/ 01 Residential & Park Private PA3 7/3/01 Crops/Turf Grass Public PU5 9/28/01 Comm. & Silv. Private PA4 7/2/01 Crops Private PU6 9/29/01 Commercial Private PA5 8/8/01 Cattle Private PU7 6/18/02 Resid. & Orchard Private PA6 8/9/01 Cattle Private PU8 6/19/02 Indust. & Silv. Private PA7 6/5/02 Cattle Private PU9 6/20/02 Residential Private PA8 8/8/02 Silviculture Public PU10 7/25/02 Institutional Private *Site Codes correspond to the region, land use category, and sample order: S = south, C = central, N = north, and P = panhandle; R = re ference, A = agriculture, and U = urban. ^Surrounding Land Use abbreviations: NWR = National Wildlife Refuge; WMD = Water Management District; Resid. = Residential; Ca t. = Cattle; Comm. = Commercial; Instit. = Institutional; Crops = Row Crops; Sil v. = Silviculture; Ease. = Easement. allow the biological assemblages dependent on inundation time to respond. Wetlands sampled without sufficient standing water were revisited later in the field season once the wetlands held at least 10 cm of water. Table 2-2 identifies data collected at each wetland. Site codes reflect the ecoregi on (S = south; C = central; N = north; P = panhandle), land use category (R = reference; A = agricultur al; U = urban), and the order they were sampled. Site codes were assigned to preser ve the anonymity of individual land owners. Gradients of Landscape Development Intensity Three independent indices of anthropoge nic activity in the landscape were calculated for the study wetlands including th e Landscape Development Intensity (LDI)

PAGE 54

39 Table 2-2. Field-data collect ed at 118 sample wetlands. Site Soil Water Diatoms Macrophytes^ Macroinvertebrates Site Soil Water Diatoms Macrophytes^ Macroinvertebrates Site Soil Water Diatoms Macrophytes^ Macroinvertebrates SA1 CR4 NU1 SA2 CR5 NU2 SA3 CR6 NU3 SA4 CR7 NU4 SA5 CR8 NU5 SA6 CR9 NU6 SA7 CR10 NU7 SA8 CR11 NU8 SA9 CU1 NU9 SR1 CU2 NU10 SR2 CU3 PA1 SR3 CU4 PA2 SR4 CU5 PA3 SR5 CU6 PA4 SR6 CU7 PA5 SR7 CU8 PA6 SR8 CU9 PA7 SR9 CU10 PA8 SU1 CU11 PA9 SU2 NA1 PA10 SU3 NA2 PR1 SU4 NA3 PR2 SU5 NA4 PR3 SU6 NA5 PR4 SU7 NA6 PR5 SU8 NA7 PR6 SU9 NA8 PR7 SU10 NA9 PR8 CA1 NA10 PU1 CA2 NA11 PU2 CA3 NA12 PU3 CA4 NR1 PU4 CA5 NR2 PU5 CA6 NR3 PU6 CA7 NR4 PU7 CA8 NR5 PU8 CA9 NR6 PU9 CR1 NR7 PU10 CR2 NR8 CR3 NR9 = Data collected ^ -sampled with >10 cm standing water; -sampled with < 10 cm standing water

PAGE 55

40 index, the Wetland Rapid Assessment Pr ocedure (WRAP), and the Minnesota disturbance index. LDI, WRAP, and Minnesota disturbance index scores for each sample wetland are listed in Appendix B . LDI scores were calculated prior to site visits using 1999 digital orthophoto imagery of Florida available from Labins , The Land Boundary Information System from the Florida Department of Environmental Protection (FDEP) (available at http://www.labins.org/2003/index.cfm ). Sample wetlands were delineated from aerial images, and a 100 m buffer was c onstructed around the edge of each wetland in ArcView GIS 3.2 ( Environmental Systems Research Institute, Inc. 1999 ). Different land uses within the 100 m buffer were hand de lineated based on the aerial images. Land uses were updated during the si te visit to reflect any cha nges in land use since the 1999 aerial images were recorded. The followi ng equation was used to calculate LDI: LDI = (LDC * % LU) (2-1) where LDC is the Landscape Development Coefficient for a particular land use based on the amount of nonrenewable energy use per unit area in the surrounding landscape (Table 2-3), and %LU is the percent of a land use within the 100 m buffer of the wetland. The LDC values and LDI equation are based on work by Lane ( 2003 )and Brown and Vivas ( 2004 ). Potential LDI scores ranged from a minimum of 1.0 (Natural Land/Open Space) to a maximum of 10.0 (Central Business District). WRAP was scored during the initial 30 mi nutes at each study wetland according to descriptions from Miller and Gunsalus ( 1997 ). The Minnesota disturbance index was scored (after the field visit) by the field crew leader us ing information obtained from ArcView GIS 3.2 ( Environmental Systems Research Institute, Inc. 1999 ) and field notes using categories established by Gernes and Helgen ( 1999 ).

PAGE 56

41 Table 2-3. Landscape Development Coefficients (LDC) used in the calculation of the Landscape Development Intensity (LDI) index. Land Use Nonrenewable Energy Use (E14 solar equivalent joules/ha/yr) LDC Natural Land / Open Water 1.0 1.0 Pine Plantation 5.1 2.0 Low Intensity Open Space / Recreational 6.7 2.1 Unimproved Pastureland (with livestock) 8.3 2.6 Improved Pasture (no livestock) 19.5 3.7 Low Intensity Pasture (with livestock) 36.9 4.5 Medium Intensity Open Space / Recreational 51.5 4.8 High Intensity Pasture (with livestock) 65.4 4.9 Citrus 67.3 5.2 Row crops 117.1 5.9 High Intensity Agriculture (dairy farm) 201.0 6.6 Recreational / Open Space (High-intensity) 1077.0 6.9 Single Family Residential (Low-density) 1230.0 7.6 Single Family Residential (Med-density) 2461.5 7.7 Single Family Residential (High-density) 3080.0 8.0 Low Intensity commercial (Comm Strip) 3729.5 8.0 Institutional 3758.0 8.1 Highway (2 lane) 4042.2 8.3 Industrial 5020.0 8.3 Multi-family residential (Low rise) 5210.6 8.7 Highway (4 lane) 7391.5 8.9 High intensity commercial (Mall) 12661.0 9.2 Multi-family residential (High rise) 12825.0 9.2 Central Business District (Avg 2 stories) 16150.3 9.4 Central Business District (Avg 4 stories) 29401.3 10.0 Field-data Collection A concise summary of field-data collection procedures follows. Appendix C provides more detailed descriptions of fielddata collection techniques in the format of Standard Operating Procedures (SOPs) for fi eld use. Field methods are described as transect establishment followed by wa ter, soil, diatom, macrophyte, and macroinvertebrate sampling techniques.

PAGE 57

42 Figure 2-2. Belted transect layout for macrophyt e sampling and the location of the water and soil samples. One soil core ( o ) was taken along each transect and compiled for each wetland. One water sample ( ) was taken in the approximate center of each wetland. Sampling Design Figure 2-2 shows the positioning of the 4 transects established at each wetland that were situated as perpendicular crossing axes running through the center of each wetland. Transect axes always corresponded to the cardin al directions (north, east, south, west). The wetland/upland boundary was determined based on a combination of wetland plant presence according to wetland plant status (for example, obligate, facultative, or upland from Tobe et al. 1998 ) and wetland hydrologic indicators. Water Samples A grab style water sample was taken in the deepest pool of each wetland when a minimum of 10 cm of standing water was pres ent throughout at least 50% of the wetland West East North SouthWetland/upland boundary Quadrats were placed back-toback and were 1m wide by 5m long. o4 belted transects of variable length were arranged along the cardinal axes. o o o

PAGE 58

43 area. The area with the deepest pool often coincided with the center of each wetland as depicted in Figure 2-2. Water samples were collected at 75 wetla nds, including 14 in the panhandle ecoregion (reference n = 5; agricultural n = 4; ur ban n = 5), 14 in the north ecoregion (reference n = 6; agricultural n = 3; urban n = 5), 23 in the central ecoregion (reference n = 9; agricultural n = 6; urban n = 8), and 24 in the south ecoregion (reference n = 8; agricultural n = 8; urban n = 8). Dissolved oxygen and water temperatur e were taken on-site using a YSI-55 Dissolved Oxygen hand meter. Grab water samples were sent to the Florida Department of Environmental Protection (FDEP) Central Ch emistry Laboratory, Tallahassee, Florida. Analysis included color (EPA 110.2), turb idity (EPA 180.1), pH (150.1), specific conductance (EPA 120.1), ammonianitrogen (E PA 350.1), nitrate/nitrite-nitrogen (EPA 353.2), total Kjeldahl nitrogen (EPA 351.2), and total phosphorus (EPA 365.4). Soil Samples A composite soil sample was collected at all 118 sample wetlands. Cores were taken using a 7.6 cm diameter PVC pipe driven 10 cm into the soil. One soil core was collected in the approximate middle of each tr ansect (Figure 2-2), and soil cores were homogenized into a composite sample per site. Soil moisture ( Gardner 1986 ), organic matter, total Kjeldahl nitrogen ( USEPA 1993 ), and total phosphorus ( USEPA 1979 ) were analyzed. Nitrogen and phosphorus samples were processed through the Institute of Food and Agricultural Sciences (IFAS) Analyt ical Research Laboratory, Gainesville, Florida. Diatoms Benthic diatom samples were collected at 50 isolated forested wetlands throughout Florida between May-September 2001, as liste d in Table 2-2. Figure 2-3 shows the

PAGE 59

44 Ê Ú# S % UÊ Ú Ê Ú# S % U # S # SÊ Ú Ê Ú Ê Ú# S # SÊ Ú% U % U # S # S % U % U # S % UÊ Ú% UÊ Ú# SÊ Ú Ê Ú Ê Ú% U # S # S % U # S % U % U % U % U % U % U # S # S # S # SÊ Ú Ê Ú Ê Ú# SÊ Ú 0200400KilometersFlorida Ecoregions Panhandle North Central South Sample Wetlands (n=50)# SReference% UAgriculturalÊ ÚUrban N Figure 2-3. Benthic diatom samples were coll ected at 50 isolated forested wetlands. The state of Florida was separated into four ecoregions ( Lane 2000 ). Sample wetlands were designated by surr ounding land use: o reference, agricultural, or urban. spatial location of the wetlands with benthic diatom samples in Florida. Sites were sampled in the panhandle (n=10), north (n =10), central (n=13), and south (n=17) ecoregions. Sample wetlands were situated in 3 a priori described land use categories (reference n=18; agricultur al n=16; urban n=16). A minimum of 10 cm of standing wate r was necessary for benthic diatom collection. Ten samples were taken thr oughout the flooded portion of the wetland. A hollow cylinder was placed on the soil surface to isolate an area of substrate with a surface area of 28 cm2. A bulb pipette was used to loos en the top 0.5 cm layer at the soil surface-water interface, and a 10 mL sample wa s extracted. This was repeated 10 times

PAGE 60

45 throughout the wetland, resulting in a final sa mple volume of 100 mL. For preservation, 5 mL of M3, an standard pres ervative for algae samples ( APHA 1995 ), was added to the 100 mL algae sample. Benthic algae samples were shipped to Michigan State University for identification and enumeration under the supervision of R. J. Stevenson. Samples were homogenized prior to sub-sampling for laboratory identifica tion. Sub-samples were digested following Hasle and Fryxell ( 1970 ), which removed the organic matter from the diatom frustules to aide in identification. Following rinsing with distilled water, the digested sub-samples were mounted on microscope slides usi ng Naphrax (Northern Biological Supplies Limited, Ipswich, England). Five hundred valves were counted along microscope transects and identified to the lowest po ssible taxonomic level, preferably species (following FDEP SOP#AB-03.1 http://www.dep.state.fl.us/labs/sop ). Macrophytes Macrophyte vegetation was sampled at al l 118 depressional isolated forested wetlands throughout Florida (Figure 2-1). Macrophyte sampling was conducted along the 4 transects situated as perpendicular cr ossing axes running thr ough the center of each wetland, shown in Figure 2-2. Along each tran sect, a series of 1 m wide by 5 m long quadrats was established back to back. Li ving macrophytes rooted within each quadrat were identified to the lowe st taxonomic level possible. Supplementary data Taxonomic information including species, genus, and family were compiled for all of the macrophytes identified. Additional characteristics were collected for use in developing potential biological in dicator metrics, including ca tegory (annual or perennial, evergreen or deciduous, indigenous or exotic) an d growth form (aquatic, fern, grass, herb,

PAGE 61

46 sedge, shrub, tree, or vine). References specific to Florida were consulted first ( Tobe et al. 1998 ; Wunderlin and Hansen 2003 ), and additional information was supplemented from other sources (in the following order: Godfrey and Wooten ( 1981 ), Wunderlin ( 1998 ), and USDA NRCS ( 2002 ). When information was still unavailable in published literature for species encountered, Florida botani sts (who also participated in the Floristic Quality Assessment Index) were consulted. Floristic Quality Index Five Florida botanists agreed to participate in creating a Floristic Quality Assessment Index (FQI) for Florida isolated forested wetlands. The Florida FQI was modeled after an earlier st udy from Chicago, Illinois, by Willhelm and Ladd ( 1988 ), which enlisted botanists to provide quantitati ve scores for vegetation based on the fidelity of each plant species to a pa rticular environment. The FQI score for an individual wetland was calculated as: Modified FQI = [ (CC for each species present) ] / species richness (2-2) where CC = Coefficient of Conservatism score. This equation is considered a modified FQI because previous studies did not account for species richness. The sum of the species CC scores was divided by species ric hness (Equation 2-2) in this study to account for potential differences in species ric hness due to differences in ecoregions, a priori land use categories, or othe r unforeseen differences. CC scores were obtained from the Florida botanists surveyed. Each botanist was sent a complete list of species found in th e isolated forested wetlands in the 2001 field season (n = 482 species), and was asked to sc ore each species based on its faithfulness to Florida isolated forested wetlands. After th e 2002 field season, one botanist scored the

PAGE 62

47 additional 79 taxa not previous ly encountered, raising the numb er of taxa with CC scores to 561 species. Potential CC scores ranged from 0 to 10: 0 exotic taxa and native taxa that act as opportunistic invaders, includes species that commonly occur in disturbed ecosystems 1-3 taxa that are widely distribut ed and occur in disturbed ecosystems 4-6 taxa with a faithfulness to a particular ecosystem, but also tolerant of moderate levels of disturbance 7-8 taxa typical of well-est ablished ecosystems that sust ain only minor disturbances 9-10 taxa that occur within a narro w set of stable ecological conditions Species with low CC scores were considered tolerant of many disturbances, whereas species with high CC scores were considered to occur within a narrow set of stable ecological condition. Appendix D lists the CC scores for 56 1 macrophytes identified in this study. Macroinvertebrates Macroinvertebrates were collected at 79 de pressional isolated forested wetlands throughout Florida as shown in Figure 2-4. Field research spanned two growing seasons with 49 wetlands sampled between June -October 2001 and an additional 30 wetlands sampled between June-October 2002. Sites were sampled in the panhandle (n=13), north (n=15), central (n=25), and sout h (n=26) ecoregions; sample wetlands were situated in 3 a priori land use categories (reference n=29; agriculture n=24; urban n=26). Samples were collected using a U.S. Sta ndard 30 mesh D-frame net. One sweep covered 0.5 m2 and was measured as 1 net width by 2 net lengths wide, which was repeated 3 times at each location to ensure adequate sampling coverage. Sweeps were always conducted over areas which had not re cently been trampled by the field crew. Twenty sweeps were proportioned among major vegetation zones throughout each sample wetland. Sample wetlands generally had between 1 to 3 vegetation zones,

PAGE 63

48 Figure 2-4. Macroinvertebrates were sampled at 79 isolated forested wetlands. The state of Florida was separated into four eco-regions ( Lane 2000 ). Sample wetlands were designated by surrounding land use: o reference, agricultural, or urban. which were defined by changes in the dominan t species cover. When herbaceous plants were included in the sweep area, the bottom of the net was swept from the bottom of the substrate up the plants. In areas with woody plants, the bottom of the net was swept from the substrate up the tree trunk and pieces of woody debris were brushed to remove attached macroinvertebrates. The contents from the sweeps were collected in a 3.8 L

PAGE 64

49 plastic jar and preserved with buffered formalin at a rate of 10% of the sample volume. Appendix C provides a more detailed descriptio n of field methods and preservation. Macroinvertebrate identification was comp leted at the Florida Department of Environmental Protection (FDEP) Central Laboratory, Tallahassee, Florida, following standard operating procedures (FDEP Standard Operating Procedure #IZ-06 http://www.dep.state.fl.us/labs/sop ). Macroinvertebrate samp les were sieved, washed, and placed on a pan with 24 individually numbered cells. One-third of the cells were randomly selected and combined in a second numbered tray. A single cell was randomly selected from the second tray for enumerati on and identification of macroinvertebrates. When fewer than 100 individuals were enc ountered in the sample, a second cell was randomly selected from the second tray, and al l of the individuals were enumerated and identified. Identification was to the lowest taxonomic level possible. Data Analysis Water and Soil Parameters Water and soil parameters within 3 a priori land use categories were compared using FisherÂ’s LSD pair wise comparison in Minitab ( Version 13.1, Minitab Statistical Software ). The non-parametric Mann-Whitney UTest was used to discern differences among medians of low and high LDI groups for the water and soil parameters. Wetlands in the low LDI group had a site LDI score less than 2.0, whereas wetlands in the high LDI group had site LDI scores greater than or equal to 2.0, corresponding to a break in the LDC coefficients of undeveloped vers us developed land uses (Table 2-3). To test for multicollinearity among the environmental variables, the variance inflation factor (VIF) and to lerance were calculated using SAS ( Version 6 from the SAS Institute, Inc., Cary, North Carolina ). Multicollinearity occurred when two or more of the

PAGE 65

50 independent variables exhibited a comparable pattern of correlation with other variables ( Zar 1999 ; Tabachnick and Fidell 1983 ). To avoid issues with multicollinearity, environmental variables with a VIF greater than 10.0 and a tolerance less than 0.10 ( Ott and Longnecker 2001 ; Pan and Stevenson 1996 ; ter Braak 1987 ) were excluded from further analyses. Summary Statistics Summary statistics for each assemblage in cluded richness (R), evenness (E), and Shannon diversity (H). For the diatom a nd macrophyte assemblages summary statistics were calculated at the speci es level; for the macroinvertebrate assemblage summary statistics were calculated at the genus level. Richness was defined as the total number of distinct taxa encountered w ithin the sample wetland. Evenness was calculated as the Shannon diversity value divided by the natural log of richness: E = H / log (R) (2-3) ( McCune and Grace 2002 ). Evenness has also been described as the fraction of maximum possible diversity in a wetland. The Shannon diversity index has been described as measuring the “information c ontent” of a sample unit where maximum diversity yields maximum uncertainty ( McCune and Grace 2002 ). For Shannon diversity calculations (H), the sample unit wa s an individual forested wetland: H = pi * log(pi) (2-4) pi = ni / N (2-5) where ni was the number of occurrences of taxo n i, and N was the total number of occurrences of all taxa at a wetland. For the diatom and m acroinvertebrate assemblages, the number of occurrence represented the enum eration of the laboratory identified sample and the total number of occurrences represen ted the sum of the number of occurrences of

PAGE 66

51 all taxa; for the macrophyte assemblage, the number of occurrences represented the number of quadrats a species occurred in, and the total number of occurrences represented the sum of the to tal number of quadrats of a ll of the species identified. For the diatom and macroinvertebrate a ssemblage, Simpson’s index (S) was also calculated as S = 1 – (pi * pi) (2-6) where p was defined in Equation 2-5. For the macrophyte assemblage firstand secondorder jackknife estimators of species richness (Jack1 and Jack2, respectively) and Whittaker’s beta diversity ( W) were calculated. Firstand second-order jackknife estimators were calculated as estimates of true species richness ( Colwell and Coddington 1994 ). Assuming that the sampling effort onl y measured a portion of the ecosystem, jackknife estimators of species richness provide estimates of actual species richness. The equation for first-order jackknife estimators of species richne ss is based on the number of species observed (S), the number of species o ccurring in only one sample unit (rl) (where one sample unit represents one quadrat), and the number of sample units (n) (quadrats): Jack1 = S + [(rl*(n-1)) / n] (2-7) The second-order jackknife estimator (Jack2) also incorporated the number of species occurring in exactly two samp le units (r2) (quadrats): Jack2 = S + [(rl*(2n-3)) / n] – [(r2*(n-2)2) / (n*(n-1))] (2-8) These estimators of total species richness ha ve shown useful in pr edicting actual species richness when only a small area of th e total ecosystem has been sampled ( McCune and Grace 2002 ). Whittaker’s beta diversity ( W) was computed as a calculation of overall beta diversity, or the compositional change represented in a sample. Whittaker’s beta

PAGE 67

52 diversity was calculated as th e number of species at a pa rticular forested wetland (Sc) divided by the average species ri chness per quadrat (S) minus one: w = [Sc / S] – 1 (2-9) The resulting value for Whitaker’s beta diversit y was described as the “number of distinct communities” ( McCune and Grace 2002 ). When w equals zero, all of the sample units contain all of the species. Some multivariate methods strongly depend on beta diversity, and as a general rule beta diversity gr eater than five is considered high ( McCune and Grace 2002 ). Summary statistics means within a priori land use categories were compared with Fisher’s Least Significant Difference (L SD) pair wise comparison test using Minitab ( Version 13.1, Minitab Statistical Software ). The strength of using Fisher’s LSD was in the comparison of unequal group sizes ( Ott and Longnecker 2001 ; Minitab 2000 ). Sample wetlands were divided into two groups based on Landscape Development Intensity (LDI) index valu es including low LDI (LDI < 2.0) and high LDI (LDI 2.0) groups. Comparisons were made using the non-parametric Mann-Whitney U-Test in Minitab ( Ott and Longnecker 2001 ). Overall calculations of beta and gamma diversity were calculated for sample wetlands in the 3 a priori land use categories. Gamma di versity was calculated as the overall number of taxa encounter ed at all sample wetlands per a priori category. A higher gamma diversity for an a priori land use category would suggest a greater difference among the species composition of wetlands within that a priori land use category, assuming a similar number of wetlands were sampled within each a priori land

PAGE 68

53 use category. Beta diversity was calculated as a priori category gamma diversity divided by the average site taxa richness. Regional Compositional Analysis The Multi-Response Permutation Procedure (M RPP) was used to test the similarity of community composition for each assemblage among the 4 Florida ecoregions (further application in Zimmerman et al. 1985 ; McCune et al. 2000 ; McCune and Grace 2002 ). MRPP is a nonparametric technique which te sts for no difference between groups (the null hypothesis) and is available in PCORD ( Version 4.1 from MJM Software, Gleneden Beach, Oregon ). It was an appropriate procedur e for ecological community data as it does not require distributional assumptions of normality and homogeneity of variances. The Sorensen distance measure was used to ca lculate the average weighted within-group distance. MRPP provides a test statistic (T), p-value, and chance-corrected within-group agreement (A), which describes within-group similarity. When A equals one, all items are identical within groups, and when A equals zero, differences with in-groups equal that expected by chance. Most values of A are less than 0.1 in community ecology ( McCune and Grace 2002 ). MRPP was calculated across all gr oups (panhandle versus north versus central versus south) as well as for multiple pair wise comparisons (panhandle versus north, panhandle versus central, panhandle ve rsus south, north versus central, north versus south, and central versus south). Community Composition Community composition of each assemblage was summarized in a non-metric multidimensional scaling (NMS) ordination to relate changes in community composition with environmental gradients. NMS is an ordination technique designed to compress multi-dimensional space, and is particularly agreeable with ecological data because it

PAGE 69

54 does not rely on linear relationships among vari ables. This has been described as a compensation for the “zero-truncation problem ” through the use of ranked distances and the use of many distance measure ( McCune and Grace 2002 ). The “zero-truncation problem” refers to the extraordinary number of zeros in community ecology data sets. Other ordination techniques depend upon a valu e for each measured variable for each sample unit. However NMS is useful for pr esence/absence or abundance data sets, where many species are not present, or receive a zero in the species by site matrix. By ranking the variables, NMS caters to non-para metric, community ecology data sets. NMS explored the dissimilarities of the community composition of sample wetlands for each assemblage. The Sorensen (Bray-Curtis) distance measure was used for ordination. The dimensionality was chos en based on an initial 6 dimensional run in autopilot mode, which suggested an optimal 3 dimensional solution for the each community composition dataset. To find th e optimal 3 dimensional solution, 50 runs with real data and 50 randomized runs were pe rformed with the instab ility criterion set at 0.00001 and the maximum number of iterations to reach a stable solution set at 500. This procedure was repeated 20 times to insure stab ility and reproducibility in results. The final run was completed with the starting point set as the results from the best experimental 3 dimensional run, with th e lowest stress and best overall fit. Water and soil parameters, LDI, latitude, and longitude were correlated with the NMS ordination axes with Pearson’s correla tion coefficients. To improve normality and decrease skewness, 11 water and soil para meters were log (base 10) transformed, including water parameters (dissolved oxygen concentration, temperature, color, turbidity, specific condu ctivity, ammonia-nitrogen concen tration, nitrate/ nitrite-nitrogen

PAGE 70

55 concentration, total Kjeldahl nitrogen (TKN) concentrati on, and total phosphorus (TP) concentration) and soil parameters (TKN a nd TP concentration). Water pH was not transformed. The remaining soil parameters were measured as percentages, and were transformed by taking the arcsin e square root. These included soil moisture and organic matter. Metric Development In the context of this study, metrics were defined as biologic al attributes which have a consistent and predictable response to anthr opogenic activities ( Karr and Chu 1997 ). Metrics were summarized in 5 main categories: tolerance metrics (indicator species or esta blished index values such as the Florida Index) autecological (metrics that explore a pr eviously described relationship between a species and an environmental parameter) community structure (metrics th at explore taxonomic structure) community balance (metrics with calculated values, such as evenness or dominance) functional group (metrics rela ted to feeding behavior) Appendix E provides tables of candidate metr ics for each assemblage, including 169 candidate diatom metrics (Table E-1), 238 candidate macrophyte metrics (Table E-2), and over 400 candidate macroinvertebrate metr ics (Table E-3). Candidate metrics were calculated at the statewide scale for both the diatom and macroinvert ebrate assemblages, as sample sizes were limited for regional metr ic development, particularly in the north and panhandle ecoregions. Candidate macr ophyte metrics were calculated at both the regional scale and statewide. Metrics for the diatom assemblage were calculated as three main forms, including the number (N), percent (P), and abundan ce (A) based on the single composite sample taken at each sample wetland. The number metr ic (N) referred to a straight count of

PAGE 71

56 species fitting the particular metric. The pe rcent metric (P) was calculated as the number metric (N) divided by the species richness (R) for each sample wetland. Pi = Ni / Ri (2-10) where i represents a sample wetland. The abundance metric (A) referred to the sum of the total number of individuals designated by the metric (m) divided by the total number of all individuals identified at a wetland (M): Ai = mi occurrence of metric species / Mi all species occurrences (2-11) where i represents a sample wetland. The macroinvertebrate metrics were calculated as the diatom metrics, with the addition of the number of taxa (T), which was calculated as the number of lower taxonomic grou ps within the metric category. Macrophyte data were collected within mu ltiple quadrats at each sample wetland, so additional metric forms were possible. Candidate metrics were constructed as the number (N), percent (P), abundance (A), and frequency of occurrence (F). The abundance metric (A) referred to the sum of the total number of species designated by the metric in each quadrat for each respective sample wetland (m) divided by the total number of all species occurrences at a wetla nd (M). The frequency of occurrence metric (F) was calculated as the number of quadrats a particular category of species occurred in (q) divided by the total number of qu adrats sampled at each wetland (Q). Fi = qi / Qi (2-12) Candidate metrics were accepted if they s howed a constant and predictable change along the LDI ( Brown and Vivas 2004) according to the strength and significance of the SpearmanÂ’s correlation coefficient calculated with Analyze-It software v. 1.67 for Microsoft Excel. The Spearman rank correlati on tests for an association between 2

PAGE 72

57 related variables, and is a nonparametric alternative to the Pearson correlation. Scatter plots were constructed for each candidate metric versus LDI to ensure correlations were visually distinguishable. The pool of poten tial candidate metrics was streamlined to reduce the redundancy of selected metrics. Candidate metrics were subjected to a 2sample t-test to detect differences between low and high LDI groups. Indicator Species Analysis For each assemblage, sample wetlands were categorized into two LDI groups and analyzed with Indicator Species Analysis (ISA) in PCORD , which evaluates the abundance and faithfulness of taxa in a defined group ( McCune and Grace 2002 ). ISA can be used to detect and describe the value of taxa indicative of environmental conditions. It requires a priori groups and data on the abundance or presence of taxa in each group. These groups are commonly define d by categorical environmental variables, levels of disturbance, experimental treatments , presence and absence of a target species, or habitat types ( McCune and Grace 2002 ). The ISA calculation combines information on the concentration of species abundance and the faithfulness of occurrence of a species in a group. Mathematical e quations are available in McCune and Grace ( 2002 ) and Dufrêne and Legendre ( 1997 ). The calculated indicator sp ecies values were based on two standards, faithfulness and exclusion. Faithfulness was defined mathematically by a particular taxa always being present in a particular group. Additionally, the perfect indicator taxa would be exclusive to that group, meaning it never occurred in other groups ( Dufrêne and Legendre 1997 ; McCune and Grace 2002 ). Calculated indicator values ranged from 0 (no indication), to 100 (a perfect indication of a particular group). Multiple ISA were conducted to determine sensitive and tolerant indicator taxa for each assemblage. Sample wetlands were cat egorized based on consecutive LDI breaks

PAGE 73

58 from 1.0 through 7.0 at each 0.25 increment. Each ISA was conducted at each LDI break once using the abundance and once using the pr esence of taxa at each wetland for each assemblage. The percent sensitive and tole rant indicator taxa at each wetland was calculated and correla ted with LDI using Spearman rank correlation. ISA was conducted for each ecoregion (panhandle, north, central, and south) as well as statewide for the macrophyte assemblage. Only statewide an alyses were run for the diatom and macroinvertebrate assemblages. Indicator values were calculated and tested for statistical significance using a Monte Carlo randomizati on technique with 1000 randomized runs. The ISA was used to identify taxa with si gnificant associations to LDI categories. Indicator taxa categorized as tolerant taxa were associat ed with the higher LDI group; indicator taxa deemed sensitive taxa we re associated with the lower LDI group. In the macrophyte assemblage, the Spearma n rank correlation was used to assess differences between statewide and regional indicator species lists for each of the 4 ecoregions. To test for equal application of the statewide indicator species list for each ecoregion, the non-parametric Kruskal-Wallis test was run with Analyse-It software ( Ott and Longnecker 2001 ). Distributional differenc es were analyzed between a priori categories for both tolerant and sensitive indicator taxa among ecoregions. Diatom metrics Diatom metrics were created in 3 cat egories including tolerance, community composition, and autecological metrics ( Bahls 1993 ; van Dam et al. 1994 ; McCormick and Cairns 1994 ; Stevenson 2001 ; Fore and Grafe 2002 ; Lane et al. 2002 ; USEPA 2002b ; Lane 2003 ;). Tolerance metrics were created wi th ISA. Community composition metrics included richness, evenness, and diversity calcul ations as described a bove. Autecological metrics were based on previous research that correlated individual diatoms with

PAGE 74

59 morphology, behavior, and the physical and ch emical water environment. Diatom species were assigned ecological indicator valu es using a coded checklist of autecological relationships ( Bahls 1993 ; van Dam et al. 1994 ). The ecological indicator values from van Dam et al. ( 1994 ) included categories diatoms according to water preference, nitrogen metabolism, pH, salinity, diss olved oxygen, saprobic condition, and trophic status. Additional autecological ecologica l indicator values were adapted from Bahls ( 1993 ) analysis of diatoms in Montana streams which included pollution tolerance classifications. Metrics based on morphology and motility were assigned from Stevenson ( Lane 2003 ). Macrophyte metrics Macrophyte metrics included to lerance, exotic species, Floristic Quality Index (FQI), longevity, plant growth fo rm, and wetland status metrics ( Adamus 1996 ; Kantrud and Newton 1996 ; Galatowitsch et al. 1999a ; Gernes and Helgen 1999 ; Carlisle et al. 1999 ; Fennessy et al. 2001 ; Mack 2001 ; and Lane 2003 ). Tolerance metrics were calculated with ISA. The exotic species metric was calculated as the percent of species that were exotic to Florida divided by the num ber of species identified at each particular isolated forested wetland. The timeline for de termining the exotic status of a species was set near the beginning of European settleme nt in North America. Many sources were consulted to determine whether a spec ies was considered exotic, including Godfrey and Wooten ( 1981 ), Tobe et al. ( 1998 ), Wunderlin ( 1998 ), USDA NRCS ( 2002 ), and Wunderlin and Hansen ( 2003 ). For each wetland the modified FQI metric was calculated. Each species was categorized as native or exotic and annual or perennial ( Godfrey and Wooten 1981 ; Tobe et al. 1998 ; Wunderlin 1998 ; USDA NRCS 2002 ; and

PAGE 75

60 Wunderlin and Hansen 2003 ). The percent native perennial species metric was calculated as the number of native perennial species encountered divided by the wetland species richness. Wienhold and Van der Valk ( 1989 ), Ehrenfeld and Schneider ( 1991 ), and Lane ( 2003 ) and determined that disturbance often favors annual species over perennial species or promotes the inva sion of nonnative perennials in wetlands. Galatowitsch et al. ( 2000 ) found that while native pere nnial cover was reduced in wetlands impacted by cultivation, the occurren ce of introduced perennials rather than annuals increased in stormw ater impacted wetlands. The wetland status metric was calculated as the percent of plants classified as obligate or facultative wetland indicator specie s divided by the number of species at each wetland. Wetland indicator status classifications was from Tobe et al. ( 1998 ), USDA NRCS ( 2002 ), and Wunderlin and Hansen ( 2003 ). There were 5 potential wetland status classifications, including oblig ate, facultative wetland, facult ative, facultative upland, and upland. When possible, the Florida specific wetland indicator status was applied. In some cases, when the Florida wetland indicator status was not available, the National Wetlands Inventory wetland indicator stat us for the United States was used. Macroinvertebrate metrics Candidate metrics for the macroinvertebra te assemblage were constructed in 4 categories, including tolerance, community st ructure, community ba lance, and functional group metrics ( Lenat 1993 ; Lenat and Barbour 1994 ; Kerans and Karr 1994 ; Wallace et al. 1996 ; Barbour et al. 1996b ; Gerristen and White 1997 ; Danielson 1998a ; Leslie et al. 1999 ; Galatowitsch et al. 1999a ; Smogor and Angermeier 2001 ; Helgen 2001 ; Cummins and Merritt 2001 ; USEPA 2002c ; Lane et al. 2002 ; Lane 2003 ; Griffith et al. 2003 ; and Butcher et al. 2003 ). In addition to ISA, other tolerance candidate metrics were

PAGE 76

61 calculated. Many of the established toleran ce metrics were created for flowing waters, which complicated the application of establis hed tolerance values (ex. the Florida Index from Barbour et al. 1996b and Beck 1954 ; and the Hilsenhoff Biotic Index from Hilsenhoff 1987 ), since the wetlands sampled in this study were isolated from flowing surface waters (except in extreme high water events when some surface flow may be detectable). Community structure metrics in cluded richness measures ( Danielson 1998a ), for example the number of distinct species or specified taxonomic units like the number of families, genera, or species in a collection. Examples of taxa richness metrics include total taxa richness, Ephemer optera, Plecoptera, and Trichopt era richness; the number of Coleoptera species; or the num ber of Insecta species ( Danielson 1998a ). The use of community balance metrics included some measure of abundance or relative abundance in an attempt to meas ure the evenness of the macroinvertebrate community ( Lenat and Barbour 1994 ). Examples of community balance metrics include the Shannon diversity index or the percen t contribution of the most abundant taxon ( Lenat and Barbour 1994 ). Macroinvertebrate taxa were grouped based on their functional relationships that overlap taxonomic categoriz ation, including func tional feeding groups, habitat groups, and voltinism groups (or life-cycle patterns). Cummins and Merritt ( 2001 ) suggest using ratios of numerical abundance or, more fa vorably, biomass of the various functional groups as indicators of ecosystem attributes, essentially considering the functional groups as surrogates of ecosystem condition. Func tional feeding group me trics were based on the morphological structures a nd behaviors responsible for food acquisition by particular

PAGE 77

62 taxa at a site ( Resh and Jackson 1993 ; Danielson 1998a ). As an example, herbivores consume algae and plant material, while pred ators consume animals, omnivores eat both plant and animal materials, and detritovores consume decomposed particulate material ( Helgen 2001 ). Wetland Condition Index Candidate metrics were selected for inclusi on in the WCI if they satisfied 3 criteria: 1. Metrics were correlated with the LDI accord ing to the strength and significance of the SpearmanÂ’s correlation coefficient 2. Displayed visually distinguishable co rrelations with LDI in scatter plots 3. Showed a significant difference between low and high LDI groups tested with the Mann-Whitney U-test. A WCI was constructed for each assemblage , including the diatom WCI, the macrophyte WCI, and the macroinvertebrate WCI. Each index was composed of individual metrics specific to the assemblage, which were scaled and added together. Metric scoring was based on an approach modified from th e Stream Condition In dex, a Florida based biological index of the macroi nvertebrate assemblage used to discern stream condition ( Fore 2003 ). Metrics with a skewed distributi on were log transformed to improve the distribution. The 5th to 95th percentile values of each me tric were normalized from 0 to 10, with 10 always representing the best biological wetland condition. The selected metrics, WCI, and LDI were correlated with water and soil parameters using SpearmanÂ’s correlation coefficient. Cluster Analysis In order to determine whether the WCI provided comparable scores for wetlands with similar community composition within ea ch assemblage, an agglomerative cluster analysis in PCORD was used to determine wetland clus ters. A further description is

PAGE 78

63 available in McCune and Grace ( 2002 ). The dissimilarity matr ix was constructed using the Sorensen distance measure and the flexible beta ( = -0.25) linkage method, which is a flexible clustering setting designed to reduce chaining in the dendrogram. The resulting dendrogram was pruned to maintain the sma llest number of si gnificantly different clusters based on FisherÂ’s LSD pair wise comparison (p < 0.05). Comparisons among Wetland Condition Index Metrics Metrics selected for inclusion in the WCI were compared using the Pearson correlation coefficient ( Analyse-it Software ).

PAGE 79

64 CHAPTER 3 RESULTS Water and Soil Parameters Water samples were analyzed for 75 wetlands, and soil samples were analyzed for all 118 isolated forested wetlands sampled. Ta ble 3-1 shows mean values ± the standard deviation for water and soil parameters for the 3 a priori land use categories (reference, agricultural, and urban). M eans with similar letters were not significantly different (Fisher’s LSD pair wise comparison, = 0.05). Water temperatur e, water nitrate/nitritenitrogen, and soil TKN were not significantly different among a priori land use categories. Reference wetlands had significantly diffe rent dissolved oxygen, turbidity, water pH, and water column total phosphorus, than agricultural and urban wetlands. The water color of urban wetlands was significantly di fferent from reference and agricultural wetlands. Specific conductivity was significan tly different between reference and urban wetlands. Water ammonia-nitr ogen (mg N/L), water TKN (mg N/L), soil moisture, and soil TP (mg P/g soil) were significantly di fferent between reference and agricultural wetlands. Soil organic matter was significantly different between agricultural and urban wetlands. Table 3-2 shows the Mann-Whitney U-te st results comparing water and soil parameters of low (LDI < 2.0) and high (LDI 2.0) LDI groups. Dissolved oxygen, turbidity, water pH, water TP, soil moisture , and soil TP were significantly different between LDI groups. Water temperature, sp ecific conductivity, and water ammonia-

PAGE 80

65 Table 3-1. Water and soil parameters among 3 a priori land use categories. Reference*Agricultural* Urban* Water parameters Dissolved oxygen (mg/L) 2.9 ± 1.7a1.6 ± 0.9b 1.9 ± 1.1b Temperature (ºC) 26.2 ± 2.8a25.2 ± 1.9a 24.9 ± 2.4a Color (PCU) 285 ± 178a346 ± 204a 198 ± 129b Turbidity (NTU) 3.8 ± 4.2a17.7 ± 40.7b 9.5 ± 11.9b pH 5.2 ± 1.2a6.2 ± 0.8b 6.4 ± 1.0b Specific conductivity (umhos/cm) 81 ± 48a136 ± 134ab 231 ± 175b Ammonia-nitrogen (mg N/L) 0.15 ± 0.33a0.33 ± 0.57b 0.19 ± 0.27ab Nitrate/nitrite-nitrogen (mg N/L) 0.09 ± 0.37a0.01 ± 0.01a 0.02 ± 0.03a TKN nitrogen (mg N/L) 1.93 ± 1.24a3.17 ± 2.20b 1.84 ± 1.06ab Total phosphorus (mg P/L) 0.08 ± 0.11a0.81 ± 1.38b 0.23 ± 0.26bSoil parameters Moisture (%) 61 ± 20a46 ± 17b 55 ± 22ab Organic matter (%) 40 ± 25ab30 ± 17a 41 ± 28b TKN nitrogen (mg N/g soil) 6.76 ± 3.68a5.53 ± 3.30a 6.70 ± 4.75a Total phosphorus (mg P/g soil) 0.38 ± 0.28a0.91 ± 1.27b 0.53 ± 0.31ab Values represent the mean ± standard deviation. *Categories with similar letter s were not significantly differe nt (Fisher's LSD pair wise comparison, =0.05). nitrogen were significantly different between the LDI groups at the less strict = 0.10 level. Water color, water nitrate/nitrite-n itrogen, water TKN, soil organic matter, and soil TKN were not significantly different between LDI groups. Environmental variables with a VIF greater than 10.0 and a tolerance less than 0.10, including so il organic matter and soil TKN, were excluded from further use to avoid issues with multicollinearity ( ter Braak 1987 ; Pan and Stevenson 1996 ; Ott and Longnecker 2001 ). Diatoms Statewide 50 wetlands were sampled with 214 diatom taxa identified at the species level or lower. Diatoms identified at the species level represented 98% of the sample. Five diatom species were identified at 50% or more of the sample wetlands (n 25) including Pinnularia subcapitata (at 66% of the wetlands), Eunotia bilunaris (60%), Nitzschia palea debilis (60%), Eunotia incisa (54%), and Gomphonema gracile (50%).

PAGE 81

66 Table 3-2. Water and soil parameters among LDI groups. Low LDIHigh LDI W^p` Water parameters Dissolved oxygen (mg/L) 2.8 ± 1.71.8 ± 1.0 1328.50.00* Temperature (ºC) 26.2 ± 2.724.9 ± 2.2 1242.00.06 * Color (PCU) 272 ± 180270 ± 178 1292.50.22 * Turbidity (NTU) 3.7 ± 4.113.5 ± 28.4 953.10.02* pH 5.4 ± 1.36.2 ± 0.9 797.50.00* Specific conductivity (umhos/cm) 117 ± 129177 ± 154 120.00.05 * Ammonia-nitrogen (mg N/L) 0.15 ± 0.320.26 ± 0.43 1007.50.07 * Nitrate/nitrite-nitrogen (mg N/L) 0.08 ± 0.360.01 ± 0.03 837.50.78 * TKN (mg N/L) 1.89 ± 1.212.45 ± 1.77 1155.50.81 * TP (mg P/L) 0.08 ± 0.110.50 ± 0.96 803.50.00* Soil parameters Moisture (%) 59 ± 2051 ± 36 2866.50.02* Organic matter (%) 39 ± 2536 ± 24 2588.00.50 * TKN (mg N/g soil) 6.49 ± 3.696.25 ± 4.15 2534.00.60 * TP (mg P/g soil) 0.36 ± 0.280.73 ± 0.94 1808.00.00* ^W = Mann-Whitney U-test statistic. `p = significance value. The 3 diatoms identified most often included Eunotia naegelii , Eunotia incisa , and Nitzschia palea debilis . Of the diatoms encountered, 94 taxa (44%) occurred at a minimum of 5% of the sample wetlands (n 3). Forty-one percent of the taxa identified (87 taxa) were encountered in only one wetland. In the panhandle ecoregi on, 10 wetlands were sampled with 4 reference, 4 agricultural, and 2 urban wetla nds hosting 73 diatom taxa. In the north ecoregion 10 wetlands were sampled (4 reference, 2 ag ricultural, and 4 ur ban) with 94 taxa encountered. The central ecoregion included 13 wetlands (5 reference, 4 agricultural, and 4 urban) with 112 taxa sampled. The s outh ecoregion had 17 sample wetlands (6 reference, 5 agricultural, and 6 urban) with 147 taxa identified. Summary Statistics Richness (R), evenness (E), Shannon divers ity (H), and Simpson’s index (S) were calculated for each sample wetland ( Appendix F ). Table 3-3 summarizes the

PAGE 82

67 Table 3-3. Diatom richness, evenness, and diversity among a priori land use categories. ReferenceAgriculturalUrban Species richness (R) 19 ± 8a19 ± 6a22 ± 8a Species evenness (E) 0.74 ± 0.09a0.75 ± 0.08a0.73 ± 0.10aShannon diversity (H) 2.13 ± 0.49a2.18 ± 0.34a2.21 ± 0.49aSimpson's index (S) 0.80 ± 0.09a0.82 ± 0.07a0.81 ± 0.10aBeta diversity 6.96.15.8 Gamma diversity 132117126 Categories with similar letters were not si gnificantly different (Fisher's LSD pair wise comparison, = 0.05). richness, evenness, and diversity calcul ations of the diatom assemblage by a priori land use category. Species richness ranged from 9 taxa at CR6 and CU1 to 39 taxa at CU3 (surrounded by commercial and residential la nd uses). Species evenness ranged from 0.57 at NU5 (an urban wetland), to 0.89 at CA 6 (surrounded by citrus groves). Shannon diversity ranged from 1.41 at NU5 to 2.95 at SR5 (surrounded by marsh and flooded flatwoods). Simpson’s index was highest at SR5 and SU2 at 0.93, and lowest at NU5 at 0.58. Richness, evenness, Shannon diversity, an d Simpson’s index were not significantly different among the 3 a priori land use categories (Table 3-3) or between LDI groups (Table 3-4). Beta and gamma diversity were similar among a priori land use categories. Beta and gamma diversity were higher for the high LDI group (beta diversity of 8.2 and gamma diversity of 167) and lower for th e low LDI group (beta diversity of 7.5 and gamma diversity of 145). Compositional Analysis MRPP was calculated across a ll groups (panhandle versus north versus central versus south) as well as for multiple pair wise comparisons (panhandle versus north, panhandle versus central, panhandle versus sout h, north versus central, north versus

PAGE 83

68 Table 3-4. Mean diatom summar y statistics between LDI groups. Low LDI High LDI W^ p` Species richness (R) 19 ± 8 20 ± 7 484 0.61 Species evenness (E) 0.73 ± 0.09 0.74 ± 0.09 497 0.80 Shannon diversity (H) 2.13 ± 0.47 2.20 ± 0.43 486 0.64 Simpson's index (S) 0.80 ± 0.09 0.82 ± 0.08 480 0.55 Beta diversity 7.5 8.2 Gamma diversity 145 167 ^W = Mann-Whitney U-test statistic. `p = significance value. south, and central versus south) in order to te st the similarity of diatom taxa composition among the ecoregions. Table 3-5 shows the resu lts for the MRPP tests, including the test statistic (T), the chance-corrected within-g roup agreement (A, a measure of within group similarity), and the significance value (p). The global comparison among all wetlands and the 3 a priori categories showed that diatom community composition at the species level was significantly different ( = 0.05). Within the pair wise comparisons, only the panhandle versus south and north versus sout h comparisons had signi ficantly different diatom community compositi on for all land use types. In the reference wetlands, the south ecore gion had a significantly different diatom community composition compared to both the panhandle and north ecoregions. Similarly, diatom community composition among pair wise comparisons of agricultural wetlands was significant differe nt for the panhandle versus south ecoregions. The only ecoregions with significantly different diatom community composition among urban wetlands were the north and south ecoregions. Community Composition Figure 3-1 shows a 2 dimensional bi-plot of the NMS axes used to explore diatom community composition with overlays of significant environmental variables

PAGE 84

69 Table 3-5. Similarity of diatom community composition using MRPP. Sites (n) T^ A` p# All wetlands All regions (P vs N vs C vs S) 50 -2.20.01 0.03* Panhandle vs north 20 0.5-0.19 0.67 Panhandle vs central 23 0.9-0.03 0.81 Panhandle vs south 27 -2.50.06 0.02* North vs central 23 -0.70.02 0.20 North vs south 27 -3.90.09 0.00* Central vs south 30 -1.70.04 0.06 Reference wetlands All regions (P vs N vs C vs S) 18 -1.30.09 0.11 Panhandle vs north 8 -0.70.07 0.21 Panhandle vs central 8 0.8-0.07 0.76 Panhandle vs south 10 -2.10.16 0.03* North vs central 8 0.9-0.08 0.82 North vs south 10 -2.40.17 0.02* Central vs south 10 0.3-0.02 0.58 Agricultural wetlands All regions (P vs N vs C vs S) 16 0.1-0.01 0.52 Panhandle vs north 6 0.6-0.06 0.70 Panhandle vs central 9 0.1-0.01 0.51 Panhandle vs south 9 -2.40.16 0.02* North vs central 7 1.5-0.19 0.93 North vs south 7 0.2-0.02 0.52 Central vs south 10 -0.30.02 0.38 Urban wetlands All regions (P vs N vs C vs S) 16 -0.70.05 0.23 Panhandle vs north 6 -0.90.15 0.18 Panhandle vs central 6 0.9-0.07 0.83 Panhandle vs south 8 -0.30.02 0.37 North vs central 8 -0.10.01 0.37 North vs south 10 -1.90.11 0.05* Central vs south 10 1.1-0.05 0.86 *A high |T| value and significant p-valu e (p<0.05) suggests a difference in species composition ^ T = the MRPP test statistic `A = the chance correcte d within-group agreement #p = the significance value

PAGE 85

70 LDI log(DO) log(Spec.Cond.) Water pH log(Turbidity) log(Water TKN) log(Water TP) Axis 1: 12.4%Axis 2: 35.5% A Priori Land Use Reference Agricultural Urban Figure 3-1. NMS ordination bi -plot of 50 wetlands in diat om species space with an overlay of environmental paramete rs. LDI, dissolved oxygen (DO), turbidity, water pH, specific conducti vity (Spec.Cond.), water TKN, and water TP (shown as radiating vectors), were significantly correlated with the NMS axes based on diatom community composition. The length of the vector represents the strength of the correlation, and the angle represents the direction of maximum change. Axis 1 explained 12.4% variance, axis 2 explained 35.5% variance, and axis 3 ( not shown) represented an additional 26.7% variance.

PAGE 86

71 Table 3-6. Pearson correlations between e nvironmental parameters and NMS ordination axes based on diatom community composition. Axis 1 Axis 2 Axis 3 Incremental r2 12.4% 35.5% 26.7% Cumulative r2 12.4% 47.9% 74.6% Latitude 0.06 0.12 0.16 Longitude 0.03 0.05 0.04 LDI 0.35 0.18 0.05 Log (DO) 0.33 0.08 0.09 Log (Temperature) 0.03 0.08 0.07 Log(Color) 0.15 0.09 0.17 Log(Turbidity) 0.22 0.01 0.00 pH 0.06 0.47 0.45 Log(Spec.Cond.) 0.09 0.27 0.06 Log(Ammonia-N) 0.13 0.00 0.03 Log (Nitrate/nitrite-N) 0.01 0.01 0.05 Log(TKN) 0.23 0.00 0.01 Log(Water TP) 0.33 0.08 0.00 Arcsin Sqrt (%Soil moisture) 0.11 0.00 0.08 Log(Soil TP) 0.07 0.05 0.02 (transformed), including LDI and 6 water parameters: log(DO), log(turbidity), pH, log(Spec.Cond.), log(TKN), and log(TP). Ta ble 3-6 provides the PearsonÂ’s r-squared correlation coefficients between the envir onmental variables and the NMS ordination axes. A 3 dimensional solution was construc ted with an overall stress of 16.3 and a final stability of 0.00001, which is borderline high bu t an acceptable stress limit for a useful ordination with community data ( Kruskal 1964 ; Clarke 1993 ; McCune and Grace 2002 ). Axis 1 explained 12.4% of the variance and wa s correlated with LDI, log(DO), log(water TKN), and log(water TP). Axis 2 explaine d 35.5% variance and was correlated with pH and log(Spec.Cond). Axis 3 explained an additional 26.7% variance and was correlated with water column pH.

PAGE 87

72 Metric Selection Twenty-eight of the candidate metrics were significantly correlated with LDI (p < 0.05). Due to the redundant nature of some candidate metrics and strong correlations among metrics (PearsonÂ’s r2 > 0.9), 7 metrics which were significant for both the SpearmanÂ’s correlation coefficient (|r| > 0.45, p < 0.01) and the Mann-Whitney U-test between LDI groups (p < 0.10) were selected fo r inclusion in the diatom WCI. Table 3-7 provides the SpearmanÂ’s correlation valu es between the 7 metrics and LDI. Metrics selected for inclusion represented 2 of the metric categories, including tolerance metrics and autecological metrics. Tolerance metrics included percent tolerant and sensitive indicator species. The 5 aut ecological metrics included pollution class 1 (very tolerant to pollution) , nitrogen uptake metabolism class 3 (need periodically elevated concentrations of organically bound nitrogen), saprobity class 4 (inhabit aquatic environments with an oxygen saturati on between 10-25% and a biological oxygen demand of approximately 13-22 mg/L), pH cla ss 3 (circumneutral, mainly occurring at pH values around 7), and dissolved oxygen cl ass 1 (requiring continuously high dissolved oxygen concentrations near 100%). Pollution class was established by Bahls ( 1993 ), and nitrogen metabolism, saprobity, pH, and dissolved oxygen classes were defined by Table 3-7. SpearmanÂ’s correlati ons for 7 diatom metrics and LDI. All correlations were significant (p < 0.01). Diatom Metrics SpearmanÂ’s r Tolerant indicator species 0.65 Sensitive indicator species -0.60 Pollution class 1 0.52 Nitrogen class 3 0.48 Saprobity class 4 0.48 pH class 3 0.48 Dissolved oxygen class 1 -0.46

PAGE 88

73 Table 3-8. Comparisons among diatom metric s and the diatom WCI for LDI groups. Metric Low LDIHigh LDIW^ p` Tolerant indicator species 1.4 ± 2.535.5 ± 15.8309.0 <0.001 Sensitive indicator species 36.6 ± 25.018.0 ± 18.5719.0 <0.001 Pollution class 1 3.7 ± 6.07.6 ± 10.1337.0 <0.001 Nitrogen class 3 7.7 ± 15.520.3 ± 21.5361.0 0.003 Saprobity class 4 2.7 ± 4.525.4 ± 25.6373.5 0.006 pH class 3 18.3 ± 22.214.1 ± 16.6344.0 0.001 Dissolved oxygen class 1 69.3 ± 24.339.9 ± 23.8658.0 0.004 Diatom WCI 55.3 ± 11.145.4 ± 27.1718.0 <0.001 Values represent the mean ± standard deviation. ^W = the Mann-Whitney U-Test statistic `p = the significance value. van Dam et al. ( 1994 ). Tolerant indicator species, pollution class 1, nitrogen metabolism class 3, saprobity class 4, and pH class 3 increased with increasing development intensity; whereas, sensitive i ndicator genera and dissolved oxygen class 1 decreased with increasing development intensity. Tabl e 3-8 shows all of the selected metrics differentiated between the 2 LDI groups. Tolerance metrics Table 3-9 shows the results of the iterative Indicato r Species Analysis (ISA) calculations using species-leve l abundance data for the diatom assemblage. Tolerant diatom indicator species were established at an LDI break of 5.0, and included 12 species representing 6 genera. Table 3-10 lists the tolerant diatom indicator species. The 3 tolerant indicator species w ith the highest indicator values were all in the genera Navicula , including N. minima , N. confervacea , and N. mutica . Figure 3-2 shows the percent tolerant di atom indicator species increased with increasing landscape development intensity. We tlands with the highe st percent tolerant indicator species were in the low LDI group, including CR5 and SU3. While CR5 was

PAGE 89

74 Table 3-9. Spearman correlations of diatom in dicator species over a range of LDI values. Highlighted areas indicate the LDI valu e selected for sensitive and tolerant diatom indicator species. Low LDI High LDI Sensitive Tolerant LDI n* = n* = r^ p` r^ p` No. Sensitive Indicators No. Tolerant Indicators 1 8 42 -0.48 0.000 0.58 0.000 14 3 1.25 16 34 -0.60 0.000 0.59 0.000 18 8 1.5 19 31 -0.57 0.000 0.58 0.000 14 8 1.75 20 30 -0.57 0.000 0.62 0.000 12 7 2 20 30 -0.57 0.000 0.62 0.000 12 7 2.25 21 29 -0.56 0.000 0.63 0.000 11 9 2.5 21 29 -0.56 0.000 0.63 0.000 11 9 2.75 21 29 -0.56 0.000 0.63 0.000 11 9 3 21 29 -0.56 0.000 0.63 0.000 11 9 3.25 22 28 -0.61 0.000 0.63 0.000 10 10 3.5 22 28 -0.61 0.000 0.63 0.000 10 10 3.75 22 28 -0.61 0.000 0.63 0.000 10 10 4 24 26 -0.56 0.000 0.56 0.000 6 13 4.25 25 25 -0.57 0.000 0.57 0.000 6 8 4.5 26 24 -0.48 0.000 0.56 0.000 5 12 4.75 30 20 -0.55 0.000 0.60 0.000 5 17 5 35 15 -0.52 0.000 0.65 0.000 2 12 5.25 39 11 x x 0.59 0.000 0 11 5.5 41 9 -0.41 0.003 0.59 0.000 1 14 5.75 42 8 -0.45 0.001 0.54 0.000 1 13 6 42 8 -0.45 0.001 0.54 0.000 1 13 6.25 45 5 x x 0.55 0.000 0 21 6.5 47 3 x x 0.41 0.003 0 13 6.75 47 3 x x 0.41 0.003 0 13 7 48 2 x x 0.45 0.001 0 23 *n = number of sites ^r = SpearmanÂ’s r correlation coefficien t of indicator species versus LDI `p = significance value categorized a reference wetland, it was located in a fragmented state park within a small, highly developed county (Seminole County n ear Orlando, Florida). SU3 was an urban wetland surrounded by nearly 100 m of marsh that received nutrient enriched water. The 2 wetlands with the highest percent tolerant indicator species in the high LDI group

PAGE 90

75 Table 3-10. Diatom tolerant indicator species. All reporte d tolerant indicator species calculated at an LDI break of 5.0 were significant (p < 0.10). Indicator Species (LDI>5.0) Indicator Value pvalue Cyclotella pseudostelliger 20.0 0.021 Diploneis elliptica 19.7 0.045 Navicula confervacea 45.3 0.006 Navicula minima 48.0 0.050 Navicula mutica 40.5 0.044 Navicula recens 12.9 0.100 Navicula subminuscula 12.5 0.083 Neidium alpinum 20.0 0.021 Nitzschia subacicularis 18.1 0.080 Pinnularia braunii 22.6 0.043 Pinnularia divergentissima 13.3 0.085 Stauroneis kriegeri 13.3 0.085 0 10 20 30 40 50 60 70 02468LDI%Tolerant Indicator Specie s Figure 3-2. Percent diatom tolerant indi cator species increased with increasing development intensity (LDI).

PAGE 91

76 included SU1 and CU5. Four agricultural a nd urban wetlands and 10 reference wetlands had no tolerant diatom indicator species present. Sensitive diatom indicator species were selected at an LDI break of 1.25, correlating to a break in the natural and developed land us es. Table 3-11 lists the 18 statewide sensitive indicator species. The 5 sensitive indicator species with the highest indicator values included Eunotia naegelii , E. rhomboidea , Frustulia rhomboides , Anomoeoneis brachysira , and Desmogonium rabenhorstianum . Figure 3-3 shows that the percent sensitive indicator species decreased wi th increasing development intensity in the surrounding landscape. Eighty-th ree percent of the high LDI wetlands hosted less than 10% sensitive indicator species. Of these 5 sites with greater than 10% sensitive Table 3-11. Diatom sensitive indicator species . All reported sensit ive indicator species calculated at an LDI break of 1.25 were significant (p<0.10). Indicator Species (LDI<1.25) Indicator Value pvalue Anomoeoneis brachysira 39.2 0.008 Cymbella microcephala 12.5 0.097 Desmogonium rabenhorstianum 33.5 0.010 Encyonema silesiacum 24.4 0.100 Eunotia flexuosa 12.5 0.094 Eunotia glacialis 17.0 0.035 Eunotia intermedia 28.1 0.016 Eunotia naegelii 59.2 0.002 Eunotia pectinalis undulata 26.9 0.038 Eunotia rhomboidea 45.9 0.013 Frustulia rhomboides 41.6 0.069 Frustulia rhomboides crassinervia 18.7 0.038 Navicula capitatoradiata 11.5 0.091 Navicula subtilissima 12.5 0.079 Nitzschia nana 17.0 0.064 Nitzschia paleacea 18.7 0.034 Pinnularia streptoraphe 18.7 0.031 Rhopalodia gibba 18.7 0.031

PAGE 92

77 0 10 20 30 40 50 60 70 80 90 02468LDI%Sensitive Indicator Species Figure 3-3. Percent diatom sensitive indicator species decreased with increasing development intensity (LDI). indicator species, 2 were agricultural wetla nds embedded in pasture including PA2 and CA4, and 3 were set in recently develope d urban landscapes including CU1, NU6, and SU6. Six agricultural and 2 urban wetland s hosted no sensitive indicator species. Autecological metrics Between 56-69% of diatoms identified r eceived scores based on established autecological relationships ( van Dam et al. 1994 ; Bahls 1993 ). Five metrics based on scoring diatoms from a coded checklist descri bing their autecology were incorporated in the diatom WCI, including the proportion of di atoms in pollution class 1, nitrogen uptake metabolism class 3, saprobity class 4, pH cl ass 3, and dissolved oxygen requirement class 1. Figure 3-4 shows that the proportion of di atoms in pollution cla ss 1 increased with increasing development intensity in the su rrounding landscape. Diatoms in pollution class 1 were very tolerant to pollution, as compared to pollution class 2 (moderately

PAGE 93

78 0 20 40 60 80 100 02468LDI%Pollution Class 1 Figure 3-4. Pollution toleran ce class 1 diatoms increased with increasing development intensity (LDI). tolerant) or pollution class 3 (sensitive to pollution; Bahls 1993 ). A background level of approximately 10% of the diatoms belonging to pollution class 1 distinguishes low and high LDI wetlands. Two exceptions in the low LDI group include CR5 (pollution class 1 = 13%) and SR5 (pollution class 1 = 23%). The wetland with the greatest percent of diatoms in pollution class 1 was CA3 (LDI 4.9, pollution class 1 = 88%), a wetland that received waters carrying wastes from a pullet farm operation. Figure 3-5 shows that the proportion of diatoms in nitrogen uptake metabolism class 3 increased with increas ing development intensity in the surrounding landscape. Membership in nitrogen uptake metabolism cl ass 3 was defined by facultative nitrogenheterotrophic taxa that need periodically elevated concen trations of organically bound nitrogen ( van Dam et al. 1994 ). Eighty percent of the low LDI group wetlands had less than 10% of the diatoms in nitrogen uptake me tabolism class 3. Four outliers in the low

PAGE 94

79 0 20 40 60 80 02468LDI%Nitrogen Class 3 Figure 3-5. Nitrogen uptake metabolism cl ass 3 diatoms increased with increasing development intensity (LDI). LDI group included SU3, SR6, SR4, and SR5. These 3 southern reference wetland outliers were within state or federal lands pr otected as part of th e Florida Everglades. The urban outlier, SU3 was surrounded by n early 100 m of marsh that has received nutrient enriched waters since the mid 1970s. The percent of diatoms in saprobity cla ss 4 increased with increasing landscape development intensity (Figure 3-6). Diatoms characterized as belonging to saprobity class 4 included mesoto poly-saprobous speci es (inhabit aquatic environments with an oxygen saturation between 10-25 % and a biological oxygen demand (BOD5 20) of 13-22 mg/L) ( van Dam et al. 1994 ). Eighty-five percent of th e wetlands in the low LDI group had less than 5% of diatoms in saprobity cl ass 4. The 3 low LDI wetlands with over the 5% threshold included SU3, PR1, and SR6. Ov er 50% of the wetlands in the high LDI

PAGE 95

80 0 20 40 60 02468LDI%Saprobity Class 4 Figure 3-6. Saprobity class 4 diatoms increased with incr easing development intensity (LDI). group had greater than 5% of diatoms in sa probity class 4. CA3 hosted the greatest percent of diatoms in saprobity cl ass 4 (saprobity class 4 = 52%). Figure 3-7 shows that the percent of diatoms in pH class 3 increased with increasing landscape development intensity. Di atoms in pH class 3 were described as circumneutral (mainly occurring at pH values of approximately 7) ( van Dam et al. 1994 ). Of the wetlands in the low LDI group, 70% ha d less than 20% of pH class 3 diatoms; whereas, 73% of wetlands in the high LDI gr oup had greater than 20% of diatoms in pH class 3. In the low LDI group, the greates t outlier was SU4 (pH class 3 = 89%). Diatoms requiring continuously high dissolved oxygen concentrations of approximately 100% saturation (dissolved oxygen class 1) decreased with increasing development intensity in the surrounding lands cape (Figure 3-8). In the low LDI group, the same 6 outliers occurred as in the pH class 3 metric; with SU3 having the lowest

PAGE 96

81 0 20 40 60 80 100 02468LDI%pH Class 3 Figure 3-7. The pH class 3 diatoms increas ed with increasing development intensity (LDI). 0 20 40 60 80 100 02468LDI%Dissolved Oxygen Class 1 Figure 3-8. Dissolved oxygen class 1 diatoms decreased with increasing development intensity (LDI).

PAGE 97

82 percent diatoms in dissolved oxygen class 1 (1 8%). Seventy percen t of wetlands in the low LDI group had greater than 60% di atoms in dissolved oxygen class 1. Diatom Wetland Condition Index The seven metrics described above were sc ored and added togeth er to create the diatom WCI. Figure 3-9 shows the relati onship between the diatom WCI and LDI. Potential scores for the diatom WCI ranged from 0-70, with higher values representing wetlands surrounded by undeveloped landscapes. Actual scores ranged from 8 at CA3 (an agricultural wetland receiving inputs from a spray field associated with pullet farm wastes), to 69 at SR2 (a wetland surrounded by flooded flatwoods and marsh). The next highest scoring wetlands receiv ed diatom WCI scores 2 points lower than SR2, with a score of 67 at both NR3 and SR1. Diatom WCI ranges varied regionally, with the highest scores in each region including PR4 (65), NR3 (67), CR6 (65), and SR2 (69), in the panha ndle, north, central, and south ecoregions, respectively. The lo west scores in the panhandle and north ecoregions were for urban wetlands embedded in residential land use, including PU4 (11) and NU2 (24). Wetlands surrounded by agricultur al land uses received the lowest scores in the central and south ecoregions, includi ng CA3 (8) and SA4 (16). The diatom WCI was robustly correlated with the LDI index (Spearman correlation |r| = 0.64, p < 0.001). A Kruskal-Wallis test between median di atom WCI values suggested a significant difference (H = 20.7, p < 0.001) among wetlands in the 3 a priori land use categories. Cluster Analysis Cluster analysis determined 4 categories based on diatom community composition. Using site descriptions, cluste rs were explained by regions, a priori land use categories, and water level including: 1: wetlands in th e panhandle to central ecoregions with low

PAGE 98

83 0 10 20 30 40 50 60 70 02468 LDIDiatom WC I Reference Agricultural Urban Figure 3-9. Diatom WCI scores decreased with increasing development intensity (LDI). Sample wetlands are designated by a priori land use category: reference, agricultural, or urban. development intensity; 2: wetlands occurri ng in mixed ecoregions with low development intensity; 3: wetlands within the southern Everglades ; and 4: wetlands within mixed regions surrounded by high development intens ity. Figure 3-10 shows that based on the diatom WCI scores, clusters 1 and 2 were not significantly different from one another, but were significantly different from both cluster 3 and cluste r 4 (p<0.05). Clusters 3 and 4 were significantly different from each othe r. Table 3-12 provides means and standard deviations for D-WCI scores and LDI of the four diatom based clusters. Cluster 4 had significantly different D-WCI and LDI scores than all other clusters. Macrophytes Statewide, 118 wetlands were sampled with 605 species, representing 323 genera and 126 families identified. The most abundant species was Taxodium ascendens , which

PAGE 99

84 0 10 20 30 40 50 60 70 Cluster 1 D-WCI Cluster 4 Cluster 3 Cluster 2 a a b c Figure 3-10. Diatom WCI scores for we tland clusters based on diatom community composition. Boxes represent the inter quartile range; solid circles represent the mean; middle lines represent the me dian; whiskers represent the range; asterisks represent outliers (> ±2 standa rd deviations). Clusters with similar letters were not significantly different (Fisher’s LSD, p<0.05). was found rooted within the vegetation quadrats at 93% of the study wetlands. The second most abundant species was Myrica cerifera found in 64% of the study wetlands.wetlands. The most common fern was Woodwardia virginica found at 53% of the wetlands; the most common vine was Toxicodendron radicans also found at 53% of the wetlands; and the most common graminoid was Panicum hemitomon found at 50% of the wetlands. Of the species encountered, 130 species (22%) occurred at a minimum of 5% of the sample wetlands (n 6). Approximately one-third of the species identified (202 species or 33.5%) were rooted in th e vegetation quadrats at only one wetland. In the panhandle ecore gion, 28 wetlands were sampled hosting 328 species, representing 191 genera and 90 families. In the north ecoregion 31 wetlands were

PAGE 100

85 Table 3-12. Diatom WCI and LDI values for wetland clusters based on diatom community composition. Cluster* 1 2 3 4 Diatom WCI 57.6 ± 6.4a 52.9 ± 11.6a 38.1 ± 4.4b 24.1 ± 11.7c LDI 2.9 ± 2.1a 2.8 ± 2.0a 2.4 ± 1.9a 5.3± 1.4b * Clusters with similar letters were not significantly different (Fisher’s LSD, p<0.05). sampled with 306 species (180 genera and 89 families) encountered. The central ecoregion included 31 wetlands with 329 species (202 genera and 94 families) sampled. The south ecoregion had 28 sample wetland s with 266 species (in 180 genera and 89 families) identified. Summary Statistics Species richness (R), first and second order jackknife estimators (Jack1 and Jack2, respectively), species evenness (E), Shannon di versity (H), and Whitta ker’s beta diversity ( W) were calculated based on the macrophyte assemblage for each sample wetland ( Appendix F ). Species richness ranged from 13 species at NA3 (embedded in silvicultural land use), to 77 species at NA12 (surrounded with pasture and row crops). The greatest estimates of species richness we re 99 species and 114 species at NA12, for first and second order jackknife estimators, respectively. Sampled species richness at NA12 was 77 species. The lowest estimates of actual species richness were for NR1, with 11 and 15 species estimated with firs t and second order jackknife estimators, respectively. Sampled species richness for NR1 was 14 species. Species evenness ranged from 0.71 at PR7 (a large, deep water wetland on a private c onservation tract), to 0.93 at PU4 (a wetland surrounded by a resi dential community and park). Shannon diversity ranged from 1.8 to 3.9 at tw o agricultural wetlands, NA3 and NA12,

PAGE 101

86 respectively, similar to species richness. WhittakerÂ’s beta diversity ranged from a low of 0.2 at PU9 (an urban wetland surrounded by resi dential land use), to a high of 9.4 at CR5 (a deep water wetland on state land). Table 3-13 summarizes compar isons of mean richness and diversity calculations by a priori land use category. Agricultural wetl ands had the greatest species richness followed closely by urban wetlands. This same trend was evident for species evenness, with agricultural wetlands having greater sp ecies evenness. Diversity indices yielded similar results, with reference wetlands having lower Shannon diversity and WhittakerÂ’s beta diversity then both agricu ltural and urban wetlands. Be ta and gamma diversity were calculated for a priori land use categories, with urban wetlands having the highest beta diversity and agricultural wetla nds the highest gamma diversity. Only WhitakerÂ’s beta diversity was significantly different among the a priori land use categories (FisherÂ’s LSD pair wise comparison, = 0.05). Species richness and WhittakerÂ’s beta diversity were not significantly different between low (LDI < 2.0) and high (LDI 2.0) LDI groups (Table 3-14); whereas, species evenness a nd Shannon diversity were significantly different (p < 0.05) between LDI groups (Mann-Whitney U-Test). Compositional Analysis MRPP was calculated across a ll groups (panhandle versus north versus central versus south) as well as for multiple pair wise comparisons (panhandle versus north, panhandle versus central, panhandle versus so uth, north versus central, north versus south, and central versus south). Table 315 shows the results for the MRPP tests, including the test statistic (T), chance-c orrected within-group agreement (A), and significance value (p). Only 2 of the MRPP comparisons (agricultural wetlands in the panhandle and north ecoregions and central and south ecoregions) were not significant at

PAGE 102

87 Table 3-13. Mean macrophyte richne ss, evenness, and diversity among a priori land use categories. Reference Agricultural Urban Species richness (R) 32 ± 11a37 ± 14a36 ± 10a Species evenness (E) 0.85 ± 0.04a0.87 ± 0.04a0.86 ± 0.03aShannon diversity (H) 2.9 ± 0.3a3.1 ± 0.4a3.1 ± 0.3aWhittaker's Beta diversity ( W) 4.0 ± 2.9a4.8 ± 1.7b4.3 ± 1.7abBeta diversity 9.510.410.5 Gamma diversity 304383378 Categories with similar letters were not significantly different (Fisher's LSD, =0.05). the = 0.05 level, suggesting that there were regionally si gnificant differences among species composition across all regions and within a priori land use categories. Community Composition Macrophyte community composition was su mmarized in 2 NMS ordinations to relate changes in macrophyte community co mposition with environmental variables. Figure 3-11 shows a two dimensional bi-plot of the NMS axes used to explore the dissimilarities of macrophyte community co mposition with overlays of significant environmental variables, including log(soil TP ), LDI, latitude, and longitude. Table 3-16 provides the Pearson correlation coefficients between environmental variables and NMS Table 3-14. Mean macrophyte richness, eve nness, and diversity between LDI groups. Low LDI High LDI W^ p` Species richness (R . ) 33 ± 10 37 ± 12 2120 0.07 * Species evenness (E) 0.85 ± 0.04 0.87 ± 0.04 2028 0.02* Shannon diversity (H) 2.9 ± 0.3 3.1 ± 0.4 2062 0.03* Whittaker's Beta diversity ( W) 4.1 ± 1.9 4.5 ± 1.7 2131 0.08 * Beta diversity 10.2 13.8 Gamma diversity 338 510 * Indicates significance at <0.05 ^W = Mann-Whitney U-test statistic `p = significance value

PAGE 103

88 Table 3-15. Macrophyte community compos ition similarity among ecoregions with MRPP. Sites (n) T^ A` p# All wetlands All regions (P vs N vs C vs S) 118 -26.80.06 0.00* Panhandle vs north 59 -7.10.03 0.00* Panhandle vs central 59 -13.10.04 0.00* Panhandle vs south 56 -23.60.09 0.00* North vs central 62 -7.90.03 0.00* North vs south 59 -24.20.09 0.00* Central vs south 59 -11.80.04 0.00* Reference wetlands All regions (P vs N vs C vs S) 37 -12.50.12 0.00* Panhandle vs north 17 -5.20.08 0.00* Panhandle vs central 19 -6.80.09 0.00* Panhandle vs south 17 -8.10.14 0.00* North vs central 20 -4.70.06 0.00* North vs south 18 -8.50.14 0.00* Central vs south 20 -7.10.07 0.00* Agricultural wetlands All regions (P vs N vs C vs S) 40 -6.60.05 0.00* Panhandle vs north 22 -0.70.01 0.21 * Panhandle vs central 19 -2.80.03 0.01* Panhandle vs south 19 -8.10.10 0.00* North vs central 21 -2.00.02 0.04* North vs south 21 -7.70.09 0.00* Central vs south 18 -1.40.02 0.09 * Urban wetlands All regions (P vs N vs C vs S) 41 -15.30.14 0.00* Panhandle vs north 20 -6.00.07 0.00* Panhandle vs central 21 -9.00.10 0.00* Panhandle vs south 20 -11.10.17 0.00* North vs central 21 -3.30.03 0.00* North vs south 20 -10.50.15 0.00* Central vs south 21 -8.60.09 0.00* *A high |T| value and significant p-valu e (p<0.05) suggests a difference in species composition. ^ T = the MRPP test statistic `A = the chance correcte d within-group agreement #p = the significance value.

PAGE 104

89 ordination axes. A 3 dimensional solution wa s constructed with an overall stress of 20.8 with a final stability of 0.06, which is a fairly high stress limit but considered useful for ordinations with community data sets ( Kruskal 1964 ; Clarke 1993 ; McCune and Grace 2002 ). Axis 1 explained 29.1% variance and wa s correlated with la titude and longitude; axis 2 explained 34.2% variance and was correla ted with LDI and log(soil TP). Axis 3 explained an additional 12.1% of the va riance and was not correlated with soil parameters. A second NMS ordination was comple ted using the macrophyte species composition at the 75 sample wetlands with measured water parameters. Figure 3-12 shows the bi-plot from the NMS ordinati on. The final stress was 16.6 with a final instability of 0.004. The ordi nation explained a cumulative 77.4% of the variance in wetland macrophyte community composition. Table 3-17 shows the Pearson r-squared correlation coefficient values for the envir onmental parameters and the three ordination axes. Axis 1 explained 31.3% of the vari ance and was correlated with latitude and longitude. Axis 2 was correlated with LDI, water pH, log(water TP), log(soil TP), and log(DO), and explained 25.7% of the variance. Axis 3 explained an additional 20.4% of the variance and was correlated with water pH and Arcsin Sqrt (soil moisture). log(Water N) concentration (ammonia, nitr ate/nitrite, and TKN) , log(Temperature), log(color), and Log(turbidity) were not strong ly correlated with the NMS ordination axes. Metric Selection Over 35 of the candidate metr ics were significantly correlated with LDI. Due to the redundant nature of some candidate metrics and the multiple forms of calculations (number, percent, proportion, frequency of o ccurrence), 6 metrics th at were significant for both the SpearmanÂ’s correlation coeffi cient (|r|>0.50, p<0.001) and the Mann

PAGE 105

90 Figure 3-11. NMS ordination bi-plot of 118 sample wetlands in macrophyte species space with an overlay of environmenta l parameters. Latitude, longitude, LDI, and log(soil TP), shown as radi ating vectors, were significantly correlated with NMS axes. Vector length represents the strength of the correlation, and the angle represents the direction of maximum change. Axis 1 explained 29.1% variance, axis 2 explained 34.2% variance, and axis 3 (not shown) represented an additional 12.1% variance. Phosphorus Latitude Longitude LDI Axis 1: 29.1% A xis 2: 34.2% Minimally Impaired Agriculture Impaired Urban Impaired Refence Agricultural Urban Log( )

PAGE 106

91 Table 3-16. Pearson correlations between e nvironmental variables and NMS axes based on macrophyte community composition at 118 wetlands. Axis 1 Axis 2 Axis 3 Incremental r2 29.1% 34.2% 12.1% Cumulative r2 29.1% 63.3% 75.4% Latitude 0.71 0.04 0.00 Longitude 0.42 0.02 0.00 LDI 0.01 0.46 0.01 Arcsin Sqrt (Soil moisture) 0.00 0.11 0.04 Log (Soil TKN) 0.03 0.06 0.00 Log(Soil TP) 0.03 0.20 0.00 macrophyte WCI. Table 3-18 provides the statewide Spearman correlation values Whitney U-test between LDI groups (p<0.001) we re selected for inclusion in the between the 6 macrophyte metrics and LDI. Metrics se lected for inclusion were tolerant and sensitive indicator species; modified Floristic Quality Index (FQI); exotic species; native perennial species; and wetland st atus species. Tolerant indicator and exotic species increased with increasing development intensity; whereas, sensitive indicator species, modified FQI, native perennial species, and wetland status species decreased with increased landscape development. Table 3-19 shows significant diffe rences of metrics between low and high LDI groups. Tolerance metrics Multiple ISAs were completed at different LDI breaks, starting at 1.0 and continuing through 7.0, at 0.25 step increments. Table 3-20 shows the results of the iterative ISA calculations. The greatest numbe r of statewide tolera nt indicator species was established at an LDI break of 4.0, and the greatest number of statewide sensitive indicator species was found at an LDI break of 2.0. These break points were used for successive ISA calculations.

PAGE 107

92 Reference Agricultural Urban LDI Latitude Longitude Log(DO) Water pH Log(TP) Axis 2: 25.7%Axis 1: 31.1 % Reference Agricultura l Urban Figure 3-12. NMS ordination bi-plot of 75 sample wetlands in macrophyte species space with an overlay of environmental parameters. Latitude, longitude, LDI, Log(TP), water pH, and Log(DO), s hown as radiating vectors, were significantly correlated with NMS axes. Vector lengt h represents the strength of the correlation, and the a ngle represents the direct ion of maximum change. Axis 1 explained 31.3% variance, axis 2 explained 25.7% variance, and axis 3 (not shown) represented an additional 20.4% variance.

PAGE 108

93 Table 3-17. Pearson correlations between e nvironmental variables and NMS axes based on macrophyte community composition at 75 wetlands. Axis 1 Axis 2 Axis 3 Incremental r2 31.3% 25.7% 20.4% Cumulative r2 31.3% 57.0% 77.4% Latitude 0.68 0.03 0.17 Longitude 0.46 0.03 0.06 LDI 0.03 0.35 0.11 Water parameters Log(DO) 0.02 0.20 0.05 Log(Temperature) 0.02 0.05 0.00 Log(Color) 0.14 0.00 0.05 Log(Turbidity) 0.03 0.06 0.03 pH 0.15 0.26 0.35 Log(Ammonia-N) 0.01 0.04 0.00 Log(Nitrate/nitrite-N) 0.06 0.00 0.06 Log(TKN) 0.02 0.01 0.01 Log(TP) 0.11 0.25 0.06 Soil parameters Arcsin Sqrt (Moisture) 0.00 0.00 0.24 Log(TKN) 0.00 0.00 0.16 Log(TP) 0.09 0.23 0.01 ISA calculations were determined for each of the 4 ecoregions and statewide. Table 3-21 provides a list of tolerant indicat or species comparing regional and statewide analyses. The same random seed number was used for each ISA. In total the ISA reported 69 statewide tolerant indicator speci es, and less for each ecoregions with 7, 28, 7, and 12 for the panhandle, north, central, and south ecoregions, respectively. The statewide ISA produced 69 tolerant indicator species and an additional 7 species were included on regional lists, but not the statewide list. No sp ecies occurred on the tolerant indicator lists statewide and in all 4 ecoregi ons. Three species occu rred on the statewide tolerant indicator speci es list and in 3 of the ecoregions, including Commelina diffusa (north, central, and south), Cynodon dactylon (panhandle, north, central), and Diodia

PAGE 109

94 Table 3-18. Spearman correlations between six macrophyte metrics and LDI. All correlations were significant (p < 0.001). Metric r Tolerant indicator species 0.75 Sensitive indicator species -0.66 Modified FQI -0.71 Exotic species 0.65 Native perennial species -0.63 Wetland status species -0.55 virginiana (panhandle, north, south). Seven speci es occurred on the tolerant indicator species list statewide and in 2 regions, and 24 were listed both statewide and in 1 ecoregion. Thirty-five of the 69 statewide tolerant indicat or species (51%) were not listed in any ecoregion. In total, the ecoregi ons shared more than two-thirds of their listed species with the statewide list: panhandl e (100%), north (93%), central (86%), and south (67%). Two species were unique to the north, 1 to the central, and 4 to the south ecoregion tolerant indicator species lists. Figure 3-13 shows the scatter plots of the percent tolerant indicator species versus LDI. The percent tolerant indicator species increased with increasing development intensit y. For the statewide tolerant indicator Table 3-19. Comparisons among 6 macrophyte metrics for LDI groups. Metric Low LDI High LDI W^ p` Tolerant indicator species 7.8 ± 7.831.2 ± 14.71116.5 <0.001 Sensitive indicator species 39.5 ± 16.79.4 ± 10.13665.0 <0.001 Modified FQI 4.81 ± 0.623.62 ± 0.803771.0 <0.001 Exotic species 3.0 ± 3.614.3 ± 10.61379.0 <0.001 Native perennial species 92.7 ± 4.579.7 ± 12.03453.0 <0.001 Wetland status species 72.0 ± 9.854.1 ± 12.53612.0 <0.001 Values represent the mean ± standard deviation ^W = the Mann-Whitney U-Test statistic `p = the significance value

PAGE 110

95 Table 3-20. Macrophyte ISA calcu lations were conducted over a range of LDI values. Highlighted areas indicate the LDI valu e selected for stat ewide sensitive and tolerant indicator species. Low LDI High LDI Sensitive Tolerant LDI n* = n* = r^ p` r^ p` No. Sensitive Indicators No. Tolerant Indicators 1 13 105 -0.52 <0.0001 0.46 <0.0001 34 4 1.25 33 85 -0.67 <0.0001 0.71 <0.0001 55 31 1.5 37 81 -0.66 <0.0001 0.69 <0.0001 58 39 1.75 39 79 -0.66 <0.0001 0.71 <0.0001 59 47 2 41 77 -0.66 <0.0001 0.70 <0.0001 61 43 2.25 48 70 -0.66 <0.0001 0.72 <0.0001 58 60 2.5 48 70 -0.66 <0.0001 0.72 <0.0001 58 60 2.75 49 69 -0.66 <0.0001 0.73 <0.0001 59 56 3 51 67 -0.66 <0.0001 0.73 <0.0001 57 56 3.25 55 63 -0.68 <0.0001 0.73 <0.0001 49 55 3.5 56 62 -0.68 <0.0001 0.74 <0.0001 53 52 3.75 56 62 -0.68 <0.0001 0.74 <0.0001 53 52 4 63 55 -0.65 <0.0001 0.75 <0.0001 35 69 4.25 65 53 -0.62 <0.0001 0.77 <0.0001 34 62 4.5 69 49 -0.63 <0.0001 0.76 <0.0001 32 47 4.75 73 45 -0.65 <0.0001 0.77 <0.0001 25 50 5 82 36 -0.65 <0.0001 0.74 <0.0001 17 41 5.25 91 27 -0.70 <0.0001 0.66 <0.0001 8 30 5.5 97 21 -0.67 <0.0001 0.58 <0.0001 4 17 5.75 99 19 -0.58 <0.0001 0.57 <0.0001 4 17 6 101 17 -0.63 <0.0001 0.63 <0.0001 3 18 6.25 106 12 -0.52 <0.0001 0.53 <0.0001 2 24 6.5 110 8 -0.52 <0.0001 0.56 <0.0001 2 27 6.75 113 5 xx xx 0.42 <0.0001 0 19 7 114 4 xx xx 0.14 0.1432 0 14 *n = number of sites. ^r = SpearmanÂ’s r correlation coefficien t of indicator spec ies versus LDI. `p = significance value. species, CA2 (tolerant = 72%) had the highest percent statewide tolerant indicator species. Ninety-three percent of the wetla nds in the low LDI group had less than 20% statewide tolerant indicator species. Three outliers included SA8 (s tatewide tolerant = 32%), CA8 (statewide tolerant = 26%), and PR7 (statewide tolerant = 25%). In the high

PAGE 111

96 Table 3-21. Statewide and re gional macrophyte tolerant indi cator species. Tolerant indicator species calculated at an LD I break of 4.0 were significant (p < 0.10). Statewide Panhandl North Central South No. of Tolerant Species 69 7 28 7 12 Acer rubrum (28.5, 0.04) (53.7, 0.01) Alternanthera philoxeroides (11.3, 0.002) Amaranthus spinosus (10.9, 0.01) (21.4, 0.08) Ampelopsis arborea (18.6, 0.06) Aster carolinianus (9.5, 0.05) Axonopus fissifolius (9.1, 0.02) Blechnum serrulatum (42.4, 0.07) Boehmeria cylindrica (37.9, 0.00) (57.9, 0.01) (43.7, 0.04) Carex longii (30.6, 0.00) (54.3, 0.00) Centella asiatica (52.0, 0.02) Colocasia esculenta (5.5, 0.09) Commelina diffusa (44.8, 0.00) (50.0, 0.00) (73.3, 0.00) (44.1, 0.05) Cuphea carthagenensis (28.0, 0.00) (28.6, 0.03) (50.0, 0.00) Cynodon dactylon (29.4, 0.00) (35.7, 0.05) (35.7, 0.01) (28.1, 0.09) Cyperus croceus (9.1, 0.02) Cyperus lanceolatus (7.3, 0.04) Cyperus polystachyos (13.1, 0.01) Cyperus retrorsus (18.2, 0.00) (35.7, 0.01) (20.0, 0.10) Cyperus virens (11.3, 0.02) (42.9, 0.01) Digitaria ciliaris (9.1, 0.02) Diodia virginiana (37.2, 0.00) (50.0, 0.01) (40.5, 0.03) (50.0, 0.06) Dioscorea bulbifera (7.3, 0.04) Echinochloa colona (5.5, 0.01) Eclipta prostrata (17.9, 0.01) Eupatorium capillifolium (37.9, 0.01) (44.1, 0.06) Galium hispidulum (5.5, 0.09) Galium tinctorium (22.6, 0.00) (51.8, 0.00) (26.7, 0.05) Hymenachne amplexicaulis (9.1, 0.02) (25.0, 0.08) Hypericum mutilum (28.6, 0.03) Juncus effusus (22.1, 0.00) (42.9, 0.00) Kyllinga brevifolia (7.3, 0.04) Leersia hexandra (7.3, 0.04) (20.0, 0.09) Lepidium virginicum (5.5, 0.01) Ligustrum sinense (10.2, 0.08) Lonicera japonica (12.7, 0.05) Ludwigia peruviana (17.7, 0.06) Ludwigia repens (14.3, 0.06) (21.4, 0.08) Luziola fluitans (5.5, 0.09) Lygodium japonicum (11.9, 0.03)

PAGE 112

97 Table 3-21. Continued. Statewide PanhandleNorth Central South Melaleuca quinquenervia (36.2, 0.06) Melothria pendula (20.8, 0.00) (28.6, 0.03) Micranthemum umbrosum (5.5, 0.09) Momordica charantia (10.2, 0.08) Oxalis corniculata (18.5, 0.00) (28.6, 0.04) Parthenocissus quinquefolia (34.6, 0.00) (40.5, 0.04) (52.0, 0.01) Paspalum notatum (24.4, 0.01) (30.7, 0.06) Paspalum urvillei (18.5, 0.00) (35.7, 0.04) (28.6, 0.03) Phyla nodiflora (15.2, 0.08) (32.1, 0.09) Phyllanthus urinaria (16.7, 0.00) (21.4, 0.07) Phytolacca americana (26.1, 0.01) (36.7, 0.09) Polygonum hydropiperoides (21.1, 0.05) (57.1, 0.00) Polygonum punctatum (28.0, 0.00) (42.9, 0.02) (44.7, 0.01) Polypremum procumbens (9.5, 0.06) Proserpinaca palustris (5.5, 0.09) Richardia brasiliensis (5.5, 0.09) Rubus argutus (25.1, 0.07) (47.1, 0.03) Rubus trivialis (17.7, 0.06) Sabal palmetto (66.7, 0.00) Sacciolepis indica (7.3, 0.05) Sambucus canadensis (24.6, 0.00) (30.7, 0.08) Sapium sebiferum (19.4, 0.02) Saururus cernuus (40.5, 0.05) Senna obtusifolia (9.1, 0.02) Sesbania vesicaria (5.5, 0.01) Setaria parviflora (7.3, 0.04) Sida rhombifolia (20.8, 0.00) (35.7, 0.01) Smilax pumila (7.3, 0.04) (21.4, 0.08) Solanum carolinense (9.1, 0.03) Solidago stricta (5.5, 0.09) Sporobolus indicus (7.3, 0.05) Stenotaphrum secundatum (9.5, 0.06) (26.7, 0.04) Toxicodendron radicans (38.0, 0.03) Trifolium repens (7.3, 0.04) (21.4, 0.07) Urena lobata (52.0, 0.01) Vitis rotundifolia (36.7, 0.01) (48.5, 0.05) Wedelia trilobata (5.5, 0.09) (25.0, 0.07)

PAGE 113

98 0 10 20 30 40 50 02468 LDI%North Tolerant 0 10 20 30 40 50 02468 LDI%Central Toleran t 0 10 20 30 40 50 02468 LDI%South Tolerant 0 20 40 60 80 02468 LDI%Statewide Tolerant Panhandle North Central South Figure 3-13. Tolerant macrophyte indicat or species increased with increasing development intensity (LDI). A) Pa nhandle tolerant indi cator species at panhandle study wetlands. B) North to lerant indicator species at north study wetlands. C) Central tolerant indicator species at central study wetlands. D) South tolerant indicator species at south study wetlands. E) Statewide tolerant indicator species at all study wetlands. 0 10 20 30 40 50 02468 LDI%Panhandle TolerantA.B. C. D. E.

PAGE 114

99 LDI group, 73% of the wetlands had over 20% st atewide tolerant indicator species. In the regional ISA calculations, the north ecore gion had the largest percent tolerant indicator species, with NA1 (north tolerant = 47%). Table 3-22 provides a list of statewide and re gional sensitive indi cator species. The statewide sensitive indicator sp ecies list included 61 species of which 16 were not listed in any of the ecoregions. Two species occurr ed on all statewide and regional lists, including Eriocaulon decangulare and Panicum erectifolium . Similarly, 6 species occurred on the statewide list and 3 regional lists including Andropogon virginicus , Aristida purpurascens , Ilex glabra , and Polygala cymosa (statewide, panhandle, north, and central); and Fuirena scirpoidea and Pinus elliottii (statewide, panhandle, central, and south). All 4 ecoregions shared over th ree-quarters of their species with the statewide list (panhandle = 79%, north = 84% , central = 92%, and south = 85% shared). Six species were unique to the panhandle sens itive indicator species analysis, 3 to the north, 2 to the central, and 2 to the south ecoregion. Figure 3-14 shows that the percent sensitive indicator species, statewide and regionally, decreased with increasing development intensity. Statewide, 85% of wetlands in the low LDI group had over 20% statewide sensitive indicator species; wher eas, 86% of wetlands in the high LDI group had less than 20% statewide indicator species. All of the indicator species metrics were significantly correlated with landscape development intensity. Table 3-23 shows Sp earman correlations calculated with both statewide and regional indicator species lis ts for each ecoregion. There was little difference between the strength of regional a nd statewide indicator species correlations. Regional indicator species metrics had a strong er correlation value, though all metrics

PAGE 115

100 Table 3-22. Statewide and re gional macrophyte sensitive indi cator species. Sensitive indicator species calculated at an LDI break of 2.0 were significant (p<0.10). Statewide Panhandle North Central South No. of Sensitive Species 61 28 19 24 13 Amphicarpum muhlenbergianum (19.3, 0.02) (52.4, 0.02) Andropogon virginicus (46.9, 0.00) (62.5, 0.01) (66.7, 0.01) (66.5, 0.00) Aristida beyrichiana (8.6, 0.05) (45.5, 0.01) Aristida patula (22.2, 0.09) Aristida purpurascens (32.9, 0.00) (50.0, 0.01) (22.2, 0.08) (53.5, 0.00) Carex verrucosa (25.0, 0.05) Cladium jamaicense (22.3, 0.01) (60.6, 0.01) Coelorachis rugosa (7.3, 0.05) Cyperus haspan (11.0, 0.02) (33.3, 0.01) Drosera brevifolia (7.3, 0.06) Erianthus giganteus (17.8, 0.01) (33.3, 0.02) Eriocaulon compressum (17.1, 0.00) (25.0, 0.07) (33.3, 0.02) Eriocaulon decangulare (37.8, 0.00) (37.5, 0.02) (22.2, 0.09) (53.5, 0.01) (33.3, 0.02) Eupatorium leptophyllum (14.6, 0.00) (25.0, 0.05) Eupatorium mohrii (8.6, 0.05) Fuirena scirpoidea (26.9, 0.00) (25.0, 0.09) (25.0, 0.05) (36.2, 0.06) Gaylussacia frondosa (25.0, 0.06) Gratiola ramosa (18.3, 0.00) (41.7, 0.00) Hypericum chapmanii (8.6, 0.04) (45.5, 0.02) Hypericum fasciculatum (38.2, 0.00) (53.5, 0.00) (44.4, 0.01) Hypericum myrtifolium (17.8, 0.01) (47.7, 0.01) Hyptis alata (10.8, 0.08) Ilex glabra (46.2, 0.00) (53.9, 0.01) (78.6, 0.00) (45.9, 0.02) Ilex myrtifolia (17.1, 0.06) (71.2, 0.01) Ipomoea sagittata (11.0, 0.02) Lachnanthes caroliniana (39.6, 0.00) (71.2, 0.00) (40.2, 0.08) Lachnocaulon anceps (25.0, 0.08) Lobelia floridana (25.0, 0.08) Lophiola aurea (12.2, 0.01) (62.5, 0.00) Ludwigia linifolia (7.3, 0.05) Lycopodiella alopecuroides (9.8, 0.02) (25.0, 0.06) (22.2, 0.07) Lyonia lucida (24.4, 0.07) Nymphaea odorata (6.2, 0.10) (22.2, 0.08) Nymphoides aquatica (12.2, 0.00) Panicum ensifolium (13.1, 0.05) Panicum erectifolium (41.5, 0.00) (50.0, 0.01) (33.3, 0.01) (45.2, 0.01) (36.2, 0.06) Panicum hemitomon (40.4, 0.01) (79.2, 0.00) Panicum rigidulum (17.1, 0.05) (25.0, 0.08) Panicum tenerum (14.6, 0.00) (33.3, 0.02)

PAGE 116

101 Table 3-22. Continued. Statewide Panhandle North Central South Pinus elliottii (33.9, 0.02) (56.2, 0.04) (37.0, 0.02) (25.0, 0.07) Pinus palustris (8.6, 0.05) Pluchea foetida (10.1, 0.06) (25.0, 0.05) Pluchea rosea (15.5, 0.07) (33.3, 0.08) Polygala cymosa (28.0, 0.00) (37.5, 0.02) (29.3, 0.06) (33.3, 0.01) Polygala lutea (7.3, 0.04) Proserpinaca pectinata (12.4, 0.02) Rhexia alifanus (13.4, 0.01) (70.3, 0.00) Rhexia lutea (13.4, 0.02) (45.5, 0.02) (22.2, 0.07) Rhexia mariana (23.1, 0.00) (47.7, 0.01) (45.2, 0.01) Rhexia petiolata (25.0, 0.07) Rhus copallinum (33.1, 0.06) Rhynchospora corniculata (25.0, 0.07) Rhynchospora filifolia (37.5, 0.02) Rhynchospora inundata (17.2, 0.00) (22.2, 0.08) Rhynchospora microcarpa (28.8, 0.06) Rhynchospora wrightiana (22.2, 0.08) Sabatia bartramii (7.3, 0.05) Sagittaria graminea (13.9, 0.08) (41.7, 0.00) Sagittaria lancifolia (17.2, 0.01) (36.2, 0.06) Salix caroliniana (44.4, 0.01) Sarracenia minor (22.2, 0.07) Scleria baldwinii (7.3, 0.04) Scleria georgiana (7.3, 0.05) Scleria triglomerata (7.3, 0.06) (25.0, 0.08) Serenoa repens (22.3, 0.05) (37.5, 0.02) (57.6, 0.01) Spartina bakeri (7.3, 0.05) (25.0, 0.05) Stillingia aquatica (13.4, 0.01) (36.2, 0.05) Syngonanthus flavidulus (12.2, 0.01) (22.2, 0.08) Utricularia purpurea (8.6, 0.06) (25.0, 0.07) Vaccinium corymbosum (20.9, 0.02) (33.1, 0.06) (33.3, 0.02) Xyris ambigua (11.0, 0.02) Xyris caroliniana (7.3, 0.05) (25.0, 0.08) Xyris elliottii (18.3, 0.00) (50.0, 0.00) Xyris jupicai (8.6, 0.05) (25.0, 0.04) were significantly correlated with LDI (p < 0.01). Two exceptions included the panhandle (statewide tolerant r = 0.73, panhandle tolerant r = 0.72) and central (statewide tolerant r = 0.74, central to lerant r = 0.68) tolerant indicator correlations.

PAGE 117

102 0 10 20 30 40 50 60 70 02468 LDI%Panhandle Sensitive 0 10 20 30 40 50 60 70 02468 LDI%North Sensitive 0 10 20 30 40 50 60 70 02468 LDI%Central Sensitiv e 0 10 20 30 40 50 60 70 02468 LDI%South Sensitive 0 10 20 30 40 50 60 70 02468 LDI%Statewide Sensitive Panhandle North Central South Figure 3-14. Macrophyte sensitive indicat or species decreased with increasing development intensity (LDI). A) Panhandle sensitive indicator species at panhandle study wetlands. B) North sensit ive indicator species at north study wetlands. C) Central sensitive indicator species at central study wetlands. D) South sensitive indicator species at south study wetlands. E) Statewide sensitive indicator species at all study wetlands. A.B. D. C. E.

PAGE 118

103 Table 3-23. Statewide and regional macrophyt e indicator species were significantly correlated with LDI (p < 0.01). Statewide ISARegional ISA n= Spearman's r Spearman's r Statewide Tolerant indicator species 118 0.75 Sensitive indicator species 118 -0.66 Panhandle Tolerant indicator species 28 0.73 0.72 Sensitive indicator species 28 -0.66 -0.68 North Tolerant indicator species 31 0.79 0.80 Sensitive indicator species 31 -0.76 -0.78 Central Tolerant indicator species 31 0.74 0.68 Sensitive indicator species 31 -0.67 -0.72 South Tolerant indicator species 28 0.78 0.86 Sensitive indicator species 28 -0.60 -0.71 It is important to note that shrub and tree species were included in the ISA for both tolerant and sensitive metrics. Metrics developed based on the macrophyte community composition included woody species rooted with the sampling quadrats, as structure was thought to play an important role in the biological condition of pondcypress domes. Excluding the tree and shrub layers would seemingly underscore their importance. However, trees comprised only a small percentage of the tolerant and sensitive indicator species lists (Tables 3-21 and 3-22). Three percent of the statewide tolerant indicator species were trees, 9% were shrubs, 14% vi nes, and 74% herbaceous (including herbs, sedges, grasses, etc.). The 2 statewide to lerant indicator tree species included the hardwood Acer rubrum and exotic Sapium sebiferum (Table 3-21). The 6 statewide shrub tolerant indicator sp ecies were of the genera Aster (a climbing species), an exotic Ligustrum , an exotic Ludwigia , Rubus , and Sambucus .

PAGE 119

104 2 3 4 5 6 7 02468 LDIPanhandle Modified FQI 2 3 4 5 6 7 02468 LDINorth Modified FQI 2 3 4 5 6 7 02468 LDICentral Modified FQI 2 3 4 5 6 7 02468 LDISouth Modified FQ I Figure 3-15. Modified FQI scores decreas ed with increasing development intensity (LDI). A) Modified FQI scores at panhandle study wetlands. B) Modified FQI scores at north study wetlands. C) Modified FQI scores at central study wetlands. D) Modified FQI sc ores at south study wetlands. Modified Floristic Quality Index metric Figure 3-15 shows modified FQI scores decreased with increasing landscape development intensity. Wetlands in the pa nhandle (maximum modified FQI = 6.25) and north (maximum modified FQI = 5.95) ecoregion s had higher modified FQI scores versus wetlands in the central (maximum modified FQI = 4.93) and south (maximum modified FQI = 5.24) ecoregions. Statewide the modifi ed FQI was significantly correlated with LDI (|r| = 0.71, p < 0.001; Table 3-18); and th ere was a significant difference between the mean modified FQI scores between low and high LDI groups (W = 3771.0, p = <0.001; A.B. C.D.

PAGE 120

105 0 10 20 30 40 50 60 02468 LDI%Panhandle Exotic Â…. 0 10 20 30 40 50 60 02468 LDI%North Exotic .... 0 10 20 30 40 50 60 02468 LDI%Central Exotic Â…. 0 10 20 30 40 50 60 02468 LDI%South Exotic Â…. Figure 3-16. Exotic species increased with increasing development intensity (LDI). A) Exotic species at panhandle study wetla nds. B) Exotic species at north study wetlands. C) Exotic species at central study wetlands. D) Exotic species at south study wetlands. Table 3-19). In the low LDI groups, 87.5% , 100%, 92%, and 83% of the wetlands had a modified FQI score greater than 4.00 in the panhandle, north, central, and south ecoregions, respectively. Wetla nds with a modified FQI sc ore less than 4.00 accounted for 70% 82%, 74%, and 87.5% of the wetlands in the high LDI group in the panhandle, north, central, and south ecoregions, respectively. A.B. D. C.

PAGE 121

106 Exotic species metric Statewide, the percent exotic species was significantly correlated with development intensity in the surroundi ng landscape (r = 0.65, p < 0.001; Table 3-18). Figure 3-16 shows that the percent of exotic species in creased with increasing LDI in each ecoregion. The north ecoregion hosted the wetland with the greatest percent e xotic species, NA1 (52.6% exotic species). NA1 was surrounded by research facility growing experimental pasture species, potentially bi asing the high percent exotic species present at this study wetland. The wetland with the second highest percent exotic species was SU8 (38.5%), a wetland embedded in urban land use (residential and commercial). One apparent outlier in south ecoregion low LDI group was SU4 (e xotics = 18.4%). All remaining wetlands in the low LDI group (n = 40) ha d less than 10% exotic species. Statewide, the percent exotic species wa s significantly different between low and high LDI groups (W = 1379.0, p < 0.001; Table 3-19). Table 3-24 lis ts the 113 exotic species encountered throughout Florida and id entifies the ecoregion(s) in which each species was found. Only 6 exotic specie s were found in all 4 ecoregions including Commelina diffusa , Cuphea carthagenensis , Cynodon dactylon , Kyllinga brevifolia , Ludwigia peruviana , and Paspalum notatum . Fourteen exotic sp ecies occurred in 3 of the 4 ecoregions. Native perennial species metric Of the 605 macrophyte species identified, 427 (71%) were classified as native perennials. Figure 3-17 shows that native perennial species decreased with increasing development intensity. Statewide there was a significant difference between the percent native perennial species between low and high LDI groups (W = 3453.0, p < 0.001; Table 3-19). The native perennial species metric was significantly correlated with LDI

PAGE 122

107 Table 3-24. Exotic macrophyte specie s identified at 118 study wetlands. Exotic Species PNCSExotic Species P N C S Albizia julibrissin Kyllinga brevifolia Aloe vera Lantana camara Alternanthera philoxeroides Ligustrum japonicum Alternanthera sessilis Ligustrum lucidum Amaranthus blitum Ligustrum sinense Amaranthus spinosus Lindernia crustacea Ardisia crenata Lolium perenne Begonia cucullata Lonicera japonica Bischofia javanica Ludwigia peruviana Blechum pyramidatum Lygodium japonicum Bromus catharticus Lygodium microphyllum Callisia repens Macroptilium lathyroides Chenopodium album Melaleuca quinquenervia Chenopodium ambrosioides Melia azedarach Cinnamomum camphora Melochia corchorifolia Citrus Xaurantium Momordica charantia Colocasia esculenta Morrenia odorata Commelina diffusa Morus alba Conyza bonariensis Murdannia nudiflora Cuphea carthagenensis Nandina domestica Cyclospermum leptophyllum Nephrolepis cordifolia Cynodon dactylon Oeceoclades maculata Cyperus iria Oxalis debilis Cyperus lanceolatus Paederia foetida Desmodium incanum Panicum maximum Digitaria bicornis Panicum repens Dioscorea bulbifera Paspalidium geminatum Duchesnea indica Paspalum acuminatum Echinochloa colona Paspalum notatum Echinochloa crusgalli Paspalum urvillei Eichhornia crassipes Phalaris angusta Eleusine indica Phyllanthus tenellus Eragrostis atrovirens Phyllanthus urinaria Eugenia uniflora Plantago lanceolata Hedychium coronarium Pouzolzia zeylanica Hedyotis corymbosa Pueraria montana Hemarthria altissima Rhodomyrtus tomentosa Hymenachne amplexicaulis Rhoeo discolor Imperata cylindrica Richardia brasiliensis Ipomoea indica Richardia scabra Ipomoea quamoclit Rumex crispus Kummerowia striata Rumex obtusifolius

PAGE 123

108 Table 3-24. Continued. Rumex pulcher Tradescantia zebrina Sacciolepis indica Trifolium repens Salvinia minima Urena lobata Sapium sebiferum Urtica dioica Schinus terebinthifolius Verbena bonariensis Senna obtusifolia Verbena brasiliensis Senna pendula Viburnum odoratissimum Solanum tampicense Vicia sativa Solanum viarum Wedelia trilobata Sonchus asper Xanthosoma sagittifolium Sorghum bicolor Xyris jupicai Spermacoce verticillata Youngia japonica Sporobolus indicus Yucca aloifolia Thelypteris dentata Zea mays Tradescantia fluminensis (Spearman |r| = 0.63, p < 0.001; Table 3-18). St atewide 78% of the wetlands in the low LDI group had greater than 90% native perenn ial species; whereas, 75% of the wetlands in the high LDI group had less than 90% native perennial species. Wetland status metric Fifty-six percent of the macrophyte species identified were included in the wetland status metric, including 160 species designa ted as obligate and 180 species designated as facultative wet species. There were an a dditional 137 facultative, 62 facultativ e upland, and 49 upland species identified in the study wetlands. Seventeen species (of the 605 macrophyte species identified in this study) were not categorized by wetland status. Figure 3-18 shows wetland status species decreased with increasing development intensity in each ecoregion. The percent wetland status species was significantly different between LDI groups (W = 3612.0, p < 0.001; Table 3-19); and significantly correlated statewide with th e LDI index (Spearman |r| = 0.55, p < 0.001; Table 3-18). Statewide 90% of the wetlands in the low LD I group had greater than 60% wetland status

PAGE 124

109 40 50 60 70 80 90 100 02468 LDI%Panhandle Native Perennia l ,,, 0 10 20 30 40 50 60 02468 LDI%North Exotic .... 40 50 60 70 80 90 100 02468 LDI%Central Native Perennial Â…. 40 50 60 70 80 90 100 02468 LDI%South Native Perennial Â…. Figure 3-17. Native perennial species decreased with increasing development intensity (LDI). A) Panhandle study wetlands. B) North study wetlands. C) Central study wetlands. D) South study wetlands. species, whereas 75% of the wetlands in the high LDI group had less than 60% wetland status species. Macrophyte Wetland Condition Index The 6 metrics described above were incl uded in the macrophyte WCI. Figure 3-19 shows that both statewide and regional m acrophyte WCI scores decrease with increasing development intensity. Table 3-25 compar es the overall macrophyte WCI calculated statewide and regionally for the low LDI group (LDI < 2.0). A comparable statewide macrophyte WCI should equally score referenc e wetlands in each region; however, the A.B. C.D.

PAGE 125

110 20 40 60 80 100 02468 LDI%Panhandle Wetland 20 40 60 80 100 02468 LDI%North Wetland (A) (B) 20 40 60 80 100 02468 LDI%Central Wetland 20 40 60 80 100 02468 LDI%South Wetland (C) (D) Figure 3-18. The percent wetland status spec ies decreased with increasing development intensity (LDI). This trend was cons istent for the (A) panhandle, (B) north, (C) central, and (D) south ecoregions. south ecoregion had significantly different overall macrophyte WCI scores for the low LDI group compared to the panhandle, north, and central ecoregions. When calculated statewide, 5 of the 6 metrics had 1 or more ecoregion with significantly different metric scores. The north and central ecoregions ha d significantly differe nt scores for the statewide tolerant indicator species, wh ereas the north and south ecoregions had significantly different scores for the statewide sensitive indicator species. The panhandle and north ecoregions were not significantly different from each other, but were significantly different from the central and south ecoregions for modified FQI scores; A.B C.D.

PAGE 126

111 0 10 20 30 40 50 60 02468 LDIStatewide Mp-WCI 0 10 20 30 40 50 60 02468 LDIRegional Mp-WC I Reference Agricultural Urban Figure 3-19. Macrophyte WCI scores decrease d with increasing development intensity (LDI). A) Statewide B) Regional. A. Statewide Macro p h y te WCI B. Re g ional Macro p h y te WCI

PAGE 127

112 Table 3-25. Macrophyte WCI and metrics sc ored statewide and regionally for study wetlands in the low LDI group. Panhandle North Central South Statewide Macrophyte WCI 50.3 ± 8.6a 51.6 ± 8.4a 45.9 ± 3.4ab 42.2 ± 9.4b State tolerant 8.1 ± 2.5ab 8.4 ± 2.2a 6.6 ± 1.2b 6.8 ± 2.3ab State sensitive 8.8 ± 2.6ab 9.0 ± 1.3a 8.6 ± 0.7ab 6.7 ± 2.9b Modified FQI 8.5 ± 2.2a 8.1 ± 1.4a 6.6 ± 0.9b 6.4 ± 1.7b Exotic species 9.3 ± 1.0a 9.4 ± 1.1a 8.4 ± 1.3a 6.6 ± 1.9b Native perennial species 8.6 ± 0.9ab 9.1 ± 1.0a 8.3 ± 0.6b 7.9 ± 1.1b Wetland status species 7.0 ± 2.0a 7.5 ± 1.9a 7.4 ± 1.6a 7.7 ± 2.3a Regional Macrophyte WCI 50.9 ± 7.2a 51.3 ± 7.9a 51.4 ± 4.0a 46.8 ± 9.8a Regional tolerant 9.6 ± 1.0a 9.2 ± 1.3ab 9.0 ± 1.5ab 7.7 ± 2.7b Regional sensitive 8.2 ± 2.6a 8.5 ± 1.7a 8.6 ± 1.6a 6.9 ± 2.4a Modified FQI 7.6 ± 2.3a 7.9 ± 1.4a 8.7 ± 1.1a 7.7 ± 2.0a Exotic species 9.3 ± 1.0a 9.5 ± 1.1a 8.3 ± 1.4a 8.3 ± 1.9a Native perennial species 8.3 ± 1.0a 9.2 ± 0.9a 9.1 ± 0.7a 8.7 ± 1.2a Wetland status species 8.0 ± 2.1a 7.0 ± 2.5a 7.8 ± 1.9a 7.6 ± 2.1a Values represent the mean score ± the standard deviation Ecoregions with similar letters were not significantly different (p<0.05) suggesting that the panhandle and north ecoregions hosted more species with a narrower set of ecological conditions found in refe rence wetlands. The south ecoregion had significantly different statewide exotic species scores than the other 3 ecoregions. The scores for the statewide native perennial speci es were significantly different for the north ecoregion. When the macrophyte WCI was scored regionally, there wa s not a significant difference between mean scores for the low LDI group (Table 3-25). The only regionally scored metric with significantly different means scores for the low LDI group was the regional tolerant indi cator species for the south ecor egion which was only significantly different from the panhandle ecoregion. Table 3-26 shows similar results for th e high LDI group. Within the macrophyte WCI calculated statewide, wetlands in the nor th ecoregion had significantly different macrophyte WCI scores, suggesting the north ecoregion high LDI wetlands had higher

PAGE 128

113 Table 3-26. Macrophyte WCI and metrics sc ored statewide and regionally for study wetlands in the high LDI group. Panhandle North Central South Statewide Macrophyte WCI 20.4 ± 13.1a 29.4 ± 15.9b 18.9 ± 11.0a 17.6 ± 10.1a State tolerant 2.1 ± 1.9ab 3.3 ± 2.8a 2.0 ± 1.6b 3.3 ± 1.9ab State sensitive 2.5 ± 2.0ab 4.1 ± 2.4a 2.2 ± 2.1ab 2.9 ± 2.7b Modified FQI 3.2 ± 3.2a 4.3 ± 2.8a 2.8 ± 1.8b 2.5 ± 1.8b Exotic species 4.0 ± 3.0a 6.3 ± 3.7a 3.6 ± 2.4a 2.7 ± 2.4b Native perennial species 5.2 ± 2.7ab 6.6 ± 3.2a 4.3 ± 2.6b 3.9 ± 2.9b Wetland status species 3.3 ± 2.4a 4.8 ± 2.6a 4.0 ± 2.5a 2.3 ± 1.8a Regional Macrophyte WCI 21.3 ± 14.5ab 27.1 ± 16.1a 20.6 ± 12.8ab 16.6 ± 9.6b Regional tolerant 5.5 ± 3.7a 3.9 ± 3.2ab 4.2 ± 3.6ab 1.8 ± 1.4b Regional sensitive 1.3 ± 1.7a 2.1 ± 2.4a 1.6 ± 2.0a 1.3 ± 1.8a Modified FQI 2.8 ± 2.8a 4.4 ± 2.7a 3.3 ± 2.5a 2.9 ± 2.1a Exotic species 3.9 ± 3.0a 6.4 ± 3.6a 3.2 ± 2.5a 3.7 ± 2.9a Native perennial species 4.2 ± 3.2a 6.8 ± 3.2a 4.5 ± 2.9a 4.3 ± 3.2a Wetland status species 3.7 ± 2.8a 3.5 ± 3.3a 3.7 ± 2.9a 2.6 ± 1.8a Ecoregions with similar letters were not significantly different (p<0.05). Values represent the mean score ± the standard deviation. ecological integrity than wetlands of other ecoregions. Five of the 6 metrics calculated statewide had at least 1 ecoregion with signi ficantly different mean scores in the high LDI group. Only the percent wetland status species metric did not have significantly different scores for both statewide and regi onal calculations among ecoregions. For the regional macrophyte WCI calculations, the north and south ecoregions had significantly different mean macrophyte WCI scores. Addition ally, only the regional tolerant indicator species metric for the panhandle and south eco regions had significan tly different mean scores for all of the regionally calculated metrics. Correlations between macrophyte WCI and 6 metrics with LDI were strong (|r| > 0.50, p < 0.01) for all of the metrics statewide and regionally (Table 3-27), except for the central ecoregion wetland status species (|r| = 0.39, p = 0.03). This metric was still significantly correlated at the more flexible p < 0.05 level. Three of the 4 ecoregions,

PAGE 129

114 Table 3-27. Spearman correlations between the macrophyte WCI, metrics, and LDI. Spearman's r p-value Statewide Macrophyte WCI -0.73 <0.0001 Tolerant indicator species 0.75 <0.0001 Sensitive indicator species -0.66 <0.0001 Modified FQI -0.71 <0.0001 Exotic species 0.65 <0.0001 Native perennial species -0.63 <0.0001 Wetland status species -0.55 <0.0001 Panhandle Macrophyte WCI -0.74 <0.0001 Tolerant indicator species 0.72 <0.0001 Sensitive indicator species -0.68 <0.0001 Modified FQI -0.68 <0.0001 Exotic species 0.72 <0.0001 Native perennial species -0.67 0.0001 Wetland status species -0.60 0.0007 North Macrophyte WCI -0.74 <0.0001 Tolerant indicator species 0.80 <0.0001 Sensitive indicator species -0.78 <0.0001 Modified FQI -0.75 <0.0001 Exotic species 0.65 <0.0001 Native perennial species -0.65 <0.0001 Wetland status species -0.55 0.0015 Central Macrophyte WCI -0.73 <0.0001 Tolerant indicator species 0.68 <0.0001 Sensitive indicator species -0.72 <0.0001 Modified FQI -0.68 <0.0001 Exotic species 0.70 <0.0001 Native perennial species -0.66 <0.0001 Wetland status species -0.39 0.0309 South Macrophyte WCI -0.88 <0.0001 Tolerant indicator species 0.86 <0.0001 Sensitive indicator species -0.71 <0.0001 Modified FQI -0.86 <0.0001 Exotic species 0.80 <0.0001 Native perennial species -0.80 <0.0001 Wetland status species -0.69 <0.0001

PAGE 130

115 including the panhandle, nort h, and south ecoregions had st ronger WCI correlations with LDI than both the statewide and central ecoregion WCIs. Cluster Analysis Cluster analysis determined 5 categor ies of wetlands based on macrophyte community composition. Clusters were roughly defined by ecoregions and a priori land use categories, including: 1: northern referenc e; 2: southern reference; 3: northern developed land use; 4: southern developed la nd use; and 5: statew ide cattle land use. Figure 3-20 shows that based on regional macr ophyte WCI scores, clusters 1 and 2 were not significantly different from one another, but were significantly di fferent from clusters 3, 4, and 5 (p<0.05). Clusters 3 and 4 were not significantly different from each 0 10 20 30 40 50 60 Cluster 1 Regional Mp-WCI Cluster 2 Cluster 3 Cluster 4 Cluster 5a b a c b Figure 3-20. Regional macrophyte WCI scor es for 5 wetland clusters based on macrophyte community composition. Re g ional Macro p h y te WCI

PAGE 131

116 Table 3-28. Macrophyte WCI scores and LDI values for wetland clusters based on macrophyte community composition. Cluster Statewide Macrophyte WCI Regional Macrophyte WCI LDI 1 46.8 ± 11.1a 47.4 ± 9.9a 2.3 ± 1.9a 2 44.5 ± 7.7a 47.4 ± 9.7a 1.7 ± 1.3a 3 24.9 ± 12.2b 25.9 ± 12.6b 4.4 ± 1.9bc 4 21.9 ± 13.8b 23.2 ± 16.2b 4.0± 2.0b 5 10.6 ± 7.0c 8.8 ± 6.6c 5.2 ± 0.5c Clusters with similar letters within columns were not significantly different (p<0.05). other. Cluster 5 was significan tly different from all other clus ters. Identical results were obtained using statewide macrophyte WCI sc ores. Table 3-28 provides means and standard deviations for cluster statew ide macrophyte WCI scores, regional macrophyte WCI scores, and LDI. Macroinvertebrates Statewide 79 wetlands were sampled for the macroinvertebrate assemblage, with 118 species, representing 169 genera, 85 families, 24 orders, 9 classes, and 5 phyla. The most common macroinvertebrat e genera identified were Polypedilum, Dero , and Goeldichironomous, comprising 19%, 18%, and 8% of all the individual macroinvertebrates identified to the genus or lower taxonomic level, respectively. Four genera, Polypedilum, Dero , Goeldichironomous, and Kiefferulus, were found at over 50% of the study wetlands. Of the genera enc ountered, 81 genera ( 48%) occurred at a minimum of 5% of the sample wetlands (n 4). Approximately one-third of the genera identified (53 genera or 31%) we re encountered at only one wetland. The most common families identifie d included Chironomidae, Naididae, Enchytraeidae, and Culicidae, represen ting 39, 19, 4, and 4% of the individuals

PAGE 132

117 identified, and occurring at 99, 81, 52, a nd 56% of the study wetlands, respectively. Macroinvertebrates in the fa mily Chironomidae were further divides into the subfamilies Chironominae (89% of Chironomidae), Ta nypodinae (10% of Chironomidae), and Orthocladiinae (1% of Chironomidae). Six orders were found at over 50% of the wetlands sampled, including Diptera (47% of individual identified to the order taxonomic level or lower), Tubificida (24%), Co leoptera (6%), Basommatophora (5%), Odonata (4%), and Hemiptera (3%). The most common classes of macroinvertebrates identified included Insecta (63%), Oligochaeta (24%), Gastropoda (6%), and Crustacea (5%), all occurring at over 50% of the study wetlands. Five phylum were identified, including Arthropoda, Annelida, Mollusca, Platyhelm inthes, and Nemertea, with Arthropoda, Annelida, Mollusca found at 100, 92, and 65% of the wetlands sampled, respectively. In the panhandle ecore gion, 13 wetlands were sampled hosting 84 genera representing 48 families and 17 orders. In the north ecoregion 15 wetlands were sampled with 87 genera (58 families and 20 orders) encountered. The central ecoregion included 25 wetlands with 109 genera (60 families and 23 orders) recognized. The south ecoregion had 26 sample wetlands with 105 genera (in 60 families and 21 orders) identified. Summary Statistics Richness (R), evenness (E), Shannon divers ity (H), and SimpsonÂ’s index (S) were calculated for each sample wetland ( Appendix F ). Richness ranged from 1 genera ( Pristina ) at PR4 (embedded in a low-intensit y silvicultural land use in a National Forest), to 26 genera at PA3 (surrounded w ith row crops). Species evenness ranged from 0.00 at PR4 to 0.97 at PR5 (surrounded by upland forest). Shannon diversity ranged from 0 at PR4 to 0.92 at CA7 (surrounded by silvicul tural operations and past ure with cattle).

PAGE 133

118 Table 3-29. Macroinvertebrate rich ness, evenness, and diversity among a priori land use categories. Reference Agricultural Urban Richness (R) 14 ± 6a14 ± 6a 13 ± 5a Evenness (E) 0.69 ± 0.24a0.69 ± 0.12a 0.68 ± 0.15aShannon Diversity (H) 0.70 ± 0.26a0.72 ± 0.15a 0.70 ± 0.17aSimpson's Index (S) 1.83 ± 0.75a1.81 ± 0.54a 1.73 ± 0.55aBeta Diversity 8.07.8 8.5 Gamma Diversity 114110 111 Categories with similar letters were not significantly different (Fisher's LSD, =0.05) Values represent mean ± standard deviation Simpson’s index was highest at CR10 at 2.79. Table 3-29 summarizes the richness, evenness, and diversity calculations by a priori land use category. No significant differences were found in richness, evenne ss, Shannon diversity, or Simpson’s index among the 3 a priori land use categories. Beta and gamma diversity were also similar among a priori land use categories. Table 3-30 shows that the no significant differences were found for richness, evenness, Shannon diversity, or Simpson’s i ndex between wetlands in low and high LDI groups. Beta and gamma diversity were higher for the high LDI groups with beta diversity at 10.9 and gamma dive rsity at 146 for the high LDI group and beta diversity at 8.7 and gamma diversity at 124 for the low LDI group wetlands. Compositional Analysis MRPP was used to test the similarity of macroinvertebrate genera composition across all ecoregions (panhandle versus north versus central ve rsus south) as well as for multiple pair wise ecoregion comparisons (panhandle versus north, panhandle versus central, panhandle versus south, north versus central, north versus south, and central versus south). Among all wetlands, the comparison across all groups and the multiple pair wise comparisons suggested macroinve rtebrate community composition at the

PAGE 134

119 Table 3-30. Macroinvertebrate richness, ev enness, and diversity for LDI groups. Low LDI High LDI W^ p` Richness (R) 14 ± 6 13 ± 5 661 0.33 Evenness (E) 0.68 ± 0.22 0.69 ± 0.14 687 0.47 Shannon Diversity (H) 0.70 ± 0.24 0.71 ± 0.17 695 0.52 Simpson's Index (S) 1.82 ± 0.71 1.77 ± 0.56 678 0.42 Beta Diversity 8.7 10.9 Gamma Diversity 124 146 Values represent mean ± standard deviation ^W = Mann-Whitney U-Test statistic `p = significance value genera level was signifi cantly different(at the = 0.05 level), with the exception of the pair wise comparison between the panhandle and north ecore gions (Table 3-31). The macroinvertebrate community composition of the panhandle and north ecoregions was not significantly different for all tests, in cluding among all wetlands and for reference, agricultural, and urban wetlands independently. In reference wetlands, the south eco region had a significantly different macroinvertebrate community composition as compared to all other ecoregions (panhandle versus south T = -3.2, p = 0.00; nor th versus south T = -3.3, p = 0.00; central versus south T = -2.1, p = 0.03). There were not significant differences in macroinvertebrate community composition between agricultural wetlands in any neighboring ecoregions; however, macroinver tebrate community composition in the south ecoregion was significantly differe nt from both the panhandle and north ecoregions; as were the central and panhandl e ecoregions. The only ecoregions with significantly different macroi nvertebrate community com position among urban wetlands were the panhandle and south ecoregions.

PAGE 135

120 Table 3-31. Macroinvertebrate comm unity composition similarity among a priori land use categories and ecoregions. Sites (n)T^ A` p# All Wetlands All regions (P vs N vs C vs S) 79 -7.10.10 0.00* Panhandle vs north 28 0.5-0.01 0.67 Panhandle vs central 38 -3.60.05 0.00* Panhandle vs south 39 -8.50.13 0.00* North vs central 40 -2.40.04 0.02* North vs south 41 -6.60.10 0.00* Central vs south 51 -3.20.04 0.00* Reference wetlands All regions (P vs N vs C vs S) 29 -3.20.13 0.00* Panhandle vs north 12 -0.20.01 0.35 Panhandle vs central 14 -0.70.03 0.21 Panhandle vs south 15 -3.80.17 0.00* North vs central 14 -0.50.03 0.27 North vs south 15 -3.30.15 0.00* Central vs south 17 -2.10.08 0.03* Agricultural wetlands All regions (P vs N vs C vs S) 24 -2.70.12 0.01* Panhandle vs north 8 -0.50.04 0.30 Panhandle vs central 12 -2.10.10 0.03* Panhandle vs south 12 -3.30.18 0.00* North vs central 12 -1.00.04 0.17 North vs south 12 -2.30.12 0.02* Central vs south 16 -0.50.02 0.30 Urban wetlands All regions (P vs N vs C vs S) 26 -1.20.05 0.11 Panhandle vs north 8 -1.00.09 0.16 Panhandle vs central 12 -0.60.03 0.27 Panhandle vs south 12 -2.10.09 0.03* North vs central 14 0.7-0.03 0.75 North vs south 14 -1.60.07 0.07 Central vs south 18 -1.10.03 0.14 *A high |T| value and significant p-valu e (p<0.05) suggests a difference in species composition ^ T = the MRPP test statistic `A = the chance correcte d within-group agreement #p = the significance value.

PAGE 136

121 Latitude LDI pH Axis 2: 14.2%Axis 3: 35.9% A priori Impact Category Minimally Impaired Agriculturally Impaired Urban Impaired A priori Land Use Category o Reference Agricultural Urban Figure 3-21. NMS ordination bi-plot for 79 wetlands in macroinvertebrate genus space with an overlay of environmental parameters. Latitude, LDI, and water column pH (shown as radiating vector s), were significantly correlated with the NMS axes based on macroinvertebrate community composition. Vector length represents the strength of the co rrelation, and the angle represents the direction of maximum change. Axis 2 explained 14.2% variance, axis 3 explained 35.9% variance, and axis 1 ( not shown) represented an additional 18.9% variance.

PAGE 137

122 Table 3-32. PearsonÂ’s r-square d correlations between envi ronmental variables and NMS ordination axes based on macroinve rtebrate community composition. Axis 1 Axis 2 Axis 3 Incremental r2 18.9% 14.2% 35.9% Cumulative r2 18.9% 33.0% 68.9% Latitude 0.17 0.03 0.22 Longitude 0.05 0.02 0.09 LDI 0.01 0.25 0.01 Log(DO) 0.01 0.13 0.00 Log(Temperature) 0.03 0.05 0.07 Log(Color) 0.02 0.02 0.10 Log(Turbidity) 0.00 0.09 0.02 pH 0.00 0.06 0.24 Log(Water ammonia-N) 0.04 0.08 0.01 Log(Water nitrate/nitrite-N) 0.01 0.02 0.05 Log(Water TKN) 0.05 0.08 0.00 Log(Water TP) 0.02 0.18 0.01 Arcsin Sqrt (Soil moisture) 0.02 0.04 0.09 Log(Soil TP) 0.10 0.04 0.01 Community Composition Macroinvertebrate community composition was summarized in an NMS ordination to relate changes in macroinvertebrate community composition with environmental variables. Figure 3-21 shows a 2 dimensiona l bi-plot of the NMS axes. Overlays of significant environmental variables include water column pH, LDI, and latitude. Table 332 provides the PearsonÂ’s r-squared correlati on coefficients between environmental variables and NMS ordination ax es. A three dimensional so lution was constructed with an overall stress of 19.8 with a final stabilit y of 0.04. Axis 1 explained 18.9% of the variance and was not correlated with any m easured environmental parameters. Axis 2 explained 14.2% variance and was correlated with LDI; axis 3 e xplained 35.9% variance and was correlated with lati tude and water column pH.

PAGE 138

123 Metric Selection Over 20 of the candidate metrics were significantly correlated with LDI (SpearmanÂ’s |r| > 0.30, p < 0.01). Table 333 provides the statewide Spearman correlation values between the 6 macroinverteb rate metrics and LDI, water column pH, dissolved oxygen, and water column TP. Macroinvertebrate metrics selected for inclusion represented toleran ce, community balance, and functional group metrics. Tolerance metrics included the tolerant indicator genera, sensitive indicator genera, and Florida Index. Community ba lance metrics included Mollusca (phylum taxonomic level) and Noteridae (family taxonomic level). On e functional groups metric was included, scrapers. The percent of tolerant indicator genera, Mollusca, and scrapers increased with increasing development intensity, whereas sens itive indicator genera , Florida Index, and Noteridae decreased with increasing developmen t intensity. Table 3-34 shows that scores of selected metrics and the macroinvertebra te WCI were significan tly different between low and high LDI groups (p < 0.05). Table 3-33. Spearman correlations between macroinvertebrate metrics and the macroinvertebrate WCI with LDI, pH, log(DO), and log(TP). Macroinvertebrate Metrics LDIpHLog(DO) Log(TP) Tolerance metrics Tolerant indicator genera 0.510.62-0.25 Sensitive indicator genera -0.470.39 -0.37 Florida Index -0.350.35 -0.24 Community balance Mollusca 0.330.54-0.28 Noteridae -0.34 Functional group Scraper 0.30 Macroinvertebrate WCI -0.62-0.560.48 -0.34 All correlations shown are significant (p<0.05)

PAGE 139

124 Table 3-34. Macroinvertebrate metric and WCI scores between LDI groups. Metric Low LDI High LDI W^ p` Tolerant indicator species 4.4 ± 11.814.2 ± 15.5904.0 <0.001 Sensitive indicator species 15.5 ± 20.32.2 ± 4.41679.0 <0.001 Florida Index 2.1 ± 2.30.9 ± 1.21572.0 0.008 Mollusca 2.0 ± 3.59.7 ± 13.41025.0 0.003 Noteridae 1.8 ± 3.70.2 ± 0.71518.0 0.012 Scrapers 4.2 ± 6.710.9 ± 12.81046.0 0.006 Macroinvertebrate WCI 36.8 ± 10.022.3 ± 8.41878.5 <0.001 Values represent the mean ± standard deviation ^W = Mann-Whitney U-Test statistic `p = significance value Tolerance metrics Multiple ISA using genus-level abundance data were completed at different LDI breaks, starting at 1.0 and continuing thr ough 7.0, at 0.25 step increments. Table 3-35 shows the results of the iter ative ISA calculations. The st atewide tolerant indicator genera were established at an LD I break of 4.0, and included 6 genera, Goeldichironomus , Micromenetus , Microvelia , Physella , Tropisternus , and Tanypus (Table 3-36). Figure 3-22 shows tolerant i ndicator genera increased w ith increasing development intensity. Two outliers were apparent in the low LDI group, including SU4 (tolerant = 61%) and SA8 (tolerant = 31%). Wetlands in the high LDI group with the highest percent tolerant indicator genera included 4 central (CA9, CA3, CU5, CA2) and 1 north (NA4) ecoregion wetland.. All 5 wetlands were surrounded by di fferent land uses, including citrus crops (CA9), pullet farm sp ray field (CA3), residential and commercial (CU5), pasture (CA2), and row crops (NA4).

PAGE 140

125 Table 3-35. Macroinvertebrate IS A calculations over a range of LDI values. Highlighted areas indicate the LDI value selected for sensitive and tolerant indicator species. Low LDI High LDI Sensitive Tolerant LDI n* = n* = r^ p` r^ p` No. Sensitive Indicators No. Tolerant Indicators 1 10 69 -0.39 0.00 xx xx 8 0 1.25 26 53 -0.43 <0.0001 0.47 <0.0001 16 3 1.5 30 49 -0.45 <0.0001 0.46 <0.0001 11 5 1.75 32 47 -0.46 <0.0001 0.48 <0.0001 14 5 2 33 46 -0.48 <0.0001 0.49 <0.0001 11 3 2.25 35 44 -0.52 <0.0001 0.49 <0.0001 12 3 2.5 35 44 -0.52 <0.0001 0.49 <0.0001 12 3 2.75 36 43 -0.54 <0.0001 0.48 <0.0001 13 4 3 37 42 -0.54 <0.0001 0.48 <0.0001 13 4 3.25 38 41 -0.53 <0.0001 0.47 <0.0001 13 4 3.5 39 40 -0.47 <0.0001 0.49 <0.0001 9 5 3.75 39 40 -0.47 <0.0001 0.49 <0.0001 9 5 4 42 37 -0.52 <0.0001 0.50 <0.0001 9 6 4.25 43 36 -0.53 <0.0001 0.50 <0.0001 7 5 4.5 46 33 -0.48 <0.0001 0.46 <0.0001 8 2 4.75 50 29 -0.40 0.00 0.51 <0.0001 7 4 5 56 23 -0.31 0.01 0.43 <0.0001 3 5 5.25 61 18 -0.20 0.07 0.55 <0.0001 3 5 5.5 64 15 -0.21 0.07 0.20 0.08 3 5 5.75 66 13 -0.04 0.70 0.13 0.25 1 3 6 66 13 -0.04 0.70 0.13 0.25 1 3 6.25 70 9 xx xx 0.06 0.58 0 7 6.5 72 7 0.03 0.82 0.21 0.07 1 13 6.75 74 5 xx xx 0.19 0.10 0 13 7 75 4 xx xx 0.38 0.00 0 6 *n = number of sites. ^r = SpearmanÂ’s r correlation coefficien t of indicator spec ies versus LDI. `p = significance value.

PAGE 141

126 PhylumClassOrderFamilyGeneraIndicator Valuep-value ArthropodaInsectaColeopteraHydrophilidae Tropisternus 17.20.053 DipteraOrthocladiinae Goeldichironomus 56.90.001 Chironomidae Tanypus 8.10.098 HeteropteraVeliidae Microvelia 12.90.033 MolluscaGastropodaBasommatophoraPlanorbidae Micromenetus 22.50.084 Physidae Physella 21.50.002 Table 3-36. Macroinvertebrate tolerant indicator genera. Tolera nt indicator genera calc ulated at an LDI break of 4.0 were significant (p < 0.10).

PAGE 142

127 0 10 20 30 40 50 60 70 02468Landscape Development Intensity Index%Toleran t Figure 3-22. Tolerant macroi nvertebrate indicator genera increased with increasing development intensity (LDI). The 14 sensitive indicator genera (Table 337) were calculated at an LDI break of 1.75. The 5 sensitive indicator genera with the highest indicator values included Bersous , Hydrocanthus , Larsia, Pristina , and Pristinella . Sensitive indicator genera included macroinvertebrates in 2 phyla, Annelida and Arthropoda. The phylum Annelida was represented by 2 genera of aquatic worms, Prisina and Pristinella , both in the family Naididae, order Haplotaxida, class Oligocha eta. The 12 remaining sensitive indicator genera fell within the phyl um Arthropoda, representing 2 cl asses Arachnida (including a water mite); and Insecta, aquatic insects in 3 orders including Coleoptera (5 genera of beetles), Diptera (4 genera of true flies), and Trichoptera (2 genera of caddis flies). Figure 3-23 shows sensitive indicator genera decreased with increasing development intensity in the landscape. Two outliers occurred, 1 in each LDI group. All of the macroinvertebrates identified at PR4 (in the low LDI group) were sensitive

PAGE 143

128 PhylumClassOrderFamilyGeneraIndicator Valuep-value AnnelidaOligochaetaHaplotaxidaNaididae Pristina 35.50.008 Pristinella 25.90.094 ArthropodaArachnidaAcariformesHydrachnidae Hydrachna 9.40.049 InsectaColeopteraDytiscidae Laccophilus 8.90.061 Haliplidae Haliplus 8.80.055 Hydrophilidae Berosus 28.80.003 Noteridae Hydrocanthus 28.50.004 Suphis 12.50.048 DipteraChironomidae Larsia 24.80.011 Paramerina 140.062 Zavreliella 17.30.008 Orthocladiinae Dicrotendipes 14.80.065 TrichopteraLeptoceridae Oecetis 12.50.023 Hydroptilidae Oxyethira 12.50.032 Table 3-37. Sensitive macroinvertebrat e indicator genera. Sensitive indica tor genera calculated at an LDI break of 1.75 were significant (p < 0.10).

PAGE 144

129 0 20 40 60 80 100 02468Landscape Development Intensity Index%Sensitive Figure 3-23. Sensitive macroinvertebrate indi cator genera decreased with increasing development intensity (LDI). indicator genera (100 %). Wetlands hosting the next highe st sensitive macroinvertebrate indicator genera were NR3 with only 42% sensitive indicator genera and CR11 with 41%. The outlier in the high LDI group with a low presence of sensitive indicator genera was SU2 (sensitive = 25%). The third tolerance metric was the Flor ida Index, an index based on the relative pollution tolerance of macroinverte brates identified in a water body ( USEPA 2002c ; Beck 1954 ). Calculations for the Florida Inde x included scoring Class I organisms, which were considered least tolerant, and Class II organisms, which were considered intolerant of pollution. Mixed taxonomic leve ls were included in the Florida Index from species (example: Polypedilum halterale ) to genus (example: all species of Elimia ) to family (example: all species of Gammaridae) to order (example: all species of Plecoptera; USEPA 2002c ; Beck 1954 ). The Florida Index value was expected to decrease with

PAGE 145

130 0 2 4 6 8 02468Landscape Development Intensity IndexFlorida Inde x Figure 3-24. Florida Index scores decreas ed with increasing development intensity (LDI). increasing development intensity in the surrounding landscape ( Barbour et al. 1996a ). Figure 3-24 shows that the Florida Index sc ore generally decreased with increasing development intensity in the surrounding lands cape, with 6 outliers including 4 urban and 2 agricultural wetlands. The 5 highest scoring wetland s in the low LDI group included wetlands in each ecoregion including CR4 (Flori da Index = 8), SR8 (Florida Index = 7), NR3 (Florida Index = 7), CR11 (Florida I ndex = 6), and PR8 (Florida Index = 5). Community balance metrics Two community balance metrics were inco rporated into the macroinvertebrate WCI including percent Mollusca an d percent Noteridae. The pe rcent of individuals in the phylum Mollusca was significantly correlated with the LDI (Table 3-33) and significantly differentiated between low and high LDI groups (Table 3-34). Figure 3-25 shows that the percent of macroinverteb rates in the phylum Mollusca increased with increasing

PAGE 146

131 0 10 20 30 40 50 02468Landscape Development Intensity Index%Mollusca Figure 3-25. Macroinvertebrates in the phylum Mollusca increas ed with increasing development intensity (LDI). development intensity. Macroinvertebrates were identified in 3 classes within the order Mollusca, including Bivalva, Gastropoda, a nd Plecypoda. Nearly two-thirds of the wetlands hosted macroinvertebrates in the phy lum Mollusca (n=51). In 5 wetlands over one-third of the macroinvertebrates that were identified belonged to the phylum Mollusca, including SU2 (44.7%), CU8 (44.4 %), SU7 (44.1%), SA4 (37.5%), and CA5 (34.9%). In the low LDI group, 4 wetlands had greater than 5% of the identified macroinvertebrates in the phylum Mollusca, including SA8 (16.7%), PR7 (9.0%), SR4 (8.1%), and CR3 (5.8%). Figure 3-26 shows that the percent of macr oinvertebrates in the family Noteridae decreased with increasing landscape de velopment intensity as expected ( Barbour et al 1996b ). Macroinvertebrates in the family Noteri dae never made up more than 15% of the

PAGE 147

132 0 2 4 6 8 10 12 14 02468Landscape Development Intensity Index%Noteridae Figure 3-26. Macroinvertebrates in the fa mily Noteridae decreased with increasing development intensity (LDI). individuals identified to the family taxonomic level or lo wer at any of the sample wetlands. The family Noteridae falls within order Coleoptera, class Insecta, phylum Arthropoda. The family Noteridae (burrowi ng water beetles), typically inhabit the shallow margins of standing or slow flowing stre ams, or in lentic habitats act as climbers associated with vascular m acrophytes or as burrowers ( Williams and Feltmate 1992 ; White and Brigham 1996 ). Both the larvae and adults of macroinvertebrates in the family Noteridae are aquatic ( Peckarsky et al. 1993 ), one of only 5 families within Coleoptera with both life stages being aqua tic. As a general rule, the percent of macroinvertebrates in the order Coleopter a was found to decrease with increasing development surrounding Florida streams ( Barbour et al. 1996b ). One apparent outlier in the high LDI group was CA4 (4.1%).

PAGE 148

133 0 10 20 30 40 50 60 02468Landscape Development Intensity Index%Scrapers Figure 3-27. Macroinvertebrates that be long to the scraper functional feeding group increased with increasing development intensity (LDI). Functional group metrics One functional group metric was selected for inclusion in the macroinvertebrate WCI, the percent scrapers functional feed ing group. Figure 3-27 shows the percent macroinvertebrates in the scraper functiona l feeding group increased with increasing landscape development intensity (LDI). Th e scraper functional f eeding group included macroinvertebrates that scrape periphyton fr om mineral and organic surfaces and those that browse or graze algal materials. Two outliers in the low LDI group included SA8 (27%), and CR3 (25%). Five wetlands with the highest percent scrapers were found in among all 4 ecoregions and represented wetlands embedded in a mix of urban and agricultural land uses, including SU7 (51%), SA4 (40%), CA5 (40%), PA6 (38%), and CU8 (31%). Nearly one-quarter of the samp le wetlands (n = 19) did not have scrapers identified in the macroinvertebrate samples.

PAGE 149

134 Macroinvertebrate Wetland Condition Index The 6 metrics described above were incl uded in the macroinvertebrate WCI. Figure 3-28 shows the relationship between the macroinvertebrate WCI and LDI. Potential scores for the macroinvertebrate WCI range from 0-60, with higher values representing reference wetlands. Actual sc ores ranged from 5.2 at SU5 (deeply flooded swamp surrounded by industrial land use, LDI = 5.2), to 57.0 at SR3 (surrounded by native pine flatwoods, LDI = 1.0). Ranges vari ed regionally, though th e regional scores were not significantly different for either lo w or high LDI groups. The highest scores in each ecoregion included 4 reference wetla nd, PR8 (40.7), NR3 (52.8), CR6 (50.4), and Figure 3-28. Macroinvertebrate WCI scor es decreased with increasing landscape development intensity index (LDI). 0 10 20 30 40 50 60 02468 LDIMi-WCI Reference Agricultural Urban Macroinverteb r ate WCI

PAGE 150

135 SR8 (57.0), in the panhandle, north, central, and south eco regions, respectively. The lowest scoring wetlands in both the panhandl e and central ecoregions were embedded in pasture, including PA6 (12.9) a nd CA5 (7.2). In the north ecoregion NA4 (surrounded by row crops) scored 10.4. In the south SU5 (s urrounded by urban land use) received the lowest score overall of 5.3. The macroinvert ebrate WCI was signifi cantly correlated with the LDI index (SpearmanÂ’s |r| = 0.62, p < 0.001). A Kruskal-Wallis test suggested a significant difference (H = 36.0, p < 0.001) amon g median macroinvertebrate WCI scores for study wetlands in a priori land use categories. Cluster Analysis Cluster analysis determined 5 categorie s based on macroinvertebrate community composition. Clusters were explained by regions, a priori land use categories, and water level including: 1: south to central low development inte nsity; 2: mixed region low development intensity; 3: north central to panhandle middle deve lopment intensity; 4: northern to panhandle middle development inte nsity; and 5: high development intensity and southern Everglades. Figure 3-29 s hows that based on macroinvertebrate WCI scores, clusters 1 and 2 were not significan tly different from one another, but were significantly different from clus ter 5. Clusters 3 and 4 were significantly different from cluster 1 and cluster 5. Table 3-38 provi des means and standard deviations for macroinvertebrate WCI and LDI scores of the 5 clusters. Wetland Condition Index In total 19 metrics were used to constr uct the WCI, including 7 metrics based on the diatom assemblage, 6 metrics based on the macrophyte assemblage, and 6 metrics based on the macroinvertebrate assemblage. Table 3-39 lists the WCI scores for 118 isolated forested wetlands for each assemblage. Scores ranged from 0-70 for the diatom

PAGE 151

136 0 10 20 30 40 50 60 Cluster 1 Macroinvertebrate WCI Cluster 5 Cluster 4 Cluster 3 Cluster 2a ab b b c Figure 3-29. Macroinvertebrate WCI sc ores for 5 wetland clusters based on macroinvertebrate community composition. Clusters with similar letters were not significantly diffe rent (FisherÂ’s LSD, p<0.05). WCI and from 0-60 for both the macrophyte a nd macroinvertebrate WCI. Different wetlands received the highest and lowest scor es for each WCI. The highest scores for each WCI were found at wetlands in the s outh and central ecoregions, including SR2 (diatom WCI = 68.9), CR11 (macrophyte WCI =59.0), and SR8 (macroinvertebrate WCI= 57.0). Minimum WCI scores we re found among 3 different ecoregions, includingCA3 (diatom WCI = 7.9), NA1 (macrophyte WCI = 0.0), and SU2 (macroinvertebrate WCI = 5.3). Within the north and central ecore gions some wetlands received the ecoregion maximum scores for multiple assemblages, including NR3 (diatom WCI = 66.8; macroinvertebrate WCI =52.8) and CR6 (diatom WCI = 65.5; macroinvertebrate WCI = 50.4). In the panhandle ecoregion, PU4 received minimum WCI scores for both the diatom and macrophyte assemblages (diatom WCI = 10.5; macrophyte WCI = 4.0). Figure 3-30 shows a thr ee dimensional scatter plot for the 50

PAGE 152

137 Table 3-38. Macroinvertebrate WCI scores a nd LDI values for wetland clusters based on macroinvertebrate community composition. Cluster 1 Cluster 2 Cluster 3 Cluster 4 Cluster 5 Mi-WCI 41.6 ± 10.7a 35.5 ± 10.8ab 29.3 ± 11.8b 28.3 ± 6.2b 21.9 ± 9.7c LDI 2.0 ± 1.5a 3.1 ± 2.5ab 3.4 ± 2.3ab 3.4± 2.3ab 4.0 ± 2.1b Clusters with similar letters were not significantly different (p<0.05) wetlands receiving scores for all three asse mblages. The maximum diatom WCI for the 50 wetlands graphed was 68.9 (of 70) at SR 2. The maximum macrophyte WCI was 58.4 (of 60) at PR6 (LDI = 1.3), and the maximu m macroinvertebrate WC I was 52.8 (of 60) at NR3 (LDI = 1.0). The 1:1:1 line is shown for convenience in interpretation. Figure 3-31 shows 2 dimensional comparis ons of wetlands scored with multiple assemblages, including (A) 50 wetlands with diatom and macrophyte WCI scores, (B) 50 wetlands with diatom and macroinvertebrate WCI scores, and (C) 79 wetlands with macrophyte and macroinvertebrate WCI scores. The most obvious outlier between the diatom and macrophyte WCI scores (Figure 3-31 A) was at CU6, a wetland surrounded by a golf course that had been developed within the past five years. CU6 had low scores for the diatom (15.1 of 70) and macroinverteb rate (23.4 of 60) but a higher score for the macrophyte (41.5 of 60) WCI. The 19 metrics incorporated into the WCI were compared across different assemblages (Table 3-40). Of the 120 poten tial metric comparis ons (42 among diatom and macrophyte metrics, 42 among diatom and macroinvertebrate metrics, and 36 among macrophyte and macroinvertebrate metrics), 53 % of the comparisons were significantly correlated at the p < 0.01 level. An add itional 20% of the poten tial comparisons were significantly correlated at the more flexible p < 0.05 level; and an additional 6% at the more flexible p < 0.10 level. Less than onequarter of the compar isons among metrics of

PAGE 153

138 Table 3-39. WCI scores for 118 wetlands base d on three assemblages including diatoms, macrophytes, and macroinvertebrates. Site Code Diatom WCI Macr ophyte WCI MacroinvertebrateWCI PA1 30.4 PA2 38.1 11.9 30.1 PA3 34.9 8.3 25.1 PA4 12.6 PA5 51.1 6.5 21.2 PA6 28.2 7.7 12.9 PA7 17.7 PA8 50.6 PA9 12.1 PA10 41.7 PR1 61.1 55.9 37.6 PR2 50.5 PR3 49.5 PR4 64.5 51.2 40.0 PR5 58.0 53.6 30.0 PR6 63.9 58.4 34.4 PR7 34.8 26.5 PR8 53.6 40.7 PU1 6.2 PU2 31.5 PU3 33.1 31.0 35.7 PU4 10.5 4.0 21.6 PU5 22.1 PU6 16.5 PU7 24.1 PU8 33.6 PU9 48.8 PU10 9.2 30.2 NA1 0.0 NA2 3.0 NA3 56.4 NA4 33.8 16.3 10.4

PAGE 154

139 Table 3-39. Continued. Site Code Diatom WCI Macrophyt e WCI Macroinvertebrate WCI NA5 2.9 NA6 56.3 18.8 16.9 NA7 37.0 NA8 46.0 NA9 37.3 NA10 51.5 30.0 NA11 32.6 28.7 NA12 8.0 NR1 52.0 NR2 65.8 34.8 30.0 NR3 66.8 58.2 52.8 NR4 58.3 42.2 39.8 NR5 52.3 NR6 57.9 55.0 48.6 NR7 52.3 NR8 58.4 30.0 NR9 56.7 33.0 NU1 35.2 NU2 24.1 23.7 15.5 NU3 25.6 NU4 54.5 35.1 31.0 NU5 60.0 40.1 24.0 NU6 48.8 20.7 28.0 NU7 11.8 NU8 38.6 NU9 37.5 NU10 17.2 23.0 CA1 8.9 CA2 10.6 0.7 19.4 CA3 7.9 7.1 20.0 CA4 56.9 38.8 31.3 CA5 43.6 26.9 7.2 CA6 22.7 7.1 21.1 CA7 9.8 32.1

PAGE 155

140 Table 3-39. Continued. Site Code Diatom WCI Macrophyt e WCI Macroinvertebrate WCI CA8 37.7 31.3 CA9 11.8 22.6 CR1 51.0 CR2 49.9 CR3 57.7 47.6 33.9 CR4 57.8 51.2 48.9 CR5 43.8 43.5 29.7 CR6 65.5 54.6 50.4 CR7 51.7 CR8 54.3 28.8 CR9 49.4 34.4 CR10 53.5 45.0 CR11 59.0 49.5 CU1 61.1 42.9 40.6 CU2 10.0 CU3 28.5 13.5 22.3 CU4 21.4 CU5 21.5 22.3 17.8 CU6 15.1 41.5 23.4 CU7 20.7 10.6 CU8 21.1 10.1 CU9 28.3 28.3 CU10 38.3 32.3 CU11 21.3 34.1 SA1 0.7 SA2 34.1 9.4 15.0 SA3 47.9 23.1 28.6 SA4 15.8 11.3 9.1 SA5 46.3 18.9 19.0 SA6 31.9 3.7 19.0 SA7 30.8 29.8 SA8 34.5 11.0 SA9 29.8 17.9 SR1 66.8 54.1 33.4

PAGE 156

141 Table 3-39. Continued. Site Code Diatom WCI Macrophyt e WCI Macroinvertebrate WCI SR2 68.9 50.8 46.6 SR3 51.6 51.2 26.4 SR4 43.7 57.9 28.2 SR5 39.4 49.8 38.4 SR6 41.0 51.8 39.0 SR7 49.9 42.7 SR8 47.5 57.0 SR9 50.1 43.4 SU1 17.2 17.8 22.1 SU2 46.2 20.3 15.2 SU3 31.7 42.6 35.4 SU4 42.3 21.8 18.9 SU5 38.9 23.9 5.3 SU6 46.1 28.1 21.0 SU7 12.5 9.1 SU8 2.7 23.3 SU9 20.4 32.3 SU10 11.7 different assemblages were not significantly co rrelated (22%). The strongest correlation among metrics of different assemblages wa s between the diatom sensitive indicator genera and the macrophyte sensitive indica tor species (PearsonÂ’ s r = 0.74, p < 0.01). Twelve of the metric comparisons between the diatom and macrophyte assemblages were strongly significant (|r| 0.60, p < 0.01). Only 2 of the comparisons between the diatom and macrophyte assemblages were not significantly correlated (p < 0.10). Comparisons between the diatom and macroi nvertebrate metrics were not as strong, with less than 50% of the metrics significan tly correlated (p < 0.10). Only 6 diatom and macroinvertebrate metrics were correlated at the more stringent p < 0.01 level. The

PAGE 157

142 0 0 10 10 20 20 30 40 30 50 60 40 5 0 70 60 50 40 30 20 10 0 0 60 Reference Agricultural Urban 1 : 1 : 1 L i n eMacroinvertebrate WCIMacrophyte WCI Diatom WCI Figure 3-30. Three dimensional scatter plot of the WCI based on three assemblages, including diatoms, macrophytes , and macroinvertebrates. macroinvertebrate metrics Noteridae and scra pers were not significantly correlated with any of the diatom metrics. Correlati ons were stronger among the macrophyte and macroinvertebrate comparisons, with 94% of the comparisons significantly correlated (p < 0.10). In fact, 20 of the metric comparis ons (56%) were correlated at the strictest significance level of p < 0.01. Two metric comparisons between the macrophyte and macroinvertebrate metrics were not signi ficantly correlated, including the macrophyte wetland status and macroinvertebrate Mo llusca metrics and between the macrophyte exotic and macroinvertebrate scrapers metrics.

PAGE 158

143 0 10 20 30 40 50 60 010203040506070 Diatom WCIMacrophyte WCI 0 10 20 30 40 50 60 010203040506070 Diatom WCIMacroinvertebrate WC I 0 10 20 30 40 50 60 0102030405060 Macrophyte WCIMacroinvertebrate WC I Reference Agricultural Urban Figure 3-31. Scatterplots of WCI scores for wetlands ba sed on diatom, macrophyte, and macroinvertebrate assemblages. A) Diatom and macrophyte WCI scores (n = 50 wetlands). B) Diatom and macroinve rtebrate WCI scores (n = 50). C) Macrophyte and macroinvertebra te WCI scores (n = 79).

PAGE 159

144 Table 3-40. Pearson correlations among 19 metrics. Macrophytes Diatoms Tolerant Sensitive MFQI Exotic Native Perennial Wetland Status Tolerant 0.60* -0.52* -0.52* 0.49* -0.60* -0.26# Sensitive -0.59* 0.74* 0.68* -0.53* 0.57* 0.42* Pollution Class 1 0.60* -0.50* -0.51* 0.59* -0.60* -0.36^ Nitrogen Class 3 0.54* -0.55* -0.46* 0.61* -0.58* Saprobity Class 4 0.51* -0.47* -0.44* 0.61* -0.60* -0.34^ pH Class 3 0.57* -0.60* -0.63* 0.62* -0.58* -0.39* Oxygen Class 1 -0.57* 0.68* 0.56* -0.57* 0.60* Macroinvertebrates Diatoms Tolerant Sensitive FL In dex Mollusca Noteridae Scrapers Tolerant 0.49* 0.29^ -0.27# Sensitive -0.48* 0.38* 0.40* -0.33^ Pollution Class 1 0.25# -0.29^ Nitrogen Class 3 0.33^ -0.29^ -0.30^ Saprobity Class 4 -0.30^ pH Class 3 0.50* -0.30^ -0.25# Oxygen Class 1 -0.45* 0.32^ 0.35^ -0.33^ Macroinvertebrates Macrophytes Tolerant Sensitive FL Index Mollusca Noteridae Scrapers Tolerant 0.47* -0.41* -0.30* 0.27^ -0.25^ 0.23^ Sensitive -0.44* 0.54* 0.31* -0.30* 0.31* -0.27^ MFQI -0.47* 0.34* 0.28^ -0.33* 0.21# -0.30* Exotic 0.48* -0.32* 0.28^ 0.27^ -0.23^ Native Perennial -0.45* 0.33* 0.29* -0.30* 0.27^ -0.22# Wetland Status -0.35* 0.37* 0.28^ 0.26^ -0.22# * p < 0.01 ^ p < 0.05 # p < 0.10

PAGE 160

145 CHAPTER 4 DISCUSSION While previous research has identifie d responses of wetland ecosystems to individual changes (such as increased nutrien ts or altered hydrology), few have combined multiple biotic components, environmental parameters, and landscape development intensity in an attempt to quantify ecological integrity. The contribu tion of this research to our understanding of changes in the comm unity composition of isolated forested wetlands (based on the diatom, macrophyte, and macroinvertebrate assemblages) in relation to different development intens ities in the surrounding landscape can be summarized in 6 main points. First, the richness, evenness, and diversity of each assemblage were not sensitive to different land uses or development intens ities in the surrounding landscape. Second, biological indicators along with physical and chemical parameters were useful in defining biological integrity. Third, the variable turnover times and sensitivities of the 3 assemblages (diatoms, macrophytes, macroinve rtebrates) suggest that a multi-metric multi-assemblage Wetland Condition Index (WC I) has more merit than a WCI based on a single assemblage. Fourth, regionalization may strengthen the WCI. Fifth, a WCI independent of wetland type may be feasible, given the strong likeness of the forested WCI to the marsh Index of Wetland Condition (IWC) ( Lane 2003 ). Sixth, urban wetlands exhibit a different vector of change than do agricultural wetlands, and while the WCI suggests low biological integrity of both agricultural and urban wetlands, these wetlands do provide services and do work in the environment.

PAGE 161

146 Richness, Evenness, and Diversity Measures of richness, evenness, and di versity of the diatom, macrophyte, and macroinvertebrate assemblage were not sensitive to difference in land use or development intensity in the surroundi ng landscape. For both the diatom and macroinvertebrate assemblages neither a priori land use classificati on nor categories of landscape development intensity showed significa nt differences in ri chness, evenness, or diversity calculations. Differences in macrophyte evenness and diversity between reference and agricultural wetlands (Table 3-13 ; 3-14) may be attributable to both direct (for example grazing by domes tic cattle) or more indirect (increased nutrients from fertilizer carried in run-off from surr ounding agricultural fields) activities in the surrounding landscape. However, macrophyte evenness and diversity were higher for wetlands surrounded by more developed land uses, contrasting earlier findings on decreases in plant diversit y from grazing pressures ( Blanch and Brock 1994 ; Grace and Jutila 1999 ) and nutrient enrichment ( Bedford et al. 1999 ). Mitsch and Gosselink ( 1993 ) report that freshwater forested wetlands have low species diversity, so perhaps macrophyte species that enter wetlands in developed landscape are merely taking advantage of available habitat and are in f act increasing the over all species diversity. The increased incidence of exotic species have long been associated with disturbed ecosystems ( Cronk and Fennessy 2001 ; Galatowitsch 1999b ), suggesting more specifically that as anthropoge nic development intensity in creases, the incidence of exotic species may escalate. An increase in the frequency of exotic species has been attributed to drainage and hydrologic alterations ( Hobbs and Heunneke 1992 ; David 1999 ; Galatowitsch et al. 1999b ), increased human development ( Cronk and Fennessy 2001 ), and ecosystem scale alteration s such as clear-cut harvests ( Devine 1998 ). Within

PAGE 162

147 the study wetlands, the percent of exotic m acrophyte species increa sed with increasing development intensity in the surrounding lands cape (Figure 3-16). The influx of exotic species added to, rather than diminished, th e species evenness and diversity within the isolated forested wetlands sampled. Describing Biological Integrity Biological indicators along with chemical a nd physical parameters were useful in determining the biological integrity of isolat ed forested wetlands. For the purposes of this study, biological integrity has been defined quantitative ly with the WCI. The WCI incorporates 19 metrics from 3 different sp ecies assemblages (dia toms, macrophytes, and macroinvertebrates). Correlations be tween the diatom, macrophyte, and macroinvertebrate WCIs and the intensity of development in the surrounding landscape (based on the use of nonrenewable ener gy and calculated with the Landscape Development Intensity (LDI) index) suggest that changes in community composition were captured by the WCI. It has been suggested that organisms respond to environmental gradients by colonizing a range of feasible conditions beyond which the organisms fail to persist ( ter Braak 1987 ). By selecting species that occur throughout the range of measurable environmental parameters , the WCI defined and detected deviations from the condition of reference wetlands base d on community composition. Each of the 19 WCI metrics addressed some disparity from the assumed range of feasible conditions. For all 3 assemblages, the tolerant indi cator species metric demonstrated the strongest correlation with LDI (Tables 3-7; 318; 3-33), suggesting that the presence of a suite of taxa characteristic of wetlands w ith low biological integrity may be the single most effective means of identifying change s in community composition. The isolated forested wetlands sampled were influenced by various anthropogeni c activities (from

PAGE 163

148 direct herbivory and trampling by domestic cattle, to increased nutrients from agricultural or stormwater run-off, to hydrological impoundm ents or drainage), yet despite the vast differences in surrounding land uses the community composition of these wetlands was similar enough to detect a universal suite of tolerant indicator species. Clustering the isolated forested wetlands based on the 3 assemblages separately suggested that differences in some agricultu ral and urban development intensities may be too subtle to detect with compositional data (Figures 3-10; 3-20; 3-29). Furthermore, greater variability in the m acroinvertebrate assemblage of reference wetlands as compared to that of agricultural and urban wetlands (Table 3-31) suggested that perturbations to the driving energies in isol ated wetlands may result in a convergence of the taxa present. Indeed, the natural com positional variability inherent among reference wetlands may be lost with increased devel opment intensity in the surrounding landscape. While the WCI can not be used to predict changes in the physical and chemical parameters of a wetland, its strength lies in providing an overview of biological integrity through the integration of changes in commun ity composition from cumulative effects. Among a priori land use categories, differences in water and soil parameters were apparent (including dissolved oxygen, color, turbidity, water column pH, specific conductivity, water ammonia-nitr ogen, water TKN, water TP, soil moisture, soil organic matter, and soil TP; Table 3-1). When soil a nd water parameters were used to explain variation in the community composition of each assemblage, water column pH was universally identified (Tables 3-6; 3-17; and 3-32). Additionally, total phosphorus concentrations explained some of the variance in both the diatom and macrophyte assemblages. Perhaps preservation and re storation strategies could focus on limiting

PAGE 164

149 activities that influence changes to water column pH and total phosphorus loading to wetlands in order to prom ote biological integrity. Merits of a Multi-Metric Multi-Assemblage WCI The variable turnover times and sensitiv ities of the 3 different assemblages (diatoms, macrophytes, macroinvertebrates) su ggest that a multi-metric multi-assemblage WCI has more merit than a WCI based on a si ngle assemblage. Diatoms have short life cycles and live within the phys ical and chemical environment of the water column. As such, they act as integrators of ecosyst em condition, with a rapid reaction time to environmental perturbation. Changes to the physical, chemical, and/or biological characteristics of a wetland influence the intr icate interactions diatoms have with their environment. Macrophytes have longer life cy cles than diatoms and macroinvertebrates, and as such they act as integrators of bot h present and historic changes in driving energies. The reaction time of the macrophyte assemblage to changes in driving energies is likely slower and more buffered than that of the other species a ssemblages. Unlike the diatom and macrophyte assemblages, macr oinvertebrates may be able to abandon unsuitable habitats, and so thei r occurrence may reflect only th e suitability of the recent wetland environment. Many macroinvertebrate s have short life cycles, and many have multiple generations per year. Still others requi re over wintering in saturated soils or on wetland vegetation, suggesting an intimate re lationship between m acroinvertebrates and their individual environment. The WCI can be used to infer influences in temporal and spatial changes to which a particular wetland has been exposed. For ex ample, diatoms have rapid turnover times and may react immediately to shifts in driv ing energies. On the other hand, perennial macrophytes may respond to changes over a long er period of time, particularly in the

PAGE 165

150 case of the woody midand over-story species. While the macrophyte WCI score may remain relatively high in a recently enriched wetland, the diatom WCI score may reflect lower biological integrity. While macrophyt es assimilate nutrients for growth, this process has a longer time frame than the ra pid growth rate ty pical of the algal assemblage. An explosion of algal growth ma y in turns alter the av ailable food resources within a wetland affecting other assemblages, for example there may be an increase in macroinvertebrate algae scra pers, and decline in the macroinvertebrate WCI score. While agreement in the ranking of the biol ogical condition of study wetlands using the WCI was anticipated, discrepanc ies among the ranking from the different assemblages may provide greater insight in to wetland condition as different species assemblages respond to changes in driving ener gies over different tim e scales. There was variation among the ranking of wetla nds for the diatom, macrophyte, and macroinvertebrate WCI; though there were no obvious outliers when the three assemblages were compared (Figure 3-30). While the a priori reference wetlands were generally differentiated from the agricultural and urban wetlands, differences between the agricultural and urban land uses were not as apparent (Figure 3-31). Many of the metrics of the different a ssemblages were significantly correlated (Table 3-40), none were correlate d at the threshold (|r| > 0.90) used to exclude candidate metrics from inclusion in the WCI. Perhaps the value of each of the 19 selected metrics is inherent in its differen tiation between categories of landscape development intensity (Tables 3-8, 3-19, and 3-34). Diatom and m acrophyte metrics were strongly correlated with one another, and yet diat om and macroinvertebrate metric s were not, reinforcing the value of including various species assemblages in an assessment of biological integrity.

PAGE 166

151 Perhaps, with regular biological monitoring it may be possible to further explore the temporal effects of chan ging development intensity. A Case for Regionalization The climate of Florida is consider ed humid subtropi cal, though pronounced differences occur in the local climate across Fl orida, such as differences in the amount of annual rainfall, seasonal maximum temp eratures, and number of freeze days ( Fernald and Purdum 1992 ; Lane 2000 ). The latitudinal ra nge of the wetlands sa mpled in this study was 31.0ºN at PA4 in Escambia County th e western most county in the Florida panhandle, to 26.0ºN at SR4 in Collier County in southwest Florida. The longitudinal range was from 87.5ºW at PA 4 to 80.1ºW at SU8 in Palm Beach County, along the southeastern coast of Florida. Despite the broad latitudinal and longitudinal ranges of sample wetlands throughout Florida, statew ide significant difference in water and soil parameters among a priori land use categories were detect ed (Table 3-1), suggesting the statewide scale may be appropriate for a phys ical and chemical assessment of wetland condition. The influence of latitude and longitu de was reflected in the compositional difference of all 3 assemblages found among th e Florida ecoregions (Table 3-5, 3-15, and 3-31). Latitude and longit ude were significantly correla ted with macrophyte community composition (Figure 3-11), and latitude expl ained partial variance in macroinvertebrate community composition (Figure 3-21). In a ddition, wetlands in the Florida Everglades were outliers in many of the diatom metrics (Fig ures 3-3; 3-5; 3-7; 3-8), and the southern Everglades wetlands formed distinguished cl usters based on diatom (Figure 3-10) and macroinvertebrate (Figure 3-29) community composition.

PAGE 167

152 Most of the human development in Flor ida has occurred along the east and west coastal areas of pe ninsular Florida ( Fernald and Purdum 1992 ), suggesting that while the reference wetlands selected in the south and central ecoregions were seemingly the best possible examples of reference type c onditions, they may be more affected by development in the surrounding landscape (suc h as compounded secondary effects) than their panhandle and far north ecoregion counterparts. Regionalization was explored for the macrophyte assemblage because of the sufficient number of wetlands sampled within each ecoregion. There were clear regional differences in the statewide macrophyte WCI scores for wetlands in the low LDI group, which down scored the reference wetlands of the south and central ecoregion (Table 325). This led to the use of regionalized sc oring of the macrophyte metrics. While the ease and utility of a single statewide WC I would seemingly prevail over 4 regional indices, the necessity of sc oring each ecoregion based on the best possible reference conditions ( Karr and Chu 1999 ) cannot be overlooked. Regi onalization of biological indices has been suggested thr oughout the literature. The main reason for classification is to compare “like to like” ( Gerristen et al. 2000 ), that is, to reduce the noise in background variability in biological data. Differences in the macroinvertebrate community composition among ecoregions may be of importance in improving the macroi nvertebrate WCI. For example, none of the sample wetlands in the panhandle ecoregi on hosted macroinvertebrates in the order Trichoptera (caddis flies), whereas no wetland in the north ecoregion hosted macroinvertebrates in the order Ephemeropter a (mayflies). While both of these orders

PAGE 168

153 are characteristic of lotic environments ( Edmunds and Waltz 1996 ), the absence of an entire order from ecoregions suggests th e value of regionalization of the WCI. WCI Independent of Wetland Type Recent work by Lane ( 2003 ) presents a 24 metric Index of Wetland Condition (IWC) for Florida marshes based on the community composition of the diatom, macrophyte, and macroinvertebrate assemb lages. The IWC was created based on 75 isolated depressional freshwater mars hes surrounded by undeveloped (n=34) and agricultural (n=40) land uses throughout peninsular Florida. Of the 14 metrics based on the diatom assemblage, the forested WCI a nd the marsh IWC share 7 metrics, including all of the diatom WCI metrics. Two of these 7 shared metr ics were based on tolerant and sensitive indicator species analyses, which were determined separately for each wetland type. Shared species were limited between wetland types, as the tolerant indicator species list had only 2 mutual species ( Navicula confervacea and N. minima ), of 12 species for the forested WCI and 21 species for the marsh IWC. Si milarly, the sensitive indicator species lists shared only 5 species ( Eunotia flexuosa, E. naegelii, E. rhomboidea, Frustulia rhomboids , and F. rhomboids crassinervia ), of 18 species for the forested WCI and 22 species for the marsh IW C. The 5 remaining diatom metrics were based on autecological relationshi ps, including pollution class 1 ( Bahls 1993 ), nitrogen class 3 ( van Dam et al. 1994 ), saprobity class 3 ( van Dan et al. 1994 ), pH class 3 ( van Dan et al. 1994 ), and dissolved oxygen class 1 ( van Dan et al. 1994 ). Of the remaining 7 marsh IWC diatom metrics, 5 were consider ed too similar to selected forested WCI metrics and were excluded to avoid redundanc y. The final 2 marsh IWC diatom metrics were based on salinity class ( van Dam et al. 1994 ), and were not significantly correlated with LDI for forested wetlands.

PAGE 169

154 Five macrophyte metrics were incorporated into the marsh IWC, and variants of these were included in the metrics of the forested WCI. Tolerant and sensitive indicator species lists were constructed separately for each wetland type. Of the 46 statewide tolerant macrophyte indicator sp ecies for the marsh IWC, 28 al so occur on the statewide tolerant macrophyte indicator sp ecies list for the forested WC I (of 61 species). Similarly, 20 statewide sensitive macrophyte indicator sp ecies were shared for the marsh IWC (of 36) and the forested WCI (of 69). The 3 a dditional metrics included in the marsh IWC were percent exotic species, annual to perennial ratio, and a metric based on scores from a Floristic Quality Assessment Index (simila r to the one conducted in this study, but specific to marshes). In the forested WCI, a variant of the annual to perennial ratio was used, the percent native perennial species (to account for anticipate d conditions at urban wetlands). The sixth forested WCI metric was the percent wetland status species. There was less similarity between the 5 macroinvertebrate marsh IWC metrics and the 6 forested macroinvertebrate WCI metrics. Tolerant and sensitive indicator metrics were constructed separately for each wetland type and were included in both indices. Three tolerant indicator ge nera occurred on both lists ( Goeldichironomus, Micromenetus, and Physella ), and only 1 sensitive indi cator genera was shared ( Larisa ). The marsh IWC included 3 additional metrics: %Preda tors, %Odonata, and %Ort hocladiinae. The forested WCI included 4 different metrics: Florida Index, %Mollusca, %Noteridae, and %Scrapers. Overall, the marsh IWC and the forested WCI were similar, with many shared metrics. Some additional variability between selected metrics was expected as the forested WCI included 2 additional sources of variability (wetla nds in the panhandle

PAGE 170

155 ecoregion and urban land uses, which were not included in the sample wetlands for the marsh IWC). Perhaps the strong similarity of metrics suggests that a universal assessment index could be constructed regard less of wetland type. However, it would likely be necessary to maintain independent indicator species lists specific to wetland type. Wetland Value Urban wetlands appear to exhibit a different vector of change than do agricultural wetlands; however the WCI did not significantl y differentiate between agricultural and urban wetlands (Figures 3-30; 3-31). The LDI also did not specifically differentiate between these land use categories, as Lands cape Development Coefficients (LDC) for agricultural land uses range from 2.6 (unimproved pastur e) to 6.6 (high intensity agriculture) (Table 2-3). Ur ban LDCs overlap that range with variants of Open Space/Recreational land uses ranging from 2.1 (low intensity) to 4.8 (middle intensity) to 6.9 (high intensity). Simila rly, other measures of anth ropogenic influence like the Wetland Rapid Assessment Procedure and the Minnesota disturbance index ( Appendix B ) did not clearly differentiate betw een agricultural and urban land uses. The main conclusion we can draw from the WCI is that both agricultural and urban wetlands have lowered biological integrity. However, this statement is not meant to imply that these wetlands lack value, as th ey provide important se rvices and do work in the environment. Wetlands embedded in a developed landscape matrix provide an abundance of potential services. For exampl e, they may store and purify stormwater, process nutrients and toxins (perhaps ac ting as a sink and prot ecting hydrologically connected systems), provide habitat for lo cal wildlife and perhaps migratory species, produce oxygen, filter the air, provide noise abatement, and act as refugia for urban

PAGE 171

156 ecologists. Specifically in the case of urban wetlands, there is a debate as to the value of small remnant wetlands embedded within in hi ghly developed landscape matrices. While wetlands do exist in highly urbanized areas, the do not appear to closely resemble wetlands in undeveloped landscapes. Under current Florida law, mitigation ratios for urban wetlands will be small, and some people may question the idea of keep ing urban wetlands of marginal biological integrity on expensive real estate parcels. Perhaps mitigating off-site into near-by areas with low development intensity would improve the chances of creating or restoring a wetland with the possibility of successfully meeting mitigation criteria. However, offsite mitigation undervalues the services provided by urban wetlands. Urban wetlands clearly provide some function, and percha nce they are doing more work processing nutrients, storing urban stormw ater run-off, and storing toxins, than wetlands in undeveloped landscapes. The continued existenc e of urban wetlands is crucial (for the maintenance of biotic diversity, buffering po llution and contamination to protect nearby environments, increasing oxygen production in an urban center, etc.). While the WCI scores for urban wetlands reflect lowered biological integrity, pe rhaps having 30-70% on average of the biological integrity of refere nce wetland is more important than having no wetland and therefore no services or free wo rk. Wetlands with the lowest biological integrity could have scored a 0 for the WCI, and yet only 1 agricultural wetland did (NA1, for the macrophyte assemblage). Ther e is no doubt that the intensity of human land use across the landscape plays a role in th e loss of biological integrity of wetlands, however we should reconsider ou r willingness to remove all of the biological integrity of a wetland by otherwise erasi ng its existence by filling.

PAGE 172

157 Limitations and Further Research Several limitations to this study should be noted, including sampling methods and drought conditions. One water sample was coll ected to represent the water environment of the entire sample wetland, and water sa mples were taken at a range of times throughout the day. While water samples were al ways taken first when a crew arrived at a sample wetland, the time of day the crew ar rived fluctuated. Additionally, while an attempt was made to avoid taking water samples immediately following extreme rain event, there is the pos sibility that the sample was take n during a period of time without rain. Therefore, there was no consistency as to recent weather conditions when water samples were taken. There were also stri ct requirements of preservation, temperature control, and shipping protoc ols associated with the wa ter samples. When these requirements were breached the sample had to be discarded. Similarly, one composite soil sample was ta ken for each sample wetland, and bulk density was not measured, which complicates the use of soil nutrien t data. As well, generally wetlands were visited only once, with a complete sample effort lasting just one day. This provided a mere snapshot of we tland condition. Revisits were conducted at some wetlands to collect water, soil, algae, or macroinvertebrates in the case of dry conditions on the initial visit or a discar ded sample (generally for quality control reasons). Visiting these wetla nds only once or twice did no t allow insight into seasonal or yearly variations in the assemblages. As an additional confounding factor, Florida experienced drought conditions in 2001, and the macrophyte assemblage at many wetlands was sampled without standing wate r, which allowed many flood intolerant species to encroach into the sample wetlands.

PAGE 173

158 While the WCI has satisfactorily distinguished between wetlands embedded in an array of land uses with varying development intensities, much needs to be done to insure accuracy and usability. First, seasonal and yearly variation should be identified for the study wetlands. Wetlands are pulsing systems, and as such wetland organisms must adapt to wide fluctuations in hydrology, temperature, salinity, and dissolved oxygen ( Evans et al. 1999 ; Leslie et al. 1999 ; Sharitz and Batzer 1999 ). A new set of wetlands should be sampled and scored based on the WCI to test the reliability of this index. The WCI was limited to nineteen metrics due to the redundant nature of many of the candidate metrics, as well as the high va riability of species composition within the dataset. A larger sample size could im prove the significance of the WCI based on ecoregions for metrics such as indicator sp ecies analysis. Regi onalization may be an important step in refining the WCI, as th is study was somewhat limited to a statewide approach due to small sample sizes within each ecoregion. Conclusions The use of 3 separate species assemblage s for a biological assessment of isolated forested wetland provided a useful tool fo r detecting changes in biological integrity associated with changes in the driving ener gies of a wetland measured through landscape development intensity. While richness, ev enness, and diversity measures were not particularly sensitive to changes in landscap e development intensity, biological indicators along with physical and chemical parameters we re useful in defining biological integrity. In the future a multi-metric multi-assemblage WCI could be constructed for all freshwater wetlands throughout th e state of Florida, with sp ecific indicator species and metric scores based on Florida ecoregions . While the WCI suggests low biological integrity of both agricultural and urban wetla nds, these wetlands pr ovide services and do

PAGE 174

159 work in the environment. Therefore, th e quantitative score of biological integrity established through the WCI should not be used as a surrogate for wetland value, but an objective, quantitative means of compari ng changes in community composition along gradients of development intensity.

PAGE 175

160 APPENDIX A ENERGY CIRCUIT LANGUAGE Table A-1. Symbols used in energy circuit diagramming. Symbol Name Description System Boundary Defines the system being di agrammed. Lines that cross the system boundary indicate inflows and outflows of the system. Energy Circuit A pathway which has a flow proportional to the quantity in the storage or source upstream. Source Outside source of energy deliv ering forces according to a program controlled from outside; a forcing function. Flow Limited Source Outside source of energy with a flow that is externally controlled. Storage Tank A compartment of energy storage within the system storing a quantity as th e balance of inflows and outflows. Sensor The sensor (tiny square box on storage) suggests the storage tank controls some other flow but does not supply the main energy for it. Producer Unit that collects and transforms low-quality energy under control interactions of high-quality flows. Consumer Unit that transforms energy qua lity, stores it, and feeds it back autocatalytically to improve inflow. Heat Sink Dispersion of potential energy into heat that accompanies all real transformation processes and storages; loss of potential energy from further use by the system.

PAGE 176

161 APPENDIX B QUANTIFYING ANTHROPOGENIC INFLUENCE Table B-1. LDI, WRAP, and Minnesota di sturbance index scores for 118 wetlands. Site Code LDI WRAP Minnesota Disturbance Index Site Code LDI WRAP Minnesota Disturbance Index SA1 6.3 6.9 16 CA1 6.1 5.8 16 SA2 4.6 4.6 14 CA2 5.0 6.3 9 SA3 4.6 5.5 9 CA3 4.9 8.1 14 SA4 6.2 7.5 17 CA4 4.3 3.4 5 SA5 4.0 4.5 11 CA5 5.1 5.5 13 SA6 5.0 6.0 9 CA6 5.4 4.8 17 SA7 2.8 2.6 2 CA7 4.4 5.7 8 SA8 1.3 3.6 5 CA8 1.9 2.7 9 SA9 4.8 3.2 6 CA9 5.4 6.3 17 SR1 1.0 1.0 0 CR1 1.0 1.1 5 SR2 1.0 1.2 0 CR2 1.0 1.8 1 SR3 1.0 1.4 3 CR3 1.0 1.5 2 SR4 1.0 1.0 1 CR4 1.1 1.4 0 SR5 1.2 1.0 1 CR5 1.0 1.0 0 SR6 1.0 1.5 2 CR6 1.0 1.3 0 SR7 1.1 1.8 3 CR7 1.1 1.8 6 SR8 1.0 1.7 2 CR8 1.0 3.0 2 SR9 1.0 3.2 3 CR9 1.0 1.3 2 SU1 5.0 6.2 14 CR10 1.5 1.8 5 SU2 4.8 6.0 8 CR11 1.0 1.5 0 SU3 1.3 5.9 15 CU1 2.1 2.4 7 SU4 1.5 4.9 8 CU2 3.9 6.0 7 SU5 5.2 6.3 17 CU3 7.1 6.4 17 SU6 3.9 5.5 15 CU4 4.5 6.0 13 SU7 5.1 7.1 13 CU5 6.4 6.1 12 SU8 7.2 8.4 17 CU6 7.2 6.0 13 SU9 6.2 4.6 16 CU7 5.6 6.4 16 SU10 3.2 6.0 8 CU8 4.4 5.1 16 CU9 7.0 7.2 14 CU10 3.8 5.8 10 CU11 3.3 6.2 15

PAGE 177

162 Table B-1 Continued. Site Code LDI WRAP Minnesota Disturbance Index Site Code LDI WRAP Minnesota Disturbance Index NA1 5.1 6.9 13 PA1 3.1 4.7 9 NA2 4.9 6.3 9 PA2 5.0 6.2 18 NA3 2.1 2.9 4 PA3 6.8 6.1 19 NA4 6.2 6.8 19 PA4 6.6 6.8 18 NA5 5.1 7.5 17 PA5 4.7 6.7 19 NA6 5.3 5.5 16 PA6 4.7 6.0 17 NA7 5.0 4.3 7 PA7 4.9 6.6 17 NA8 2.0 3.1 4 PA8 2.2 4.9 10 NA9 2.2 4.2 5 PA9 5.8 5.4 7 NA10 2.2 3.0 6 PA10 2.0 4.9 6 NA11 2.5 3.9 9 PR1 1.0 1.6 0 NA12 5.5 5.1 7 PR2 1.0 1.3 0 NR1 1.1 2.1 1 PR3 1.1 1.9 0 NR2 1.0 1.5 1 PR4 1.5 1.7 2 NR3 1.0 1.2 1 PR5 1.0 1.0 0 NR4 1.0 1.2 2 PR6 1.3 1.3 0 NR5 1.0 1.4 1 PR7 1.0 1.3 0 NR6 1.1 1.1 0 PR8 1.0 1.2 0 NR7 1.8 1.8 1 PU1 5.3 8.3 17 NR8 1.0 2.5 0 PU2 5.9 5.3 9 NR9 1.0 1.9 1 PU3 6.3 5.6 12 NU1 2.8 3.9 7 PU4 4.8 7.5 17 NU2 4.2 4.4 9 PU5 4.0 6.4 13 NU3 5.3 6.6 10 PU6 4.8 5.1 8 NU4 3.2 5.4 11 PU7 3.8 5.4 10 NU5 6.2 6.2 9 PU8 5.0 5.4 9 NU6 5.6 4.8 15 PU9 3.1 3.9 5 NU7 4.2 5.3 10 PU10 6.5 8.6 17 NU8 3.8 4.8 11 NU9 6.3 5.9 12 NU10 6.6 6.4 17

PAGE 178

163 APPENDIX C STANDARD OPERATING PROCEDURES Standard Operating Procedures (SOPs) have been included for sampling methods employed for the entire project, which included more data collection than that included in this dissertation. These additional methods were included in Appendix C to provide readers with a complete picture of fiel d methodology, and to understand the order of events during field sampling. Data omitted from this dissertation include tree basal area along transects, fisheye canopy photography, algae analysis of epiphyton, metaphyton, and phytoplankton, and dry bent hic algae sampling. Canopy phot o analysis was explored in an earlier thesis by Spurrier ( 2000 ). Laboratory identification of additional algae samples was not completed due to the enorm ous expense associated with enumeration and identification of each sample. While vegetation zone descriptions are pr ovided for soil sampling, these procedures were initially created for use in freshwater isolated marshes ( Lane 2003 ) . Zonation for soil samples was only employed at 3 of 118 sa mple wetlands that were characterized by open centers (where no canopy trees occurred in the deep pooled center area of these 3 wetlands). As such, soil samples were taken in both the outer forest ed zone and the inner marsh zone and analyzed separately. As sugge sted in the soil SOPs , the soil data values were weighted based on the area occupied by each vegetation zone.

PAGE 179

164 SOPs for Isolated Forested Wetlands Water Quality Site/Habitat Characterization Vegetation o Herbaceous – 1 x 5 m qu adrats along transect o Trees – variable area plots ev ery 10 meters along transect Algae o Wet Sites Only o Dry Sites – benthic algae only Macroinvertebrates – wet sites only Soils ORDER OF FIELD EVENTS FOR ISOLATED CYPRESS DOMES: 1. Water quality is ALWAYS taken first. Two field crew members take the water samples; one records the data while the other takes the sample. The data are recorded on BOTH the FDEP lab submittal form and on the wetland characterization form. 2. While two crewmembers are collecting water, the other(s) unlo ads the vehicle and prepares the field equipment. 3. After the water samples are obtained (follow SOP for water quality), complete the Site/Habitat Characterization Data Sheet & WRAP assessment. 4. When completed, start the vegetation transects. This includes delineating the wetland and running all four transe cts (follow SOP for vegetation). 5. The remaining field crew should: 6. Collect algae samples (f ollow SOP for algae) 7. Collect macroinvertebrates (follo w SOP for macroinvertebrates) 8. Collect soil samples (follow SOP for soil) 9. Take site photographs 10. Establish stakes for canopy photos CHECKLIST OF MATERIALS /FIELD EQUIPMENT: Miscellaneous o SOPs o Large cooler with frozen ice bottles for soils and vegetation o Camera o 3.5” floppy disks o Waders o Garmin III GPS unit o Florida Gazetteer o Machete o Aerial photo & FLUCCS codes of site Water Quality o Small cooler with ice o YSI meter (DO/temp) o 2 500-mL bottles – turbidity /color/conductivity/pH & NH3/NOX/TKN/TP o Pipette for H2SO4

PAGE 180

165 o Bottle of 1:1 H2SO4 o Clear tape o FedEx air bills o Zip-loc bags o FDEP Central Lab submittal form Vegetation Transects o 2 100m transect tapes o 1 m PVC o 2-3 compasses o 2 pieces of 1.5 m rebar o Clipboards o Field data sheets – a minimum of 8 per site o Site Characterization & WRAP sheets – 1 per person per site o Pencils & sharpie o Bag for unknown plants o Plant press, newspaper, and cardboard o Masking tape o Field ID manuals o Aerial photos o Prism for basal area o Hand lens o Index cards Macroinvertebrates o US Std 30 mesh sweep net o Large 1 gallon jar for sample o Bottle of formalin for preserving sample Algae o Collecting jars – 3 100-mL pea cups & 1 1-L sample bottle o Collection jar with bottom missing for benthic algae o Large pipette – aka turkey baste o Knife o Zip-lock freezer bags o Masking tape o Sharpie – black permanent maker o Falcon phytoplankton sampler – aka 50-mL centrifuge tube o Bottle brush & scraper o M3 preservative o Pipette for M3 preservative o 1-L deionized water for dry sites Canopy Photos o Digital camera o Spare batteries o Film disks o Tripod o Compass o Height pole

PAGE 181

166 Soils o 3-inch diameter PVC pipe o Knife o Piece of wood o Dampened hammer o Buckets o Small & medium sized freezer Zip-loc bags o Stainless steel spoon o Permanent marker o De-ionized water SOPs for Forested Wetlands: WATER QUALITY 1. The YSI meter must be on and calibrated for 15 minutes before using. 2. Always take this collection first !! 3. Water samples can ONLY be collected Monday, Tuesday, Wednesday, or Thursday. Samples are sent to FDEP Central Lab overnight. 4. One 500 mL bottle for turbidity/color/c onductivity/pH and one 500 mL bottle for NH3/NOx/TKN/TP are collected per site. Th ese have labels provided by FDEP. 5. Only take water samples if the water depth is greater than 10 cm. 6. Carefully enter the water without s tirring up organic material and silt. 7. Remove cap from each 500 mL bottle w ithout touching the lip or interior surfaces. 8. Rinse the bottle three times in wetland water, dumping the water away from collection area. 9. Place the sample bottle upside down in the standing wetland water. 10. Carefully tilt back end into the water a nd press on bottom of bottle to allow water to slowly flow inside. 11. Be deliberate, making sure that no susp ended organic matter enters the sample bottle. If organic materials do enter the sample, dump the sample and begin again. 12. When all exiting air bubbles have stoppe d, carefully lift bottle out of water. 13. Repeat, so both 500 mL bottles are full. 14. After the water is collected, take dissol ved oxygen and temperature readings using the YSI meter. Take measurements with in the top 10 cm of the water column. Apply constant, gentle motion to the di ssolved oxygen probe, as the meter is consuming oxygen during measurem ent. Measure water depth. 15. Preserve only the bottle for NH3/NOx/TKN/TP, using 2 mL of 1:1 H2SO4 per 500 mL sample. 16. Place both sample bottles on ice in a six-pack cooler. For transport reasons, the ice should be in a bag atop the samples. 17. Fill out the FDEP Central Lab Sample Subm ittal Form. Place in a zip-lock bag in cooler. 18. If 2 sites are sampled in 1 day, place all 4 sample bottles in the cooler along with the forms (one form is sufficient if properly filled out). 19. Tape the cooler shut and make sure the air bill is filled out properly. Call 1-800GO-FEDEX to find a nearby office.

PAGE 182

167 20. If you cannot get to FedEx in time, dump the samples taken. Repeat procedure another day. SOPs for Forested Wetlands: VEGETATION 1. Using a compass, locate the 4 cardinal point directions (nor th, south, east, and west). The 4 transects will begin at each cardinal point running from the edge of the wetland into the interior/middle of the wetland. These 4 transects will intersect in the middle and divide the we tland into 4 approximately equal sections. 2. At the beginning of each transect, delineate the edge of the wetland using a combination of wetland plants and hydric soils. Be conservative on the side of the wetland. 3. Establish the transect using the meter tapes. Start with 0 meters at the wetland edge, and increase distance to wards the wetland interior. 4. Use a separate field data sheet for each cardinal direction. If the number of species located on a transect exceeds the number of columns on the data sheet, start a new data sheet. 5. Creating quadrats that are 0.5 m on either si de of the transect (1 m wide) and 5 m long, record all species present wi thin these elongated quadrats. 6. Plant species names are recorded on the data sheets using the full genus and species names. Each unknown species is given a unique ID code using the transect location (ex. N-1). 7. Voucher specimens for all unknown species ar e collected, being sure to get plant inflorescence and roots, tagged with prope rly labeled masking tape, and put into a labeled collection bag. Note the color of the inflorescence on the label, as the flowers often do not preserve well. Index cards can be us ed to protect especially sensitive parts. 8. All collected plants are identified in the field on the day of sa mpling and placed in a plant press for further clarification and identification. Plant nomenclature follows FDEP’s Florida Wetland Plant Identification Manual (Tobe et al. 1998). 9. At each 10 m along each transect, starting at 10 m, 20 m, 30 m, etc., tree basal area will be recorded. Use the data sheet for basal area, and record basal area per species using variable area plots and a 10 factor prism. Hold the prism at eye level, with a bent elbow and count the num ber of trees per species that fall within the variable area plot. The prism shall be centered over the sampling point at all times, with the field person rotating around the prism so that the entire circular area (360o) around the point of sampling is included. 10. As the sun lowers on the horizon, take canopy photos at 1 point along each transect. Placement of tripod will be 10 paces out from the center of the wetland along each transect. In those instances wh en the cypress dome is a “hole in the doughnut” and there are no cypress trees in the center of the dome, tripod placement will be 10 paces along the transect out from the ring of trees. Follow directions according to A Manual for the Analysis of Hemispherical Photography ( Rich 1989 ). 11. At each photograph spot, insert a wooden stak e so that photo sites can be revisited in the future.

PAGE 183

168 12. Center the tripod over the stak e with the top of the lens cap at the height of the provided height pole [which is at breast height 1.3 m] . The top of the camera should face south, so that the photographer’s back is to the north. 13. Level the tripod, so that the bubble on top of the lens cap is centered within the circle. 14. Turn the camera on to automatic. Set the camera to XGA Fine, using the dial at the front right and the button on the back of the camera. 15. Zoom out the camera all of the way so th at the back display shows the canopy as a circle, surrounded by a dark/black border. 16. Record the photo number for the positi on (ex. north, south, east, or west). 17. Complete the canopy photo data sheet, noting time of day, cloud condition, surrounding vegetation, etc. SOPs for Forested Wetlands: ALGAE AT WET SITES: Separate samples by substrates of the site you are working on (i.e. epiphyton, benthic algae, metaphyton, and phytoplankton). For each substrate, collect 10 aliquots, and k eep each substrate type separate in their own collection jars. At the end of the co llection there should be between 100-120 mL of wetland algae-water mix in the cups, except for phytoplankton which should have approximately 1000 mL. Rinse all sample equipment in wetland water prior to sampling. EPIPHYTON – divide appropriately among he rbaceous and woody debris based of the proportion of the area of wetland of each: 1. For herbaceous vegetation: Cut plant stems under water and place in zip-lock bag with wetland water; shake and knead vigorously in zip-loc bag; use turkey baste to extract 10 mL of algal suspension and place in labeled pea cup; distribute the aliquots appropriately throughout the different ve getation/habitat zones. 2. For woody debris (roots, snags): Using a brush, brush the wood for alg ae. Place brush in bag with water and shake algae off of the brush. Pipe tte the algae into the collection jar; -orUse bottomless pea cup to isolate a spot on the debris. Use turkey baste to stir algae from surface of debris; ex tract 10 mL of algal suspension and place in pea cup; since woody and he rbaceous are within the same 10 aliquots, they must be divide d appropriately between the two. BENTHIC ALGAE: 1. Use bottomless pea cup to isolate a spot on the sediment; 2. Use turkey baste to gently stir al gae from the surface of the sediment; 3. Extract 10 mL of algal susp ension and place in pea cup. METAPHYTON: 1. Collect approximately 100 mL of wetland water in a pea cup;

PAGE 184

169 2. Using your fingers, collect a thumbna il size portion of the algal mat; 3. Obtain aliquots from 10 diffe rent areas of the wetland. PHYTOPLANKTON (1 L is collected): 1. In total use the 50 mL centrifuge tube (x2) to collect ten 100 mL aliquots; 2. Divide aliquots proportionately between the major vegetation zones; 3. Rinse tube with wetland water. With th e cap on the tube, lower into the water column and then remove the cap to allow the tube to fill with water; 4. Cap the tube under the water and th en bring tube out of water; 5. Carefully pour contents into th e dark algae collection bottle; 6. Since the tube is only 50 mL, you will need to do this twice in each of the 10 aliquots. Preserve the samples. o Add 5 mL of M3 per 100 mL of algal suspension in pea cups. o Add 20 mL of M3 per 1 L of algal suspension in column algae bottle. Properly label collection jars, identifying site, date, collector, and sample type. Carefully clean equipment with deionized wate r to avoid cross-contamination at future sites. When at the Center for Wetlands, clean al l equipment with Clorox/water solution. Return full collection jars to room 8 at the Center for Wetlands to await laboratory analysis. AT DRY SITES ONLY BENTHIC ALGAE IS TO BE COLLECTED: Use bottomless pea cup to isolate a spot on the sediment. Extract the upper 0.5 cm of so il into an additional pea cup, this depth is marked on the collection pea cup. Add 100 mL deionized water and st ir well with turkey baste. Extract 10 mL of algal suspensi on and place in sample pea cup. Repeat, so that you have collected 10 aliquo ts representative of the vegetation/habitat zones in the wetland. Preserve the sample with 5 mL of M3 per 100 mL of algal suspension. Properly label collection jar, identifying site, date, collector, and sample type. Carefully clean equipment with DI water to a void cross-contamination at future sites. When at the Center for Wetlands, clean al l equipment with Clorox/water solution. Return full collection jars to room 8 at the Center for Wetlands. DIRECTIONS FOR M3 FIXATIVE PREPARATION: Materials 10 g Iodine 5 g Potassium iodide (KI) 50 mL Glacial acetic acid 250 mL Formalin (37% W/W formaldehyde) 1 L Deionized water

PAGE 185

170 Methods 1. Dissolve 10 g iodine in a sm all quantity of deionized water to aid in solution of iodine. 2. Dissolve 5 g potassium iodide, 10 g disso lved iodine (from st ep 1), 50 mL glacial acetic acid and 250 mL formalin (37% W/ W formaldehyde) in 1 L deionized water. 3. Store in the dark. (p.10-8, Standard Methods for the Exam ination of Water and Wastewater, 17th edition) SOPs for Forested Wetlands: MACROINVERTEBRATES There are to be 20 sweeps (evenly divided into the vegetation/habitat zones) to send to FDEP unpicked for identification. Always do your sweeps in undisturbed areas where you have not walked through yet. A single sweep is one net width and two net lengths to equal 0.5 m2. Using a U.S. Standard 30 mesh net, sweep from the bottom of the substrate up the plant stalks. Use your hands to strip the plant of a ll material into the ne t. If you are in a forested site, use a brush to clea n any snags and roots of material. Vigorously sample the area repeatedly (3 times) to ensure good coverage. Dip net into water repeatedly, without letting the sample out, to try and sift the muck and silt through the net. Do not sample in the muck! Place the contents of each sweep into the 3.8 L jar. When all 20 samples are complete, preserve the sample by adding Formalin at a rate of 10% of the sample volume. Seal the jar. Shake to ensure thorough mixing. Place masking tape over the lid to prevent l eakage during travel & shipment. Properly label the jar with the site name, date, and collector. Thoroughly clean all equipment off with water. Return samples to room 8 at the Center fo r Wetlands for later shipment to the FDEP. DIRECTIONS FOR BUFFERED FORMALIN PREPARATION: Materials Sodium bicarbonate (sodium borate may also be used) Formalin (37% W/W formaldehyde) pH meter Methods Note that formalin in the common name for 37% W/W formaldehyde. “Formalin” and “Buffered Formalin” are 2 separate things in this recipe. 1. Calibrate the pH meter (directions follow) . The pH electrode and temperature probe should rest in a beaker of deionized water between measurements. Rinse the electrode and probe off with a spray bottle of deionized water before submerging in other solutions. 2. Select a container for prep aring the buffered formalin. Usually this is simply the plastic container that the formalin was shipped in. 3. Fill the container with formalin to just below (1-2 cm) where the top of the desired buffered formalin solution level. 4. Scoop a small amount of sodium bicarbonat e into the container, close, and shake vigorously (at least 1 minute to ensure prope r mixing). All of the sodium bicarbonate may not dissolve into the formalin, as this is a supersaturated solution.

PAGE 186

171 5. Measure the pH of the resulting solution. 6. Repeat steps 4 and 5 until the pH is at least 7.5 but not higher than 8.0. 7. If desired, transfer the buffered forma lin to smaller containers for field use. 8. Make sure that all containers are cl early labeled “Buffered Formalin pH 7.5-8.0.” 9. When disposing of used buffered formalin, deposit it in an appropriately labeled waste container. The waste container should have a yellow Hazardous Materials sticker, call the Environmental Health and Safety department at 392-1591. Original recipe from the Florida Depart ment of Environmen tal Protection Biology Section, Standard Operating Procedure (SOP) #IZ-10, “Preparation of Buffered Formalin.” Expanded upon by Melissa Yonteck on May 2, 2001. CALIBRATION OF pH METER (H anna Instruments HI 9025C) Materials pH meter pH 4.00 buffer solution pH 7.00 buffer solution 3 beakers deionized water bottle Methods *When not being used, rest the electrode and pr obe in a beaker of deionized water. Rinse the electrode and probe off w ith a spray bottle of deionized water before submerging in other solutions. 1. Pour small quantities of the pH 4.00 and pH 7.00 buffer solutions into each of 2 clean beakers. 2. Immerse the pH electrode and temperatur e probe into the pH 4.0 buffer solution, stir briefly. The electrode and probe should be close together, and they should be submerged approximately 4 cm (1½ inch) into the solution. 3. Press the CAL key. The “CAL” and bu ffer indicators will be displayed. The secondary LCD display should read “4.01.” If not, adjust it using the “ c” key. 4. After the pH reading becomes stable, th e “READY” and “CON” indicators will blink. Once this happens, press the CON key to confirm the calibration. 5. After rinsing with deioni zed water, immerse the pH electrode and temperature probe into the pH 7.00 solution, and stir briefly. 6. Select the second buffer value (“7.01” ) on the secondary display using the “ c” key. 7. After the pH reading becomes stable, th e “READY” and “CON” indicators will blink. Once this happens, press the CON key to confirm the calibration. 8. Press “CAL” key to end calibration process and begin measuring. 9. When finished using the pH meter, pat th e electrode and probe dr y with a KimWipe. Place the pH electrode in the yellow/orange cap with a small amount of deionized water. The pH meter should be recalibrates: whenever the pH electrode or temperature probe is replaced at least once a month after testing aggressive chemicals

PAGE 187

172 if greatest accuracy is required whenever the batteries have been replaced. SOPs for Forested Wetlands: SOIL 1. The wetland will be visually divided into vegetation zones. Cores will be taken within each zone and combined, so that each vegetation zone has one cumulative soil sample. The number of co res taken per zone is generally 4, 1 along each transect. In some cases th ere will be fewer than 4 cores per vegetation zone as the zones may not al l fall along the established transects. 2. To sample soils: a. Clean off the detritus from the site that will be sampled. This means removal of plant material that app ears less than 6 months old, or the recognizable fallen plant material. b. Place the 7.9 cm diameter PVC pipe on the soil surface at the sample location. c. Using the knife carefully cut a circul ar shape around the sampling pipe, so that the pipe will easily slide through the soil and r oots. This reduces soil compaction. d. Using the dampened hammer, gently pound the sampling pipe into the soil. Hammer the core 10 cm into the soil. There is a black line indicating this depth on the soil core. e. With the core in place, dig down to the bottom of the core and extract the core into a bucket that has been ma rked with the name of the vegetation zone. 3. Repeat along each transect for each vege tation zone, making sure that the 10 cm soil sample is placed into the properly marked bucket (to assure vegetation zones are not mixed). 4. Thoroughly mix each bucket of soil with th e large stainless steel spoon. Clean and rinse the spoon with de-ion ized water between buckets. 5. Gather several quart-sized freezer zip-loc bags and a permanent marker. Label each small zip-loc bag with the site name, vegetation zone, number of cores, the bag number (i.e. 4 of 7), date, and name of collector. 6. Using a clean stainless steel spoon, ta ke enough randomly selected spoonfuls of soil to fill the labeled quart size Zip-loc bag. 7. Seal the Zip-loc bag and place in a larg er zip-loc freezer bag labeled with the site name, date, and number of smaller bags contained. Place in cooler and ice down. 8. Since the resultant nutrient and % organic matter will be weighted based on the % area of each vegetation zone, it is imperative that the vegetation zones marked on the soil bags are also marked on the vegetation zone map that is part of the wetland characterization sh eet. Do not forget to include the approximate % of each area in the wetland. 9. Rinse field equipment with deionized water. 10. Return the samples to the Center for We tlands, and store in the refrigerator in the back lab pending laboratory analysis.

PAGE 188

173 APPENDIX D COEFFICIENT OF CONSERVATISM SCORES Table D-1. Coefficient of Conservatism ( CC) scores for 561 macrophytes identified in isolated depressional freshwater forested wetlands in Florida. Species CC Species CC Acalypha gracilens 3.3 Begonia cucullata 1.5 Acer rubrum 5.2 Berchemia scandens 5.1 Acrostichum danaeifolium 6.2 Betula nigra 4.8 Agalinis filifolia 6.7 Bidens alba 1.0 Agrostis hyemalis 5.4 Bidens discoidea 4.8 Albizia julibrissin 0.0 Bidens mitis 3.8 Aloe vera 0.0 Bignonia capreolata 4.8 Alternanthera philoxeroides 0.0 Bischofia javanica 0.0 Alternanthera sessilis 0.7 Blechnum serrulatum 5.5 Amaranthus australis 2.6 Blechum pyramidatum 0.0 Amaranthus blitum 0.0 Boehmeria cylindrica 4.5 Amaranthus spinosus 0.0 Boltonia diffusa 3.8 Ambrosia artemisiifolia 0.7 Bromus catharticus 0.0 Ampelopsis arborea 3.3 Bulbostylis stenophylla 4.4 Amphicarpum muhlenbergianum 5.0 Callicarpa americana 2.4 Andropogon glomeratus 3.1 Callisia repens 0.0 Andropogon virginicus 2.6 Campsis radicans 3.3 Annona glabra 6.8 Canna flaccida 5.7 Anthaenantia villosa 7.1 Caperonia castaneifolia 2.4 Apios americana 3.1 Carex debilis 6.5 Ardisia crenata 1.0 Carex frankii 6.0 Ardisia escallonioides 0.0 Carex gigantea 6.4 Aristida beyrichiana 9.8 Carex glaucescens 7.1 Aristida patula 6.3 Carex longii 3.6 Aristida purpurascens 6.0 Carex striata 5.7 Aristida spiciformis 6.4 Carex verrucosa 7.1 Asplenium platyneuron 4.8 Carphephorus odoratissimus 7.6 Aster carolinianus 6.9 Carphephorus paniculatus 6.0 Aster dumosus 3.6 Celtis laevigata 5.0 Aster elliottii 4.2 Centella asiatica 1.9 Aster pilosus 5.4 Cephalanthus occidentalis 6.0 Aster subulatus 4.5 Cercis canadensis 4.0 Aster tenuifolius 7.1 Chamaecrista fasciculata 0.0 Axonopus fissifolius 2.8 Chamaecrista nictitans 2.9 Axonopus furcatus 2.4 Chamaesyce hypericifolia 0.0 Azolla caroliniana 2.6 Chaptalia tomentosa 7.9 Baccharis halimifolia 2.1 Chenopodium album 0.0 Bacopa caroliniana 6.0 Chiococca alba 5.6 Bacopa monnieri 4.3 Chrysobalanus icaco 6.3

PAGE 189

174 Table D-1. Continued. Species CC Species CC Cicuta maculata 5.0 Diodia teres 1.9 Cinnamomum camphora 0.2 Diodia virginiana 2.4 Cirsium nuttallii 4.8 Dioscorea bulbifera 0.0 Cissus trifoliata 4.2 Diospyros virginiana 4.0 Citrus Xaurantium 0.0 Drosera brevifolia 6.7 Cladium jamaicense 5.5 Drosera capillaris 6.7 Cleistes bifaria 7.1 Drymaria cordata 1.2 Clethra alnifolia 5.2 Duchesnea indica 3.6 Cliftonia monophylla 5.0 Dulichium arundinaceum 6.8 Coelorachis cylindrica 5.6 Echinochloa colona 0.7 Coelorachis rugosa 6.3 Echinochloa crusgalli 0.0 Coelorachis tuberculosa 6.5 Echinochloa walteri 3.1 Colocasia esculenta 0.0 Eclipta prostrata 1.7 Commelina diffusa 1.7 Eleocharis baldwinii 2.1 Commelina erecta 4.8 Eleocharis flavescens 3.6 Commelina virginica 4.8 Eleocharis interstincta 5.5 Conoclinium coelestinum 4.3 Eleocharis microcarpa 3.0 Conyza canadensis 0.3 Eleocharis vivipara 2.4 Cornus foemina 4.8 Elephantopus nudatus 4.0 Crataegus viridis 8.6 Eleusine indica 0.0 Crinum americanum 7.6 Elymus virginicus 4.0 Ctenium aromaticum 10.0 Eragrostis atrovirens 1.8 Cuphea carthagenensis 1.4 Erechtites hieracifolia 2.1 Cyclospermum leptophyllum 1.2 Erianthus giganteus 6.0 Cynanchum scoparium 4.8 Erigeron quercifolius 2.9 Cynodon dactylon 0.0 Erigeron strigosus 2.4 Cyperus croceus 1.8 Erigeron vernus 4.3 Cyperus distinctus 3.8 Eriocaulon compressum 6.7 Cyperus erythrorhizos 4.2 Eriocaulon decangulare 6.7 Cyperus haspan 2.6 Eriocaulon ravenelii 4.8 Cyperus iria 1.2 Eryngium prostratum 4.0 Cyperus lanceolatus 2.4 Eugenia uniflora 0.0 Cyperus odoratus 3.6 Eupatorium capillifolium 0.5 Cyperus polystachyos 2.4 Eupatorium leptophyllum 3.6 Cyperus retrorsus 1.7 Eupatorium mohrii 5.5 Cyperus surinamensis 1.9 Eupatorium rotundifolium 6.2 Cyperus virens 3.9 Eupatorium serotinum 4.8 Cyrilla racemiflora 4.5 Eustachys glauca 2.4 Desmodium incanum 0.0 Eustachys petraea 0.0 Desmodium lineatum 6.0 Euthamia caroliniana 2.6 Desmodium paniculatum 3.6 Euthamia minor 3.6 Dichondra caroliniensis 1.9 Ficus aurea 5.7 Digitaria bicornis 0.0 Fimbristylis dichotoma 4.0 Digitaria ciliaris 0.3 Fraxinus caroliniana 7.1 Digitaria serotina 1.8 Fuirena scirpoidea 3.8

PAGE 190

175 Table D-1. Continued. Species CC Species CC Galactia elliottii 3.8 Ixora chinensis 0.0 Galactia volubilis 3.6 Jacquemontia tamnifolia 0.0 Galium hispidulum 3.3 Juncus coriaceus 5.1 Galium tinctorium 3.1 Juncus dichotomus 2.9 Galium uniflorum 5.1 Juncus effusus 1.9 Gaylussacia frondosa 6.7 Juncus marginatus 2.4 Gaylussacia mosieri 7.4 Juncus megacephalus 3.3 Gelsemium sempervirens 4.0 Juncus polycephalus 3.3 Gnaphalium falcatum 1.9 Juncus repens 5.2 Gnaphalium obtusifolium 2.4 Juncus tenuis 2.4 Gordonia lasianthus 6.7 Juniperus virginiana 5.2 Gratiola ramosa 5.0 Justicia angusta 6.0 Gratiola virginiana 7.1 Justicia ovata 5.5 Habenaria repens 4.8 Kummerowia striata 0.0 Hedychium coronarium 0.0 Kyllinga brevifolia 0.3 Hedyotis corymbosa 2.0 Kyllinga pumila 3.3 Hedyotis uniflora 3.6 Lachnanthes caroliniana 3.1 Hemarthria altissima 0.0 Lachnocaulon anceps 5.5 Hydrocotyle bonariensis 3.3 Lachnocaulon engleri 4.8 Hydrocotyle ranunculoides 3.1 Lachnocaulon minus 6.0 Hydrocotyle umbellata 2.9 Lactuca graminifolia 2.7 Hydrocotyle verticillata 3.1 Lantana camara 0.0 Hymenachne amplexicaulis 0.0 Leersia hexandra 4.8 Hypericum brachyphylum 6.8 Lemna minor 1.0 Hypericum chapmanii 7.1 Lepidium virginicum 0.2 Hypericum cistifolium 5.0 Leptochloa uninervia 3.0 Hypericum fasciculatum 5.7 Leucothoe axillaris 6.0 Hypericum gallioides 6.0 Leucothoe racemosa 6.2 Hypericum hypericoides 4.0 Ligustrum japonicum 0.0 Hypericum mutilum 3.6 Ligustrum lucidum 0.0 Hypericum myrtifolium 5.5 Ligustrum sinense 0.0 Hypoxis curtissii 6.0 Limnobium spongia 4.8 Hyptis alata 4.3 Linaria canadensis 0.3 Hyptis mutabilis 0.0 Lindernia crustacea 0.6 Ilex cassine 8.1 Lindernia grandiflora 3.6 Ilex coriacea 6.0 Liquidambar styraciflua 3.3 Ilex glabra 4.3 Litsea aestivalis 9.8 Ilex myrtifolia 8.3 Lobelia floridana 6.5 Ilex opaca 6.0 Lolium perenne 0.0 Ilex vomitoria 4.8 Lonicera japonica 0.0 Ilex x attenuata 7.1 Lophiola aurea 6.5 Ipomoea indica 0.6 Ludwigia alata 4.5 Ipomoea sagittata 5.4 Ludwigia curtissii 4.4 Iris hexagona 7.1 Ludwigia hirtella 6.0 Itea virginica 7.9 Ludwigia linifolia 4.5

PAGE 191

176 Table D-1. Continued. Species CC Species CC Ludwigia maritima 3.3 Nymphaea odorata 5.5 Ludwigia microcarpa 3.1 Nymphoides aquatica 5.7 Ludwigia octovalvis 2.4 Nyssa aquatica 3.6 Ludwigia palustris 4.0 Nyssa biflora 7.4 Ludwigia peruviana 1.2 Oeceoclades maculata 0.4 Ludwigia repens 2.9 Oplismenus hirtellus 3.3 Ludwigia virgata 3.9 Osmunda cinnamomea 5.5 Luziola fluitans 4.8 Osmunda regalis 6.9 Lycopodiella alopecuroides 6.7 Oxalis corniculata 1.2 Lycopodiella prostrata 7.1 Oxalis debilis 0.0 Lycopus rubellus 5.2 Oxypolis filiformis 6.7 Lycopus virginicus 5.2 Paederia foetida 0.0 Lygodium japonicum 0.0 Panicum aciculare 4.8 Lygodium microphyllum 0.0 Panicum chamaelonche 4.8 Lyonia ligustrina 6.9 Panicum ciliatum 4.5 Lyonia lucida 6.0 Panicum commutatum 4.5 Lythrum alatum 3.0 Panicum dichotomum 4.0 Magnolia grandiflora 3.6 Panicum ensifolium 5.0 Magnolia virginiana 8.1 Panicum erectifolium 5.7 Malus angustifolia 6.0 Panicum hemitomon 5.0 Matelea floridana 6.7 Panicum repens 0.0 Mecardonia acuminata 3.9 Panicum rigidulum 4.5 Melaleuca quinquenervia 0.0 Panicum scabriusculum 5.0 Melia azedarach 0.0 Panicum sphaerocarpon 5.1 Melochia corchorifolia 1.5 Panicum spretum 5.4 Melothria pendula 1.8 Panicum tenerum 5.0 Micranthemum glomeratum 3.6 Panicum tenue 4.2 Micranthemum umbrosum 4.3 Panicum verrucosum 4.3 Micromeria brownei 4.8 Parietaria floridana 1.8 Mikania scandens 2.4 Parthenocissus quinquefolia 3.0 Mitchella repens 6.7 Paspalidium geminatum 3.6 Mitreola petiolata 5.4 Paspalum acuminatum 2.0 Mitreola sessilifolia 5.4 Paspalum conjugatum 3.1 Modiola caroliniana 3.2 Paspalum laeve 3.8 Momordica charantia 0.0 Paspalum monostachyum 9.1 Morrenia odorata 0.0 Paspalum notatum 0.0 Morus alba 1.2 Paspalum plicatulum 2.4 Morus rubra 3.6 Paspalum repens 4.0 Myrica cerifera 3.1 Paspalum setaceum 2.1 Myrica heterophyla 7.9 Paspalum urvillei 1.2 Myrica inodora 9.0 Passiflora incarnata 3.0 Nandina domestica 0.0 Passiflora suberosa 3.0 Nephrolepis biserrata 5.2 Peltandra virginica 3.6 Nephrolepis exaltata 4.8 Pentodon pentandrus 6.0 Nuphar luteum 5.2 Persea borbonia 6.3

PAGE 192

177 Table D-1. Continued. Species CC Species CC Persea palustris 7.4 Quercus phellos 7.4 Phalaris angusta 0.0 Quercus virginiana 4.2 Phanopyrum gymnocarpon 6.0 Rapanea punctata 5.2 Phlebodium aureum 6.8 Rhexia alifanus 6.9 Photinia pyrifolia 5.7 Rhexia lutea 6.5 Phyla nodiflora 1.4 Rhexia mariana 3.8 Phyllanthus tenellus 0.0 Rhexia nashii 6.2 Phyllanthus urinaria 0.0 Rhexia petiolata 6.2 Physalis angulata 1.2 Rhexia virginica 5.0 Phytolacca americana 1.2 Rhododendron viscosum 7.6 Pieris phillyreifolia 9.5 Rhodomyrtus tomentosa 0.0 Pinus clausa 5.6 Rhoeo discolor 0.0 Pinus elliottii 4.0 Rhus copallinum 2.4 Pinus palustris 7.1 Rhynchospora capitelatta 6.0 Pinus serotina 7.1 Rhynchospora cephalantha 4.3 Pinus taeda 3.3 Rhynchospora chalarocephala 4.8 Plantago lanceolata 1.2 Rhynchospora chapmanii 6.0 Pluchea camphorata 4.3 Rhynchospora colorata 5.5 Pluchea carolinensis 3.6 Rhynchospora corniculata 6.0 Pluchea foetida 3.8 Rhynchospora decurrens 6.3 Pluchea longifolia 2.8 Rhynchospora fascicularis 4.5 Pluchea odorata 3.8 Rhynchospora filifolia 6.0 Pluchea rosea 3.6 Rhynchospora gracilenta 6.0 Polygala cymosa 9.0 Rhynchospora inundata 6.0 Polygala lutea 3.6 Rhynchospora latifolia 6.9 Polygonum hydropiperoides 2.6 Rhynchospora microcarpa 4.5 Polygonum punctatum 2.6 Rhynchospora microcephala 4.8 Polygonum sagittatum 4.8 Rhynchospora miliacea 7.1 Polypremum procumbens 1.2 Rhynchospora odorata 6.7 Pontederia cordata 5.0 Rhynchospora plumosa 6.4 Populus deltoides 1.2 Rhynchospora pusilla 6.7 Pouzolzia zeylanica 0.4 Rhynchospora tracyi 8.3 Proserpinaca palustris 3.8 Rhynchospora wrightiana 7.1 Proserpinaca pectinata 3.8 Richardia brasiliensis 0.0 Prunus caroliniana 3.0 Rivina humilis 1.2 Prunus serotina 3.6 Rosa carolina 7.1 Psilotum nudum 3.6 Rosa palustris 6.9 Psychotria nervosa 3.6 Rubus argutus 2.1 Psychotria sulzneri 3.6 Rubus cuneifolius 1.9 Pteridium aquilinum 3.6 Rubus trivialis 1.9 Ptilimnium capillaceum 3.1 Ruellia caroliniensis 4.3 Pueraria montana 0.0 Rumex crispus 0.2 Quercus geminata 5.2 Rumex obtusifolius 0.7 Quercus laurifolia 3.6 Rumex pulcher 0.6 Quercus nigra 2.1 Sabal palmetto 4.5

PAGE 193

178 Table D-1. Continued. Species CC Species CC Sabatia bartramii 6.8 Solanum americanum 1.4 Sacciolepis indica 1.9 Solanum capsicoides 1.4 Sacciolepis striata 3.6 Solanum carolinense 1.2 Sageretia minutiflora 7.9 Solanum nigrum 0.0 Sagittaria graminea 5.5 Solanum tampicense 0.7 Sagittaria lancifolia 4.5 Solanum viarum 0.0 Sagittaria latifolia 5.0 Solidago canadensis 3.0 Salix caroliniana 2.1 Solidago fistulosa 3.6 Salix nigra 3.3 Solidago gigantea 3.2 Salvia lyrata 0.0 Solidago latissimifolia 1.8 Sambucus canadensis 1.7 Solidago sempervirens 5.0 Samolus ebracteatus 5.7 Sonchus asper 1.5 Sapium sebiferum 0.0 Sorghum bicolor 0.0 Sarcostemma clausum 2.4 Spartina bakeri 5.5 Sarracenia flava 9.3 Spermacoce assurgens 1.2 Sarracenia minor 4.8 Spermacoce verticillata 0.0 Saururus cernuus 5.5 Sporobolus floridanus 7.1 Schinus terebinthifolius 0.0 Sporobolus indicus 0.2 Scirpus cyperinus 4.5 Stachys floridana 1.4 Scleria baldwinii 6.7 Stenotaphrum secundatum 0.8 Scleria georgiana 6.2 Stillingia aquatica 7.4 Scleria reticularis 5.1 Styrax americanus 6.9 Scleria triglomerata 4.8 Syngonanthus flavidulus 5.2 Scoparia dulcis 2.4 Taxodium ascendens 8.8 Scutellaria integrifolia 5.7 Thalia geniculata 6.2 Senecio glabellus 4.0 Thelypteris dentata 6.0 Senna obtusifolia 0.0 Thelypteris hispidula 4.5 Senna pendula 0.0 Thelypteris interrupta 5.2 Serenoa repens 4.5 Thelypteris kunthii 5.2 Sesbania herbacea 1.0 Thelypteris palustris 3.6 Sesbania vesicaria 0.5 Tilia americana 5.5 Setaria parviflora 3.1 Toxicodendron radicans 1.9 Seymeria cassioides 6.0 Tradescantia fluminensis 0.0 Sida acuta 1.0 Tradescantia ohiensis 0.9 Sida rhombifolia 1.0 Tradescantia zebrina 0.0 Sideroxylon celastrinum 6.0 Triadenum virginicum 5.0 Sideroxylon reclinatum 6.0 Trifolium repens 0.0 Smilax auriculata 3.8 Tripsacum dactyloides 4.0 Smilax bona-nox 2.6 Typha domingensis 1.2 Smilax glauca 3.3 Typha latifolia 1.2 Smilax laurifolia 5.2 Ulmus americana 7.4 Smilax rotundifolia 3.2 Urena lobata 0.0 Smilax smallii 4.5 Urochloa mutica 0.0 Smilax tamnoides 3.6 Utricularia gibba 3.6 Smilax walteri 6.0 Utricularia purpurea 6.7

PAGE 194

179 Table D-1. Continued. Species CC Species CC Vaccinium arboreum 6.4 Vitis shuttleworthii 1.2 Vaccinium corymbosum 5.7 Vittaria lineata 1.2 Vaccinium darrowii 6.2 Waltheria indica 2.4 Vaccinium elliottii 6.7 Wedelia trilobata 0.0 Vaccinium myrsinites 4.8 Woodwardia areolata 5.7 Valeriana scandens 7.1 Woodwardia virginica 4.8 Verbena bonariensis 0.0 Xanthosoma sagittifolium 0.0 Verbena brasiliensis 0.0 Xyris ambigua 5.7 Viburnum nudum 3.6 Xyris caroliniana 5.7 Viburnum obovatum 1.2 Xyris elliottii 5.7 Viburnum odoratissimum 0.0 Xyris fimbriata 5.7 Vicia sativa 0.4 Xyris jupicai 1.7 Vigna luteola 3.6 Xyris platylepis 3.6 Viola lanceolata 4.8 Youngia japonica 0.0 Vitis aestivalis 2.9 Yucca aloifolia 1.2 Vitis cinerea 2.0 Zea mays 0.0 Vitis rotundifolia 2.1

PAGE 195

180 APPENDIX E CANDIDATE METRICS Table E-1. Candidate metrics ba sed on the diatom assemblage. N P A Tolerance Metrics Indicator Species Sensitive Taxa Abundance 1 1 1 Indicator Species Sensitive Taxa Presence/Absence 1 1 1 Indicator Species Tole rant Taxa Abundance 1 1 1 Indicator Species Tolerant Taxa Presence/Absence 1 1 1 Community Composition Metrics Richness 1 Evenness 1 Shannon Diversity 1 Simpson's Index 1 Autecological Metrics Morphological Guild Erect 1 1 1 Morphological Guild Stalked 1 1 1 Morphological Guild Unattached 1 1 1 Morphological Guild Prostrate/Adnate 1 1 1 Morphological Guild Variable 1 1 1 Motility Highly Motile 1 1 1 Motility Moderately Motile 1 1 1 Motility Highly & Moderately Motile 1 1 1 Motility Not Motile 1 1 1 Motility Variable 1 1 1 Pollution Tolerance Very Tolerant 1 1 1 Pollution Tolerance Moderately Tolerant 1 1 1 Pollution Tolerance Very & Moderately Tolerant 1 1 1 Pollution Tolerance Sensitive / Intolerant 1 1 1 Dissolved Oxygen Class 1 (Bahls 1993) 1 1 1 Dissolved Oxygen Class 2 (Bahls 1993) 1 1 1 Dissolved Oxygen Class 3 (Bahls 1993) 1 1 1 Dissolved Oxygen Class 4 (Bahls 1993) 1 1 1 Dissolved Oxygen Class 5 (Bahls 1993) 1 1 1 Wet/Dry Preference Class 1 (van Dam et al. 1994) 1 1 1 Wet/Dry Preference Class 2 (van Dam et al. 1994) 1 1 1

PAGE 196

181 Table E-1. Continued. N P A Autecological Metrics continued Wet/Dry Preference Class 3 (van Dam et al. 1994) 1 1 1 Wet/Dry Preference Class 4 (van Dam et al. 1994) 1 1 1 Wet/Dry Preference Class 5 (van Dam et al. 1994) 1 1 1 Nitrogen Metabolism Class 1 (van Dam et al. 1994) 1 1 1 Nitrogen Metabolism Class 2 (van Dam et al. 1994) 1 1 1 Nitrogen Metabolism Class 3 (van Dam et al. 1994) 1 1 1 Nitrogen Metabolism Class 4 (van Dam et al. 1994) 1 1 1 pH Class 1 (van Dam et al. 1994) 1 1 1 pH Class 2 (van Dam et al. 1994) 1 1 1 pH Class 3 (van Dam et al. 1994) 1 1 1 pH Class 4 (van Dam et al. 1994) 1 1 1 pH Class 5 (van Dam et al. 1994) 1 1 1 pH Class 6 (van Dam et al. 1994) 1 1 1 Salinity Class 1 (van Dam et al. 1994) 1 1 1 Salinity Class 2 (van Dam et al. 1994) 1 1 1 Salinity Class 3 (van Dam et al. 1994) 1 1 1 Salinity Class 4 (van Dam et al. 1994) 1 1 1 Saprobity Class 1 (van Dam et al. 1994) 1 1 1 Saprobity Class 2 (van Dam et al. 1994) 1 1 1 Saprobity Class 3 (van Dam et al. 1994) 1 1 1 Saprobity Class 4 (van Dam et al. 1994) 1 1 1 Saprobity Class 5 (van Dam et al. 1994) 1 1 1 Trophic Class 1 (van Dam et al. 1994) 1 1 1 Trophic Class 2 (van Dam et al. 1994) 1 1 1 Trophic Class 1 & 2 (van Dam et al. 1994) 1 1 1 Trophic Class 3 (van Dam et al. 1994) 1 1 1 Trophic Class 4 (van Dam et al. 1994) 1 1 1 Trophic Class 5 (van Dam et al. 1994) 1 1 1 Trophic Class 6 (van Dam et al. 1994) 1 1 1 Trophic Class 7 (van Dam et al. 1994) 1 1 1

PAGE 197

182 Table E-2. Candidate metrics based on th e macrophyte assemblage. Metrics were calculated in multiple forms including N number, P percent, A abundance, F frequency of occurrence, and O other. N P A F O Wetland Plant Status Obligate Species 1 1 1 1 Facultative Wetland Species 1 1 1 1 Facultative Species 1 1 1 1 Facultative Upland Species 1 1 1 1 Upland Species 1 1 1 1 Obligate + Facultative Wetland Species 1 1 1 1 Obligate + Facultative Wetland + Facultative Species 1 1 1 1 Facultative Upland + Upland Species 1 1 1 1 Upland + Facultative Upland + Facultative Species 1 1 1 1 Plant Growth Form & Taxa Metrics Graminoid Species 1 1 1 1 Carex sp. 1 1 1 1 Herbaceous Species 1 1 1 1 Species in Asteraceae 1 1 1 1 Polygonum sp. 1 1 1 1 Graminoids to Herbaceous 1 1 1 Vine Species 1 1 1 1 Vines that are Woody 1 1 1 1 Shrub Species 1 1 1 1 Tree Species 1 1 1 1 Tree and Shrub Species 1 1 1 1 Salix sp. 1 1 1 1 Hardwoods 1 1 1 1 Trees as Hardwoods 1 1 1 1 Nyssa sp. 1 1 1 1 Trees as Nyssa sp. 1 1 1 1 Acer rubrum 1 1 1 1 Trees as Acer rubrum 1 1 1 1 Trees as Conifers 1 1 1 1 Trees as Taxodium sp. 1 1 1 1 Native Evergreen Shrubs 1 1 1 1 Native Ferns 1 1 1 1 Native Perennial Graminoids 1 1 1 1 Native Perennial Herbs 1 1 1 1

PAGE 198

183 Table E-2. Continued. N P A F O Indicator Species Sensitive (R:I Ratio, where R equals the number of Reference Sites and I equals number of agricultural and urban sites a species was found at) 1 1 1 1 Tolerants (R:I Ratio, where R equals the number of Reference Sites and I equals number of agricultural and urban sites a species was found at) 1 1 1 1 TWINSPAN Sensitive Species 1 1 1 1 TWINSPAN Sensitive Species by Presence Absence 1 1 1 1 TWINSPAN Tolerant Species 1 1 1 1 TWINSPAN Tolerant Species by Presence Absence 1 1 1 1 TWINSPAN Sensitive Species, Excluding Exotic Species 1 1 1 1 TWINSPAN Sensitive Species by Presence Absence, Excluding Exotic Species 1 1 1 1 TWINSPAN Tolerant Species, Excluding Exotic Species 1 1 1 1 TWINSPAN Tolerant Species by Presence Absence, Excluding Exotic Species 1 1 1 1 Indicator Species Sensitive Taxa – Occurrence 1 1 1 1 Indicator Species Sensitive Ta xa Presence/Absence 1 1 1 1 Indicator Species Tolerant Taxa – Occurrence 1 1 1 1 Indicator Species Tolerant Ta xa Presence/Absence 1 1 1 1 Indicator Species Sensitive Taxa, Excluding Exotic Species – Occurrence 1 1 1 1 Indicator Species Sensitive Taxa, Excluding Exotic Species Presence/Absence 1 1 1 1 Indicator Species Tolerant Taxa, Excluding Exotic Species – Occurrence 1 1 1 1 Indicator Species Tolerant Taxa, Excluding Exotic Species Presence/Absence 1 1 1 1 Modified FQI Score 1 1 Exotic Species Metric 1 1 1 1 Longevity Metrics Annuals 1 1 1 1 Native Annuals 1 1 1 1 Perennials 1 1 1 1 Native Perennials 1 1 1 1

PAGE 199

184 Table E-2. Continued. N P A F O Longevity Metrics continued Annual to Perennial Ratio 1 1 Native Annual to Native Perennial Ratio 1 1 Richness Metrics Species Richness by Site 1 Species Richness by Quadrat 1 Species Richness by Occurrence 1 Species Richness by Transect 1 Mean Site Evenness 1 Dominant Species 1 Log (Proportion of Dominant Species) 1 Vascular Genera 1 Nonvascular Genera 1

PAGE 200

185 Table E-3. Candidate metrics based on the m acroinvertebrate assemblage. Metrics were calculated in multiple forms including N – number, P – percent, A – abundance, and T – number of taxa. N P A T Tolerance Metrics Indictor Species Analysis Sensitive Taxa Abundance 1 1 1 Sensitive Taxa Presence/Absence 1 1 1 Tolerant Taxa Abundance 1 1 1 Tolerant Taxa Presence/Absence 1 1 1 Florida Index 1 Lake Condition Index 1 Community Structure & Balance Metrics Mixed Taxonomic Levels Crustacea + Mollusca 1 1 1 Dominant Taxa 1 1 1 Exotic Richness 1 Taxa Richness 1 1 Tubificida/Insecta 1 Phylum Phylum Richness 1 1 Annelida 1 1 1 Arthropoda 1 1 1 Mollusca 1 1 1 Platyhelminthes 1 1 1 Class Class Richness 1 1 Arachnida 1 1 1 Bivalva 1 1 1 Crustacea 1 1 1 Gastropoda 1 1 1 1 Insecta 1 1 1 Oligochaeta 1 1 1 1 Plecypoda 1 1 1 1 Turbellaria 1 1 1 1 Order Order Richness 1 1 Acariforrmes 1 1 1 Amphipoda 1 1 1 1 Anostraca 1 1 1 Basommatophora 1 1 1 Coleoptera 1 1 1 Collembola 1 1 1 Decapoda 1 1 1 1 Diptera 1 1 1 1

PAGE 201

186 Table E-3. Continued. N P A T Community Structure & Balance Metrics continued Order continued Diptera non-Chironomid 1 1 1 Ephemeroptera 1 1 1 1 Ephemeroptera + Plecoptera + Trichoptera 1 1 1 1 Ephemeroptera + Trichoptera + Odonata 1 1 1 1 Haplotaxida 1 1 1 Hemiptera 1 1 1 1 Heteroptera 1 1 1 Hoplonemertea 1 1 1 Isopoda 1 1 1 1 Lepidoptera 1 1 1 Lumbriculida 1 1 1 Megaloptera 1 1 1 Mesogastropoda 1 1 1 Odonata 1 1 1 1 Oribatei 1 1 1 Plecoptera 1 1 1 1 Trichoptera 1 1 1 1 Tricladida 1 1 1 Tromibidiformes 1 1 1 1 Tubificida 1 1 1 Veneroida 1 1 1 Zygoptera 1 1 1 1 Family Family Richness 1 1 Aeshnidae 1 1 1 Ancylidae 1 1 1 Arrenuridae 1 1 1 Asellidae 1 1 1 Baetidae 1 1 1 Belostomatidae 1 1 1 Cambaridae 1 1 1 Ceratopogonidae 1 1 1 Chaoboridae 1 1 1 Chironomidae 1 1 1 1 Coenagrionidae 1 1 1 Corixidae 1 1 1 Crangonyctidae 1 1 1

PAGE 202

187 Table E-3. Continued. N P A T Community Structure & Balance Metrics continued Family continued Culicidae 1 1 1 Curulionidae 1 1 1 Dryopidae 1 1 1 Dytiscidae 1 1 1 Enchytraeidae 1 1 1 Haliplidae 1 1 1 Helodidae 1 1 1 Hydrophilidae 1 1 1 Libellulidae 1 1 1 Lumbriculidae 1 1 1 Naididae 1 1 1 Noteridae 1 1 1 Notonectidae 1 1 1 Physidae 1 1 1 Planorbidae 1 1 1 Tabanidae 1 1 1 Tipulidae 1 1 1 Tubificidae 1 1 1 Sub-Families of Chironomidae Chironominae 1 1 1 Orthocladiinae 1 1 1 1 Tanypodinae 1 1 1 Ratio Tanypodinae/Orthocladiinae 1 Ratio Chironominae/Orthocladiinae 1 Ratio (Tanypodinae + Chironominae)/Orthocladiinae 1 Genus Genus Richness 1 1 Ablabesmyia 1 1 1 Anopheles 1 1 1 Arrenurus 1 1 1 Atrichopogon 1 1 1 Beardius 1 1 1 Belostoma 1 1 1 Berosus 1 1 1 Bratislavia 1 1 1 Buenoa 1 1 1 Caecidotea 1 1 1 Callibaetis 1 1 1 Chaoborus 1 1 1 Chironomus 1 1 1 Crangonyx 1 1 1

PAGE 203

188 Table E-3. Continued. N P AT Community Structure & Balance Metrics continued Genus continued Culex 1 1 1 Dero 1 1 1 Desmopachria 1 1 1 Eclipidrilus 1 1 1 Goeldichironomus 1 1 1 Haemonais 1 1 1 Hydrocanthus 1 1 1 Hydrochus 1 1 1 Ischnura 1 1 1 Kiefferulus 1 1 1 Labrundinia 1 1 1 Larsia 1 1 1 Micromenetus 1 1 1 Monopelopia 1 1 1 Notonecta 1 1 1 Ochlerotatus 1 1 1 Pachydiplax 1 1 1 Pachydrus 1 1 1 Parachironomus 1 1 1 Paramerina 1 1 1 Pelonomus 1 1 1 Physella 1 1 1 Polypedilum 1 1 1 Pristina 1 1 1 Pristinella 1 1 1 Scirtes 1 1 1 Tanytarsus 1 1 1 Tropisternus 1 1 1 Zavreliella 1 1 1 Functional Feeding Group Metrics Browsers and Grazers of Periphyton 1 1 1 Collector-Filterers/Sus pension Feeders 1 1 1 Collector-Gatherers/Deposit Feeders 1 1 1 Macrophyte Piercers 1 1 1 Macrophyte Shredders 1 1 1 Parasites 1 1 1 Periphyton Scrapers 1 1 1 Predators & Carnivores 1 1 1 Scavenger (animals) 1 1 1

PAGE 204

189 APPENDIX F SUMMARY STATISTICS Table F-1. Summary statistics of richness (R ), evenness (E), Shannon diversity (H), and SimpsonÂ’s index (S) for the diat om assemblage (genus level). Site R E HS Site RE H S PR1 12 0.82 2.030.82 CA3 180.63 1.81 0.73 PR4 16 0.63 1.750.69 CA4 170.78 2.21 0.86 PR5 12 0.71 1.770.77 CA5 220.73 2.27 0.83 PR6 11 0.60 1.440.68 CA6 130.89 2.28 0.87 NR2 22 0.68 2.090.77 SA2 310.75 2.57 0.85 NR3 13 0.76 1.960.79 SA3 250.84 2.69 0.90 NR4 12 0.83 2.070.84 SA4 200.78 2.33 0.84 NR6 16 0.81 2.250.87 SA5 240.61 1.95 0.73 CR3 30 0.84 2.870.92 SA6 200.86 2.56 0.90 CR4 22 0.74 2.280.85 PU3 280.83 2.78 0.90 CR5 26 0.81 2.630.86 PU4 200.72 2.17 0.82 CR6 9 0.66 1.450.68 NU2 230.83 2.60 0.89 SR1 19 0.65 1.900.77 NU4 140.66 1.73 0.73 SR2 13 0.59 1.500.63 NU5 120.57 1.41 0.58 SR3 26 0.81 2.630.90 NU6 230.74 2.31 0.85 SR4 22 0.77 2.370.86 CU1 90.76 1.66 0.78 SR5 35 0.84 2.980.93 CU3 390.79 2.89 0.90 SR6 36 0.81 2.890.91 CU5 310.69 2.37 0.78 PA2 12 0.80 1.980.83 CU6 260.84 2.73 0.91 PA3 10 0.70 1.620.73 SU1 170.62 1.77 0.72 PA5 19 0.74 2.170.83 SU2 280.88 2.92 0.93 PA6 17 0.74 2.080.83 SU3 240.73 2.31 0.84 NA4 34 0.78 2.750.89 SU4 210.62 1.87 0.67 NA6 14 0.64 1.680.70 SU5 230.83 2.59 0.89 CA2 14 0.76 2.000.80 SU6 150.59 1.59 0.70

PAGE 205

190 Table F-2. Summary statistics of richness (R ), jackknife estimator s of species richness (Jack1, Jack2), evenness (E), Shannon diversity (H), and WhittakerÂ’s beta diversity ( W) for the macrophyte assemb lage (species level). Site R Jack1 Jack2 E H W PR1 37 43 45 0.89 3.2 3.5 PR2 31 37 41 0.87 3.0 2.6 PR3 37 50 56 0.80 2.9 5.5 PR4 24 30 30 0.79 2.5 4.8 PR5 23 28 31 0.88 2.7 4.0 PR6 27 37 41 0.83 2.7 4.1 PR7 32 43 42 0.71 2.5 8.3 PR8 34 47 54 0.83 2.9 2.0 NR1 14 15 11 0.83 2.2 2.8 NR2 40 55 62 0.84 3.1 6.8 NR3 28 37 44 0.84 2.8 5.1 NR4 32 43 49 0.85 2.9 4.8 NR5 29 38 43 0.85 2.9 6.4 NR6 42 48 49 0.89 3.3 3.6 NR7 35 44 45 0.83 3.0 3.1 NR8 31 38 40 0.86 2.9 3.8 NR9 15 17 16 0.79 2.1 1.2 CR1 31 35 36 0.91 3.1 2.3 CR2 31 40 46 0.82 2.8 6.7 CR3 53 72 84 0.86 3.4 5.0 CR4 40 54 58 0.83 3.1 3.4 CR5 31 46 56 0.89 3.1 9.4 CR6 27 31 32 0.92 3.0 1.4 CR7 49 62 69 0.88 3.4 4.9 CR8 22 28 30 0.79 2.4 1.4 CR9 53 69 76 0.87 3.5 5.0 CR10 35 42 42 0.89 3.2 3.4 CR11 46 56 58 0.91 3.5 3.2 SR1 27 36 40 0.85 2.8 4.1 SR2 25 31 31 0.86 2.8 3.3 SR3 25 31 33 0.86 2.8 4.6 SR4 20 25 28 0.85 2.6 3.1 SR5 29 38 42 0.88 3.0 3.0 SR6 16 20 21 0.91 2.5 1.5 SR7 60 77 79 0.89 3.6 3.8 SR8 40 53 61 0.84 3.1 3.1 SR9 26 34 40 0.85 2.8 1.1

PAGE 206

191 Table F-2. Continued. Site R Jack1 Jack2 E H W PA1 36 45 47 0.85 3.0 7.9 PA2 29 37 39 0.88 3.0 4.9 PA3 29 38 40 0.82 2.7 7.5 PA4 25 34 39 0.87 2.8 6.7 PA5 50 68 75 0.87 3.4 6.2 PA6 34 39 42 0.89 3.1 4.5 PA7 52 62 64 0.89 3.5 5.6 PA8 22 29 33 0.87 2.7 1.6 PA9 44 56 58 0.89 3.4 3.3 PA10 35 53 68 0.88 3.1 3.8 NA1 19 24 25 0.86 2.5 2.8 NA2 36 49 57 0.82 2.9 3.6 NA3 13 17 17 0.72 1.8 5.6 NA4 53 74 85 0.90 3.6 4.7 NA5 45 60 71 0.89 3.4 6.1 NA6 41 50 55 0.86 3.2 4.6 NA7 44 55 60 0.91 3.4 4.0 NA8 21 25 27 0.90 2.7 3.4 NA9 60 73 80 0.92 3.8 3.4 NA10 36 45 46 0.88 3.2 4.1 NA11 53 73 85 0.89 3.5 5.9 NA12 77 99 114 0.90 3.9 6.6 CA1 44 56 64 0.89 3.4 2.8 CA2 18 23 26 0.91 2.6 1.4 CA3 34 45 49 0.88 3.1 4.2 CA4 43 51 53 0.89 3.3 4.8 CA5 26 33 37 0.80 2.6 2.3 CA6 26 34 39 0.80 2.6 6.2 CA7 60 81 91 0.85 3.5 4.6 CA8 47 63 69 0.85 3.3 4.4 CA9 31 41 47 0.90 3.1 6.1 SA1 21 26 29 0.87 2.7 4.3 SA2 34 43 45 0.90 3.2 5.5 SA3 38 45 46 0.91 3.3 2.1 SA4 31 40 45 0.83 2.9 5.8 SA5 27 35 36 0.83 2.7 7.0 SA6 50 65 74 0.86 3.4 4.8 SA7 20 27 31 0.91 2.7 7.2 SA8 40 52 55 0.87 3.2 7.5 SA9 36 47 51 0.87 3.1 5.9

PAGE 207

192 Table F-2. Continued. Site R Jack1 Jack2 E H W PU1 37 47 54 0.89 3.2 5.9 PU2 34 42 43 0.87 3.1 5.8 PU3 42 59 68 0.86 3.2 5.4 PU4 43 57 66 0.93 3.5 6.1 PU5 42 61 76 0.88 3.3 4.4 PU6 29 39 46 0.85 2.8 5.3 PU7 24 30 34 0.91 2.9 0.8 PU8 35 50 59 0.83 2.9 1.6 PU9 16 22 26 0.79 2.2 0.2 PU10 38 55 63 0.87 3.1 7.5 NU1 37 46 50 0.89 3.2 3.8 NU2 46 51 50 0.86 3.3 5.3 NU3 48 60 62 0.88 3.4 5.4 NU4 27 34 38 0.87 2.9 4.0 NU5 34 44 49 0.86 3.0 3.9 NU6 26 39 49 0.78 2.5 7.7 NU7 41 55 65 0.89 3.3 5.1 NU8 41 50 50 0.88 3.3 4.2 NU9 35 49 58 0.86 3.1 1.7 NU10 42 55 61 0.86 3.2 5.5 CU1 46 56 61 0.85 3.2 5.5 CU2 44 54 58 0.88 3.3 3.1 CU3 33 40 43 0.86 3.0 5.4 CU4 42 58 67 0.84 3.1 7.0 CU5 34 40 37 0.85 3.0 4.5 CU6 23 27 24 0.83 2.6 4.0 CU7 45 59 64 0.87 3.3 4.3 CU8 63 92 111 0.85 3.5 4.5 CU9 32 40 45 0.86 3.0 1.7 CU10 26 36 40 0.77 2.5 2.8 CU11 38 49 56 0.90 3.3 3.0 SU1 55 73 81 0.88 3.5 6.4 SU2 38 50 59 0.89 3.2 3.3 SU3 24 32 36 0.85 2.7 4.5 SU4 38 53 63 0.85 3.1 4.8 SU5 16 20 21 0.90 2.5 2.6 SU6 21 30 33 0.83 2.5 3.8 SU7 48 59 60 0.91 3.5 3.5 SU8 26 37 45 0.86 2.8 2.8 SU9 47 59 64 0.90 3.5 5.3 SU10 39 50 56 0.87 3.2 3.9

PAGE 208

193 Table F-3. Summary statistics of richness (R ), evenness (E), Shannon diversity (H), and SimpsonÂ’s index (S) for the macroinvert ebrate assemblage (genus level). Site R E H S Site R E H S PR1 11 0.58 0.62 1.40 CA5 14 0.68 0.71 1.80 PR4 1 0.00 0.00 0.00 CA6 11 0.59 0.60 1.42 PR5 5 0.97 0.78 1.56 CA7 20 0.90 0.92 2.70 PR6 10 0.87 0.84 2.01 CA8 12 0.65 0.70 1.62 PR7 18 0.84 0.89 2.41 CA9 9 0.55 0.59 1.20 PR8 15 0.69 0.74 1.86 SA2 14 0.79 0.84 2.09 NR2 8 0.63 0.65 1.30 SA3 17 0.73 0.78 2.07 NR3 24 0.86 0.91 2.72 SA4 7 0.53 0.48 1.03 NR4 21 0.82 0.89 2.50 SA5 15 0.89 0.89 2.41 NR6 19 0.88 0.90 2.59 SA6 14 0.68 0.73 1.79 NR8 4 0.15 0.08 0.21 SA7 11 0.60 0.63 1.45 NR9 11 0.56 0.59 1.34 SA8 19 0.73 0.81 2.14 CR3 11 0.76 0.80 1.81 SA9 8 0.51 0.46 1.05 CR4 17 0.82 0.87 2.33 PU3 23 0.84 0.90 2.63 CR5 16 0.76 0.84 2.12 PU4 16 0.79 0.82 2.19 CR6 20 0.86 0.90 2.57 PU10 13 0.83 0.85 2.13 CR8 8 0.29 0.23 0.60 NU2 7 0.76 0.71 1.47 CR9 7 0.37 0.29 0.72 NU4 11 0.49 0.47 1.16 CR10 25 0.87 0.92 2.79 NU5 8 0.83 0.77 1.72 CR11 21 0.84 0.89 2.55 NU6 10 0.72 0.73 1.66 SR1 17 0.67 0.72 1.90 NU10 8 0.68 0.66 1.41 SR2 10 0.87 0.84 2.01 CU1 25 0.82 0.90 2.64 SR3 18 0.76 0.82 2.21 CU3 9 0.68 0.69 1.49 SR4 11 0.78 0.79 1.87 CU5 9 0.57 0.63 1.25 SR5 21 0.84 0.89 2.56 CU6 11 0.51 0.54 1.22 SR6 15 0.68 0.71 1.84 CU7 11 0.72 0.75 1.72 SR7 11 0.61 0.60 1.46 CU8 21 0.89 0.91 2.72 SR8 22 0.85 0.89 2.62 CU9 13 0.69 0.70 1.78 SR9 14 0.45 0.44 1.19 CU10 5 0.19 0.12 0.31 PA2 24 0.66 0.77 2.09 CU11 17 0.71 0.77 2.02 PA3 26 0.78 0.87 2.54 SU1 11 0.63 0.69 1.50 PA5 18 0.79 0.85 2.28 SU2 13 0.79 0.82 2.04 PA6 8 0.63 0.63 1.32 SU3 19 0.65 0.73 1.91 NA4 13 0.83 0.84 2.13 SU4 12 0.58 0.60 1.44 NA6 9 0.73 0.77 1.60 SU5 16 0.87 0.89 2.42 NA10 5 0.38 0.32 0.61 SU6 20 0.63 0.75 1.89 NA11 18 0.81 0.87 2.35 SU7 14 0.67 0.74 1.76 CA2 20 0.73 0.78 2.20 SU8 8 0.65 0.65 1.36 CA3 8 0.67 0.65 1.40 SU9 10 0.47 0.45 1.09 CA4 18 0.76 0.84 2.20

PAGE 209

194 LIST OF REFERENCES Adams, S.M. 2002. Biological indicators of aquatic ecosystem st ress: introduction and overview. Pages 1-11 in S.M. Adams, ed itor. Biological indi cators of aquatic ecosystem stress. American Fisheries Society. Bethesda, Maryland, USA. Adamus, P.R. 1996. Bioindicators for assessi ng ecological integrity of prairie wetlands. EPA600-R-96/082. National Health and Environmental Effects Laboratory, Western Ecology Division, United States Environmental Protection Agency. Corvallis, Oregon, USA. Adamus, P.R. and K. Brandt. 1990. Impacts on quality of inland wetlands of the United States: a survey of indicators, techniques, and applications of community level biomonitoring data. EPA/600/3-90/073. Offi ce of Research and Development, United States Environmental Protection Ag ency. Washington, D.C., USA. American Public Health Association (APHA). 1995. Standard methods for the examination of water and wastewater, 20th edition. L.S. Clesceri, A.E. Greenberg, and A.D. Easton, editors. Washington, D.C., USA. Analyse-it Software, Ltd. 1997-2003. vers ion 1.67. Leeds, England, United Kingdom. Andreas, B.K. and R.W. Lichvar. 1995. A fl oristic assessment system for northern Ohio. Wetlands Research Program Technical Re port WRP-DE-8. U.S. Army Corps of Engineers Waterways Experiment Sta tion, Vicksburg, Mississippi, USA. Apfelbeck, R. 2000. Developing preliminary bioassessment protocols for Montana wetlands, State of Montana Department of Environmental Quality. Helena, Montana, USA. Arcview GIS 3.2 Environmental Systems Res earch Institute, Inc. 1999. Neuron Data, Inc. 1991-1996. Portions copyright 19911995 Arthur D. Applegate. Found at: http://www.esri.com/ . Redlands, California, USA. Bahls, L. 1993. Periphyton bioassessment me thods for Montana streams. Water Quality Bureau, Department of Health and Envir onmental Science, Helena, Montana, USA. Barbour, M.T., J. Gerritsen, G.E. Griffith, R. Frydenborg, E. McCarron, J.S. White, and M.L. Bastian. 1996a. A framework for biol ogical criteria for Florida streams using benthic macroinvertebrates. Journal of the North American Benthological Society 15(2): 185-211.

PAGE 210

195 Barbour, M.T., J. Gerritsen, B.D. Snyder, and J.B. Stribling. 1999. Rapid bioassessment protocols for use in streams and wadeab le rivers, 2nd edition. EPA 841-B-99-002. United States Environmental Protection Agency, Office of Water, Washington, D.C., USA. Barbour, M.T., J. Gerritsen, and J.S. White. 1996b. Development of the Stream Condition Index (SCI) for Florida. A Report to the Florida Department of Environmental Protection, Stormwater and Nonpoint Source Management Section. Tetra Tech, Inc. Owing Mills, Maryland, USA. Batzer, D.P. and S.A. Wissinger. 1996. Ecology of insect communities in nontidal wetlands. Annual Review of Entomology 41: 75-100. Beck, W.M. 1954. Studies in stream po llution biology I: a simplified ecological classification of organisms. Quarterly Journal of the Florida Academy of Science 17(4): 211-227. Bedford, B.L., M.R. Wabridge , and A. Aldous. 1999. Pattern s in nutrient availability and plant diversity of temperate Nort h American wetlands. Ecology 80(7): 21512169. Blanch, S.J. and M.A. Brock. 1994. Effects of grazing and depth on two wetland plant species. Australian Journal of Marine and Freshwater Research 45: 1387-1394. Brady, N.C. and R.R. Weil. 2004. Elements of the Nature and Properties of Soils Prentice-Hall, Inc. Englewood Cliffs, New Jersey, USA. Brandt, K. and K.C. Ewel. 1989. Ecology and management of cypress swamps: a review. University of Florida Extension Bulletin 252. Gainesvill e, Florida, USA. Brightman, R.S. 1984. Benthic macroinvertebrate response to secondarily treated wastewater in north-central Florida cypress domes. Pages 186-196 in K.C. Ewel and H.T. Odum, editors. Cypress swam ps. University Presses of Florida, Gainesville, Florida, USA. Brown, M.T. and M.B. Vivas. 2004, accepte d. A Landscape Development Intensity Index. Environmental Monitoring and Assessment. Brown, R.B., E.L. Stone, and V.W. Carlisle. 1990. Soils. Pages 35-69 in R.L. Myers and J.J. Ewel, editors. Ecosystems of Flor ida. University of Central Florida Press, Orlando, Florida, USA. Butcher, J.T., P.M. Stewart, and T.P. Simon. 2003. A benthic community index for streams in the northern lakes and forest s ecoregion. Ecological Indicators 3: 181193.

PAGE 211

196 Cairns, J. and J.R. Pratt. 1993. A hist ory of biological moni toring using benthic macroinvertebrates. Pages 10-27 in D.M. Rosenberg and V.H. Resh, editors. Freshwater biomonitoring and benthic m acroinvertebrates. Chapman and Hall, New York, New York, USA. Carlisle, B.K., A.L. Hicks, J.P. Smith, S.R. Garcia, and B.G. Largay. 1999. Plants and aquatic invertebrates as indicators of wetland biological integrity in Waquoit Bay watershed, Cape Cod. Environm ent Cape Cod 2(2): 30-60. Charles, D., F. Aker, and N.A. Roberts. 1996. Diatom periphyton in Montana lakes and wetlands: ecology and potential as bioassessm ent indicators. Patrick Center of Environmental Research, Environmental Research Division. The Academy of Natural Sciences, Philadel phia, Pennsylvania, USA. Clarke, K.R. 1993. Non-parametric multivar iate analyses of changes in community structure. Australian Journal of Ecology 18: 117-143. Cohen, M.J., S.M. Carstenn, and C.R. Lane. 2004, accepted. Floristic quality assessment of isolated marshes in Florida along agricu ltural disturbance gradients. Ecological Applications. Colwell, R.K. and J.A. Coddington. 1994. Estimating terrestrial biodiversity through extrapolation. Philosophical Transactions of the Royal Society of London, Series B 345: 101-118. Coultas, C.L. and M.J. Duever. 1984. Soils of Cypress Swamps. Pages 51-59. K.C. Ewel and H.T. Odum, editors. Cypress sw amps. University Presses of Florida, Gainesville, Florida, USA. Cowardin, L.M., V. Carter, F.C. Goulet, and E.T. LaRoe. 1979. Classification of wetlands and deepwater habitats of the Unite d States. United States Department of the Interior, Fish and Wildlife Service, Washington, D.C., USA. Cronk, J. K. and M.S. Fennessy. 2001. We tland plants: biology and ecology. Lewis Publishers. Boca Raton, Florida, USA. Crowder, A. and D.S. Painter. 1991. Submer ged macrophytes in Lake Ontario: current knowledge, importance, threats to stability, and needed studies. Canadian Journal of Fisheries and Aqua tic Sciences 48:1539-1545. Cummins, K.W. and R.W. Merri tt. 2001. Application of inve rtebrate functional groups to wetland ecosystem function and biomon itoring. Pages 85-111 in R.B. Rader, D.P. Batzer, and S.A. Wissinger, editors. Bioassessment and management of North American freshwater wetlands. John W iley and Sons, Inc., New York, New York, USA.

PAGE 212

197 Dahl, T.E. 2000. Status and trends of we tlands in the conterminous United States 1986 to 1997. United States Department of the Interior, Fish and Wildlife Service, Washington, D.C., USA. Danielson, T.J. 1998a. Indicators for monito ring and assessing biol ogical integrity of inland freshwater wetlands. EPA 843R-98-002. Wetlands Division Office of Water, United States Environmental Pr otection Agency, Washington, D.C., USA. Danielson, T.J. 1998b. Wetland bioassessmen t fact sheets. EPA 846-98-001. Wetlands Division Office of Water, United Stat es Environmental Protection Agency, Washington, D.C., USA. David, P.G. 1999. Response of exotics to restored hydroperiod at Dupuis Reserve, Florida. Restoration Ecology 7(4): 407-410. de Szalay F.A. and V.H. Resh. 1996. Sp atial and temporal variability of trophic relationships among aquatic macroinvertebrat es in a seasonal marsh. Wetlands 16: 458-466. Demaree, D. 1932. Submerging experiments with Taxodium . Ecology 13:258-262. Devall, M.S. 1998. An interim old-growth definition for cypress-tupelo communities in the southeast. U.S. Department of Agricu lture, Forest Service, Southern Research Station-19. Asheville, NC, USA. Devine, R. 1998. Alien invasion. Nati onal Geographic Society, Washington, D.C., USA. Dierberg, F.E. and P.L. Berzonik. 1984. Nitrogen and phosphorus mass balance in a cypress dome receiving wastewater. Pa ges 112-118 K.C. Ewel and H.T. Odum, editors. Cypress swamps. University Pre sses of Florida, Gainesville, Florida, USA. Doherty, S.M., M. Cohen, C. Lane, L. Line, an d J. Surdick. 2000. Biol ogical criteria for inland freshwater wetlands in Florida: a revi ew of technical and scientific literature (1990-1999). Report to the United States Environmental Protection Agency, Center for Wetlands, University of Fl orida, Gainesville, Florida, USA. Dufrêne, M. and P. Legendre. 1997. Species assemblages and indicator species the need for a flexible asymmetrical approach . Ecological Monographs 67(3): 345-366. Edmunds, Jr., G.F. and R.D. Waltz. 1996. E phemeroptera. Pages 126-163 in R.W. Merritt and K.W. Cummins, editors. An introduction to the aquatic insects of North America, 3rd edition. Kendall /Hunt Publishing Company, Dubuque, Iowa, USA.

PAGE 213

198 Ehrenfeld, J.G. and J.P. Schneider. 1991. Chamaecyparis thyoides wetlands and suburbanization: effects of non-point source water pollution on hydrology and plant community structure. Journal of Applied Ecology 28(2): 467-490. Evans, D.L., W.J. Streever, and T.L. Cris man. 1999. National flatwoods marshes and created freshwater marshes of Florida: factors influencing aq uatic invertebrate distribution and comparisons between na tural and created marsh communities. Pages 81-104 in D.P. Batzer, R.B. Ra der, and S.A. Wissinger, editors. Invertebrates in freshwater wetlands of North America: ecology and management. John Wiley and Sons, Inc. New York, New York, USA. Ewel, K.C. 1984. Effects of fire and wast ewater on understory vegetation in cypress domes. Pages 119-126 in K.C. Ewel and H.T. Odum, editors. Cypress swamps. University Presses of Florida, Gainesville, Florida, USA. Ewel, K.C. 1990. Swamps. Pages 281-323 in R.L. Myers and J.J. Ewel, editors. Ecosystems of Florida. University of Ce ntral Florida Press, Orlando, Florida, USA. Ewel, K.C., H.J. Davis, and J.E. Smith. 1989. Recovery of Florid a cypress swamps from clearcutting. Southern Journal of Applied Forestry 13: 123-126. Fennessy, S., R. Geho, B. Elifritz and R. Lopez. 1998. Testing the floristic quality assessment index as an indicator of ripari an wetland quality. Final report to U.S. EPA. Ohio Environmental Protection Agency, Division of Surface Water, Columbus, Ohio, USA. Fennessy, S., M. Gernes, J. Mack, and D. H. Waldrop. 2001. Methods for evaluating wetland condition: using vegetation to asse ss environmental conditions in wetlands. EPA 822-R-01-007j. U.S. Environmental Protection Agency, Office of Water, Washington, D.C., USA. Fernald, E.A. and E.D. Purdum, editors. 1992. Atlas of Florida. University Press of Florida. Gainesville, Florida, USA. Findlay, C.S., and J. Houlahan. 1997. Anth ropogenic correlates of species richness in Southeastern Ontario wetlands. Conservation Biology, 11(4): 1000-1009. Florida Department of Natural Resources (FDNR). 1988. Wetlands in Florida. An Addendum to Department of Land C onservation and Development and the Division of State Lands. Fishman Envi ronmental Services. Portland, Oregon, USA. Fore, L.S. 2003. Development and testi ng of biomonitoring tools for stream macroinvertebrates in Florida streams, draf t report 2. Statisti cal Design, Seattle, Washington. A report for the Florida De partment of Environmental Protection, Tallahassee, Florida, USA.

PAGE 214

199 Fore, L.S. and C. Grafe. 2002. Using diatom s to assess the biologi cal condition of large rivers in Idaho (U.S.A.). Freshwater Biology 47: 2015-2037. Francis, C.M., M.J.W. Austen, J.M. Bowl es, W.B. Draper. 2000. Assessing floristic quality in southern Ontario woodlan ds. Natural Areas Journal 20:66-77. Frappier B. and R.T. Eckert. 2003. Utilizi ng the USDA PLANTS database to predict exotic woody plant invasiveness in New Hampshire. Forest Ecology and Management 185(1-2): 207-215. Galatowitsch, S.M., N.O. Anderson, and P.D. Ascher. 1999b. Invasiveness in wetland plants in temperate North Amer ica. Wetlands 19(4): 733-755. Galatowitsch, S.M., D.C. Whited, R. Lehtinen, J. Husveth, and K. Schik. 2000. The vegetation of wet meadows in relation to their land-use. Ecological Monitoring and Assessment 60: 121-144. Galatowitsch, S.M., D.C. Whited, and J.R. Tester. 1999a. Development of community metrics to evaluate recovery of Minnesot a wetlands. Journal of Aquatic Ecosystem Stress and Recovery 6: 217-234. Gardner, W.H. 1986. Water content. Pages 493-544 in Methods of so il analysis: part 1 physical and mineralogical methods, 2nd ed ition. American Society of Agronomy, Inc. and Soil Science Society of America, Inc. Madison, Wisconsin, USA. Gernes, M.C. and J.C. Helgen. 1999. Indexes of biotic integrity for wetlands, section B: wetland vegetation IBI for depressional wetla nds. Final Report to the United States Environmental Protection Agency A ssistance Number CD995525-01, April 1999. Minnesota Pollution Control Agency, St. Paul, Minnesota, USA. Gerristen, J., M.T. Barbour, and K. King. 2000. Apples, oranges, and ecoregions: on determining pattern in aquatic assembla ges. Journal of the North American Benthological Association 19(3): 487-496. Gerristen, J. and J. White. 1997. Development of a biological index fo r Florida lakes. A Report to the Florida Department of Envi ronmental Protection. Tetra Tech, Inc. Owing Mills, Maryland, USA. Godfrey, R.K. and J.W. Wooten. 1981. Aquatic and wetland plants of the southeastern United States. University of Geor gia Press, Athens, Georgia, USA. Goldsborough, G. 2001. Sampling algae in wetlands. Pages 263-295 in R.B. Rader, D.P. Batzer, and S.A. Wissinger, editors. Bioassessment and management of North American freshwater wetlands. John W iley and Sons, Inc., New York, New York, USA.

PAGE 215

200 Gore, J.A., H.L. Griffith, III. and D.S. Addison. 1998. Inventory of the freshwater macroinvertebrates in hydric pine flatw oods for the district's isolated wetland monitoring program. South Flor ida Water Management District. West Palm Beach, Florida, USA. Grace, J.B. and H. Jutila. 1999. The relationship between species density and community biomass in grazed and ungrazed coastal meadows. Oikos 85: 398-408. Graves, G.A., D.G. Strom and B.E. Robson. 1998. Stormwater impact to the freshwater Savannas Preserve marsh, Florid a. Hydrobiologia 379: 111-122. Griffith, M.B., P. Husby, R.K. Hall, P.P. Kaufmann, and B.H. Hill. 2003. Analysis of macroinvertebrate assemblages in relati on to environmental gradients among lotic habitats of CaliforniaÂ’s Ce ntral Valley. Environmenta l Monitoring and Assessment 82: 281-309. Haag, K.H, J.C. Joyce, W.M. Hetrick, a nd J.C. Jordan. 1987. Predation on Water hyacinth weevils and other aquatic insect s by three wetland birds in Florida. Florida Entomologist 70(4): 457-471. Harms, W.R., H.T. Schreuder, D.D. Hook, C. L. Brown, and F.W. Schropshire. 1980. The effects of flooding on the swamp forest in Lake Oklawaha, Florida. Ecology 61: 1412-1421. Harper, H.H. 1994. Stormwater loading rate parameters for central and south Florida. Environmental Research and Design, Inc. Orlando, Florida, USA. Harris, L.D. and C.R. Vickers. 1984. Some faunal community characteristics of cypress ponds and the changes produced by perturba tions. Pages 171-185 in K.C. Ewel and H.T. Odum, editors. Cypress swamps . University Presses of Florida, Gainesville, Florida, USA. Hasle, G.R., and G.A. Fryxell. 1970. Di atoms: cleaning and mounting for light and electron microscopy. Transactions of th e American Microscopy Society 89 (4): 469-474. Helgen, J. 2001. Methods for evalua ting wetland condition: #9 developing an invertebrate index of bi ological integrity for wetl ands. EPA 843-B-00-002h. Office of Water, United States Environmental Protection Agency, Washington, D.C., USA. Herman, K.D., A.A. Reznicek, L.A. Masters, G.S. Wilhelm, M.R. Penskar and W.W. Brodowicz. 1997. Floristic quality assessm ent: development and application in the state of Michigan (USA). Natural Areas Journal 17:265-279.

PAGE 216

201 Hilsenhoff, W.L. 1987. An improved biotic index of organic str eam pollution. Great Lakes Entomology 20: 31-39. Hobbs, R.J. and L.F. Hueneke. 1992. Disturba nce, diversity, and invasion: implications for conservation. Conservation Biology 6(3): 324-337. James, M.O. and K.M. Kleinow. 1994. Trophi c transfers of chemicals in the aquatic environment. Pages 1-35 in D.C. Malins and G.K. Ostrander, editors. Aquatic toxicology and cellular pers pectives. Lewis Publishe rs. Boca Raton, Florida, USA. 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. Karr, J.R. 1981. Assessment of biotic integr ity using fish communities. Fisheries 6: 2127. Karr, J.R. 1993. Defining and assessing ecological integrity: beyond water quality. Environmental Toxicology and Chemistry 12: 1521-1531. Karr, J.R. and E.W. Chu. 1997. Biol ogical monitoring and assessment: using multimetric indexes effectively. EPA 235-R-97-001. University of Washington, Seattle, Washington, USA. Karr, J.R. and E.W. Chu. 1999. Restoring life in running waters . Island Press. Washington, D.C., USA. Karr, J.R. and D.R. Dudley. 1981. Ecologica l perspectives on water quality goals. Environmental Management 5: 55-68. Keddy, P.A. 2000. Wetlands ecology: prin ciples and conservation. Cambridge University Press, Cambridge, United Kingdom. Kelly, M.G. and B.A. Whitton. 1998. Biologica l monitoring of eutrophication in rivers. Hydrobiologia 384: 55-67. Kent, D.M. 2000. Evaluating wetland functions and values. Chapter 3 in D.M. Kent, editor. Applied wetlands science and t echnology. Lewis Publishers. Boca Raton, Florida, USA. Kerans, B.L. and J.R. Karr. 1994. A benthic index of biotic integrity (B-IBI) for rivers in the Tennessee valley. Ecologi cal Applications 4(4): 768-785. Kruskal, J.B. 1964. Multidimensional scaling by optimizing goodness of fit to a nonmetric hypothesis. Psychometrika 29: 115-129.

PAGE 217

202 Kushlan, J.A. 1990. Freshwater marshes. Pages 324-363 in Myers, R.L. and J.J. Ewel, editors. Ecosystems of Florida. Univer sity of Central Florida Press. Orlando, Florida, USA. Lane, C.R. 2000. Proposed ecological regions for freshwater wetlands of Florida. Masters Thesis, University of Florida, Gainesville, Florida, USA. Lane, C.R. 2003. Biological indicators of wetland condition for isolated depressional herbaceous wetlands in Florida. Ph.D. Dissertation, University of Florida, Gainesville, Florida, USA. Lane, C.R., M.T. Brown, M. Murray-Hudson, and M.B. Vivas. 2002. The development of the Wetland Condition Index for Florida. A Report to the United States Environmental Protection Agency, Biologi cal Assessment of Wetlands Workgroup. Center for Wetlands, University of Fl orida, Gainesville, Florida, USA. Lange-Bertalot, H. 1979. Pollution tolerance of diatoms as a criterion for water quality estimation. Nova Hedwigia 54: 285-304. Lemlich, S.K. and K.C. Ewel. 1984. Effects of wastewater disposal on growth rates of cypress trees. Journal of Envi ronmental Quality 13(4): 602-604. Lenat, D.R. 1993. A biotic index for the sout heastern United States : derivations and list of tolerance values, with criteria for assi gning water-quality ratings. Journal of the North American Benthologi cal Society 12(3): 279-290. Lenat, D.R. and M.T. Barbour 1994. Us ing benthic macroinvertebrate community structure for rapid, cost-effective, water qu ality monitoring: rapid bioassessment. Pages 187-215 in S.L. Loeb and A. Spacie, editors. Biolog ical monitoring of aquatic systems. Lewis Publishers, Boca Raton, Florida, USA. Leslie, A.J., J.P. Prenger, and T.L. Crisma n. 1999. Cypress dome s in North Florida: invertebrate ecology and response to huma n disturbance. Pages 105-119 in D.P. Batzer, R.B. Rader, and S.A. Wissinger, editors. Invertebra tes in freshwater wetlands of North America. John Wile y and Sons, Inc., New York, New York, USA. Lugo, A.E., and S.L. Brown. 1986. The Ockl awaha River forested wetlands and their response to chronic flooding. Pages 365-373 in K.C. Ewel and H.T. Odum, editors. Cypress swamps. University Presses of Florida. Gainesville, Florida, USA.

PAGE 218

203 Mack, J. 2001. Vegetation Index of Bi ological Integrity (VIBI) for wetlands: ecoregional, hydrogeomorphologic, and pl ant community comparisons with preliminary wetland aquatic life use designati ons. Final Report to the United States Environmental Protection Agency Grant No. CD985875, Volume 1. Wetland Ecology Group, Division of Surface Wate r, Ohio Environmental Protection Agency, Columbus Ohio, USA. Found at: http://www.epa.state.oh.us/dsw.w etlands/wetlands_bioasses.html . Mack, J.J., M. Micacchion, L.D. Augusta, and G.R. Sablak. 2000. Vegetation Indices of Biotic Integrity (VIBI) for wetlands and calibration of the Ohio Rapid Assessment Method for Wetlands v 5.0. Final Report to the United States Environmental Protection Agency Grant No. CD985 276, Interim report Grant No. CD985875, Volume 1. Wetland Ecology Group, Division of Surface Water, Ohio Environmental Protection Agency, Columbus, Ohio, USA. Found at: http://www.epa.state.oh.us/dsw.w etlands/wetlands_bioasses.html . Marois, K.C. and K.C. Ewel. 1983. Natural a nd management related variation in cypress domes. Forest Science 29(3):627-640. McCormick, P.V. and J. Cairns, Jr. 1994. Al gae as indicators of environmental change. Journal of Applied Phychology 6: 509-526. McCune, B. and J.B. Grace. 2002. Analysis of ecological communities. MJM Software Design. Gleneden Beach, Oregon, USA. McCune, B., R. Rosentreter, J.M. Ponzetti, and D.C. Shaw. 2000. Epiphyte habitats in an old conifer forest in western Wa shington, USA. Br yologist 103: 417-427. Means, D.B., J.G. Palis, and M. Baggett. 1998. Effects of slash pi ne silviculture on a Florida population of flatwoods salama nder. Conservation Biology 10(2): 426-437. Merritt, R.W. and K.W. Cummins. 1996. In troduction. Pages 1-4 in R.W. Merritt and K.W. Cummins, editors. An introduction to the aquatic insects of North America, 3rd edition. Kendall/Hunt Publishing Company. Dubuque, Iowa, USA. Merritt, R. W., J.R. Wallace, M.J. Higgins, M.K. Alexander, M.B. Berg, W.T. Morgan, K.W. Cummins, and B. Va ndeneeden. 1996. Procedures for the functional analysis of invertebrate communitie s of the Kissimmee River-floodplain ecosystem. Florida Scientist 59(4): 216-274. Miller, R.E., Jr. and B.E. Boyd. 1999. Wetla nd rapid assessment procedure. South Florida Water Management District, Tec hnical Publication REG-001. West Palm Beach, Florida, USA.

PAGE 219

204 Miller, R.E., Jr. and B.E. Gunsalus. 1997. Es tuarine wetland rapid assessment procedure for mitigation banks in Florida. South Florida Water Management District, Technical Publication. West Palm Beach, Florida, USA. Minitab Statistical Software , version 13.1. 2000. Found at: http://www.minitab.com . State College, Pennsylvania, USA. Mitsch, W.J. and J.G. Gosselink. 1993. We tlands, 2nd edition. John Wiley and Sons, Inc. New York, New York, USA. Munn, M.D., R.W. Black, and S.J. Gruber. 2002. Response of benthic algae to environmental gradients in an agricultura lly dominated landscape. Journal of the North American Benthologi cal Society 21 (2): 221-237. Mushet, D.M., N.H. Euliss, and T.L. Shaffer. 2002. Floristic qual ity assessment of one natural and three restored wetland comple xes in North Dakota, USA. Wetlands 22(1): 126-138. Nessel, J.K., K.C. Ewel, and M.S. Burnet t. 1982. Wastewater en richment increases mature pondcypress growth rates. Forest Science 28(2):400-403. OÂ’Connell, T.J., L.E. Jackson, and R.P. Brooks . 1998. A bird community index of biotic integrity for the mid-Atlantic highl ands. Environmental Monitoring and Assessment 51: 145-156. Odum, H.T. 1978. Principles for interfacing wetlands with development. Pages 29-56 in M.A. Drew, editor. Environmental quality through wetland utilization. Symposium proceeding by the Coordinati ng Council on the Restoration of the Kissimmee River Valley and Taylor Cr eek-Nubbins Slough Basin. Tallahassee, Florida, USA. Odum, H.T. 1995. Environmental Accoun ting: Emergy and Environmental Decision Making. John Wiley and Sons, New York, New York, USA. Odum, H.T. 1994. Ecological and general syst ems: an introduction to systems ecology. University Press of Colorado, Niwot, Colorado, USA. Ott, R.L. and M. Longnecker. 2001. An in troduction to statistical methods and data analysis, 5th edition. Duxbury, Wadswo rth Group. Pacific Grove, California, USA. Pan, Y. and R.J. Stevenson. 1996. Gradient an alysis of diatom assemblages in western Kentucky wetlands. Journal of Phycology 32: 222-232. Pan, Y., R.J. Stevenson, B.H. Hill, A.T. Her lihy, and G.B. Collins. 1996. Using diatoms as indicators of ecological conditions in lotic systems: a regional assessment. Journal of the North American Be nthological Society 15(4): 481-495.

PAGE 220

205 PCORD, version 4.1. MJM Software.. Found at: http://home.centurytel.net/~mjm/ . Gleneden Beach, Oregon, USA Peckarsky, B.L., P.R. Fraissinet, M.A. Pent on, and D.J. Conklin, Jr. 1993. Freshwater macroinvertebrates of northeastern Nort h America. Comstock Publishing Associates, Cornell University Pr ess. Ithaca, New York, USA. Pickard, D.P. and A.C. Benke. 1996. Production dynamics of Hyalella azteca (Amphipoda) among different habitats in a sm all wetland in the southeastern USA. Journal of the North American Be nthological Society 15(4): 537-550. Rader, R.B. and C.J. Richardson. 1992. The effects of nutrient enrichment on algae and macroinvertebrates in the Everglades: A review. Wetlands 12(2): 121-135. Rader, R.B. and C.J. Richardson. 1994. Re sponse of macroinvertebrates and small fish to nutrient enrichment in the norther n Everglades. Wetlands 14(2): 134-146. Raschke, R.L. 1993. Diatom (Bacillariophyta ) community response to phosphorus in the Everglades National Park, USA. Phycologia 32 (1): 48-58. Resh, V.H. and J.K. Jackson. 1993. Ra pid assessment approaches in benthic macroinvertebrate biomonitoring studies. Pages 195-233 in D.M. Rosenberg and V.H. Resh, editors. Freshwater biomon itoring and benthic macroinvertebrates. Chapman and Hall, New York, New York, USA. Rich, P.M. 1989. A manual for analysis of hemispherical canopy photography. Los Alamos National Laboratory. Lo s Alamos, New Mexico, USA. SAS Institute, Inc. (SAS). 1990. SAS userÂ’s guide, version 6, 4th edition. SAS Institute, Inc. Cary, North Carolina, USA. Schindler, D.W. 1987. Detecting ecosyste m responses to anthropogenic stress. Canadian Journal of Fisherie s and Aquatic Sciences 44:6-25. Schulz, E.J., M.V. Hoyer, and D.E. Canfiel d, Jr. 1999. An index of biotic integrity: a test with limnological and fish data from si xty Florida lakes. Transactions of the American Fisheries Society 128: 564-577. Sharitz, R. and D. Batzer. 1999. An introduction to freshwater wetlands in North America and their invertebrates. Pages 122 in Batzer, D.P., R.B. Rader, and S.A. Wissinger, editors. Invertebrates in fres hwater wetlands of North America. John Wiley and Sons, Inc. New York, New York, USA. Smogor, R.A. and P.L. Angermeier. 2001. Determining a regional framework for assessing biotic integrity of Virginia st reams. Transactions of the American Fisheries Society 130:18-35.

PAGE 221

206 Snyder, B.D., M.T. Barbour and E.W. Le ppo. 1998. Development of a watershed-based approach for biomonitoring of fresh surf ace waters in southern Florida canal systems. Prepared for Metro-Dade Environmental Resources Management, Environmental Monitoring Di vision, Miami, FL, USA. Spurrier, E. 2000. Hemispherical canopy photogr aphy in isolated forested wetlands in Florida. Masters Thesis, University of Florida, Gainesville, Florida, USA. Stansly, P.A., J.A. Gore, D.W. Ceilley and M.B. Main. 1997. Inventory of freshwater macroinvertebrates. Contract # C-7949 Fi nal Report for the South Florida Water Management District Isolated Wetland Monitoring Program. West Palm Beach. Florida, USA. Stevenson, R.J. 2001. Using algae to asse ss wetlands with multivariate statistics, multimetric indices, and an ecological risk assessment framework. Pages 113-140 in D.P. Batzer, R.B. Rader, and S.A. Wissinger, editors. Bioassessment and management of North American freshwat er wetlands. John Wiley and Sons, New York, New York, USA. Stevenson, R.J. , P.R. Sweets, Y. Pan, and R.E. Schultz. 1999. Algal community patterns in wetlands and their use as indicators of ecological conditions. Pages 517-527 in A. J. McComb and J. A. Davis, editors . Proceedings of INTECOLÂ’s 5th International Wetland Conference. Gleneagles Press, Adelaide, Australia. Stevenson, R.J., and B. Wang. 2001. Developi ng and testing algal indicators of nutrient status in Florida streams. A report for the Florida Department of Environmental Protection, Tallahassee, Florida, USA. Stewart, P.M., J.T. Butcher, and P.J. Gerovac. 1999. Diatom (Bacillariophyta) community response to water quality and la nd use. Natural Areas Journal 19: 155165. Tabachnick, B.G. and L.S. Fidell. 1983. Us ing multivariate statistics. Harper and Rox Publishers. Philadelphi a, Pennsylvania, USA. ter Braak, C.J.F. 1987. The analysis of vegetation-environmental relationships by canonical correspondence analys is. Vegetatio 69: 69-77. Terwilliger, V.J. and K.C. Ewel. 1986. Rege neration and growth after logging in Florida cypress domes. Forest Science 32: 493-506. Tobe, J.D., K. Craddock Burks, R.W. Cantre ll, M.A. Garland, M.E. Sweeley, D.W. Hall, P. Wallace, G. Anglin, G. Nelson, J.R. Coope r, D. Bickner, K. Gilbert, N. Aymond, K. Greenwood, and N. Raymond. 1998. Flor ida wetland plants: an identification manual. Florida Department of Envir onmental Protection, Tallahassee, Florida, USA.

PAGE 222

207 Toth, L.A. 1993. The ecological basis of the Kissimmee River Restor ation Plan. Florida Scientist 56(1): 25-51. United States Department of Agriculture, Natural Resource Conservation Service (USDA NRCS). 2002. The PLANTS Database, Vers ion 3.5. National Plant Data Center, Baton Rouge, Louisiana, US A. Available on-line at: http://plants.usda.gov . Accessed 2001-2004. United States Environmental Protection Agency (USEPA). 1979. 351.2. Nitrogen, Kjeldahl, total (colorimetric; semi-autom ated digester, AAII) . CAS # N Nitrogen 7727-37-9. EPA/600/4-79/020 Methods for the chemical analysis of water and wastes. Found at: http://www.epa.gov/epahome/index/ . Boston, Massachusetts, USA. United States Environmental Protection Agency (USEPA). 1990. Feasibility report on environmental indicators for surface water programs. Office of Water Regulations and Standards and Office of Policy, Pla nning and Evaluation. Washington, D.C., USA. United States Environmental Protection Agen cy (USEPA). 1993. 365.1 Determination of phosphorus by semi-automated colorimetry methods for the determination of inorganic substances in environmenta l samples. EPA-600-R-93-100. Found at: http://www.epa.gov/epahome/index/ . Boston, Massachusetts, USA. United States Environmental Protection Agen cy (USEPA). 1998. Lake and reservoir bioassessment and biocriteria: EPA 841B-98-007 Technical Guidance Document. Washington, D.C., USA. Available on-line at: http://www.epa.gov/owow/m onitoring/tech/lakes.html . Accessed 2002-2004. United States Environmental Protection Ag ency (USEPA). 2002c. Methods for evaluating wetland condition: developing an invertebrate index of biological integrity for wetlands. EPA-822-R-02-019. Office of Water, Washington, D.C., USA. United States Environmental Protection Ag ency (USEPA). 2002a. Methods for evaluating wetland condition: introduction to biological assessment. EPA-822-R02-014. Office of Water, Washington, D.C., USA. United States Environmental Protection Ag ency (USEPA). 2002b. Methods for evaluating wetland condition: using algae to assess environmental conditions in wetlands. EPA-822-R-02-021. Office of Water, Washington, D.C., USA. United States Environmental Protection Agency (USEPA). 2003. Bi ological Indicators of Watershed Health. Available on-line at: http://www.epa.gov/bioindicators . Accessed 2003-2004.

PAGE 223

208 van Dam, H., A. Mertens, and J. Sinkelda m. 1994. A coded checklist and ecological indicator values of freshwat er diatoms from the Nether lands. Netherlands Journal of Aquatic Ecology 28: 117-133. Vernon. R.O. 1947. Cypress domes. Science 105:97-99. Voshell, J.R., Jr. 2002. A guide to common fr eshwater invertebrates of North America. The McDonald and Woodward Publishing Company, Blacksburg, Virginia, USA. Vymazal, J. and C. J. Richardson. 1995. Sp ecies composition, biomass, and nutrient content of periphyton in the Florida Ever glades. Journal of Phycology 31(3): 343354. Wallace, J.B., J.W. Grubauch, and M.R. Wh iles. 1996. Biotic indices and stream ecosystem processes: results from an e xperimental study. Ecological Applications 6(1): 140-151. Wharton, C.H., W.M. Kitchens, E.C. Pendl eton, and T.W. Sipe. 1982. The ecology of bottomland hardwood swamps of the Sout heast: a community profile. FWS/OBS81/37. Fish and Wildlife Service, United St ates Department of the Interior. Washington, D.C., USA. Wharton, C.H., H.T. Odum, K. Ewel, M. Duever, A. Lugo, R. Boyt, J. Bartholomew, E. DeBellevue, S. Brown, M. Brown, and L. Duever. 1977. Forested wetlands of Florida – their management and use. DSP-BCP-19-77. Center for Wetlands, University of Florida. Gainesville, Florida, USA. White, D.S. and W.U. Brigham. 1996. Aqua tic Coleoptera. Pages 399-473 in R.W. Merritt and K.W. Cummins, editors. An introduction to the aquatic insects of North America, 3rd edition. Kendall /Hunt Publishing Company, Dubuque, Iowa, USA. Whitmore, T.J. 1989. Florida diatom assemblage s as indicators of trophic state and pH. Limnology and Oceanography 34 (5): 882-895. Wienhold, C.E. and A.G. Van der Valk. 1989. The impact of duration of drainage on the seed banks of northern prairie wetlands . Canadian Journal of Botany 67: 18781884. Wilhelm, G. and D. Ladd. 1988. Natural Area Assessment in the Chicago Region. Pages 361-375 in Transactions of the 53r d North American Wildlife and Natural Resource Conference, Louisville, Kentuc ky. Wildlife Management Institute, Washington D.C., USA.

PAGE 224

209 Wilhite, L.P. and J.R. Toliver. 1990. Taxodium distichum (L.) Rich. Baldcypress. In Silvics of North America. Hardwoods. Burns, R.M. and B.H. Honkala, technical coordinators. United States Departme nt of Agriculture Handbook. Washington, D.C., USA. Williams, D.D. and B.W. Feltmate. 1992. A quatic insects. C-A-B International, Redwood Press Ltd., Melksham, United Kingdom. Winter, J.G. and H.C. Duthie. 2000. Epilith ic diatoms as indicators of stream total N and total P concentration. Journal of the North American Benthological Society 19 (1): 32-49. Wunderlin, R.P. 1998. Guide to the Vascular Plants of Florida. University Press of Florida, Gainesville, Florida, USA. Wunderlin, R. P., and B. F. Hansen. 2003. Atla s of Florida Vascular Plants. S. M. Landry and K. N. Campbell, applicati on development. Florida Center for Community Design and Research. Institute for Systematic Botany, University of South Florida. Found at: http://www.plantatlas.usf.edu/ . Tampa, Florida, USA. Young, P.J., B.D. Keeland, and R.R. Sharitz. 1995. Growth Response of baldcypress [ Taxodium distichum (L.) Rich.] to an altered hydrologic regime. American Midland Naturalist 133(2): 206-212. Zar, J.H. 1999. Biostatistical analysis, 4t h ed. Prentice Hall. Englewood Cliffs, New Jersey, USA. Zimmerman, G.M., H. Goetz, and P.W. Mielke , Jr. 1985. Use of an improved statistical method for group comparisons to study effect s of prairie fire. Ecology 66(2): 606611.

PAGE 225

210 BIOGRAPHICAL SKETCH Kelly Ann Chinners Reiss was born March, 12, 1976, in West Palm Beach, Florida. She grew up in the boat, on the beach, or ot herwise outdoors. Her parents recount that Kelly and her sister Casey learned to swim be fore they could walk. Kelly spent weeks at a time camping, canoeing, kayaking, and whitewa ter rafting with her extended family, which sparked her intrigue in the natural world around her. She attended Palm Beach Lakes High School, lettered in 3 varsity spor ts (volleyball, soccer, and softball), and graduated as valedictorian in 1994. Kelly completed her Bachelors of Scie nce degree through the School of Forest Resources and Conservation at the University of Florida in 1998, gra duating with honors. During her undergraduate year s, Kelly was active in the Environmental Action Groups (serving as Chair of the Endangered Species Committee) and University Habitat for Humanity (serving as Co-President). She spent one formative summer in Asheville, North Carolina, working at ReCreation Experiences and honing her home repair, leadership, and rock climbing/belaying skills. In May of 1998 she began studying reclaimed wetlands in the Cent ral Florida Phosphate District . She completed her Masters of Science degree in systems ecology th rough the Department of Environmental Engineering Sciences, College of Engineeri ng, University of Florida, in 2000. More recently, Kelly has studied forested wetlands throughout Florida, where she glimpsed her first Florida black bear.