PAGE 1

The Florida Wetland Condition Index (FWCI): Developing Biological Indicat ors for Isolated Depressional Forested Wetlands Report Submitted to the Florida Department of Environmental Protection Under Contract #WM-683 By Kelly Chinners Reiss and Mark T. Brown June 2005 Howard T. Odum Center for Wetlands University of Florida Gainesville, Florida 32611-6350

PAGE 2

ACKNOWLEDGMENTS This research on biological indicators was supported by a grant to Mark T. Brown, principle investigator, from the Florida Department of Environmental Protection (FDEP). The FDEP staff provided support for this research, particularly Russ Frydenborg, Ellen McCarron, Ashley ONeal, Julie Espy, Tom Frick, Joy Jackson, Liz Miller, Johnny Richardson, and Lori Wolfe. FDEP staff served as reviewers for the draft report, including Julie Espy, Denise Miller, Vicki Tauxe, Nia Wellendorf, and David Whiting. Additionally, acknowledgement is due to the systems ecology research group at the H.T. Odum Center for Wetlands for stimulating discussion and valuable input along the way. In particular, Chuck Lane (who worked on biological indicators for Florida isolated marshes) provided groundwork for this analysis. Assistance in field-data collection, laboratory analysis, data entry, and/or feedback 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 Vivas, and Melissa Yonteck, was particularly valuable. This project and the preparation of this report were funded in part by a Section 319 Nonpoint Source Management grant from the U.S. Environmental Protection Agency through a contract with the Florida Department of Environmental Protection. ii

PAGE 3

TABLE OF CONTENTS Page ACKNOWLEDGEMENTS................................................................................................ii LIST OF TABLES.............................................................................................................v i LIST OF FIGURES...........................................................................................................i x EXECUTIVE SUMMARY...............................................................................................x i 1 INTRODUCTION AND OVERVIEW.........................................................................1 Review of Isolated Depressional Forested Wetlands...............................................1 Changes in Hydrology.................................................................................4 Increased Inflows of Nutrients and/or Toxins.............................................5 Physical Disturbance....................................................................................6 Biological Indicators of Ecosystem Integrity..........................................................7 Diatoms as Biological Indicators.................................................................9 Macrophytes as Biological Indicators........................................................10 Macroinvertebrates as Biological Indicators.............................................12 Quantifying Anthropogenic Influence...................................................................15 Landscape Development Intensity Index...................................................15 Wetland Rapid Assessment Procedure......................................................16 Minnesota Disturbance Index....................................................................16 Project Overview...................................................................................................17 2 METHODS..................................................................................................................18 Site Selection.........................................................................................................18 Gradients of Anthropogenic Activity.....................................................................22 Field-data Collection..............................................................................................23 Water Samples...........................................................................................23 Soil Samples...............................................................................................24 Sampling Design........................................................................................25 Data Analysis.........................................................................................................30 Water and Soil Parameters.........................................................................30 Summary Statistics.....................................................................................30 Regional Compositional Analysis..............................................................32 Community Composition...........................................................................32 Metric Development..............................................................................................33 Indicator Species Analysis.........................................................................34 Diatom Metrics..........................................................................................34 iii

PAGE 4

Macrophyte Metrics...................................................................................35 Macroinvertebrate Metrics.........................................................................36 Florida Wetland Condition Index..........................................................................37 3 RESULTS....................................................................................................................38 Gradients of Anthropogenic Activity ....................................................................38 Water and Soil Parameters.....................................................................................38 Diatom Assemblage...............................................................................................42 Summary Statistics.....................................................................................42 Compositional Analysis.............................................................................42 Community Composition...........................................................................43 Metric Selection.........................................................................................46 Florida Wetland Condition Index Diatoms.............................................52 Cluster Analysis.........................................................................................55 Macrophyte Assemblage........................................................................................57 Summary Statistics.....................................................................................57 Compositional Analysis.............................................................................58 Community Composition...........................................................................60 Metric Selection.........................................................................................62 Macrophyte Florida Wetland Condition Index .........................................80 Cluster Analysis.........................................................................................84 Macroinvertebrate Assemblage.............................................................................86 Summary Statistics.....................................................................................86 Compositional Analysis.............................................................................87 Community Composition...........................................................................89 Metric Selection.........................................................................................89 Macroinvertebrate Florida Wetland Condition Index................................99 Cluster Analysis............................................................................................. Florida Wetland Condition Index........................................................................103 4 DISCUSSION............................................................................................................111 Describing Biological Integrity............................................................................112 Merits of a Multi-Metric Florida Wetland Condition Index................................113 Regionalization of the Florida Wetland Condition Index....................................114 Florida Wetland Condition Index Independent of Wetland Type.......................115 Landscape Development Intensity Index.............................................................116 Agricultural and Urban Wetlands........................................................................117 Richness, Evenness, and Diversity......................................................................118 Limitations and Further Research........................................................................118 Conclusions..........................................................................................................119 APPENDIX A STANDARD OPERATING PROCEDURES.....................................................120 iv

PAGE 5

B COEFFICIENT OF CONSERVATISM SCORES..............................................131 C CANDIDATE METRICS....................................................................................138 D QUANTIFYING ANTHROPOGENIC INFLUENCE........................................147 E SUMMARY STATISTICS..................................................................................149 F METRIC SCORING FOR THE FLORIDA FORESTED WETLAND CONDITION INDEX..........................................................................................154 REFERENCES................................................................................................................158 v

PAGE 6

LIST OF TABLES Table Page 2-1 Surrounding land use, land ownership, and sample date for 118 study wetlands in Florida.................................................................................................20 2-2 Field-data collected at 118 sample wetlands.........................................................22 2-3 Landscape Development Intensity coefficients (LDI i ) used in the calculation of the Landscape Development Intensity index.....................................................24 3-1 Comparison among three measures of anthropogenic influence, including the Landscape Development Intensity (LDI) index, the Wetland Rapid Assessment Procedure (WRAP), and the Minnesota disturbance index (DI).......39 3-2 Water and soil parameters among three a priori land use categories....................40 3-3 Water and soil parameters among LDI groups......................................................41 3-4 Diatom richness, evenness, and diversity among a priori land use categories......43 3-5 Mean diatom summary statistics among LDI groups............................................43 3-6 Similarity of diatom community composition using MRPP..................................44 3-7 Pearson correlations between environmental parameters and NMS ordination axes based on diatom community composition.............................................................46 3-8 Spearmans correlations for seven diatom metrics and LDI...................................47 3-9 Comparisons among diatom metrics and the diatom FWCI for LDI groups......... 47 3-10 Diatom Tolerant Indicator Species........................................................................48 3-11 Diatom Sensitive Indicator Species.......................................................................49 3-12 Mean diatom FWCI and LDI values for wetland clusters based on diatom community composition.........................................................................................56 3-13 Mean macrophyte richness, evenness, and diversity among a priori land use categories...............................................................................................................58 3-14 Mean macrophyte richness, evenness, and diversity between LDI groups...........58 3-15 Macrophyte community composition similarity among wetland regions..............59 3-16 Pearson correlations between environmental variables and NMDS axes based on macrophyte community composition at 118 wetlands...........................62 3-17 Pearson correlations between environmental variables and NMDS axes based on macrophyte community composition at 75 wetlands.............................64 3-18 Spearman correlations between six macrophyte metrics and LDI.........................64 3-19 Comparison among six macrophyte metrics for LDI groups.................................65 3-20 Statewide and regional macrophyte Sensitive Indicator Species...........................66 vi

PAGE 7

3-21 Statewide and regional macrophyte Sensitive Indicator Species...........................69 3-22 The percent of statewide and regional macrophyte Indicator Species metrics were significantly correlated with LDI..................................................................73 3-23 Exotic macrophyte species identified at 118 study wetlands.................................76 3-24 Macrophyte FWCI and component metrics scored statewide and regionally for study wetlands in the low LDI group...............................................................82 3-25 Macrophyte FWCI and component metrics scored statewide and regionally for study wetlands in the high LDI group..............................................................82 3-26 Spearman correlations between the macrophyte FWCI, component metrics, and LDI....................................................................................................83 3-27 Macrophyte FWCI scores and LDI values for wetland clusters based on macrophyte community composition.....................................................................85 3-28 Macroinvertebrate richness, evenness, and diversity among a priori land use categories...............................................................................................................87 3-29 Macroinvertebrate richness, evenness, and diversity for LDI groups...................87 3-30 Macroinvertebrate community composition similarity among a priori land use categories and ecoregions......................................................................................88 3-31 Pearson correlations between environmental variables and NMDS ordination axes based on macroinvertebrate community composition...........................................91 3-32 Spearman correlations between macroinvertebrate metrics and the macroinvertebrate FWCI with LDI, pH, dissolved oxygen (DO), and total phosphorus (TP).....................................................................................................91 3-33 Macroinvertebrate metric and FWCI scores between LDI groups........................92 3-34 Macroinvertebrate Tolerant Indicator Genera.......................................................93 3-35 Macroinvertebrate Sensitive Indicator Genera......................................................95 3-36 Macroinvertebrate FWCI scores and LDI values for wetland clusters based on the macroinvertebrate community composition..............................................102 3-37 FWCI scores for 118 wetlands based on three assemblages including diatoms, macrophytes, and macroinvertebrates..................................................................104 3-38 Pearson correlations among 19 metrics...............................................................110 4-1 The 19 metrics of the Florida Wetland Condition Index, including seven diatom, six macrophyte, and six macroinvertebrate metrics................................112 B-1 Coefficient of Conservatism (CC) scores for 561 macrophytes identified in isolated depressional freshwater forested wetlands in Florida............................131 C-1 Candidate metrics based on the diatom assemblage............................................138 C-2 Candidate metrics based on the macrophyte assemblage....................................140 C-3 Candidate metrics based on the macroinvertebrate assemblage..........................143 vii

PAGE 8

D-1 LDI, WRAP, and Minnesota disturbance index scores for 118 wetlands...........147 E-1 Summary statistics of richness (R), evenness (E), Shannon diversity (H), and Simpsons index (S) for the diatom assemblage (genus level)............................149 E-2 Summary statistics of richness (R), jackknife estimators of species richness (Jack1, Jack2), evenness (E), Shannon diversity (H), and Whittakers beta diversity (W) for the macrophyte assemblage (species level)...........................150 E-3 Summary statistics of richness (R), evenness (E), Shannon diversity (H), and Simpsons index (S) for the macroinvertebrate assemblage (genus level)..........153 viii

PAGE 9

LIST OF FIGURES Figure Page 1-1 Characteristic silhouette shape of three isolated depressional wetlands sampled in this study................................................................................2 2-1 Study site location of 118 isolated depressional forested wetlands in Florida................................................................................................................19 2-2 Belted transect layout for macrophyte sampling and the location of the water and soil samples...........................................................................................25 2-3 Benthic diatom samples were collected at 50 isolated depressional forested wetlands...................................................................................................26 2-4 Macroinvertebrates were sampled at 79 isolated depressional forested wetlands.................................................................................................................29 3-1 NMS ordination bi-plot of 50 wetlands in diatom species space with an overlay of environmental parameters................................................................45 3-2 Percent diatom Tolerant Indicator Species increased with increasing development intensity............................................................................................49 3-3 Percent diatom Sensitive Indicator Species decreased with increasing development intensity............................................................................................50 3-4 Pollution Tolerance Class 1 diatoms increased with increasing development intensity............................................................................................51 3-5 Nitrogen Uptake Metabolism Class 3 diatoms increased with increasing development intensity...........................................................................51 3-6 Saprobity Class 4 diatoms increased with increasing development intensity..................................................................................................................53 3-7 The pH Class 3 diatoms increased with increasing development intensity..................................................................................................................53 3-8 Dissolved Oxygen Class 1 diatoms decreased with increasing development intensity............................................................................................54 3-9 Diatom FWCI scores decrease with increasing development intensity.................54 3-10 Diatom FWCI scores for wetland clusters based on diatom community composition ...........................................................................................................55 3-11 NMDS ordination bi-plot of 118 sample wetlands in macrophyte species space with an overlay of environmental parameters..............................................61 3-12 NMDS ordination bi-plot of 75 sample wetlands in macrophyte species space with an overlay of environmental parameters..............................................63 ix

PAGE 10

3-13 Macrophyte Tolerant Indicator Species increased with increasing development intensity............................................................................................68 3-14 Macrophyte Sensitive Indicator Species decreased with increasing development intensity............................................................................................72 3-15 FQAI scores decreased with increasing development intensity............................74 3-16 Exotic Species increased with increasing development intensity..........................75 3-17 Native Perennial Species decreased with increasing development intensity.........78 3-18. The Percent Wetland Status species decreased with increasing development intensity..................................................................................................................79 3-19 Macrophyte FWCI scores decreased with increasing development intensity.......81 3-20 Regional macrophyte FWCI scores for five wetland clusters based on macrophyte community composition.....................................................................84 3-21 NMDS ordination bi-plot for 79 wetlands in macroinvertebrate genus space with an overlay of environmental parameters........................................................90 3-22 Macroinvertebrate Tolerant Indicator Genera increased with increasing development intensity............................................................................................94 3-23 Macroinvertebrate Sensitive Indicator Genera decreased with increasing development intensity............................................................................................96 3-24 Florida Index scores decreased with increasing development intensity................97 3-25 Macroinvertebrates in the phylum Mollusca increased with increasing development intensity............................................................................................98 3-26 Macroinvertebrates in the family Noteridae decreased with increasing development intensity............................................................................................99 3-27 Macroinvertebrates that belong to the scraper functional feeding group increased with increasing development intensity................................................100 3-28 Macroinvertebrate FWCI scores decreased with increasing landscape development intensity index................................................................................100 3-29. Macroinvertebrate FWCI scores for five wetland clusters based on macroinvertebrate community composition........................................................102 3-30 Three dimensional scatter plot of the FWCI based on three assemblages, including diatoms, macrophytes, and macroinvertebrates...................................108 3-31 Scatterplots of FWCI scores for wetlands based on diatom, macrophyte, and macroinvertebrate assemblages................................................109 x

PAGE 11

EXECUTIVE SUMMARY THE FLORIDA WETLAND CONDITION INDEX (FWCI): DEVELOPING BIOLOGICAL INDICATORS FOR ISOLATED DEPRESSIONAL FORESTED WETLANDS Over 30 years ago, the federal Water Pollution and Control Act obliged states to protect and restore the chemical, physical, and biological integrity of waters, and charged states with establishing water-quality standards for all waters within state boundaries including wetlands. Criteria for defining water-quality could be narrative or numeric; and it could be addressed through chemical, physical, or biological standards. Initially, states used chemical and physical 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 ). The United States Environmental Protection Agency (USEPA) recognized the potential of biological criteria 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-protection agencies over the past 2 decades ( Gerristen et al. 2000 ). Such biological criteria and monitoring programs 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 ( Mack 2001 ; USEPA 2002a ). Within Florida, biological indices have been created based on macroinvertebrate community composition for streams ( Barbour et al. 1996a ; Fore 2003 ), lakes ( Gerristen and White 1997 ), and isolated depressional freshwater herbaceous wetlands (Lane et al. 2003), and based on the community composition of the diatom and macroinvertebrate assemblages for freshwater herbaceous wetlands ( Lane et al. 2003 ). The primary objective of this research was to develop a Florida Wetland Condition Index (FWCI) for isolated depressional forested wetlands in Florida. Wetland study sites were sought in various landscape settings that included natural, agricultural, and urban land uses. Three independent measures of anthropogenic influence were calculated for each wetland including the Landscape Development Intensity index (LDI) ( Lane et al. 2003 ; Brown and Vivas 200 5), the Wetland Rapid Assessment Procedure (WRAP) ( Miller and Boyd 1999 ), and the Minnesota disturbance index ( Gernes and Helgen 1999 ). Strong correlations among LDI, WRAP, and MDI led us to adopt the LDI to represent the gradient of anthropogenic influence for further analyses. Compositional differences among the diatom, macrophyte, and macroinvertebrate assemblages were identified and related to the LDI measure of anthropogenic influence. Each assemblage (diatom, macrophyte, and macroinvertebrate) was used to construct an assemblage specific FWCI for isolated depressional forested wetlands in Florida. While previous research has identified responses of wetland ecosystems to particular system impacts (such as increased nutrients or altered hydrology), few have combined multiple biotic components, environmental parameters, and landscape development intensity in an attempt to quantify ecological integrity. The contribution of this research to our understanding of changes in the community composition of isolated xi

PAGE 12

depressional forested wetlands (based on the diatom, macrophyte, and macroinvertebrate assemblages) in relation to different development intensities in the surrounding landscape can be summarized in seven main points. Biological indicators along with physical and chemical parameters were useful in defining biological integrity; The variable turnover times and sensitivities of the three assemblages (diatoms, macrophytes, macroinvertebrates) suggest that a multi-metric multi-assemblage FWCI has more merit than a FWCI based on a single assemblage; Regionalization may strengthen the FWCI; A FWCI independent of wetland type may be feasible, given the strong likeness of the forested FWCI to the herbaceous FWCI ( Lane et al. 2003 ); The Landscape Development Intensity (LDI) index was a useful tool correlating with the measured biological condition for isolated forested wetlands; Urban wetlands exhibit a different vector of change than do agricultural wetlands, and while the FWCI suggests low biological integrity of both agricultural and urban wetlands, these wetlands do provide services and do work in the environment. Richness, evenness, and diversity (for each assemblage) were not sensitive to different land uses or development intensities in the surrounding landscape. The assemblage specific FWCIs provide a quantitative measure of the biological integrity of isolated depressional forested wetlands in Florida. The FWCI includes three separate measures of wetland biological integrity based on the community composition of three species assemblages. In total 19 metrics are used to address deviations from the wetland biological integrity potential based on the suite of reference wetlands, including seven diatom, six macrophyte, and six macroinvertebrate based metrics. Metrics were selected for inclusion in the FWCI based on the strength and significance (Spearmans correlation coefficient) of each metric in correlation with a quantitative gradient of LDI, the ability of a metric to visually distinguish correlations with LDI in scatter plots, and a significant difference between low and high LDI groups (Mann-Whitney U-test). The FWCI for each assemblage was composed of individual metrics specific to the assemblage, which were scaled and added together, creating the diatom FWCI (0-70 scale), macrophyte FWCI (0-60 scale), and the macroinvertebrate FWCI (0-60 scale), with the highest values reflecting the highest biological integrity whereas a score of 0 would reflect a lack of biological integrity or no similarity to the reference wetland condition. The seven diatom metrics included in the FWCI, which were significant for both the Spearmans correlation coefficient (|r| > 0.45, p < 0.01) and the Mann-Whitney U-test between LDI groups (p < 0.10), represented tolerance and autecological metrics. Tolerance metrics included tolerant and sensitive indicator species. The five autecological 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 saturation between 10-25% and a biological oxygen demand of approximately 13-22 mg/L), pH Class 3 (circumneutral, mainly occurring at pH values around 7), and Dissolved Oxygen Class 1 (requiring continuously high dissolved oxygen concentrations xii

PAGE 13

near 100%). Pollution Class was established by Bahls ( 1993 ), and Nitrogen Metabolism, Saprobity, pH, and Dissolved Oxygen Classes were defined by van Dam et al. ( 1994 ). Six macrophyte metrics that were significant for both the Spearmans correlation coefficient (|r|>0.39, p<0.001) and the Mann-Whitney U-test between LDI groups (p<0.001) included in the FWCI were Tolerant Indicator Species, Sensitive Indicator species, Floristic Quality Assessment Index (FQAI), Exotic Species, Native Perennial Species, and Wetland Status Species. FQAI scores were obtained for species identified specific to this study. The six macroinvertebrate metrics included in the FWCI, which were significant for both the Spearmans correlation coefficient (|r| > 0.3, p < 0.05) and the Mann-Whitney U-test between LDI groups (p < 0.05), represented tolerance, community balance, and functional group metrics. Tolerance metrics included Tolerant Indicator Genera, Sensitive Indicator Genera, and Florida Index scores. Community balance metrics included Mollusca (phylum taxonomic level) and Noteridae (family taxonomic level). One functional groups metric was included, the percent of macroinvertebrates identified as scrapers. Diatom FWCI Metrics Macrophyte FWCI Metrics Macroinvertebrate FWCI Metrics % Tolerant species % Tolerant species % Tolerant species % Sensitive species % Sensitive species % Sensitive species % in Pollution Class 1 Modified FQI Score FL Index Score % in Nitrogen Class 3 % Exotic % Mollusca % in Saprobity Class 4 % Native Perennial % Noterida % in pH Class 3 % Wetland Status % Scraper % in Oxygen Class 1 The variable turnover times and sensitivities of the three different assemblages (diatoms, macrophytes, macroinvertebrates) suggest that use of multiple assemblage specific FWCIs has more sagacity than a FWCI based on a single assemblage. The three species assemblage specific FWCIs can be used to infer influences in temporal and spatial changes to which a particular wetland has been exposed. While agreement in the ranking of the biological condition of study wetlands using the three species assemblage FWCIs was anticipated, discrepancies among the ranking from the different assemblages may provide great insight into wetland condition as different species assemblages respond to changes in driving energies over different time scales. 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. Ranges of FWCI scores for the a priori wetland categories (based on land use in the surrounding landscape) show the difficulties inherent in attempting to discern a specific mode of impact based on FWCI scores. Though the FWCI suggested low biological integrity of both agricultural and urban wetlands, these wetlands provide services and do work in the environment. Therefore, a quantitative score of biological integrity established through the FWCI should not be used as a surrogate for wetland value, but rather as an objective, quantitative means of comparing changes in community composition along gradients of xiii

PAGE 14

landscape development intensity. In the future, an integrative multi-metric FWCI could be constructed for wetlands throughout the state, with lists of Indicator Species and metric scores dependent on Florida ecoregions and specific to wetland type. While the FWCI 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 community 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-nitrogen, water TKN, water TP, soil moisture, soil organic matter, and soil TP). When soil and water parameters were used to explain variation in the community composition for each assemblage based on Non-metric Multidimensional Scaling (NMDS) ordination techniques, water column pH was universally identified as a keystone parameter significantly correlating the community composition ordination. Additionally, total phosphorus concentrations correlated with the ordinations based on both the diatom and macrophyte assemblages. Perhaps preservation and restoration strategies could focus on limiting activities (in the surrounding landscape) that influence changes to water column pH and total phosphorus loading to wetlands in order to promote biological integrity. xiv

PAGE 15

CHAPTER 1 INTRODUCTION AND OVERVIEW With even casual observation, it is apparent that ecosystems change with increasing levels of human development, and that the extent of change is observably related to the magnitude of human activity. While previous research has identified the responses of forested wetlands to human induced changes 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 effects of anthropogenic activities. This is particularly true for wetlands located within urban settings. Our study focused on understanding the effects of landscape development from anthropogenic activities on isolated depressional forested wetlands in Florida. More specifically, our goal was to define the condition of wetlands surrounded by different land use activities by describing the biological integrity of the ecosystem. Determining wetland ecosystem integrity through the use of biological indicators requires an accepted definition of integrity. Karr and Dudley ( 1981 ) defined integrity as the ability of an aquatic ecosystem to support and maintain a balanced, integrated, adaptive community of organisms having a species composition, diversity, and functional organization comparable to that of the natural habitats of the region. This definition requires two things: a definition of the natural habitat (or reference standard condition) and appropriate regionalization. Reference standard wetlands have been defined as those wetlands surrounded by natural landscapes, with no apparent anthropogenic alterations. Ecosystem condition has been commonly defined by data collected at a suite of reference wetlands, which include a suite of reference standard wetlands as well as a suite of impaired wetland representing the full range of impaired conditions from minimal to severe impairment. Gerristen et al. ( 2000 ) concur that biological assessment relies on a characterization of the reference condition. Review of Isolated Depressional Forested Wetlands Throughout the world, wetlands have been categorized in many different ways (Mitsch and Gosselink 1993; Keddy 2000; Kent 2000). Probably the most widely recognized classification system in North America is that by Cowardin et al. (1979). Our study focuses on what Cowardin et al. (1979) categorized forested palustrine wetlands. In more general terms, the wetlands targeted in our study have been called pondcypress domes with reference to Vernon (1947), who was the first to name these systems after their characteristic silhouette in the landscape (Figure 1-1). Pondcypress domes range in size from less than 1 hectare to more than 10 hectares (Wharton et al. 1977), and host Taxodium ascendens (pondcypress) as the principal tree species (Devall 1998). Other tree species associated with Taxodium ascendens include: Nyssa sylvatica var. biflora tupelo (swamp tupelo), Pinus spp. (many southern pines), Acer rubrum (red maple), and Magnolia virginiana var. australis (sweetbay magnolia) (Wilhite and Toliver 1990; Devall 1998). Pondcypress trees characteristically dominate the center, with pondcypress 1

PAGE 16

2 (A) (B) (C) Figure 1-1. Characteristic silhouette shape of three isolated depressional wetlands sampled in this study. These pondcypress domes named by Vernon (1947) include: (A) CU2 in Orange County, (B) CR4 in Sumter County, and (C) PR2 in Santa Rosa County.

PAGE 17

3 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 number of seedlings in the drier edges (Odum 1978). While the target ecosystem type was the pondcypress dome, many of the study wetlands hosted co-dominant species or were dominated by canopy species other than pondcypress and may more appropriately be described as isolated depress-in from a relatively small watershl 1998). In drained pondcypress domes in northern Florida, Marois and Ewel (1983) found an increase in the densities of hardwood and shrub species. lorida, fire suppression has resulted in a shift in the canopy ommunity composition to readily regenerating hardwood species over pondcypress dominae time during the year necessary to prevent germination of more terrestrial faster-growing pines and hardwoods that are not tolerant ional mixed-forested wetlands. The principal commonalities among the study wetlands include the isolated depressional nature of these wetlands and the occurrence of mature forest canopy. The use of the word isolated to describe a wetland has been scrutinized because it has been interpreted to suggest separation of a wetland system from its landscape. However, the term isolated in this study simply refers to the lack of a regular or permanent surface water connection between the wetland and the surrounding landscape. However, in periods of high water and excessive rainfall, many of these isolated wetland systems receive surface run-in from nearby wetlands or discharge to other nearby wetlands or surface water features. Under typical conditions, inflows into isolated depressional wetlands are limited to sunlight, wind, water (rain, surface run-off, groundwater), and recruitment of plant and animal species. Water inflow comes almost entirely from rainwater, both as a direct input and as run ed. As such these wetlands are often termed isolated due to their somewhat limited hydrologic connections. Standing water is present in most isolated depressional forested wetlands much of the year (Odum 1978; Mitsch and Gosselink 1993). Some have deep central pools staying wet year round, while others go dry annually. There is also variation in maximum flooding depth and length of standing water between years, which reflects the larger scale climatic and physiogeographic influences to these wetland systems. Typically, the wettest period is summer and the driest spring and fall (Mitsch and Gosselink 1993). Isolated depressional forested wetlands dominated with pondcypress are characterized by a tolerance of low nutrient levels and intermittent fire (Brandt and Ewel 1989), with major system inputs 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 (Deval Furthermore, in south F c nce (Ewel 1990). Perhaps many of the depressional mixed-forested wetlands identified throughout the state were historically dominated by pondcypress, but have undergone a shift in canopy species composition due to drainage, fire suppression, timber harvesting, and other indeterminate anthropogenic activities. Pondcypress and the most characteristic co-dominate swamp tupelo are deciduous, shedding their leaves from October to December. Ewel (1984) and Wharton et al. (1977) found that the perpetuation of pondcypress domes depends on fluctuating water levels, with a dry period without standing water necessary for pondcypress generation and higher water levels som

PAGE 18

4 of stanmechanisms within the developed landscape leading to potentie impeded which is otherwise necessaloodplain tree species to sustain ding water. Altering the typical hydroperiod of a pondcypress dome has a direct effect on species composition, resulting in encroachment of terrestrial species in drained pondcypress domes and a lack of regeneration of pondcypress in artificially flooded pondcypress domes. Anthropogenic activities in the surrounding up-slope landscape can create a wide array of changes to the inflows of these wetland systems. Some of the potential alterations to isolated depressional wetlands located within developed landscapes include changes in the seasonality and depth of flooding, increased nutrient inputs, increased toxin inputs, and physical impacts (canals, retaining walls, stormwater box culverts, etc.). Changes in Hydrology There are two important al hydrologic alterations to an isolated depressional wetland ecosystem surrounded by developed land uses. First, increased run-off is considered a factor of the amount of increased impervious surface in the watershed supporting the wetland system and the amount of rainfall. This would be particularly apparent in an urban landscape, where previously vegetated lands are paved creating increased water flow during rain events, which might otherwise have been intercepted by the vegetation or permeated into the soil. A second mechanism of the developed landscape is an increased outflow of water storage from the wetland as a result of drainage. In Florida, mature cypress trees (Taxodium spp.) are considered 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), though the tress suffered decreased growth rates and no evidence of regeneration. This suggests that while mature cypress ecosystems may be able to withstand some threshold level of long-term flooding, regeneration may b ry to ensure the long-term continued existence of the ecosystem. Ultimately, removing the structure of the wetland will predictably alter other ecosystem components. For example, by removing the tree canopy, algae may flourish at the water surface where it would have otherwise been shaded out. This would in term alter the 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 f ed increases in water depth after the damming of the Ocklawaha River in Florida. They found that while the larger trees survived some depths of flooding, in the deepest areas, where mean water depth was 1 m, there was 100% tree mortality within 5 years of flooding. Where as, in the control system, tree mortality was less than 1% per year. Additionally, flooded trees responded with a dieback of terminal 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 increased. 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

PAGE 19

5 growth increases: decreased competition due to the death of less flood tolerant species, or increased nutrient levels immediately following flooding. Marois and Ewel (1983) studied the effects of ditches and berms on 15 pondcypress domes situated within an intensively managed slash pine plantation. In the pondcypress domes not ditched and bermed, the lengths of flooding and mean water depth were generally greater. Alterna tively in the drier ditched and bermed pondcypress domes tion by blocking sunlight from reaching the forest floor. Marois and Ewel (1983) found the highest percentage of light The pondcypress domes with greater ght also had an abundance of grasses and sedges. Ewel (1990) noted that drainage allows ious surface run-off) sources. An increase in run-in from imperv the density of hardwoods, shrubs, and vines increased. They concluded that while pondcypress tree growth increased in the years directly following the drying of the pondcypress domes, pondcypress regeneration may have been inhibited due to changes 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 pondcypress dome may inhibit or seriously diminish potential germination. The higher density of hardwoods, shrubs, and vines may also inhibit pondcypress regenera transmittance in unaltered pondcypress domes. li species with low flood tolerance to become established, resulting in an increased density of shrubs and hardwoods, poor pondcypress regeneration, increased fire potential, and a dramatic shift to arboreal wildlife species from aquatic 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, dahoon holly; Lyonia lucida, fetterbush; Magnolia virginiana var. autralis, sweetbay magnolia; and Persea palustris, swamp bay), became more common in swamps when water levels were lowered. Harris and Vickers (1984) speculated that shifting species in the vegetation layer equates to altered 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 pondcypress domes for regeneration. Benthic invertebrates may not withstand increased dry periods, and reproduction may be difficult or unattainable. Often forming the base of swamp food chains (Ewel 1990), eliminating or decreasing the population of benthic macroinvertebrates could have repercussions throughout the food chain. Fish, amphibians, and reptiles may be eliminated from pondcypress domes with decreased hydroperiods due to reproduction difficulties or altered food availability (Means et al. 1998; Ewel 1990). Increased Inflows of Nutrients and/or Toxins In undeveloped landscapes isolated depressional wetlands receive limited nutrient inputs from rainwater and surface water run-in (Wharton et al. 1977). An increase in the inflow of nutrients and toxins may come from both point (eg. wastewater additions) and non-point (ex. imperv ious surfaces in the surrounding landscape may increase the loading of nutrients and toxins to a wetland system (Harper 1994). Surface run-off carrying fertilizer used on agricultural crops or home lawns are examples of non-point source contributions. As nutrients flow into the wetland, the growth of living biomass increases, and nutrients

PAGE 20

6 accumulate in the water and soil organic matter storages. Conversely, as toxins flow into and accumulate in the wetland, there is a deleterious effect on biomass. Two nutrients of primary importance in pondcypress domes are phosphorus and nitrogen. Phosphorus, an element critical to plant growth, is mostly bound into forms unavailable to plants at pH levels below 5.7 (Brady and Weil 2004), higher than the average pH of pondcypress domes embedded in undeveloped landscapes (Coultas and Duevers needles at a pondcypress dome embedded in silvicultural land use and a cypress strand receiving sewage for more than 40 years. he silvicultural wetland had a lower average phosphorus oncentration than cypress needles in the wetland receiving wastewater. Additionally, the toprstory vegetation (from increased leaf area or coverinxamples of physical changes in depressional forested wetlands include trampling and grazing by dome barriers such roads 1984). Dierberg and Brezonik (1984) found an annual average pH of 4.63 for pondcypress wetland with the primary source of water being rainfall, and an annual average pH of 5.4 for a pondcypress wetland with the primary source of water being shallow groundwater for wetlands in north-central Florida. Phosphorus is known to accumulate in the clay layers found beneath pondcypress domes, which makes pondcypress dome ecosystems dependent on a constant input of available phosphorus from rainfall. Raising the pH of a wetland increases the concentration of available phosphorus. In contrast, nitrogen does not accumulate in the clay layer or organic sediments at the bottom of pondcypress domes due to denitrification processes, and the rate of the nitrogen cycle seems dependent on the rate of decomposition of organic matter (Wharton et al. 1977). Pondcypress trees respond to increased nutrient loading with increased tree growth rates ( Nessel et al.1982 ; Lemlich and Ewel 1984 ). Nessel et al. ( 1982 ) measured phosphorus concentrations in live cypres Cypress needles in t c 20 cm of sediment in the wetland receiving wastewater had nearly 5.5 times as much phosphorus per square meter as in the silvicultural wetland. They concluded that the trees in the cypress strand were in fact responding to the increased nutrient inputs (from the sewage) with increased growth rates. Traditionally pondcypress domes are 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. (duckweed), Spirodela spp. (giant duckweed), and/or Azolla caroliniana (Carolina mosquito fern) on the water surface. Ewel ( 1984 ) also noted that a nutrient enriched pondcypress dome had similar understory species composition compared to a pondcypress dome not receiving wastewater, however the leaf area was significantly higher in the nutrient enriched pondcypress dome. Changes in the unde g of the water surface with a layer of vegetation) can have an effect on other trophic levels within the wetland ecosystem. In pondcypress domes receiving wastewater additions, Harris and Vickers ( 1984 ) reported an increase in the number 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 pondcypress dome were different from the control wetland. Physical Disturbance E stic cattle, rooting of feral pigs, construction of

PAGE 21

7 and retention walls, and conduits such as sere found ric; and it could be addent undetected. This same wunt, persistence, and exposure of chemical tormwater culverts. Few publications w that quantified the effects of physical modifications to wetlands. Findlay and Houlahan ( 1997 ) used existing biological surveys for 30 Ontario, Canada, wetlands, comparing species richness of plants, amphibians, birds, and reptiles with wetland areas, density of paved roads, 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 wetland. They concluded that increasing the density of paved road surface or decreasing the forest cover by 20% within 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 modification to isolated depressional forested wetlands is the removal of a portion or all of the canopy layer. Florida has a long history of timber harvesting, and Ewel ( 1990 ) suggested that nearly all of the pondcypress domes in north Florida have been logged since the late 1800s. Studies show that logged pondcypress domes maintain their defining characteristics after regeneration ( Terwilliger and Ewel 1986 ; Ewel et al. 1989 ); however, during regeneration, there are shifts in the flora and fauna of logged wetlands. Physical modifications such as roads, canals, and stormwater culverts also act as direct conduits for the introduction of exotic species ( Frappier and Eckert 2003 ). Biological Indicators of Ecosystem Integrity Over 30 years ago, the Water Pollution and Control Act (later referred to as the Clean Water Act, 1972) required states to restore and maintain the chemical, physical, and biological integrity of the Nations waters ( USEPA 1990 ). This legislation included establishing water-quality standards for all waters within state boundaries, including wetlands. Criteria for defining water-quality could be narrative or nume ressed through chemical, physical, or biological standards. Initially, states used chemical and physical 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 ). There are several shortcomings in deriving ecosystem integrity based on exceeding established limits for chemical and physical parameters. Such criteria have been considered rudimentary in their ability to reflect more than the temporal concentration of chemicals 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 w ater body could have undesirable levels of other nontarget 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 occur during specific loading events and may therefore incompletely describe the biological and ecological condition of the system. Adams ( 2002 ) points out other environmental factors (such as sedimentation, alterations to habitat, varying temperature and oxygen levels, and changes in ecological aspects like food availability and predator-prey relationships) are not reflected with chemical criteria alone. James and Kleinow ( 1994 ) note that different organisms respond in different ways to the amo

PAGE 22

8 compougical criteria to assess water-quality standards and in the late 1980s requireed widely in ecological research. One trend in biological monitoring has led to the development of dices of biological integrity (often referred to as IBIs), for different species ssemblages including diatoms (Fore and Grafe 2002); macrophytes (Galatowitsch et al. Mack 2001; Lane 2003); macroinvertebrates (Kerans and Karr 1994; Barbour et al. 1996b); fish (Schulz et al. 1999); and birds (OConnell et al. 199 nds otherwise foreign to an organism, and single-valued chemical and physical criteria of water quality may overlook important biological implications. Alternatively, biological indicators integrate the spatial and temporal effects of the environment on resident organisms and are suitable 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 reflect chemical exposure and also integrate changes in the community composition of the ecosystem (from physical, chemical, and biological changes) ( Adams 2002 ). The United States Environmental Protection Agency (USEPA) recognized the potential of biolo d 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-protection agencies over the past 2 decades ( Gerristen et al. 2000 ). Biological criteria and monitoring 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 ). Biological monitoring to assess ecosystem condition has been appli in a 1999a; Gernes and Helgen 1999; 8 ). Such indices have been applied to ecosystems throughout the world including in Europe ( Kelly and Whitton 1998 ); widely throughout the United States ( Karr 1981 ; Lenat 1993 ; Fore and Grafe 2002 ; Lane et al. 2003 ); and is beginning in Australia by J.E. Ling of the Royal Botanical Gardens, University of Western Sydney. The primary aim of biological monitoring is to detect changes in abundance, structure, and diversity of target species assemblages. Danielson ( 1998a ) notes that biological signals are effective mainly because biological monitoring incorporates changes from various collective constant or pulsing sources. Many studies have created multi-metric indices of biological condition, incorporating individual metrics into a quantitative value of community condition or ecosystem integrity. Karr and Chu ( 1997 ) defined metrics as biological attributes that have a consistent and predictable response to anthropogenic activities. Anthropogenic activities can alter the integrity of wetland ecosystems by causing one or more of the following conditions: eutrophication, contaminant toxicity, acidification, salinization, sedimentation, burial, thermal alteration, vegetation removal, turbidity, shading, dehydration or inundation, and/or habitat fragmentation ( Danielson 1998a ).

PAGE 23

9 Diatoms as Biological Indicators Diatoms are unicellular or colonial algae 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: provide a food source for organisms at higher trophic levels, contribute to n utrient and biogeochemical cycling, oxygene high diversity within the algal community (particularly of diatoentify based on well-established taxonomic keys of their decay resistanation can complicate assessments based on scales of anthrop ate the water column, regulate water chemistry, create habitat for other organisms, and act as physical barriers to erosion. Because of their rapid turnover times, algae have a short response time to perturbations including nutrient and toxic contaminant inputs; and algae continue production throughout the winter, taking advantage of available nutrients when higher plants are dormant ( Cronk and Fennessy 2001 ). While the standing stock of algae is typically lower than that of the macrophyte assemblage, algae can constitute a higher proportion of primary productivity within 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 advantages of using algae for biological assessment include th m species) in aquatic environments ( Stevenson 2001 ). There is a depth of knowledge as to the sensitivity of many species to different environmental conditions based on their autecological characteristics, including two published tables of autecological relationships by Bahls ( 1993 ) and van Dam et al. ( 1994 ). Additionally, the rapid-response time of the algal community to changing 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 considered easy to id t siliceous structures ( Stevenson et al. 1999 ), and there are well-tested protocols for sampling aquatic habitats ( Goldsborough 2001 ). Few significant disadvantages of using algae in biological assessment methodologies have been described. Among them is the necessity of a high-powered microscope for identification ( Doherty et al. 2000 ), though identification is relatively easy, and good taxonomic keys have been established ( Stevenson et al. 1999 ). However, taxonomic identification to lower taxonomic levels such as genus or species should be reserved for an expert taxonomist. Additionally, while most algae are not readily motile, wind and current transloc ogenic activity in the surrounding 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, diatoms are noted as a useful assemblage ( Stevenson 2001 ; Doherty et al. 2000 ). Previous research has correlated the response of diatoms in streams, lakes, and wetlands, to changes in surrounding land use and to changes in water-column characteristics including nutrient loading ( van Dam et al. 1994 ); pH ( Pan and Stevenson 1996 ); heavy metal loading ( Charles et al. 1996 ); and saprobity levels ( Lange-Bertalot 1979 ). The USEPA ( 2002b ) reported that diatoms are one of the most

PAGE 24

10 commonly used assemblages in aquatic ecosystems for assessing biological, physical, and chemical conditions. Research correlating changes in the diatom 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 biological 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 et al ( 2003 ) incorporated 14 metrics into the Diatom Index of Wetland Condition (DIWC). These included tolerant indicator species, sensitive indicator species, diatoms requiring low pH, diatoms requiring low salinity, diatoms tolerant of high salinity, diatoms tolerant of hi gh H, diatoms sensitive to high nitrogen, diatoms tolerant of high nitrogen, diatoms iatoms tolerant of low dissolved oxygen, mesond polysaprobous diatoms, diatoms characteristic of oligotrophic environments, diatoms charactrelated the response of diatom community composition to different land use and water p p requiring elevated dissolved oxygen, d a eristic of eutrophic environments, and pollution-tolerant diatoms. Environmental parameters correlating with diatom community composition included specific conductivity, water-column pH, water ammonia-nitrogen concentration, water total Kjeldahl nitrogen (TKN) concentration, water total phosphorus concentration (TP), soil pH, and soil TP ( Lane 2003 ). In a study of lotic (relating to moving water) systems in the Mid-Atlantic States, Pan et al. ( 1996 ) found the strongest correlation with diatom community composition and changes in water-column pH. Additional water-column parameters correlating with diatom community composition included turbidity, aluminum concentration, chlorine concentration, TP, total suspended solids, and dissolved organic-carbon concentration. Similarly, in a study of emergent permanently flooded floodplain wetlands in western Kentucky, Pan and Stevenson ( 1996 ) found significant correlations between diatom community composition and eight water variables, including alkalinity, conductivity, ammonia-nitrogen concentration, pH, silicon concentration, nitrate-nitrogen concentration, chlorine concentration, and TP. Another study of streams in Michigan also cor hysical and chemical parameters ( Stewart et al. 1999 ). These findings reiterate that usefulness of the algal assemblage in reflecting changes in the water environment. Macrophytes as Biological Indicators Wetland macrophytes are defined as aquatic emergent, submergent, or floating plants growing in or near water ( USEPA 1998 ) and are described as distinguishing landscape features. The spatial 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 composition of wetland macrophytes typifies the physical, chemical, and biological wetland dynamics in time and space. Macrophytes

PAGE 25

11 play a vital role in supporting the structure and function of wetlands by providing food and habitat for other assemblages including algae, macroinvertebrates, fish, amphibi ans, reptiles; ease of identification, to at least some uun using macropin wetlands, 12 for low-order river floodplain wetlands, eight for mid-order river floodplain wetlands, seven for calcareous littoral wetlands, six for noncalcareous littoral ws, four for forest glacial marshes, and a single metric for prairie glacial marshes. Another comprehensive biological assessment used to birds, and mammals. As such, macrophyte populations can be used as diagnostic tools to assess other aspects of the wetland ecosystem. 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 overgrowth of particular macrophytes may signify increased nutrient loading ( USEPA 1998 ). Many advantages of studying macrophytes as indicators of wetland condition have been noted, including their large, obvious size seful taxonomic level; known response to toxicity tests; and general lack of ability to move to avoid unfavorable conditions ( Danielson 1998a ; Cronk and Fennessy 2001 ). Additionally, macrophytes readily respond to changes in nutrient, light, toxic contaminant, metal, herbicide, turbidity, water, and salinity levels. They can also be sampled in the field with transects, or from the office with aerial photography, and well-established field methods of sampling macrophytes exist ( USEPA 2003 ). Furthermore, the USEPA ( 2003 ) states that macrophytes do not require laboratory analysis, can easily be used for calculating simple abundance metrics, and are superb integrators of environmental condition. In general, macrophytes 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 noted shortcomings of using macrophytes as biological indicators. These include the potential delay in response time for perennial vegetation, difficulty identifying taxa to the species level in certain seasons and for some genera, different herbivory patterns, and varied pest-management practices ( Cronk and Fennessy 2001 ). Despite these limitations, macrophytes have provided strong signals of anthropogenic influence ( USEPA 2003 ). In fact, many states have beg hytes in their wetland biological assessment programs, including 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 wetland type and bioregion. Lane et al. ( 2003 ) calculated five metrics based on the macrophyte community composition for inclusion in the marsh Vegetative Index of Wetland Condition (VIWC). The five core metrics of the VIWC included tolerant indicator species, sensitive indicator species, exotic species, annual to perennial ratio, and average coefficient of conservatism score ( Lane 2003 ). In Minnesota, Vegetative Indices of Biotic Integrity (V-IBIs) have been created for eight wetland types ( Galatowitsch et al. 1999a ), and included 15 metrics for high-order river floodpla wetlands, seven for wet prairie-sedge meado construct multi-metric indices of biotic integrity for Ohio wetlands was designed by Mack et al. ( 2000 ). Separate biological multi-metric indices were developed for

PAGE 26

12 emergent, forested, and shrub wetlands. Twelve metrics were incorporated, including Carex species, dicot species, shrub species, hydrophyte species, Rosaceae species, Floristic Quality Assessment Index (FQAI), tolerant species, intolerant species, invasive graminoids, shrub density, small-tree density, and maximum importance value. The FQAI has been included in many of the multi-metric indices created for the macrophyte assemblage. The concept of FQAI was developed by Wilhelm and Ladd ( 1988 ) for vegetation around Chicago, Illinois. This method of scoring plant species based ortebrates as Biological Indicators ing waters), though less is known about lorth Dakota n expert 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 et al. 2003 ; Cohen et al. 2004 ). The FQAI provides a quantitative means of assessing the fidelity of a plant to a particular environment 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 ). Macroinve Biological assessment based on the macroinvertebrate assemblage has been widely applied for indications of environmental quality, and often more specifically water quality ( Lenat 1993 ; Cummins and Merritt 2001 ). Invertebrates participate in many fundamental ecological processes, including the breakdown of organic matter and recycling of nutrients. Additionally, invertebrates are a vital component of the food web, making up a large portion of the diets of other organisms (such as fish, amphibians, and birds) ( Cummins and Merritt 2001 ; Helgen 2001 ). In fact, Voshell ( 2002 ) recognized that freshwater invertebrates have been used more often than any other group of organisms for assessing freshwater ecosystems. A great deal of research has been completed on the specific ecology of lotic macroinvertebrates (of, relating to, or living in flow entic macroinvertebrates (of, relating to, or living in still waters). Williams and Feltmate ( 1992 ) noted that, while not well studied, the communities of aquatic insects in wetlands include species from most of the major aquatic groups. The community composition of wetland macroinvertebrates differs from that of flowing waters, because of differences in substrate, dissolved-oxygen level in the water column, hydroperiod, and annual water fluctuations. Macroinvertebrates have been useful indicators of environmental condition in streams; and Karr and Chu ( 1997 ) speculate macroinvertebrates also may be appropriate indicators of environmental 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 wetland biological assessment methodologies have been initiated in many states, including Florida ( Lane et al. 2003 ), Minnesota, Montana, N and Ohio ( Danielson 1998b ). Within the state of Florida and throughout the southeastern Coastal Plain, ecological research on the macroinvertebrate community has included many ecosystem types from isolated marshes of peninsular Florida ( Kushlan

PAGE 27

13 1990 ; Lane 2003 ); isolated wetlands in south Florida ( Stansly et al. 1997 ); 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 structure and function of the macroinvertebrate community accurately reflects the biological condition of a wetland, and thavironmental Protection (FDEP) has initiated the develop; community structure (percentage dominance of the most abundant taxon and pertus. In the northern Everglades, Rader and Richardson (1994) found that macroinvertebrates responded to nutrient enrichment t the macroinvertebrate community composition changes in predictable ways with increased human influence. Because wetland macroinvertebrates complete part or all of their lives in the wetland, they are directly exposed to conditions 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 environment. Their quick response time, reliance on water (both for the water quality and duration of inundation), and ease of collection make macroinvertebrates a favorable assemblage for use as biological indicators. Noted disadvantages to using macroinvertebrates include the amount of time and knowledge necessary for identification to lower taxonomic levels ( Cummins and Merritt 2001 ). The Florida Department of En ment of biological indices based on the macroinvertebrate assemblage for freshwater bodies in Florida. Macroinvertebrate-based biological indices have been created for isolated marshes through the Macroinvertebrate Index of Wetland Condition (MIWC; Lane et al. 2003 ), for Florida streams through the 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 restoration in the Kissimmee River Basin ( Merritt et al. 1996 ), and in freshwater lakes through the Lake Condition Index (LCI; Gerristen and White 1997 ). Different core metrics comprise each multi-metric biological index. Lane ( 2003 ) incorporated five core metrics as biological indicators of wetland condition for isolated marshes in the MIWC, including sensitive taxa, tolerant taxa, predators, Odonata, and Orthocladiinae. The SCI was developed with ten core metrics within six categories of biological organization including were taxonomic richness (total number of taxa, number of Trichoptera taxa, and number of Ephemeroptera taxa); feeding group (percentage filterer individuals); voltinism (long-lived taxa richness); habit (clinger taxa richness) centage Tanytarsini midges); and sensitivity and tolerance (sensitive taxa richness and percentage very tolerant individuals) ( Fore 2004 ). The LCI incorporated seven metrics, including taxa richness, Shannon diversity, Hulbert index, ETO taxa (Ephemeroptera, Trichoptera, and Odonata), percent dominance, filter feeders, and gatherers ( Gerristen and White 1997 ). Numerous studies have documented the response 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 sta

PAGE 28

14 with ased on the effects of adding stormwater to freshwater wetlands the macroinvertebrate community composition. Toth (1993) reported a response of the macroinvertebrate community to water-level manipulation in a Kissimmee River demonstration project. Hmmunity composition, as some macroinvertebrates either temporarily relocated or coped with behavioral and biologi greater number of Coleopteran species present in nutrient-enriched and intermediate areas (especially those in the families Hydrophilidae and Dystcidae) than in non-enriched areas. With shifts in trophic status, the structure of other assemblages also changed, affecting the benthic macroinvertebrate community composition. For example, Adamus and Brandt ( 1990 ) found that shading from dense stands of emergent vegetation altered the distribution among functional feeding groups by limiting the production of benthic algae (thus favoring detritovores over grazers). De Szalay and Resh ( 1996 ) similarly found that increased shading 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 alters the natural hydrology of an isolated wetland, it also increases the inflow of nutrients, sediments, and toxic metals. Harris and Vickers ( 1984 ) found that adding wastewater to pondcypress domes shifted the macroinvertebrate community toward a less-complex trophic structure. Similarly, when wastewater was directed into Florida pondcypress domes, Lemna spp. (duckweed) mats covered the water surface, blocking sunlight from the water column, and creating anoxic conditions ( Dierberg and Brezonik 1984 ). This reduced the diversity and biomass of benthic invertebrates, leaving only a few pollution-tolerant organisms ( Brightman 1984 ). In Florida streams, Barbour et al. ( 1996b ) found that the occurrence of tubificid oligochaetes increased with organic enrichment. Other studies have focu s. Freshwater marshes in Savannas Preserve State Park, Florida, receiving stormwater additions showed increased phosphorus levels, lowered 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 chironomids of the family Orthocladiinae, including those in the genus Cricotopus, were found to be tolerant of metal pollution; while other Orthocladiinae, including Rheocricotopus spp. and Corynoneura spp., were thought to be sensitive to metal pollution in Florida streams. Aside from the nutrient, sediments, and toxics loading, wastewater or stormwater additions also alter the water level and hydroperiod of a wetland system, which affect ydroperiod had a distinct influence on co cal adaptations to changing water conditions. Macroinvertebrates with adaptations to wetland hydroperiods demonstrate both behavioral and physiological adaptations for draw-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 sediments 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.

PAGE 29

15 2000 ). Stansly et al. ( 1997 ) concluded that in isolated wetlands of south Florida, the presence of macroinvertebrates with long life cycles or predatory behavior may indicate hydroperiod stability. Snyder et al. ( 1998 ) found that macroinvertebrates with comparatively short life cycles that are capable of rapid colonization were typical of canals surrounded by urban land uses, whereas the occurrence of macroinvertebrates with longer life cycles were more common in canals surrounded by more natural landscapes. Quantifying Anthropogenic Influence Wetlands occupy a large portion of the Florida landscape. An estimate from the 1780s reported 8 ,225,000 ha of wetlands in Florida (Dahl 2000). By the mid-1980s, the Nation otentially affected by land use practices such as herbicide and pesticide application, oil erosion, and deposition. An additional 17% f the remaining wetlands were adjacent to urban or rural development. Freshwater non-tidal wscales. Three I (quantifying the onrenewable energy use per unit area in the surrounding landscape) stems from earlier al Wetlands Inventory estimated Florida had 4,467,000 ha of wetlands remaining, translating into a loss in Florida of 46% of the pre-1780s wetland area ( Mitsch and Gosselink 1993 ; Dahl 2000 ). Throughout the continental United States, similar trends were apparent, with a drastic decline in the surface area of wetlands. 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 sampled were adjacent to agricultural lands and therefore p irrigation, livestock watering and wastes, s o etlands experienced the greatest development pressure just inland from coastlines as the demand for housing, transportation 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 both 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 in Florida 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 wetlands were harvested and succeeded into shrub wetlands. Anthropogenic activities can influence an array of changes to the physical, chemical, and biological characteristics in surrounding ecosystems. There have been numerous attempts at quantifying anthropogenic influence based on varying primary indices of anthropogenic influences were incorporated throughout our study to compare wetland condition, including the Landscape Development Intensity (LDI) index ( Brown and Vivas 2005 ), the Wetland Rapid Assessment Procedure (WRAP; Miller and Boyd 1999 ), and the Minnesota disturbance index ( Gernes and Helgen 1999 ). Landscape Development Intensity Index The LDI index has been used as a gauge of human activity based on a development intensity measure derived from nonrenewable energy use in the surrounding landscape. The underlying concept behind calculating the LD n

PAGE 30

16 works by Odum (1995), who pioneered emergy analysis for environmental accounting. mergy is an environmental accounting term referring to expressing energy use in solar equivalels, increased flooding), or a ombination of such factors, all of which enter the natural ecological system from the pe (Brown and Vivas 2005). Wetlands surrounded by ore intense activities such as highways and multi-family residential land uses receive higher alus 1999). Minnesota Disturbance Index Wetlands only receive a score in one of the primary categories, and reference standard [E ents ( Odum 1995 ).] Brown and Ulgiati ( 2005 ) suggest that landscape condition, or ecosystem health, is strongly related to the surrounding intensity of human activity, and that ecological communities are affected by the direct, secondary, and cumulative impacts of activities in the surrounding landscape. Healthy ecosystems are defined as those with integrity and sustainability, which correlate to limited development in the surrounding landscape and the maintenance of ecosystem structure and function, even when stressors (e.g. flooding, drought, etc.) are present ( Brown and Ulgiati 2005 ). The LDI does not account for any individual causal agent directly, but instead, may represent the combined effects of air and water pollutants, physical damage, changes in the suite of environmental conditions (ex. groundwater lev c surrounding developed landsca m LDI index values. The highest LDI coefficient of 10.0 is assigned to the urban land use category of Central Business District. Undeveloped land uses such as wetlands, lakes, and upland forests, are assigned an LDI coefficient of 1.0, the lowest possible value, based on no use of nonrenewable energy in these ecosystems. Wetland Rapid Assessment Procedure The Wetland Rapid Assessment Procedure (WRAP) was designed to provide accurate and consistent evaluations of wetland sites, and relies on an evaluator with an adequate understanding of the functions of and species found throughout Florida ecosystems ( Miller and Gunsalus 1999 ). WRAP consists of a qualitative score describing the functional capacity of a wetland. Scores range from 0.0 to 3.0, in 0.5 increments. The primary scoring categories for WRAP include: 1) Wildlife 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 indicates a wetland with a reduced functional capacity ( Miller and Guns The Minnesota disturbance index is considered a gradient of human disturbance, or a measure of land use disturbance based on investigator knowledge, observations, and best professional judgment about the degree of influence to the ecosystem. Gernes and Helgen (1999) used the Minnesota disturbance index as a baseline for creating an index of vegetative biotic integrity for depressional wetlands. There are two primary categories and three secondary categories used to calculate the Minnesota disturbance index score. The primary categories include stormwater and agricultural influence, and are weighted twice as high as the secondary categories. Wetlands receive scores assigned according to significantly affected (S = 8), moderately affected (M = 4), least affected (L = 2), and not applicable (NA = 0) depending on the scorers opinion as to the degree of influence.

PAGE 31

17 wetlands receive a score of zero in both primary categories. The three secondary categorie s include hydrologic/miscellaneous influence, historical influence, and buffer, receiving scores of significantly affected (S = 3), moderately affected (M = 2), least 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, with a score of 0 representing the reference standard condition. Project Overview Physical and chemical environmental parameters and the community composition of the diatom, macrophyte, and macroinvertebrate assemblages were sampled to develop the Florida Wetland Condition Index (FWCI) for isolated depressional forested wetlands for Florida. Wetland study sites were sought in various landscape settings that included undeveloped, agricultural, and urban land uses. Compositional differences in the diatom, macrophyte, and macroinvertebrate assemblages among wetlands were identified and related to the Landscape Development Intensity (LDI) index, a quantitative calculation of anthropogenic activity in the landscape surrounding a wetland which is based on a development intensity measure derived from nonrenewable energy use in the surrounding landscape. Each assemblage was used to construct the three assemblage specific Florida Wetland Condition Indices (FWCIs) for isolated depressional forested wetlands in Florida.

PAGE 32

CHAPTER 2 METHODS Biological, physical, and chemical parameters were sampled in 118 forested wetlands less than 2 ha in size. This chapter describes site selection, calculations of the gradient of anthropogenic activity including the Landscape Development Intensity (LDI) index, field-data collection, and laboratory analyses. Statistical analyses are described for each assemblage and for the creation of the assemblage specific Florida Wetland Condition Indices (FWCIs). Site Selection Field research spanned two growing seasons 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, agricultural, urban). Hereafter wetlands surrounded by greater than 50% undeveloped landscapes were called reference (short for reference standard), wetlands surrounded by greater than 50% agricultural land uses were called agricultural and wetlands surrounded by greater than 50% urban land uses were called urban. Random site selection was not feasible given 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 accomplished with the aide of the Natural Resources Conservation Service under the United States Department of Agriculture and University of Florida Institute of Food and Agricultural Sciences extension agents. Sample wetlands were targeted spatially throughout Florida, so that a nearly equal distribution of wetlands was sampled within each of the four Florida wetland region. Boundaries of the Florida wetland regions were determined with a hydrologic model by Lane ( 2000 ). Florida freshwater palustrine wetlands were classified using a hierarchical classification technique, and physical (surficial geology, soils, digital elevation model, slope) and climatic (precipitation, potential evapotranspiration, runoff, annual days of freezing) variables were tested for correlation with wetland clusters. Final wetland region 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 reference, n = 9 agricultural, n = 10 urban), 31 in the central (n = 11 reference, n = 9 agricultural, 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) wetland regions. All 72 wetlands sampled in 2001 and 39 (of the 46) wetlands sampled in 2002 hosted Taxodium ascendens (pondcypress) as a dominant or co-dominant canopy species. The remaining seven wetlands sampled in 2002 had a canopy layer comprised of a mixture of species that was not dominated or co-dominated by Taxodium ascendens. These wetlands exhibited the characteristic depressional shape in the landscape and from GIS based digital orthophotography. Wetlands surrounded by natural landscapes were generally located on conservation lands 18

PAGE 33

19 #S#S%% #S#S % #S#S %%#S#S% %% #S#S#S #S%%% %#S%#S %#S #S% #S%#S#S#S #S%#S% %%#S #S#S%#S%#S% %% %% % #S#S %%#S#S#S%#S#S%%% %%%%#S#S #S#S #S 0200400KilometersFlorida Ecoregions Panhandle North Central SouthSample Wetlands (n=118)#SReference%Agricultural Urban N Figure 2-1. Study site location of 118 isolated depressional forested wetlands in Florida. The state of Florida was separated into four ecoregions ( Lane 2000 ). Sample wetlands were designated by a priori surrounding land use categories: o reference, agricultural, or urban. including state and national parks and forests, county 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 the widespread historic loss of wetlands throughout Florida ( FDNR 1988 ) and early incentives to drain swamp lands, few isolated depressional forested wetlands were found in the oldest urban areas. Many of the urban wetlands sampled were suspected to previously been embedded in agricultural land uses. Table 2-1 provides some general information about each sample wetland, including sample date, surrounding land use, and land ownership. The sample date provided is the earliest sample date, and correlates to macrophyte 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 after initial hydration for the growing season, which would not 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 wetland region (S = south; C = central;

PAGE 34

20 Table 2-1. Surrounding land use, land ownership, 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. & Res. 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 35

21 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 PR1 6/15/01 National Forest Public NU2 6/11/01 Resid. & Golf Private PR2 7/3/01 WMD Public NU3 6/26/01 Residential Private PR3 7/4/01 Military Public NU4 6/27/01 Residential Private PR4 8/9/01 State Forest Public NU5 6/28/01 Residential Private PR5 8/10/01 State Forest Public NU6 8/1/01 Resid. & Inst. Private PR6 8/18/01 National Forest Public NU7 5/15/02 Comm. & Res. Private PR7 6/4/02 Conserv. Tract Private NU8 6/3/02 Res. & Golf Private PR8 8/7/02 NWR Public NU9 6/12/02 Industrial Private PU1 6/14/01 Residential Private NU10 7/29/02 Res. & Instit. Private PU2 7/5/01 Residential Private PA1 5/24/01 Cattle Private PU3 8/17/01 Res. & Comm. Private PA2 5/29/01 Cattle Private PU4 8/17/01 Res. & 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 Res. & Orchard Private PA6 8/9/01 Cattle Private PU8 6/19/02 Ind. & 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 = reference, A = agricultural, and U = urban. ^Surrounding Land Use abbreviations: NWR = National Wildlife Refuge; WMD = Water Management District; Res. = Residential; Cat. = Cattle; Comm. = Commercial; Inst. = Institutional; Ind. = Industrial; Crops = Row Crops; Silv. = Silviculture; Ease. = Easement. N = north; P = panhandle), land use category, and order of sampling. Site codes were assigned to preserve the anonymity of land owners. Gradients of Anthropogenic Activity Three independent indices of anthropogenic activity were calculated including the Landscape Development Intensity (LDI) index, the Wetland Rapid Assessment Procedure (WRAP), and the Minnesota disturbance index (DI). LDI scores were calculated prior to site visits using 1999 digital orthophoto imagery 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 constructed around the edge of each wetland in ArcView GIS 3.2 ( Environmental Systems Research Institute, Inc. 1999 ). Land uses within the 100 m wetland buffer were delineated based on digital orthophoto

PAGE 36

22 Table 2-2. Field-data collected at 118 sample wetlands. Site Soil Water Diatoms Macrophytes^ Macro-invertebrates Site Soil Water Diatoms Macrophytes^ Macro-invertebrates Site Soil Water Diatoms Macrophytes^ Macro-invertebrates 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 37

23 imagery, and were updated during the site visit to reflect changes in land use since the 1999 images were recorded. The following equation was used to calculate LDI: LDI Total = % LU i LDI i (2-1) where %LU i is the percent of a land use within wetland buffer and LDI i is the LDI coefficient for a particular land use based on the amount of nonrenewable energy use per unit area in the surrounding landscape (Table 2-3). The LDI coefficient values and LDI equation are based on work by Brown and Vivas ( 2005 ). 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 minutes at each study wetland according to descriptions from Miller and Gunsalus ( 1999 ). The DI was scored in the office after the field visit by the field crew leader using information obtained from ArcView GIS 3.2 ( Environmental Systems Research Institute, Inc. 1999 ) and field notes according to categories established by Gernes and Helgen ( 1999 ). The three measures of anthropogenic activity were compared using the Pearson correlation coefficient. Field-data Collection A concise summary of field-data collection procedures follows. Appendix A provides more detailed descriptions of field-data collection techniques in the format of Standard Operating Procedures (SOPs) for field use. Field methods are described as transect establishment followed by water, soil, diatom, macrophyte, and macroinvertebrate sampling techniques. Figure 2-2 shows the positioning of the four 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 cardinal directions (north, east, south, west). The upland/wetland boundary was determined using the Florida Unified Wetland Delineation Methodology (Chapter 62-340, F.A.C.), using a combination of wetland plant presence according to wetland plant status (e.g. obligate, facultative wetland, facultative, or upland) and wetland hydrologic indicators (e.g. lichen lines, moss collars). 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 present throughout at least 50% of the wetland surface 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 wetlands, including 14 in the panhandle wetland region (reference n = 5; agricultural n = 4; urban n = 5), 14 in the north wetland region (reference n = 6; agricultural n = 3; urban n = 5), 23 in the central wetland region (reference n = 9; agricultural n = 6; urban n = 8), and 24 in the south wetland region (reference n = 8; agricultural n = 8; urban n = 8). Dissolved oxygen and water temperature were taken on-site using an YSI-55 Dissolved Oxygen hand meter. Grab water samples were sent to the FDEP Central Chemistry Laboratory, Tallahassee, Florida. Analysis included color (EPA 110.2), turbidity (EPA 180.1), pH (150.1), specific conductance (EPA 120.1), ammonianitrogen

PAGE 38

24 Table 2-3. Landscape Development Intensity coefficients (LDI i ) used in the calculation of the Landscape Development Intensity (LDI) index. Land Use (i) Nonrenewable Energy Use (E14 solar equivalent joules/ha/yr) LDI i 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 (EPA 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 transect (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

PAGE 39

25 West East North South Wetland/upland boundary Quadrats were placed back-to-back and were 1m wide by 5m long. o 4 belted transects of variable length were arranged along the cardinal axes. o o o Figure 2-2. Belted transect layout for macrophyte 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. Food and Agricultural Sciences (IFAS) Analytical Research Laboratory, Gainesville, Florida. Sampling Design Benthic diatoms Benthic diatom samples were collected at 50 isolated depressional forested wetlands throughout Florida between May-September 2001, as listed in Table 2-2. Figure 2-3 shows the 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) wetland regions. Sample wetlands were situated in three a priori described land use categories (reference n=18; agricultural n=16; urban n=16). A minimum of 10 cm of standing water was necessary for benthic diatom collection. A hollow cylinder was placed on the soil surface to isolate an area of substrate with a surface area of 28 cm 2 A bulb pipette was used to loosen the top 0.5 cm layer at the soil surface-water interface, and a 10 mL sample was extracted. This was repeated 10 times throughout the flooded portion of the wetland, resulting in a final sample volume of 100 mL. For preservation, 5 mL of M 3 a standard preservative for

PAGE 40

26 #S%U #S%U#S#S #S#S %U%U#S#S%U%U#S%U %U #S %U#S#S%U#S%U%U%U%U%U%U#S#S#S#S #S 0200400KilometersFlorida Ecoregions Panhandle North Central SouthSample Wetlands (n=50)#SReference%UAgricultural Urban N Figure 2-3. Benthic diatom samples were collected at 50 isolated depressional forested wetlands. The state of Florida was separated into four wetland regions ( Lane 2000 ). Sample wetlands were designated by surrounding land use: o reference, agricultural, or urban. algae samples ( APHA 1995 ), was added to the 100 mL algae sample. Appendix A provides the recipe for creating a batch of M 3 preservative. Benthic diatom 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 identification. 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 using Naphrax (Northern Biological Supplies Limited, Ipswich, England). Five hundred valves were counted along microscope transects and identified to the lowest possible taxonomic level, preferably species (following FDEP SOP#AB-03.1 http://www.dep.state.fl.us/labs/sop ). Macrophytes Macrophyte vegetation was sampled at all 118 wetlands throughout Florida (Figure 2-1). Sampling was conducted along four transects situated as perpendicular crossing axes running through the center of each wetland (Figure 2-2). Along each

PAGE 41

27 transect, a series of 1 m wide by 5 m long quadrats was established back to back. Living macrophytes rooted within each quadrat were identified to the lowest 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 metrics, including category (annual or perennial, evergreen or deciduous, indigenous or exotic) and growth form (aquatic, fern, grass, herb, sedge, shrub, tree, or vine). References specific to Florida were consulted first ( Tobe et al. 1998 ; Wunderlin and Hansen 2003 ). 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, Florida botanists (who also participated in the Floristic Quality Assessment Index) were consulted. Floristic Quality Assessment Index. A Floristic Quality Assessment Index (FQAI) has been included in many of the multi-metric biotic indices created for the macrophyte assemblage. The concept of FQAI was developed by Wilhelm and Ladd ( 1988 ) for vegetation around Chicago, Illinois. This method of scoring plant species based on expert 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 et al. 2003 ; Cohen et al. 2004). The FQAI provides a quantitative means of assessing the fidelity of a plant to a particular environment through the Delphi technique ( Kent 2000 ), where individual botanists assigned coefficients (termed coefficients of conservatism) to individual species. 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 ). The Florida FQAI enlisted regional expert botanists to provide quantitative scores for vegetation in the form of a coefficient of conservatism assigned individually to each species. The FQAI score for an individual wetland was calculated as: FQAI = [ (C 1 + C 2 + . C n ) ] / N (2-2) where C = species specific coefficient of conservatism score (CC), and N = species richness. This equation was considered a modified FQAI because previous studies used the square root of native species richness as the denominator. Theories on the importance of species richness suggest that higher species richness signifies a more valuable ecosystem, which can be quantified by using the square root function ( Fennessy et al 1998 ). However, a recent study by Cohen et al. ( 2004 ) found a stronger relationship along a disturbance gradient for the FQAI of Eq. 2-2 than with the traditional FQAI equation (using the square root of native species richness in the denominator). They also reported that using total species richness (i.e. including exotic species in the calculation of species richness) improved the relationship of mean CC with the human disturbance gradient (measured with LDI). The sum of the species CC scores was divided by total species richness ( Eq. 2-2 ) in this study to account for potential differences in species richness due to differences in wetland regions ( Lane 2000 ), bioregions ( Griffith et al. 1994 ), a priori land use categories, or other unspecified differences.

PAGE 42

28 CC scores were obtained from the Florida botanists surveyed. Each botanist was sent a complete list of species identified in the isolated forested wetlands in the 2001 field season (n = 482 species), and was asked to score each species based on its faithfulness to Florida isolated forested wetlands. After the 2002 field season, one botanist scored the additional 79 taxa not previously encountered, raising the number of taxa with CC scores to 561 species. Potential CC scores ranged from 0 to 10: 0 Exotic and native species that act as opportunistic invaders, included species that commonly occur in disturbed ecosystems 1-3 Species that were widely distributed and occurred in disturbed ecosystems 4-6 Species with a faithfulness to a particular ecosystem, but also tolerant of moderate levels of disturbance 7-8 Species typical of well-established ecosystems that sustain only minor disturbances 9-10 Species occurring within a narrow set of 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 B lists the CC scores for the macrophyte species identified in this study. Macroinvertebrates Macroinvertebrates were collected at 79 isolated depressional forested wetlands throughout Florida (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. Wetlands were sampled in the panhandle (n = 13), north (n = 15), central (n = 25), and south (n = 26) wetland regions. Sample wetlands were situated in 3 a priori land use categories (reference n = 29, agricultural n = 24, and urban n = 26). Samples were collected using a U.S. Standard 30 mesh D-frame net. One sweep covered 0.5 m 2 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 recently been trampled by the field crew. Twenty sweeps were proportioned among major vegetation zones throughout each sample wetland. Sample wetlands generally had only one vegetation zone, although some wetlands did have two or three distinct vegetation zones. Vegetation zones were defined by clear changes in the dominant species. When herbaceous plants were included in the sweep area, the bottom of the net was swept from the substrate up the plant stalks. In areas with woody plant structure, 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 plastic jar and preserved with buffered formalin at a rate of 10% of the sample volume. Appendix A provides a more detailed description of field methods and preservation. Macroinvertebrate identification was completed at the FDEP Central Laboratory, Tallahassee, Florida, following standard operating procedures (FDEP Standard Operating

PAGE 43

29 Figure 2-4. Macroinvertebrates were sampled at 79 isolated depressional forested wetlands. The state of Florida was separated into four wetland regions ( Lane 2000 ). Sample wetlands were designated by surrounding land use: o reference, agricultural, or urban. Procedure #IZ-06 http://www.dep.state.fl.us/labs/sop ). Macroinvertebrate samples were sieved, washed, and placed on a pan with 24 individually numbered cells. Eight of the cells were randomly selected and combined in a second numbered tray. A single cell was randomly selected from the second tray for enumeration and identification of macroinvertebrates. When fewer than 100 individuals were encountered in the sample, a second cell was randomly selected from the second tray, and all of the individuals were enumerated and identified. Identification was to the lowest taxonomic level possible.

PAGE 44

30 Data Analysis Water and Soil Parameters Water and soil parameters within three a priori land use categories were compared using Fishers LSD pair wise comparison in Minitab ( Version 13.1, Minitab Statistical Software ). The non-parametric Mann-Whitney U-Test was used to discern differences among medians of low (LDI < 2.0) and high (LDI 2.0) LDI groups for the water and soil parameters. An LDI break of 2.0 corresponds to a break in the LDI coefficients of undeveloped versus developed land uses (Table 2-3). Summary Statistics Summary statistics for each assemblage included richness (R), evenness (E), and Shannon diversity (H). For the diatom and macrophyte assemblages summary statistics were calculated at the species level, and for the macroinvertebrate assemblage summary statistics were calculated at the genus level. Richness was defined as the total number of distinct taxa encountered within the sample wetland. Evenness has been described as the fraction of maximum possible diversity in a wetland and was calculated as the Shannon diversity index value divided by the natural log of taxa richness: E = H / log (R) (2-3) ( McCune and Grace 2002 ). The Shannon diversity index has been described as measuring the information content of a sample unit where maximum diversity yields maximum uncertainty ( McCune and Grace 2002 ). For Shannon diversity calculations (H), the sample unit was an individual isolated depressional forested wetland: H = p i log(p i ) (2-4) p i = n i / N (2-5) where n i was the number of occurrences of taxon i, and N was the total number of occurrences of all taxa at a wetland. For the diatom and macroinvertebrate assemblages, the number of occurrence represented the enumeration of the laboratory identified sample and the total number of occurrences represented the sum of the number of occurrences of 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 total number of quadrats of all of the species identified. For the diatom and macroinvertebrate assemblage, Simpsons index (S) was also calculated, and the equation: S = 1 (p i p i ) (2-6) where p i was defined in Equation 2-5 as the number of occurrences of taxon i (n i )divided by the total number of occurrences of all taxa at a wetland (N). For the macrophyte assemblage, firstand second-order jackknife estimators of species richness (Jack 1 and Jack 2 respectively) and Whittakers beta diversity (W) were calculated. Firstand second-order jackknife estimators were calculated as estimates of true species richness ( Colwell and Coddington 1994 ). Jackknife estimators are used to gauge the thoroughness

PAGE 45

31 of sampling intensity and the inherent variability of the target system. Assuming that the sampling effort only 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 richness is based on the number of species observed (S), the number of species occurring in only one sample unit (rl) (where one sample unit represents one quadrat), and the number of sample units (n) (quadrats): Jack 1 = S + [(rl*(n-1)) / n] (2-7) The second-order jackknife estimator (Jack 2 ) also incorporated the number of species occurring in exactly two sample units (r2), where a sample unit was defined as a quadrat: Jack 2 = S + [(rl*(2n-3)) / n] [(r2*(n-2)2) / (n*(n-1))] (2-8) These estimators of total species richness have shown useful in predicting actual species richness when only a small area of the total ecosystem has been sampled ( McCune and Grace 2002 ). Whittakers beta diversity (W) was computed as a calculation of overall beta diversity, or the compositional change represented in a sample. Whittakers beta diversity was calculated as the number of species at a particular forested wetland (S c ) divided by the average species richness per quadrat (S) minus one: w = [S c / S] 1 (2-9) The resulting value for Whitakers beta diversity 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 greater than five is considered high ( McCune and Grace 2002 ). Summary statistic means within a priori land use categories were compared with Fishers Least Significant Difference (LSD) pair wise comparison test using Minitab ( Version 13.1, Minitab Statistical Software ). The strength of using Fishers LSD is 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 values 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 three a priori land use categories (reference, agricultural, urban). Gamma diversity was calculated as the overall number of taxa encountered at all sample wetlands per a priori land use category. A higher value of gamma diversity for an a priori land use category suggests a greater difference among the species composition of wetlands within that a priori land use category, assuming equal sample sizes among a priori land use categories. 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 (MRPP) was used to test the similarity of community composition for each assemblage among the four Florida

PAGE 46

32 wetland regions (further application in Zimmerman et al. 1985 ; McCune et al. 2000 ; McCune and Grace 2002 ). MRPP is a nonparametric technique which tests for no difference between groups (the null hypothesis) and is available in PCORD ( Version 4.1 from MJM Software, Gleneden Beach, Oregon ). It is an appropriate procedure for ecological community data as it does not require distributional assumptions of normality and homogeneity of variances. The Srensen distance measure was used to calculate 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 within-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 groups (panhandle versus north versus central versus south) as well as for multiple pair wise comparisons (panhandle versus north, panhandle versus central, panhandle versus south, north versus central, north versus south, and central versus south) for the Florida wetland regions ( Lane 2000 ). Community Composition Community composition of each assemblage was summarized in a non-metric multidimensional scaling (NMDS) ordination to relate changes in community composition with environmental gradients. NMDS is an ordination technique designed to compress multi-dimensional space, and is particularly agreeable with ecological data because it does not rely on linear relationships among variables. This has been described as a compensation for the zero-truncation problem through the use of ranked distances and the application 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 (i.e. Principal Components Analysis, PCA) depend upon a value for each measured variable for each sample unit, interpreting shared zeros as a positive relationship. NMDS is useful for presence/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, NMDS caters to non-parametric, community ecology data sets. NMDS explored the dissimilarities of the community composition of sample wetlands for each assemblage. The Srensen (Bray-Curtis) distance measure was used for ordination. The dimensionality was chosen based on an initial six dimensional run in autopilot mode, which suggested an optimal three dimensional solution for the each community composition dataset. To find the optimal three dimensional solution, 50 runs with real data and 50 randomized runs were performed with the instability 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 stability and reproducibility in results. The final run was completed with the starting point set as the results from the best experimental with dimensional run, considered the run with the lowest stress and best overall fit. Water and soil parameters, LDI, latitude, and longitude were correlated with the NMDS ordination axes with Pearson correlation coefficients. To improve normality and decrease skewness, 11 water and soil parameters were log (base 10) transformed, including water parameters (dissolved oxygen concentration, temperature, color,

PAGE 47

33 turbidity, specific conductivity, ammonia-nitrogen concentration, nitrate/nitrite-nitrogen concentration, total Kjeldahl nitrogen (TKN) concentration, and total phosphorus (TP) concentration) and soil parameters (TKN and TP concentration). Water pH was not transformed as it is already on a log scale (pH is calculated as the log base 10 of the hydronium ion concentration). The remaining parameters were measured as percentages, and were transformed by taking the arcsine square root, including soil moisture and soil organic matter. Metric Development In the context of this study, metrics were defined as biological attributes which have a consistent and predictable response to anthropogenic activities ( Karr and Chu 1997 ). Metrics were summarized in five main categories: 1. Tolerance metrics (determined with indicator species or established index values) 2. Autecological (metrics that explore a previously described relationship between taxa and an environmental gradient) 3. Community structure (metrics that explore taxonomic structure) 4. Community balance (metrics with calculated values, such as evenness or dominance) 5. Functional group (metrics related to feeding behavior) Appendix C provides tables of candidate metrics for each assemblage, including 59 candidate diatom metrics (Table C-1), 68 candidate macrophyte metrics (Table C-2), and over 100 candidate macroinvertebrate metrics (Table C-3). Candidate metrics were calculated at the statewide scale for both the diatom and macroinvertebrate assemblages, as sample sizes were limited for regional metric development, particularly in the north and panhandle wetland regions. Candidate macrophyte metrics were calculated at both the regional scale and statewide. Metrics were calculated in multiple forms including richness (R) and proportion (P) metrics. The richness metric (R) simply represents a count of the taxa present within a metric category. The proportion metric (P) was calculated as the richness metric (R) divided by the total species richness (N) for each sample wetland: P i = R i / N i (2-10) Candidate metrics were accepted if they showed a constant and predictable change along the LDI ( Brown and Vivas 2005) according to the strength and significance of the Spearmans correlation coefficient ( Analyze-It software v. 1.67 for Microsoft Excel). The Spearman rank correlation tests for an association between two related variables, and is a non-parametric alternative to the Pearson correlation coefficient. Scatter plots were constructed for each candidate metric versus LDI to ensure correlations were visually distinguishable. Candidate metrics were subjected to a Mann-Whitney U-test to detect differences between low and high LDI groups.

PAGE 48

34 Indicator Species Analysis For each assemblage, sample wetlands were categorized into two LDI groups and analyzed with Indicator Species Analysis (ISA) in PCORD ISA 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 sites be assigned into groups and species data on the abundance or presence of taxa in each group. These groups are commonly defined by categorical environmental variables, levels of disturbance, experimental treatments, presence and absence of a target species, or habitat types ( McCune and Grace 2002 ). Mathematical equations are available in Dufrne and Legendre ( 1997 ) and McCune and Grace ( 2002 ). The calculated indicator species values are based on two standards, faithfulness and exclusion. Faithfulness is defined mathematically by a particular taxa always being present in a particular group. The perfect indicator taxon is exclusive to a group, meaning it would never occur in other groups. Indicator values range from 0 (no indication) to 100 (a perfect indication of a particular group). ISA was conducted to determine sensitive and tolerant indicator taxa separately for each species assemblage. Statewide ISAs were run for the diatom and macroinvertebrate assemblages. ISA was conducted for each wetland region (panhandle, north, central, and south) as well as statewide for the macrophyte assemblage. Since correlations between LDI and each species assemblage were uncertain, iterative ISAs were calculated for each species assemblage. ISA depends on the predetermined categorization of sites, and sites were divided into groups based upon the calculated LDI scores. ISA was run for each LDI break from 1.0 to 7.0 at each 0.5 increment. For example, at an LDI break of 1.0, all sites with a calculated LDI of 1.0 were assigned to Group A, and all sites with a calculated LDI > 1.0 were assigned to Group B, and an ISA calculation completed. Then, for a separate iteration, sites were divided in Group A (LDI < 1.5) and Group B (LDI 1.5) and an ISA calculation was completed. Next, sites were assigned to Group A (LDI < 2.0) and Group B (LDI 2.0) and an ISA calculation was completed. This process of assigning sites into groups based on their calculated LDI scores was repeated at LDI breaks of 2.5, 3.0, 3.5, 4.0, 4.5, 5.0, 5.5, 6.0, 6.5, and 7.0. The idea behind the iterative ISA calculations was to broaden the application of the LDI to detect changes in species composition among wetland sites for different species assemblages and for different anthropogenic influences which may be specific or more dramatic to one species assemblage. The final species lists for tolerant and sensitive indicator species were determined for the LDI break yielding the greatest number of indicator species and the strongest significance based on the Monte Carlo randomization technique. Calculated indicator values were tested for statistical significance using a Monte Carlo randomization technique with 1000 randomized runs. Indicator taxa categorized as tolerant taxa were associated with the higher LDI group (Group B); indicator taxa deemed sensitive taxa were associated with the lower LDI group (Group A). In the macrophyte assemblage, the Spearman rank correlation was used to assess differences between statewide and regional indicator species lists for each of the four wetland regions. To test for equal application of the statewide indicator species list for each wetland region, the non-parametric Kruskal-Wallis test was run with Analyse-It

PAGE 49

35 software ( Ott and Longnecker 2001 ). Distributional differences were analyzed between a priori categories for both tolerant and sensitive indicator taxa among wetland regions. Diatom Metrics Diatom metrics (Appendix C Table C-1) were adopted from previous studies and could be categorized in three categories 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 with ISA. Community composition metrics included richness, evenness, and diversity calculations. Autecological metrics were based on available research that correlated individual diatoms with morphology, behavior, and the physical and chemical water environment. Diatom species were assigned ecological indicator values 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 categorizing diatoms according to water preference (moisture tolerance), nitrogen metabolism, pH, salinity, dissolved oxygen, saprobic condition, and trophic status. Pollution tolerance classification values were taken from Bahls ( 1993 ) analysis of diatoms in Montana streams. Metrics based on morphology and motility were assigned from Stevenson ( Lane 2003 ). Macrophyte Metrics Macrophyte metrics (Appendix C Table C-2) were adopted from previous studies and included tolerance, exotic species, Floristic Quality Assessment Index (FQAI), longevity and plant growth form (native perennial), 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. 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 Wunderlin and Hansen 2003 ). The exotic species metric was calculated as the proportion of species that were exotic to Florida divided by the number of species identified at each particular isolated depressional forested wetland. The timeline for determining the exotic status of a species was set near the beginning of European settlement in North America. 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 ) determined that disturbance often favors annual species over perennial species and promotes the invasion of nonnative perennials in wetlands. Galatowitsch et al. ( 2000 ) found that while native perennial cover was reduced in wetlands impacted by cultivation, the occurrence of introduced perennials rather than annuals increased in stormwater impacted wetlands. The wetland status metric was calculated as the percent of plants classified as obligate or facultative wetland indicator species divided by the number of species at each wetland. Wetland indicator status classifications was from FDEPs vegetative index as presented in section 62-340.450, F.A.C ., which can be found in Tobe et al. ( 1998 ), USDA

PAGE 50

36 NRCS ( 2002 ), and Wunderlin and Hansen ( 2003 ). There are five potential species classifications, including obligate, facultative wetland, facultative, upland, and invisible. In some cases, when the Florida wetland indicator status was not available, the National Wetlands Inventory wetland indicator status for the United States was used. Macroinvertebrate Metrics Candidate metrics for the macroinvertebrate assemblage (Appendix C Table C-3) were adopted from previous studies and could be classified in four categories, including tolerance, community structure, community balance, 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 calculated. Many of the established tolerance metrics were created for flowing waters, which complicated the application of established tolerance values (ex. the Florida Index from Beck 1954 and Barbour et al. 1996b ; 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 included richness measures ( Danielson 1998a ), for example the number of distinct species or specified taxonomic units such as the number of families, genera, or species in a sample. Examples of taxa richness metrics include total taxa richness, Ephemeroptera richness, number of Coleoptera species, or number of Insecta species. The use of community balance metrics included some measure of abundance or relative abundance in an attempt to measure the evenness of the macroinvertebrate community ( Lenat and Barbour 1994 ). Examples of community balance metrics include the Shannon diversity index or the percent contribution of the most abundant taxon ( Lenat and Barbour 1994 ). Macroinvertebrate taxa were grouped based on their functional relationships that overlap taxonomic categorization, including functional feeding groups, habitat groups, and voltinism groups (life-cycle patterns). Cummins and Merritt ( 2001 ) suggest using ratios of numerical abundance or, more favorably, biomass of the various functional groups as indicators of ecosystem attributes, essentially considering the functional groups as surrogates of ecosystem condition. Functional feeding group metrics were based on the morphological structures and behaviors responsible for food acquisition by particular taxa at a site ( Resh and Jackson 1993 ; Danielson 1998a ). For example, herbivores consume algae and plant material, while predators consume animals, omnivores eat both plant and animal materials, and detritovores consume decomposed particulate material ( Helgen 2001 ).

PAGE 51

37 Florida Wetland Condition Index Candidate metrics were selected for inclusion in the Florida Wetland Condition Index (FWCI) if they satisfactorily met three criteria: Metrics were correlated with the LDI according to the strength and significance of the Spearmans correlation coefficient Metrics displayed visually distinguishable correlations with LDI in scatter plots Metrics showed significant differences between low (LDI < 2.0) and high (LDI 2.0) LDI groups tested with the Mann-Whitney U-test. An FWCI was constructed for each assemblage, including the diatom FWCI, the macrophyte FWCI, and the macroinvertebrate FWCI. Each FWCI was composed of individual metrics specific to the assemblage, which were scaled and added together. Metric scoring was based on an approach modified from the SCI, a Florida based biological index of the macroinvertebrate assemblage used to discern stream condition ( Fore 2004 ). Metrics with a skewed distribution were log transformed to improve the distribution. The 5 th to 95 th percentile values of each metric were normalized from 0 to 10, with 10 always representing the best biological wetland condition. The selected metrics, FWCI, and LDI were correlated with water and soil parameters using Spearmans correlation coefficient. In order to determine whether the FWCI provided comparable scores for wetlands with similar community composition within each assemblage, an agglomerative cluster analysis in PCORD was used to determine wetland clusters ( McCune and Grace 2002 ). The dissimilarity matrix was constructed using the Srensen distance measure and the flexible beta linkage method ( = -0.25), which is a flexible clustering setting designed to reduce chaining in the dendrogram. The resulting dendrogram was pruned to maintain the smallest number of significantly different clusters based on Fishers LSD pair wise comparison (p < 0.05). To compare differences in wetland condition between the three species assemblages, metrics selected for inclusion in the three FWCIs among different species assemblages were compared using the Pearson correlation coefficient ( Analyse-it Software ).

PAGE 52

CHAPTER 3 RESULTS Quantifying Anthropogenic Influence Three independent quantitative measures of anthropogenic activity were calculated for the 118 isolated depressional forested wetlands including the Landscape Development Intensity (LDI) index, the Wetland Rapid Assessment Procedure (WRAP), and the Minnesota disturbance index (DI). Individual scores for each wetland are presented in Appendix D. Potential LDI scores ranged from a minimum of 1.0 to a maximum of 10.0 representing a highly impacted wetland. Actual LDI scores ranged from 1.0 at 26 wetlands to 7.2 at two wetlands including SU8 and CU6, two urban wetlands surrounded by transportation and commercial land uses. The mean LDI score for a priori reference wetlands was 1.1 ( = 0.2). A priori agricultural and urban wetlands had similar mean LDI scores at 4.5 ( = 1.4) and 4.8 ( = 1.5), respectively. Scores for the WRAP potentially range from 0.0 to 3.0, with the lowest score of 0.0 representing a severely impacted wetland. Actual WRAP scores ranged from 3.0 at seven a priori reference wetlands to 0.5 at two urban wetlands including SU8 and PU10. The mean WRAP score for a priori reference wetlands was 2.8 ( = 0.2), with a priori agricultural and urban wetlands having similar mean scores of 1.6 ( = 0.5) and 1.4 ( = 0.4), respectively. The third gradient of anthropogenic influence, the Minnesota DI ranges from zero to 17 with unimpacted wetlands receiving a score of zero. Actual Minnesota DI scores ranged from zero at 15 a priori reference wetlands to 17 at three a priori agricultural wetlands. The mean Minnesota DI score for a priori reference wetlands was 1.6 ( = 2.1). A priori agricultural and urban wetlands had higher mean Minnesota DI scores at 11.3 ( = 5.4) and 12.2. ( = 3.6), respectively. Comparisons among the three measures of anthropogenic influence showed strong and significant correlations (Pearsons correlation coefficient |r| 0.84, p < 0.01) (Table 3-1). WRAP and the Minnesota DI had the greatest correlation coefficient (r = 0.89). The comparisons between LDI and WRAP and LDI and the Minnesota DI also had large correlation coefficients (0.87 and 0.84, respectively). The LDI index was selected for further analysis and metric selection for the development of the Florida Wetland Condition Index (FWCI) because it is an objective and repeatable measure of anthropogenic activity and it is not based on the subjective scoring of an assessor. Water and Soil Parameters Physical and chemical properties of 75 water samples and 118 soil samples were analyzed for isolated depressional forested wetlands. Table 3-2 shows mean values for water and soil parameters for the three a priori land use categories (reference, agricultural, and urban). Means with similar letters were not significantly different (Fishers LSD pair wise comparison, = 0.05), including water temperature, water 38

PAGE 53

39 Table 3-1. Comparison among three measures of anthropogenic influence, including the Landscape Development Intensity (LDI) index, the Wetland Rapid Assessment Procedure (WRAP), and the Minnesota disturbance index (DI). LDI WRAP WRAP 0.87 DI 0.84 0.89 Values are Pearson correlation coefficients. All correlations were significant (p < 0.01). nitrate/nitrite-nitrogen, and soil TKN, which were not significantly different among a priori land use categories. Of the 14 water and soil parameters measured, eight water and three soil parameters showed significant difference (p < 0.05) among reference, agricultural and/or urban wetlands, categorized by the three a priori land use categories. Reference wetlands had significantly different dissolved oxygen (x = 2.9 mg O 2 /L), turbidity (x = 3.8 NTU), water pH (x = 5.2), and water column TP (x = 0.08 mg P/L), than agricultural and urban wetlands. Water ammonia-nitrogen (mg N/L), water TKN (mg N/L), soil moisture, and soil TP (mg P/g soil) were significantly different between reference and agricultural wetlands. Specific conductivity was significantly different between reference (x = 81 umhos/cm) and urban wetlands (x = 231 umhos/cm). The water color of urban wetlands (x = 198 PCU) was significantly different from reference (x = 285 PCU) and agricultural (x = 346 PCU) wetlands. Soil organic matter was significantly different between agricultural (x = 30%) and urban wetlands (x = 41%). Measures of four water and two soil parameters were significantly different (p<0.05) between sample wetlands in low (LDI < 2.0) and high (LDI 2.0) LDI groups (Table 3-3). Dissolved oxygen levels were higher in less developed wetlands (x = 2.9 mg O 2 /L) as compared to wetlands surrounded by higher development intensity (x = 1.8 mg O 2 /L), and wetlands surrounded by lower development intensity had lower water column pH (x = 5.2) than wetlands in higher development intensity landscapes (x = 6.3). Low LDI wetlands had significantly less TP in both the water column (x = 0.08 mg P/L) and soil (x = 0.39 mg P/L), than the high LDI wetlands. The physical measure of turbidity showed that wetlands surrounded by low development intensity had lower turbidity (x = 3.8 NTU) than those surrounded by higher development intensity (x = 13.2 NTU). Additionally, low LDI wetlands had higher soil moisture (x = 59) than high LDI wetlands (x = 50).

PAGE 54

Table 3-2. Water and soil parameters among three a priori land use categories, including reference, agricultural, and urban wetlands. Reference Agricultural Urban n x 95%CI n x 95%CI n x 95%CI Water Parameters Dissolved oxygen (mg/L) 29 2.9 1.7 (2.3 to 3.6) a 18 1.6 0.9 (1.2 to 2.1) b 24 1.9 1.1 (1.4 to 2.4) b Temperature (C) 29 26.2 2.8 (25.1 to 27.3) a 18 25.2 1.9 (24.2 to 26.2) a 24 24.9 2.4 (23.9 to 25.9) a Color (PCU) 30 285 178 (218 to 352) a 19 346 204 (248 to 445) a 26 198 129 (146 to 251) b Turbidity (NTU) 30 3.8 4.2 (2.3 to 5.4) a 19 17.7 40.7 (0.0 to 37.3) b 26 9.5 11.9 (4.7 to 14.3) b pH 30 5.2 1.2 (4.8 to 5.6) a 19 6.2 0.8 (5.8 to 6.6) b 26 6.4 1.0 (5.9 to 6.8) b Specific conductivity (umhos/cm) 10 81 48 (46 to 115) a 10 136 134 (39 to 232) ab 13 231 175 (126 to 337) b Ammonia-nitrogen (mg N/L) 30 0.15 0.33 (0.03 to 0.28) a 19 0.33 0.57 (0.06 to 0.61) b 26 0.19 0.27 (0.08 to 0.30) ab Nitrate/nitrite-nitrogen (mg N/L) 26 0.09 0.37 (0.00 to 0.24) a 15 0.01 0.01 (0.00 to 0.01) a 21 0.02 0.03 (0.00 to 0.03) a TKN nitrogen (mg N/L) 30 1.93 1.24 (1.47 to 2.39) a 19 3.17 2.20 (2.11 to 4.23) b 26 1.84 1.06 (1.41 to 2.27) ab Total phosphorus (mg P/L) 30 0.08 0.11 (0.04 to 0.12) a 19 0.81 1.38 (0.15 to 1.47) b 26 0.24 0.26 (0.13 to 0.34) b Soil Parameters Moisture (%) 38 61 20 (54 to 67) a 39 46 17 (40 to 51) b 41 55 20 (48 to 61) ab Organic matter (%) 38 40 25 (32 to 48) ab 39 30 17 (24 to 35) a 41 41 28 (32-50) b TKN nitrogen (mg N/g soil) 38 6.76 3.68 (5.55 to 7.97) a 39 5.53 3.30 (4.46 to 6.59) a 41 6.71 4.75 (5.21 to 8.20) a Total phosphorus (mg P/g soil) 38 0.38 0.28 (0.29 to 0.47) a 39 0.91 1.27 (050 to 1.32) b 41 0.53 0.31 (0.43 to 0.63) ab *Minimum Detection Limits: Ammonia-nitrogen 0.010 mg N/L; Nitrate/nitrite-nitrogen 0.004 mg N/L; TKN nitrogen 0.060 mg N/L; Total phosphorus 0.015 mg P/L x = mean; = standard deviation; 95% CI = 95% confidence interval Categories with dissimilar letters were significantly different (Fisher's LSD pair wise comparison, =0.05).

PAGE 55

Table 3-3. Water and soil parameters among LDI groups. The Low LDI Group includes wetlands with an LDI < 2.0, and the High LDI Group includes wetlands with an LDI 2.0. Low LDI Group High LDI Group n x 95% CI n x 95% CI W^ p` Water parameters Dissolved oxygen (mg/L) 30 2.9 1.7 (2.2 to 3.5) 41 1.8 1.0 (1.5 to 2.1) 367 <0.01* Temperature (C) 30 26.1 2.8 (25.1 to 27.2) 41 25.0 2.2 (24.4 to 25.7) 453 0.06* Color (PCU) 31 292 179 (226 to 358) 44 255 177 (202 to 309) 568 0.21* Turbidity (NTU) 31 3.8 4.1 (2.3 to 5.3) 44 13.2 28.1 (4.6 to 21.7) 907 0.02* pH 31 5.2 1.1 (5.0 to 5.6) 44 6.3 0.9 (6.0 to 6.6) 1063 <0.01* Specific conductivity (umhos/cm) 10 81 48 (46 to 115) 23 190 163 (119 to 260) 165 0.05* Ammonia-nitrogen (mg N/L)* 31 0.15 0.33 (0.03 to 0.27) 44 0.25 0.43 (0.12 to 0.38) 853 0.07* Nitrate/nitrite-nitrogen (mg N/L)* 26 0.09 0.37 (0.00 to 0.24) 36 0.01 0.03 (0.01 to 0.02) 450 0.77* TKN nitrogen (mg N/L)* 31 1.99 1.27 (1.53 to 2.46) 44 2.37 1.75 (1.84 to 2.90) 705 0.81* Total phosphorus (mg P/L)* 31 0.08 0.11 (0.04 to 0.12) 44 0.48 0.96 (0.19 to 0.77) 1057 <0.01* Soil parameters Moisture (%) 45 59 19 (54 to 65) 73 50 20 (46 to 55) 1163 <0.01* Organic matter (%) 45 40 24 (33 to 48) 73 35 24 (29 to 41) 1357 0.11* TKN nitrogen (mg N/g soil) 45 6.88 3.66 (5.78 to 7.98) 73 6.00 4.15 (5.03 to 6.97) 1366 0.13* Total phosphorus (mg P/g soil) 45 0.39 0.27 (0.31 to 0.48) 73 0.74 0.97 (0.51 to 0.96) 2140 <0.01* *Minimum Detection Limits: Ammonia-nitrogen 0.010 mg N/L; Nitrate/nitrite-nitrogen 0.004 mg N/L; TKN nitrogen 0.060 mg N/L; Total phosphorus 0.015 mg P/L ^W = Mann-Whitney U-test statistic. `p = significance value.

PAGE 56

42 Diatom Assemblage Statewide 50 wetlands were sampled with 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%). The three 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 ecoregion, 10 wetlands were sampled with four reference, four agricultural, and two urban wetlands hosting 73 diatom taxa. In the north ecoregion 10 wetlands were sampled (four reference, two agricultural, and four urban) with 94 taxa encountered. The central ecoregion included 13 sample wetlands (five reference, four agricultural, and four urban) with 112 taxa sampled. The south ecoregion included 17 sample wetlands (six reference, five agricultural, and six urban) with 147 taxa identified. Summary Statistics Richness (R), evenness (E), Shannon diversity (H), and Simpsons index (S) were calculated for each sample wetland (Appendix E). Table 3-4 summarizes the richness, evenness, and diversity calculations of the diatom assemblage by a priori land use category. Species richness ranged from nine taxa at CR6 (embedded within upland pine forest) and CU1 (bordered on three sides by an arboretum and on the remaining side by a paved road) to 39 taxa at CU3 (surrounded by commercial and residential land uses). Species evenness ranged from 0.57 at NU5 (surrounded by residential and transportation), to 0.89 at CA6 (surrounded by citrus groves). Shannon diversity ranged from 1.41 at NU5 to 2.95 at SR5 (surrounded by marsh and flooded flatwoods). Simpsons index was highest at SR5 and SU2 (adjacent to a recently built public school and transportation) at 0.93, and lowest at NU5 at 0.58. Richness, evenness, Shannon diversity, and Simpsons index were not significantly different among the three a priori land use categories (Table 3-4) or between LDI groups (Table 3-5). 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 the low LDI group (beta diversity of 7.5 and gamma diversity of 145). Compositional Analysis MRPP was calculated across all groups (panhandle versus north versus central versus south) as well as for multiple pair wise comparisons (panhandle versus north, panhandle versus central, panhandle versus south, north versus central, north versus south, and central versus south) in order to test the similarity of diatom taxa composition among the wetland regions. Table 3-6 shows the results for the MRPP tests, including the test statistic (T), the chance-corrected within-group agreement (A, a measure of within group similarity), and the significance value (p). The global comparison among all wetlands and the three a priori categories showed that diatom community composition

PAGE 57

43 Table 3-4. Diatom richness, evenness, and diversity among a priori land use categories. Reference Agricultural Urban Species richness (R) 19 (8) a 19 (6) a 22 (8) a Species evenness (E) 0.74 (0.09) a 0.75 (0.08) a 0.73 (0.10) a Shannon diversity (H) 2.13 (0.49) a 2.18 (0.34) a 2.21 (0.49) a Simpson's index (S) 0.80 (0.09) a 0.82 (0.07) a 0.81 (0.10) a Beta diversity 6.9 6.1 5.8 Gamma diversity 132 117 126 Values represent the mean standard deviation. Categories with dissimilar letters are significantly different (Fisher's LSD pair wise comparison, = 0.05). at the species level was significantly different ( = 0.05). Within the pair wise comparisons, only the panhandle versus south and north versus south comparisons had significantly different diatom community composition for all land use types. Reference wetlands in the south wetland region had a significantly different diatom community composition compared to wetlands in both the panhandle and north ecoregions. Similarly, diatom community composition among pair wise comparisons of agricultural wetlands was significantly different for the panhandle versus south ecoregions. Only the north and south ecoregions had significantly different diatom community composition among urban wetlands. Community Composition Figure 3-1 shows a two dimensional bi-plot of the NMS axes used to explore diatom community composition with overlays of significant (transformed) environmental variables, including LDI and six water parameters: DO (log), water turbidity (log), pH, specific conductivity (log), TKN (log), and TP (log). Table 3-7 provides the Pearson rTable 3-5. Mean diatom summary 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 9(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 Values represent the mean (standard deviation). ^W = Mann-Whitney U-test statistic. `p = significance value.

PAGE 58

44 Table 3-6. Similarity of diatom community composition using MRPP. Sites (n) T ^ A` p # All wetlands All regions (P vs N vs C vs S) 50 -2.2 0.01 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 -2.5 0.06 0.02* North vs central 23 -0.7 0.02 0.20* North vs south 27 -3.9 0.09 0.00* Central vs south 30 -1.7 0.04 0.06* Reference wetlands All regions (P vs N vs C vs S) 18 -1.3 0.09 0.11* Panhandle vs north 8 -0.7 0.07 0.21* Panhandle vs central 8 0.8 -0.07 0.76* Panhandle vs south 10 -2.1 0.16 0.03* North vs central 8 0.9 -0.08 0.82* North vs south 10 -2.4 0.17 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 -2.4 0.16 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 -0.3 0.02 0.38* Urban wetlands All regions (P vs N vs C vs S) 16 -0.7 0.05 0.23* Panhandle vs north 6 -0.9 0.15 0.18* Panhandle vs central 6 0.9 -0.07 0.83* Panhandle vs south 8 -0.3 0.02 0.37* North vs central 8 -0.1 0.01 0.37* North vs south 10 -1.9 0.11 0.05* Central vs south 10 1.1 -0.05 0.86* *A high |T| value and significant p-value (p<0.05) suggests a difference in species composition ^ T = the MRPP test statistic `A = the chance corrected within-group agreement #p = the significance value

PAGE 59

45 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 UseReferenceAgriculturalUrban Figure 3-1. NMS ordination bi-plot of 50 wetlands in diatom species space with an overlay of environmental parameters. LDI, dissolved oxygen-DO (log), turbidity (log), water column pH, specific conductivity-Spec.Cond. (log), water TKN (log), and water TP (log), 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 60

46 Table 3-7. Pearson r-squared correlation coefficients between environmental parameters and NMS ordination axes based on diatom community composition. Axis 1 Axis 2 Axis 3 Incremental r 2 12.4% 35.5% 26.7% Cumulative r 2 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 DO (log) 0.33 0.08 0.09 Temperature (log) 0.03 0.08 0.07 Color (log) 0.15 0.09 0.17 Turbidity (log) 0.22 0.01 0.00 pH 0.06 0.47 0.45 Spec.Cond. (log) 0.09 0.27 0.06 Ammonia-N (log) 0.13 0.00 0.03 Nitrate/nitrite-N (log) 0.01 0.01 0.05 TKN (log) 0.23 0.00 0.01 Water TP (log) 0.33 0.08 0.00 %Soil moisture (arcsine sqroot) 0.11 0.00 0.08 Soil TP (log) 0.07 0.05 0.02 squared correlation coefficients between the environmental variables and the NMS ordination axes. A three dimensional solution was constructed with an overall stress of 16.3 and a final stability of 0.00001, which is borderline high but 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 was correlated with LDI, DO (log), water TKN (log), and water TP (log). Axis 2 explained 35.5% variance and was correlated with pH and specific conductivity (log). Axis 3 explained an additional 26.7% variance and was correlated with water column pH. Metric Selection The pool of potential candidate metrics was streamlined to reduce the redundancy of selected metrics. Metrics selected for inclusion represented two of the metric categories, including tolerance metrics and autecological metrics. In general, at least one metric was selected from each of the autecological categories according to which metric was most highly correlated with LDI. If two metrics were calculated from the same set of taxa representing redundant metrics only one metric was included (i.e. metrics based on the same autecological category were not included, for example if Dissolved Oxygen Class 1 were included, no other metric based on Dissolved Oxygen Class would be included). Seven metrics which were significant according to the Spearmans correlation coefficient (|r| 0.46, p < 0.01) (Table 3-8), showed visually distinguishable trends with

PAGE 61

47 Table 3-8. Spearmans correlation coefficients for seven diatom metrics with LDI. Diatom Metrics Spearmans 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 All correlations were significant (p < 0.01) LDI, and were significant for the Mann-Whitney U-test between LDI groups (p < 0.10) (Table 3-9) were selected for inclusion in the diatom FWCI. Metrics selected for the FWCI were calculated as the percent of species richness using presence/absence data. Tolerance metrics included Tolerant Indicator Species and Sensitive Indicator Species. The five autecological 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 saturation between 10-25% and a biological oxygen demand of approximately 13-22 mg O 2 /L), pH Class 3 (circumneutral, mainly occurring at pH values around 7), and Dissolved Oxygen Class 1 (requiring continuously high dissolved oxygen concentrations near 100%). Pollution Class was established by Bahls ( 1993 ), and Nitrogen Uptake Metabolism, Saprobity, pH, and Dissolved Oxygen Classes were defined by van Dam et al. ( 1994 ). Tolerant Indicator Species, Pollution Class 1, Nitrogen Uptake Metabolism Class 3, Saprobity Class 4, and pH Class 3 increased with increasing LDI. Sensitive Indicator Species and Dissolved Oxygen Class 1 decreased with increasing LDI. Table 3-9 shows that all of the selected metrics significantly differentiated between the high (LDI 2.0) and low (LDI < 2.0) LDI groups (p < 0.01). Table 3-9. Comparisons among diatom metrics and the diatom FWCI for LDI groups. Metric Low LDI High LDI W ^ p` Tolerant Indicator Species 1.4 (2.5) 35.5 (15.8) 309.0 <0.001 Sensitive Indicator Species 36.6 (25.0) 18.0 (18.5) 719.0 <0.001 Pollution Class 1 3.7 (6.0) 7.6 (10.1) 337.0 <0.001 Nitrogen Class 3 7.7 (15.5) 20.3 (21.5) 361.0 0.003 Saprobity Class 4 2.7 (4.5) 25.4 (25.6) 373.5 0.006 pH Class 3 18.3 (22.2) 14.1 (16.6) 344.0 0.001 Dissolved Oxygen Class 1 69.3 (24.3) 39.9 (23.8) 658.0 0.004 Diatom FWCI 55.3 (11.1) 45.4 (27.1) 718.0 <0.001 Values represent the mean (standard deviation). W ^ = the Mann-Whitney U-Test statistic p` = the significance value.

PAGE 62

48 Tolerance metrics The lists of Tolerant and Sensitive Indicator Species were determined using Indicator Species Analysis (ISA) using species-level abundance data for the diatom assemblage. Tolerant diatom indicator species were established at an LDI break of 4.75, and included 17 species representing nine genera. Table 3-10 lists the diatom Tolerant Indicator Species. The three tolerant indicator species with 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 Indicator Species increased with increasing LDI. The two wetlands with the highest percent Tolerant Indicator Species in the high LDI group included SU1 (a wetland surrounded by residential land use and a golf course, which was previously embedded in pasture) and CU5 (surrounded by transportation and stormwater retention basins). Four agricultural and urban wetlands and 10 reference wetlands had zero percent tolerant diatom indicator species. Diatom Sensitive Indicator Species were selected at an LDI break of 1.25, correlating to a break in the natural and developed land uses. Table 3-11 lists the 18 Sensitive Indicator Species. The five 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 with increasing development intensity in the surrounding landscape. Six agricultural and 2 urban wetlands hosted no sensitive indicator species. Table 3-10. Diatom Tolerant Indicator Species. Tolerant Indicator Species Indicator Value Cyclotella pseudostelliger 20.0 Diploneis elliptica 19.7 Navicula confervacea 45.3 Navicula minima 48.0 Navicula mutica 40.5 Navicula recens 12.9 Navicula subminuscula 12.5 Neidium alpinum 20.0 Nitzschia subacicularis 18.1 Pinnularia braunii 22.6 Pinnularia divergentissima 13.3 Stauroneis kriegeri 13.3 All reported indicator values were significant (p < 0.10).

PAGE 63

49 0102030405060700246LDI%Tolerant Indicator Specie 8 s Figure 3-2. Percent diatom Tolerant Indicator Species increased with increasing development intensity (LDI). Table 3-11. Diatom Sensitive Indicator Species. Indicator Species Indicator Value Anomoeoneis brachysira 39.2 Cymbella microcephala 12.5 Desmogonium rabenhorstianum 33.5 Encyonema silesiacum 24.4 Eunotia flexuosa 12.5 Eunotia glacialis 17.0 Eunotia intermedia 28.1 Eunotia naegelii 59.2 Eunotia pectinalis undulate 26.9 Eunotia rhomboidea 45.9 Frustulia rhomboids 41.6 Frustulia rhomboides crassinervia 18.7 Navicula capitatoradiata 11.5 Navicula subtilissima 12.5 Nitzschia nana 17.0 Nitzschia paleacea 18.7 Pinnularia streptoraphe 18.7 Rhopalodia gibba 18.7 All reported indicator values were significant (p<0.10).

PAGE 64

50 01020304050607080900246LDI%Sensitive Indicator Species 8 Figure 3-3. Percent diatom sensitive indicator species decreased with increasing development intensity (LDI). Autecological metrics Between 56-69% of diatoms identified received scores based on established autecological relationships ( Bahls 1993 ; van Dam et al. 1994 ). Five metrics based on scoring diatoms from a coded checklist describing their autecology were incorporated in the diatom FWCI, including the proportion of diatoms in Pollution Class 1, Nitrogen Uptake Metabolism Class 3, Saprobity Class 4, pH Class 3, and Dissolved Oxygen Requirement Class 1. Figure 3-4 shows that the proportion of diatoms in Pollution Class 1 increased with increasing development intensity in the surrounding landscape. Diatoms in Pollution Class 1 were very tolerant to pollution, as compared to Pollution Class 2 (moderately tolerant) or Pollution Class 3 (sensitive to pollution) ( Bahls 1993 ). In general a background level of approximately 10% of the diatoms belonging to Pollution Class 1 distinguishes low (LDI < 2.0) and high (LDI 2.0) LDI wetlands. 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 dispersed from a pullet farm operation. Figure 3-5 shows that the proportion of diatoms in Nitrogen Uptake Metabolism Class 3 increased with increasing development intensity in the surrounding landscape. Membership in Nitrogen Uptake Metabolism Class 3 was defined by facultative nitrogenheterotrophic taxa, those needing periodically elevated concentrations of organically bound nitrogen ( van Dam et al. 1994 ). Four wetlands in the low LDI (LDI < 2.0) group that had higher than anticipated percent diatoms in Nitrogen Uptake Metabolism Class 3 included SR4, SR5, SR6, and SU3. The three southern reference wetlands (SR4, SR5, and SR6) were located within state or federal lands protected as part of the Florida Everglades. The urban outlier, SU3 was surrounded by nearly 100 m of marsh that has received nutrient enriched waters since the mid 1970s.

PAGE 65

51 0204060801000246LDI%Pollution Class 1 8 Figure 3-4. Pollution Tolerance Class 1 diatoms increased with increasing development intensity (LDI). 0204060800246LDI%Nitrogen Class 3 8 Figure 3-5. Nitrogen Uptake Metabolism Class 3 diatoms increased with increasing development intensity (LDI).

PAGE 66

52 The percent of diatoms in Saprobity Class 4 increased with increasing landscape development intensity (Figure 3-6). Diatoms characterized as belonging to Saprobity Class 4 included mesoto poly-saprobous species (that inhabit aquatic environments with an oxygen saturation between 10-25% and a biological oxygen demand (BOD520) of 13-22 mg/L) ( van Dam et al. 1994 ). Eighty-five percent of the wetlands in the low LDI group had less than 5% of diatoms in Saprobity Class 4. CA3 (a wetland receiving waters dispersed from a pullet farm operation) hosted the greatest percent of diatoms in Saprobity Class 4 (saprobity class 4 = 52%). Figure 3-7 shows that the percent of diatoms in pH Class 3 increased with increasing LDI. Diatoms categorized in pH Class 3 were described as circumneutral (mainly occurring at pH values of approximately 7) ( van Dam et al. 1994 ). In the low LDI group, the wetland with the highest percent diatoms in pH Class 3 was SU4, a wetland surrounded by low intensity residential, previously in pasture (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 landscape (Figure 3-8). Six wetlands with low LDI scores (LDI < 2.0) hosted a smaller percent of diatoms requiring high dissolved oxygen than anticipated. These were the same six wetlands that hosted a higher percent pH Class 3 diatoms (Figure 3-7). Diatom Wetland Condition Index The seven metrics described above were scored and added together to create the diatom FWCI. Appendix F provides detailed instructions on metric scoring. Figure 3-9 shows the relationship between the diatom FWCI and LDI. Potential scores for the diatom FWCI 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 received diatom FWCI scores 2 points lower than SR2, with a score of 67 at both NR3 and SR1. Diatom FWCI scores varied regionally, with the highest scores in each wetland region including PR4 (65), NR3 (67), CR6 (65), and SR2 (69), in the panhandle, north, central, and south wetland regions, respectively. The lowest scores in the panhandle and north wetland regions were for urban wetlands embedded in residential land use, including PU4 (11) and NU2 (24). Wetlands surrounded by agricultural land uses received the lowest scores in the central and south wetland regions, including CA3 (8) and SA4 (16). The diatom FWCI was robustly correlated with the LDI index (Spearman correlation |r| = 0.64, p < 0.001). A Kruskal-Wallis test between median diatom FWCI score values suggested a significant difference (H = 20.7, p < 0.001) among wetlands in the three a priori land use categories.

PAGE 67

53 02040600246LDI%Saprobity Class 4 8 Figure 3-6. Saprobity Class 4 diatoms increased with increasing development intensity (LDI). 0204060801000246LDI%pH Class 3 8 Figure 3-7. The pH Class 3 diatoms increased with increasing development intensity (LDI).

PAGE 68

54 0204060801000246LDI%Dissolved Oxygen Class 1 8 Figure 3-8. Dissolved Oxygen Class 1 diatoms decreased with increasing development intensity (LDI). 01020304050607002468LDIDiatom WC I Reference Agricultural Urban Figure 3-9. Diatom FWCI scores decreased with increasing development intensity (LDI). Sample wetlands are designated by a priori land use category: reference, agricultural, or urban.

PAGE 69

55 Cluster Analysis Cluster analysis determined four categories based on diatom community composition. Using site descriptions, clusters were explained by wetland regions and a priori land use categories including: 1: wetlands in the panhandle to central wetland regions surrounded by low development land use activities; 2: wetlands occurring in mixed wetland regions surrounded by low development land use activities; 3: wetlands within the southern Everglades; and 4: wetlands within mixed wetland regions surrounded by high development land use activities. Figure 3-10 shows that based on the diatom FWCI scores, clusters 1 and 2 were not significantly different from one another, but were significantly different from both cluster 3 and cluster 4 (p<0.05). Clusters 3 and 4 were significantly different from each other. Table 3-12 provides means and standard deviations for diatom FWCI scores and LDI of the four diatom based clusters. Cluster 4 had significantly different diatom FWCI and LDI scores than all other clusters. Figure 3-10. Diatom FWCI scores for wetland clusters based on diatom community composition. Boxes represent the interquartile range; solid circles represent the mean; middle lines represent the median; whiskers represent the range; asterisks represent outliers (> standard deviations). Clusters with similar letters were not significantly different (Fishers LSD, p<0.05). 010203040506070 Cluster 1 D-WCI Cluster 4Cluster 3Cluster 2 aabc Diatom FWCI

PAGE 70

56 Table 3-12. Mean diatom FWCI and LDI values for wetland clusters based on diatom community composition. Cluster* 1 2 3 4 Diatom FWCI 57.6 (6.4) a 52.9 (11.6) a 38.1 (4.4) b 24.1 (11.7) c LDI 2.9 (2.1) a 2.8 (2.0) a 2.4 (1.9) a 5.3 (1.4) b Clusters with similar letters were not significantly different (Fishers LSD, p<0.05). Values represent mean (standard deviation).

PAGE 71

57 Macrophyte Assemblage Statewide, 118 wetlands were sampled with 605 species, representing 323 genera and 126 families identified. The most abundant species was Taxodium ascendens 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. 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 the vegetation quadrats at only one wetland. In the panhandle wetland region, 28 wetlands were sampled hosting 328 species, representing 191 genera and 90 families. In the north wetland region 31 wetlands were sampled with 306 species (180 genera and 89 families) encountered. The central wetland region included 31 wetlands with 329 species (202 genera and 94 families) sampled. The south wetland region had 28 sample wetlands 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 diversity (H), and Whittakers beta diversity (W) were calculated based on the macrophyte assemblage for each sample wetland ( Appendix E). 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 were 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 (surrounded by pine flatwoods), with 11 and 15 species estimated with first 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 wetland on a private conservation tract), to 0.93 at PU4 (surrounded by residential land use including housing and a community park). Shannon diversity ranged from 1.8 to 3.9 at two agricultural wetlands, NA3 (surrounded by silviculture) and NA12 (surrounded by pasture and row crops), respectively, similar to species richness. Whittakers beta diversity ranged from a low of 0.2 at PU9 (surrounded by residential land use), to a high of 9.4 at CR5 (located in a fragmented state park within highly developed Seminole County). Table 3-13 summarizes comparisons of mean richness and diversity calculations by a priori land use category. Agricultural wetlands had the greatest species richness followed closely by urban wetlands. This same trend was evident for species evenness, with agricultural wetlands having greater species evenness. Diversity indices yielded similar results, with reference wetlands having lower Shannon diversity and Whittakers beta diversity then both agricultural and urban wetlands. Beta and gamma diversity were calculated for a priori land use categories, with urban wetlands having the highest beta diversity and agricultural wetlands the highest gamma diversity. Only Whitakers beta

PAGE 72

58 Table 3-13. Mean macrophyte richness, evenness, and diversity among a priori land use categories. Reference Agricultural Urban Species richness (R) 32 (11) a 37 (14) a 36 (10) a Species evenness (E) 0.85 (0.04) a 0.87 (0.04) a 0.86 (0.03) a Shannon diversity (H) 2.9 (0.3) a 3.1 (0.4) a 3.1 (0.3) a Whittaker's Beta diversity (W) 4.0 (2.9) a 4.8 (1.7) b 4.3 (1.7) ab Beta diversity 9.5 10.4 10.5 Gamma diversity 304 383 378 Categories with similar letters were not significantly different (Fisher's LSD, =0.05). Values represent mean (standard deviation). diversity was significantly different among a priori land use categories (Fishers LSD pair wise comparison, = 0.05). Species richness and Whittakers beta diversity were not significantly different between low (LDI < 2.0) and high (LDI 2.0) LDI groups (Table 3-14); whereas, species evenness and Shannon diversity were significantly different (p < 0.05) between LDI groups (Mann-Whitney U-Test). Compositional Analysis MRPP was calculated across all wetland regions (panhandle versus north versus central versus south) as well as for multiple pair wise comparisons (panhandle versus north, panhandle versus central, panhandle versus south, north versus central, north versus south, and central versus south). Table 3-15 shows the results for the MRPP tests, including the test statistic (T), chance-corrected within-group agreement (A), and significance value (p). Only two of the MRPP comparisons (agricultural wetlands in the panhandle and north, and central and south wetland regions) were not significant at the Table 3-14. Mean macrophyte richness, evenness, 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 Values represent mean (standard deviation).

PAGE 73

59 Table 3-15. Macrophyte community composition similarity among wetland regions with MRPP. Sites (n) T ^ A` p # All wetlands All regions (P vs N vs C vs S) 118 -26.8 0.06 0.00* Panhandle vs north 59 -7.1 0.03 0.00* Panhandle vs central 59 -13.1 0.04 0.00* Panhandle vs south 56 -23.6 0.09 0.00* North vs central 62 -7.9 0.03 0.00* North vs south 59 -24.2 0.09 0.00* Central vs south 59 -11.8 0.04 0.00* Reference wetlands All regions (P vs N vs C vs S) 37 -12.5 0.12 0.00* Panhandle vs north 17 -5.2 0.08 0.00* Panhandle vs central 19 -6.8 0.09 0.00* Panhandle vs south 17 -8.1 0.14 0.00* North vs central 20 -4.7 0.06 0.00* North vs south 18 -8.5 0.14 0.00* Central vs south 20 -7.1 0.07 0.00* Agricultural wetlands All regions (P vs N vs C vs S) 40 -6.6 0.05 0.00* Panhandle vs north 22 -0.7 0.01 0.21* Panhandle vs central 19 -2.8 0.03 0.01* Panhandle vs south 19 -8.1 0.10 0.00* North vs central 21 -2.0 0.02 0.04* North vs south 21 -7.7 0.09 0.00* Central vs south 18 -1.4 0.02 0.09* Urban wetlands All regions (P vs N vs C vs S) 41 -15.3 0.14 0.00* Panhandle vs north 20 -6.0 0.07 0.00* Panhandle vs central 21 -9.0 0.10 0.00* Panhandle vs south 20 -11.1 0.17 0.00* North vs central 21 -3.3 0.03 0.00* North vs south 20 -10.5 0.15 0.00* Central vs south 21 -8.6 0.09 0.00* *A high |T| value and significant p-value (p<0.05) suggests a difference in species composition. T ^ = the MRPP test statistic A` = the chance corrected within-group agreement p # = the significance value.

PAGE 74

60 = 0.05 level, suggesting that there were regionally significant differences among species composition across wetland regions and within a priori land use categories. Community Composition Macrophyte community composition was summarized in two separate NMDS ordinations to relate changes in macrophyte community composition with environmental variables. Figure 3-11 shows a two dimensional bi-plot of the NMDS axes used to explore the dissimilarities of macrophyte community composition with overlays of significant environmental variables, including soil TP (log), LDI, latitude, and longitude. Table 3-16 provides the Pearson correlation coefficients between environmental variables and NMDS ordination axes. A three dimensional solution was constructed with an overall stress of 20.8 and a final stability of 0.06, which is a fairly high stress value but considered useful for ordinations with community data sets ( Kruskal 1964 ; Clarke 1993 ; McCune and Grace 2002 ). Axis 1 explained 29.1% variance and was correlated with latitude and longitude; axis 2 explained 34.2% additional variance and was correlated with LDI and soil TP (log). Axis 3 explained an additional 12.1% of the variance and was not significantly correlated with any of the measured soil parameters, LDI, latitude, or longitude. A second NMS ordination was completed using the macrophyte species composition at the 75 sample wetlands with measured water chemical and physical parameters. Figure 3-12 shows the bi-plot from the NMS ordination. The final stress was 16.6 with a final instability of 0.004. The ordination 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 environmental parameters and the three ordination axes. Axis 1 explained 31.3% of the variance and was correlated with latitude and longitude. Axis 2 was correlated with LDI, water pH, water TP (log), soil TP (log), and DO (log), and explained 25.7% additional variance. Axis 3 explained an additional 20.4% of the variance and was correlated with water pH and soil moisture (arcsine square root). Water ammonia-nitrogen concentration (log), water nitrate/nitrite-nitrogen concentration (log), water TKN-nitrogen concentration (log), water temperature (log), water color (log), and water turbidity (log) were not strongly correlated with the NMDS ordination axes.

PAGE 75

61 Phosphorus Latitude Longitude LDI Axis 1: 29.1% A xis 2: 34.2% Minimally ImpairedAgriculture ImpairedUrban Impaired RefenceAgriculturalUrban Log( ) Figure 3-11. NMDS ordination bi-plot of 118 sample wetlands in macrophyte species space with an overlay of environmental parameters. Latitude, longitude, LDI, and log(soil TP), shown as radiating vectors, were significantly correlated with the 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% additional variance, and axis 3 (not shown) represented an additional 12.1% variance.

PAGE 76

62 Table 3-16. Pearson correlation coefficients between environmental variables and NMDS axes based on macrophyte community composition at 118 wetlands. Axis 1 Axis 2 Axis 3 Incremental r 2 29.1% 34.2% 12.1% Cumulative r 2 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 Arcsine 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 Metric Selection The pool of potential candidate metrics was streamlined to reduce the redundancy of selected metrics. If two metrics were calculated from the same set of taxa representing redundant metrics only one metric was included (i.e. metrics based on the same category were not included, for example if percent Obligate Species was included, no other metric based on wetland indicator status). Six metrics that were significant for the Spearmans correlation coefficient (|r|>0.50, p<0.001) (Table 3-18), showed visually distinguishable trends with LDI, and were significant for the Mann-Whitney U-test between LDI groups (p<0.001) (Table 3-19) were selected for inclusion in the macrophyte FWCI. Metrics selected for the FWCI were calculated as the percent of species richness using presence/absence data. Metrics selected for inclusion were percent Tolerant Indicator Species, percent Sensitive Indicator Species; Floristic Quality Assessment Index (FQAI); percent Exotic Species; percent Native Perennial Species; and percent Wetland Status Species (including both obligate and facultative wetland species). The percent of Tolerant Indicator Species and percent Exotic Species increased with increasing development intensity. Whereas, percent Sensitive Indicator Species, FQAI, percent Native Perennial Species, and percent Wetland Status Species values decreased with increased landscape development. Table 3-19 shows that all six metrics significantly differentiated between low (LDI < 2.0) and high (LDI 2.0) LDI groups. Tolerance metrics Macrophyte Tolerant and Sensitive Indicator Species lists were determined using Indicator Species Analysis (ISA) for each of the four wetland regions and statewide using species-level abundance data. Macrophyte Tolerant Indicator Species were established at an LDI break of 4.0. Table 3-20 provides a list of Tolerant Indicator Species comparing regional and statewide analyses. The same random seed number was used for each ISA for statewide and regional analyses.

PAGE 77

63 ReferenceAgriculturalUrban LDI Latitude Longitude Log(DO) Water pH Log(TP) Axis 2: 25.7%Axis 1: 31.1 % ReferenceAgricultura l Urban Figure 3-12. NMDS ordination bi-plot of 75 sample wetlands in macrophyte species space with an overlay of environmental parameters. Latitude, longitude, LDI, TP (log), water pH, and DO (log), shown as radiating vectors, were significantly correlated with the NMDS axes. Vector length represents the strength of the correlation, and the angle represents the direction of maximum change. Axis 1 explained 31.3% variance, axis 2 explained 25.7% additional variance, and axis 3 (not shown) represented an additional 20.4% variance.

PAGE 78

64 Table 3-17. Pearson correlations between environmental variables and NMDS axes based on macrophyte community composition at 75 wetlands. Axis 1 Axis 2 Axis 3 Incremental r 2 31.3% 25.7% 20.4% Cumulative r 2 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 DO (log) 0.02 0.20 0.05 Temperature (log) 0.02 0.05 0.00 Color (log) 0.14 0.00 0.05 Turbidity (log) 0.03 0.06 0.03 pH 0.15 0.26 0.35 Ammonia-N (log) 0.01 0.04 0.00 Nitrate/nitrite-N (log) 0.06 0.00 0.06 TKN (log) 0.02 0.01 0.01 TP (log) 0.11 0.25 0.06 Soil parameters Moisture (arcsine sqrt) 0.00 0.00 0.24 TKN (log) 0.00 0.00 0.16 TP (log) 0.09 0.23 0.01 Table 3-18. Spearman correlations between six statewide macrophyte metrics with LDI. Metric Spearmans r Percent Tolerant Indicator Species 0.75 Percent Sensitive Indicator Species -0.66 FQAI -0.71 Percent Exotic Species 0.65 Percent Native Perennial Species -0.63 Percent Wetland Status Species -0.55 All correlations were significant (p < 0.01).

PAGE 79

65 Table 3-19. Comparisons among six statewide macrophyte metrics between LDI groups. Metric Low LDI High LDI W ^ p` Percent Tolerant Indicator Species 7.8 (7.8) 31.2 (14.7) 1116.5 <0.001 Percent Sensitive Indicator Species 39.5 (16.7) 9.4 (10.1) 3665.0 <0.001 FQAI 4.81 (0.62) 3.62 (0.80) 3771.0 <0.001 Percent Exotic Species 3.0 (3.6) 14.3 (10.6) 1379.0 <0.001 Percent Native Perennial Species 92.7 (4.5) 79.7 (12.0) 3453.0 <0.001 Percent Wetland Status Species 72.0 (9.8) 54.1 (12.5) 3612.0 <0.001 Values represent the mean (standard deviation) W ^ = the Mann-Whitney U-Test statistic p` = the significance value In total the ISA reported 69 statewide Tolerant Indicator Species, and less for each wetland region with 7, 28, 7, and 12 for the panhandle, north, central, and south wetland regions, respectively. The statewide ISA produced 69 Tolerant Indicator Species with an additional seven species included on regional lists, but not the statewide list. No species occurred on the Tolerant Indicator Species lists statewide and in all four wetland regions. Three species occurred on the statewide Tolerant Indicator Species list and on three of the wetland region lists, including Commelina diffusa (north, central, and south), Cynodon dactylon (panhandle, north, central), and Diodia virginiana (panhandle, north, south). Seven species occurred on the Tolerant Indicator Species list statewide and in two wetland regions, and 24 were listed both statewide and in one wetland region. Thirty-five of the 69 statewide Tolerant Indicator Species (51%) were not listed in any wetland region. In total, the wetland regions shared more than twothirds of their listed species with the statewide list: panhandle (100%), north (93%), central (86%), and south (67%). Two species were unique to the north, one to the central, and four to the south wetland region Tolerant Indicator Species lists. Figure 3-13 shows the scatterplots of the percent Tolerant Indicator Species versus LDI. The percent Tolerant Indicator Species increased with increasing development intensity. For the statewide Tolerant Indicator Species, CA2 (surrounded by pasture; tolerant = 72%) had the highest percent statewide Tolerant Indicator Species. Three wetlands in the low LDI group with high percents Tolerant Indicator Species included SA8 (a suburban park receiving run-off from a pasture; 32%), CA8 (in a preserve but receiving run-off from dairy activities; 26%), and PR7 (located on a private conservation tract; 25%). In the regional ISA calculations, a wetland in the north wetland region, NA1 (surrounded by pasture; north tolerant = 47%) had the largest percent Tolerant Indicator Species. Macrophyte Sensitive Indicator Species were established at an LDI break of 2.0. Table 3-21 provides a list of statewide and regional Sensitive Indicator Species. The statewide Sensitive Indicator Species list included 61 species of which 16 were not listed in any of the wetland regions. Two species occurred on all statewide and wetland region lists, including Eriocaulon decangulare and Panicum erectifolium. Similarly, six species occurred on the statewide list and three wetland region lists including Andropogon

PAGE 80

66 Table 3-20. Statewide and regional macrophyte Tolerant Indicator Species. Statewide Panhandle North Central South No. of Tolerant Species 69 7 28 7 12 Acer rubrum 28.5 53.7 Alternanthera philoxeroides 11.3 Amaranthus spinosus 10.9 21.4 Ampelopsis arborea 18.6 Aster carolinianus 9.5 Axonopus fissifolius 9.1 Blechnum serrulatum 42.4 Boehmeria cylindrica 37.9 57.9 43.7 Carex longii 30.6 54.3 Centella asiatica 52.0 Colocasia esculenta 5.5 Commelina diffusa 44.8 50.0 73.3 44.1 Cuphea carthagenensis 28.0 28.6 50.0 Cynodon dactylon 29.4 35.7 35.7 28.1 Cyperus croceus 9.1 Cyperus lanceolatus 7.3 Cyperus polystachyos 13.1 Cyperus retrorsus 18.2 35.7 20.0 Cyperus virens 11.3 42.9 Digitaria ciliaris 9.1 Diodia virginiana 37.2 50.0 40.5 50.0 Dioscorea bulbifera 7.3 Echinochloa colona 5.5 Eclipta prostrata 17.9 Eupatorium capillifolium 37.9 44.1 Galium hispidulum 5.5 Galium tinctorium 22.6 51.8 26.7 Hymenachne amplexicaulis 9.1 25.0 Hypericum mutilum 28.6 Juncus effusus 22.1 42.9 Kyllinga brevifolia 7.3 Leersia hexandra 7.3 20.0 Lepidium virginicum 5.5 Ligustrum sinense 10.2 Lonicera japonica 12.7 Ludwigia peruviana 17.7 Ludwigia repens 14.3 21.4 Luziola fluitans 5.5 Lygodium japonicum 11.9

PAGE 81

67 Table 3-20. Continued. Statewide Panhandle North Central South Melaleuca quinquenervia 36.2 Melothria pendula 20.8 28.6 Micranthemum umbrosum 5.5 Momordica charantia 10.2 Oxalis corniculata 18.5 28.6 Parthenocissus quinquefolia 34.6 40.5 52.0 Paspalum notatum 24.4 30.7 Paspalum urvillei 18.5 35.7 28.6 Phyla nodiflora 15.2 32.1 Phyllanthus urinaria 16.7 21.4 Phytolacca americana 26.1 36.7 Polygonum hydropiperoides 21.1 57.1 Polygonum punctatum 28.0 42.9 44.7 Polypremum procumbens 9.5 Proserpinaca palustris 5.5 Richardia brasiliensis 5.5 Rubus argutus 25.1 47.1 Rubus trivialis 17.7 Sabal palmetto 66.7 Sacciolepis indica 7.3 Sambucus canadensis 24.6 30.7 Sapium sebiferum 19.4 Saururus cernuus 40.5 Senna obtusifolia 9.1 Sesbania vesicaria 5.5 Setaria parviflora 7.3 Sida rhombifolia 20.8 35.7 Smilax pumila 7.3 21.4 Solanum carolinense 9.1 Solidago stricta 5.5 Sporobolus indicus 7.3 Stenotaphrum secundatum 9.5 26.7 Toxicodendron radicans 38.0 Trifolium repens 7.3 21.4 Urena lobata 52.0 Vitis rotundifolia 36.7 48.5 Wedelia trilobata 5.5 25.0 Values represent Indicator Values. All reported Indicator Values were significant (p < 0.10).

PAGE 82

68 0102030405002%North Tolerant 468LDI 0102030405002468LDI%Central Toleran t 0102030405002468LDI%South Tolerant 0204060800246LDI%Statewide Tolerant 8 Panhandle North Central South 0102030405002468LDI%Panhandle Tolerant A. B. C. D. E. Figure 3-13. Macrophyte Tolerant Indicator Species increased with increasing development intensity (LDI). A) panhandle Tolerant Indicator Species at panhandle study wetlands. B) north Tolerant 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 118 study wetlands.

PAGE 83

69 Table 3-21. Statewide and regional macrophyte Sensitive Indicator Species. Statewide Panhandle North Central South No. of Sensitive Species 61 28 19 24 13 Amphicarpum muhlenbergianum 19.3 52.4 Andropogon virginicus 46.9 62.5 66.7 66.5 Aristida beyrichiana 8.6 45.5 Aristida patula 22.2 Aristida purpurascens 32.9 50.0 22.2 53.5 Carex verrucosa 25.0 Cladium jamaicense 22.3 60.6 Coelorachis rugosa 7.3 Cyperus haspan 11.0 33.3 Drosera brevifolia 7.3 Erianthus giganteus 17.8 33.3 Eriocaulon compressum 17.1 25.0 33.3 Eriocaulon decangulare 37.8 37.5 22.2 53.5 33.3 Eupatorium leptophyllum 14.6 25.0 Eupatorium mohrii 8.6 Fuirena scirpoidea 26.9 25.0 25.0 36.2 Gaylussacia frondosa 25.0 Gratiola ramose 18.3 41.7 Hypericum chapmanii 8.6 45.5 Hypericum fasciculatum 38.2 53.5 44.4 Hypericum myrtifolium 17.8 47.7 Hyptis alata 10.8 Ilex glabra 46.2 53.9 78.6 45.9 Ilex myrtifolia 17.1 71.2 Ipomoea sagittata 11.0 Lachnanthes caroliniana 39.6 71.2 40.2 Lachnocaulon anceps 25.0 Lobelia floridana 25.0 Lophiola aurea 12.2 62.5 Ludwigia linifolia 7.3 Lycopodiella alopecuroides 9.8 25.0 22.2 Lyonia lucida 24.4 Nymphaea odorata 6.2 22.2 Nymphoides aquatica 12.2 Panicum ensifolium 13.1 Panicum erectifolium 41.5 50.0 33.3 45.2 36.2 Panicum hemitomon 40.4 79.2 Panicum rigidulum 17.1 25.0 Panicum tenerum 14.6 33.3

PAGE 84

70 Table 3-21. Continued. Statewide Panhandle North Central South Pinus elliottii 33.9 56.2 37.0 25.0 Pinus palustris 8.6 Pluchea foetida 10.1 25.0 Pluchea rosea 15.5 33.3 Polygala cymosa 28.0 37.5 29.3 33.3 Polygala lutea 7.3 Proserpinaca pectinata 12.4 Rhexia alifanus 13.4 70.3 Rhexia lutea 13.4 45.5 22.2 Rhexia mariana 23.1 47.7 45.2 Rhexia petiolata 25.0 Rhus copallinum 33.1 Rhynchospora corniculata 25.0 Rhynchospora filifolia 37.5 Rhynchospora inundata 17.2 22.2 Rhynchospora microcarpa 28.8 Rhynchospora wrightiana 22.2 Sabatia bartramii 7.3 Sagittaria graminea 13.9 41.7 Sagittaria lancifolia 17.2 36.2 Salix caroliniana 44.4 Sarracenia minor 22.2 Scleria baldwinii 7.3 Scleria Georgiana 7.3 Scleria triglomerata 7.3 25.0 Serenoa repens 22.3 37.5 57.6 Spartina bakeri 7.3 25.0 Stillingia aquatica 13.4 36.2 Syngonanthus flavidulus 12.2 22.2 Utricularia purpurea 8.6 25.0 Vaccinium corymbosum 20.9 33.1 33.3 Xyris ambigua 11.0 Xyris caroliniana 7.3 25.0 Xyris elliottii 18.3 50.0 Xyris jupicai 8.6 25.0 Values represent Indicator Values. All reported Indicator Values were significant (p < 0.10).

PAGE 85

71 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 four wetland regions shared over three-quarters of their species with the statewide list (panhandle = 79%, north = 84%, central = 92%, and south = 85%). Six species were unique to the panhandle Sensitive Indicator Species analysis, three to the north, two to the central, and two to the south wetland region. Figure 3-14 shows that the percent Sensitive Indicator Species, statewide and regionally, decreased with increasing development intensity. All of the indicator species metrics were significantly correlated with the LDI index. Table 3-22 shows Spearman correlations calculated with both statewide and regional Indicator Species lists for each region. Regional Indicator Species metrics had a stronger correlation value, though all metrics were significantly correlated with LDI (p < 0.01). Two exceptions included the panhandle Tolerant Indicator Species correlations (statewide Tolerant Indicator Species, r = 0.73, panhandle Tolerant Indicator Species, r = 0.72) and central (statewide Tolerant Indicator Species, r = 0.74, central Tolerant Indicator Species, r = 0.68). Shrub and tree species were included in the ISA for both Tolerant and Sensitive Indicator Species 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 isolated forested wetlands. Excluding the tree and shrub layers would underscore their importance in providing structure for the isolated depressional forested wetlands. However, trees comprised only a small percentage of the Tolerant and Sensitive Indicator Species lists (Tables 3-20 and 3-21). Three percent of the statewide Tolerant Indicator Species were trees, 9% were shrubs, 14% vines, and 74% herbaceous (including herbs, sedges, grasses, etc.). The two statewide tree Tolerant Indicator Species included the hardwood Acer rubrum and exotic Sapium sebiferum (Table 3-20). The six statewide shrub Tolerant Indicator Species were of the genera Aster (a climbing species), an exotic Ligustrum, an exotic Ludwigia, Rubus, and Sambucus. Of the statewide Sensitive Indicator Species, 3% were trees, 15% shrubs, 2% vines, and 80% herbaceous. Floristic Quality Assessment Index metric Figure 3-15 shows wetland FQAI scores decreased with increasing landscape development intensity. Wetlands in the panhandle (maximum modified FQI = 6.25) and north (maximum FQAI = 5.95) wetland regions had higher FQAI scores versus wetlands in the central (maximum FQAI = 4.93) and south (maximum FQAI = 5.24) wetland regions. Statewide the FQAI was significantly correlated with LDI (|r| = 0.71, p < 0.001; Table 3-18); and there was a significant difference between the mean FQAI scores between low (LDI < 2.0) and high (LDI 2.0) LDI groups (W = 3771.0, p = <0.001; Table 3-19). CU6, an urban wetland in the high LDI group in the central wetland region surrounded by recently developed and under construction sections of residential land use (FQAI = 4.43; LDI = 7.2), had a higher FQAI score than anticipated. The wetland with the lowest FQAI score was NA1, an agricultural wetland surrounded by pasture (FQAI = 2.06).

PAGE 86

72 01020304050607002468LDI%Panhandle Sensitive 01020304050607002468LDI%North Sensitive 01020304050607002468LDI%Central Sensitiv e 01020304050607002468LDI%South Sensitive 0102030405060700246LDI%Statewide Sensitive 8 Panhandle North Central South Figure 3-14. Macrophyte Sensitive Indicator Species decreased with increasing development intensity (LDI). A) panhandle Sensitive Indicator Species at panhandle study wetlands. B) north Sensitive 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 118 study wetlands. A. B. D. C. E.

PAGE 87

73 Table 3-22. The percent of statewide and regional macrophyte Indicator Species metrics were significantly correlated with LDI (p < 0.01). %Statewide ISA %Regional ISA Wetland Region n= Spearmans r Spearmans r Sta tewide Tolerant Indicator Species 118 0.75 Sensitive Indicator Species 118 -0.66 Pan handle North Central South Tolerant Indicator Species 28 0.73 0.72 Sensitive Indicator Species 28 -0.66 -0.68 Tolerant Indicator Species 31 0.79 0.80 Sensitive Indicator Species 31 -0.76 -0.78 Tolerant Indicator Species 31 0.74 0.68 Sensitive Indicator Species 31 -0.67 -0.72 Tolerant Indicator Species 28 0.78 0.86 Sensitive Indicator Species 28 -0.60 -0.71 Values represent Spearmans correlation coefficients, all reported correlations significant (P < 0.01). Exotic Species metric Statewide, the percent Exotic Species was significantly correlated with development intensity in the surrounding landscape (r = 0.65, p < 0.001; Table 3-18). Figure 3-16 shows that the percent of exotic species increased with increasing LDI in each wetland region. The north wetland region hosted the wetland with the greatest percent Exotic Species, NA1 with 52.6% Exotic Species. NA1 was surrounded by a research facility growing experimental pasture grass species, biasing 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 irregularity in south wetland region low LDI group was SU4 a wetland surrounded by low density residential land use, previously embedded in pasture with 18.4% Exotic Species. All remaining wetlands in the low LDI group (n = 40) had less than 10% Exotic Species. Statewide, the percent Exotic Species was significantly different between low and high LDI groups (W = 1379.0, p < 0.001; Table 3-19). Table 3-23 lists the 113 Exotic Species encountered throughout Florida and identifies the wetland region(s) in which each species was found. Only six Exotic Species were found in all four wetland regions including Commelina diffusa, Cuphea carthagenensis, Cynodon dactylon, Kyllinga brevifolia, Ludwigia peruviana, and Paspalum notatum. Fourteen Exotic Species occurred in three of the four wetland regions.

PAGE 88

74 23456702468LDIPanhandle Modified FQI 23456702468LDINorth Modified FQI 23456702468LDICentral Modified FQI 23456702468LDISouth Modified FQ I Figure 3-15. FQAI scores decreased with increasing development intensity (LDI). A) FQAI scores at panhandle study wetlands. B) FQAI scores at north study wetlands. C) FQAI scores at central study wetlands. D) FQAI scores at south study wetlands. A. B. C. D.

PAGE 89

75 010203040506002468LDI%Panhandle Exotic. 010203040506002468LDI%North Exotic.... 010203040506002468LDI%Central Exotic. 010203040506002468LDI%South Exotic. Figure 3-16. Exotic Species increased with increasing development intensity (LDI). A) Exotic Species at panhandle study wetlands. B) Exotic Species at north study wetlands. C) Exotic Species at central study wetlands. D) Exotic Species at south study wetlands. A. B. D. C.

PAGE 90

76 Table 3-23. Macrophyte Exotic Species identified at 118 study wetlands throughout the four Florida wetland regions. Exotic Species P N C S Exotic 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 91

77 Table 3-23. Continued. Exotic Species P N C S Exotic Species P N C S 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 Ecoregions determined by Lane (2000): P = panhandle wetland region; N = north wetland region; C = central wetland region; S = south wetland region. Native Perennial Species metric Of the 605 macrophyte species identified, 427 (71%) were classified as native perennials. Figure 3-17 shows that the percent Native Perennial Species decreased with increasing development intensity. The Native Perennial Species metric was significantly correlated with LDI (Spearman |r| = 0.63, p < 0.001; Table 3-18). 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). Wetland Status Species metric Fifty-six percent of the macrophyte species identified were included in the Wetland Status Species metric, including 160 species designated as obligate and 180 species designated as facultative wetland species. Figure 3-18 shows that the percent Wetland Status Species decreased with increasing development intensity in each wetland region. 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 the LDI index (Spearman |r| = 0.55, p < 0.001; Table 3-18).

PAGE 92

78 40506070809010002468LDI%Panhandle Native Perennia l ,,, 40506070809010002468LDI%North Native Perennial 40506070809010002468LDI%Central Native Perennial 40506070809010002468LDI%South Native Perennial C. D. Figure 3-17. The percent 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. A. B.

PAGE 93

79 2040608010002468LDI%Panhandle Wetland 2040608010002468LDI%North Wetland 2040608010002468LDI%Central Wetland 2040608010002468LDI%South Wetland C. D. Figure 3-18. The percent Wetland Status Species decreased with increasing development intensity (LDI). This trend was consistent for the (A) panhandle, (B) north, (C) central, and (D) south wetland regions. A. B

PAGE 94

80 Macrophyte Florida Wetland Condition Index The six metrics described above were included in the macrophyte FWCI. Appendix F provides detailed instructions on metric scoring. Figure 3-19 shows that both statewide and regional macrophyte FWCI scores decrease with increasing development intensity. Table 3-24 compares the overall macrophyte FWCI calculated statewide and regionally for the low LDI group (LDI < 2.0). A comparable statewide macrophyte FWCI should equally score reference wetlands in each wetland region. However, the south wetland region had significantly different overall macrophyte FWCI scores for the low LDI group compared to the panhandle, north, and central wetland regions. When calculated statewide, five of the six metrics had one or more wetland region with significantly different metric scores. The north and central wetland regions had significantly different scores for the statewide Tolerant Indicator Species metric, whereas the north and south wetland regions had significantly different scores for the statewide Sensitive Indicator Species metric. The panhandle and north wetland regions were not significantly different from each other, but were significantly different from the central and south wetland regions for FQAI scores; suggesting that the panhandle and north wetland regions hosted more species with a narrower set of ecological conditions found in reference wetlands. The south wetland region had significantly different statewide Exotic Species metric scores than the other three wetland regions. The scores for the statewide Native Perennial Species metric were significantly different for the north wetland region. When the macrophyte FWCI was scored regionally, there was not a significant difference between mean scores for the low LDI group (Table 3-24). The only regionally scored metric with significantly different mean scores for the low LDI group was the regional Tolerant Indicator Species metric for the south wetland region which was only significantly different from the panhandle wetland region. Table 3-25 shows similar results for the high LDI group (LDI 2.0). For the statewide macrophyte FWCI wetlands in the north wetland region had significantly different macrophyte FWCI scores, suggesting the north wetland region high LDI wetlands had higher ecological integrity than wetlands of other wetland regions. Five of the six metrics calculated statewide had at least one wetland region with significantly different mean scores in the high LDI group. Only the Wetland Status Species metric did not have significantly different scores for both statewide and regional calculations among wetland regions. For the regional macrophyte FWCI calculations, the north and south wetland regions had significantly different mean macrophyte FWCI scores. Only the regional Tolerant Indicator Species metric for the panhandle and south wetland regions had significantly different mean scores for all of the regionally calculated metrics. Correlations between macrophyte FWCI and the six component metrics with LDI were strong (|r| > 0.50, p < 0.01) for all of the metrics statewide and regionally (Table 3-26), except for the central wetland region Wetland Status Species metric (|r| = 0.39, p = 0.03). This metric was still significantly correlated at a more flexible significance level (p < 0.05). Three of the four wetland regions, including the panhandle, north, and south wetland regions had stronger FWCI correlations with LDI than both the statewide and central wetland region FWCIs. The regional macrophyte FWCI was significantly correlated with the LDI index (Spearmans |r| = 0.73, p < 0.001). A Kruskal-Wallis test

PAGE 95

81 102030405060Statewide Mp-WCI 002468LDI 002468LDI 102030405060Regional Mp-WC I Reference Agricultural Urban Figure 3-19. Macrophyte FWCI scores decreased with increasing development intensity (LDI). A) Statewide FWCI. B) Regional FWCI. A. Statewide Macro p h y te FWCI B. Re g ional Macro p h y te FWCI

PAGE 96

82 Table 3-24. Macrophyte FWCI and component metrics scored statewide and regionally for study wetlands in the low LDI group (LDI < 2.0). PanhandleNorthCentral South Statewide Macrophyte FWCI 50.3 (8.6)a51.6 (8.4)a45.9 (3.4)ab42.2 (9.4)b State Tolerant 8.1 (2.5)ab8.4 (2.2)a6.6 (1.2)b6.8 (2.3)ab State Sensitive 8.8 (2.6)ab9.0 (1.3)a8.6 (0.7)ab6.7 (2.9)b FQAI Score 8.5 (2.2)a8.1 (1.4)a6.6 (0.9)b6.4 (1.7)b Exotic Species 9.3 (1.0)a9.4 (1.1)a8.4 (1.3)a6.6 (1.9)b Native Perennial Species 8.6 (0.9)ab9.1 (1.0)a8.3 (0.6)b7.9 (1.1)b Wetland Status Species 7.0 (2.0)a7.5 (1.9)a7.4 (1.6)a7.7 (2.3)aRegional Macrophyte FWCI 50.9 (7.2)a51.3 (7.9)a51.4 (4.0)a46.8 (9.8)a Regional Tolerant 9.6 (1.0)a9.2 (1.3)ab9.0 (1.5)ab7.7 (2.7)b Regional Sensitive 8.2 (2.6)a8.5 (1.7)a8.6 (1.6)a6.9 (2.4)a FQAI Score 7.6 (2.3)a7.9 (1.4)a8.7 (1.1)a7.7 (2.0)a Exotic Species 9.3 (1.0)a9.5 (1.1)a8.3 (1.4)a8.3 (1.9)a Native Perennial Species 8.3 (1.0)a9.2 (0.9)a9.1 (0.7)a8.7 (1.2)a Wetland Status Species 8.0 (2.1)a7.0 (2.5)a7.8 (1.9)a7.6 (2.1)a Values represent the mean (standard deviation). Wetland regions with similar letters were not significantly different (p<0.05) Table 3-25. Macrophyte FWCI and component metrics scored statewide and regionally for study wetlands in the high LDI group (LDI 2.0). PanhandleNorthCentral South Statewide Macrophyte FWCI 20.4 13.1a29.4 15.9b18.9 11.0a17.6 10.1a State Tolerant 2.1 1.9ab3.3 2.8a2.0 1.6b3.3 1.9ab State Sensitive 2.5 2.0ab4.1 2.4a2.2 2.1ab2.9 2.7b FQAI Score 3.2 3.2a4.3 2.8a2.8 1.8b2.5 1.8b Exotic Species 4.0 3.0a6.3 3.7a3.6 2.4a2.7 2.4b Native Perennial Species 5.2 2.7ab6.6 3.2a4.3 2.6b3.9 2.9b Wetland Status Species 3.3 2.4a4.8 2.6a4.0 2.5a2.3 1.8aRegional Macrophyte FWCI 21.3 14.5ab27.1 16.1a20.6 12.8ab16.6 9.6b Regional Tolerant 5.5 3.7a3.9 3.2ab4.2 3.6ab1.8 1.4b Regional Sensitive 1.3 1.7a2.1 2.4a1.6 2.0a1.3 1.8a FQAI 2.8 2.8a4.4 2.7a3.3 2.5a2.9 2.1a Exotic Species 3.9 3.0a6.4 3.6a3.2 2.5a3.7 2.9a Native Perennial Species 4.2 3.2a6.8 3.2a4.5 2.9a4.3 3.2a Wetland Status Species 3.7 2.8a3.5 3.3a3.7 2.9a2.6 1.8a Values represent the mean (standard deviation). Wetland regions with similar letters were not significantly different (p<0.05).

PAGE 97

83 Table 3-26. Spearman correlations between the macrophyte FWCI and component metrics with LDI. Spearmans rp-value Statewide Macrophyte FWCI -0.73<0.0001 Tolerant Indicator Species 0.75<0.0001 Sensitive Indicator Species -0.66<0.0001 FQAI Score -0.71<0.0001 Exotic Species 0.65<0.0001 Native Perennial Species -0.63<0.0001 Wetland Status Species -0.55<0.0001Panhandle Macrophyte FWCI -0.74<0.0001 Tolerant Indicator Species 0.72<0.0001 Sensitive Indicator Species -0.68<0.0001 FQAI Score -0.68<0.0001 Exotic Species 0.72<0.0001 Native Perennial Species -0.670.0001 Wetland Status Species -0.600.0007North Macrophyte FWCI -0.74<0.0001 Tolerant Indicator Species 0.80<0.0001 Sensitive Indicator Species -0.78<0.0001 FQAI Score -0.75<0.0001 Exotic Species 0.65<0.0001 Native Perennial Species -0.65<0.0001 Wetland Status Species -0.550.0015Central Macrophyte FWCI -0.73<0.0001 Tolerant Indicator Species 0.68<0.0001 Sensitive Indicator Species -0.72<0.0001 FQAI Score -0.68<0.0001 Exotic Species 0.70<0.0001 Native Perennial Species -0.66<0.0001 Wetland Status Species -0.390.0309South Macrophyte FWCI -0.88<0.0001 Tolerant Indicator Species 0.86<0.0001 Sensitive Indicator Species -0.71<0.0001 FQAI Score -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 98

84suggested a significant difference (H = 66.3, p < 0.001) among median regional macrophyte FWCI scores for study wetlands among the three a priori land use categories (reference, agricultural, urban). Cluster Analysis Cluster analysis determined five categories of wetlands based on macrophyte community composition. Clusters were roughly defined by wetland regions and a priori land use categories, including: 1: northern reference; 2: southern reference; 3: northern developed land use; 4: southern developed land use; and 5: statewide pasture land use. Figure 3-20 shows that based on regional macrophyte FWCI scores, clusters 1 and 2 were not significantly different from one another, but were significantly different from clusters 3, 4, and 5 (p < 0.05). Clusters 3 and 4 were not significantly different from each other. Cluster 5 was significantly different from all other clusters. Identical results were obtained using statewide macrophyte FWCI scores. Table 3-27 provides means and standard deviations for cluster statewide macrophyte FWCI scores, regional macrophyte FWCI scores, and LDI. 0102030405060 Cluster 1 Regional Mp-WCI Cluster 2Cluster 3Cluster 4Cluster 5abacb Figure 3-20. Regional macrophyte FWCI scores for five wetland clusters based on macrophyte community composition. Regional Macrophyte FWCI

PAGE 99

85 Table 3-27. Macrophyte FWCI scores and LDI values for wetland clusters based on macrophyte community composition. Cluster Statewide Macrophyte FWCI Regional Macrophyte FWCI LDI 1 46.8 (11.1)a47.4 (9.9)a2.3 (1.9)a 2 44.5 (7.7)a47.4 (9.7)a1.7 (1.3)a3 24.9 (12.2)b25.9 (12.6)b4.4 (1.9)bc4 21.9 (13.8)b23.2 (16.2)b4.0 (2.0)b5 10.6 (7.0)c8.8 (6.6)c5.2 (0.5)c Values represent mean (standard deviation). Clusters with similar letters within columns were not significantly different (p<0.05).

PAGE 100

86 Macroinvertebrate Assemblage Statewide 79 wetlands were sampled for the macroinvertebrate assemblage, with 118 species, 169 representing genera, 85 families, 24 orders, nine classes, and five phyla. The most common macroinvertebrate 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 encountered, 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%) were encountered at only one wetland. The most common families identified included Chironomidae, Naididae, Enchytraeidae, and Culicidae, representing 39, 19, 4, and 4% of the individuals identified, and occurring at 99, 81, 52, and 56% of the study wetlands, respectively. Macroinvertebrates in the family Chironomidae were further divides into the subfamilies Chironominae (89% of Chironomidae), Tanypodinae (10% of Chironomidae), and Orthocladiinae (1% of Chironomidae). Six orders were found at over 50% of the wetlands sampled, including Diptera (47% of individuals identified to the order taxonomic level or lower), Tubificida (24%), Coleoptera (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 phyla were identified, including Arthropoda, Annelida, Mollusca, Platyhelminthes, and Nemertea, with Arthropoda, Annelida, Mollusca found at 100, 92, and 65% of the wetlands sampled, respectively. In the panhandle wetland region, 13 wetlands were sampled hosting 84 genera representing 48 families and 17 orders. In the north wetland region 15 wetlands were sampled with 87 genera (58 families and 20 orders) encountered. The central wetland region included 25 wetlands with 109 genera (60 families and 23 orders) identified. The south wetland region had 26 sample wetlands with 105 genera (in 60 families and 21 orders) identified. Summary Statistics Richness (R), evenness (E), Shannon diversity (H), and Simpsons index (S) were calculated for each sample wetland (Appendix E). Richness ranged from one genus (Pristina) at PR4, a wetland embedded in a low-intensity silvicultural land use, to 26 genera at PA3, a wetland surrounded with row crops. Species evenness ranged from 0.00 at PR4 to 0.97 at PR5, a wetland surrounded by upland forest. Shannon diversity ranged from 0 at PR4 to 0.92 at CA7, a wetland surrounded by silvicultural operations and pasture with cattle. Simpsons index was highest at CR10 at 2.79. Table 3-28 summarizes the richness, evenness, and diversity calculations by a priori land use category (reference, agricultural, or urban). No significant differences were found in richness, evenness, Shannon diversity, or Simpsons index among the three a priori land use categories. Beta and gamma diversity were also similar among a priori land use categories. Table 3-29 shows that the no significant differences were found for richness,

PAGE 101

87 Table 3-28. Macroinvertebrate richness, evenness, and diversity among a priori land use categories. Reference Agricultural Urban Richness (R) 14 (6)a14 (6)a13 (5)a Evenness (E) 0.69 (0.24)a0.69 (0.12)a0.68 (0.15)aShannon Diversity (H) 0.70 (0.26)a0.72 (0.15)a0.70 (0.17)aSimpson's Index (S) 1.83 (0.75)a1.81 (0.54)a1.73 (0.55)aBeta Diversity 8.0 7.8 8.5 Gamma Diversity 114 110 111 Categories with similar letters were not significantly different (Fisher's LSD, =0.05) Values represent mean (standard deviation). evenness, Shannon diversity, or Simpsons index between wetlands in low (LDI < 2.0) and high (LDI 2.0) LDI groups. Beta and gamma diversity were higher for the high LDI groups with beta diversity at 10.9 and gamma diversity 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 wetland regions (panhandle versus north versus central versus south) as well as for multiple pair wise region 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 macroinvertebrate community composition at the genera level was significantly different (at the = 0.05 level), with the exception of the pair wise comparison between the panhandle and north wetland regions (Table 3-30). The macroinvertebrate community composition of the panhandle and north wetland regions was not significantly different for all tests, including among all wetlands and for reference, agricultural, and urban wetlands independently. Table 3-29. Macroinvertebrate richness, evenness, 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 W^ = Mann-Whitney U-Test statistic p` = significance value Values represent mean (standard deviation).

PAGE 102

88 Table 3-30. Macroinvertebrate community composition similarity among a priori land use categories and wetland regions. 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 280.5-0.01 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 140.7-0.03 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-value (p<0.05) suggests a difference in species composition T^ = the MRPP test statistic A` = the chance corrected within-group agreement p# = the significance value.

PAGE 103

89 In reference wetlands, the south wetland region had a significantly different macroinvertebrate community composition as compared to all other wetland regions (panhandle versus south T = -3.2, p = 0.00; north 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 wetland regions. However, macroinvertebrate community composition in the south wetland region was significantly different from both the panhandle and north regions; as were the central and panhandle regions. The only wetland regions with significantly different macroinvertebrate community composition among urban wetlands were the panhandle and south regions. Community Composition Macroinvertebrate community composition was summarized in an NMDS ordination to relate changes in macroinvertebrate community composition with environmental variables. Figure 3-21 shows a two dimensional bi-plot of the NMDS axes. Overlays of significant environmental variables include water column pH, LDI, and latitude. Table 3-31 provides the Pearsons r-squared correlation coefficients between environmental variables and NMDS ordination axes. A three dimensional solution was constructed with an overall stress of 19.8 with a final stability of 0.04. Axis 1 explained 18.9% of the variance and was not correlated with any measured environmental parameter. Axis 2 explained 14.2% variance and was correlated with LDI; axis 3 explained 35.9% variance and was correlated with latitude and water column pH. Metric Selection The pool of potential candidate metrics was streamlined to reduce the redundancy of selected metrics. If two metrics were calculated from the same set of taxa representing redundant metrics only one metric was included. Six metrics that were significant for the Spearmans correlation coefficient with LDI (|r|>0.30, p<0.01) (Table 3-32), showed visually distinguishable trends with LDI, and were significant for the Mann-Whitney U-test between LDI groups (p<0.001) (Table 3-33) were selected for inclusion in the macrophyte FWCI. Table 3-32 provides the statewide Spearman correlation values between the six macroinvertebrate metrics and LDI, water column pH, dissolved oxygen, and TP. Correlations with additional measured environmental variables were used to corroborate meaningful ecological trends in metrics despite the Spearmans correlation coefficient (|r|>0.30) for metrics and LDI. Metrics selected for the macroinvertebrate FWCI were calculated based on the presence/absence of individual macroinvertebrates. Macroinvertebrate metrics selected for inclusion in the FWCI represented tolerance, community balance, and functional group metrics. Tolerance metrics included the Tolerant Indicator Genera, Sensitive Indicator Genera, and Florida Index. Community balance metrics included Mollusca (phylum taxonomic level) and Noteridae (family taxonomic level). One functional groups metric was included, percent Scrapers. The percent of Tolerant Indicator Genera, Mollusca, and Scrapers increased with increasing development intensity, whereas Sensitive Indicator Genera, Florida Index, and Noteridae decreased with increasing development intensity. Table 3-33 shows that scores

PAGE 104

90 Latitude LDI pH Axis 2: 14.2%Axis 3: 35.9% A priori Impact CategoryMinimally ImpairedAgriculturally ImpairedUrban 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 vectors), were significantly correlated with the NMS axes based on macroinvertebrate community composition. Vector length represents the strength of the correlation, 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 105

91 Table 3-31. Pearsons r-squared correlations between environmental variables and NMDS ordination axes based on macroinvertebrate community composition. Axis 1 Axis 2 Axis 3 Incremental r218.9% 14.2% 35.9% Cumulative r218.9% 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 DO (log) 0.01 0.13 0.00 Temperature (log) 0.03 0.05 0.07 Color (log) 0.02 0.02 0.10 Turbidity (log) 0.00 0.09 0.02 pH 0.00 0.06 0.24 Water ammonia-N (log) 0.04 0.08 0.01 Water nitrate/nitrite-N (log) 0.01 0.02 0.05 Water TKN (log) 0.05 0.08 0.00 Water TP (log) 0.02 0.18 0.01 Soil Moisture (arcsine sqrt) 0.02 0.04 0.09 Soil TP (log) 0.10 0.04 0.01 Table 3-32. Spearman correlations between macroinvertebrate metrics and the macroinvertebrate FWCI with LDI, pH, DO (log), and TP (log). Macroinvertebrate Metrics LDIpHDO (log) TP (log) Tolerance metrics Tolerant Indicator Genera 0.510.62-0.25 Sensitive Indicator Genera -0.470.39 -0.37 Florida Index -0.350.35 -0.24Community balance Mollusca 0.330.54-0.28 Noteridae -0.34 Functional group Scraper 0.30 Macroinvertebrate FWCI -0.62-0.560.48 -0.34 All correlations shown are significant (p<0.05)

PAGE 106

92 Table 3-33. Macroinvertebrate metric and FWCI scores between LDI groups. Metric Low LDIHigh LDIW^p` Tolerant indicator species 4.4 (11.8)14.2 (15.5)904.0 <0.001 Sensitive indicator species 15.5 (20.3)2.2 (4.4)1679.0 <0.001Florida Index 2.1 (2.3)0.9 (1.2)1572.0 0.008Mollusca 2.0 (3.5)9.7 (13.4)1025.0 0.003Noteridae 1.8 (3.7)0.2 (0.7)1518.0 0.012Scrapers 4.2 (6.7)10.9 (12.8)1046.0 0.006Macroinvertebrate FWCI 36.8 (10.0)22.3 (8.4)1878.5 <0.001 W^ = Mann-Whitney U-Test statistic. p` = significance value. Values represent the mean (standard deviation). of selected metrics and the macroinvertebrate FWCI were significantly different between low and high LDI groups (p < 0.05). Tolerance metrics The statewide Tolerant Indicator Genera were established at an LDI break of 4.0, and included six genera Goeldichironomus, Micromenetus, Microvelia, Physella, Tropisternus, and Tanypus (Table 3-34). Figure 3-22 shows Tolerant Indicator Genera increased with increasing development intensity. Two irregularities were apparent in the low LDI group, including SU4 (Tolerant Indicator Genera = 61%) and SA8 (Tolerant Indicator Genera = 31%). Wetlands in the high LDI group with the highest percent Tolerant Indicator Genera included four central (CA9, CA3, CU5, CA2) and one north (NA4) wetland region wetland.. All five wetlands were surrounded by different land uses, including citrus crops (CA9), pullet farm spray field (CA3), residential and commercial (CU5), pasture (CA2), and row crops (NA4). The 14 sensitive indicator genera (Table 3-35) were calculated at an LDI break of 1.75. The five Sensitive Indicator Genera with the highest Indicator Values included Bersous, Hydrocanthus, Larsia, Pristina, and Pristinella. Sensitive Indicator Genera included macroinvertebrates in two phyla, Annelida and Arthropoda. The phylum Annelida was represented by two genera of aquatic worms, Pristina and Pristinella, both in the family Naididae, order Haplotaxida, class Oligochaeta. The 12 remaining Sensitive Indicator Genera fell within the phylum Arthropoda, representing two classes Arachnida (including a water mite); and Insecta, aquatic insects in three orders including Coleoptera (five genera of beetles), Diptera (four genera of true flies), and Trichoptera (two genera of caddis flies). Figure 3-23 shows the percent Sensitive Indicator Genera decreased with increasing development intensity in the landscape. Two wetlands hosted an unexpected percent of Sensitive Indicator Genera including one in each LDI group. All of the macroinvertebrates identified at PR4 (in the low LDI group) were categorized as Sensitive Indicator Genera (100%). Wetlands hosting the next highest percent macroinvertebrate Sensitive Indicator Genera were NR3 with only 42% sensitive

PAGE 107

93 Table 3-34. Macroinvertebrate tolerant indicator genera. Toleraof 4.0 were significant (p < 0.10). nt indicator genera calculated at an LDI break PhylumClassOrderFamilyGeneraIndicator Valuep-valueArthropodaInsectaColeopteraHydrophilidaeTropisternus17.20.053DipteraOrthocladiinaeGoeldichironomus56.90.001ChironomidaeTanypus8.10.098HeteropteraVeliidaeMicrovelia12.90.033MolluscaGastropodaBasommatophoraPlanorbidaeMicromenetus22.50.084PhysidaePhysella21.50.002

PAGE 108

94 0102030405060700246Landscape Development Intensity Index%Toleran 8 t Figure 3-22. Macroinvertebrate Tolerant Indicator Genera increased with increasing development intensity (LDI).

PAGE 109

95 Table 3-35. Sensitive macroinvertebrate indicator genera. Sensitive indicator genera calculated at an LDI break of 1.75 were significant (p < 0.10). PhylumClassOrderFamilyGeneraIndicator Valuep-valueAnnelidaOligochaetaHaplotaxidaNaididaePristina35.50.008Pristinella25.90.094ArthropodaArachnidaAcariformesHydrachnidaeHydrachna9.40.049InsectaColeopteraDytiscidaeLaccophilus8.90.061HaliplidaeHaliplus8.80.055HydrophilidaeBerosus28.80.003NoteridaeHydrocanthus28.50.004Suphis12.50.048DipteraChironomidaeLarsia24.80.011Paramerina140.062Zavreliella17.30.008OrthocladiinaeDicrotendipes14.80.065TrichopteraLeptoceridaeOecetis12.50.023HydroptilidaeOxyethira12.50.032

PAGE 110

96 0204060801000246Landscape Development Intensity Index%Sensitive 8 Figure 3-23. Sensitive macroinvertebrate indicator genera decreased with increasing development intensity (LDI). Indicator Genera and CR11 (surrounded by pine flatwoods) with 41%. The inconsistent score in the high LDI group with a low presence of Sensitive Indicator Genera was SU2 (situated on the edge of a public school campus and adjacent to a high-intensity roadway with urban marsh on more than half of the wetland boundary; Sensitive Indicator Genera = 25%). The third tolerance metric was the Florida Index, an index based on the relative pollution tolerance of macroinvertebrates identified in a water body (Beck 1954; USEPA 2002c). Calculations for the Florida Index included scoring Class I organisms, which were considered least tolerant, and Class II organisms, which were considered intolerant of pollution. Mixed taxonomic levels 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; Beck 1954; USEPA 2002c). The Florida Index value was expected to decrease with increasing development intensity in the surrounding landscape (Barbour et al. 1996a). Figure 3-24 shows that the Florida Index score generally decreased with increasing development intensity in the surrounding landscape, with six wetlands showing an inconsistent trend including four urban and two agricultural wetlands. The five highest scoring wetlands in the low LDI group included wetlands in each wetland region including CR4 (Florida Index = 8), SR8 (Florida Index = 7), NR3 (Florida Index = 7), CR11 (Florida Index = 6), and PR8 (Florida Index = 5).

PAGE 111

97 0246802468Landscape Development Intensity IndexFlorida Inde x Figure 3-24. Florida Index scores decreased with increasing development intensity (LDI). Community balance metrics Two community balance metrics were incorporated into the macroinvertebrate FWCI including percent Mollusca and percent Noteridae. The percent of individuals in the phylum Mollusca was significantly correlated with LDI (Table 3-32) and significantly differentiated between low and high LDI groups (Table 3-33). Figure 3-25 shows that the percent of macroinvertebrates in the phylum Mollusca increased with increasing development intensity. Macroinvertebrates were identified in three classes within the order Mollusca, including Bivalva, Gastropoda, and Plecypoda. Nearly two-thirds of the wetlands hosted macroinvertebrates in the phylum Mollusca (n=51). In five wetlands over one-third of the macroinvertebrates that were identified belonged to the phylum Mollusca, including SU2 (situated on the edge of a public school campus and adjacent to a high-intensity roadway with urban marsh on more than half of the wetland boundary; 44.7%), CU8 (surrounded by industrial and transportation land use; 44.4%), SU7 (surrounded most recently by a high-intensity shopping mall, previously in row crops; 44.1%), SA4 (surrounded by row crops; 37.5%), and CA5 (surrounded by pasture; 34.9%). In the low LDI group, four wetlands had greater than 5% of the identified macroinvertebrates in the phylum Mollusca, including SA8 (surrounded by a suburban park and receiving run-off from animal husbandry; 16.7%), PR7 (a large, deep wetland

PAGE 112

98 010203040500246Landscape Development Intensity Index%Mollusca 8 Figure 3-25. Macroinvertebrates in the phylum Mollusca increased with increasing development intensity (LDI). on a private conservation tract ; 9.0%), SR4 (nested in the Everglades; 8.1%), and CR3 (surrounded by uplands undergoing restoration; 5.8%). Figure 3-26 shows that the percent of macroinvertebrates in the family Noteridae decreased with increasing landscape development intensity as expected (Barbour et al 1996b). Macroinvertebrates in the family Noteridae never made up more than 15% of the individuals identified to the family taxonomic level or lower at any of the sample wetlands. The family Noteridae falls within order Coleoptera, class Insecta, phylum Arthropoda. Macroinvertebrates in the family Noteridae (burrowing water beetles) typically inhabit the shallow margins of standing or slow flowing streams, or in lentic habitats, act as climbers associated with vascular macrophytes 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 five families within Coleoptera with both life stages being aquatic. As a general rule, the percent of macroinvertebrates in the order Coleoptera was found to decrease with increasing development surrounding Florida streams (Barbour et al. 1996b). One wetland not following the expected trend in the high LDI group was CA4 (surrounded by pasture; 4.1%).

PAGE 113

99 024681012140246Landscape Development Intensity Index%Noteridae 8 Figure 3-26. Macroinvertebrates in the family Noteridae decreased with increasing development intensity (LDI). Functional group metrics One functional group metric was selected for inclusion in the macroinvertebrate FWCI, the percent scrapers functional feeding group. Figure 3-27 shows the percent macroinvertebrates in the scraper functional feeding group increased with increasing LDI. The scraper functional feeding group included macroinvertebrates that scrape periphyton from mineral and organic surfaces and those that browse or graze algal materials. Five wetlands with the highest percent scrapers were found in among all four wetland regions and represented wetlands embedded in a mix of urban and agricultural land uses, including SU7 (surrounded most recently by a high-intensity shopping mall, previously in row crops; 51%), SA4 (surrounded by row crops; 40%), CA5 (surrounded by pasture; 40%), PA6 (surrounded by pasture; 38%), and CU8 (surrounded by industrial and transportation land use; 31%). Nearly one-quarter of the sample wetlands (n = 19) did not have scrapers identified in the macroinvertebrate samples. Macroinvertebrate Florida Wetland Condition Index The six metrics described above were included in the macroinvertebrate FWCI. Appendix F provides detailed instructions on metric scoring. Figure 3-28 shows the relationship between the macroinvertebrate FWCI and LDI. Potential scores for the macroinvertebrate FWCI range from 0-60, with higher values representing reference wetlands. Actual scores ranged from 5.2 at SU5 (a deeply flooded swamp surrounded by industrial land use, LDI = 5.2), to 57.0 at SR3 (surrounded by pine flatwoods, LDI = 1.0).

PAGE 114

100 01020304050600246Landscape Development Intensity Index%Scrapers 8 Figure 3-27. The percent of macroinvertebrates that belong to the scraper functional feeding group increased with increasing development intensity (LDI). Figure 3-28. Macroinvertebrate FWCI scores decreased with increasing landscape development intensity index (LDI). 01020304050600246LDIMi-WCI 8 Reference Agricultural Urban Macroinvertebrate FWCI

PAGE 115

101 Ranges varied regionally, though the regional scores were not significantly different for either low or high LDI groups. The highest scores in each wetland region included four reference wetland, PR8 (40.7), NR3 (52.8), CR6 (50.4), and SR8 (57.0), in the panhandle, north, central, and south wetland regions, respectively. The lowest scoring wetlands in both the panhandle and central wetland regions were embedded in pasture, including PA6 (12.9) and CA5 (7.2). In the north wetland region NA4 (surrounded by row crops) scored 10.4. In the south SU5 (surrounded by commercial land use) received the lowest score overall of 5.3. The macroinvertebrate FWCI was significantly correlated with the LDI index (Spearmans |r| = 0.62, p < 0.001). A Kruskal-Wallis test suggested a significant difference (H = 36.0, p < 0.001) among median macroinvertebrate FWCI scores for study wetlands in a priori land use categories. Cluster Analysis Cluster analysis determined five categories based on macroinvertebrate community composition. Clusters were explained by regions and a priori land use categories including: 1: south to central low development intensity; 2: mixed region low development intensity; 3: north central to panhandle middle development intensity; 4: northern to panhandle middle development intensity; and 5: high development intensity and southern Everglades. Figure 3-29 shows that based on macroinvertebrate FWCI scores, clusters 1 and 2 were not significantly different from one another, but were significantly different from cluster 5. Clusters 3 and 4 were significantly different from cluster 1 and cluster 5. Table 3-36 provides means and standard deviations for macroinvertebrate FWCI and LDI scores of the five clusters.

PAGE 116

102 0102030405060 Cluster 1 Macroinvertebrate WCI Cluster 5Cluster 4Cluster 3Cluster 2aabbbc Macroinvertebrate FWCI Figure 3-29. Macroinvertebrate FWCI scores for five wetland clusters based on macroinvertebrate community composition. Clusters with similar letters were not significantly different (Fishers LSD, p<0.05). Table 3-38. Macroinvertebrate FWCI scores and LDI values for wetland clusters based on macroinvertebrate community composition. Cluster 1 Cluster 2 Cluster 3 Cluster 4 Cluster 5 FWCI 41.6 (10.7)a35.5 (10.8)ab29.3 (11.8)b28.3 (6.2)b21.9 (9.7)c LDI 2.0 (1.5)a3.1 (2.5)ab3.4 (2.3)ab3.4(2.3)ab4.0 (2.1)b Clusters with similar letters were not significantly different (p<0.05). Values represent the mean (standard deviation).

PAGE 117

103 Florida Wetland Condition Index In total 19 metrics were used to construct the FWCI, including seven metrics based on the diatom assemblage, six metrics based on the macrophyte assemblage, and six metrics based on the macroinvertebrate assemblage. Table 3-37 lists the FWCI scores for 118 isolated forested wetlands for each assemblage. Scores ranged from 0-70 for the diatom FWCI and from 0-60 for both the macrophyte and macroinvertebrate FWCI. Different wetlands received the highest and lowest scores for each FWCI. The highest scores for each FWCI were found at wetlands in the south and central wetland regions, including SR2 (diatom FWCI = 68.9), CR11 (macrophyte FWCI =59.0), and SR8 (macroinvertebrate FWCI= 57.0). Minimum FWCI scores were found among three different wetland regions, including CA3 (receiving wastewaters from pullet farm operations; diatom FWCI = 7.9), NA1 (surrounded by pasture at an experimental research facility; macrophyte FWCI = 0.0), and SU2 (situated on the edge of a public school campus and adjacent to a high-intensity roadway with urban marsh on more than half of the wetland boundary; macroinvertebrate FWCI = 5.3). Within the north and central wetland regions some wetlands received the wetland region maximum scores for multiple assemblages, including NR3 (surrounded by pine flatwoods; diatom FWCI = 66.8; macroinvertebrate FWCI =52.8) and CR6 (surrounded by pine flatwoods; diatom FWCI = 65.5; macroinvertebrate FWCI = 50.4). In the panhandle wetland region, PU4 (surrounded by residential and commercial land uses) received minimum FWCI scores for both the diatom and macrophyte assemblages (diatom FWCI = 10.5; macrophyte FWCI = 4.0). Figure 3-30 shows a three dimensional scatter plot for the 50 wetlands receiving scores for all three assemblages. The maximum diatom FWCI for the 50 wetlands graphed was 68.9 (of 70) at SR2. The maximum macrophyte FWCI was 58.4 (of 60) at PR6 (LDI = 1.3), and the maximum macroinvertebrate FWCI 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 two dimensional comparisons of FWCI scores for wetlands with multiple assemblages sampled, including (A) 50 wetlands with diatom and macrophyte FWCI scores, (B) 50 wetlands with diatom and macroinvertebrate FWCI scores, and (C) 79 wetlands with macrophyte and macroinvertebrate FWCI scores. The most obvious irregularity between the diatom and macrophyte FWCI scores (Figure 3-31 A) was at CU6, a wetland surrounded by residential (both single and multifamily) and a golf course that had been developed within the past five years. CU6 had low scores for the diatom (15.1 of 70) and macroinvertebrate (23.4 of 60) but a considerably higher score for the macrophyte (41.5 of 60) FWCI. The 19 metrics incorporated into the here assemblage specific FWCIs were compared across different assemblages (Table 3-38). Of the 120 potential metric comparisons (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 additional 20% of the potential 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 one-quarter of the comparisons among metrics of different assemblages were not significantly correlated (22%). The strongest correlation among metrics of different assemblages was

PAGE 118

104 Table 3-37. Florida Wetland Condition Index scores for 118 wetlands based on three assemblages including diatoms, macrophytes, and macroinvertebrates. Site Code Diatom FWCI Macrophyte FWCI MacroinvertebrateFWCI 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 119

105 Table 3-37. Continued. Site Code Diatom FWCI Macrophyte FWCI Macroinvertebrate FWCI 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 120

106 Table 3-37. Continued. Site Code Diatom FWCI Macrophyte FWCI Macroinvertebrate FWCI 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 121

107 Table 3-37. Continued. Site Code Diatom FWCI Macrophyte FWCI Macroinvertebrate FWCI 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 between the diatom Sensitive Indicator Genera and the macrophyte Sensitive Indicator Species (Pearsons 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 two of the comparisons between the diatom and macrophyte assemblages were not significantly correlated at the most flexible significance level (p < 0.10). Comparisons between the diatom and macroinvertebrate metrics were not as strong, with less than 50% of the metrics significantly correlated (p < 0.10). Only six comparisons among diatom and macroinvertebrate metrics were correlated at the more stringent significance level (p < 0.01). The macroinvertebrate metrics Noteridae and scrapers were not significantly correlated with any of the diatom metrics. Correlations were stronger among the macrophyte and macroinvertebrate comparisons, with 94% of the comparisons significantly correlated (p < 0.10). In fact, 20 of the metric comparisons (56%) were correlated at the strictest significance level (p < 0.01). Two metric comparisons between the macrophyte and macroinvertebrate metrics were not significantly correlated, including the macrophyte Wetland Status Species and macroinvertebrate Mollusca metrics and between the macrophyte Exotic Species and macroinvertebrate Scrapers metrics.

PAGE 122

108 0 0 10 10 20 20 30 40 30 50 6040 5 0 70 60 50 40 30 20 10 0 0 60 Reference Agricultural Urban 1:1:1 LineMacroinvertebrate WCIMacrophyte WCIDiatom WCI Figure 3-30. Three dimensional scatter plot of the FWCI scores based on three assemblages, including diatoms, macrophytes, and macroinvertebrates.

PAGE 123

109 0102030405060010203040506070Diatom WCIMacrophyte WCI 0102030405060010203040506070Diatom WCIMacroinvertebrate WC I 01020304050600102030405060Macrophyte WCIMacroinvertebrate WC I Reference Agricultural Urban Figure 3-31. Scatterplots of FWCI scores for wetlands based on diatom, macrophyte, and macroinvertebrate assemblages. A) Diatom and macrophyte FWCI scores (n = 50 wetlands). B) Diatom and macroinvertebrate FWCI scores (n = 50). C) Macrophyte and macroinvertebrate FWCI scores (n = 79).

PAGE 124

110 Table 3-38. Pearson correlations among 19 metrics included in the three assemblage specific Florida Wetland Condition Indices. 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 Index 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 125

CHAPTER 4 DISCUSSION The three assemblage Florida Wetland Condition Indices (FWCIs) for isolated depressional forested wetlands of Florida are comprised of 19 metrics, based on the community composition of three separate species assemblages including seven diatom, six macrophyte, and six macroinvertebrate based metrics (Table 4-1). Metrics were selected for inclusion in the FWCIs based on three criteria: Strength and significance (Spearmans correlation coefficient) of each metric in correlation with a quantitative gradient of Landscape Development Intensity (LDI); Ability of a metric to visually distinguish correlations with LDI in scatter plots; Calculated significant difference (Mann-Whitney U-test) between low and high LDI groups. The index for each assemblage was composed of individual metrics specific to the assemblage, which were scaled and added together. While previous research has identified responses of wetland ecosystems to individual changes (such as increased nutrients or altered hydrology), few have combined multiple biotic components, environmental parameters, and landscape development intensity in an attempt to quantify ecological integrity. The contribution of this research to our understanding of changes in the community composition of isolated forested wetlands (based on the diatom, macrophyte, and macroinvertebrate assemblages) in relation to different development intensities in the surrounding landscape can be summarized in seven main points. Biological indicators along with physical and chemical parameters were useful in defining biological integrity; The variable turnover times and sensitivities of the three assemblages (diatoms, macrophytes, macroinvertebrates) suggest that multiple assemblage specific multi-metric FWCIs have more merit than an FWCI based on a single assemblage; Regionalization may strengthen the FWCI; A FWCI independent of wetland type may be feasible, given the strong likeness of the forested FWCI to the marsh Wetland Condition Index (Lane et al. 2003); The LDI index was a useful tool correlating with the measured biological condition for isolated forested wetlands; Urban wetlands exhibit a different vector of change than do agricultural wetlands, and while the FWCI suggests low biological integrity of both agricultural and urban wetlands, these wetlands do provide services and do work in the environment; Richness, evenness, and diversity of each assemblage were not sensitive to different land uses or development intensities in the surrounding landscape. 111

PAGE 126

Table 4-1. The 19 metrics of the Florida Wetland Condition Indices for isolated depressional forested wetlands, including seven diatom, six macrophyte, and six macroinvertebrate metrics. Diatom Metrics Macrophyte Metrics Macroinvertebrate Metrics % Tolerant % Tolerant % Tolerant % Sensitive % Sensitive % Sensitive % in Pollution Class 1 Modified FQI Score FL Index Score % in Nitrogen Class 3 % Exotic % Mollusca % in Saprobity Class 4 % Native Perennial % Noterida % in pH Class 3 % Wetland Status % Scraper % in Oxygen Class 1 Describing Biological Integrity Biological indicators along with chemical and physical parameters were useful in determining the biological integrity of isolated depressional forested wetlands. For the purposes of this study, biological integrity has been defined quantitatively with the FWCIs. The FWCIs incorporated 19 metrics from three different species assemblages (diatoms, macrophytes, and macroinvertebrates). Correlations between the diatom, macrophyte, and macroinvertebrate FWCIs and the intensity of development in the surrounding landscape (based on the use of nonrenewable energy and calculated with the Landscape Development Intensity (LDI) index advocate that changes in community composition were captured by the FWCIs. 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 FWCIs defined and detected deviations from the condition of reference wetlands based on community composition. Each of the 19 FWCI metrics addressed some disparity from the assumed range of feasible conditions. For all three assemblages, the Tolerant Indicator Species metric demonstrated the strongest correlation with LDI, suggesting that the presence of a suite of taxa characteristic of wetlands with low biological integrity may be the single most effective means of identifying changes in community composition. The isolated depressional forested wetlands sampled were influenced by various anthropogenic activities (from direct herbivory and trampling by domestic cattle, to increased nutrients from agricultural or stormwater run-off, to hydrological impoundments 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 three assemblages separately suggested that differences in some agricultural and urban development intensities may be too subtle to detect with compositional data. Furthermore, greater variability in the macroinvertebrate assemblage of reference wetlands as compared to that of agricultural and urban wetlands suggested that perturbations to the driving energies in isolated wetlands may result in a convergence of the taxa present. Indeed, the natural 112

PAGE 127

113 compositional variability inherent among reference wetlands may be lost with increased development intensity in the surrounding landscape. While the FWCIs can not be used to predict changes in the physical and chemical parameters of a wetland, their strength lies in providing an overview of biological integrity through the integration of changes in community 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-nitrogen, water TKN, water TP, soil moisture, soil organic matter, and soil TP). When soil and water parameters were used to explain variation in the community composition of each assemblage, water column pH was universally identified. Additionally, total phosphorus concentrations explained some of the variance in both the diatom and macrophyte assemblages. Perhaps preservation and restoration strategies could focus on limiting activities (in the surrounding landscape) that influence changes to water column pH and total phosphorus loading to wetlands in order to promote biological integrity. Merits of a Multi-Metric FWCI The variable turnover times and sensitivities of the three different assemblages (diatoms, macrophytes, macroinvertebrates) suggest that species specific multi-metric FWCIs have more merit than an FWCI based on a single assemblage. The FWCIs can be used to infer influences in temporal and spatial changes to which a particular wetland has been exposed. For example, diatoms have rapid turnover times and may react immediately to shifts in driving energies. On the other hand, perennial macrophytes may respond to changes over a longer period of time, particularly in the case of the woody midand over-story species. While the macrophyte FWCI score may remain relatively high in a recently enriched wetland, the diatom FWCI score may reflect lower biological integrity. Whereas macrophytes assimilate nutrients for growth, this process has a longer time frame than the rapid growth rate typical of the algal assemblage. An explosion of algal growth may in turn alter the available food resources within a wetland affecting other assemblages, for example there may be an increase in macroinvertebrate that scrape algae (scrapers), and therefore a decline in the macroinvertebrate FWCI score. While agreement in the ranking of the biological condition of study wetlands using the FWCIs was anticipated, discrepancies among the ranking from the different assemblages may provide great insight into wetland condition as different species assemblages respond to changes in driving energies over different time scales. There was variation among the ranking of wetlands for the diatom, macrophyte, and macroinvertebrate FWCIs; though there were no obvious outliers when the three assemblages were compared. 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. Many of the metrics of the different assemblages were significantly correlated. Perhaps the value of each of the 19 selected metrics is inherent in the differentiation between categories of landscape development intensity. Diatom and macrophyte metrics were strongly correlated with one another, and yet diatom and macroinvertebrate metrics were not, reinforcing the value of including various species assemblages in an assessment

PAGE 128

114 of biological integrity. Perhaps, with regular biological monitoring it may be possible to further explore the temporal effects of changing development intensity on the biological integrity of the community composition of different species assemblages and the entire wetland community. Regionalization of the Florida Wetland Condition Index The climate of Florida is considered humid subtropical, though pronounced differences occur in the local climate across Florida, such as differences in the amount of annual rainfall, seasonal maximum temperatures, and number of freeze days (Fernald and Purdum 1992; Lane 2000). Despite the broad latitudinal and longitudinal ranges of sample wetlands throughout Florida (26.0N -31.0N latitude, 80.1W-87.5W longitude), statewide significant difference in water and soil parameters among a priori land use categories were detected, suggesting the statewide scale may be appropriate for a physical and chemical assessment of wetland condition. Nevertheless, the influence of latitude and longitude was reflected in the compositional difference of all three species assemblages found among the Florida wetland regions (Lane 2000). Latitude and longitude were significantly correlated with macrophyte community composition, and latitude explained partial variance in macroinvertebrate community composition. In addition, wetlands in the Florida Everglades were outliers in many of the diatom metrics, and the southern Everglades wetlands formed distinguished clusters based on diatom and macroinvertebrate community composition. Most of the human development in Florida has occurred along the east and west coastal areas of peninsular Florida (Fernald and Purdum 1992), suggesting that while the reference wetlands selected in the south and central ecoregions were selected as the best possible examples of reference type conditions, they may be more affected by development in the surrounding landscape (such as compounded secondary effects) than their panhandle and far north wetland region counterparts. Regionalization was explored for the macrophyte assemblage because of the sufficient number of wetlands sampled within each wetland region. There were clear regional differences in the statewide macrophyte FWCI scores for wetlands in the low LDI group, which down scored the reference wetlands of the south and central ecoregion This led to the use of regionalized scoring of the macrophyte metrics. While the ease and utility of statewide species lists for metrics such as Indicator Species would seemingly prevail over four regional species lists, the necessity of scoring each wetland region based on the best possible reference conditions (Karr and Chu 1999) cannot be overlooked. Regionalization of biological indices has been suggested throughout the literature, as the main premise behind indices of biological integrity is based on a comparison of like to like (Gerristen et al. 2000), that is, to reduce the noise in background variability in biological data (which could be accomplished through regional based indices). For instance, differences in the macroinvertebrate community composition among wetland regions may be of importance in improving the macroinvertebrate FWCI. For example, none of the sample wetlands in the panhandle wetland region hosted macroinvertebrates in the order Trichoptera (caddis flies), whereas no wetland in the north wetland region hosted macroinvertebrates in the order Ephemeroptera (mayflies). While both of these

PAGE 129

115 orders are characteristic of lotic environments (Edmunds and Waltz 1996), the absence of an entire order from wetland regions suggests the value of regionalization of the FWCIs. Florida Wetland Condition Index Independent of Wetland Type Recent work by Lane et al. (2003) presents a 24 metric herbaceous Wetland Condition Index (FWCI) for isolated depressional herbaceous wetlands in Florida based on the community composition of the diatom, macrophyte, and macroinvertebrate assemblages. The herbaceous FWCI was created based on 75 freshwater marshes surrounded by reference (n=34) and agricultural (n=40) land uses throughout peninsular Florida. Of the 14 metrics for the diatom assemblage, the forested and marsh FWCIs share five metrics, including all of the forested diatom FWCI metrics except for species lists for the Sensitive and Tolerant Indicator Species. Though both the forested and herbaceous FWCIs include Sensitive and Tolerant Indicator Species metrics, shared species were limited between wetland types, as the Tolerant Indicator Species list had only two mutual species (Navicula confervacea and N. minima), of 12 species for the forested FWCI and 21 species for the herbaceous FWCI. Similarly, the sensitive indicator species lists shared only five species (Eunotia flexuosa, E. naegelii, E. rhomboidea, Frustulia rhomboids, and F. rhomboids crassinervia), of 18 species for the forested FWCI and 22 species for the herbaceous FWCI. The five remaining diatom metrics were based on autecological relationships, including Pollution Class 1 (Bahls 1993), Nitrogen Uptake Metabolism 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 seven marsh FWCI diatom metrics, five were considered too similar to selected forested FWCI metrics and were excluded to avoid redundancy. The final two marsh FWCI diatom metrics were based on Salinity Class (van Dam et al. 1994), and were not significantly correlated with LDI for isolated depressional forested wetlands. Five macrophyte metrics were incorporated into the depressional herbaceous FWCI, and variants of these were included in the metrics of the isolated depressional forested FWCI. Tolerant and Sensitive Indicator Species lists were constructed separately for each wetland type. Of the 46 statewide tolerant macrophyte indicator species for the herbaceous FWCI, 28 also occur on the statewide macrophyte Tolerant Indicator Species list for the forested FWCI (of 61 species). Similarly, 20 statewide macrophyte Sensitive Indicator Species were shared for the herbaceous FWCI (of 36) and the forested FWCI (of 69). The three additional metrics included in the herbaceous FWCI were percent Exotic Species, Annual to Perennial Ratio, and a metric based on scores from a Floristic Quality Assessment Index (similar to the one conducted in this study, but with scores specific to Florida depressional herbaceous wetlands). In the isolated depressional forested FWCI, a variant of the Annual to Perennial Ratio was used, the percent Native Perennial Species (to account for anticipated conditions at urban wetlands). The sixth forested FWCI metric was the percent Wetland Status Species. There was less similarity between the five macroinvertebrate depressional herbaceous FWCI metrics and the six isolated depressional forested macroinvertebrate FWCI metrics. Tolerant and Sensitive Indicator metrics were constructed separately for each wetland type and were included in both indices. Three Tolerant Indicator Genera

PAGE 130

116 occurred on both lists (Goeldichironomus, Micromenetus, and Physella), and only one Sensitive Indicator Genera was shared (Larisa). The herbaceous FWCI included three additional metrics: %Predators, %Odonata, and %Orthocladiinae. The forested FWCI included four different metrics: Florida Index, %Mollusca, %Noteridae, and %Scrapers. Overall, the herbaceous and forested FWCIs were similar, with many shared metrics. Some additional variability between selected metrics was expected as the isolated depressional forested FWCI included two additional sources of variability (wetlands in the panhandle wetland region and urban land uses, which were not included in the sample wetlands for the herbaceous FWCI). Perhaps the strong similarity of metrics suggests that a universal assessment index could be constructed regardless of wetland type. However, it would likely be necessary to maintain independent indicator species lists specific and metric scoring criteria for different wetland types. Landscape Development Intensity Index The Landscape Development Intensity index (LDI) is a useful tool in approximating ecological integrity. Its primary power is the reproducible, objective, and quantitative methods employed to obtain a score based on the use on non-renewable energy in the surrounding landscape, which in turn corresponds strongly and significantly with the FWCI for diatoms, macrophytes, and macroinvertebrates. By correlating the community composition based FWCI and the landscape based LDI, we are better able to understand driving influences to ecological integrity, while also identifying areas where landscape scale changes in surrounding land use may not be manifest as apparent changes in species assemblage level community composition. Some urban wetlands received somewhat contradictory LDI and FWCI scores, which were concurrent to expected trends. However, many of these urban wetlands had been exposed to an assortment of additional land uses in the recent past. For example, in areas of Florida just inland of the extensive coastal development and that were once highly agrarian, development has spread, and so wetlands that would have been considered agricultural within the past decade are now categorized as urban. Many of these previous agricultural fields and pasture lands have only recently been transformed for residential and commercial uses, suggesting that there may be limitations to the predictive power of LDI for wetlands situated in rapidly changing landscapes. Further modifiers for time-since-disturbance factors, or perhaps weighted averages based on previous land use may be appropriate means of modifying LDI, though further work is needed. Species assemblage specific response time appears extremely important, particularly for longer-lived species, where reproduction may be inhibited by current conditions, and so ecosystem organization would currently reflect a certain higher level of ecological integrity despite impeded possibilities for system perpetuity. There are some additional deviations from expected LDI and FWCI correlations. For example, agricultural wetlands generally receive lower LDI scores than urban wetlands, however in the case of wetlands with active cattle grazing within the wetland ecosystem FWCI scores can be lower for grazed wetlands than for urban wetlands. However, as a whole, the LDI is a surrogate for the FWCI, taking into account landscape scale impacts that would alter the community composition of a target species assemblage. While the sites grazed by cattle reflected low macrophyte FWCI scores, sample wetlands

PAGE 131

117 that were actively grazed by cattle were not clustered into separate groups based on the diatom or macroinvertebrate assemblages. Agricultural and Urban Wetlands Urban wetlands appear to exhibit a different vector of change than do agricultural wetlands; however the FWCIs did not significantly differentiate between agricultural and urban wetlands. The LDI also did not specifically differentiate between these land use categories, as LDI coefficients for agricultural land uses range from 2.6 (unimproved pasture) to 6.6 (high intensity agriculture). Urban LDI coefficients 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). Similarly, other measures of anthropogenic influence like the Wetland Rapid Assessment Procedure and the Minnesota disturbance index did not clearly differentiate between agricultural and urban land uses. The main land use specific conclusion we can draw from the FWCIs 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 they provide important services and do considerable work in the environment. Wetlands embedded in a developed landscape matrix provide an abundance of potential services. For example, they may store and purify stormwater, process nutrients and toxins (perhaps acting as a sink and protecting hydrologically connected systems), provide habitat for local wildlife and perhaps migratory species, produce oxygen, filter the air, provide noise abatement, and act as refugia for urban ecologists. Specifically in the case of urban wetlands, there is a debate as to the value of small remnant wetlands embedded within highly developed landscape matrices. While wetlands do exist in highly urbanized areas, they do not appear to closely resemble wetlands in undeveloped landscapes. This leaves us to question just how much of the inherent function of the wetland has been lost. Under current Florida law, mitigation ratios for urban wetlands will be small, and some people may dispute the idea of keeping urban wetlands of reduced 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, off-site mitigation undervalues the services provided by urban wetlands. Urban wetlands clearly provide some function, and perchance they are doing more work processing nutrients, storing urban stormwater run-off, and storing toxins, than wetlands in undeveloped landscapes. The continued existence of urban wetlands is crucial (for the maintenance of biotic diversity, buffering pollution and contamination to protect nearby environments, increasing oxygen production in an urban center, etc.). While the FWCI scores for urban wetlands reflect lowered biological integrity, perhaps maintaining 30-70% on average of the biological integrity of reference wetland is more important than having no wetland and therefore no services or free work. Wetlands with the lowest biological integrity could have scored a 0 for the FWCI, and yet only three heavily grazed wetlands did and only based on one of three the species assemblages assessed. There is no doubt that the intensity of human land use across the landscape plays a role in the loss of biological integrity of wetlands, however we should reconsider our willingness to

PAGE 132

118 remove all of the biological integrity of a wetland by otherwise erasing its existence by filling. Richness, Evenness, and Diversity Measures of richness, evenness, and diversity of the diatom, macrophyte, and macroinvertebrate assemblage were not sensitive to difference in land use or development intensity in the surrounding landscape. For both the diatom and macroinvertebrate assemblages neither a priori land use classification nor categories of landscape development intensity showed significant differences in richness, evenness, or diversity calculations. Differences in macrophyte evenness and diversity between reference and agricultural wetlands may be attributable to both direct (for example grazing by domestic cattle) or more indirect (increased nutrients from fertilizer carried in run-off from surrounding 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 diversity 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 fact increasing the overall 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 anthropogenic development intensity increases, 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 alterations such as clear-cut harvests (Devine 1998). Within the study wetlands, the percent of exotic macrophyte species increased with increasing development intensity in the surrounding landscape. The influx of exotic species added to, rather than diminished, the species evenness and diversity within the isolated forested wetlands sampled. Limitations and Further Research Several limitations to this study should be noted, including sampling methods and drought conditions. One water sample was collected to represent the water environment of the entire sample wetland, and water samples were taken at a range of times throughout the day. While water samples were always taken when a crew initially arrived at a sample wetland, the time of day the crew arrived fluctuated. Additionally, while an attempt was made to avoid taking water samples immediately following extreme rain event, there is the possibility that the sample was taken during a period of time without rain when the wetland was drying down. Therefore, there was no consistency as to recent weather conditions when water samples were taken. There were also strict requirements of preservation, temperature control, and shipping protocols associated with water samples. When these requirements were breached the sample had to be discarded.

PAGE 133

119 Similarly, one composite soil sample was taken for each sample wetland, and bulk density was not measured, which complicates the use of soil nutrient data. As well, generally wetlands were visited only once, with a complete sample effort lasting just one day. This provided a mere snapshot of wetland 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 discarded sample (generally for quality control reasons). Visiting these wetlands only once or twice did not 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 water, which allowed many flood intolerant species to encroach into the sample wetlands. While the FWCIs have 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 FWCIs to test the reliability of these indices. The FWCI was limited to 19 metrics due to the redundant nature of many of the candidate metrics, as well as the high variability of community composition within the dataset. A larger sample size could improve the significance of the FWCI based on ecoregions for metrics such as indicator species analysis. Regionalization may be an important step in refining the FWCI, as this study was somewhat limited to a statewide approach due to small sample sizes within each ecoregion. Conclusions The use of three separate species assemblages for a biological assessment of isolated depressional forested wetland provided a useful tool for detecting changes in biological integrity associated with changes in the driving energies of a wetland measured through landscape development intensity. While richness, evenness, and diversity measures were not particularly sensitive to changes in landscape development intensity, biological indicators along with physical and chemical parameters were useful in defining biological integrity. In the future multiple assemblage specific multi-metric FWCIs could be constructed for all freshwater wetlands throughout the state of Florida, with specific Indicator Species and metric scoring criteria based on Florida wetland regions. While the FWCI suggests low biological integrity of both agricultural and urban wetlands, these wetlands provide services and do work in the environment. Therefore, the quantitative score of biological integrity established through the FWCI should not be used as a surrogate for wetland value, but an objective, quantitative means of comparing changes in community composition along gradients of development intensity.

PAGE 134

120 APPENDIX A 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 report. These additional methods were included in Appendix A to provide readers with a complete picture of field methodology, and to understand the order of events during field sampling. Data omitted from this report include tree basal area along transects, fisheye canopy photography, algae analysis of epiphyton, metaphyton, and phytoplankton, and dry benthic algae analysis. Canopy photo analysis was explored in a non-thesis project by Lisa Spurrier at the University of Florida in 2000. Laboratory identification of additional algae samples was not completed at the time of publication. While vegetation zone descriptions are provided for soil sampling, these procedures were initially created for use in freshwater isolated depressional herbaceous wetlands (Lane et al. 2003). Zonation for soil samples was only employed at three of the 118 forested wetlands included in this study. These three wetlands were characterized by open centers with no canopy trees occurred in the deep pooled center area of the wetland. In these instances soil samples were taken in both the outer forested zone and the inner marsh zone and analyzed separately. As suggested in the soil SOPs, the soil data values were weighted for each vegetation zone based on the approximate area occupied by each vegetation zone. ORDER OF FIELD EVENTS FOR ISOLATED DEPRESSIONAL FORESTED WETLANDS: 1. Water quality is ALWAYS taken first. Two field crew members take the water samples. One records data while the other takes the sample. 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) unloads the vehicle and prepare(s) field equipment. 3. After the water samples are obtained (follow SOPs for water quality), complete the Site/Habitat Characterization Data Sheet & WRAP assessment. 4. When completed, establish vegetation transects. This includes delineating the upland/wetland boundary and running all four transects (follow SOPs for vegetation). 5. The remaining field crew should: Collect algae samples (follow SOPs for algae) Collect macroinvertebrates (follow SOPs for macroinvertebrates) Collect soil samples (follow SOP for soil) 6. Take site photographs and GPS point

PAGE 135

121 CHECKLIST OF MATERIALS/FIELD EQUIPMENT: Miscellaneous o SOPs o Large cooler with frozen ice bottles for soils and vegetation o Digital camera o Waders (knee, hip, etc.) o Garmin III GPS unit o Aerial photo & FLUCCS codes of site o Clipboard, pencils, sharpies, masking tape, plastic bags, index cards, hand lens o Data Sheets: Water Quality: FDEP Central Lab submittal form Site Characterization & WRAP sheets 1 per person per site Bioassessment macrophyte data sheets a minimum of 8 per site Water Quality o Small cooler with ice (can purchase for shipping) o YSI meter (DO/temp) o 2 500-mL water sample bottles turbidity/color/conductivity/pH & NH3/NOX/TKN/TP (or other sizes as supplied by FDEP) o Pipette for H2SO4 o Bottle of 1:1 H2SO4 o FedEx or DHL air bills Transects & Macrophytes o 2 100m transect tapes o 1 m PVC o 2-3 compasses o Plant press, newspaper, and cardboard o Field ID manuals o Prism for basal area 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 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

PAGE 136

122 Canopy Photos o Digital camera o Spare batteries o Film disks o Tripod o Compass o Height pole Soils o 3-inch diameter PVC pipe o Knife o Piece of wood o Dampened hammer o Buckets o Stainless steel spoon o De-ionized water

PAGE 137

123 SOPs for Isolated Depressional 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/conductivity/pH and one 500 mL bottle for NH3/NOx/TKN/TP are collected per site. These 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 stirring up organic material and silt. 7. Remove cap from each 500 mL bottle without touching the lip or interior surfaces. 8. Rinse the bottle 3 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 and press on bottom of bottle to allow water to slowly flow inside. 11. Be deliberate, making sure that no suspended 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 stopped, carefully lift bottle out of water. 13. Repeat, so both 500 mL bottles are full. 14. After the water is collected, take dissolved oxygen and temperature readings using the YSI meter. Take measurements within the top 10 cm of the water column. Apply constant, gentle motion to the dissolved oxygen probe, as the meter is consuming oxygen during measurement. 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 Submittal 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 1-800-GO-FEDEX to find a nearby office. 20. If you cannot get to FedEx in time, dump the samples taken. Repeat procedure another day.

PAGE 138

124 SOPs for Isolated Depressional Forested Wetlands: VEGETATION 1. Using a compass, locate the 4 cardinal point directions (north, 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 wetland 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 towards 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 side of the transect (1 m wide) and 5 m long, record all species present within 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 are collected, being sure to get plant inflorescence and roots, tagged with properly 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 used to protect especially sensitive parts. 8. All collected plants are identified in the field on the day of sampling and placed in a plant press for further clarification and identification. Plant nomenclature follows FDEPs 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 number 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 when 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 stake so that photo sites can be revisited in the future. 12. Center the tripod over the stake 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 photographers back is to the north.

PAGE 139

125 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 that the back display shows the canopy as a circle, surrounded by a dark/black border. 16. Record the photo number for the position (ex. north, south, east, or west). 17. Complete the canopy photo data sheet, noting time of day, cloud condition, surrounding vegetation, etc.

PAGE 140

126 SOPs for Isolated Depressional 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 keep each substrate type separate in their own collection jars. At the end of the collection 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 herbaceous 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 vegetation/habitat zones. 2. For woody debris (roots, snags): Using a brush, brush the wood for algae. Place brush in bag with water and shake algae off of the brush. Pipette 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; extract 10 mL of algal suspension and place in pea cup; since woody and herbaceous are within the same 10 aliquots, they must be divided 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 algae from the surface of the sediment; 3. Extract 10 mL of algal suspension and place in pea cup. METAPHYTON: 1. Collect approximately 100 mL of wetland water in a pea cup; 2. Using your fingers, collect a thumbnail size portion of the algal mat; 3. Obtain aliquots from 10 different 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 the 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 then bring tube out of water; 5. Carefully pour contents into the 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.

PAGE 141

127 ALL SAMPLES: 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 water to avoid cross-contamination at future sites. When at the Center for Wetlands, clean all 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: 1.Use bottomless pea cup to isolate a spot on the sediment. 2.Extract the upper 0.5 cm of soil into an additional pea cup, this depth is marked on the collection pea cup. 3.Add 100 mL deionized water and stir well with turkey baste. 4.Extract 10 mL of algal suspension and place in sample pea cup. 5.Repeat, so that you have collected 10 aliquots representative of the vegetation/habitat zones in the wetland. 6.Preserve the sample with 5 mL of M3 per 100 mL of algal suspension. 7.Properly label collection jar, identifying site, date, collector, and sample type. 8.Carefully clean equipment with DI water to avoid cross-contamination at future sites. When at the Center for Wetlands, clean all equipment with Clorox/water solution. 9.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 Methods 1. Dissolve 10 g iodine in a small quantity of deionized water to aid in solution of iodine. 2. Dissolve 5 g potassium iodide, 10 g dissolved iodine (from step 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 Examination of Water and Wastewater, 17th edition)

PAGE 142

128 SOPs for Isolated Depressional Forested Wetlands: MACROINVERTEBRATES 1. There are to be 20 sweeps (evenly divided into the vegetation/habitat zones) to send to FDEP unpicked for identification. 2. Always do your sweeps in undisturbed areas where you have not walked through yet. 3. A single sweep is one net width and two net lengths to equal 0.5 m2. 4. 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 all material into the net. If you are in a forested site, use a brush to clean any snags and roots of material. 5. Vigorously sample the area repeatedly (3 times) to ensure good coverage. 6. Dip net into water repeatedly, without letting the sample out, to try and sift the muck and silt through the net. 7. Do not sample in the muck! 8. 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. 9. Place masking tape over the lid to prevent leakage during travel & shipment. Properly label the jar with the site name, date, and collector. 10. Thoroughly clean all equipment off with water. 11. Return samples to room 8 at the Center for 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 preparing 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 bicarbonate into the container, close, and shake vigorously (at least 1 minute to ensure proper mixing). All of the sodium bicarbonate may not dissolve into the formalin, as this is a supersaturated solution. 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 formalin to smaller containers for field use. 8. Make sure that all containers are clearly labeled Buffered Formalin pH 7.5-8.0.

PAGE 143

129 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 Department of Environmental 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 (Hanna 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 probe in a beaker of deionized water. Rinse the electrode and probe off with 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 temperature 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 (1 inch) into the solution. 3. Press the CAL key. The CAL and buffer indicators will be displayed. The secondary LCD display should read .01. If not, adjust it using the c key. 4. After the pH reading becomes stable, the READY and CON indicators will blink. Once this happens, press the CON key to confirm the calibration. 5. After rinsing with deionized water, immerse the pH electrode and temperature probe into the pH 7.00 solution, and stir briefly. 6. Select the second buffer value (.01) on the secondary display using the c key. 7. After the pH reading becomes stable, the 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 the electrode and probe dry 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 if greatest accuracy is required whenever the batteries have been replaced.

PAGE 144

130 SOPs for Isolated Depressional 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 cores taken per zone is generally 4, 1 along each transect. In some cases there will be fewer than 4 cores per vegetation zone as the zones may not all 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 appears 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 circular shape around the sampling pipe, so that the pipe will easily slide through the soil and roots. 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 marked with the name of the vegetation zone. 3. Repeat along each transect for each vegetation 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 the large stainless steel spoon. Clean and rinse the spoon with de-ionized 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, take 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 larger 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 sheet. 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 Wetlands, and store in the refrigerator in the back lab pending laboratory analysis.

PAGE 145

APPENDIX B COEFFICIENT OF CONSERVATISM SCORES Table B-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 6.3 131

PAGE 146

132 Table B-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 147

133 Table B-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 148

134 Table B-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 149

135 Table B-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 150

136 Table B-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 151

137 Table B-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 152

APPENDIX C CANDIDATE METRICS Table C-1. Candidate metrics based on the diatom assemblage. Tolerance Metrics Indicator Species Sensitive Taxa Abundance Indicator Species Sensitive Taxa Presence/Absence Indicator Species Tolerant Taxa Abundance Indicator Species Tolerant Taxa Presence/Absence Community Composition Metrics Richness Evenness Shannon Diversity Simpson's Index Autecological Metrics Morphological Guild Erect Morphological Guild Stalked Morphological Guild Unattached Morphological Guild Prostrate/Adnate Morphological Guild Variable Motility Highly Motile Motility Moderately Motile Motility Highly & Moderately Motile Motility Not Motile Motility Variable Pollution Tolerance Very Tolerant Pollution Tolerance Moderately Tolerant Pollution Tolerance Very & Moderately Tolerant Pollution Tolerance Sensitive / Intolerant Dissolved Oxygen Class 1 (Bahls 1993) Dissolved Oxygen Class 2 (Bahls 1993) Dissolved Oxygen Class 3 (Bahls 1993) Dissolved Oxygen Class 4 (Bahls 1993) Dissolved Oxygen Class 5 (Bahls 1993) Wet/Dry Preference Class 1 (van Dam et al. 1994) Wet/Dry Preference Class 2 (van Dam et al. 1994) 138

PAGE 153

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

PAGE 154

140 Table C-2. Candidate metrics based on the macrophyte assemblage. Wetland Plant Status Obligate Species Facultative Wetland Species Facultative Species Facultative Upland Species Upland Species Obligate + Facultative Wetland Species Obligate + Facultative Wetland + Facultative Species Facultative Upland + Upland Species Upland + Facultative Upland + Facultative Species Plant Growth Form & Taxa Metrics Graminoid Species Carex sp. Herbaceous Species Species in Asteraceae Polygonum sp. Graminoids to Herbaceous Vine Species Vines that are Woody Shrub Species Tree Species Tree and Shrub Species Salix sp. Hardwoods Trees as Hardwoods Nyssa sp. Trees as Nyssa sp. Acer rubrum Trees as Acer rubrum Trees as Conifers Trees as Taxodium sp. Native Evergreen Shrubs Native Ferns Native Perennial Graminoids Native Perennial Herbs

PAGE 155

141 Table C-2. Continued. 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) 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) TWINSPAN Sensitive Species TWINSPAN Sensitive Species by Presence Absence TWINSPAN Tolerant Species TWINSPAN Tolerant Species by Presence Absence TWINSPAN Sensitive Species, Excluding Exotic Species TWINSPAN Sensitive Species by Presence Absence, Excluding Exotic Species TWINSPAN Tolerant Species, Excluding Exotic Species TWINSPAN Tolerant Species by Presence Absence, Excluding Exotic Species Indicator Species Sensitive Taxa Occurrence Indicator Species Sensitive Taxa Presence/Absence Indicator Species Tolerant Taxa Occurrence Indicator Species Tolerant Taxa Presence/Absence Indicator Species Sensitive Taxa, Excluding Exotic Species Occurrence Indicator Species Sensitive Taxa, Excluding Exotic Species Presence/Absence Indicator Species Tolerant Taxa, Excluding Exotic Species Occurrence Indicator Species Tolerant Taxa, Excluding Exotic Species Presence/Absence Modified FQI Score Exotic Species Metric Longevity Metrics Annuals Native Annuals Perennials Native Perennials

PAGE 156

142 Table C-2. Continued. Longevity Metrics continued Annual to Perennial Ratio Native Annual to Native Perennial Ratio Richness Metrics Species Richness by Site Species Richness by Quadrat Species Richness by Occurrence Species Richness by Transect Mean Site Evenness Dominant Species Log (Proportion of Dominant Species) Vascular Genera Nonvascular Genera

PAGE 157

143 Table C-3. Candidate metrics based on the macroinvertebrate assemblage. Tolerance Metrics Indictor Species Analysis Sensitive Taxa Abundance Sensitive Taxa Presence/Absence Tolerant Taxa Abundance Tolerant Taxa Presence/Absence Florida Index Lake Condition Index Community Structure & Balance Metrics Mixed Taxonomic Levels Crustacea + Mollusca Dominant Taxa Exotic Richness Taxa Richness Tubificida/Insecta Phylum Phylum Richness Annelida Arthropoda Mollusca Platyhelminthes Class Class Richness Arachnida Bivalva Crustacea Gastropoda Insecta Oligochaeta Plecypoda Turbellaria Order Order Richness Acariforrmes Amphipoda Anostraca Basommatophora Coleoptera Collembola Decapoda Diptera

PAGE 158

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

PAGE 159

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

PAGE 160

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

PAGE 161

APPENDIX D QUANTIFYING ANTHROPOGENIC INFLUENCE Table D-1. LDI, WRAP, and Minnesota disturbance 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 15 147

PAGE 162

148 Table D-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 163

APPENDIX E SUMMARY STATISTICS Table E-1. Summary statistics of richness (R), evenness (E), Shannon diversity (H), and Simpsons index (S) for the diatom 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 0.70 149

PAGE 164

150 Table E-2. Summary statistics of richness (R), jackknife estimators of species richness (Jack1, Jack2), evenness (E), Shannon diversity (H), and Whittakers beta diversity (W) for the macrophyte assemblage (species level). Site R Jack1Jack2E 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 165

151 Table E-2. Continued. Site R Jack1Jack2E 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 166

152 Table E-2. Continued. Site R Jack1Jack2E 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 167

153 Table E-3. Summary statistics of richness (R), evenness (E), Shannon diversity (H), and Simpsons index (S) for the macroinvertebrate 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 168

154 APPENDIX F METRIC SCORING FOR THE FLORIDA WETLAND CONDITION INDEX FOR DEPRESSIONAL FORESTED WETLANDS Scoring the Diatom Florida Wetland Condition Index 1. Calculate values for the 7 metrics: 1 Percent tolerant indicator species 2 Percent sensitive indicator species 3 Percent pollution class 1 4 Percent nitrogen class 3 5 Percent saprobity class 4 6 Percent pH class 3 7 Percent oxygen class 1 2. Take the natural log of metrics to improve distribution. = ln (metric value + 10) 3. Use the scoring equations to normalize scores between 0 and 10. Metrics that increase with increasing LDI tolerant, pollution class 1, nitrogen class 3, saprobity class 4, pH class 3 metrics: = 10 ( ( metric 5th percentile) ( 10 / ( 95th percentile 5th percentile))) Metrics that decrease with increasing LDI sensitive, oxygen class 1: = (( metric 5th percentile) ( 10 / ( 95th percentile 5th percentile))) Below are the 5th and 95th percentiles for each metric (transformed values are presented, see step 2 above): 5th Percentile 95th Percentile %Tolerant indicator species 2.30 2.35 %Sensitive indicator species 2.30 2.37 %Pollution class 1 2.30 4.11 %Nitrogen class 3 2.30 4.31 %Saprobity class 4 2.30 3.96 %pH class 3 2.36 4.48 %Oxygen class 1 3.06 4.66 4. Rescore, so that the metrics in the outer 5th percentiles receive scores of 0 or 10. = IF ( score < 0, 0, ( IF ( score <= 10, score, 10)))

PAGE 169

Scoring the Macrophyte Florida Wetland Condition Index 1. Calculate values for the 6 metrics: 1 Percent tolerant indicator species 2 Percent sensitive indicator species 3 Percent exotic species 4 Modified FQI Score 5 Percent native perennial species 6 Percent wetland status species 2. Take the natural log of metrics to improve distribution. For tolerant, sensitive, and exotics species metrics: = ln (metric value + 10) For modified FQI, native perennial, and wetland status use the straight metric values (do not take the natural log) 3. Use the scoring equations to normalize scores between 0 and 10. Metrics that increase with increasing LDI tolerant and exotic metrics: = 10 ( ( metric 5th percentile) ( 10 / ( 95th percentile 5th percentile))) Metrics that decrease with increasing LDI sensitive, modified FQI, native perennial, and wetland status: = (( metric 5th percentile) ( 10 / ( 95th percentile 5th percentile))) Below are the regionalized 5th and 95th percentiles for each metric (transformed values are presented, see step 2 above): %Tolerant Indicator Species Region 5th Percentile 95th Percentile Panhandle 2.30 4.16 North 2.30 4.13 Central 2.64 4.02 South 2.42 3.91 %Sensitive Indicator Species Region 5th Percentile 95th Percentile Panhandle 2.30 4.28 North 2.51 4.17 Central 2.30 4.04 South 2.30 4.10 155

PAGE 170

156 %Exotic Species Region 5th Percentile 95th Percentile Panhandle 2.30 3.63 North 2.30 3.72 Central 2.30 3.58 South 2.59 3.71 Modified FQI Score Region 5th Percentile 95th Percentile Panhandle 2.68 6.11 North 2.55 5.70 Central 2.83 4.83 South 2.73 5.08 %Native Perennial Species Region 5th Percentile 95th Percentile Panhandle 67.72 100.00 North 60.00 100.00 Central 63.57 95.55 South 61.94 95.73 %Wetland Status Species Region 5th Percentile 95th Percentile Panhandle 38.28 77.09 North 47.66 83.01 Central 41.96 80.01 South 33.33 87.05 4. Rescore, so that the metrics in the outer 5th percentiles receive scores of 0 or 10. = IF ( score < 0, 0, ( IF ( score <= 10, score, 10)))

PAGE 171

157 Scoring the Macroinvertebrate Florida Wetland Condition Index 1. Calculate values for the 6 metrics: 1 Percent tolerant indicator genera 2 Percent sensitive indicator genera 3 Florida Index 4 Percent Mollusca 5 Percent Noteridae 6 Percent scrapers 2. Take the natural log of metrics to improve distribution. = ln (metric value + 10) 3. Use the scoring equations to normalize scores between 0 and 10. Metrics that increase with increasing LDI tolerant, Mollusca, scrapers: = 10 ( ( metric 5th percentile) ( 10 / ( 95th percentile 5th percentile))) Metrics that decrease with increasing LDI sensitive, Florida Index, Noteridae: = (( metric 5th percentile) ( 10 / ( 95th percentile 5th percentile))) Below are the 5th and 95th percentiles for each metric (transformed values are presented, see step 2 above): 5th Percentile 95th Percentile %Tolerant Indicator Genera 2.30 4.04 %Sensitive Indicator Genera 2.30 3.73 Florida Index 2.30 2.71 %Mollusca 2.30 3.73 %Noteridae 2.30 3.81 %Scraper 2.30 2.66 4. Rescore, so that the metrics in the outer 5th percentiles receive scores of 0 or 10. = IF ( score < 0, 0, ( IF ( score <= 10, score, 10)))

PAGE 172

LIST OF REFERENCES Adams, S.M. 2002. Biological indicators of aquatic ecosystem stress: introduction and overview. Pages 1-11 in S.M. Adams, editor. Biological indicators of aquatic ecosystem stress. American Fisheries Society. Bethesda, Maryland, USA. Adamus, P.R. 1996. Bioindicators for assessing 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 bio-monitoring data. EPA/600/3-90/073. Office of Research and Development, United States Environmental Protection Agency. 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. version 1.67. Leeds, England, United Kingdom. Andreas, B.K. and R.W. Lichvar. 1995. A floristic assessment system for northern Ohio. Wetlands Research Program Technical Report WRP-DE-8. U.S. Army Corps of Engineers Waterways Experiment Station, 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 Research Institute, Inc. 1999. Neuron Data, Inc. 1991-1996. Portions copyright 1991-1995 Arthur D. Applegate. Found at: http://www.esri.com/. Redlands, California, USA. Bahls, L. 1993. Periphyton bioassessment methods for Montana streams. Water Quality Bureau, Department of Health and Environmental 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 biological criteria for Florida streams using benthic macroinvertebrates. Journal of the North American Benthological Society 15(2): 185-211. 158

PAGE 173

159 Barbour, M.T., J. Gerritsen, B.D. Snyder, and J.B. Stribling. 1999. Rapid bioassessment protocols for use in streams and wadeable 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 pollution 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. Patterns in nutrient availability and plant diversity of temperate North American wetlands. Ecology 80(7): 2151-2169. 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. Gainesville, 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 swamps. University Presses of Florida, Gainesville, Florida, USA. Brown, M.T. and M.B. Vivas. 2005. A Landscape Development Intensity Index. Environmental Monitoring and Assessment 101: 289-309. Brown, M.T. and S. Ulgiati. 2005. Emergy, transformity, and ecosystem health. Pages 333-352 in S.E. Jrgensen, R. Costanza, and F. Xu, editors. Handbook of ecological indicators for assessment of ecosystem health. Taylor and Francis, Boca Raton, Florida, USA. 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 Florida. 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 forests ecoregion. Ecological Indicators 3: 181-193. Cairns, J. and J.R. Pratt. 1993. A history of biological monitoring using benthic macroinvertebrates. Pages 10-27 in D.M. Rosenberg and V.H. Resh, editors. Freshwater biomonitoring and benthic macroinvertebrates. 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. Environment 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 bioassessment indicators. Patrick Center of Environmental Research, Environmental Research Division. The Academy of Natural Sciences, Philadelphia, Pennsylvania, USA. Clarke, K.R. 1993. Non-parametric multivariate 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 agricultural 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.

PAGE 174

160 Coultas, C.L. and M.J. Duever. 1984. Soils of Cypress Swamps. Pages 51-59. K.C. Ewel and H.T. Odum, editors. Cypress swamps. 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 United States. United States Department of the Interior, Fish and Wildlife Service, Washington, D.C., USA. Cronk, J. K. and M.S. Fennessy. 2001. Wetland plants: biology and ecology. Lewis Publishers. Boca Raton, Florida, USA. Crowder, A. and D.S. Painter. 1991. Submerged macrophytes in Lake Ontario: current knowledge, importance, threats to stability, and needed studies. Canadian Journal of Fisheries and Aquatic Sciences 48:1539-1545. Cummins, K.W. and R.W. Merritt. 2001. Application of invertebrate functional groups to wetland ecosystem function and biomonitoring. Pages 85-111 in R.B. Rader, D.P. Batzer, and S.A. Wissinger, editors. Bioassessment and management of North American freshwater wetlands. John Wiley and Sons, Inc., New York, New York, USA. Dahl, T.E. 2000. Status and trends of wetlands 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 monitoring and assessing biological integrity of inland freshwater wetlands. EPA 843-R-98-002. Wetlands Division Office of Water, United States Environmental Protection Agency, Washington, D.C., USA. Danielson, T.J. 1998b. Wetland bioassessment fact sheets. EPA 846-98-001. Wetlands Division Office of Water, United States 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. Spatial and temporal variability of trophic relationships among aquatic macroinvertebrates 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 Agriculture, Forest Service, Southern Research Station-19. Asheville, NC, USA. Devine, R. 1998. Alien invasion. National 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. Pages 112-118 K.C. Ewel and H.T. Odum, editors. Cypress swamps. University Presses of Florida, Gainesville, Florida, USA. Doherty, S.M., M. Cohen, C. Lane, L. Line, and J. Surdick. 2000. Biological criteria for inland freshwater wetlands in Florida: a review of technical and scientific literature (1990-1999). Report to the United States Environmental Protection Agency, Center for Wetlands, University of Florida, Gainesville, Florida, USA. Dufrne, 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. Ephemeroptera. 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. 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.

PAGE 175

161 Evans, D.L., W.J. Streever, and T.L. Crisman. 1999. National flatwoods marshes and created freshwater marshes of Florida: factors influencing aquatic invertebrate distribution and comparisons between natural and created marsh communities. Pages 81-104 in D.P. Batzer, R.B. Rader, 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 wastewater 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 Central Florida Press, Orlando, Florida, USA. Ewel, K.C., H.J. Davis, and J.E. Smith. 1989. Recovery of Florida 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 riparian 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 assess 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. Anthropogenic 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 Conservation and Development and the Division of State Lands. Fishman Environmental Services. Portland, Oregon, USA. Fore, L.S. 2004. Development and testing of biomonitoring tools for macroinvertebrates in Floirda streams. Statistical Design, Seattle, Washington. A report for the Florida Department of Environmental Protection, Tallahassee, Florida, USA. Fore, L.S. and C. Grafe. 2002. Using diatoms to assess the biological condition of large rivers in Idaho (U.S.A.). Freshwater Biology 47: 2015-2037. Francis, C.M., M.J.W. Austen, J.M. Bowles, W.B. Draper. 2000. Assessing floristic quality in southern Ontario woodlands. Natural Areas Journal 20:66-77. Frappier B. and R.T. Eckert. 2003. Utilizing 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 America. 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 Minnesota wetlands. Journal of Aquatic Ecosystem Stress and Recovery 6: 217-234. Gardner, W.H. 1986. Water content. Pages 493-544 in Methods of soil analysis: part 1 physical and mineralogical methods, 2nd edition. 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 wetlands. Final Report to the United States Environmental Protection Agency Assistance Number CD995525-01, April 1999. Minnesota Pollution Control Agency, St. Paul, Minnesota, USA.

PAGE 176

162 Gerristen, J., M.T. Barbour, and K. King. 2000. Apples, oranges, and ecoregions: on determining pattern in aquatic assemblages. Journal of the North American Benthological Association 19(3): 487-496. Gerristen, J. and J. White. 1997. Development of a biological index for Florida lakes. A Report to the Florida Department of Environmental 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 Georgia 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 Wiley and Sons, Inc., New York, New York, USA. Gore, J.A., H.L. Griffith, III. and D.S. Addison. 1998. Inventory of the freshwater macroinvertebrates in hydric pine flatwoods for the district's isolated wetland monitoring program. South Florida 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, Florida. Hydrobiologia 379: 111-122. Griffith, G.E., J.M. Omernik, C.M. Rohm, and S.M. Pierson. 1994. Florida Regionalization Project. Environmental Research Laboratory, United States Environmental Protection Agency, Corvallis, OR. Griffith, M.B., P. Husby, R.K. Hall, P.P. Kaufmann, and B.H. Hill. 2003. Analysis of macroinvertebrate assemblages in relation to environmental gradients among lotic habitats of Californias Central Valley. Environmental Monitoring and Assessment 82: 281-309. Haag, K.H, J.C. Joyce, W.M. Hetrick, and J.C. Jordan. 1987. Predation on Water hyacinth weevils and other aquatic insects 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 perturbations. 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. Diatoms: cleaning and mounting for light and electron microscopy. Transactions of the American Microscopy Society 89 (4): 469-474. Helgen, J. 2001. Methods for evaluating wetland condition: #9 developing an invertebrate index of biological integrity for wetlands. 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 assessment: development and application in the state of Michigan (USA). Natural Areas Journal 17:265-279. Hilsenhoff, W.L. 1987. An improved biotic index of organic stream pollution. Great Lakes Entomology 20: 31-39. Hobbs, R.J. and L.F. Hueneke. 1992. Disturbance, diversity, and invasion: implications for conservation. Conservation Biology 6(3): 324-337. James, M.O. and K.M. Kleinow. 1994. Trophic transfers of chemicals in the aquatic environment. Pages 1-35 in D.C. Malins and G.K. Ostrander, editors. Aquatic toxicology and cellular perspectives. Lewis Publishers. Boca Raton, Florida, USA.

PAGE 177

163 Kantrud, H.A. and W.E. Newton. 1996. A test 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 integrity using fish communities. Fisheries 6: 21-27. 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. Biological 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. Ecological perspectives on water quality goals. Environmental Management 5: 55-68. Keddy, P.A. 2000. Wetlands ecology: principles and conservation. Cambridge University Press, Cambridge, United Kingdom. Kelly, M.G. and B.A. Whitton. 1998. Biological 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 technology. 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. Ecological Applications 4(4): 768-785. Kruskal, J.B. 1964. Multidimensional scaling by optimizing goodness of fit to a non-metric hypothesis. Psychometrika 29: 115-129. Kushlan, J.A. 1990. Freshwater marshes. Pages 324-363 in Myers, R.L. and J.J. Ewel, editors. Ecosystems of Florida. University 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. 2003. The development of the Wetland Condition Index for Florida. A Report to the United States Environmental Protection Agency, Biological Assessment of Wetlands Workgroup. Center for Wetlands, University of Florida, 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 Environmental Quality 13(4): 602-604. Lenat, D.R. 1993. A biotic index for the southeastern United States: derivations and list of tolerance values, with criteria for assigning water-quality ratings. Journal of the North American Benthological Society 12(3): 279-290. Lenat, D.R. and M.T. Barbour 1994. Using benthic macroinvertebrate community structure for rapid, cost-effective, water quality monitoring: rapid bioassessment. Pages 187-215 in S.L. Loeb and A. Spacie, editors. Biological monitoring of aquatic systems. Lewis Publishers, Boca Raton, Florida, USA. Leslie, A.J., J.P. Prenger, and T.L. Crisman. 1999. Cypress domes in North Florida: invertebrate ecology and response to human disturbance. Pages 105-119 in D.P. Batzer, R.B. Rader, and S.A. Wissinger, editors. Invertebrates in freshwater wetlands of North America. John Wiley and Sons, Inc., New York, New York, USA.

PAGE 178

164 Lugo, A.E., and S.L. Brown. 1986. The Ocklawaha 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. Mack, J. 2001. Vegetation Index of Biological Integrity (VIBI) for wetlands: ecoregional, hydrogeomorphologic, and plant community comparisons with preliminary wetland aquatic life use designations. Final Report to the United States Environmental Protection Agency 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.wetlands/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. CD985276, 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.wetlands/wetlands_bioasses.html. Marois, K.C. and K.C. Ewel. 1983. Natural and management related variation in cypress domes. Forest Science 29(3):627-640. McCormick, P.V. and J. Cairns, Jr. 1994. Algae 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 Washington, USA. Bryologist 103: 417-427. Means, D.B., J.G. Palis, and M. Baggett. 1998. Effects of slash pine silviculture on a Florida population of flatwoods salamander. Conservation Biology 10(2): 426-437. Merritt, R.W. and K.W. Cummins. 1996. Introduction. 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. Vandeneeden. 1996. Procedures for the functional analysis of invertebrate communities of the Kissimmee River-floodplain ecosystem. Florida Scientist 59(4): 216-274. Miller, R.E., Jr. and B.E. Gunsalus. 1999. Wetland rapid assessment procedure. South Florida Water Management District, Technical Publication REG-001. 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. Wetlands, 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 agriculturally dominated landscape. Journal of the North American Benthological Society 21 (2): 221-237. Mushet, D.M., N.H. Euliss, and T.L. Shaffer. 2002. Floristic quality assessment of one natural and three restored wetland complexes in North Dakota, USA. Wetlands 22(1): 126-138. Nessel, J.K., K.C. Ewel, and M.S. Burnett. 1982. Wastewater enrichment increases mature pondcypress growth rates. Forest Science 28(2):400-403. OConnell, T.J., L.E. Jackson, and R.P. Brooks. 1998. A bird community index of biotic integrity for the mid-Atlantic highlands. Environmental Monitoring and Assessment 51: 145-156.

PAGE 179

165 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 Coordinating Council on the Restoration of the Kissimmee River Valley and Taylor Creek-Nubbins Slough Basin. Tallahassee, Florida, USA. Odum, H.T. 1995. Environmental Accounting: Emergy and Environmental Decision Making. John Wiley and Sons, New York, New York, USA. Odum, H.T. 1994. Ecological and general systems: an introduction to systems ecology. University Press of Colorado, Niwot, Colorado, USA. Ott, R.L. and M. Longnecker. 2001. An introduction to statistical methods and data analysis, 5th edition. Duxbury, Wadsworth Group. Pacific Grove, California, USA. Pan, Y. and R.J. Stevenson. 1996. Gradient analysis of diatom assemblages in western Kentucky wetlands. Journal of Phycology 32: 222-232. Pan, Y., R.J. Stevenson, B.H. Hill, A.T. Herlihy, and G.B. Collins. 1996. Using diatoms as indicators of ecological conditions in lotic systems: a regional assessment. Journal of the North American Benthological Society 15(4): 481-495. 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. Penton, and D.J. Conklin, Jr. 1993. Freshwater macroinvertebrates of northeastern North America. Comstock Publishing Associates, Cornell University Press. Ithaca, New York, USA. Pickard, D.P. and A.C. Benke. 1996. Production dynamics of Hyalella azteca (Amphipoda) among different habitats in a small wetland in the southeastern USA. Journal of the North American Benthological 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. Response of macroinvertebrates and small fish to nutrient enrichment in the northern 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. Reiss, K.C. 2004. Developing Biological Inidicators for Isolated Depressional Forested Wetlands in Florida. Ph.D. Dissertation, University of Florida, Gainesville, Florida, USA. Resh, V.H. and J.K. Jackson. 1993. Rapid assessment approaches in benthic macroinvertebrate biomonitoring studies. Pages 195-233 in D.M. Rosenberg and V.H. Resh, editors. Freshwater biomonitoring 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. Los Alamos, New Mexico, USA. SAS Institute, Inc. (SAS). 1990. SAS users guide, version 6, 4th edition. SAS Institute, Inc. Cary, North Carolina, USA. Schindler, D.W. 1987. Detecting ecosystem responses to anthropogenic stress. Canadian Journal of Fisheries and Aquatic Sciences 44:6-25. Schulz, E.J., M.V. Hoyer, and D.E. Canfield, Jr. 1999. An index of biotic integrity: a test with limnological and fish data from sixty 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 1-22 in Batzer, D.P., R.B. Rader, and S.A. Wissinger, editors. Invertebrates in freshwater wetlands of North America. John Wiley and Sons, Inc. New York, New York, USA.

PAGE 180

166 Smogor, R.A. and P.L. Angermeier. 2001. Determining a regional framework for assessing biotic integrity of Virginia streams. Transactions of the American Fisheries Society 130:18-35. Snyder, B.D., M.T. Barbour and E.W. Leppo. 1998. Development of a watershed-based approach for biomonitoring of fresh surface waters in southern Florida canal systems. Prepared for Metro-Dade Environmental Resources Management, Environmental Monitoring Division, Miami, FL, USA. Stansly, P.A., J.A. Gore, D.W. Ceilley and M.B. Main. 1997. Inventory of freshwater macroinvertebrates. Contract # C-7949 Final Report for the South Florida Water Management District Isolated Wetland Monitoring Program. West Palm Beach. Florida, USA. Stevenson, R.J. 2001. Using algae to assess 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 freshwater 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 INTECOLs 5th International Wetland Conference. Gleneagles Press, Adelaide, Australia. Stevenson, R.J., and B. Wang. 2001. Developing 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 land use. Natural Areas Journal 19: 155-165. Tabachnick, B.G. and L.S. Fidell. 1983. Using multivariate statistics. Harper and Rox Publishers. Philadelphia, Pennsylvania, USA. ter Braak, C.J.F. 1987. The analysis of vegetation-environmental relationships by canonical correspondence analysis. Vegetatio 69: 69-77. Terwilliger, V.J. and K.C. Ewel. 1986. Regeneration and growth after logging in Florida cypress domes. Forest Science 32: 493-506. Tobe, J.D., K. Craddock Burks, R.W. Cantrell, M.A. Garland, M.E. Sweeley, D.W. Hall, P. Wallace, G. Anglin, G. Nelson, J.R. Cooper, D. Bickner, K. Gilbert, N. Aymond, K. Greenwood, and N. Raymond. 1998. Florida wetland plants: an identification manual. Florida Department of Environmental Protection, Tallahassee, Florida, USA. Toth, L.A. 1993. The ecological basis of the Kissimmee River Restoration Plan. Florida Scientist 56(1): 25-51. United States Department of Agriculture, Natural Resource Conservation Service (USDA NRCS). 2002. The PLANTS Database, Version 3.5. National Plant Data Center, Baton Rouge, Louisiana, USA. 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-automated 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, Planning and Evaluation. Washington, D.C., USA. United States Environmental Protection Agency (USEPA). 1993. 365.1 Determination of phosphorus by semi-automated colorimetry methods for the determination of inorganic substances in environmental samples. EPA-600-R-93-100. Found at: http://www.epa.gov/epahome/index/. Boston, Massachusetts, USA. United States Environmental Protection Agency (USEPA). 1998. Lake and reservoir bioassessment and biocriteria: EPA 841-B-98-007 Technical Guidance Document. Washington, D.C., USA. Available on-line at: http://www.epa.gov/owow/monitoring/tech/lakes.html. Accessed 2002-2004.

PAGE 181

167 United States Environmental Protection Agency (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 Agency (USEPA). 2002a. Methods for evaluating wetland condition: introduction to biological assessment. EPA-822-R-02-014. Office of Water, Washington, D.C., USA. United States Environmental Protection Agency (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. Biological Indicators of Watershed Health. Available on-line at: http://www.epa.gov/bioindicators. Accessed 2003-2004. van Dam, H., A. Mertens, and J. Sinkeldam. 1994. A coded checklist and ecological indicator values of freshwater diatoms from the Netherlands. 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 freshwater invertebrates of North America. The McDonald and Woodward Publishing Company, Blacksburg, Virginia, USA. Vymazal, J. and C. J. Richardson. 1995. Species composition, biomass, and nutrient content of periphyton in the Florida Everglades. Journal of Phycology 31(3): 343-354. Wallace, J.B., J.W. Grubauch, and M.R. Whiles. 1996. Biotic indices and stream ecosystem processes: results from an experimental study. Ecological Applications 6(1): 140-151. Wharton, C.H., W.M. Kitchens, E.C. Pendleton, and T.W. Sipe. 1982. The ecology of bottomland hardwood swamps of the Southeast: a community profile. FWS/OBS-81/37. Fish and Wildlife Service, United States 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. Aquatic 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 assemblages 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: 1878-1884. Wilhelm, G. and D. Ladd. 1988. Natural Area Assessment in the Chicago Region. Pages 361-375 in Transactions of the 53rd North American Wildlife and Natural Resource Conference, Louisville, Kentucky. Wildlife Management Institute, Washington D.C., USA. 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 Department of Agriculture Handbook. Washington, D.C., USA. Williams, D.D. and B.W. Feltmate. 1992. Aquatic insects. C-A-B International, Redwood Press Ltd., Melksham, United Kingdom. Winter, J.G. and H.C. Duthie. 2000. Epilithic 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.

PAGE 182

168 Wunderlin, R. P., and B. F. Hansen. 2003. Atlas of Florida Vascular Plants. S. M. Landry and K. N. Campbell, application 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, 4th 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 effects of prairie fire. Ecology 66(2): 606-611.


The Florida Wetland Condition Index (FWCI): Developing Biological Indicators for Isolated Depressional Forested Wetlands
CITATION THUMBNAILS PDF VIEWER PAGE IMAGE ZOOMABLE
Full Citation
STANDARD VIEW MARC VIEW
Permanent Link: http://ufdc.ufl.edu/AA00004284/00001
 Material Information
Title: The Florida Wetland Condition Index (FWCI): Developing Biological Indicators for Isolated Depressional Forested Wetlands
Physical Description: Report
Language: English
Creator: Reiss, Kelly Chinners
Brown, Mark T.
Publisher: Center for Wetlands
Publication Date: 2005
 Subjects
Subjects / Keywords: forested wetlands
Florida Wetland Condition Index (FWCI)
bioassessment
biological indicators
swamps
Landscape Development Intensity Index
Spatial Coverage: United States -- Florida
 Notes
General Note: 168 Pages
 Record Information
Source Institution: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
System ID: AA00004284:00001

Downloads

This item has the following downloads:

( PDF )


Full Text











The Florida Wetland Condition Index (FWCI):
Developing Biological Indicators for Isolated Depressional Forested Wetlands







Report Submitted to the
Florida Department of Environmental Protection
Under Contract #WM-683










By

Kelly Chinners Reiss
and
Mark T. Brown

June 2005


Howard T. Odum Center for Wetlands
University of Florida
Gainesville, Florida 32611-6350













ACKNOWLEDGMENTS


This research on biological indicators was supported by a grant to Mark T.
Brown, principle investigator, from the Florida Department of Environmental Protection
(FDEP). The FDEP staff provided support for this research, particularly Russ
Frydenborg, Ellen McCarron, Ashley O'Neal, Julie Espy, Tom Frick, Joy Jackson, Liz
Miller, Johnny Richardson, and Lori Wolfe. FDEP staff served as reviewers for the draft
report, including Julie Espy, Denise Miller, Vicki Tauxe, Nia Wellendorf, and David
Whiting.
Additionally, acknowledgement is due to the systems ecology research group at
the H.T. Odum Center for Wetlands for stimulating discussion and valuable input along
the way. In particular, Chuck Lane (who worked on biological indicators for Florida
isolated marshes) provided groundwork for this analysis. Assistance in field-data
collection, laboratory analysis, data entry, and/or feedback 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 Vivas, and Melissa Yonteck, was
particularly valuable.
























This project and the preparation of this report were funded in part by a Section 319
Nonpoint Source Management grant from the U.S. Environmental Protection Agency
through a contract with the Florida Department of Environmental Protection.












TABLE OF CONTENTS

Page

A C K N O W L E D G E M E N T S ......... .. ............................................................... .............. ii

LIST OF TABLES ......... ........ ................................... .......... .... ............ vi

LIST OF FIGURES ......... ......................... ...... ........ ............ ix

EXECUTIVE SUMMARY .. ...................................................... ............... xi

1 INTRODUCTION AND OVERVIEW ..................................... ...................

Review of Isolated Depressional Forested Wetlands.......................... ...............1....
Changes in H ydrology .......................................... ........................... 4
Increased Inflows of Nutrients and/or Toxins ..........................................5
Physical D isturbance............... ........................................ ...... ........ .. 6
Biological Indicators of Ecosystem Integrity ........................................ ...............7
Diatoms as Biological Indicators............... ......... ....... .. ........ .....9
M acrophytes as Biological Indicators................................... ............... 10
M acroinvertebrates as Biological Indicators .......................................... 12
Quantifying Anthropogenic Influence .... .......... ........................................ 15
Landscape Development Intensity Index................................................ 15
Wetland Rapid Assessment Procedure ............................................... 16
M innesota D disturbance Index ....................................... ............... 16
Project O overview .............................. ............ ........... .. ............ 17

2 M E T H O D S ................................................................18

Site Selection ........................................18
Gradients of Anthropogenic Activity.............. ............................................... 22
Field-data Collection............................................. 23
W after Sam ples ........................ .. ....................... .. ..... ........... 23
Soil Samples.............. . .. ................ ..... .. ......24
Sam pling D design ......... ..... .............. ..... ........ .. .... .. ............ 25
Data Analysis ............ ......... .. .................. 30
Water and Soil Parameters...... ..................... ...............30
Summary Statistics ............ ............. ................ .... 30
Regional Com positional Analysis................................... ............... 32
Com m unity Com position...................................... ......................... 32
M etric D evelopm ent ........................................... .................. ............... 33
Indicator Species A nalysis..................................... ......... ............... 34
D iatom M etrics ............................ .............. ............. .... ...........34









M acrophyte M etrics ......................................................... ..................35
M acroinvertebrate M etrics..................................... ......... ............... 36
Florida W etland Condition Index ........................................ ....... ............... 37

3 R E S U L T S ................................................................... 3 8

Gradients of Anthropogenic Activity .................... ..................... ............... .38
W ater and Soil Param eters......................................................... ............... 38
D iatom A ssem blage .................. ................. .... ...... .. ........ .............. 42
Sum m ary Statistics .... ..... .... ........................ ............ .. .. .. ........ .... 42
C om positional A nalysis....................................... .......................... 42
Community Composition...................... .... ........................... 43
M etric Selection ........................ .................... ..................... .. ...... .. 46
Florida Wetland Condition Index Diatoms....................................... 52
C luster A analysis ................. ................. .......... .... .. .... ...... .. ........ .... 55
M acrophyte A ssem blage............................................... ............................. 57
Sum m ary Statistics.............................. ......... ................ ............... 57
C om positional A nalysis....................................... .......................... 58
Community Composition...................... .... ........................... 60
M etric Selection .......................... .... ... .... ....... .. .............. 62
Macrophyte Florida Wetland Condition Index .......................................80
C luster A analysis ................. ................. .......... .... .. .... ...... .. ........ .... 84
M acroinvertebrate A ssem blage ........................................ ........................ 86
Sum m ary Statistics...................................................................................86
C om positional A nalysis....................................... .......................... 87
Community Composition...................... .... ........................... 89
M etric Selection ............................... .. ..... ....... .............. .... 89
Macroinvertebrate Florida Wetland Condition Index.............................99
Cluster Analysis .................... ...........................
Florida W etland Condition Index ............................................. ............... 103

4 DISCU SSION ............... ..... ................. ........ .......... ............... 111

D escribing Biological Integrity........................................... ............ ............... 112
Merits of a Multi-Metric Florida Wetland Condition Index.............................13
Regionalization of the Florida Wetland Condition Index..................................114
Florida Wetland Condition Index Independent of Wetland Type .......................115
Landscape D evelopm ent Intensity Index............................................................116
A agricultural and U rban W wetlands ................................. ............. .................. 117
Richness, Evenness, and Diversity ............. ................................ ...............118
Lim stations and Further Research............................................... .................. 118
C on clu sion s ...................................... ............................................ . 1 19

APPENDIX

A STANDARD OPERATING PROCEDURES ....................................................120













B COEFFICIENT OF CONSERVATISM SCORES................................131

C C A N D ID A TE M E TR IC S ...................................................................................138

D QUANTIFYING ANTHROPOGENIC INFLUENCE ............. .... ...............147

E SU M M A R Y STA TISTIC S....................................................... .....................149

F METRIC SCORING FOR THE FLORIDA FORESTED WETLAND
CON D ITION IN D EX ............................................................. .............. 154


REFERENCES ..................................... ................ ............. 158









LIST OF TABLES


Table Page

2-1 Surrounding land use, land ownership, and sample date for 118 study
w wetlands in F lorida.......... .......................................... ................ .. .... ... . . 20
2-2 Field-data collected at 118 sam ple wetlands ............................... ............... .22
2-3 Landscape Development Intensity coefficients (LDIi) used in the calculation
of the Landscape Development Intensity index............... ................ ..............24
3-1 Comparison among three measures of anthropogenic influence, including
the Landscape Development Intensity (LDI) index, the Wetland Rapid
Assessment Procedure (WRAP), and the Minnesota disturbance index (DI). ......39
3-2 Water and soil parameters among three apriori land use categories....................40
3-3 Water and soil parameters among LDI groups ............................................... 41
3-4 Diatom richness, evenness, and diversity among apriori land use categories......43
3-5 Mean diatom summary statistics among LDI groups ........................................43
3-6 Similarity of diatom community composition using MRPP.............................44
3-7 Pearson correlations between environmental parameters and NMS ordination axes
based on diatom community composition .................................. ............... 46
3-8 Spearmans correlations for seven diatom metrics and LDI...............................47
3-9 Comparisons among diatom metrics and the diatom FWCI for LDI groups.........47
3-10 Diatom Tolerant Indicator Species ............................................. ............... 48
3-11 D iatom Sensitive Indicator Species ............................................ ............... 49
3-12 Mean diatom FWCI and LDI values for wetland clusters based on diatom
community composition...................... ...... ............................... 56
3-13 Mean macrophyte richness, evenness, and diversity among apriori land use
c a te g o rie s ...............................................................................................................5 8
3-14 Mean macrophyte richness, evenness, and diversity between LDI groups ..........58
3-15 Macrophyte community composition similarity among wetland regions..............59
3-16 Pearson correlations between environmental variables and NMDS axes
based on macrophyte community composition at 118 wetlands ........................62
3-17 Pearson correlations between environmental variables and NMDS axes
based on macrophyte community composition at 75 wetlands ...........................64
3-18 Spearman correlations between six macrophyte metrics and LDI.......................64
3-19 Comparison among six macrophyte metrics for LDI groups..............................65
3-20 Statewide and regional macrophyte Sensitive Indicator Species.........................66









3-21 Statewide and regional macrophyte Sensitive Indicator Species.........................69
3-22 The percent of statewide and regional macrophyte Indicator Species metrics
were significantly correlated with LDI.................. ......... ... ... ............ 73
3-23 Exotic macrophyte species identified at 118 study wetlands........................ 76
3-24 Macrophyte FWCI and component metrics scored statewide and regionally
for study wetlands in the low LDI group ................ ....................... ......... 82
3-25 Macrophyte FWCI and component metrics scored statewide and regionally
for study wetlands in the high LDI group............... ........ .. ............... ......... 82
3-26 Spearman correlations between the macrophyte FWCI, component
m etrics, and L D I ...................... .... ................ ................. .... ....... 83
3-27 Macrophyte FWCI scores and LDI values for wetland clusters based on
macrophyte community composition............................................... 85
3-28 Macroinvertebrate richness, evenness, and diversity among apriori land use
c a te g o rie s ...............................................................................................................8 7
3-29 Macroinvertebrate richness, evenness, and diversity for LDI groups ...................87
3-30 Macroinvertebrate community composition similarity among apriori land use
categories and ecoregions ............................................. ............................. 88
3-31 Pearson correlations between environmental variables and NMDS ordination axes
based on macroinvertebrate community composition .......................................91
3-32 Spearman correlations between macroinvertebrate metrics and the
macroinvertebrate FWCI with LDI, pH, dissolved oxygen (DO), and total
phosphorus (T P ) ................................................................... 9 1
3-33 Macroinvertebrate metric and FWCI scores between LDI groups......................92
3-34 M acroinvertebrate Tolerant Indicator Genera ....................................... .......... 93
3-35 Macroinvertebrate Sensitive Indicator Genera ............................................... 95
3-36 Macroinvertebrate FWCI scores and LDI values for wetland clusters based
on the macroinvertebrate community composition........................................102
3-37 FWCI scores for 118 wetlands based on three assemblages including diatoms,
macrophytes, and macroinvertebrates................ ....................... 104
3-38 Pearson correlations among 19 metrics ............. .............................................. 110
4-1 The 19 metrics of the Florida Wetland Condition Index, including seven
diatom, six macrophyte, and six macroinvertebrate metrics.............................112
B-l Coefficient of Conservatism (CC) scores for 561 macrophytes identified in
isolated depressional freshwater forested wetlands in Florida ..........................131
C-1 Candidate metrics based on the diatom assemblage ................................ 138
C-2 Candidate metrics based on the macrophyte assemblage ..................................140
C-3 Candidate metrics based on the macroinvertebrate assemblage..........................143









D-1 LDI, WRAP, and Minnesota disturbance index scores for 118 wetlands ...........147
E-1 Summary statistics of richness (R), evenness (E), Shannon diversity (H), and
Simpson's index (S) for the diatom assemblage (genus level)..........................149
E-2 Summary statistics of richness (R), jackknife estimators of species richness
(Jackl, Jack2), evenness (E), Shannon diversity (H), and Whittaker's beta
diversity (3W) for the macrophyte assemblage (species level)........................150
E-3 Summary statistics of richness (R), evenness (E), Shannon diversity (H), and
Simpson's index (S) for the macroinvertebrate assemblage (genus level)..........153









LIST OF FIGURES


Figure Page


1-1 Characteristic silhouette shape of three isolated depressional
w wetlands sam pled in this study ........................................... ......................... 2
2-1 Study site location of 118 isolated depressional forested wetlands
in F lorida ........................ ................................... .. .... ............... 19
2-2 Belted transect layout for macrophyte sampling and the location of the
w ater and soil sam ples. .......................... .... ............ .............. ....... .......... 25
2-3 Benthic diatom samples were collected at 50 isolated depressional
forested w wetlands. .............................. ....... ...... ........ .... ...... ...... 26
2-4 Macroinvertebrates were sampled at 79 isolated depressional forested
w etla n d s ......................................................................... . 2 9
3-1 NMS ordination bi-plot of 50 wetlands in diatom species space with
an overlay of environmental parameters..... .......... ..................................45
3-2 Percent diatom Tolerant Indicator Species increased with increasing
develop ent intensity ............................................ ......................... 49
3-3 Percent diatom Sensitive Indicator Species decreased with increasing
develop ent intensity ...................... .. .................... .................. ...........50
3-4 Pollution Tolerance Class 1 diatoms increased with increasing
develop ent intensity .................. ............................. .... .... .. ........ .... 51
3-5 Nitrogen Uptake Metabolism Class 3 diatoms increased with
increasing develop ent intensity...................................... ......................... 51
3-6 Saprobity Class 4 diatoms increased with increasing development
intensity..................................... .................. .............. ........... 53
3-7 The pH Class 3 diatoms increased with increasing development
intensity..................................... .................. .............. ........... 53
3-8 Dissolved Oxygen Class 1 diatoms decreased with increasing
develop ent intensity ...................... .. .................... .................. ...........54
3-9 Diatom FWCI scores decrease with increasing development intensity ...............54
3-10 Diatom FWCI scores for wetland clusters based on diatom community
com position .................................... ............................... ........ 55
3-11 NMDS ordination bi-plot of 118 sample wetlands in macrophyte species
space with an overlay of environmental parameters............................................61
3-12 NMDS ordination bi-plot of 75 sample wetlands in macrophyte species
space with an overlay of environmental parameters ................ ........ ..........63









3-13 Macrophyte Tolerant Indicator Species increased with increasing
develop ent intensity ............................................ ......................... 68
3-14 Macrophyte Sensitive Indicator Species decreased with increasing
develop ent intensity ............................................ ......................... 72
3-15 FQAI scores decreased with increasing development intensity ..........................74
3-16 Exotic Species increased with increasing development intensity..........................75
3-17 Native Perennial Species decreased with increasing development intensity.........78
3-18. The Percent Wetland Status species decreased with increasing development
intensity..................................... .................. .............. ........... 79
3-19 Macrophyte FWCI scores decreased with increasing development intensity. ......81
3-20 Regional macrophyte FWCI scores for five wetland clusters based on
macrophyte community composition............................................................84
3-21 NMDS ordination bi-plot for 79 wetlands in macroinvertebrate genus space
with an overlay of environmental parameters..................................................90
3-22 Macroinvertebrate Tolerant Indicator Genera increased with increasing
develop ent intensity. ...................... .. .................... ......................... 94
3-23 Macroinvertebrate Sensitive Indicator Genera decreased with increasing
develop ent intensity ............................................ ......................... 96
3-24 Florida Index scores decreased with increasing development intensity ...............97
3-25 Macroinvertebrates in the phylum Mollusca increased with increasing
develop ent intensity ............................................ ......................... 98
3-26 Macroinvertebrates in the family Noteridae decreased with increasing
develop ent intensity. ...................... .. .................... ......................... 99
3-27 Macroinvertebrates that belong to the scraper functional feeding group
increased with increasing development intensity. .............................................100
3-28 Macroinvertebrate FWCI scores decreased with increasing landscape
develop ent intensity index ........................................... ......................... 100
3-29. Macroinvertebrate FWCI scores for five wetland clusters based on
macroinvertebrate community composition. ............................... ......... ...... 102
3-30 Three dimensional scatter plot of the FWCI based on three assemblages,
including diatoms, macrophytes, and macroinvertebrates................................108
3-31 Scatterplots of FWCI scores for wetlands based on diatom,
macrophyte, and macroinvertebrate assemblages ..............................................109










EXECUTIVE SUMMARY


THE FLORIDA WETLAND CONDITION INDEX (FWCI):
DEVELOPING BIOLOGICAL INDICATORS
FOR ISOLATED DEPRESSIONAL FORESTED WETLANDS

Over 30 years ago, the federal Water Pollution and Control Act obliged states to
protect and restore the chemical, physical, and biological integrity of waters, and charged
states with establishing water-quality standards for all waters within state boundaries
including wetlands. Criteria for defining water-quality could be narrative or numeric;
and it could be addressed through chemical, physical, or biological standards. Initially,
states used chemical and physical 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). The United States Environmental
Protection Agency (USEPA) recognized the potential of biological criteria 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-
protection agencies over the past 2 decades (Gerristen et al. 2000). Such biological
criteria and monitoring programs 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 (Mack 2001; USEPA
2002a). Within Florida, biological indices have been created based on macroinvertebrate
community composition for streams (Barbour et al. 1996a; Fore 2003), lakes (Gerristen
and White 1997), and isolated depressional freshwater herbaceous wetlands (Lane et al.
2003), and based on the community composition of the diatom and macroinvertebrate
assemblages for freshwater herbaceous wetlands (Lane et al. 2003).
The primary objective of this research was to develop a Florida Wetland
Condition Index (FWCI) for isolated depressional forested wetlands in Florida. Wetland
study sites were sought in various landscape settings that included natural, agricultural,
and urban land uses. Three independent measures of anthropogenic influence were
calculated for each wetland including the Landscape Development Intensity index (LDI)
(Lane et al. 2003; Brown and Vivas 2005), the Wetland Rapid Assessment Procedure
(WRAP) (Miller and Boyd 1999), and the Minnesota disturbance index (Gernes and
Helgen 1999). Strong correlations among LDI, WRAP, and MDI led us to adopt the LDI
to represent the gradient of anthropogenic influence for further analyses. Compositional
differences among the diatom, macrophyte, and macroinvertebrate assemblages were
identified and related to the LDI measure of anthropogenic influence. Each assemblage
(diatom, macrophyte, and macroinvertebrate) was used to construct an assemblage
specific FWCI for isolated depressional forested wetlands in Florida.
While previous research has identified responses of wetland ecosystems to
particular system impacts (such as increased nutrients or altered hydrology), few have
combined multiple biotic components, environmental parameters, and landscape
development intensity in an attempt to quantify ecological integrity. The contribution of
this research to our understanding of changes in the community composition of isolated









depressional forested wetlands (based on the diatom, macrophyte, and macroinvertebrate
assemblages) in relation to different development intensities in the surrounding landscape
can be summarized in seven main points.
Biological indicators along with physical and chemical parameters were useful in
defining biological integrity;
The variable turnover times and sensitivities of the three assemblages diatomss,
macrophytes, macroinvertebrates) suggest that a multi-metric multi-assemblage
FWCI has more merit than a FWCI based on a single assemblage;
Regionalization may strengthen the FWCI;
A FWCI independent of wetland type may be feasible, given the strong likeness
of the forested FWCI to the herbaceous FWCI (Lane et al. 2003);
The Landscape Development Intensity (LDI) index was a useful tool correlating
with the measured biological condition for isolated forested wetlands;
Urban wetlands exhibit a different vector of change than do agricultural
wetlands, and while the FWCI suggests low biological integrity of both
agricultural and urban wetlands, these wetlands do provide services and do work
in the environment.
Richness, evenness, and diversity (for each assemblage) were not sensitive to
different land uses or development intensities in the surrounding landscape.

The assemblage specific FWCIs provide a quantitative measure of the biological
integrity of isolated depressional forested wetlands in Florida. The FWCI includes three
separate measures of wetland biological integrity based on the community composition of
three species assemblages. In total 19 metrics are used to address deviations from the
wetland biological integrity potential based on the suite of reference wetlands, including
seven diatom, six macrophyte, and six macroinvertebrate based metrics. Metrics were
selected for inclusion in the FWCI based on the strength and significance (Spearmans
correlation coefficient) of each metric in correlation with a quantitative gradient of LDI,
the ability of a metric to visually distinguish correlations with LDI in scatter plots, and a
significant difference between low and high LDI groups (Mann-Whitney U-test). The
FWCI for each assemblage was composed of individual metrics specific to the
assemblage, which were scaled and added together, creating the diatom FWCI (0-70
scale), macrophyte FWCI (0-60 scale), and the macroinvertebrate FWCI (0-60 scale),
with the highest values reflecting the highest biological integrity whereas a score of 0
would reflect a lack of biological integrity or no similarity to the reference wetland
condition.
The seven diatom metrics included in the FWCI, which were significant for both
the Spearmans correlation coefficient (|r| > 0.45, p < 0.01) and the Mann-Whitney U-test
between LDI groups (p < 0.10), represented tolerance and autecological metrics.
Tolerance metrics included tolerant and sensitive indicator species. The five
autecological 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
saturation between 10-25% and a biological oxygen demand of approximately 13-22
mg/L), pH Class 3 (circumneutral, mainly occurring at pH values around 7), and
Dissolved Oxygen Class 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 van Dam et al. (1994).
Six macrophyte metrics that were significant for both the Spearmans correlation
coefficient (|r|>0.39, p<0.001) and the Mann-Whitney U-test between LDI groups
(p<0.001) included in the FWCI were Tolerant Indicator Species, Sensitive Indicator
species, Floristic Quality Assessment Index (FQAI), Exotic Species, Native Perennial
Species, and Wetland Status Species. FQAI scores were obtained for species identified
specific to this study.
The six macroinvertebrate metrics included in the FWCI, which were significant
for both the Spearmans correlation coefficient (|r| > 0.3, p < 0.05) and the Mann-Whitney
U-test between LDI groups (p < 0.05), represented tolerance, community balance, and
functional group metrics. Tolerance metrics included Tolerant Indicator Genera,
Sensitive Indicator Genera, and Florida Index scores. Community balance metrics
included Mollusca (phylum taxonomic level) and Noteridae (family taxonomic level).
One functional groups metric was included, the percent of macroinvertebrates identified
as scrapers.

Macrophyte FWCI Macroinvertebrate
Diatom FWCI Metrics Metrics FWCI Metrics
% Tolerant species % Tolerant species % Tolerant species
% Sensitive species % Sensitive species % Sensitive species
% in Pollution Class 1 Modified FQI Score FL Index Score
% in Nitrogen Class 3 % Exotic % Mollusca
% in Saprobity Class 4 % Native Perennial % Noterida
% in pH Class 3 % Wetland Status % Scraper
% in Oxygen Class 1

The variable turnover times and sensitivities of the three different assemblages
diatomss, macrophytes, macroinvertebrates) suggest that use of multiple assemblage
specific FWCIs has more sagacity than a FWCI based on a single assemblage. The three
species assemblage specific FWCIs can be used to infer influences in temporal and
spatial changes to which a particular wetland has been exposed. While agreement in the
ranking of the biological condition of study wetlands using the three species assemblage
FWCIs was anticipated, discrepancies among the ranking from the different assemblages
may provide great insight into wetland condition as different species assemblages
respond to changes in driving energies over different time scales. 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.
Ranges of FWCI scores for the a priori wetland categories (based on land use in the
surrounding landscape) show the difficulties inherent in attempting to discern a specific
mode of impact based on FWCI scores.
Though the FWCI suggested low biological integrity of both agricultural and
urban wetlands, these wetlands provide services and do work in the environment.
Therefore, a quantitative score of biological integrity established through the FWCI
should not be used as a surrogate for wetland value, but rather as an objective,
quantitative means of comparing changes in community composition along gradients of









landscape development intensity. In the future, an integrative multi-metric FWCI could
be constructed for wetlands throughout the state, with lists of Indicator Species and
metric scores dependent on Florida ecoregions and specific to wetland type.
While the FWCI 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 community 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-nitrogen, water TKN, water TP, soil moisture, soil organic
matter, and soil TP). When soil and water parameters were used to explain variation in
the community composition for each assemblage based on Non-metric Multidimensional
Scaling (NMDS) ordination techniques, water column pH was universally identified as a
keystone parameter significantly correlating the community composition ordination.
Additionally, total phosphorus concentrations correlated with the ordinations based on
both the diatom and macrophyte assemblages. Perhaps preservation and restoration
strategies could focus on limiting activities (in the surrounding landscape) that influence
changes to water column pH and total phosphorus loading to wetlands in order to
promote biological integrity.











CHAPTER 1
INTRODUCTION AND OVERVIEW

With even casual observation, it is apparent that ecosystems change with
increasing levels of human development, and that the extent of change is observably
related to the magnitude of human activity. While previous research has identified the
responses of forested wetlands to human induced changes 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 effects of
anthropogenic activities. This is particularly true for wetlands located within urban
settings. Our study focused on understanding the effects of landscape development from
anthropogenic activities on isolated depressional forested wetlands in Florida. More
specifically, our goal was to define the condition of wetlands surrounded by different
land use activities by describing the biological integrity of the ecosystem.
Determining wetland ecosystem integrity through the use of biological indicators
requires an accepted definition of integrity. Karr and Dudley (1981) defined integrity as
"the ability of an aquatic ecosystem to support and maintain a balanced, integrated,
adaptive community of organisms having a species composition, diversity, and functional
organization comparable to that of the natural habitats of the region." This definition
requires two things: a definition of the natural habitat (or reference standard condition)
and appropriate regionalization. Reference standard wetlands have been defined as those
wetlands surrounded by natural landscapes, with no apparent anthropogenic alterations.
Ecosystem condition has been commonly defined by data collected at a suite of reference
wetlands, which include a suite of reference standard wetlands as well as a suite of
impaired wetland representing the full range of impaired conditions from minimal to
severe impairment. Gerristen et al. (2000) concur that biological assessment relies on a
characterization of the reference condition.

Review of Isolated Depressional Forested Wetlands

Throughout the world, wetlands have been categorized in many different ways
(Mitsch and Gosselink 1993; Keddy 2000; Kent 2000). Probably the most widely
recognized classification system in North America is that by Cowardin et al. (1979). Our
study focuses on what Cowardin et al. (1979) categorized forested palustrine wetlands.
In more general terms, the wetlands targeted in our study have been called pondcypress
domes with reference to Vernon (1947), who was the first to name these systems after
their characteristic silhouette in the landscape (Figure 1-1). Pondcypress domes range in
size from less than 1 hectare to more than 10 hectares (Wharton et al. 1977), and host
Taxodium ascendens (pondcypress) as the principal tree species (Devall 1998). Other
tree species associated with Taxodium ascendens include: Nyssa sylvatica var. biflora
tupelo (swamp tupelo), Pinus spp. (many southern pines), Acer rubrum (red maple), and
Magnolia virginiana var. australis (sweetbay magnolia) (Wilhite and Toliver 1990;
Devall 1998). Pondcypress trees characteristically dominate the center, with pondcypress























































Figure 1-1.


Characteristic silhouette shape of three isolated depressional wetlands
sampled in this study. These pondcypress domes named by Vernon
(1947) include: (A) CU2 in Orange County, (B) CR4 in Sumter County,
and (C) PR2 in Santa Rosa County.









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 number of seedlings in the
drier edges (Odum 1978). While the target ecosystem type was the pondcypress dome,
many of the study wetlands hosted co-dominant species or were dominated by canopy
species other than pondcypress and may more appropriately be described as isolated
depressional mixed-forested wetlands. The principal commonalities among the study
wetlands include the isolated depressional nature of these wetlands and the occurrence of
mature forest canopy.
The use of the word isolated to describe a wetland has been scrutinized because it
has been interpreted to suggest separation of a wetland system from its landscape.
However, the term isolated in this study simply refers to the lack of a regular or
permanent surface water connection between the wetland and the surrounding landscape.
However, in periods of high water and excessive rainfall, many of these isolated wetland
systems receive surface run-in from nearby wetlands or discharge to other nearby
wetlands or surface water features. Under typical conditions, inflows into isolated
depressional wetlands are limited to sunlight, wind, water (rain, surface run-off,
groundwater), and recruitment of plant and animal species. Water inflow comes almost
entirely from rainwater, both as a direct input and as run-in 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 isolated depressional forested wetlands much of
the year (Odum 1978; Mitsch and Gosselink 1993). Some have deep central pools
staying wet year round, while others go dry annually. There is also variation in
maximum flooding depth and length of standing water between years, which reflects the
larger scale climatic and physiogeographic influences to these wetland systems.
Typically, the wettest period is summer and the driest spring and fall (Mitsch and
Gosselink 1993).
Isolated depressional forested wetlands dominated with pondcypress are
characterized by a tolerance of low nutrient levels and intermittent fire (Brandt and Ewel
1989), with major system inputs 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 pondcypress domes in northern Florida, Marois
and Ewel (1983) found an increase in the densities of hardwood and shrub species.
Furthermore, in south Florida, fire suppression has resulted in a shift in the canopy
community composition to readily regenerating hardwood species over pondcypress
dominance (Ewel 1990). Perhaps many of the depressional mixed-forested wetlands
identified throughout the state were historically dominated by pondcypress, but have
undergone a shift in canopy species composition due to drainage, fire suppression, timber
harvesting, and other indeterminate anthropogenic activities.
Pondcypress and the most characteristic co-dominate swamp tupelo are
deciduous, shedding their leaves from October to December. Ewel (1984) and Wharton
et al. (1977) found that the perpetuation of pondcypress domes depends on fluctuating
water levels, with a dry period without standing water necessary for pondcypress
generation and higher water levels some time during the year necessary to prevent
germination of more terrestrial faster-growing pines and hardwoods that are not tolerant









of standing water. Altering the typical hydroperiod of a pondcypress dome has a direct
effect on species composition, resulting in encroachment of terrestrial species in drained
pondcypress domes and a lack of regeneration of pondcypress in artificially flooded
pondcypress domes. Anthropogenic activities in the surrounding up-slope landscape can
create a wide array of changes to the inflows of these wetland systems. Some of the
potential alterations to isolated depressional wetlands located within developed
landscapes include changes in the seasonality and depth of flooding, increased nutrient
inputs, increased toxin inputs, and physical impacts (canals, retaining walls, stormwater
box culverts, etc.).

Changes in Hydrology

There are two important mechanisms within the developed landscape leading to
potential hydrologic alterations to an isolated depressional wetland ecosystem surrounded
by developed land uses. First, increased run-off is considered a factor of the amount of
increased impervious surface in the watershed supporting the wetland system and the
amount of rainfall. This would be particularly apparent in an urban landscape, where
previously vegetated lands are paved creating increased water flow during rain events,
which might otherwise have been intercepted by the vegetation or permeated into the soil.
A second mechanism of the developed landscape is an increased outflow of water storage
from the wetland as a result of drainage.
In Florida, mature cypress trees (Taxodium spp.) are considered 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), though the tress suffered decreased growth rates and no evidence of regeneration.
This suggests that while mature cypress ecosystems may be able to withstand some
threshold level of long-term flooding, regeneration may be impeded which is otherwise
necessary to ensure the long-term continued existence of the ecosystem. Ultimately,
removing the structure of the wetland will predictably alter other ecosystem components.
For example, by removing the tree canopy, algae may flourish at the water surface where
it would have otherwise been shaded out. This would in term alter the 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 damming of the Ocklawaha River in Florida.
They found that while the larger trees survived some depths of flooding, in the deepest
areas, where mean water depth was 1 m, there was 100% tree mortality within 5 years of
flooding. Where as, in the control system, tree mortality was less than 1% per year.
Additionally, flooded trees responded with a dieback of terminal 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 increased. 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 tolerant species, or
increased nutrient levels immediately following flooding.
Marois and Ewel (1983) studied the effects of ditches and berms on 15
pondcypress domes situated within an intensively managed slash pine plantation. In the
pondcypress domes not ditched and bermed, the lengths of flooding and mean water
depth were generally greater. Alternatively in the drier ditched and bermed pondcypress
domes the density of hardwoods, shrubs, and vines increased. They concluded that while
pondcypress tree growth increased in the years directly following the drying of the
pondcypress domes, pondcypress regeneration may have been inhibited due to changes 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 pondcypress dome may
inhibit or seriously diminish potential germination. The higher density of hardwoods,
shrubs, and vines may also inhibit pondcypress regeneration by blocking sunlight from
reaching the forest floor. Marois and Ewel (1983) found the highest percentage of light
transmittance in unaltered pondcypress domes. The pondcypress 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, resulting in an increased
density of shrubs and hardwoods, poor pondcypress regeneration, increased fire potential,
and a dramatic shift to arboreal wildlife species from aquatic 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,
dahoon holly; Lyonia lucida, fetterbush; Magnolia virginiana var. autralis, sweetbay
magnolia; and Persea palustris, swamp bay), became more common in swamps when
water levels were lowered. Harris and Vickers (1984) speculated that shifting species in
the vegetation layer equates to altered 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 pondcypress domes for regeneration. Benthic
invertebrates may not withstand increased dry periods, and reproduction may be difficult
or unattainable. Often forming the base of swamp food chains (Ewel 1990), eliminating
or decreasing the population of benthic macroinvertebrates could have repercussions
throughout the food chain. Fish, amphibians, and reptiles may be eliminated from
pondcypress domes with decreased hydroperiods due to reproduction difficulties or
altered food availability (Means et al. 1998; Ewel 1990).

Increased Inflows of Nutrients and/or Toxins

In undeveloped landscapes isolated depressional wetlands receive limited nutrient
inputs from rainwater and surface water run-in (Wharton et al. 1977). An increase in the
inflow of nutrients and toxins may come from both point (eg. wastewater additions) and
non-point (ex. impervious surface run-off) sources. An increase in run-in from
impervious surfaces in the surrounding landscape may increase the loading of nutrients
and toxins to a wetland system (Harper 1994). Surface run-off carrying fertilizer used on
agricultural crops or home lawns are examples of non-point source contributions. As
nutrients flow into the wetland, the growth of living biomass increases, and nutrients









accumulate in the water and soil organic matter storage. Conversely, as toxins flow into
and accumulate in the wetland, there is a deleterious effect on biomass.
Two nutrients of primary importance in pondcypress domes are phosphorus and
nitrogen. Phosphorus, an element critical to plant growth, is mostly bound into forms
unavailable to plants at pH levels below 5.7 (Brady and Weil 2004), higher than the
average pH of pondcypress domes embedded in undeveloped landscapes (Coultas and
Duever 1984). Dierberg and Brezonik (1984) found an annual average pH of 4.63 for
pondcypress wetland with the primary source of water being rainfall, and an annual
average pH of 5.4 for a pondcypress wetland with the primary source of water being
shallow groundwater for wetlands in north-central Florida. Phosphorus is known to
accumulate in the clay layers found beneath pondcypress domes, which makes
pondcypress dome ecosystems dependent on a constant input of available phosphorus
from rainfall. Raising the pH of a wetland increases the concentration of available
phosphorus. In contrast, nitrogen does not accumulate in the clay layer or organic
sediments at the bottom of pondcypress domes due to denitrification processes, and the
rate of the nitrogen cycle seems dependent on the rate of decomposition of organic matter
(Wharton et al. 1977).
Pondcypress trees respond to increased nutrient loading with increased tree
growth rates (Nessel et al.1982; Lemlich and Ewel 1984). Nessel et al. (1982) measured
phosphorus concentrations in live cypress needles at a pondcypress dome embedded in
silvicultural land use and a cypress strand receiving sewage for more than 40 years.
Cypress needles in the silvicultural wetland had a lower average phosphorus
concentration than cypress needles in the wetland receiving wastewater. Additionally,
the top 20 cm of sediment in the wetland receiving wastewater had nearly 5.5 times as
much phosphorus per square meter as in the silvicultural wetland. They concluded that
the trees in the cypress strand were in fact responding to the increased nutrient inputs
(from the sewage) with increased growth rates.
Traditionally pondcypress domes are 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. (duckweed), Spirodela spp.
(giant duckweed), and/or Azolla caroliniana (Carolina mosquito fern) on the water
surface. Ewel (1984) also noted that a nutrient enriched pondcypress dome had similar
understory species composition compared to a pondcypress dome not receiving
wastewater, however the leaf area was significantly higher in the nutrient enriched
pondcypress dome. Changes in the understory vegetation (from increased leaf area or
covering of the water surface with a layer of vegetation) can have an effect on other
trophic levels within the wetland ecosystem. In pondcypress domes receiving wastewater
additions, Harris and Vickers (1984) reported an increase in the number 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 pondcypress dome
were different from the control wetland.

Physical Disturbance

Examples of physical changes in depressional forested wetlands include trampling
and grazing by domestic cattle, rooting of feral pigs, construction of barriers such roads









and retention walls, and conduits such as stormwater culverts. Few publications were
found that quantified the effects of physical modifications to wetlands. Findlay and
Houlahan (1997) used existing biological surveys for 30 Ontario, Canada, wetlands,
comparing species richness of plants, amphibians, birds, and reptiles with wetland areas,
density of paved roads, 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
wetland. They concluded that increasing the density of paved road surface or decreasing
the forest cover by 20% within 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 modification to isolated depressional forested wetlands is the
removal of a portion or all of the canopy layer. Florida has a long history of timber
harvesting, and Ewel (1990) suggested that nearly all of the pondcypress domes in north
Florida have been logged since the late 1800s. Studies show that logged pondcypress
domes maintain their defining characteristics after regeneration (Terwilliger and Ewel
1986; Ewel et al. 1989); however, during regeneration, there are shifts in the flora and
fauna of logged wetlands. Physical modifications such as roads, canals, and stormwater
culverts also act as direct conduits for the introduction of exotic species (Frappier and
Eckert 2003).

Biological Indicators of Ecosystem Integrity

Over 30 years ago, the Water Pollution and Control Act (later referred to as the
Clean Water Act, 1972) required states to "restore and maintain the chemical, physical,
and biological integrity of the Nation's waters" (USEPA 1990). This legislation included
establishing water-quality standards for all waters within state boundaries, including
wetlands. Criteria for defining water-quality could be narrative or numeric; and it could
be addressed through chemical, physical, or biological standards. Initially, states used
chemical and physical 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).
There are several shortcomings in deriving ecosystem integrity based on
exceeding established limits for chemical and physical parameters. Such criteria have
been considered rudimentary in their ability to reflect more than the temporal
concentration of chemicals 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 undetected. This
same water body could have undesirable levels of other nontarget 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 occur during specific
loading events and may therefore incompletely describe the biological and ecological
condition of the system. Adams (2002) points out other environmental factors (such as
sedimentation, alterations to habitat, varying temperature and oxygen levels, and changes
in ecological aspects like food availability and predator-prey relationships) are not
reflected with chemical criteria alone. James and Kleinow (1994) note that different
organisms respond in different ways to the amount, persistence, and exposure of chemical









compounds otherwise foreign to an organism, and single-valued chemical and physical
criteria of water quality may overlook important biological implications.
Alternatively, biological indicators integrate the spatial and temporal effects of
the environment on resident organisms and are suitable 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 reflect chemical
exposure and also integrate changes in the community composition of the ecosystem
(from physical, chemical, and biological changes) (Adams 2002).
The United States Environmental Protection Agency (USEPA) recognized the
potential of biological criteria 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-protection agencies over the past 2 decades (Gerristen
et al. 2000). Biological criteria and monitoring 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).
Biological monitoring to assess ecosystem condition has been applied widely in
ecological research. One trend in biological 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 (O'Connell et
al. 1998). Such indices have been applied to ecosystems throughout the world including
in Europe (Kelly and Whitton 1998); widely throughout the United States (Karr 1981;
Lenat 1993; Fore and Grafe 2002; Lane et al. 2003); and is beginning in Australia by J.E.
Ling of the Royal Botanical Gardens, University of Western Sydney. The primary aim of
biological monitoring is to detect changes in abundance, structure, and diversity of target
species assemblages. Danielson (1998a) notes that biological signals are effective mainly
because biological monitoring incorporates changes from various collective constant or
pulsing sources.
Many studies have created multi-metric indices of biological condition,
incorporating individual metrics into a quantitative value of community condition or
ecosystem integrity. Karr and Chu (1997) defined metrics as biological attributes that
have a consistent and predictable response to anthropogenic activities. Anthropogenic
activities can alter the integrity of wetland ecosystems by causing one or more of the
following conditions: eutrophication, 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 algae 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: provide a food source
for organisms at higher trophic levels, contribute to nutrient and biogeochemical cycling,
oxygenate the water column, regulate water chemistry, create habitat for other organisms,
and act as physical barriers to erosion. Because of their rapid turnover times, algae have
a short response time to perturbations including nutrient and toxic contaminant inputs;
and algae continue production throughout the winter, taking advantage of available
nutrients when higher plants are dormant (Cronk and Fennessy 2001). While the
standing stock of algae is typically lower than that of the macrophyte assemblage, algae
can constitute a higher proportion of primary productivity within 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 advantages of using algae for
biological assessment include the high diversity within the algal community (particularly
of diatom species) in aquatic environments (Stevenson 2001).
There is a depth of knowledge as to the sensitivity of many species to different
environmental conditions based on their autecological characteristics, including two
published tables of autecological relationships by Bahls (1993) and van Dam et al.
(1994). Additionally, the rapid-response time of the algal community to changing
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
considered easy to identify based on well-established taxonomic keys of their decay
resistant siliceous structures (Stevenson et al. 1999), and there are well-tested protocols
for sampling aquatic habitats (Goldsborough 2001).
Few significant disadvantages of using algae in biological assessment
methodologies have been described. Among them is the necessity of a high-powered
microscope for identification (Doherty et al. 2000), though identification is relatively
easy, and good taxonomic keys have been established (Stevenson et al. 1999). However,
taxonomic identification to lower taxonomic levels such as genus or species should be
reserved for an expert taxonomist. Additionally, while most algae are not readily motile,
wind and current translocation can complicate assessments based on scales of
anthropogenic activity in the surrounding 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, diatoms are noted as a useful assemblage (Stevenson
2001; Doherty et al. 2000). Previous research has correlated the response of diatoms in
streams, lakes, and wetlands, to changes in surrounding land use and to changes in water-
column characteristics including nutrient loading (van Dam et al. 1994); pH (Pan and
Stevenson 1996); heavy metal loading (Charles et al. 1996); and saprobity levels (Lange-
Bertalot 1979). The USEPA (2002b) reported that diatoms are one of the most









commonly used assemblages in aquatic ecosystems for assessing biological, physical, and
chemical conditions.
Research correlating changes in the diatom 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 biological 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 et al. (2003)
incorporated 14 metrics into the Diatom Index of Wetland Condition (DIWC). These
included tolerant indicator species, sensitive indicator species, diatoms requiring low pH,
diatoms requiring low salinity, diatoms tolerant of high salinity, diatoms tolerant of high
pH, diatoms sensitive to high nitrogen, diatoms tolerant of high nitrogen, diatoms
requiring elevated dissolved oxygen, diatoms tolerant of low dissolved oxygen, meso-
and polysaprobous diatoms, diatoms characteristic of oligotrophic environments, diatoms
characteristic of eutrophic environments, and pollution-tolerant diatoms. Environmental
parameters correlating with diatom community composition included specific
conductivity, water-column pH, water ammonia-nitrogen concentration, water total
Kjeldahl nitrogen (TKN) concentration, water total phosphorus concentration (TP), soil
pH, and soil TP (Lane 2003).
In a study of lotic (relating to moving water) systems in the Mid-Atlantic States,
Pan et al. (1996) found the strongest correlation with diatom community composition and
changes in water-column pH. Additional water-column parameters correlating with
diatom community composition included turbidity, aluminum concentration, chlorine
concentration, TP, total suspended solids, and dissolved organic-carbon concentration.
Similarly, in a study of emergent permanently flooded floodplain wetlands in western
Kentucky, Pan and Stevenson (1996) found significant correlations between diatom
community composition and eight water variables, including alkalinity, conductivity,
ammonia-nitrogen concentration, pH, silicon concentration, nitrate-nitrogen
concentration, chlorine concentration, and TP. Another study of streams in Michigan
also correlated the response of diatom community composition to different land use and
water physical and chemical parameters (Stewart et al. 1999). These findings reiterate
that usefulness of the algal assemblage in reflecting changes in the water environment.

Macrophytes as Biological Indicators

Wetland macrophytes are defined as aquatic emergent, submergent, or floating
plants growing in or near water (USEPA 1998) and are described as distinguishing
landscape features. The spatial 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 composition of wetland macrophytes typifies the
physical, chemical, and biological wetland dynamics in time and space. Macrophytes









play a vital role in supporting the structure and function of wetlands by providing food
and habitat for other assemblages including algae, macroinvertebrates, fish, amphibians,
reptiles, birds, and mammals. As such, macrophyte populations can be used as diagnostic
tools to assess other aspects of the wetland ecosystem. 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
overgrowth of particular macrophytes may signify increased nutrient loading (USEPA
1998).
Many advantages of studying macrophytes as indicators of wetland condition
have been noted, including their large, obvious size; ease of identification, to at least
some useful taxonomic level; known response to toxicity tests; and general lack of ability
to move to avoid unfavorable conditions (Danielson 1998a; Cronk and Fennessy 2001).
Additionally, macrophytes readily respond to changes in nutrient, light, toxic
contaminant, metal, herbicide, turbidity, water, and salinity levels. They can also be
sampled in the field with transects, or from the office with aerial photography, and well-
established field methods of sampling macrophytes exist (USEPA 2003). Furthermore,
the USEPA (2003) states that macrophytes do not require laboratory analysis, can easily
be used for calculating simple abundance metrics, and are superb integrators of
environmental condition. In general, macrophytes 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 noted shortcomings of using macrophytes as biological
indicators. These include the potential delay in response time for perennial vegetation,
difficulty identifying taxa to the species level in certain seasons and for some genera,
different herbivory patterns, and varied pest-management practices (Cronk and Fennessy
2001). Despite these limitations, macrophytes have provided strong signals of
anthropogenic influence (USEPA 2003). In fact, many states have begun using
macrophytes in their wetland biological assessment programs, including 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 wetland type and bioregion. Lane et al. (2003)
calculated five metrics based on the macrophyte community composition for inclusion in
the marsh Vegetative Index of Wetland Condition (VIWC). The five core metrics of the
VIWC included tolerant indicator species, sensitive indicator species, exotic species,
annual to perennial ratio, and average coefficient of conservatism score (Lane 2003). In
Minnesota, Vegetative Indices of Biotic Integrity (V-IBIs) have been created for eight
wetland types (Galatowitsch et al. 1999a), and included 15 metrics for high-order river
floodplain wetlands, 12 for low-order river floodplain wetlands, eight for mid-order river
floodplain wetlands, seven for calcareous littoral wetlands, six for noncalcareous littoral
wetlands, seven for wet prairie-sedge meadows, four for forest glacial marshes, and a
single metric for prairie glacial marshes. Another comprehensive biological assessment
used to construct multi-metric indices of biotic integrity for Ohio wetlands was designed
by Mack et al. (2000). Separate biological multi-metric indices were developed for









emergent, forested, and shrub wetlands. Twelve metrics were incorporated, including
Carex species, dicot species, shrub species, hydrophyte species, Rosaceae species,
Floristic Quality Assessment Index (FQAI), tolerant species, intolerant species, invasive
graminoids, shrub density, small-tree density, and maximum importance value.
The FQAI has been included in many of the multi-metric indices created for the
macrophyte assemblage. The concept of FQAI was developed by Wilhelm and Ladd
(1988) for vegetation around Chicago, Illinois. This method of scoring plant species
based on expert 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 et al. 2003; Cohen et al.
2004). The FQAI provides a quantitative means of assessing the fidelity of a plant to a
particular environment 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 macroinvertebrate assemblage has been
widely applied for indications of environmental quality, and often more specifically water
quality (Lenat 1993; Cummins and Merritt 2001). Invertebrates participate in many
fundamental ecological processes, including the breakdown of organic matter and
recycling of nutrients. Additionally, invertebrates are a vital component of the food web,
making up a large portion of the diets of other organisms (such as fish, amphibians, and
birds) (Cummins and Merritt 2001; Helgen 2001). In fact, Voshell (2002) recognized
that freshwater invertebrates have been used more often than any other group of
organisms for assessing freshwater ecosystems.
A great deal of research has been completed on the specific ecology of lotic
macroinvertebrates (of, relating to, or living in flowing waters), though less is known
about lentic macroinvertebrates (of, relating to, or living in still waters). Williams and
Feltmate (1992) noted that, while not well studied, the communities of aquatic insects in
wetlands include species from most of the major aquatic groups. The community
composition of wetland macroinvertebrates differs from that of flowing waters, because
of differences in substrate, dissolved-oxygen level in the water column, hydroperiod, and
annual water fluctuations. Macroinvertebrates have been useful indicators of
environmental condition in streams; and Karr and Chu (1997) speculate
macroinvertebrates also may be appropriate indicators of environmental 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 wetland biological assessment methodologies have been
initiated in many states, including Florida (Lane et al. 2003), Minnesota, Montana, North
Dakota, and Ohio (Danielson 1998b). Within the state of Florida and throughout the
southeastern Coastal Plain, ecological research on the macroinvertebrate community has
included many ecosystem types from isolated marshes of peninsular Florida (Kushlan









1990; Lane 2003); isolated wetlands in south Florida (Stansly et al. 1997); 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 structure and function of the
macroinvertebrate community accurately reflects the biological condition of a wetland,
and that the macroinvertebrate community composition changes in predictable ways with
increased human influence. Because wetland macroinvertebrates complete part or all of
their lives in the wetland, they are directly exposed to conditions 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 environment. Their quick response time, reliance on
water (both for the water quality and duration of inundation), and ease of collection make
macroinvertebrates a favorable assemblage for use as biological indicators. Noted
disadvantages to using macroinvertebrates include the amount of time and knowledge
necessary for identification to lower taxonomic levels (Cummins and Merritt 2001).
The Florida Department of Environmental Protection (FDEP) has initiated the
development of biological indices based on the macroinvertebrate assemblage for
freshwater bodies in Florida. Macroinvertebrate-based biological indices have been
created for isolated marshes through the Macroinvertebrate Index of Wetland Condition
(MIWC; Lane et al. 2003), for Florida streams through the 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 restoration in the Kissimmee River
Basin (Merritt et al. 1996), and in freshwater lakes through the Lake Condition Index
(LCI; Gerristen and White 1997). Different core metrics comprise each multi-metric
biological index.
Lane (2003) incorporated five core metrics as biological indicators of wetland
condition for isolated marshes in the MIWC, including sensitive taxa, tolerant taxa,
predators, Odonata, and Orthocladiinae. The SCI was developed with ten core metrics
within six categories of biological organization including were taxonomic richness (total
number of taxa, number of Trichoptera taxa, and number of Ephemeroptera taxa); feeding
group (percentage filterer individuals); voltinism (long-lived taxa richness); habit (clinger
taxa richness); community structure (percentage dominance of the most abundant taxon
and percentage Tanytarsini midges); and sensitivity and tolerance (sensitive taxa richness
and percentage very tolerant individuals) (Fore 2004). The LCI incorporated seven
metrics, including taxa richness, Shannon diversity, Hulbert index, ETO taxa
(Ephemeroptera, Trichoptera, and Odonata), percent dominance, filter feeders, and
gatherers (Gerristen and White 1997).
Numerous studies have documented the response 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 responded to nutrient enrichment









with a greater number of Coleopteran species present in nutrient-enriched and
intermediate areas (especially those in the families Hydrophilidae and Dystcidae) than in
non-enriched areas. With shifts in trophic status, the structure of other assemblages also
changed, affecting the benthic macroinvertebrate community composition. For example,
Adamus and Brandt (1990) found that shading from dense stands of emergent vegetation
altered the distribution among functional feeding groups by limiting the production of
benthic algae (thus favoring detritovores over grazers). De Szalay and Resh (1996)
similarly found that increased shading 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 alters the natural hydrology of an
isolated wetland, it also increases the inflow of nutrients, sediments, and toxic metals.
Harris and Vickers (1984) found that adding wastewater to pondcypress domes shifted
the macroinvertebrate community toward a less-complex trophic structure. Similarly,
when wastewater was directed into Florida pondcypress domes, Lemna spp. (duckweed)
mats covered the water surface, blocking sunlight from the water column, and creating
anoxic conditions (Dierberg and Brezonik 1984). This reduced the diversity and biomass
of benthic invertebrates, leaving only a few pollution-tolerant organisms (Brightman
1984). In Florida streams, Barbour et al. (1996b) found that the occurrence 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 increased phosphorus levels, lowered 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
chironomids of the family Orthocladiinae, including those in the genus Cricotopus, were
found to be tolerant of metal pollution; while other Orthocladiinae, including
Rheocricotopus spp. and Corynoneura spp., were thought to be sensitive to metal
pollution in Florida streams.
Aside from the nutrient, sediments, and toxics loading, wastewater or stormwater
additions also alter the water level and hydroperiod of a wetland system, which affects
the macroinvertebrate community composition. Toth (1993) reported a response of the
macroinvertebrate community to water-level manipulation in a Kissimmee River
demonstration project. Hydroperiod had a distinct influence on community composition,
as some macroinvertebrates either temporarily relocated or coped with behavioral and
biological adaptations to changing water conditions. Macroinvertebrates with adaptations
to wetland hydroperiods demonstrate both behavioral and physiological adaptations for
draw-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
sediments 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 behavior may indicate
hydroperiod stability. Snyder et al. (1998) found that macroinvertebrates with
comparatively short life cycles that are capable of rapid colonization were typical of
canals surrounded by urban land uses, whereas the occurrence of macroinvertebrates with
longer life cycles were more common in canals surrounded by more natural landscapes.


Quantifying Anthropogenic Influence

Wetlands occupy a large portion of the Florida 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 Florida had 4,467,000 ha of wetlands remaining,
translating into a loss in Florida of 46% of the pre-1780s wetland area (Mitsch and
Gosselink 1993; Dahl 2000). Throughout the continental United States, similar trends
were apparent, with a drastic decline in the surface area of wetlands. 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 sampled were adjacent to agricultural lands and therefore
potentially affected by land use practices such as herbicide and pesticide application,
irrigation, livestock watering and 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, transportation 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 both 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 in Florida 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 wetlands were harvested and succeeded into
shrub wetlands.
Anthropogenic activities can influence an array of changes to the physical,
chemical, and biological characteristics in surrounding ecosystems. There have been
numerous attempts at quantifying anthropogenic influence based on varying scales.
Three primary indices of anthropogenic influences were incorporated throughout our
study to compare wetland condition, including the Landscape Development Intensity
(LDI) index (Brown and Vivas 2005), the Wetland Rapid Assessment Procedure (WRAP;
Miller and Boyd 1999), and the Minnesota disturbance index (Gernes and Helgen 1999).

Landscape Development Intensity Index

The LDI index has been used as a gauge of human activity based on a
development intensity measure derived from nonrenewable energy use in the surrounding
landscape. The underlying concept behind calculating the LDI (quantifying the
nonrenewable energy use per unit area in the surrounding landscape) stems from earlier









works by Odum (1995), who pioneered emergy analysis for environmental accounting.
[Emergy is an environmental accounting term referring to expressing energy use in solar
equivalents (Odum 1995).] Brown and Ulgiati (2005) suggest that landscape condition,
or ecosystem health, is strongly related to the surrounding intensity of human activity,
and that ecological communities are affected by the direct, secondary, and cumulative
impacts of activities in the surrounding landscape. Healthy ecosystems are defined as
those with integrity and sustainability, which correlate to limited development in the
surrounding landscape and the maintenance of ecosystem structure and function, even
when stressors (e.g. flooding, drought, etc.) are present (Brown and Ulgiati 2005).
The LDI does not account for any individual causal agent directly, but instead,
may represent the combined effects of air and water pollutants, physical damage, changes
in the suite of environmental conditions (ex. groundwater levels, increased flooding), or a
combination of such factors, all of which enter the natural ecological system from the
surrounding developed landscape (Brown and Vivas 2005). Wetlands surrounded by
more intense activities such as highways and multi-family residential land uses receive
higher LDI index values. The highest LDI coefficient of 10.0 is assigned to the urban
land use category of Central Business District. Undeveloped land uses such as wetlands,
lakes, and upland forests, are assigned an LDI coefficient of 1.0, the lowest possible
value, based on no use of nonrenewable energy in these ecosystems.

Wetland Rapid Assessment Procedure

The Wetland Rapid Assessment Procedure (WRAP) was designed to provide
accurate and consistent evaluations of wetland sites, and relies on an evaluator with an
adequate understanding of the functions of and species found throughout Florida
ecosystems (Miller and Gunsalus 1999). WRAP consists of a qualitative score describing
the functional capacity of a wetland. Scores range from 0.0 to 3.0, in 0.5 increments.
The primary scoring categories for WRAP include: 1) Wildlife 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 indicates a wetland with
a reduced functional capacity (Miller and Gunsalus 1999).

Minnesota Disturbance Index

The Minnesota disturbance index is considered a gradient of human disturbance,
or a measure of land use disturbance based on investigator knowledge, observations, and
best professional judgment about the degree of influence to the ecosystem. Gernes and
Helgen (1999) used the Minnesota disturbance index as a baseline for creating an index
of vegetative biotic integrity for depressional wetlands. There are two primary categories
and three secondary categories used to calculate the Minnesota disturbance index score.
The primary categories include stormwater and agricultural influence, and are weighted
twice as high as the secondary categories. Wetlands receive scores assigned according to
significantly affected (S = 8), moderately affected (M = 4), least 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 one of the primary categories, and reference standard









wetlands receive a score of zero in both primary categories. The three secondary
categories include hydrologic/miscellaneous influence, historical influence, and buffer,
receiving scores of significantly affected (S = 3), moderately affected (M = 2), least
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, with a score of 0
representing the reference standard condition.

Project Overview

Physical and chemical environmental parameters and the community composition
of the diatom, macrophyte, and macroinvertebrate assemblages were sampled to develop
the Florida Wetland Condition Index (FWCI) for isolated depressional forested wetlands
for Florida. Wetland study sites were sought in various landscape settings that included
undeveloped, agricultural, and urban land uses. Compositional differences in the diatom,
macrophyte, and macroinvertebrate assemblages among wetlands were identified and
related to the Landscape Development Intensity (LDI) index, a quantitative calculation of
anthropogenic activity in the landscape surrounding a wetland which is based on a
development intensity measure derived from nonrenewable energy use in the surrounding
landscape. Each assemblage was used to construct the three assemblage specific Florida
Wetland Condition Indices (FWCIs) for isolated depressional forested wetlands in
Florida.











CHAPTER 2
METHODS

Biological, physical, and chemical parameters were sampled in 118 forested
wetlands less than 2 ha in size. This chapter describes site selection, calculations of the
gradient of anthropogenic activity including the Landscape Development Intensity (LDI)
index, field-data collection, and laboratory analyses. Statistical analyses are described for
each assemblage and for the creation of the assemblage specific Florida Wetland
Condition Indices (FWCIs).

Site Selection

Field research spanned two growing seasons 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, agricultural, urban). Hereafter wetlands
surrounded by greater than 50% undeveloped landscapes were called reference (short for
reference standard), wetlands surrounded by greater than 50% agricultural land uses were
called agricultural and wetlands surrounded by greater than 50% urban land uses were
called urban.
Random site selection was not feasible given 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 accomplished with the aide of the
Natural Resources Conservation Service under the United States Department of
Agriculture and University of Florida Institute of Food and Agricultural Sciences
extension agents. Sample wetlands were targeted spatially throughout Florida, so that a
nearly equal distribution of wetlands was sampled within each of the four Florida wetland
region. Boundaries of the Florida wetland regions were determined with a hydrologic
model by Lane (2000). Florida freshwater palustrine wetlands were classified using a
hierarchical classification technique, and physical surficiall geology, soils, digital
elevation model, slope) and climatic (precipitation, potential evapotranspiration, runoff,
annual days of freezing) variables were tested for correlation with wetland clusters. Final
wetland region boundaries were based on a spatial water balance model.
The number of wetlands sampled per apriori land use category per region varied,
with 28 wetlands in the south (n = 9 reference, n = 9 agricultural, n = 10 urban), 31 in the
central (n = 11 reference, n = 9 agricultural, 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) wetland regions. All 72 wetlands sampled in 2001 and
39 (of the 46) wetlands sampled in 2002 hosted Taxodium ascendens (pondcypress) as a
dominant or co-dominant canopy species. The remaining seven wetlands sampled in
2002 had a canopy layer comprised of a mixture of species that was not dominated or co-
dominated by Taxodium ascendens. These wetlands exhibited the characteristic
depressional shape in the landscape and from GIS based digital orthophotography.
Wetlands surrounded by natural landscapes were generally located on conservation lands





















Sample Wetlands (n=118)
o Reference
Agricultural
SUrban
Florida Ecoregions
Panhandle
| | North
m Central
[- South
0 200


400 Kilometers


Figure 2-1. Study site location of 118 isolated depressional forested wetlands in Florida.
The state of Florida was separated into four ecoregions (Lane 2000). Sample
wetlands were designated by a priori surrounding land use categories:
o reference, agricultural, or urban.

including state and national parks and forests, county 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 the widespread historic loss of wetlands
throughout Florida (FDNR 1988) and early incentives to drain swamp lands, few isolated
depressional forested wetlands were found in the oldest urban areas. Many of the urban
wetlands sampled were suspected to previously been embedded in agricultural land uses.
Table 2-1 provides some general information about each sample wetland,
including sample date, surrounding land use, and land ownership. The sample date
provided is the earliest sample date, and correlates to macrophyte 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 after initial hydration for the growing season,
which would not 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 wetland region (S = south; C = central;


I r I











Table 2-1. Surrounding land
wetlands in Florida.


Site Sample Surrounding


use, land ownership, and sample date for 118 study


Code*
SA1
SA2
SA3
SA4
SA5
SA6
SA7
SA8
SA9
SR1
SR2
SR3
SR4
SR5
SR6
SR7
SR8
SR9
SU1
SU2
SU3
SU4
SU5
SU6
SU7
SU8
SU9
SU10
CA1
CA2
CA3
CA4
CA5
CA6
CA7
CA8
CA9
CR1
CR2


Date
6/5/01
6/6/01
6/27/01
7/30/01
7/31/01
9/5/01
7/31/02
7/31/02
8/1/02
6/28/01
7/3/01
7/24/01
8/1/01
8/21/01
9/18/01
7/15/02
7/17/02
7/24/02
6/6/01
6/29/01
7/4/01
8/22/01
8/23/01
9/30/01
7/16/02
7/16/02
7/23/02
7/30/02
5/23/01
5/30/01
6/7/01
6/21/01
7/10/01
7/23/01
7/3/02
7/19/02
7/24/02
5/30/01
6/14/01


Land Use^
Cattle & Crops
Citrus
Cattle
Crops
Cattle & Crops
Cattle
Woodland
County Park
Cattle
County Park
State Park
State Reserve
National Park
State Preserve
NWR
County Park
County Airport
County Park
Resid. & Golf
School Campus
Residential
Residential
Industrial
Industrial
Commercial
Comm. & Res.
Residential
Roads & Canals
Crops
Cattle
Pullet Farm
Cattle
Cattle
Citrus
Silv. & Cattle
Dairy Farm
Citrus
Conserv. Tract
Conserv. Tract


Land
Owner
Public
Private
Public
Private
Public
Private
Public
Public
Private
Public
Public
Public
Public
Public
Public
Public
Public
Public
Private
Public
Public
Private
Private
Private
Private
Private
Private
Public
Private
Private
Private
Public
Private
Private
Public
Public
Private
Private
Private


Site
Code*
CR3
CR4
CR5
CR6
CR7
CR8
CR9
CR10
CR11
CU1
CU2
CU3
CU4
CU5
CU6
CU7
CU8
CU9
CU10
CU11
NA1
NA2
NA3
NA4
NA5
NA6
NA7
NA8
NA9
NA10
NA11
NA12
NR1
NR2
NR3
NR4
NR5
NR6
NR7


Sample
Date
6/20/01
8/10/01
8/13/01
8/15/01
5/30/02
7/2/02
7/11/02
10/9/02
10/9/02
5/31/01
6/15/01
7/16/01
8/14/01
9/11/01
9/12/01
5/30/02
7/1/02
7/8/02
8/7/02
8/8/02
5/21/01
6/4/01
6/19/01
7/20/01
7/27/01
7/31/01
5/22/02
5/21/02
6/10/02
7/12/02
7/24/02
7/26/02
5/26/01
6/18/01
7/10/01
7/11/01
8/6/01
8/21/01
5/28/02


Surrounding
Land Use^
State Park
WMD
State Park
State Forest
City Owned
State Forest
State Preserve
State Park
State Park
Univ. Campus
Residential
Commercial
Road Side
Road Side
Golf Course
City Owned
Industrial
Commercial
Park
Park
Cattle
Cattle
Silviculture
Crops
Cattle
Silv., Cat.,Crops
Crops
Silviculture
Silviculture
Silviculture
Cattle
Cattle & Crops
University Land
City Park
State Forest
WMD
Military
State Park
State Park


Land
Owner
Public
Public
Public
Public
Public
Public
Public
Public
Public
Public
Private
Private
Private
Private
Private
Public
Private
Private
Public
Public
Public
Private
Public
Private
Private
Private
Public
Private
Ease.
Ease.
Public
Public
Public
Public
Public
Public
Private
Public
Public










Table 2-1. Continued.
Site Sample Surrounding Land Site Sample Surrounding Land
Code* Date Land Use^ Owner Code* Date Land Use^ 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 PR1 6/15/01 National Forest Public
NU2 6/11/01 Resid. & Golf Private PR2 7/3/01 WMD Public
NU3 6/26/01 Residential Private PR3 7/4/01 Military Public
NU4 6/27/01 Residential Private PR4 8/9/01 State Forest Public
NU5 6/28/01 Residential Private PR5 8/10/01 State Forest Public
NU6 8/1/01 Resid. & Inst. Private PR6 8/18/01 National Forest Public
NU7 5/15/02 Comm. & Res. Private PR7 6/4/02 Conserv. Tract Private
NU8 6/3/02 Res. & Golf Private PR8 8/7/02 NWR Public
NU9 6/12/02 Industrial Private PU1 6/14/01 Residential Private
NU10 7/29/02 Res. & Instit. Private PU2 7/5/01 Residential Private
PA1 5/24/01 Cattle Private PU3 8/17/01 Res. & Comm. Private
PA2 5/29/01 Cattle Private PU4 8/17/01 Res. & 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 Res. & Orchard Private
PA6 8/9/01 Cattle Private PU8 6/19/02 Ind. & 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 = reference, A = agricultural, and U = urban.
ASurrounding Land Use abbreviations: NWR = National Wildlife Refuge; WMD = Water Management
District; Res. = Residential; Cat. = Cattle; Comm. = Commercial; Inst. = Institutional; Ind. = Industrial;
Crops = Row Crops; Silv. = Silviculture; Ease. = Easement.


N = north; P = panhandle), land use category, and order of sampling. Site codes were
assigned to preserve the anonymity of land owners.

Gradients of Anthropogenic Activity

Three independent indices of anthropogenic activity were calculated including the
Landscape Development Intensity (LDI) index, the Wetland Rapid Assessment Procedure
(WRAP), and the Minnesota disturbance index (DI). LDI scores were calculated prior to
site visits using 1999 digital orthophoto imagery 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 constructed around the edge of
each wetland in ArcView GIS 3.2 (Environmental Systems Research Institute, Inc. 1999).
Land uses within the 100 m wetland buffer were delineated based on digital orthophoto










Table 2-2. Field-data collected at 118 sample wetlands.


SA1
SA2
SA3
SA4
SA5
SA6
SA7
SA8
SA9
SR1
SR2
SR3
SR4
SR5
SR6
SR7
SR8
SR9
SU1
SU2
SU3
SU4
SU5
SU6
SU7
SU8
SU9
SU10
CA1
CA2
CA3
CA4
CA5
CA6
CA7
CA8
CA9
CR1
CR2
CR3


0


/ = Data collected
A 6-sampled with >10 cm standing water; --sampled with < 10 cm standing water


o
vl


o
vl





<




c+ CR4
S/ 6 CR5
S 0 + CR6
V V 6 V CR7
/ / 6 V/ CR8
V V 6 V CR9
/ 6 V CR10
/ 6 CR11
/ 6 / CU1
/ 6 CU2
/ /6 V/ CU3
S/ 6 / CU4
/ 6 C/ CU5
/ 6 / CU6
/ 6 / CU7
/ 6 v CU8
/ 6 / CU9
/ 6 / CU10
// + / CU11
/ / + NA1
/ / 6 / NA2
S/ 6 V NA3
/ / 6 V NA4
/ / 6 / NA5
/ 6 / NA6
/ 6 NA7
6 / NA8
+ NA9
+ NA10
/ / + / NA1
/ / + V NA12
S/ + / NR1
/ / 6 / NR2
/ 6 / NR3
/ 6 / NR4
/ + NR5
6 / NR6
/ + NR7
+ NR8
/ / / NR9


/ / 6 / NU1
/ 6 / J NU2
/ /6 / NU3
+ NU4
/ 6 / NU5
/ 6 / NU6
/ 6 / NU7
/ 6 / NU8
/ / + / NU9
+ NU10
S/ 6 / PA1
+ PA2
/ /6 v PA3
/ / 6 PA4
/ + / PA5
/ 6 / PA6
/ 6 / PA7
/ 6 / PA8
+ PA9
+ PA10
+ PR1
+ PR2
/ 6 / PR3
+ PR4
/ / 6 V PR5
+ PR6
+ PR7
+ PR8
/ 6 PU1
/ 6 / PU2
+ PU3
+ PU4
/ / + / PU5

/ / 6 / PU7
+ PU8
" 6 PU9
+ PU10
/ 6 /:
/ 6 /
v V/^
J A ^


+

/ / +

/ / +
/ / +




/ 6

/ / +





/ ,/6



/ 6
,/ 6






,/
/ 6
+





/ 6
/ 6
/ +



+
" +



" 6
J
J
J


J J
JJ
J









imagery, and were updated during the site visit to reflect changes in land use since the
1999 images were recorded. The following equation was used to calculate LDI:

LDITotal = Y % LUi LDIi (2-1)

where %LUi is the percent of a land use within wetland buffer and LDIi is the LDI
coefficient for a particular land use based on the amount of nonrenewable energy use per
unit area in the surrounding landscape (Table 2-3). The LDI coefficient values and LDI
equation are based on work by Brown and Vivas (2005). 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 minutes at each study wetland according
to descriptions from Miller and Gunsalus (1999). The DI was scored in the office after
the field visit by the field crew leader using information obtained from ArcView GIS 3.2
(Environmental Systems Research Institute, Inc. 1999) and field notes according to
categories established by Gernes and Helgen (1999). The three measures of
anthropogenic activity were compared using the Pearson correlation coefficient.

Field-data Collection

A concise summary of field-data collection procedures follows. Appendix A
provides more detailed descriptions of field-data collection techniques in the format of
Standard Operating Procedures (SOPs) for field use. Field methods are described as
transect establishment followed by water, soil, diatom, macrophyte, and
macroinvertebrate sampling techniques.
Figure 2-2 shows the positioning of the four 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 cardinal directions (north, east, south,
west). The upland/wetland boundary was determined using the Florida Unified Wetland
Delineation Methodology (Chapter 62-340, F.A.C.), using a combination of wetland
plant presence according to wetland plant status (e.g. obligate, facultative wetland,
facultative, or upland) and wetland hydrologic indicators (e.g. lichen lines, moss collars).

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 present throughout at least 50% of the wetland
surface 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 wetlands,
including 14 in the panhandle wetland region (reference n = 5; agricultural n = 4; urban
n = 5), 14 in the north wetland region (reference n = 6; agricultural n = 3; urban n = 5), 23
in the central wetland region (reference n = 9; agricultural n = 6; urban n = 8), and 24 in
the south wetland region (reference n = 8; agricultural n = 8; urban n = 8).
Dissolved oxygen and water temperature were taken on-site using an YSI-55
Dissolved Oxygen hand meter. Grab water samples were sent to the FDEP Central
Chemistry Laboratory, Tallahassee, Florida. Analysis included color (EPA 110.2),
turbidity (EPA 180.1), pH (150.1), specific conductance (EPA 120.1), ammonia- nitrogen









Table 2-3. Landscape Development Intensity coefficients (LDIi) used in the calculation
of the Landscape Development Intensity (LDI) index.
Nonrenewable Energy Use
(E14 solar equivalent
Land Use (i) joules/ha/yr) LDIi
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


(EPA 350.1), nitrate/nitrite-nitrogen (EPA 353.2), total Kjeldahl
and total phosphorus (EPA 365.4).


nitrogen (EPA 351.2),


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 transect (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


































Figure 2-2. Belted transect layout for macrophyte 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 (6) was taken in the
approximate center of each wetland.


Food and Agricultural Sciences (IFAS) Analytical Research Laboratory, Gainesville,
Florida.

Sampling Design

Benthic diatoms
Benthic diatom samples were collected at 50 isolated depressional forested
wetlands throughout Florida between May-September 2001, as listed in Table 2-2.
Figure 2-3 shows the 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) wetland regions. Sample wetlands were situated in three apriori described
land use categories (reference n=18; agricultural n=16; urban n=16).
A minimum of 10 cm of standing water was necessary for benthic diatom
collection. 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 loosen the top 0.5 cm
layer at the soil surface-water interface, and a 10 mL sample was extracted. This was
repeated 10 times throughout the flooded portion of the wetland, resulting in a final
sample volume of 100 mL. For preservation, 5 mL of M3, a standard preservative for



















Sample Wetlands (n=50) *
o Reference o
Agricultural
Urban
Florida Ecoregions o
Panhandle a
I -- North N |
|LI Central
Li South
0 200 400 Kilometers


Figure 2-3. Benthic diatom samples were collected at 50 isolated depressional forested
wetlands. The state of Florida was separated into four wetland regions
(Lane 2000). Sample wetlands were designated by surrounding land use:
o reference, agricultural, or urban.


algae samples (APHA 1995), was added to the 100 mL algae sample. Appendix A
provides the recipe for creating a batch of M3 preservative.
Benthic diatom 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 identification. 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 using Naphrax (Northern
Biological Supplies Limited, Ipswich, England). Five hundred valves were counted
along microscope transects and identified to the lowest possible taxonomic level,
preferably species (following FDEP SOP#AB-03.1 http://www.dep.state.fl.us/labs/sop).

Macrophytes
Macrophyte vegetation was sampled at all 118 wetlands throughout Florida
(Figure 2-1). Sampling was conducted along four transects situated as perpendicular
crossing axes running through the center of each wetland (Figure 2-2). Along each









transect, a series of 1 m wide by 5 m long quadrats was established back to back. Living
macrophytes rooted within each quadrat were identified to the lowest 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 metrics, including category (annual or
perennial, evergreen or deciduous, indigenous or exotic) and growth form (aquatic, fern,
grass, herb, sedge, shrub, tree, or vine). References specific to Florida were consulted
first (Tobe et al. 1998; Wunderlin and Hansen 2003). 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, Florida botanists (who also participated in the Floristic Quality
Assessment Index) were consulted.

Floristic Quality Assessment Index. A Floristic Quality Assessment Index
(FQAI) has been included in many of the multi-metric biotic indices created for the
macrophyte assemblage. The concept of FQAI was developed by Wilhelm and Ladd
(1988) for vegetation around Chicago, Illinois. This method of scoring plant species
based on expert 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 et al. 2003; Cohen et al.
2004). The FQAI provides a quantitative means of assessing the fidelity of a plant to a
particular environment through the Delphi technique (Kent 2000), where individual
botanists assigned coefficients (termed coefficients of conservatism) to individual
species. 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).
The Florida FQAI enlisted regional expert botanists to provide quantitative scores
for vegetation in the form of a coefficient of conservatism assigned individually to each
species. The FQAI score for an individual wetland was calculated as:

FQAI = [ (C+ C2 +. .. Cn) ]/N (2-2)

where C = species specific coefficient of conservatism score (CC), and N = species
richness. This equation was considered a modified FQAI because previous studies used
the square root of native species richness as the denominator. Theories on the importance
of species richness suggest that higher species richness signifies a more valuable
ecosystem, which can be quantified by using the square root function (Fennessy et al
1998). However, a recent study by Cohen et al. (2004) found a stronger relationship
along a disturbance gradient for the FQAI of Eq. 2-2 than with the traditional FQAI
equation (using the square root of native species richness in the denominator). They also
reported that using total species richness (i.e. including exotic species in the calculation
of species richness) improved the relationship of mean CC with the human disturbance
gradient (measured with LDI). The sum of the species CC scores was divided by total
species richness (Eq. 2-2) in this study to account for potential differences in species
richness due to differences in wetland regions (Lane 2000), bioregions (Griffith et al.
1994), apriori land use categories, or other unspecified differences.









CC scores were obtained from the Florida botanists surveyed. Each botanist was
sent a complete list of species identified in the isolated forested wetlands in the 2001 field
season (n = 482 species), and was asked to score each species based on its faithfulness to
Florida isolated forested wetlands. After the 2002 field season, one botanist scored the
additional 79 taxa not previously encountered, raising the number of taxa with CC scores
to 561 species. Potential CC scores ranged from 0 to 10:
0 Exotic and native species that act as opportunistic invaders,
included species that commonly occur in disturbed ecosystems
1-3 Species that were widely distributed and occurred in disturbed
ecosystems
4-6 Species with a faithfulness to a particular ecosystem, but also
tolerant of moderate levels of disturbance
7-8 Species typical of well-established ecosystems that sustain only
minor disturbances
9-10 Species occurring within a narrow set of 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 B lists the CC scores for the macrophyte species
identified in this study.

Macroinvertebrates
Macroinvertebrates were collected at 79 isolated depressional forested wetlands
throughout Florida (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. Wetlands were sampled in the panhandle (n = 13), north (n
= 15), central (n = 25), and south (n = 26) wetland regions. Sample wetlands were
situated in 3 apriori land use categories (reference n = 29, agricultural n = 24, and urban
n = 26).
Samples were collected using a U.S. Standard 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 recently been trampled by the field crew.
Twenty sweeps were proportioned among major vegetation zones throughout each
sample wetland. Sample wetlands generally had only one vegetation zone, although
some wetlands did have two or three distinct vegetation zones. Vegetation zones were
defined by clear changes in the dominant species. When herbaceous plants were included
in the sweep area, the bottom of the net was swept from the substrate up the plant stalks.
In areas with woody plant structure, 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 plastic jar
and preserved with buffered formalin at a rate of 10% of the sample volume. Appendix
A provides a more detailed description of field methods and preservation.
Macroinvertebrate identification was completed at the FDEP Central Laboratory,
Tallahassee, Florida, following standard operating procedures (FDEP Standard Operating


































300 Kilometers


Figure 2-4.


Macroinvertebrates were sampled at 79 isolated depressional forested
wetlands. The state of Florida was separated into four wetland regions
(Lane 2000). Sample wetlands were designated by surrounding land use:
o reference, agricultural, or urban.


Procedure #IZ-06 http://www.dep.state.fl.us/labs/sop). Macroinvertebrate samples were
sieved, washed, and placed on a pan with 24 individually numbered cells. Eight of the
cells were randomly selected and combined in a second numbered tray. A single cell was
randomly selected from the second tray for enumeration and identification of
macroinvertebrates. When fewer than 100 individuals were encountered in the sample, a
second cell was randomly selected from the second tray, and all of the individuals were
enumerated and identified. Identification was to the lowest taxonomic level possible.


N
N


Land Use C at.ori.s Florida Regions
0 Refer r .ce Pa Ar.h.f-l-
.sr1icut1rf1e m Nor.:th
Ur-ar. | | Cartal
souI


200


I I r I









Data Analysis

Water and Soil Parameters

Water and soil parameters within three a priori land use categories were
compared using Fisher's LSD pair wise comparison in Minitab (Version 13.1, Minitab
Statistical Software). The non-parametric Mann-Whitney U-Test was used to discern
differences among medians of low (LDI < 2.0) and high (LDI > 2.0) LDI groups for the
water and soil parameters. An LDI break of 2.0 corresponds to a break in the LDI
coefficients of undeveloped versus developed land uses (Table 2-3).

Summary Statistics

Summary statistics for each assemblage included richness (R), evenness (E), and
Shannon diversity (H). For the diatom and macrophyte assemblages summary statistics
were calculated at the species level, and for the macroinvertebrate assemblage summary
statistics were calculated at the genus level. Richness was defined as the total number of
distinct taxa encountered within the sample wetland. Evenness has been described as the
fraction of maximum possible diversity in a wetland and was calculated as the Shannon
diversity index value divided by the natural log of taxa richness:
E = H / log (R) (2-3)
(McCune and Grace 2002). The Shannon diversity index has been described as
measuring the "information content" of a sample unit where maximum diversity yields
maximum uncertainty (McCune and Grace 2002). For Shannon diversity calculations
(H), the sample unit was an individual isolated depressional forested wetland:
H = pi log(pi) (2-4)
pi= ni / N (2-5)
where ni was the number of occurrences of taxon i, and N was the total number of
occurrences of all taxa at a wetland. For the diatom and macroinvertebrate assemblages,
the number of occurrence represented the enumeration of the laboratory identified sample
and the total number of occurrences represented the sum of the number of occurrences of
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 total number of quadrats of all of the species identified.
For the diatom and macroinvertebrate assemblage, Simpson's index (S) was also
calculated, and the equation:
S = 1 I (pi pi) (2-6)
where pi was defined in Equation 2-5 as the number of occurrences of taxon i (ni)divided
by the total number of occurrences of all taxa at a wetland (N). For the macrophyte
assemblage, first- and second-order jackknife estimators of species richness (Jacki and
Jack2, respectively) and Whittaker's beta diversity (PW) were calculated. First- and
second-order jackknife estimators were calculated as estimates of true species richness
(Colwell and Coddington 1994). Jackknife estimators are used to gauge the thoroughness









of sampling intensity and the inherent variability of the target system. Assuming that the
sampling effort only 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 richness is based on the number of species observed (S),
the number of species occurring in only one sample unit (rl) (where one sample unit
represents one quadrat), and the number of sample units (n) (quadrats):
Jack = S + [(rl*(n-1)) / n] (2-7)
The second-order jackknife estimator (Jack2) also incorporated the number of species
occurring in exactly two sample units (r2), where a sample unit was defined as a quadrat:
Jack2 = S + [(rl*(2n-3)) / n] [(r2*(n-2)2) / (n*(n-1))] (2-8)
These estimators of total species richness have shown useful in predicting actual species
richness when only a small area of the total ecosystem has been sampled (McCune and
Grace 2002).
Whittaker's beta diversity (PW) was computed as a calculation of overall beta
diversity, or the compositional change represented in a sample. Whittaker's beta
diversity was calculated as the number of species at a particular forested wetland (Sc)
divided by the average species richness per quadrat (S) minus one:
pw = [Se / S] 1 (2-9)
The resulting value for Whitaker's beta diversity was described as the "number of distinct
communities" (McCune and Grace 2002). When Ow 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 greater than five is considered high (McCune and
Grace 2002).
Summary statistic means within a priori land use categories were compared with
Fisher's Least Significant Difference (LSD) pair wise comparison test using Minitab
(Version 13.1, Minitab Statistical Software). The strength of using Fisher's LSD is 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 values 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 three apriori land use categories (reference, agricultural, urban). Gamma
diversity was calculated as the overall number of taxa encountered at all sample wetlands
per a priori land use category. A higher value of gamma diversity for an a priori land
use category suggests a greater difference among the species composition of wetlands
within that a priori land use category, assuming equal sample sizes among a priori land
use categories. 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 (MRPP) was used to test the
similarity of community composition for each assemblage among the four Florida









wetland regions (further application in Zimmerman et al. 1985; McCune et al. 2000;
McCune and Grace 2002). MRPP is a nonparametric technique which tests for no
difference between groups (the null hypothesis) and is available in PCORD (Version 4.1
from MJM Software, Gleneden Beach, Oregon). It is an appropriate procedure for
ecological community data as it does not require distributional assumptions of normality
and homogeneity of variances. The Sorensen distance measure was used to calculate 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 within-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 groups (panhandle versus north versus
central versus south) as well as for multiple pair wise comparisons (panhandle versus
north, panhandle versus central, panhandle versus south, north versus central, north
versus south, and central versus south) for the Florida wetland regions (Lane 2000).

Community Composition

Community composition of each assemblage was summarized in a non-metric
multidimensional scaling (NMDS) ordination to relate changes in community
composition with environmental gradients. NMDS is an ordination technique designed to
compress multi-dimensional space, and is particularly agreeable with ecological data
because it does not rely on linear relationships among variables. This has been described
as a compensation for the "zero-truncation problem" through the use of ranked distances
and the application 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 (i.e. Principal Components Analysis, PCA) depend
upon a value for each measured variable for each sample unit, interpreting shared zeros
as a positive relationship. NMDS is useful for presence/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, NMDS caters to non-parametric, community ecology data sets.
NMDS 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 chosen based on an initial six dimensional run in
autopilot mode, which suggested an optimal three dimensional solution for the each
community composition dataset. To find the optimal three dimensional solution, 50 runs
with real data and 50 randomized runs were performed with the instability 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 stability and reproducibility in results. The
final run was completed with the starting point set as the results from the best
experimental with dimensional run, considered the run with the lowest stress and best
overall fit.
Water and soil parameters, LDI, latitude, and longitude were correlated with the
NMDS ordination axes with Pearson correlation coefficients. To improve normality and
decrease skewness, 11 water and soil parameters were log (base 10) transformed,
including water parameters (dissolved oxygen concentration, temperature, color,









turbidity, specific conductivity, ammonia-nitrogen concentration, nitrate/nitrite-nitrogen
concentration, total Kjeldahl nitrogen (TKN) concentration, and total phosphorus (TP)
concentration) and soil parameters (TKN and TP concentration). Water pH was not
transformed as it is already on a log scale (pH is calculated as the log base 10 of the
hydronium ion concentration). The remaining parameters were measured as percentages,
and were transformed by taking the arcsine square root, including soil moisture and soil
organic matter.

Metric Development

In the context of this study, metrics were defined as biological attributes which
have a consistent and predictable response to anthropogenic activities (Karr and Chu
1997). Metrics were summarized in five main categories:
1. Tolerance metrics (determined with indicator species or established
index values)
2. Autecological metricss that explore a previously described relationship
between taxa and an environmental gradient)
3. Community structure metricss that explore taxonomic structure)
4. Community balance metricss with calculated values, such as evenness
or dominance)
5. Functional group metricss related to feeding behavior)
Appendix C provides tables of candidate metrics for each assemblage, including
59 candidate diatom metrics (Table C-l), 68 candidate macrophyte metrics (Table C-2),
and over 100 candidate macroinvertebrate metrics (Table C-3). Candidate metrics were
calculated at the statewide scale for both the diatom and macroinvertebrate assemblages,
as sample sizes were limited for regional metric development, particularly in the north
and panhandle wetland regions. Candidate macrophyte metrics were calculated at both
the regional scale and statewide.
Metrics were calculated in multiple forms including richness (R) and proportion
(P) metrics. The richness metric (R) simply represents a count of the taxa present within
a metric category. The proportion metric (P) was calculated as the richness metric (R)
divided by the total species richness (N) for each sample wetland:
Pi = Ri / Ni (2-10)
Candidate metrics were accepted if they showed a constant and predictable change along
the LDI (Brown and Vivas 2005) according to the strength and significance of the
Spearmans correlation coefficient (Analyze-It software v. 1.67 for Microsoft Excel). The
Spearman rank correlation tests for an association between two related variables, and is a
non-parametric alternative to the Pearson correlation coefficient. Scatter plots were
constructed for each candidate metric versus LDI to ensure correlations were visually
distinguishable. Candidate metrics were subjected to a Mann-Whitney U-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. ISA 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 sites be assigned into groups and species data on the abundance or presence of
taxa in each group. These groups are commonly defined by categorical environmental
variables, levels of disturbance, experimental treatments, presence and absence of a target
species, or habitat types (McCune and Grace 2002). Mathematical equations are
available in Dufrene and Legendre (1997) and McCune and Grace (2002). The
calculated indicator species values are based on two standards, faithfulness and
exclusion. Faithfulness is defined mathematically by a particular taxa always being
present in a particular group. The perfect indicator taxon is exclusive to a group,
meaning it would never occur in other groups. Indicator values range from 0 (no
indication) to 100 (a perfect indication of a particular group).
ISA was conducted to determine sensitive and tolerant indicator taxa separately
for each species assemblage. Statewide ISAs were run for the diatom and
macroinvertebrate assemblages. ISA was conducted for each wetland region (panhandle,
north, central, and south) as well as statewide for the macrophyte assemblage. Since
correlations between LDI and each species assemblage were uncertain, iterative ISAs
were calculated for each species assemblage. ISA depends on the predetermined
categorization of sites, and sites were divided into groups based upon the calculated LDI
scores. ISA was run for each LDI break from 1.0 to 7.0 at each 0.5 increment. For
example, at an LDI break of 1.0, all sites with a calculated LDI of 1.0 were assigned to
Group A, and all sites with a calculated LDI > 1.0 were assigned to Group B, and an ISA
calculation completed. Then, for a separate iteration, sites were divided in Group A (LDI
< 1.5) and Group B (LDI > 1.5) and an ISA calculation was completed. Next, sites were
assigned to Group A (LDI < 2.0) and Group B (LDI > 2.0) and an ISA calculation was
completed. This process of assigning sites into groups based on their calculated LDI
scores was repeated at LDI breaks of 2.5, 3.0, 3.5, 4.0, 4.5, 5.0, 5.5, 6.0, 6.5, and 7.0.
The idea behind the iterative ISA calculations was to broaden the application of
the LDI to detect changes in species composition among wetland sites for different
species assemblages and for different anthropogenic influences which may be specific or
more dramatic to one species assemblage. The final species lists for tolerant and
sensitive indicator species were determined for the LDI break yielding the greatest
number of indicator species and the strongest significance based on the Monte Carlo
randomization technique. Calculated indicator values were tested for statistical
significance using a Monte Carlo randomization technique with 1000 randomized runs.
Indicator taxa categorized as tolerant taxa were associated with the higher LDI group
(Group B); indicator taxa deemed sensitive taxa were associated with the lower LDI
group (Group A).
In the macrophyte assemblage, the Spearman rank correlation was used to assess
differences between statewide and regional indicator species lists for each of the four
wetland regions. To test for equal application of the statewide indicator species list for
each wetland region, the non-parametric Kruskal-Wallis test was run with Analyse-It









software (Ott and Longnecker 2001). Distributional differences were analyzed between a
priori categories for both tolerant and sensitive indicator taxa among wetland regions.

Diatom Metrics

Diatom metrics (Appendix C Table C-1) were adopted from previous studies and
could be categorized in three categories 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 with ISA. Community composition metrics included
richness, evenness, and diversity calculations. Autecological metrics were based on
available research that correlated individual diatoms with morphology, behavior, and the
physical and chemical water environment. Diatom species were assigned ecological
indicator values 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
categorizing diatoms according to water preference (moisture tolerance), nitrogen
metabolism, pH, salinity, dissolved oxygen, saprobic condition, and trophic status.
Pollution tolerance classification values were taken from Bahls' (1993) analysis of
diatoms in Montana streams. Metrics based on morphology and motility were assigned
from Stevenson (Lane 2003).

Macrophyte Metrics

Macrophyte metrics (Appendix C Table C-2) were adopted from previous studies
and included tolerance, exotic species, Floristic Quality Assessment Index (FQAI),
longevity and plant growth form (native perennial), 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. 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 Wunderlin and Hansen 2003). The exotic species metric was
calculated as the proportion of species that were exotic to Florida divided by the number
of species identified at each particular isolated depressional forested wetland. The
timeline for determining the exotic status of a species was set near the beginning of
European settlement in North America. 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) determined that disturbance often favors annual species over perennial
species and promotes the invasion of nonnative perennials in wetlands. Galatowitsch et
al. (2000) found that while native perennial cover was reduced in wetlands impacted by
cultivation, the occurrence of introduced perennials rather than annuals increased in
stormwater impacted wetlands.
The wetland status metric was calculated as the percent of plants classified as
obligate or facultative wetland indicator species divided by the number of species at each
wetland. Wetland indicator status classifications was from FDEP's vegetative index as
presented in section 62-340.450, F.A.C., which can be found in Tobe et al. (1998), USDA









NRCS (2002), and Wunderlin and Hansen (2003). There are five potential species
classifications, including obligate, facultative wetland, facultative, upland, and invisible.
In some cases, when the Florida wetland indicator status was not available, the National
Wetlands Inventory wetland indicator status for the United States was used.

Macroinvertebrate Metrics

Candidate metrics for the macroinvertebrate assemblage (Appendix C Table C-3)
were adopted from previous studies and could be classified in four categories, including
tolerance, community structure, community balance, 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 calculated. Many of
the established tolerance metrics were created for flowing waters, which complicated the
application of established tolerance values (ex. the Florida Index from Beck 1954 and
Barbour et al. 1996b; 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 included richness measures (Danielson 1998a), for
example the number of distinct species or specified taxonomic units such as the number
of families, genera, or species in a sample. Examples of taxa richness metrics include
total taxa richness, Ephemeroptera richness, number of Coleoptera species, or number of
Insecta species.
The use of community balance metrics included some measure of abundance or
relative abundance in an attempt to measure the evenness of the macroinvertebrate
community (Lenat and Barbour 1994). Examples of community balance metrics include
the Shannon diversity index or the percent contribution of the most abundant taxon
(Lenat and Barbour 1994).
Macroinvertebrate taxa were grouped based on their functional relationships that
overlap taxonomic categorization, including functional feeding groups, habitat groups,
and voltinism groups (life-cycle patterns). Cummins and Merritt (2001) suggest using
ratios of numerical abundance or, more favorably, biomass of the various functional
groups as indicators of ecosystem attributes, essentially considering the functional groups
as surrogates of ecosystem condition. Functional feeding group metrics were based on
the morphological structures and behaviors responsible for food acquisition by particular
taxa at a site (Resh and Jackson 1993; Danielson 1998a). For example, herbivores
consume algae and plant material, while predators consume animals, omnivores eat both
plant and animal materials, and detritovores consume decomposed particulate material
(Helgen 2001).









Florida Wetland Condition Index

Candidate metrics were selected for inclusion in the Florida Wetland Condition
Index (FWCI) if they satisfactorily met three criteria:
Metrics were correlated with the LDI according to the strength and
significance of the Spearmans correlation coefficient
Metrics displayed visually distinguishable correlations with LDI in
scatter plots
Metrics showed significant differences between low (LDI < 2.0) and
high (LDI > 2.0) LDI groups tested with the Mann-Whitney U-test.
An FWCI was constructed for each assemblage, including the diatom FWCI, the
macrophyte FWCI, and the macroinvertebrate FWCI. Each FWCI was composed of
individual metrics specific to the assemblage, which were scaled and added together.
Metric scoring was based on an approach modified from the SCI, a Florida based
biological index of the macroinvertebrate assemblage used to discern stream condition
(Fore 2004). Metrics with a skewed distribution were log transformed to improve the
distribution. The 5th to 95th percentile values of each metric were normalized from 0 to
10, with 10 always representing the best biological wetland condition. The selected
metrics, FWCI, and LDI were correlated with water and soil parameters using Spearmans
correlation coefficient.
In order to determine whether the FWCI provided comparable scores for wetlands
with similar community composition within each assemblage, an agglomerative cluster
analysis in PCORD was used to determine wetland clusters (McCune and Grace 2002).
The dissimilarity matrix was constructed using the Sorensen distance measure and the
flexible beta linkage method (P = -0.25), which is a flexible clustering setting designed to
reduce chaining in the dendrogram. The resulting dendrogram was pruned to maintain
the smallest number of significantly different clusters based on Fisher's LSD pair wise
comparison (p < 0.05). To compare differences in wetland condition between the three
species assemblages, metrics selected for inclusion in the three FWCIs among different
species assemblages were compared using the Pearson correlation coefficient (Analyse-it
Software).











CHAPTER 3
RESULTS

Quantifying Anthropogenic Influence

Three independent quantitative measures of anthropogenic activity were
calculated for the 118 isolated depressional forested wetlands including the Landscape
Development Intensity (LDI) index, the Wetland Rapid Assessment Procedure (WRAP),
and the Minnesota disturbance index (DI). Individual scores for each wetland are
presented in Appendix D. Potential LDI scores ranged from a minimum of 1.0 to a
maximum of 10.0 representing a highly impacted wetland. Actual LDI scores ranged
from 1.0 at 26 wetlands to 7.2 at two wetlands including SU8 and CU6, two urban
wetlands surrounded by transportation and commercial land uses. The mean LDI score
for a priori reference wetlands was 1.1 (a = 0.2). A priori agricultural and urban
wetlands had similar mean LDI scores at 4.5 (a = 1.4) and 4.8 (a = 1.5), respectively.
Scores for the WRAP potentially range from 0.0 to 3.0, with the lowest score of
0.0 representing a severely impacted wetland. Actual WRAP scores ranged from 3.0 at
seven apriori reference wetlands to 0.5 at two urban wetlands including SU8 and PU10.
The mean WRAP score for a priori reference wetlands was 2.8 (a = 0.2), with a priori
agricultural and urban wetlands having similar mean scores of 1.6 (a = 0.5) and 1.4 (a =
0.4), respectively.
The third gradient of anthropogenic influence, the Minnesota DI ranges from zero
to 17 with unimpacted wetlands receiving a score of zero. Actual Minnesota DI scores
ranged from zero at 15 a priori reference wetlands to 17 at three a priori agricultural
wetlands. The mean Minnesota DI score for a priori reference wetlands was 1.6 (a =
2.1). A priori agricultural and urban wetlands had higher mean Minnesota DI scores at
11.3 (a = 5.4) and 12.2. (a = 3.6), respectively.
Comparisons among the three measures of anthropogenic influence showed
strong and significant correlations (Pearsons correlation coefficient Irl > 0.84, p < 0.01)
(Table 3-1). WRAP and the Minnesota DI had the greatest correlation coefficient (r =
0.89). The comparisons between LDI and WRAP and LDI and the Minnesota DI also
had large correlation coefficients (0.87 and 0.84, respectively). The LDI index was
selected for further analysis and metric selection for the development of the Florida
Wetland Condition Index (FWCI) because it is an objective and repeatable measure of
anthropogenic activity and it is not based on the subjective scoring of an assessor.

Water and Soil Parameters

Physical and chemical properties of 75 water samples and 118 soil samples were
analyzed for isolated depressional forested wetlands. Table 3-2 shows mean values for
water and soil parameters for the three a priori land use categories (reference,
agricultural, and urban). Means with similar letters were not significantly different
(Fisher's LSD pair wise comparison, a = 0.05), including water temperature, water









Table 3-1. Comparison among three measures of anthropogenic influence, including the
Landscape Development Intensity (LDI) index, the Wetland Rapid Assessment Procedure
(WRAP), and the Minnesota disturbance index (DI).
LDI WRAP
WRAP 0.87
DI 0.84 0.89
Values are Pearson correlation coefficients.
All correlations were significant (p < 0.01).


nitrate/nitrite-nitrogen, and soil TKN, which were not significantly different among a
priori land use categories.
Of the 14 water and soil parameters measured, eight water and three soil
parameters showed significant difference (p < 0.05) among reference, agricultural and/or
urban wetlands, categorized by the three apriori land use categories. Reference wetlands
had significantly different dissolved oxygen (x = 2.9 mg 02/L), turbidity (x = 3.8 NTU),
water pH (x = 5.2), and water column TP (x = 0.08 mg P/L), than agricultural and urban
wetlands. Water ammonia-nitrogen (mg N/L), water TKN (mg N/L), soil moisture, and
soil TP (mg P/g soil) were significantly different between reference and agricultural
wetlands. Specific conductivity was significantly different between reference (x = 81
umhos/cm) and urban wetlands (x = 231 umhos/cm). The water color of urban wetlands
(x = 198 PCU) was significantly different from reference (x = 285 PCU) and agricultural
(x = 346 PCU) wetlands. Soil organic matter was significantly different between
agricultural (x = 30%) and urban wetlands (x = 41%).
Measures of four water and two soil parameters were significantly different
(p<0.05) between sample wetlands in low (LDI < 2.0) and high (LDI > 2.0) LDI groups
(Table 3-3). Dissolved oxygen levels were higher in less developed wetlands (x = 2.9 mg
02/L) as compared to wetlands surrounded by higher development intensity (x = 1.8 mg
02/L), and wetlands surrounded by lower development intensity had lower water column
pH (x = 5.2) than wetlands in higher development intensity landscapes (x = 6.3). Low
LDI wetlands had significantly less TP in both the water column (x = 0.08 mg P/L) and
soil (x = 0.39 mg P/L), than the high LDI wetlands. The physical measure of turbidity
showed that wetlands surrounded by low development intensity had lower turbidity (x =
3.8 NTU) than those surrounded by higher development intensity (x = 13.2 NTU).
Additionally, low LDI wetlands had higher soil moisture (x = 59) than high LDI wetlands
(x = 50).













Table 3-2. Water and soil parameters among three apriori land use categories, including reference, agricultural, and urban wetlands.

Reference Agricultural Urban
n x a 95% CI n x a 95% CI n x a 95% CI
Water Parameters
Dissolved oxygen 29 2.9 1.7 (2.3 to 3.6) a 18 1.6 0.9 (1.2 to 2.1) b 24 1.9 1.1 (1.4 to 2.4) b
(mg/L)
Temperature (OC) 29 26.2 2.8 (25.1 to 27.3) a 18 25.2 1.9 (24.2 to 26.2) a 24 24.9 2.4 (23.9 to 25.9) a
Color (PCU) 30 285 178 (218 to 352) a 19 346 204 (248 to 445) a 26 198 129 (146 to 251) b
Turbidity (NTU) 30 3.8 4.2 (2.3 to 5.4) a 19 17.7 40.7 (0.0 to 37.3) b 26 9.5 11.9 (4.7 to 14.3) b
pH 30 5.2 1.2 (4.8 to 5.6) a 19 6.2 0.8 (5.8 to 6.6) b 26 6.4 1.0 (5.9 to 6.8) b
Specific
conductivity 10 81 48 (46 to 115) a 10 136 134 (39 to 232) ab 13 231 175 (126 to 337) b
(umhos/cm)
Ammonia-itrogen 30 0.15 0.33 (0.03 to 0.28) a 19 0.33 0.57 (0.06 to 0.61) b 26 0.19 0.27 (0.08 to 0.30) ab
(mg N/L)
Nitrate/nitrite-
nitrogen 26 0.09 0.37 (0.00 to 0.24) a 15 0.01 0.01 (0.00 to 0.01) a 21 0.02 0.03 (0.00 to 0.03) a
(mg N/L)
TKNnitrogen 30 1.93 1.24 (1.47 to 2.39) a 19 3.17 2.20 (2.11to 4.23) b 26 1.84 1.06 (1.41 to 2.27) ab
(mg N/L)
Total phosphorus 30 0.08 0.11 (0.04 to 0.12) a 19 0.81 1.38 (0.15 to 1.47) b 26 0.24 0.26 (0.13 to 0.34) b
(mg P/L)
Soil Parameters
Moisture (%) 38 61 20 (54 to 67) a 39 46 17 (40 to 51) b 41 55 20 (48 to 61) ab
Organic matter (%) 38 40 25 (32 to 48) ab 39 30 17 (24 to 35) a 41 41 28 (32-50) b
TKN nitrogen
(mg38 6.76 3.68 (5.55 to 7.97) 39 5.53 3.30 (4.46 to 6.59) a 41 6.71 4.75 (5.21 to 8.20) a
Total phosphorus 38 0.38 0.28 (0.29 to 0.47) a 39 0.91 1.27 (050 to 1.32) b 41 0.53 0.31 (0.43 to 0.63) ab
(mg P/g soil)
*Minimum Detection Limits: Ammonia-nitrogen 0.010 mg N/L; Nitrate/nitrite-nitrogen 0.004 mg N/L; TKN nitrogen 0.060 mg N/L; Total phosphorus 0.015 mg P/L
x = mean; a = standard deviation; 95% CI = 95% confidence interval
Categories with dissimilar letters were significantly different (Fisher's LSD pair wise comparison, a=0.05).












Table 3-3. Water and soil parameters among LDI groups. The Low LDI Group includes wetlands with an LDI < 2.0, and the High
LDI Group includes wetlands with an LDI > 2.0.

Low LDI Group High LDI Group
n x G 95% CI n x G 95% CI WA^
Water parameters
Dissolved oxygen (mg/L) 30 2.9 1.7 (2.2 to 3.5) 41 1.8 1.0 (1.5 to 2.1) 367 <0.01*
Temperature (C) 30 26.1 2.8 (25.1 to 27.2) 41 25.0 2.2 (24.4 to 25.7) 453 0.06
Color (PCU) 31 292 179 (226 to 358) 44 255 177 (202 to 309) 568 0.21
Turbidity (NTU) 31 3.8 4.1 (2.3 to 5.3) 44 13.2 28.1 (4.6 to 21.7) 907 0.02*
pH 31 5.2 1.1 (5.0 to 5.6) 44 6.3 0.9 (6.0 to 6.6) 1063 <0.01*
Specific conductivity (umhos/cm) 10 81 48 (46 to 115) 23 190 163 (119 to 260) 165 0.05
Ammonia-nitrogen (mg N/L)* 31 0.15 0.33 (0.03 to 0.27) 44 0.25 0.43 (0.12 to 0.38) 853 0.07
Nitrate/nitrite-nitrogen (mg N/L)* 26 0.09 0.37 (0.00 to 0.24) 36 0.01 0.03 (0.01 to 0.02) 450 0.77
TKN nitrogen (mg N/L)* 31 1.99 1.27 (1.53 to 2.46) 44 2.37 1.75 (1.84 to 2.90) 705 0.81
Total phosphorus (mg P/L)* 31 0.08 0.11 (0.04 to 0.12) 44 0.48 0.96 (0.19 to 0.77) 1057 <0.01*
Soil parameters
Moisture (%) 45 59 19 (54 to 65) 73 50 20 (46 to 55) 1163 <0.01*
Organic matter (%) 45 40 24 (33 to 48) 73 35 24 (29 to 41) 1357 0.11
TKN nitrogen (mg N/g soil) 45 6.88 3.66 (5.78 to 7.98) 73 6.00 4.15 (5.03 to 6.97) 1366 0.13
Total phosphorus (mg P/g soil) 45 0.39 0.27 (0.31 to 0.48) 73 0.74 0.97 (0.51 to 0.96) 2140 <0.01*
*Minimum Detection Limits: Ammonia-nitrogen 0.010 mg N/L; Nitrate/nitrite-nitrogen 0.004 mg N/L; TKN nitrogen 0.060 mg N/L; Total phosphorus 0.015 mg P/L
AW = Mann-Whitney U-test statistic.
p = significance value.









Diatom Assemblage

Statewide 50 wetlands were sampled with 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%). The three 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 ecoregion, 10 wetlands were sampled with four reference, four
agricultural, and two urban wetlands hosting 73 diatom taxa. In the north ecoregion 10
wetlands were sampled (four reference, two agricultural, and four urban) with 94 taxa
encountered. The central ecoregion included 13 sample wetlands (five reference, four
agricultural, and four urban) with 112 taxa sampled. The south ecoregion included 17
sample wetlands (six reference, five agricultural, and six urban) with 147 taxa identified.

Summary Statistics

Richness (R), evenness (E), Shannon diversity (H), and Simpson's index (S) were
calculated for each sample wetland (Appendix E). Table 3-4 summarizes the richness,
evenness, and diversity calculations of the diatom assemblage by a priori land use
category. Species richness ranged from nine taxa at CR6 (embedded within upland pine
forest) and CU1 (bordered on three sides by an arboretum and on the remaining side by a
paved road) to 39 taxa at CU3 (surrounded by commercial and residential land uses).
Species evenness ranged from 0.57 at NU5 (surrounded by residential and
transportation), to 0.89 at CA6 (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 (adjacent to a recently built public school
and transportation) at 0.93, and lowest at NU5 at 0.58. Richness, evenness, Shannon
diversity, and Simpson's index were not significantly different among the three a priori
land use categories (Table 3-4) or between LDI groups (Table 3-5). 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 the low LDI group (beta diversity of 7.5 and gamma diversity of 145).

Compositional Analysis

MRPP was calculated across all groups (panhandle versus north versus central
versus south) as well as for multiple pair wise comparisons (panhandle versus north,
panhandle versus central, panhandle versus south, north versus central, north versus
south, and central versus south) in order to test the similarity of diatom taxa composition
among the wetland regions. Table 3-6 shows the results for the MRPP tests, including
the test statistic (T), the chance-corrected within-group agreement (A, a measure of
within group similarity), and the significance value (p). The global comparison among
all wetlands and the three apriori categories showed that diatom community composition










Table 3-4. Diatom richness, evenness, and diversity among a priori land use categories.
Reference Agricultural Urban
Species richness (R) 19 (8)a 19 (6)a 22 (8)a
Species evenness (E) 0.74 (0.09)a 0.75 (0.08)a 0.73 (0.10)a
Shannon diversity (H) 2.13 (0.49)a 2.18 (0.34)a 2.21 (0.49)a
Simpson's index (S) 0.80 (0.09)a 0.82 (0.07)a 0.81 (0.10)a
Beta diversity 6.9 6.1 5.8
Gamma diversity 132 117 126
Values represent the mean + standard deviation.
Categories with dissimilar letters are significantly different (Fisher's LSD pair wise comparison, a = 0.05).


at the species level was significantly different (a = 0.05). Within the pair wise
comparisons, only the panhandle versus south and north versus south comparisons had
significantly different diatom community composition for all land use types. Reference
wetlands in the south wetland region had a significantly different diatom community
composition compared to wetlands in both the panhandle and north ecoregions.
Similarly, diatom community composition among pair wise comparisons of agricultural
wetlands was significantly different for the panhandle versus south ecoregions. Only the
north and south ecoregions had significantly different diatom community composition
among urban wetlands.

Community Composition

Figure 3-1 shows a two dimensional bi-plot of the NMS axes used to explore
diatom community composition with overlays of significant (transformed) environmental
variables, including LDI and six water parameters: DO (log), water turbidity (log), pH,
specific conductivity (log), TKN (log), and TP (log). Table 3-7 provides the Pearson r-


Table 3-5. Mean diatom summary statistics between LDI groups.
Low LDI High LDI WA p


Species richness (R) 19(8) 20 (7) 484 0.61
Species evenness (E) 0.73 (0.09) 0.74 9(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


Values represent the mean (standard deviation).
AW = Mann-Whitney U-test statistic. p = significance value.









Table 3-6. Similarity of diatom community composition using MRPP.
Sites (n) T^ A' p#
All wetlands
All regions (P vs N vs C vs S) 50 -2.2 0.01 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 -2.5 0.06 0.02*
North vs central 23 -0.7 0.02 0.20*
North vs south 27 -3.9 0.09 0.00*
Central vs south 30 -1.7 0.04 0.06*
Reference wetlands
All regions (P vs N vs C vs S) 18 -1.3 0.09 0.11*
Panhandle vs north 8 -0.7 0.07 0.21*
Panhandle vs central 8 0.8 -0.07 0.76*
Panhandle vs south 10 -2.1 0.16 0.03*
North vs central 8 0.9 -0.08 0.82*
North vs south 10 -2.4 0.17 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 -2.4 0.16 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 -0.3 0.02 0.38*
Urban wetlands
All regions (P vs N vs C vs S) 16 -0.7 0.05 0.23*
Panhandle vs north 6 -0.9 0.15 0.18*
Panhandle vs central 6 0.9 -0.07 0.83*
Panhandle vs south 8 -0.3 0.02 0.37*
North vs central 8 -0.1 0.01 0.37*
North vs south 10 -1.9 0.11 0.05*
Central vs south 10 1.1 -0.05 0.86*
*A high ITI value and significant p-value (p<0.05) suggests a difference in species
composition
^ T = the MRPP test statistic
'A = the chance corrected within-group agreement
#p = the significance value























log(DO) (c


0 A Priori Land Use
0 Reference
V Agricultural
Urban


o V



V

0 V
S t *V Axis 1: 12.4%


\ 'LDI

* log(Spec.Cond.)
V


Water pH


Figure 3-1. NMS ordination bi-plot of 50 wetlands in diatom species space with an
overlay of environmental parameters. LDI, dissolved oxygen-DO (log),
turbidity (log), water column pH, specific conductivity-Spec.Cond. (log),
water TKN (log), and water TP (log), 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.









Table 3-7. Pearson r-squared correlation coefficients between environmental 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
DO (log) 0.33 0.08 0.09
Temperature (log) 0.03 0.08 0.07
Color (log) 0.15 0.09 0.17
Turbidity (log) 0.22 0.01 0.00
pH 0.06 0.47 0.45
Spec.Cond. (log) 0.09 0.27 0.06
Ammonia-N (log) 0.13 0.00 0.03
Nitrate/nitrite-N (log) 0.01 0.01 0.05
TKN (log) 0.23 0.00 0.01
Water TP (log) 0.33 0.08 0.00
%Soil moisture (arcsine sqroot) 0.11 0.00 0.08
Soil TP (log) 0.07 0.05 0.02


squared correlation coefficients between the environmental variables and the NMS
ordination axes. A three dimensional solution was constructed with an overall stress of
16.3 and a final stability of 0.00001, which is borderline high but 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 was correlated with LDI,
DO (log), water TKN (log), and water TP (log). Axis 2 explained 35.5% variance and
was correlated with pH and specific conductivity (log). Axis 3 explained an additional
26.7% variance and was correlated with water column pH.

Metric Selection

The pool of potential candidate metrics was streamlined to reduce the redundancy
of selected metrics. Metrics selected for inclusion represented two of the metric
categories, including tolerance metrics and autecological metrics. In general, at least one
metric was selected from each of the autecological categories according to which metric
was most highly correlated with LDI. If two metrics were calculated from the same set
of taxa representing redundant metrics only one metric was included (i.e. metrics based
on the same autecological category were not included, for example if Dissolved Oxygen
Class 1 were included, no other metric based on Dissolved Oxygen Class would be
included). Seven metrics which were significant according to the Spearmans correlation
coefficient (Irl > 0.46, p < 0.01) (Table 3-8), showed visually distinguishable trends with









Table 3-8. Spearmans correlation coefficients for seven diatom metrics with LDI.
Diatom Metrics Spearmans 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
All correlations were significant (p < 0.01)


LDI, and were significant for the Mann-Whitney U-test between LDI groups (p < 0.10)
(Table 3-9) were selected for inclusion in the diatom FWCI. Metrics selected for the
FWCI were calculated as the percent of species richness using presence/absence data.
Tolerance metrics included Tolerant Indicator Species and Sensitive Indicator
Species. The five autecological 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 saturation between 10-25% and a biological oxygen
demand of approximately 13-22 mg 02/L), pH Class 3 (circumneutral, mainly occurring
at pH values around 7), and Dissolved Oxygen Class 1 (requiring continuously high
dissolved oxygen concentrations near 100%). Pollution Class was established by Bahls
(1993), and Nitrogen Uptake Metabolism, Saprobity, pH, and Dissolved Oxygen Classes
were defined by van Dam et al. (1994). Tolerant Indicator Species, Pollution Class 1,
Nitrogen Uptake Metabolism Class 3, Saprobity Class 4, and pH Class 3 increased with
increasing LDI. Sensitive Indicator Species and Dissolved Oxygen Class 1 decreased
with increasing LDI. Table 3-9 shows that all of the selected metrics significantly
differentiated between the high (LDI > 2.0) and low (LDI < 2.0) LDI groups (p < 0.01).


Table 3-9. Comparisons among diatom metrics and the diatom FWCI for LDI groups.
Metric Low LDI High LDI W p
Tolerant Indicator Species 1.4(2.5) 35.5 (15.8) 309.0 <0.001
Sensitive Indicator Species 36.6 (25.0) 18.0 (18.5) 719.0 <0.001
Pollution Class 1 3.7(6.0) 7.6(10.1) 337.0 <0.001
Nitrogen Class 3 7.7 (15.5) 20.3 (21.5) 361.0 0.003
Saprobity Class 4 2.7(4.5) 25.4(25.6) 373.5 0.006
pH Class 3 18.3 (22.2) 14.1 (16.6) 344.0 0.001
Dissolved Oxygen Class 1 69.3 (24.3) 39.9 (23.8) 658.0 0.004
Diatom FWCI 55.3 (11.1) 45.4(27.1) 718.0 <0.001
Values represent the mean (standard deviation).
W' = the Mann-Whitney U-Test statistic
p' = the significance value.









Tolerance metrics

The lists of Tolerant and Sensitive Indicator Species were determined using
Indicator Species Analysis (ISA) using species-level abundance data for the diatom
assemblage. Tolerant diatom indicator species were established at an LDI break of 4.75,
and included 17 species representing nine genera. Table 3-10 lists the diatom Tolerant
Indicator Species. The three tolerant indicator species with 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 Indicator Species increased with increasing LDI.
The two wetlands with the highest percent Tolerant Indicator Species in the high LDI
group included SU1 (a wetland surrounded by residential land use and a golf course,
which was previously embedded in pasture) and CU5 (surrounded by transportation and
stormwater retention basins). Four agricultural and urban wetlands and 10 reference
wetlands had zero percent tolerant diatom indicator species.
Diatom Sensitive Indicator Species were selected at an LDI break of 1.25,
correlating to a break in the natural and developed land uses. Table 3-11 lists the 18
Sensitive Indicator Species. The five 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 with increasing development intensity in
the surrounding landscape. Six agricultural and 2 urban wetlands hosted no sensitive
indicator species.


Table 3-10. Diatom Tolerant Indicator Species.
Tolerant Indicator Species Indicator Value
Cyclotella pseudostelliger 20.0
Diploneis elliptica 19.7
Navicula confervacea 45.3
Navicula minima 48.0
Navicula mutica 40.5
Navicula recens 12.9
Navicula subminuscula 12.5
Neidium alpinum 20.0
Nitzschia subacicularis 18.1
Pinnularia braunii 22.6
Pinnularia divergentissima 13.3
Stauroneis kriegeri 13.3
All reported indicator values were significant (p < 0.10).











70 -

60 -

S50

40

30

20

10

0 **
0 2 4 6 8
LDI

Figure 3-2. Percent diatom Tolerant Indicator Species increased with increasing
development intensity (LDI).


Table 3-11. Diatom Sensitive Indicator Species.
Indicator Species Indicator Value
Anomoeoneis brachysira 39.2
Cymbella microcephala 12.5
Desmogonium rabenhorstianum 33.5
Encyonema silesiacum 24.4
Eunotiaflexuosa 12.5
Eunotia glacialis 17.0
Eunotia intermedia 28.1
Eunotia naegelii 59.2
Eunotiapectinalis undulate 26.9
Eunotia rhomboidea 45.9
Frustulia rhomboids 41.6
Frustulia rhomboides crassinervia 18.7
Navicula capitatoradiata 11.5
Navicula subtilissima 12.5
Nitzschia nana 17.0
Nitzschia paleacea 18.7
Pinnularia streptoraphe 18.7
Rhopalodia gibba 18.7
All reported indicator values were significant (p<0.10).









90

80 -
S70 -

S60 -

50

S40 -

3 30

20

0 ** *

0 2 4 6 8
LDI

Figure 3-3. Percent diatom sensitive indicator species decreased with increasing
development intensity (LDI).


Autecological metrics

Between 56-69% of diatoms identified received scores based on established
autecological relationships (Bahls 1993; van Dam et al. 1994). Five metrics based on
scoring diatoms from a coded checklist describing their autecology were incorporated in
the diatom FWCI, including the proportion of diatoms in Pollution Class 1, Nitrogen
Uptake Metabolism Class 3, Saprobity Class 4, pH Class 3, and Dissolved Oxygen
Requirement Class 1. Figure 3-4 shows that the proportion of diatoms in Pollution Class
1 increased with increasing development intensity in the surrounding landscape. Diatoms
in Pollution Class 1 were very tolerant to pollution, as compared to Pollution Class 2
(moderately tolerant) or Pollution Class 3 (sensitive to pollution) (Bahls 1993). In
general a background level of approximately 10% of the diatoms belonging to Pollution
Class 1 distinguishes low (LDI < 2.0) and high (LDI > 2.0) LDI wetlands. 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 dispersed from a pullet farm operation.
Figure 3-5 shows that the proportion of diatoms in Nitrogen Uptake Metabolism
Class 3 increased with increasing development intensity in the surrounding landscape.
Membership in Nitrogen Uptake Metabolism Class 3 was defined by facultative nitrogen-
heterotrophic taxa, those needing periodically elevated concentrations of organically
bound nitrogen (van Dam et al. 1994). Four wetlands in the low LDI (LDI < 2.0) group
that had higher than anticipated percent diatoms in Nitrogen Uptake Metabolism Class 3
included SR4, SR5, SR6, and SU3. The three southern reference wetlands (SR4, SR5,
and SR6) were located within state or federal lands protected as part of the Florida
Everglades. The urban outlier, SU3 was surrounded by nearly 100 m of marsh that has
received nutrient enriched waters since the mid 1970s.



























*


*


Pollution Tolerance Class 1 diatoms increased with increasing development
intensity (LDI).


he


r~* ~


Figure 3-5.


Nitrogen Uptake Metabolism
development intensity (LDI).


Class 3 diatoms increased with increasing


Figure 3-4.


ren
m
w
U,

0
I40

o









The percent of diatoms in Saprobity Class 4 increased with increasing landscape
development intensity (Figure 3-6). Diatoms characterized as belonging to Saprobity
Class 4 included meso- to poly-saprobous species (that inhabit aquatic environments with
an oxygen saturation between 10-25% and a biological oxygen demand (BOD520) of 13-
22 mg/L) (van Dam et al. 1994). Eighty-five percent of the wetlands in the low LDI
group had less than 5% of diatoms in Saprobity Class 4. CA3 (a wetland receiving
waters dispersed from a pullet farm operation) hosted the greatest percent of diatoms in
Saprobity Class 4 (saprobity class 4 = 52%).
Figure 3-7 shows that the percent of diatoms in pH Class 3 increased with
increasing LDI. Diatoms categorized in pH Class 3 were described as circumneutral
(mainly occurring at pH values of approximately 7) (van Dam et al. 1994). In the low
LDI group, the wetland with the highest percent diatoms in pH Class 3 was SU4, a
wetland surrounded by low intensity residential, previously in pasture (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 landscape (Figure 3-8). Six wetlands with low
LDI scores (LDI < 2.0) hosted a smaller percent of diatoms requiring high dissolved
oxygen than anticipated. These were the same six wetlands that hosted a higher percent
pH Class 3 diatoms (Figure 3-7).

Diatom Wetland Condition Index

The seven metrics described above were scored and added together to create the
diatom FWCI. Appendix F provides detailed instructions on metric scoring. Figure 3-9
shows the relationship between the diatom FWCI and LDI. Potential scores for the
diatom FWCI 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
received diatom FWCI scores 2 points lower than SR2, with a score of 67 at both NR3
and SR1.
Diatom FWCI scores varied regionally, with the highest scores in each wetland
region including PR4 (65), NR3 (67), CR6 (65), and SR2 (69), in the panhandle, north,
central, and south wetland regions, respectively. The lowest scores in the panhandle and
north wetland regions were for urban wetlands embedded in residential land use,
including PU4 (11) and NU2 (24). Wetlands surrounded by agricultural land uses
received the lowest scores in the central and south wetland regions, including CA3 (8)
and SA4 (16). The diatom FWCI was robustly correlated with the LDI index (Spearman
correlation Irl = 0.64, p < 0.001). A Kruskal-Wallis test between median diatom FWCI
score values suggested a significant difference (H = 20.7, p < 0.001) among wetlands in
the three apriori land use categories.










60





3 40
0




S20


V
*
*


Figure 3-6. Saprobity
(LDI).


Class 4 diatoms increased with increasing development intensity


*


Figure 3-7.


The pH Class 3 diatoms
(LDI).


increased with increasing development intensity


**











100



80 -


60



40


20


Figure 3-8. Dissolved Oxygen
intensity (LDI).





70 oo

60 o

0


0
0 0
0


S40

S30
,5


Class 1 diatoms decreased with increasing development










IA
A
A *



A*
A


A


o Reference A Agricultural Urban


Diatom FWCI scores decreased with increasing development intensity (LDI).
Sample wetlands are designated by a priori land use category: reference,
agricultural, or urban.


* *


Figure 3-9.


* *









Cluster Analysis

Cluster analysis determined four categories based on diatom community
composition. Using site descriptions, clusters were explained by wetland regions and a
priori land use categories including: 1: wetlands in the panhandle to central wetland
regions surrounded by low development land use activities; 2: wetlands occurring in
mixed wetland regions surrounded by low development land use activities; 3: wetlands
within the southern Everglades; and 4: wetlands within mixed wetland regions
surrounded by high development land use activities. Figure 3-10 shows that based on the
diatom FWCI scores, clusters 1 and 2 were not significantly different from one another,
but were significantly different from both cluster 3 and cluster 4 (p<0.05). Clusters 3 and
4 were significantly different from each other. Table 3-12 provides means and standard
deviations for diatom FWCI scores and LDI of the four diatom based clusters. Cluster 4
had significantly different diatom FWCI and LDI scores than all other clusters.


70

60

50 -

40 -

30 -

20_

10_

0_


a


b


C


Cluer 1
Cluster 1


Cluster 2
Cluster 2


Cluster 3
Cluster 3


Cluster 4
Cluster 4


Figure 3-10. Diatom FWCI scores for wetland clusters based on diatom community
composition. Boxes represent the interquartile range; solid circles represent
the mean; middle lines represent the median; whiskers represent the range;
asterisks represent outliers (> +2 standard deviations). Clusters with similar
letters were not significantly different (Fisher's LSD, p<0.05).


I I






56


Table 3-12. Mean diatom FWCI and LDI values for wetland clusters based on diatom
community composition.
Cluster*
1 2 3 4
Diatom FWCI 57.6 (6.4)a 52.9 (11.6)a 38.1 (4.4)b 24.1 (11.7)c

LDI 2.9 (2.1)a 2.8 (2.0)a 2.4 (1.9)a 5.3 (1.4)b
Clusters with similar letters were not significantly different (Fisher's LSD, p<0.05).
Values represent mean (standard deviation).









Macrophyte Assemblage

Statewide, 118 wetlands were sampled with 605 species, representing 323 genera
and 126 families identified. The most abundant species was Taxodium ascendens 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. 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 the vegetation quadrats at only one wetland. In the panhandle
wetland region, 28 wetlands were sampled hosting 328 species, representing 191 genera
and 90 families. In the north wetland region 31 wetlands were sampled with 306 species
(180 genera and 89 families) encountered. The central wetland region included 31
wetlands with 329 species (202 genera and 94 families) sampled. The south wetland
region had 28 sample wetlands with 266 species (in 180 genera and 89 families)
identified.

Summary Statistics

Species richness (R), first and second order jackknife estimators (Jackl and
Jack2, respectively), species evenness (E), Shannon diversity (H), and Whittaker's beta
diversity (PW) were calculated based on the macrophyte assemblage for each sample
wetland (Appendix E). 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 were 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
(surrounded by pine flatwoods), with 11 and 15 species estimated with first 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 wetland on a private
conservation tract), to 0.93 at PU4 (surrounded by residential land use including housing
and a community park). Shannon diversity ranged from 1.8 to 3.9 at two agricultural
wetlands, NA3 (surrounded by silviculture) and NA12 (surrounded by pasture and row
crops), respectively, similar to species richness. Whittaker's beta diversity ranged from a
low of 0.2 at PU9 (surrounded by residential land use), to a high of 9.4 at CR5 (located in
a fragmented state park within highly developed Seminole County).
Table 3-13 summarizes comparisons of mean richness and diversity calculations
by a priori land use category. Agricultural wetlands had the greatest species richness
followed closely by urban wetlands. This same trend was evident for species evenness,
with agricultural wetlands having greater species evenness. Diversity indices yielded
similar results, with reference wetlands having lower Shannon diversity and Whittaker's
beta diversity then both agricultural and urban wetlands. Beta and gamma diversity were
calculated for a priori land use categories, with urban wetlands having the highest beta
diversity and agricultural wetlands the highest gamma diversity. Only Whitaker's beta









Table 3-13. Mean macrophyte richness, evenness, and diversity among apriori land use
categories.
Reference Agricultural Urban
Species richness (R) 32 (11)a 37 (14)a 36 (10)a
Species evenness (E) 0.85 (0.04)a 0.87 (0.04)a 0.86 (0.03)a
Shannon diversity (H) 2.9 (0.3)a 3.1 (0.4)a 3.1 (0.3)a
Whittaker's Beta diversity (PW) 4.0 (2.9)a 4.8 (1.7)b 4.3 (1.7)ab
Beta diversity 9.5 10.4 10.5
Gamma diversity 304 383 378
Categories with similar letters were not significantly different (Fisher's LSD, a=0.05).
Values represent mean (standard deviation).


diversity was significantly different among a priori land use categories (Fisher's LSD
pair wise comparison, a = 0.05). Species richness and Whittaker'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 and Shannon diversity were significantly
different (p < 0.05) between LDI groups (Mann-Whitney U-Test).

Compositional Analysis

MRPP was calculated across all wetland regions (panhandle versus north versus
central versus south) as well as for multiple pair wise comparisons (panhandle versus
north, panhandle versus central, panhandle versus south, north versus central, north
versus south, and central versus south). Table 3-15 shows the results for the MRPP tests,
including the test statistic (T), chance-corrected within-group agreement (A), and
significance value (p). Only two of the MRPP comparisons (agricultural wetlands in the
panhandle and north, and central and south wetland regions) were not significant at the


Table 3-14. Mean macrophyte richness, evenness, 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 (PW) 4.1(1.9) 4.5 (1.7) 2131 0.08
Beta diversity 10.2 13.8
Gamma diversity 338 510
Indicates significance at a<0.05
W = Mann-Whitney U-test statistic
p' = significance value
Values represent mean (standard deviation).









Table 3-15. Macrophyte community composition similarity among wetland regions with
MRPP.
Sites (n) T^ A' p#
All wetlands
All regions (P vs N vs C vs S) 118 -26.8 0.06 0.00*
Panhandle vs north 59 -7.1 0.03 0.00*
Panhandle vs central 59 -13.1 0.04 0.00*
Panhandle vs south 56 -23.6 0.09 0.00*
North vs central 62 -7.9 0.03 0.00*
North vs south 59 -24.2 0.09 0.00*
Central vs south 59 -11.8 0.04 0.00*
Reference wetlands
All regions (P vs N vs C vs S) 37 -12.5 0.12 0.00*
Panhandle vs north 17 -5.2 0.08 0.00*
Panhandle vs central 19 -6.8 0.09 0.00*
Panhandle vs south 17 -8.1 0.14 0.00*
North vs central 20 -4.7 0.06 0.00*
North vs south 18 -8.5 0.14 0.00*
Central vs south 20 -7.1 0.07 0.00*
Agricultural wetlands
All regions (P vs N vs C vs S) 40 -6.6 0.05 0.00*
Panhandle vs north 22 -0.7 0.01 0.21
Panhandle vs central 19 -2.8 0.03 0.01*
Panhandle vs south 19 -8.1 0.10 0.00*
North vs central 21 -2.0 0.02 0.04*
North vs south 21 -7.7 0.09 0.00*
Central vs south 18 -1.4 0.02 0.09
Urban wetlands
All regions (P vs N vs C vs S) 41 -15.3 0.14 0.00*
Panhandle vs north 20 -6.0 0.07 0.00*
Panhandle vs central 21 -9.0 0.10 0.00*
Panhandle vs south 20 -11.1 0.17 0.00*
North vs central 21 -3.3 0.03 0.00*
North vs south 20 -10.5 0.15 0.00*
Central vs south 21 -8.6 0.09 0.00*
*A high ITI value and significant p-value (p<0.05) suggests a difference in species
composition.
T = the MRPP test statistic
A' = the chance corrected within-group agreement
p# = the significance value.









a = 0.05 level, suggesting that there were regionally significant differences among
species composition across wetland regions and within apriori land use categories.

Community Composition

Macrophyte community composition was summarized in two separate NMDS
ordinations to relate changes in macrophyte community composition with environmental
variables. Figure 3-11 shows a two dimensional bi-plot of the NMDS axes used to
explore the dissimilarities of macrophyte community composition with overlays of
significant environmental variables, including soil TP (log), LDI, latitude, and longitude.
Table 3-16 provides the Pearson correlation coefficients between environmental variables
and NMDS ordination axes. A three dimensional solution was constructed with an
overall stress of 20.8 and a final stability of 0.06, which is a fairly high stress value but
considered useful for ordinations with community data sets (Kruskal 1964; Clarke 1993;
McCune and Grace 2002). Axis 1 explained 29.1% variance and was correlated with
latitude and longitude; axis 2 explained 34.2% additional variance and was correlated
with LDI and soil TP (log). Axis 3 explained an additional 12.1% of the variance and
was not significantly correlated with any of the measured soil parameters, LDI, latitude,
or longitude.
A second NMS ordination was completed using the macrophyte species
composition at the 75 sample wetlands with measured water chemical and physical
parameters. Figure 3-12 shows the bi-plot from the NMS ordination. The final stress
was 16.6 with a final instability of 0.004. The ordination 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 environmental parameters and the
three ordination axes. Axis 1 explained 31.3% of the variance and was correlated with
latitude and longitude. Axis 2 was correlated with LDI, water pH, water TP (log), soil TP
(log), and DO (log), and explained 25.7% additional variance. Axis 3 explained an
additional 20.4% of the variance and was correlated with water pH and soil moisture
(arcsine square root). Water ammonia-nitrogen concentration (log), water nitrate/nitrite-
nitrogen concentration (log), water TKN-nitrogen concentration (log), water temperature
(log), water color (log), and water turbidity (log) were not strongly correlated with the
NMDS ordination axes.











O



A



0
OA


0 AG 0
0
A 0
0 0 0 0


AI Latitude


A A
A AO
A A AA
hPhosp orusa A Axis 1 :


0 0 0 0L A A AA
0 A9
O A A*

*
0 0 A



AO O
OA
S0 Refence
*r A Agricultural
O Urban



Figure 3-11. NMDS ordination bi-plot of 118 sample wetlands in macrophyte species
space with an overlay of environmental parameters. Latitude, longitude,
LDI, and log(soil TP), shown as radiating vectors, were significantly
correlated with the 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% additional
variance, and axis 3 (not shown) represented an additional 12.1% variance.









Table 3-16. Pearson correlation coefficients between environmental variables and
NMDS 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
Arcsine 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


Metric Selection

The pool of potential candidate metrics was streamlined to reduce the redundancy
of selected metrics. If two metrics were calculated from the same set of taxa representing
redundant metrics only one metric was included (i.e. metrics based on the same category
were not included, for example if percent Obligate Species was included, no other metric
based on wetland indicator status). Six metrics that were significant for the Spearmans
correlation coefficient (I|r>0.50, p<0.001) (Table 3-18), showed visually distinguishable
trends with LDI, and were significant for the Mann-Whitney U-test between LDI groups
(p<0.001) (Table 3-19) were selected for inclusion in the macrophyte FWCI. Metrics
selected for the FWCI were calculated as the percent of species richness using
presence/absence data.
Metrics selected for inclusion were percent Tolerant Indicator Species, percent
Sensitive Indicator Species; Floristic Quality Assessment Index (FQAI); percent Exotic
Species; percent Native Perennial Species; and percent Wetland Status Species (including
both obligate and facultative wetland species). The percent of Tolerant Indicator Species
and percent Exotic Species increased with increasing development intensity. Whereas,
percent Sensitive Indicator Species, FQAI, percent Native Perennial Species, and percent
Wetland Status Species values decreased with increased landscape development. Table
3-19 shows that all six metrics significantly differentiated between low (LDI < 2.0) and
high (LDI > 2.0) LDI groups.

Tolerance metrics

Macrophyte Tolerant and Sensitive Indicator Species lists were determined using
Indicator Species Analysis (ISA) for each of the four wetland regions and statewide using
species-level abundance data. Macrophyte Tolerant Indicator Species were established at
an LDI break of 4.0. Table 3-20 provides a list of Tolerant Indicator Species comparing
regional and statewide analyses. The same random seed number was used for each ISA
for statewide and regional analyses.





















A' 0
0 0 0
0 0
Log(DO)_


0 OOA


Latitude
I A


A A


Log(TP)

LDI


* 0


Axis 2: 25.7%


Water pH


O Reference
A Agricultural
* Urban


Figure 3-12. NMDS ordination bi-plot of 75 sample wetlands in macrophyte species
space with an overlay of environmental parameters. Latitude, longitude,
LDI, TP (log), water pH, and DO (log), shown as radiating vectors, were
significantly correlated with the NMDS axes. Vector length represents the
strength of the correlation, and the angle represents the direction of
maximum change. Axis 1 explained 31.3% variance, axis 2 explained
25.7% additional variance, and axis 3 (not shown) represented an additional
20.4% variance.









Table 3-17. Pearson correlations between environmental variables and NMDS 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
DO (log) 0.02 0.20 0.05
Temperature (log) 0.02 0.05 0.00
Color (log) 0.14 0.00 0.05
Turbidity (log) 0.03 0.06 0.03
pH 0.15 0.26 0.35
Ammonia-N (log) 0.01 0.04 0.00
Nitrate/nitrite-N (log) 0.06 0.00 0.06
TKN (log) 0.02 0.01 0.01
TP (log) 0.11 0.25 0.06
Soil parameters
Moisture (arcsine sqrt) 0.00 0.00 0.24
TKN (log) 0.00 0.00 0.16
TP (log) 0.09 0.23 0.01







Table 3-18. Spearman correlations between six statewide macrophyte metrics with LDI.
Metric Spearmans r
Percent Tolerant Indicator Species 0.75
Percent Sensitive Indicator Species -0.66
FQAI -0.71
Percent Exotic Species 0.65
Percent Native Perennial Species -0.63
Percent Wetland Status Species -0.55
All correlations were significant (p < 0.01).









Table 3-19. Comparisons among six statewide macrophyte metrics between LDI groups.
Metric Low LDI High LDI W p
Percent Tolerant Indicator Species 7.8 (7.8) 31.2 (14.7) 1116.5 <0.001
Percent Sensitive Indicator Species 39.5 (16.7) 9.4 (10.1) 3665.0 <0.001
FQAI 4.81(0.62) 3.62(0.80) 3771.0 <0.001
Percent Exotic Species 3.0(3.6) 14.3(10.6) 1379.0 <0.001
Percent Native Perennial Species 92.7 (4.5) 79.7 (12.0) 3453.0 <0.001
Percent Wetland Status Species 72.0 (9.8) 54.1 (12.5) 3612.0 <0.001
Values represent the mean (standard deviation)
W' = the Mann-Whitney U-Test statistic
p' = the significance value


In total the ISA reported 69 statewide Tolerant Indicator Species, and less for
each wetland region with 7, 28, 7, and 12 for the panhandle, north, central, and south
wetland regions, respectively. The statewide ISA produced 69 Tolerant Indicator Species
with an additional seven species included on regional lists, but not the statewide list. No
species occurred on the Tolerant Indicator Species lists statewide and in all four wetland
regions. Three species occurred on the statewide Tolerant Indicator Species list and on
three of the wetland region lists, including Commelina diffusa (north, central, and south),
Cynodon dactylon (panhandle, north, central), and Diodia virginiana (panhandle, north,
south). Seven species occurred on the Tolerant Indicator Species list statewide and in
two wetland regions, and 24 were listed both statewide and in one wetland region.
Thirty-five of the 69 statewide Tolerant Indicator Species (51%) were not listed in any
wetland region. In total, the wetland regions shared more than two- thirds of their listed
species with the statewide list: panhandle (100%), north (93%), central (86%), and south
(67%). Two species were unique to the north, one to the central, and four to the south
wetland region Tolerant Indicator Species lists.
Figure 3-13 shows the scatterplots of the percent Tolerant Indicator Species
versus LDI. The percent Tolerant Indicator Species increased with increasing
development intensity. For the statewide Tolerant Indicator Species, CA2 (surrounded
by pasture; tolerant = 72%) had the highest percent statewide Tolerant Indicator Species.
Three wetlands in the low LDI group with high percent Tolerant Indicator Species
included SA8 (a suburban park receiving run-off from a pasture; 32%), CA8 (in a
preserve but receiving run-off from dairy activities; 26%), and PR7 (located on a private
conservation tract; 25%). In the regional ISA calculations, a wetland in the north wetland
region, NA1 (surrounded by pasture; north tolerant = 47%) had the largest percent
Tolerant Indicator Species.
Macrophyte Sensitive Indicator Species were established at an LDI break of 2.0.
Table 3-21 provides a list of statewide and regional Sensitive Indicator Species. The
statewide Sensitive Indicator Species list included 61 species of which 16 were not listed
in any of the wetland regions. Two species occurred on all statewide and wetland region
lists, including Eriocaulon decangulare and Panicum erectifolium. Similarly, six species
occurred on the statewide list and three wetland region lists including Andropogon











Table 3-20. Statewide an


d regional macrophyte Tolerant Indicator Species.
Statewide Panhandle North Central South
69 7 28 7 12
28.5 53.7
11.3
10.9 21.4
18.6
9.5
9.1


No. of Tolerant Species
Acer rubrum
Alternanthera philoxeroides
Amaranthus spinosus
Ampelopsis arborea
Aster carolinianus
Axonopus fissifolius
Blechnum serrulatum
Boehmeria cylindrica
Carex longii
Centella asiatica
Colocasia esculenta
Commelina diffusa
Cuphea co d l,, i ..i,. i,,,
Cynodon dactylon
Cyperus croceus
Cyperus lanceolatus
Cyperus polystachyos
Cyperus retrorsus
Cyperus virens
Digitaria ciliaris
Diodia virginiana
Dioscorea bulbifera
Echinochloa colona
Eclipta prostrata
Eupatorium capillifolium
Galium hispidulum
Galium tinctorium
Hymenachne amplexicaulis
Hypericum mutilum
Juncus ettIl.\'i.\
Kyllinga brevifolia
Leersia hexandra
Lepidium virginicum
Ligustrum sinense
Lonicerajaponica
Ludwigia peruviana
Ludwigia repens
Luziola fluitans
Lygodiumjaponicum


35.7
42.9


40.5


51.8


28.6
42.9


52.0


44.1
50.0


28.1




20.0


50.0


26.7


25.0


20.0


21.4


43.7
54.3



50.0
28.6
35.7











Table 3-20. Continued.
Statewide Panhandle North Central South


36.2


52.0


Melaleuca quinquenervia
Melothria pendula 20.8
Micranthemum umbrosum 5.5
Momordica charantia 10.2
Oxalis corniculata 18.5
Parthenocissus quinquefolia 34.6
Paspalum notatum 24.4
Paspalum urvillei 18.5 35.7
Phyla nodiflora 15.2
Phyllanthus urinaria 16.7
Phytolacca americana 26.1 36.7
Polygonum hydropiperoides 21.1 57.1
Polygonum punctatum 28.0 42.9
Polypremum procumbens 9.5
Proserpinaca palustris 5.5
Richardia brasiliensis 5.5
Rubus argutus 25.1
Rubus trivialis 17.7
Sabal palmetto
Sacciolepis indica 7.3
Sambucus canadensis 24.6
Sapium sebiferum 19.4
Saururus cernuus
Senna obtusifolia 9.1
Sesbania vesicaria 5.5
Setaria parviflora 7.3
Sida rhombifolia 20.8
Smilaxpumila 7.3
Solanum carolinense 9.1
Solidago stricta 5.5
Sporobolus indicus 7.3
Stenotaphrum secundatum 9.5
Toxicodendron radicans 38.0
Trifolium repens 7.3
Urena lobata
Vitis rotundifolia 36.7
Wedelia trilobata 5.5
Values represent Indicator Values.
All reported Indicator Values were significant (p < 0.10).


26.7


52.0
48.5
25.0


66.7





























0 2 4


40

o 30

20

10

0


50

S40

S 30

t 20

o 10


* U

* U
a.C


6 8


0


gog~O


A
a b

A


0 2 4
LDI


40

-' 30

20

00 10


x x
) I


6 8


x X
X


0 2 4
LDI


60

-







0


6 8


X 0
* 0 x

AX)K a A^
8 A
AAA x


0 2 4
LDI


A I A

00 A *

0 l
Ax0 oA


U Panhandle A North 0 Central X South



Figure 3-13. Macrophyte Tolerant Indicator Species increased with increasing
development intensity (LDI). A) panhandle Tolerant Indicator Species at
panhandle study wetlands. B) north Tolerant 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 118 study wetlands.


6 8











Table 3-21. Statewide and regional macrophyte Sensitive Indicator Species.
Statewide Panhandle North Central South
No. of Sensitive Species 61 28 19 24 13
Amphicarpum 19.3 52.4
muhlenbergianum
Andropogon virginicus 46.9 62.5 66.7 66.5
Aristida beyrichiana 8.6 45.5
Aristida patula 22.2
Aristida purpurascens 32.9 50.0 22.2 53.5
Carex verrucosa 25.0
Cladium jamaicense 22.3 60.6
Coelorachis rugosa 7.3
Cyperus haspan 11.0 33.3
Drosera brevifolia 7.3
Erianthus giganteus 17.8 33.3
Eriocaulon compressum 17.1 25.0 33.3
Eriocaulon decangulare 37.8 37.5 22.2 53.5 33.3
Eupatorium leptophyllum 14.6 25.0
Eupatorium mohrii 8.6
Fuirena scirpoidea 26.9 25.0 25.0 36.2
Gaylussaciafrondosa 25.0
Gratiola ramose 18.3 41.7
Hypericum chapmanii 8.6 45.5
Hypericumfasciculatum 38.2 53.5 44.4
Hypericum myrtifolium 17.8 47.7
Hyptis alata 10.8
Ilex glabra 46.2 53.9 78.6 45.9
Ilex myrtifolia 17.1 71.2
Ipomoea sagittata 11.0
Lachnanthes caroliniana 39.6 71.2 40.2
Lachnocaulon anceps 25.0
Lobelia floridana 25.0
Lophiola aurea 12.2 62.5
Ludwigia linifolia 7.3
Lycopodiella alopecuroides 9.8 25.0 22.2
Lyonia lucida 24.4
Nymphaea odorata 6.2 22.2
Nymphoides aquatica 12.2
Panicum ensifolium 13.1
Panicum erectifolium 41.5 50.0 33.3 45.2 36.2
Panicum hemitomon 40.4 79.2
Panicum rigidulum 17.1 25.0
Panicum tenerum 14.6 33.3







70


Table 3-21. Continued.


Statewide


Panhandle North Central South


Pinus elliottii 33.9 56.2 37.0 25.0
Pinus palustris 8.6
Plucheafoetida 10.1 25.0
Pluchea rosea 15.5 33.3
Polygala cymosa 28.0 37.5 29.3 33.3
Polygala lutea 7.3
Proserpinaca pectinata 12.4
Rhexia alifanus 13.4 70.3
Rhexia lutea 13.4 45.5 22.2
Rhexia mariana 23.1 47.7 45.2
Rhexia petiolata 25.0
Rhus copallinum 33.1
Rhynchospora corniculata 25.0
Rhynchosporafilifolia 37.5
Rhynchospora inundata 17.2 22.2
Rhynchospora microcarpa 28.8
Rhynchospora -, il, 22.2
Sabatia bartramii 7.3
Sagittaria graminea 13.9 41.7
Sagittaria lancifolia 17.2 36.2
Salix caroliniana 44.4
Sarracenia minor 22.2
Scleria baldwinii 7.3
Scleria Georgiana 7.3
Scleria triglomerata 7.3 25.0
Serenoa repens 22.3 37.5 57.6
Spartina bakeri 7.3 25.0
'\, iiMl,,i, aquatica 13.4 36.2
l i.. .,i:,i,,iii ,flavidulus 12.2 22.2
Utricularia purpurea 8.6 25.0
Vaccinium corymbosum 20.9 33.1 33.3
Xyris ambigua 11.0
Xyris caroliniana 7.3 25.0
Xyris elliottii 18.3 50.0
Xyrisjupicai 8.6 25.0
Values represent Indicator Values.
All reported Indicator Values were significant (p < 0.10).










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 four wetland regions shared over three-quarters of
their species with the statewide list (panhandle = 79%, north = 84%, central = 92%, and
south = 85%). Six species were unique to the panhandle Sensitive Indicator Species
analysis, three to the north, two to the central, and two to the south wetland region.
Figure 3-14 shows that the percent Sensitive Indicator Species, statewide and regionally,
decreased with increasing development intensity.
All of the indicator species metrics were significantly correlated with the LDI
index. Table 3-22 shows Spearman correlations calculated with both statewide and
regional Indicator Species lists for each region. Regional Indicator Species metrics had a
stronger correlation value, though all metrics were significantly correlated with LDI (p <
0.01). Two exceptions included the panhandle Tolerant Indicator Species correlations
(statewide Tolerant Indicator Species, r = 0.73, panhandle Tolerant Indicator Species, r =
0.72) and central (statewide Tolerant Indicator Species, r = 0.74, central Tolerant
Indicator Species, r = 0.68).
Shrub and tree species were included in the ISA for both Tolerant and Sensitive
Indicator Species 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 isolated forested wetlands.
Excluding the tree and shrub layers would underscore their importance in providing
structure for the isolated depressional forested wetlands. However, trees comprised only
a small percentage of the Tolerant and Sensitive Indicator Species lists (Tables 3-20 and
3-21). Three percent of the statewide Tolerant Indicator Species were trees, 9% were
shrubs, 14% vines, and 74% herbaceous (including herbs, sedges, grasses, etc.). The two
statewide tree Tolerant Indicator Species included the hardwood Acer rubrum and exotic
Sapium sebiferum (Table 3-20). The six statewide shrub Tolerant Indicator Species were
of the genera Aster (a climbing species), an exotic Ligustrum, an exotic Ludwigia, Rubus,
and Sambucus. Of the statewide Sensitive Indicator Species, 3% were trees, 15% shrubs,
2% vines, and 80% herbaceous.

Floristic Quality Assessment Index metric

Figure 3-15 shows wetland FQAI scores decreased with increasing landscape
development intensity. Wetlands in the panhandle (maximum modified FQI = 6.25) and
north (maximum FQAI = 5.95) wetland regions had higher FQAI scores versus wetlands
in the central (maximum FQAI = 4.93) and south (maximum FQAI = 5.24) wetland
regions. Statewide the FQAI was significantly correlated with LDI (I|r = 0.71, p < 0.001;
Table 3-18); and there was a significant difference between the mean FQAI scores
between low (LDI < 2.0) and high (LDI > 2.0) LDI groups (W = 3771.0, p = <0.001;
Table 3-19). CU6, an urban wetland in the high LDI group in the central wetland region
surrounded by recently developed and under construction sections of residential land use
(FQAI = 4.43; LDI = 7.2), had a higher FQAI score than anticipated. The wetland with
the lowest FQAI score was NA1, an agricultural wetland surrounded by pasture (FQAI=
2.06).













70
. 60
E 50
40
S30-
- 20
10
0


70
60
S50
S40
S30
20
10


% 'L *_- -


0 2 4
LDI


6 8


A A


A A



0 2 4 6
LDI


0
0
ta'daAo_ 2,


C. 70
60
S50
S40
S30
20
S 10
0


0 2 4
LDI


0 A



oA
A


SX
" X


0 2 4
LDI


A X U
0A 0 0
II, %4A


LDI

* Panhandle a North 0 Central X South


Figure 3-14. Macrophyte Sensitive Indicator Species decreased with increasing
development intensity (LDI). A) panhandle Sensitive Indicator Species at
panhandle study wetlands. B) north Sensitive 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 118 study wetlands.


70
S60
I 50
-
S40
30
0 20
o o
10
0


0
0 o


. X









Table 3-22. The percent of statewide and regional macrophyte Indicator Species metrics
were significantly correlated with LDI (p < 0.01).
%Statewide ISA %Regional ISA


Wetland Region
Statewide


n= Spearmans r Spearmans r


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
Values represent Spearmans correlation coefficients, all reported correlations significant (P < 0.01).


Exotic Species metric

Statewide, the percent Exotic Species was significantly correlated with
development intensity in the surrounding landscape (r = 0.65, p < 0.001; Table 3-18).
Figure 3-16 shows that the percent of exotic species increased with increasing LDI in
each wetland region. The north wetland region hosted the wetland with the greatest
percent Exotic Species, NA1 with 52.6% Exotic Species. NA1 was surrounded by a
research facility growing experimental pasture grass species, biasing 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 irregularity in south wetland region low LDI
group was SU4 a wetland surrounded by low density residential land use, previously
embedded in pasture with 18.4% Exotic Species. All remaining wetlands in the low LDI
group (n = 40) had less than 10% Exotic Species. Statewide, the percent Exotic Species
was significantly different between low and high LDI groups (W = 1379.0, p < 0.001;
Table 3-19). Table 3-23 lists the 113 Exotic Species encountered throughout Florida and
identifies the wetland regions) in which each species was found. Only six Exotic
Species were found in all four wetland regions including Commelina diffusa, Cuphea
ct, ith1genes\i\, Cynodon dactylon, Kyllinga brevifolia, Ludwigia peruviana, and
Paspalum notatum. Fourteen Exotic Species occurred in three of the four wetland
regions.












7

6

5

4

3


0 2 4
LDI


I
0
0


6 8


0 0
0 *00 0
00 0 0


0 2 4
LDI


0 2 4
LDI


X X

X x
SK


6 8


0 2 4
LDI


Figure 3-15. FQAI scores decreased with increasing development intensity (LDI). A)
FQAI scores at panhandle study wetlands. B) FQAI scores at north study
wetlands. C) FQAI scores at central study wetlands. D) FQAI scores at south
study wetlands.


E U
U U U
Umm I


A

A

AAA Aa


A

AA
A
A


6

S5

S4

u 3


6 8




D.







X
X
X K


























0 2


50

40

30

20

10 o

0 0


mU


Um


6 8


0 o


A A
A A
A A

0 2 4


X X XX
X I


0

0


0 2 4
LDI


Figure 3-16.


6 8


0 2 4
LDI


6 8


Exotic Species increased with increasing development intensity (LDI). A)
Exotic Species at panhandle study wetlands. B) Exotic Species at north
study wetlands. C) Exotic Species at central study wetlands. D) Exotic
Species at south study wetlands.







76


Table 3-23. Macrophyte Exotic Species identified at 118 study wetlands throughout the
four Florida wetland regions.


Exotic Species
Albiziajulibrissin
Aloe vera
Alternanthera philoxeroides
Alternanthera sessilis
Amaranthus blitum
Amaranthus spinosus
Ardisia crenata
Begonia cucullata
Bischofiajavanica
Blechum pyramidatum
Bromus catharticus
Callisia repens
Chenopodium album
Chenopodium ambrosioides
Cinnamomum camphora
Citrus Xaurantium
Colocasia esculenta
Commelina dillm.sa
Conyza bonariensis
Cuphea c.,, ri ,, i i, i,. a
Cyclospermum leptophyllum
Cynodon dactylon
Cyperus iria
Cyperus lanceolatus
Desmodium incanum
Digitaria bicornis
Dioscorea bulbifera
Duchesnea indica
Echinochloa colona
Echinochloa crusgalli
Eichhornia crassipes
Eleusine indica
E, 1,, .-, atrovirens
Eugenia uniflora
Hedychium coronarium
Hedyotis corymbosa
Hemarthria altissima
Hymenachne amplexicaulis
Imperata cylindrica
Ipomoea indica
Ipomoea quamoclit
Kummerowia striata


*










*
*
*


P
*


*



*
*










*


P N C S
r r r


*
*
* *


N C S Exotic Species
Kyllinga brevifolia
*Lantana camera
* Ligustrumjaponicum
Ligustrum lucidum
* Ligustrum sinense
* Lindernia crustacea
Lolium perenne
* Lonicerajaponica
Ludwigia peruviana
Lygodiumjaponicum
* Lygodium microphyllum
Macroptilium lathyroides
Melaleuca quinquenervia
* Melia azedarach
S* Melochia corchorifolia
Momordica charantia
* Morrenia odorata
* * Morus alba
Murdannia nudiflora
S* Nandina domestic
Nephrolepis cordifolia
* * Oeceoclades maculata
Oxalis debilis
* Paederiafoetida
Panicum maximum
Panicum repens
* Paspalidium geminatum
Paspalum acuminatum
* Paspalum notatum
* Paspalum urvillei
Phalaris angusta
* Phyllanthus tenellus
Phyllanthus urinaria
Plantago lanceolata
Pouzolzia zeylanica
* Pueraria montana
Rhodomyrtus tomentosa
* Rhoeo discolor
Richardia brasiliensis
Richardia scabra
Rumex crispus
Rumex obtusifolius


* *
*


* *
*


* *


* *


*
* *


*
* *
*
*










Table 3-23. Continued.
Exotic Species P N C S Exotic Species P N C S
Rumexpulcher 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 ,l.-,tt, ;.'lii
Sorghum bicolor Xyrisjupicai
Spermacoce verticillata Youngiajaponica
Sporobolus indicus Yucca aloifolia
Thelypteris dentata Zea mays
Tradescantiafluminensis
Ecoregions determined by Lane (2000): P = panhandle wetland region; N = north wetland region; C =
central wetland region; S = south wetland region.


Native Perennial Species metric

Of the 605 macrophyte species identified, 427 (71%) were classified as native
perennials. Figure 3-17 shows that the percent Native Perennial Species decreased with
increasing development intensity. The Native Perennial Species metric was significantly
correlated with LDI (Spearman Irl = 0.63, p < 0.001; Table 3-18). 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).

Wetland Status Species metric

Fifty-six percent of the macrophyte species identified were included in the
Wetland Status Species metric, including 160 species designated as obligate and 180
species designated as facultative wetland species. Figure 3-18 shows that the percent
Wetland Status Species decreased with increasing development intensity in each wetland
region. 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
the LDI index (Spearman Irl = 0.55, p < 0.001; Table 3-18).












- 100 m

S90- IL

80 -

S70

60 60

-= 50

S40


'I.


100 d A

S90

80

70

I 60

50

40


0 2 4
LDI


6 8


0 2 4
LDI


C.
o
o o
0

0 0
00 0 0
<6


100

S90

I 80

. 70

S60

4 50

o 40


0 2 4 6 8


0 2 4
LDI


Figure 3-17. The percent 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.


A A
AB
a a1


6 8


100

S 90

S80

S70

Z 60

I 50

o 40


X
x x
X X
X XK


x
X
X x


6 8







79



A. 100


* U

1
*- -


0 2 4
LDI


80

S60


40


6 8





C.


A


A A
a AA AA
A ~a A


0 2 4
LDI


80

0 60
60

7 40


0 2 4 6 8
LDI


x X
X )KX


0 2 4
LDI


Figure 3-18. The percent Wetland Status Species decreased with increasing development
intensity (LDI). This trend was consistent for the (A) panhandle, (B) north,
(C) central, and (D) south wetland regions.


-a

80

2 60


a 40
=


6 8


80

60


U 40


6 8


o
0

0000
00 0 < 0









Macrophyte Florida Wetland Condition Index

The six metrics described above were included in the macrophyte FWCI.
Appendix F provides detailed instructions on metric scoring. Figure 3-19 shows that both
statewide and regional macrophyte FWCI scores decrease with increasing development
intensity. Table 3-24 compares the overall macrophyte FWCI calculated statewide and
regionally for the low LDI group (LDI < 2.0). A comparable statewide macrophyte
FWCI should equally score reference wetlands in each wetland region. However, the
south wetland region had significantly different overall macrophyte FWCI scores for the
low LDI group compared to the panhandle, north, and central wetland regions. When
calculated statewide, five of the six metrics had one or more wetland region with
significantly different metric scores. The north and central wetland regions had
significantly different scores for the statewide Tolerant Indicator Species metric, whereas
the north and south wetland regions had significantly different scores for the statewide
Sensitive Indicator Species metric. The panhandle and north wetland regions were not
significantly different from each other, but were significantly different from the central
and south wetland regions for FQAI scores; suggesting that the panhandle and north
wetland regions hosted more species with a narrower set of ecological conditions found
in reference wetlands. The south wetland region had significantly different statewide
Exotic Species metric scores than the other three wetland regions. The scores for the
statewide Native Perennial Species metric were significantly different for the north
wetland region. When the macrophyte FWCI was scored regionally, there was not a
significant difference between mean scores for the low LDI group (Table 3-24). The
only regionally scored metric with significantly different mean scores for the low LDI
group was the regional Tolerant Indicator Species metric for the south wetland region
which was only significantly different from the panhandle wetland region.
Table 3-25 shows similar results for the high LDI group (LDI > 2.0). For the
statewide macrophyte FWCI wetlands in the north wetland region had significantly
different macrophyte FWCI scores, suggesting the north wetland region high LDI
wetlands had higher ecological integrity than wetlands of other wetland regions. Five of
the six metrics calculated statewide had at least one wetland region with significantly
different mean scores in the high LDI group. Only the Wetland Status Species metric did
not have significantly different scores for both statewide and regional calculations among
wetland regions. For the regional macrophyte FWCI calculations, the north and south
wetland regions had significantly different mean macrophyte FWCI scores. Only the
regional Tolerant Indicator Species metric for the panhandle and south wetland regions
had significantly different mean scores for all of the regionally calculated metrics.
Correlations between macrophyte FWCI and the six component metrics with LDI
were strong (Irl > 0.50, p < 0.01) for all of the metrics statewide and regionally (Table 3-
26), except for the central wetland region Wetland Status Species metric (Irl = 0.39, p =
0.03). This metric was still significantly correlated at a more flexible significance level
(p < 0.05). Three of the four wetland regions, including the panhandle, north, and south
wetland regions had stronger FWCI correlations with LDI than both the statewide and
central wetland region FWCIs. The regional macrophyte FWCI was significantly
correlated with the LDI index (Spearman's Irl = 0.73, p < 0.001). A Kruskal-Wallis test











u0 0 A

50 -

40:
40 -

30 00o
o


A *


A
A A


* A .

A A*:**


AV.L A
~A^


LDI


60 -

50 -

OA
40 -j
00
30

20 -


A

A A
IAA
"S j *


0 2 4 6 8


LDI


o Reference A Agricultural Urban


Figure 3-19. Macrophyte FWCI scores decreased with increasing development intensity
(LDI). A) Statewide FWCI. B) Regional FWCI.


*









Table 3-24. Macrophyte FWCI and component metrics scored statewide and regionally
for study wetlands in the low LDI group (LDI < 2.0).
Panhandle North Central South
Statewide
Macrophyte FWCI 50.3 (8.6)a 51.6 (8.4)a 45.9 (3.4)ab 42.2 (9.4)b
State Tolerant 8.1 (2.5)a 8.4 (2.2)a 6.6 (1.2) 6.8 (2.3)ab
State Sensitive 8.8 (2.6)a 9.0 (1.3)a 8.6 (0.7)ab 6.7 (2.9)b
FQAI Score 8.5 (2.2)a 8.1 (1.4)a 6.6 (0.9)b 6.4 (1.7)b
Exotic Species 9.3 (1.0)a 9.4 (1.1)a 8.4 (1.3)a 6.6 (1.9)b
Native Perennial Species 8.6 (0.9)a 9.1 (1.0)a 8.3 (0.6)b 7.9 (1.1)b
Wetland Status Species 7.0 (2.0)a 7.5 (1.9)a 7.4 (1.6)a 7.7 (2.3)a
Regional
Macrophyte FWCI 50.9 (7.2)a 51.3 (7.9)a 51.4 (4.0)a 46.8 (9.8)a
Regional Tolerant 9.6 (1.0)a 9.2 (1.3)ab 9.0 (1.5)ab 7.7 (2.7)b
Regional Sensitive 8.2 (2.6)a 8.5 (1.7)a 8.6 (1.6)a 6.9 (2.4)a
FQAI Score 7.6 (2.3)a 7.9 (1.4)a 8.7 (1.1)a 7.7 (2.0)a
Exotic Species 9.3 (1.0)a 9.5 (1.1)a 8.3 (1.4)a 8.3 (1.9)a
Native Perennial Species 8.3 (1.0)a 9.2 (0.9)a 9.1 (0.7)a 8.7 (1.2)a
Wetland Status Species 8.0 (2.1)a 7.0 (2.5)a 7.8 (1.9)a 7.6 (2.1)a
Values represent the mean (standard deviation).
Wetland regions with similar letters were not significantly different (p<0.05)




Table 3-25. Macrophyte FWCI and component metrics scored statewide and regionally
for study wetlands in the high LDI group (LDI > 2.0).
Panhandle North Central South
Statewide
Macrophyte FWCI 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
FQAI Score 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 FWCI 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
FQAI 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
Values represent the mean (standard deviation).
Wetland regions with similar letters were not significantly different (p<0.05).









Table 3-26. Spearman correlations between the macrophyte FWCI and component
metrics with LDI.
Spearmans r p-value
Statewide
Macrophyte FWCI -0.73 <0.0001
Tolerant Indicator Species 0.75 <0.0001
Sensitive Indicator Species -0.66 <0.0001
FQAI Score -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 FWCI -0.74 <0.0001
Tolerant Indicator Species 0.72 <0.0001
Sensitive Indicator Species -0.68 <0.0001
FQAI Score -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 FWCI -0.74 <0.0001
Tolerant Indicator Species 0.80 <0.0001
Sensitive Indicator Species -0.78 <0.0001
FQAI Score -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 FWCI -0.73 <0.0001
Tolerant Indicator Species 0.68 <0.0001
Sensitive Indicator Species -0.72 <0.0001
FQAI Score -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 FWCI -0.88 <0.0001
Tolerant Indicator Species 0.86 <0.0001
Sensitive Indicator Species -0.71 <0.0001
FQAI Score -0.86 <0.0001
Exotic Species 0.80 <0.0001
Native Perennial Species -0.80 <0.0001
Wetland Status Species -0.69 <0.0001










suggested a significant difference (H = 66.3, p < 0.001) among median regional
macrophyte FWCI scores for study wetlands among the three apriori land use categories
(reference, agricultural, urban).

Cluster Analysis

Cluster analysis determined five categories of wetlands based on macrophyte
community composition. Clusters were roughly defined by wetland regions and a priori
land use categories, including: 1: northern reference; 2: southern reference; 3: northern
developed land use; 4: southern developed land use; and 5: statewide pasture land use.
Figure 3-20 shows that based on regional macrophyte FWCI scores, clusters 1 and 2 were
not significantly different from one another, but were significantly different from clusters
3, 4, and 5 (p < 0.05). Clusters 3 and 4 were not significantly different from each other.
Cluster 5 was significantly different from all other clusters. Identical results were
obtained using statewide macrophyte FWCI scores. Table 3-27 provides means and
standard deviations for cluster statewide macrophyte FWCI scores, regional macrophyte
FWCI scores, and LDI.


60-


50


40


30


20


10 _


a

*


I
Cluster 1


a


I
Cluster 2


b


T


I
Cluster 3


Jb


U


I
Cluster 4


Cluster
Cluster 5


Figure 3-20. Regional macrophyte FWCI scores for
macrophyte community composition.


five wetland clusters based on






85

Table 3-27. Macrophyte FWCI scores and LDI values for wetland clusters based on
macrophyte community composition.
Statewide Regional
Cluster Macrophyte FWCI Macrophyte FWCI LDI
1 46.8 (11.1)a 47.4 (9.9)a 2.3 (1.9)a
2 44.5 (7.7)a 47.4 (9.7)a 1.7(1.3)a
3 24.9 (12.2)b 25.9 (12.6)b 4.4 (1.9)bc
4 21.9 (13.8)b 23.2 (16.2)b 4.0 (2.0)b
5 10.6 (7.0)c 8.8 (6.6)c 5.2 (0.5)c
Values represent mean (standard deviation).
Clusters with similar letters within columns were not significantly different (p<0.05).









Macroinvertebrate Assemblage

Statewide 79 wetlands were sampled for the macroinvertebrate assemblage, with
118 species, 169 representing genera, 85 families, 24 orders, nine classes, and five phyla.
The most common macroinvertebrate 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 encountered, 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%) were encountered at only one wetland.
The most common families identified included Chironomidae, Naididae,
Enchytraeidae, and Culicidae, representing 39, 19, 4, and 4% of the individuals
identified, and occurring at 99, 81, 52, and 56% of the study wetlands, respectively.
Macroinvertebrates in the family Chironomidae were further divides into the subfamilies
Chironominae (89% of Chironomidae), Tanypodinae (10% of Chironomidae), and
Orthocladiinae (1% of Chironomidae). Six orders were found at over 50% of the
wetlands sampled, including Diptera (47% of individuals identified to the order
taxonomic level or lower), Tubificida (24%), Coleoptera (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 phyla were identified,
including Arthropoda, Annelida, Mollusca, Platyhelminthes, and Nemertea, with
Arthropoda, Annelida, Mollusca found at 100, 92, and 65% of the wetlands sampled,
respectively.
In the panhandle wetland region, 13 wetlands were sampled hosting 84 genera
representing 48 families and 17 orders. In the north wetland region 15 wetlands were
sampled with 87 genera (58 families and 20 orders) encountered. The central wetland
region included 25 wetlands with 109 genera (60 families and 23 orders) identified. The
south wetland region had 26 sample wetlands with 105 genera (in 60 families and 21
orders) identified.

Summary Statistics

Richness (R), evenness (E), Shannon diversity (H), and Simpson's index (S) were
calculated for each sample wetland (Appendix E). Richness ranged from one genus
(Pristina) at PR4, a wetland embedded in a low-intensity silvicultural land use, to 26
genera at PA3, a wetland surrounded with row crops. Species evenness ranged from 0.00
at PR4 to 0.97 at PR5, a wetland surrounded by upland forest. Shannon diversity ranged
from 0 at PR4 to 0.92 at CA7, a wetland surrounded by silvicultural operations and
pasture with cattle. Simpson's index was highest at CR10 at 2.79. Table 3-28
summarizes the richness, evenness, and diversity calculations by a priori land use
category (reference, agricultural, or urban). No significant differences were found in
richness, evenness, Shannon diversity, or Simpson's index among the three a priori land
use categories. Beta and gamma diversity were also similar among a priori land use
categories. Table 3-29 shows that the no significant differences were found for richness,




University of Florida Home Page
© 2004 - 2011 University of Florida George A. Smathers Libraries.
All rights reserved.

Acceptable Use, Copyright, and Disclaimer Statement
Last updated May 24, 2011 - Version 3.0.0 - mvs