<%BANNER%>

An Evaluation

Permanent Link: http://ufdc.ufl.edu/UFE0041309/00001

Material Information

Title: An Evaluation the Florida Department of Environmental Protections Lake Vegetation Index (lvi)
Physical Description: 1 online resource (48 p.)
Language: english
Creator: Thomas, Eric
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2009

Subjects

Subjects / Keywords: fish, index, lvi, temporal, water
Forest Resources and Conservation -- Dissertations, Academic -- UF
Genre: Fisheries and Aquatic Sciences thesis, M.S.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: Abstract of Thesis Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Master of Science AN EVALUATION: THE FLORIDA DEPARTMENT OF ENVIRONMENTAL PROTECTIONS? LAKE VEGETATION INDEX (LVI) By Eric Flynt Thomas December 2009 Chair: Daniel E. Canfield, Jr. Major: Fisheries and Aquatic Sciences The Florida Department of Environmental Protection (FDEP) developed in 2005 an ecological assessment index for Florida lakes (Fore 2005) called the Lake Vegetation Index (LVI). The index uses aquatic plant (macrophyte) species as an indicator of human disturbance to lakes. The purpose of this project was to evaluate the effectiveness of the LVI for predicting water chemistry and to determine how LVI relates to fish communities in Florida lakes. Florida LAKEWATCH, a citizen volunteer monitoring program, has extensive long-term data for water chemistry, as well as data for aquatic plant and fish communities for a diverse group of Florida lakes. It was determined, using 20 lakes sampled in 2008, that LAKEWATCH data could be used to calculate LVI scores that were comparable to scores calculated using the FDEP protocol. Weak relationships (R2 values < 0.35) were established between LVI scores and total phosphorus, total nitrogen, chlorophyll, pH, total alkalinity, specific conductance, and color measured on the same day. The same pattern also existed for some of the long-term total phosphorus, total nitrogen, chlorophyll, and Secchi depth, surface area, and several fish community metrics (Simpson's diversity index, Simpson's evenness, Shannon-Weiner diversity index, species richness, relative sportfish biomass, relative non-native species biomass, and non-native species presence) collected by LAKEWATCH. Based on the low R2 values, the LVI does not seem to be a valid method for assessing ecological integrity or assessing the impact of human disturbance on Florida lakes.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Eric Thomas.
Thesis: Thesis (M.S.)--University of Florida, 2009.
Local: Adviser: Canfield, Daniel E.

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2009
System ID: UFE0041309:00001

Permanent Link: http://ufdc.ufl.edu/UFE0041309/00001

Material Information

Title: An Evaluation the Florida Department of Environmental Protections Lake Vegetation Index (lvi)
Physical Description: 1 online resource (48 p.)
Language: english
Creator: Thomas, Eric
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2009

Subjects

Subjects / Keywords: fish, index, lvi, temporal, water
Forest Resources and Conservation -- Dissertations, Academic -- UF
Genre: Fisheries and Aquatic Sciences thesis, M.S.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: Abstract of Thesis Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Master of Science AN EVALUATION: THE FLORIDA DEPARTMENT OF ENVIRONMENTAL PROTECTIONS? LAKE VEGETATION INDEX (LVI) By Eric Flynt Thomas December 2009 Chair: Daniel E. Canfield, Jr. Major: Fisheries and Aquatic Sciences The Florida Department of Environmental Protection (FDEP) developed in 2005 an ecological assessment index for Florida lakes (Fore 2005) called the Lake Vegetation Index (LVI). The index uses aquatic plant (macrophyte) species as an indicator of human disturbance to lakes. The purpose of this project was to evaluate the effectiveness of the LVI for predicting water chemistry and to determine how LVI relates to fish communities in Florida lakes. Florida LAKEWATCH, a citizen volunteer monitoring program, has extensive long-term data for water chemistry, as well as data for aquatic plant and fish communities for a diverse group of Florida lakes. It was determined, using 20 lakes sampled in 2008, that LAKEWATCH data could be used to calculate LVI scores that were comparable to scores calculated using the FDEP protocol. Weak relationships (R2 values < 0.35) were established between LVI scores and total phosphorus, total nitrogen, chlorophyll, pH, total alkalinity, specific conductance, and color measured on the same day. The same pattern also existed for some of the long-term total phosphorus, total nitrogen, chlorophyll, and Secchi depth, surface area, and several fish community metrics (Simpson's diversity index, Simpson's evenness, Shannon-Weiner diversity index, species richness, relative sportfish biomass, relative non-native species biomass, and non-native species presence) collected by LAKEWATCH. Based on the low R2 values, the LVI does not seem to be a valid method for assessing ecological integrity or assessing the impact of human disturbance on Florida lakes.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Eric Thomas.
Thesis: Thesis (M.S.)--University of Florida, 2009.
Local: Adviser: Canfield, Daniel E.

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2009
System ID: UFE0041309:00001


This item has the following downloads:


Full Text

PAGE 1

1 AN EVALUATION: THE FLORIDA DEPARTMENT OF ENVIRONMENTAL PROTECTION’S LAKE VEGETATION INDEX (LVI) By ERIC FLYNT THOMAS A THESIS PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORID A IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE UNIVERSITY OF FLORIDA 2009

PAGE 2

2 2009 Eric Flynt Thomas

PAGE 3

3 TABLE OF CONTENTS page LIST OF TABLES ............................................................................................................ 4LIST OF FI GURES .......................................................................................................... 5LIST OF ABBR EVIATIONS ............................................................................................. 6ABSTRACT ..................................................................................................................... 8CHAPTER 1 INTRODUC TION .................................................................................................... 102 METHOD S .............................................................................................................. 14LVI Calculat ions ...................................................................................................... 14LVI Variab ility .......................................................................................................... 16LVI and Water Chemistry Comparison .................................................................... 16Laboratory A nalyses ............................................................................................... 18LVI and Fish Communi ty Compar isons ................................................................... 18LVI and Human Dis turbance ................................................................................... 19Statistical Pr ocedures ............................................................................................. 203 RESULT S ............................................................................................................... 224 DISCUSSI ON ......................................................................................................... 40LIST OF RE FERENCES ............................................................................................... 45BIOGRAPHICAL SKETCH ............................................................................................ 48

PAGE 4

4 LIST OF TABLES Table page 2-1 List of Florida lakes sampled in su mmer 2008 for LVI calculations using the FDEP and LAKEWATCH protocol s. ................................................................... 213-1 Percent agreement before and after ex clusion of plants not considered by LAKEWATCH as aquatic specie s. ...................................................................... 353-2 Coefficients of variation (%) for scores calculated using the LAKEWATCH method, with number of sampli ng events (N) in cluded. ...................................... 363-3 Lakes used to compare long-term to tal phosphorus (TP), total nitrogen (TN), chlorophyll (chl), and Se cchi depth to LVI. ......................................................... 373-4 Florida lakes used to compare fish community metrics to LVI scores, along with electrofishing and LVI score years used in analysis. ................................... 39

PAGE 5

5 LIST OF FIGURES Figure page 3-1 Linear regression of LAKEWATCH and FDEP calculated LVI scores with E Lake incl uded ..................................................................................................... 263-2 Linear regression of LAKEWATCH and FDEP calculated LVI scores with E Lake exclu ded. ................................................................................................... 263-3 Linear regression of percent native species (PNS) as calculated by FDEP and LAKEWAT CH. ............................................................................................. 273-4 Linear regression of percent invasive species (PIS) as calculated by FDEP and LAKEWAT CH .............................................................................................. 273-5 Linear regression of percent sensitiv e species (PSS) as calculated by FDEP and LAKEWAT CH .............................................................................................. 283-6 Linear regression of dominant species coefficient of conservatism (DCC) as calculated by FDEP and LAKEWAT CH .............................................................. 283-7 Linear regression of LAKEWATCH LVI score against long-term total phosphorus ( g/L). ............................................................................................. 293-8 Linear regression of LAKEWATCH LV I score against long-term total nitrogen (g/L) ................................................................................................................. 293-9 Linear regression of LAKEWATCH LVI score against long-term chlorophyll concentration (g/L). ........................................................................................... 303-10 Linear regression LAKEWATCH LVI score against long-term Secchi depth (ft). ...................................................................................................................... 303-11 Linear regression of LVI scores agai nst Simpson’s Divers ity Index D. ............... 313-12 Linear regression of LVI scores a gainst Simpson’s Ev enness values. ............... 313-13 Linear regression of LVI scores agai nst sportfish relati ve biomass. ................... 323-14 Linear regression of LVI scores against non-native species relative biomass. ... 323-15 Linear regression of LVI score s against specie s richnes s. ................................. 333-16 Linear regression of LVI scores a gainst Shannon-Weiner Diversity Index values. ................................................................................................................ 333-17 Plot of LVI scores of the va rious human distur bance leve ls. .............................. 34

PAGE 6

6 LIST OF ABBREVIATIONS ANOVA analysis of variance APHA American Public Health Association CaCO3 calcium carbonate chl chlorophyll concentration (g/L) CV coefficient of variation D Simpsons Diversity Index FDEP Florida Department of Environmental Protection FFWCC Florida Fish and Wildlife Conservation Commission GPS global positioning system IBI Indices of Biological Integrity LAKEWATCH Florida LAKEWATCH LVI Lake Vegetation Index m meters mL milliliters mm millimeters m2 meters squared PLEX Plant Lake Ecotype Index PtCo platinum-cobalt units r correlation coefficient R2 coefficient of determination s number of fish species in a transect TL total length TN total nitrogen concentration (g/L)

PAGE 7

7 TP total phosphorus concentration (g/L) USEPA United States Environmental Protection Agency statistical significance level g/L micrograms per liter S microsiemens C degrees Celsius § section

PAGE 8

8 Abstract of Thesis Presented to the Graduate School of the University of Florida in Partial Fulf illment of the Requirements for t he Degree of Master of Science AN EVALUATION: THE FLORIDA DE PARTMENT OF ENVIRONMENTAL PROTECTIONS’ LAKE VEGETATION INDEX (LVI) By Eric Flynt Thomas December 2009 Chair: Daniel E. Canfield, Jr. Major: Fisheries and Aquatic Sciences The Florida Department of Environmental Protection (FDEP) developed in 2005 an ecological assessment index for Florida lakes (Fore 2005) called the Lake Vegetation Index (LVI). The index uses aquatic pl ant (macrophyte) species as an indicator of human disturbance to lakes. The purpose of this project was to evaluate the effectiveness of the LVI for predicting water chemistry and to determine how LVI relates to fish communities in Florida lakes. Florida LAKEWATCH, a citizen volunteer monitoring program, has extensive long-term data for water chemistry, as well as data for aquatic plant and fish communi ties for a diverse group of Florida lakes. It was determined, using 20 lakes sampled in 2008, that LAKEWATCH data could be used to calculate LVI scores that were compar able to scores calculated using the FDEP protocol. Weak relationships (R2 values < 0.35) were established between LVI scores and total phosphorus, total nitrogen, chloroph yll, pH, total alkalinity, specific conductance, and color measured on the same day. The same pattern also existed for some of the long-term total phosphorus, to tal nitrogen, chlorophyll, and Secchi depth, surface area, and several fish community me trics (Simpson’s diversity index, Simpson’s evenness, Shannon-Weiner diversity index, species richness, relative sportfish

PAGE 9

9 biomass, relative non-native species biomass, and non-native species presence) collected by LAKEWATCH. Based on the low R2 values, the LVI does not seem to be a valid method for assessing ecological in tegrity or assessing the impact of human disturbance on Florida lakes.

PAGE 10

10 CHAPTER 1 INTRODUCTION The Clean Water Act (33 United States Congress §1251 et seq. 1972) requires protection of the biological, physical, and c hemical integrity of waters in the United States to insure that those waters can support “the protection and propagation of fish, shellfish, wildlife, and recreation in and on t he water.” Assessment of the integrity of aquatic systems is therefore a desirable goal for water resource managers (USEPA, 1998). Many ways of assessing the biologica l condition of lakes have been proposed using different biocriteria (Fore 2005, Stel zer et al. 2005, Bourdaughs et al. 2006). One biocriterium has been the use of vegetation (e .g., aquatic macrophytes) as an indicator of the ecological condition of a lake. Such indices have been proposed widely in Europe. For example, Stel zer et al. (2005) developed a macrophyte-based assessment system for use in German lakes. In their index, classification was based on macrophyte species abundance. Duigan et al. (2007) pr esented a Plant Lake Ecotype Index (PLEX) for use in British lakes. This index, as well as the index develo ped by Stelzer et al. (2005), involved estimating the abundance of macr ophytes by classifying them on a 1-5 scale, ranging from rare to dominant. In the United States, vegetation indi ces have been most extensively proposed for wetland integrity studies. Bourdaughs et al. ( 2006) reported that the “Floristic Quality Index” was a good indicator for co astal wetland ecological c onditions in the Great Lakes region. Other authors have used indices in si milar ways at other locations, such as northern Ohio (Andreas and Lichvar 1995), Florida (Cohen et al. 2004), and Illinois (Matthews 2003). Nichols et al. (2000), usin g Wisconsin lakes, proposed an index for assessing the biological quality of lakes. The components of their index included

PAGE 11

11 maximum depth of plant growth, percentage of littoral zone vegetated, Simpson's diversity index, relative frequencies of subm ersed, sensitive, and exotic species, and the number of taxa. Bachmann et al. ( 2002) showed, however, that there was no relationship between biomass of aquatic pl ants and nutrient concentrations in Florida lakes. Furthermore, they concluded that t he role of macrophytes in clearing lakes may be primarily due to reduced nutrient concentrati ons for a given level of nutrient loading rather than nutrient concentrations contro lling macrophyte abundance. Their finding suggests that indices based on aquatic macr ophyte communities may not relate well to water chemistry. The use of indices of biological integr ity (IBIs) has been chal lenged in Florida. Schulz et al. (1999) examined an IBI that us ed eight metrics of fish assemblages to estimate anthropogenic impacts. In their study of 60 Florida lakes, Schulz et al. (1999) demonstrated that the IBI was unable to predi ct measures of anthropogenic impact. Given the results of the Schulz et al. ( 1999) and Bachmann et al. (2002) studies, the use of any IBIs, even those that use pl ant communities as metrics, should be scrutinized. However, in 2009, the Florida D epartment of Envir onmental Protection (FDEP) proposed using the lake vegetation index (LVI) to assess the ecological condition of Florida lakes and anthropogenic impact to lakes (Fore 2005). FDEP had conducted a study (Fore 2005) where candidate me trics of lake plant communities were identified, tested against i ndependent gradients of human disturbance, and combined into the lake vegetation index. In FDEP’ s statewide study, one data set, composed of 95 lakes, was used to test the candidate me trics and construct t he index. A second data set, composed of 63 lakes, was used to validate the correlation between LVI and

PAGE 12

12 independent measures of hum an disturbance (Fore 2005). The four plant community metrics that were included in the index were: percent native taxa, percent invasive taxa, percent sensitive taxa, and the coefficient of conservatism score for the dominant plant species (Fore 2005). LVI scores range from 0 to 100, with lower values indicating a degraded or disturbed lake. Bachmann et al. (2009) cha llenged the LVI concept and concluded that the index could not be used to determine if lakes were impaired by human disturbance. Bachmann et al. (2009) showed that the index could not s eparate out natural processes that determine the vegetation qua lity in a lake from human fact ors, similar to the results of Schulz et al. (1999). Florida LAKEWATCH is a citizen volunteer-based monitoring program that monitors water chemistry trends in Florida lakes. LAKEWATCH also conducts aquatic plant surveys and does long-ter m fish monitoring for the Florida Fish and Wildlife Conservation Commission (FFWCC) This information is public, and provides a large quantity of data regardi ng water chemistry, aquatic macrophyte communities, and fish communities, which can be used to test the relationship of the LVI to water quality and fish communities in a diverse group of Florida lakes. The use of this dataset also provides a way to ev aluate the LVI using dat a collected independent of the Fore (2005) study. The primary pur pose of this study was to examine the effectiveness of the LVI as an indicator of lake condition using water quality and fish community data. There were several goals for this study, including: 1) to determine if plant survey information, collected by LAKEWATCH, can be used to calculate an LVI value comparable to the FDEP LVI value for an individual lake, 2) to determine how much temporal variation is associated with a lake’s LVI, 3) to determine if there is a

PAGE 13

13 relationship between LVI scores and concu rrent (single sampling event) long-term (multiple years) water chemis try as measured by LAKEWATCH, 4) to determine if there is a relationship between LVI scores and me trics of fish communities, and, 5) to determine if there is a relationship between LVI scores and human disturbance.

PAGE 14

14 CHAPTER 2 METHODS LVI Calculations The protocol developed by FDEP for LVI sa mpling involves dividing a lake into 12 sections (like the numbers on a clock), and selecting sections 1, 4, 7, and 10, sections 2, 5, 8, and 11, or sections 3, 6, 9, and 12 for sampling by rolling a die (Fore 2005). During sampling of each section (For e 2005), a transect is set up perpendicular to the shore where a frotus plant sampling unit is deployed at least five times to sample submersed plants within 2.5 mete rs of both sides of the bo at (to create a five-meterwide belt transect). Then, the boat is driven parallel to the shore, where all plants that can be identified through visual observation ar e recorded. Plant species that cannot be identified, but can be reached from the boat are harvested fo r later identification. Finally, a single dominant or two co-dominant species are identified based on visual observation and recorded. Florida LAKEWATCH has a different prot ocol for aquatic plant sampling (Florida LAKEWATCH 2007). The above-ground standing crop of emergent, floating-leaved, and submerged vegetation are m easured along uniformly placed transects (10 to 30, depending on lake size) inside a 0.25-m2 quadrat that is randomly placed in each plant zone (one deployment per zone per transect). Harvested plants are placed into a nylonmesh bag, hand-spun to remove excess water, and weighed to the nearest tenth of a kilogram. The mean above-grou nd biomass of each plant zone is then averaged across all transects to get a lakewide value. T he combined width of the emergent and floatingleaved plant zones (i.e., the di stance from the outermost edge of the combined zones to shore), measured using a hand-held range finder at each transect, are then averaged

PAGE 15

15 for the lake. All plant species, observed during sampling, are recorded, and frequency of occurrence is determined by dividing the num ber of transects in which a plant species is found by the total num ber of transects. To determine if plant survey information collected by LAKEWATCH can be used to calculate LVI scores, 20 lakes were sampled during the summer of 2008 for aquatic plants using the FDEP and LAKEWATCH protoc ols. The study lakes represented a variety of human disturbance levels and trophic states (Table 2-1). LVI scores for each lake were separately calculated usi ng both the LAKEWATCH protocol and FDEP protocol. To calculate an LVI score usi ng the LAKEWATCH protocol the plant species list for all combined transects was used to calculate percent native species, percent invasive species, and percent sensitive spec ies. The dominant plant coefficient of conservatism was determined as the coefficien t of conservatism for the one or multiple species with the highest percent occurrenc e in the LAKEWATCH transects. If more than one species co-dominated, t he average of the species c oefficients of conservatism was used. One lake (Juniper, Walton County) was divided into two halves (Juniper East and Juniper West) for LAKEWATCH sampling. To account for this division, each half was scored for an LVI using the LAKEWATCH protocol, then the two LVI scores were averaged for comparison to the LVI score ca lculated using the FD EP protocol. Linear regression analysis was used to compar e calculated LAKEWATCH and FDEP LVIs from the 20 lakes, and provided a basis for determining whether the long-term LAKEWATCH plant database could be used to calculate LVI scores for past years and other lakes where LAKEWAT CH sampled plants.

PAGE 16

16 LVI Variability To examine sources of variation between LVI scores calculated using the two protocols, % agreement in recorded plant species was ca lculated by dividing the number of plant species identified by the total number of species discovered using the FDEP and LAKEWATCH protocols. Also, the individual me trics used to calculate the LVIs (i.e., percent native species, percent invasive species, percent sensitive species, and dominant species coefficient of conservati sm) were compared using linear regression for the FDEP and LAKEWATCH protocols. To examine the magnitude of temporal va riation in calculated LVI scores, data from the 20 lakes (examined in summer 2008) were used with data from 26 additional LAKEWATCH lakes. Combined, these 46 lake s had multiple years of aquatic plant survey data, thus an LVI score was calcul ated for each year of available data. Coefficients of variation for i ndividual lakes were then calculated for the time series of data by averaging the score for an individual lake across all years, then dividing the standard deviation of the LVI scores for an indivi dual lake by the mean of the LVI scores (Krebs 1999). LVI and Water Chemistry Comparison At the time of plant samp ling, a water sample was collected according to Florida LAKEWATCH protocol (see below) at each lake to determine concurrent water chemistry (sampled at the time of pl ant sampling) and water transparency was measured using a Secchi disc at an open-wate r location chosen randomly. The water sample was placed on ice until analyzed at the University of Florida’s Fisheries and Aquatic Sciences water quality laborato ry where all LAKEWATCH samples are analyzed (see below).

PAGE 17

17 To determine if there was a relationshi p between the calculated LVIs and longterm water chemistry in Florida lakes, wate r chemistry data were extracted from the long-term LAKEWATCH database for each study lake (Florida LAKEWATCH 2007). At each lake, surface water (0.5-m) samples were collected at three open-water stations. Secchi depth was also measured at each station. Water for TP and TN analyses were collected in 250-mL, acid cleaned, triple rins ed Nalgene bottles. A dditional water was collected at each station in rinsed 4-L plasti c milk jugs for chlorophyll analyses. To estimate algal biomass as measured by chlo rophyll, a measured volume of water from these jugs was filter ed through a Gelman Type A-E glass fiber filter. These filters were stored over silica gel desiccant and frozen. All samples were then transported to the laboratory for analyses by Florida LAKEWAT CH. Long-term water quality values for each of the study lakes were calculated by averaging values for each sampling date, and then averaging all sampling dates for a lake to obtain a single mean value (the grand mean). Florida LAKEWATCH sampled aquatic plants on 50 lakes in 2007 and 2008. Of the 50 lakes, 41 of these lakes had concurr ent water chemistry dat a available at the time of plant sampling, which permitted an examination of the relationship between LVI and limited (i.e., same-day sampling) water c hemistry. For lakes where samples were not taken during the plant sampling event, su pplemental data were used from the next closest sampling date. These data were used to compare the LVI scores to pH, color, total alkalinity, specific conductance, TP, TN and chl using linear regression. Finally, the LVI scores from the 50 lakes were then re lated to long-term (i.e ., multiple dates of sampling) means of TP, TN, chl, and Secchi depth using linear regression.

PAGE 18

18 Laboratory Analyses At the laboratory, the concurrent and long-term wa ter samples were analyzed to determine TP, TN, and chl concentrations (g/L) Additional analyses included pH, total alkalinity (mg/L as CaCO3), specific conductance (S/cm at 25 C), and color (Pt-Co units). TP concentrations were determined using the procedures of Murphy and Riley (1962) with a persulfate digesti on (Menzel and Corwin 1965). TN concentrations were determined by oxidizing water samples with persulfate and determining nitrate-nitrogen with a second derivative spectroscopy (D’Eli a et al. 1977, Simal et al. 1985, Wollin 1987). Chl concentrations were determi ned spectophotometrica lly (APHA 2005) following pigment extraction with ethanol (S artory and Grobbelaar, 1984). To determine pH, an Accumet model 10 pH meter calibrat ed with buffers of 4.0, 7.0, and 10.0 was used. Total alkalinity was measured by titrat ion with 0.02 N sulfuric acid (APHA 2005). Specific conductance was meas ured at 25 C using a Yellow Springs Instrument Model 35 conductance meter. Color was determined by spectroscopic comparison to platinum-cobalt standard solutions based on 500 APHA color units (Bowling et al. 1986). LVI and Fish Community Comparisons Fish were collected by Florida LAKEWATCH using electrofishing during the fall of each year from 1999-2008 (Florida LAKEWATCH 2007). The same six uniformly spaced 10-minute transects were sampled by electrofishing on each lake each year. GPS locations were used to ensure consistent sampling of the same area throughout years. One person dipped fish on these surveys, and electrical current was constantly applied for 10 minutes. Collected fish were placed into an aerated tank until they were identified to species and measured (mm TL). Fish were released immediately after

PAGE 19

19 measurement. Length-weight relati onships obtained by LAKEWATCH from unpublished FFWCC data (Schaeffer 2007) were used to estimate individual fish weights. From t hese data, LAKEWATCH calculated catch per unit effort for abundance (number of fish) and biomass (total estima ted weight), relative abundance per species and biomass (weight per species), species ri chness, species diversity (including three forms of the Simpson’s dive rsity index; D, 1/D, and 1-D), and Simpson’s evenness (calculated as (1/D)/s, where s equals the number of species in a transect). To examine the relationships between LVI and fish community metrics, data from 31 lakes, sampled between 2005 and 2009, were used. Twenty-five of the lakes had electrofishing and aquatic macrophyte data fo r the same years (Table 3-4). For the remaining six lakes, the aquatic plant dat a were collected within one year of the electrofishing surveys, thus t he calculated LVI scores were offset by one year from the time of fish sampling. Seve ral “lakes” connected by water (e.g., streams, canals) were considered as separate lakes for the LVI ca lculations. These lakes, however, were considered as a single unit by LAKEWATCH fo r the purpose of electrofishing effort. To account for this difference, a single mean LV I score was calculated for comparison with LAKEWATCH’s electrofishing data (Iv anhoe, representing Ivanhoe East, Ivanhoe Middle, and Ivanhoe West; Conway represent ing Conway North and Conway South; Josephine representing Josephine East, Jo sephine Center, and Josephine West; and Juniper representing Juniper East and Juniper West). LVI and Human Disturbance Finally, LVI scores were compared to human disturbance levels using the 20 lakes sampled in summer 2008 (Table 2-1). On each of these lakes, any human disturbance (e.g. houses, artificial canals, agriculture, silviculture, etc.) was

PAGE 20

20 documented, and lakes were subsequently classified as: Categoty-1, low human disturbance; Category-2, inte rmediate human disturbance; or Category-3, high human disturbance (Table 2-1). These classification s were compared to LVI scores calculated using the FDEP protocol with an anal ysis of variance (ANOVA) test. Statistical Procedures JMP 8.0 was used in all statistical analyses. Log 10 transformations were performed on all parameters (except pH) to normalize the data. Long-term electrofishing data, dating back to 1999, were used to determine if any non-native fish species had ever been captur ed since 1999 on each lake. A Welch’s two-tailed t-test was used to test for differences in LV I scores for lakes with and without non-native species presence. A significance level of = 0.05 was chosen for all analyses comparing LVIs to water chemis try and fish community metrics.

PAGE 21

21 Table 2-1. List of Florida lakes sampled in summer 2008 for LVI calculations using the FDEP and LAKEWATCH protocols. Lake County Surface Area (ha) Trophic State Human Disturbance Level** Apopka Orange 12412Hypereutrophic2 Cherry Lake 248Mesotrophic3 Crescent Putnam 6458Eutrophic2 Dorr Lake 759Mesotrophic1 E Miami-Dade 39Oligotrophic3 Eloise Polk 469Eutrophic3 George Putnam 18907Eutrophic1 Griffin Lake 6679Eutrophic2 Harris Lake 5579Eutrophic2 June Highlands 2316Oligotrophic3 Juniper Walton 271Oligotrophic1 Mill Dam Marion 125Oligotrophic1 Minneola Lake 764Eutrophic3 Sampson Bradford 754Mesotrophic1 Spring Walton 97Oligotrophic1 Tarpon Pinellas 1025Eutrophic3 Tohopekaliga East Osceola 5540Mesotrophic3 Wauberg Alachua 149Hypereutrophic2 Weir Marion 2861Oligotrophic3 Wildcat Lake 142Oligotrophic1* Trophic state determined using long-term TP (g/L) means and the classification system of Forsberg and Ryding (1980). **Human disturbance classified by visual observation as: Category-1, low human presence/disturbance, Category-2, interme diate human presence/ disturbance, or Category-3, high human presence/disturbance.

PAGE 22

22 CHAPTER 3 RESULTS LVI scores calculated for the 20 Florida la kes in this study (using aquatic plant information collected using the LAKEWATCH plant sampling protocol) were significantly correlated (r = 0.83) to LVI scores calc ulated using the FDEP protocol. Linear regression analysis demonstr ated that there was nearly a 1:1 slope between the two LVI scores (slope = 0.97, p < 0. 001) (Figure 3-1). The c oefficient of determination (R2 = 0.69), however, indicated that there were other sources of variance associated with the relationship. For example, one of the study lakes was actually an urbanized, barrow-pit lake (E Lake, Miami-Dade County, Florida). Excluding E Lake from the analysis (Figure 3-2), yielded a 1:1 line (slope of 1.02) and an R2 of 0.72, providing justification for excluding E Lake from other anal yses where other sources of va riation were identified. Both the LAKEWATCH and the FDEP prot ocols identified similar numbers of plant species, and no particular plant types were routinely missed by one or the other protocols. Percent agreem ent between the plant species documented, however, was only about 61% (Table 3-1). Lack of co mplete agreement was not unexpected because some species, considered to be aquat ic species by FDEP, are not recognized by LAKEWATCH as aquatic species (e.g., dog fennel, Eupatorium leprophyllum ). Excluding these species from the analysis, however, only improved the agreement to 67%, suggesting that this discrepancy in t he protocols accounted for only a small amount of the overall variance (Table 3-1). Of the 20 lakes sampled, 10 of the lakes had their percent agreement improved by 5% or less, nine improved by 6% to 10%, but one lake had its value improved by 20%, su ggesting differences between the calculated LVIs for individual lakes can be substantial. A paired t-test failed to show a significant

PAGE 23

23 difference between percent agr eement of the two methods before and after the removal of species not considered to be a quatic by Florida LAKEWATCH. Another source(s) of variation may be related to how the individual metrics (percent native species, percent invasive species, percent sensitive species, and dominant species coefficient of conservati sm) were calculated using the LAKEWATCH and FDEP protocols (Figures 3-3 through 3-6). R2 values for percent native species, percent invasive species, percent sensitive species, and dominant species coefficient of conservatism were 0.53, 0.53, 0.66, and 0.46, respectively, again indicating methodology is not greatly influencing the ov erall calculation of LVI for a group of Florida lakes. Producing statistically similar results, LVIs can be calculated using either LAKEWATCH or FDEP protocols. Analysis of LAKEWATCH’s macrophyte information permitted an assessment of the magnitude of temporal variati on in the LVI scores over multiple years. The mean c oefficient of variation (CV) for the data was 16%, with a maximum CV was 61% (see Table 3-2) The minimum CV was 1%, the 25th quartile was 8%, and the 75th quartile was 20%. The analysis of LVI scores (indep endent variable) and water chemistry (dependent variable) information (using non-tr ansformed data), obtained concurrently with the plant sampling for the 41 Florida lakes in this study, demonstrated weak correlations between LVI and pH (R2 = 0.23), total alkalinity (R2 = 0.07), specific conductance (R2 = 0.01), TP (R2 = 0.08), TN (R2 = 0.04), chl (R2 = 0.06), and color (R2 = 0.02). Transforming (log base 10) the data improved the statisti cal relationships slightly, but the relationships remained weak (R2 < 0.1).

PAGE 24

24 To determine if there mi ght be a stronger relations hip between LVI scores and long-term water chemistry data, information from 50 Florida lakes (Table 3-3) were analyzed. Using non-transformed data, statistically significant, but weak relationships, between the LVI scores and the primary trophic st ate indicators TP, TN, chl, or Secchi depth were found (Figures 3-7 to 3-10). R2 values for the major tr ophic state indicators were 0.33, 0.10, 0.05, and 0.13, respectively. Transfo rming (log base 10) the data did not improve the relationships with TP and Secchi depth, but slightly strengthened the relationships with TN and chl (R2 values of 0.12 and 0.11, respectively). Analyses of fish community metric s (dependent variable) and LVI scores (independent variable) provided evidenc e for only weak relationships (R2 <0.25) between the parameters (Figur es 3-11 through 3-16). Ther e were no statistically significant relationships between the LVI scores and Simpson’s D, Simpson’s Evenness, sportfish relative biomass, and nonnative species relative biomass (Figures 3-11 through 3-16). The only significant re lationships were between LVI scores and species richness and the Shannon-Weiner dive rsity index (Figures 3-15 and 3-16). The correlation coefficients for species richness and Shannon-Weiner were r = -0.5 and r = 0.37, respectively. The Welch’s two-tailed t-test showed no statistically significant differences in LVI scores for lakes with and without non-native fish present. To assess the relationship between the LVI scores and human disturbance on the 20 lakes sampled in 2008, an ANOVA demons trated significant differences for the LVI scores among the three human di sturbance classifications used in this study (Figure 3-17). Lakes with higher LVI scores, as a group, had less human disturbance than lakes with low LVI scores. However, t here was considerable overlap (LVI scores

PAGE 25

25 between 50 and 70) in the scores for the three human disturbance classifications (Figure 3-17).

PAGE 26

26 0 10 20 30 40 50 60 70 80 90 100 020406080100 FDEPLAKEWAT C LVI score E Lake Linear (1:1 Line) Linear (LVI score) Figure 3-1. Linear regression of LAKEWAT CH and FDEP calculated LVI scores with E Lake included 0 10 20 30 40 50 60 70 80 90 100 020406080100 FDEPLAKEWAT C LVI score Linear (1:1 Line) Linear (LVI score) Figure 3-2. Linear regression of LAKEW ATCH and FDEP calculated LVI scores with E Lake excluded.

PAGE 27

27 0 10 20 30 40 50 60 70 80 90 100 020406080100 FDEP PNSLAKEWATCH P Percent Native species. Linear (1:1 line) Linear (Percent Native species.) Figure 3-3. Linear regression of percent nat ive species (PNS) as calculated by FDEP and LAKEWATCH. 0 10 20 30 40 50 60 70 80 90 100 020406080100 FDEP PISLAKEWATCH P Percent Invasive species. Linear (1:1 line) Linear (Percent Invasive species.) Figure 3-4. Linear regression of percent invasive species (PIS) as calculated by FDEP and LAKEWATCH

PAGE 28

28 0 10 20 30 40 50 60 70 80 90 100 020406080100 FDEP PSSLAKEWATCH P S Percent Sensitive species. Linear (1:1 line) Linear (Percent Sensitive species.) Figure 3-5. Linear regression of percent s ensitive species (PSS) as calculated by FDEP and LAKEWATCH 0 1 2 3 4 5 6 7 8 9 10 0246810 FDEP DCCLAKEWATCH DC Dominant Coefficient of Conservatism Linear (1:1 line) Linear (Dominant Coefficient of Conservatism) Figure 3-6. Linear regression of dominant species coefficient of conservatism (DCC) as calculated by FDEP and LAKEWATCH

PAGE 29

29 -20 0 20 40 60 80 100 120 140 020406080100 LVI scoreT P Figure 3-7. Linear regression of LAKEW ATCH LVI score against long-term total phosphorus (g/L). 0 500 1000 1500 2000 2500 3000 3500 4000 020406080100 LVI scoreT N Figure 3-8. Linear regression of LAKEW ATCH LVI score against long-term total nitrogen (g/L).

PAGE 30

30 0 20 40 60 80 100 120 140 160 020406080100 LVI scoreCHLa Figure 3-9. Linear regression of LAKEWAT CH LVI score against long-term chlorophyll concentration (g/L). 0 2 4 6 8 10 12 14 16 18 20 020406080100 LVI scoreSecchi ( ft Figure 3-10. Linear regression LAKEWATCH LVI score against long-term Secchi depth (ft).

PAGE 31

31 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0102030405060708090100 LVI D Figure 3-11. Linear regression of LVI sco res against Simpson’s Diversity Index D. 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0102030405060708090100 LVIEvenne s Figure 3-12. Linear regression of LVI scores against Simpson’s Evenness values.

PAGE 32

32 0 20 40 60 80 100 120 0102030405060708090100 LVIRelative Biom a Figure 3-13. Linear regression of LVI sco res against sportfish relative biomass. 0 10 20 30 40 50 60 0102030405060708090100 LVIRelative Biom a Figure 3-14. Linear regression of LVI scores against non-native species relative biomass.

PAGE 33

33 0 2 4 6 8 10 12 14 16 18 20 0102030405060708090100 LVISpecies Richn e Figure 3-15. Linear regression of LVI scores against species richness. 0 0.5 1 1.5 2 2.5 0102030405060708090100 LVIShannon-Wei n Figure 3-16. Linear regression of LVI sco res against Shannon-Weiner Diversity Index values.

PAGE 34

34 0 10 20 30 40 50 60 70 80 90LVI sco r Category 1 Category 2 Category 3 Figure 3-17. Plot of LVI scores of the various human disturbance levels.

PAGE 35

35 Table 3-1. Percent agreement before and after exclusion of plants not considered by LAKEWATCH as aquatic species. Lake County % agreement before % agreement after Apopka Orange 73 80 Cherry Lake 51 59 Crescent Putnam 60 69 Dorr Lake 68 70 Eloise Polk 53 55 George Putnam 59 62 Griffin Lake 66 71 Harris Lake 70 74 June Highlands 63 70 Juniper Walton 52 61 Mill Dam Marion 61 81 Minneola Lake 36 42 Sampson Bradford 68 70 Spring Walton 59 69 Tarpon Pinellas 66 72 Tohopekaliga East Osceola 72 75 Wauberg Alachua 63 67 Weir Marion 57 61 Wildcat Lake 62 69

PAGE 36

36 Table 3-2. Coefficients of variation (%) for scores calculated using the LAKEWATCH method, with number of samp ling events (N) included. Lake County N lakes CV Alligator Osceola 3 12 Apopka Orange 2 6 Butler Orange 5 17 Cherry Lake 4 6 Conway North Orange 5 23 Conway South Orange 5 19 Crescent Putnam 2 8 Dexter Polk 4 25 Dorr Lake 3 9 E Miami-Dade 4 15 Eloise Polk 2 4 Farm 13 Indian River 4 51 George Putnam 2 1 Grasshopper Lake 4 14 Griffin Lake 2 20 Harris Lake 3 5 Istokpoga Highlands 5 33 Ivanhoe East Orange 5 25 Ivanhoe Middle Orange 5 18 Ivanhoe West Orange 5 13 Johns Orange 4 25 Josephine Center Highlands 5 26 Josephine East Highlands 5 15 Josephine West Highlands 5 23 June Highlands 5 11 Juniper East Walton 5 15 Juniper West Walton 5 8 Kissimmee Osceola 4 7 Lochloosa Alachua 5 11 Mill Dam Marion 5 10 Minneola Lake 3 17 Orange Alachua 3 6 Panasoffkee Sumter 3 13 Sampson Bradford 3 10 Santa Fe Alachua 4 18 Sellers Lake 4 6 Spring Walton 5 9 Starke Orange 5 17 Stick Marsh Indian River 3 61 Tarpon Pinellas 2 5 Tohopekaliga Osceola 3 23 Tohopekaliga East Osceola 5 8 Wauberg Alachua 5 18 Weir Marion 4 15 Weohyakapka Polk 5 13 Wildcat Lake 5 9 Wilson Hillsborough 3 25

PAGE 37

37 Table 3-3. Lakes used to compare long-te rm total phosphorus (TP) total nitrogen (TN), chlorophyll (chl), and Secchi depth to LVI. Lake County Mean TP (g/L) Mean TN (g/L) Mean CHL (g/L) Mean SECCHI (ft) Alligator Osceola 146394 5 Apopka Orange 833650111 1 Butler Orange 145683 12 Cherry Lake 159286 6 Conway North Orange 104446 13 Conway South Orange 103895 14 Crescent Putnam 83146852 2 Dead Gulf 1770213 4 Deer Point Bay 72452 7 Dexter Polk 104453 13 Dorr Lake 1747611 3 E Miami-Dade 53472 18 Eloise Polk 36130144 3 Farm 13 Indian River 130178145 2 George Putnam 58123343 2 Grasshopper Lake 54593 7 Griffin Lake 663042140 1 Harris Lake 35177659 2 Istokpoga Highlands 56132538 3 Ivanhoe East Orange 2773928 4 Ivanhoe Middle Orange 2864025 5 Ivanhoe West Orange 3163229 4 Johns Orange 41107515 4 Josephine Center Highlands 68103525 2 Josephine East Highlands 4999135 2 Josephine West Highlands 102106023 2 June Highlands 1356310 7 Juniper Walton 114936 7 Kissimmee Osceola 53130333 3 Lochloosa Alachua 70220993 2 Mill Dam Marion 125034 8 Minneola Lake 2610966 4 Orange Alachua 75176152 3 Panasoffkee Sumter 3279814 4 Poinsett Brevard 108231922 1 Sampson Bradford 237047 5 Santa Fe Alachua 114657 7 Sellers Lake 31022 18 Spring Walton 145179 7 Starke Orange 2699924 3 Stick Marsh Indian River 123178449 2 Talquin Gadsden 5581137 3 Tarpon Pinellas 36110043 2 Tohopekaliga Osceola 51106530 3 Tohopekaliga East Osceola 216725 6 Wauberg Alachua 126191997 2

PAGE 38

38 Table 3-3. cont. Lake County Mean TP (g/L) Mean TN (g/L) Mean CHL (g/L) Mean SECCHI (ft) Weir Marion 1175311 6 Weohyakapka Polk 2371211 5 Wildcat Lake 73264 8 Wilson Hillsborough 1879210 7

PAGE 39

39 Table 3-4. Florida lakes used to compare fi sh community metrics to LVI scores, along with electrofishing and LVI score years used in analysis. County Lake Year of Electrofishing Year of LVI Osceola Alligator 2007 2007 Orange Butler 2007 2007 Lake Cherry 2007 2007 Orange Conway 2007 2007 Polk Dexter 2008 2007 Lake Dorr 2007 2007 Miami-Dade E 2008 2008 Lake Grasshopper 2007 2007 Highlands Istokpoga 2005 2005 Orange Ivanhoe 2007 2007 Orange John's 2007 2007 Highlands Josephine 2007 2007 Highlands June 2008 2008 Walton Juniper 2008 2008 Osceola Kissimmee 2007 2007 Alachua Lochloosa 2006 2007 Marion Mill Dam 2008 2008 Alachua Orange 2006 2007 Sumter Panasoffkee 2007 2007 Alachua Santa Fe 2006 2007 Lake Sellers 2007 2007 Walton Spring 2008 2008 Orange Starke 2007 2007 Indian River Stick Marsh 2006 2007 Osceola Tohopekaliga 2005 2005 Osceola Tohopekaliga East 2008 2008 Alachua Wauberg 2008 2008 Marion Weir 2008 2008 Polk Weohyakapka 2006 2007 Lake Wildcat 2008 2008 Hillsborough Wilson 2007 2007

PAGE 40

40 CHAPTER 4 DISCUSSION The Clean Water Act requires protection of t he biological integrity of waters in the United States; therefore dev elopment of an adaptable, r obust method to assess the impact of human disturbance is needed. FD EP attempted to address the problem in Florida by developing the LVI (Fore 2005). The results from this study demonstrate that the LVI is adaptable to the extent that aquatic macrophyte information collected by another group can be used to calculate com parable LVI scores. For example, LVI scores (calculated using macrophyte data co llected via the LAKEWATCH protocol) for the 20 lakes sampled in this study in the summer of 2008 were significantly correlated (R2 =0.69) to LVI scores calculated using FD EP protocol (Figure 3-1), and there was a nearly 1:1 relationship between scores calcul ated using each protocol. Based on this result, it can be reasonably concluded that LAKEWATCH plant sampling data can be used to calculate LVI scores that are comparable to LVI scores calculated using the FDEP protocol. The robustness of an LVI score for an indi vidual lake has uncertainty associated with the score. When regression analysis wa s used to compare LVIs calculated using LAKEWATCH and FDEP protocols, 31% of the variation could not be explained by the model. This variance was reduced by 3% w hen an artificial lake created by limestone mining (E Lake, Miami-Dade County) was excluded from the analysis, but excluding lake types reduces the practical use of the LVI in lake assessment. Whenever different methodologies are used, differences can arise. Factors contributing to these differenc es must be considered. Afte r examining other sources of variation for the lakes used in this study, it is evident that a user of the LVI needs to

PAGE 41

41 understand how the LVI is calc ulated before interpreting the score. For example, both the LAKEWATCH and FDEP protocols identifi ed similar numbers of plant species, and neither protocol routinely mi ssed a particular plant type. However, there was a lack of complete agreement (only 61%) for the plant species identified by each protocol because some species that are considered a quatic by FDEP are not considered aquatic by LAKEWATCH (e.g., dog fennel, Eupatorium leprophyllum ). There are also methodological differences for each protocol in the calculation of the four metrics used in the LVI (percent native species, percent in vasive species, percent sensitive species, and dominant species coefficient of conser vatism). These differences, however, contributed little to the observed differences in LVIs calculated using the FDEP and LAKEWATCH protocols. This again suggests that LAKEWATCH plant sampling data can be used to calculate LVI scores comparable to scores calculated using the FDEP protocol. Rather than focusing on methodology of t he LVI protocol, potential users may have a greater problem associated with the te mporal variation in the index. This variation can be caused by natural envir onmental factors or plant management activities. For the lakes used in this study, the coefficient of variation for LVI scores calculated for individual lakes was about 16%, but was as high as 60%. This type of variation is not unexpected because the index is merely a snap shot of the aquatic plant community of a lake, which varies due to c limate conditions (e.g., drought) or natural disasters (hurricanes). For example, many sedges (e.g., plants in the genera Cyperus Rhynchospora etc.), that would not be able to establish under high water conditions could become prominent on a lake during dr ought, and could therefore affect the LVI

PAGE 42

42 score, depending on the species that establis h under these conditions. Aquatic plant management practices can also change the LV I score considerably for an individual lake. For example, Lake Tohope kaliga (Osceola County, Florida) is a large lake that is the subject of intensive hydr illa management (W. Haller, Univer sity of Florida, October 2008, pers. comm.). This lake had a series of LVI scores consisting of: 55 (2000), 34 (2005), and 51 (2007). The large change bet ween 2005 and 2007 was the result of differences in the abundance of hydrilla, an invasive aquatic plant, which was the dominant species in 2005 surveys, but not in 2007, following a major herbicide treatment. This is an ex ample of one potential problem in the index, where management activities such as invasive pl ant management can drastically affect the outcome of an LVI score, thus requiring coordination among m anagement agencies if the LVI is to be used to a ssess biotic integrity. While the methodological and tempor al problems call into question the practicality of the LVI, t he overall utility of the LVI for assessing human impacts on Florida lakes must also be questioned. For example, eutrophication (e.g., phosphorus enrichment) has been identified as a major pr oblem by FDEP. There, however, were only weak relationships between LVI sco res and long-term wate r chemistry, the strongest relationship being l ong-term TP concentrations (R2=0.33). However, Bachmann et al. (2009) demonstrated pH is a keystone environmental factor influencing the relationships, highlighting the problem of separ ating natural factor s (i.e., naturally low pH causing naturally low TP concentrati ons) from human impac ts. Other factors, not related to human disturbance, may also be influencing water chemistry. For example, internal loading was suggested to be the cause of increased TP levels in Lake

PAGE 43

43 Okeechobee (Florida) by Canfield and Hoye r (1988b). Canfield and Hoyer (1988a) showed that the mineral composition and trophi c states of Florida lakes are strongly related to their physiograpical region’s geology, and water chemistry affected the assemblage of plant species present in a waterbody (Hoyer et al. 1996). It could therefore be inferred that since water chemis try affects the plant a ssemblages of lakes, it is therefore affecting the LVI scores. Th is suggests that the location of a lake and the chemistry of its underlying soils may be having more of an affect than human disturbance on the lake’s long-term water c hemistry. When taki ng the findings of Bachmann et al. (2009) and the effect of physi ographic regions into consideration, along with the high variability associated with all of the LVI/water chemistry relationships, the practicality of the LVI is questionable. Bachmann et al. (2009) did not, however, examine the relationship between LVI scores and biological communities (except for chlorophyll) of lakes. The Clean Water Act specifically requires the pr otection of the biological int egrity of waters in the United States to insure that those waters can s upport the protection and propag ation of fish. In this study, relationships between various fi sh community metrics and LVI scores were weak. The only statistically significant relationships were between LVI scores and species richness and Shannon-Weiner Diversity Index values. In contrast to what would be expected (i.e., reduced species richness at lower LVI scores, which supposedly indicates more human disturbance ) LVI scores were inversely related to species richness (R2 = 0.25). The R2 value for the regre ssion comparing ShannonWeiner values to LVI scores was small (R2 = 0.25), suggesting that most of the variation

PAGE 44

44 could not be explained by the model. C onsequently, the LVI approach has little merit for assessing the impact of anthropogenic activities on fish communites. A two-tailed t-test failed to show diffe rences in LVI scores of lakes with and without non-native fish species. The presenc e of non-native fish species is generally associated with human action, so it would be expected that the LVI should show a difference between scores of lakes with and wit hout these species. This finding and the overall inability to relate the LVI to different fish metrics should not be surprising given that Schulz, et al. (1999) found no relati onship between human disturbance and an IBI that used fish assemblages as m easures of human disturbance. The LVI was developed as a different approach to gauge human disturbance (using the plant community of a lake to ca lculate LVI scores). Based on the available evidence for water chemistry and fish assemb lages, it must be concluded that the LVI, as a management tool, shows little potential for us e in Florida. The lack of relationships to water chemistry and fish communities renders it an impractical assessment or management tool. Its use as an index of human disturbance for Florida lakes, therefore, should be reconsider ed by FDEP. In conclusion, the LVI is a questionable way of examining the ecological condition of Florida lakes.

PAGE 45

45 LIST OF REFERENCES American Public Health Association ( APHA). 2005. Standard methods for the examination of wate r and wastewater. 21st Edition. American Public Health Association. Washi ngton, D.C., U.S.A. Andreas, B. K., and R. W. Lichvar. 1995. Floristic index for establishing assessment standards: a case study for northern Ohio. U.S. Army Corps of Engineers Waterways Experiment Station, Vicksburg, MS, USA. Technical Report WRP-DE-8. Bachmann, R. W., et al. 2009. Comment to the Florida Department of Environmental Protection. Bachmann, R. W., Horsburgh, C. A., Hoyer, M. V., et al. 2002. Relations between trophic state indicators and plant biomass in Florida lakes. Hydrobiologia, 470 219234. Bourdaghs, M., Johnston, C. A, and R. R. R egal. 2006. Properties and performance of the Floristic Quality Index in Great Lakes coastal wetlands. Wetlands, 26 (3), 718-735. Bowling, L. M, Steane, M., and P. Tyler. 1986. Spectral distribution and attenuation of underwater irradiance in Tasmanian inland waters. Freshwater Biology, 16 313-335. Canfield Jr., D. E. and M. V. Hoyer. 1988a. Regional geology and trophic state characteristics of Florida lakes. Lake and Reservoir Management, 4 (1), 21-31. Canfield Jr., D. E. and M. V. Hoyer. 1988b. The eutr ophication of Lake Okeechobee. Lake and Reservoir Management, 4 (2), 91-99. Carpenter, S. R., Caraco, N. F., R. W. Correll., et al. 1998. Nonpoint pollution of surface waters with phosphorus and nitrogen. Ecological Applications, 8 (3), 559-568. Cohen, M. J., Carstenn, S., and C. R. Lane. 2004. Floristic quality indices for biotic assessment of depressional mars h condition in Florida. Ecological Applications, 14 784-794. D’Elia, C. F., Steudler, P. A ., and N. Corwin. 1977. Determination of total nitrogen in aqueous samples using persulfate digestion. Limnology and Oceanography, 22 (4), 760-764. Duigan, C., Kovach, W., and M. Palmer. 200 7. Vegetation communities of British lakes: a revised classification scheme for conservation. Aquatic Conservation: Marine and Freshwater Ecosystems, 17 147-173. Florida LAKEWATCH. 2007. Long-term fish, plants, a nd water quality monitoring program: 2006-2007 data. Department of Fisher ies and Aquatic Sciences, University of

PAGE 46

46 Florida/Institute of Food and Agricultural Scie nces. Library, University of Florida. Gainesville, Florida. Fore, L. S. 2005. Assessing the biological c ondition of Florida lakes: development of the Lake Vegetation Index (LVI). Florida Depa rtment of Environm ental Protection. Tallahassee, FL. Forsberg, C., and S. O. Rydi ng. 1980. Eutrophication parameters and trophic state indices in 30 Swedish wa ste-receiving lakes. Hydrobiologia, 89 189-207. Hoyer, M. V., Brown, C. D., and D. E. Canfie ld. 2004. Relations between water chemistry and water quality as defined by la ke users in Florida. Lake and Reservoir Management, 20 (3), 240248. Hoyer, M. V., Canfield, Jr., D. E., Horsburgh C. A., and K. Brown. 1996. Florida Freshwater Plants. A Handbook of Common Aquatic Plants in Flor ida Lakes. University of Florida. Institute of Food and Agricultural Scie nces. Gainesville, Florida. Krebs, C. J. 1999. Ecological Methodology. Benjamin/Cummings, Menlo Park, California. Matthews, J. W. 2003. Asse ssment of the floristic quality assessment index for use in Illinois, USA, wetlands. Natural Areas Journal, 23 53-60. Menzel, D. W., and N. Corwin. The measurement of total phosphorus in seawater based on the liberation of or ganically bound fractions by persulfate oxidation. Limnology and Oceanography, 10 (2), 280-282. Murphy, J., and J. P. Riley. 1962. A modified single soluti on for the determination of phosphate in natural waters. Analytica Chimica Acta, 27 (30). Nichols, S., Weber, S., and B. Shaw. 2000. A proposed aquatic plant community biotic index for Wisconsin Lakes. Environmental Management, 26 (5), 491-502. Sartory, D. P., and J. U. Grobbelaar. 1984. Extraction of chlorophyll a from freshwater phytoplankton for spectrophotometric analysis. Hydrobiologia, 114 117-187. Schaeffer, B. 2007. Personal communication. Florida Fish and Wildlife Conservation Commission. Tallahassee, Florida. Schulz, E. J., Hoyer, M. V., 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.

PAGE 47

47 Simal, J., Lage, M. A., and I. Iglesias. 1985. Second derivative ultraviolet spectroscopy and sulfamic acid method for determi nation of nitrates in water. Journal of Analytical Chemistry, 68 962-964. Stelzer, D., Schneider, S., and A. Melzer. 2005. Macrophyte-based assessment of lakes-a contribution to the implementation of the European Water Framework Directive in Germany. United States Environmental Protection Agency. 1998. Lake and Reservoir Bioassessment and Biocriteria Technical Guidance Document. Wollin, K. M. 1987. Nitrate determination in surfac e waters as an example of the application of UV derivative spectr oscopy to environmental analysis. Acta Hydrochimica et Hydrobiologia, 15 (5), 459-469.

PAGE 48

48 BIOGRAPHICAL SKETCH Eric Flynt Thomas (also known as Bubba) was born in Gainesville, Florida. He was raised by his mother and stepfather, Stephen and Rebecca Parker. Thomas was raised in Lake Butler, Florida, and was heavily influenced by his uncle (Alray Harvey), his grandmother (Annie Harvey), and his gra ndfather (Alfred Harvey). Thomas was taught from an early age a respect for nature and learned a compassion for the outdoors that translated into his education a nd career. After gr aduating from Union County High School in 2003, Thomas a ttended Lake City Community College, and received an Associate in Arts degree in 2005. He then transferred to the University of Florida, where he received a bachelor’s degree in wildlife ecology and conservation in 2007. He entered the University of Florida graduate school in 2007 in the Program for Fisheries and Aquatic Sciences, and complet ed the course requirements in May 2009. Thomas moved to Pinedale, Wyoming to work as a fisheries technician while completing his thesis, which he submitted in the fall of 2009.