xml version 1.0 encoding UTF-8 standalone no
record xmlns http:www.loc.govMARC21slim xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.loc.govstandardsmarcxmlschemaMARC21slim.xsd
leader nam Ka
controlfield tag 001 002021461
007 cr mnu|||uuuuu
008 090729s2008 flu s 000 0 eng d
datafield ind1 8 ind2 024
subfield code a E14-SFE0002586
Madrian, Max Jacobo Moreno.
Eutrophication trend of lakes in the Tampa Bay watershed and the role of submerged aquatic vegetation in buffering lake water phosphorus concentration
h [electronic resource] /
by Max Max Jacobo Moreno Madrian.
[Tampa, Fla] :
b University of South Florida,
Title from PDF of title page.
Document formatted into pages; contains 137 pages.
Dissertation (Ph.D.)--University of South Florida, 2008.
Includes bibliographical references.
Text (Electronic dissertation) in PDF format.
Mode of access: World Wide Web.
System requirements: World Wide Web browser and PDF reader.
Co-advisor: Noreen D. Poor, Ph.D.
Co-advisor: Mark Ross, Ph.D.
ABSTRACT: Twentieth century human settlement within the Tampa Bay watershed was linked to a dramatic mid-century decline in bay water quality and loss of seagrass acreage. Decades of direct and indirect nutrient discharges to the bay from phosphorus mining, fertilizer manufacturing, and wastewater treatment, as examples, impaired the estuary. In the past twenty years, regional stakeholders have worked to improve the bay water quality by reducing point and non-point source nutrient loading to the bay. Lakes within the Tampa Bay watershed may play an important role in attenuating the flow of nutrients into the bay. This study hypothesized that between 1990 and 2007 lake water concentrations of total phosphorus (TP) and chlorophyll-, as well as the ratio of total nitrogen to total phosphorus (TN:TP), have changed for selected lakes in the Tampa Bay watershed.During this period, the watershed underwent a rapid shift in land use as groves and farms became shopping malls and new homes. A two-way analysis of variance (ANOVA) revealed that for 10 lakes clustered in the northern portion of the Tampa Bay watershed and classified as oligotrophic or mesotrophic, observed increases in water concentrations of TP and chlorophyll- were statistically significant. For 6 lakes classified as hypereutrophic and scattered across the watershed, observed decreases in water TP concentrations were statistically significant, while chlorophyll- concentrations did not change. For both groups of lakes, the TN:TP ratio declined significantly; however, oligotrophic and mesotrophic lakes were phosphorus-limited but hypereutrophic lakes were nitrogen-limited, based on this ratio.A second hypothesis of this study was that lake water concentrations of TP, total nitrogen (TN) and chlorophyll- were lower in lakes that had more coverage of submerged aquatic vegetation, as vegetation suppresses re-suspension of sediments and is a reservoir for nitrogen and phosphorus and a surface for biofilms. The results of a one-way ANOVA showed that for 34 lakes within the Tampa Bay watershed, lakes with a greater than 20 percent volume infested by macrophytes (PVI), water concentrations of TP and chlorophyll- but not TN were statistically lower than for lakes with a less than 20 PVI.
x Environmental and Occupational Health
t USF Electronic Theses and Dissertations.
Eutrophication Trend of Lakes in the Tampa Bay Watershed and the Role of Submerged Aquatic Vegetation in Buffering Lake Water Phosphorus Concentration by Max Jacobo Moreno Madrian A dissertation submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy Department of Environmental and Occupational Health College of Public Health University of South Florida Co-Major Professor: Noreen D. Poor, Ph.D. Co-Major Professor: Mark Ross, Ph.D. Connie Mizak, Ph.D. Amy Stuart, Ph.D. Date of Approval: July 1, 2008 Keywords: aquatic macrophytes, development, nutrient enrichment, land use, phytoplankton, population, Copyright 2008, Max Jacobo Moreno Madrian
DEDICATION This dissertation is dedicated to my son, Rafael Maximiliano MorenoSchiaffino, an instrument of the LORD. Through his innoc ence, purity, and love I have been given the inspirati on, strength and faith to complete my doctoral degree.
ACKNOWLEDGEMENTS First and foremost I thank GOD for arranging all the circumstances of my life in such a way that I could meet the wonderful professors, advisors, coworkers, family and friends that have m ade unique and important contributions to the achievement of this goal. I want to so lemnly express my deep gratitude to my co-major professor, Dr. Noreen Poor, who offered me knowledge and support through-out my entire academ ic program and research. I thank my committee members for their insight and guidance. I would also like to thank Dr. Jim Griffin and David Eilers for their assistance in data management, and Dr. Lar ge and Dr. Mbah for their assistance in statistical analysis. I acknowledge t he support provided by Dr. Bernard and the entire Department of Envi ronmental and Occupational Health: they created the working conditions that were essential to the accomplishment of this goal. My friend Ligia Cruz with her strength, t enacity and positive attitude, provided me with faith and encouragement in my pursuit. Other friends including Monica Gray, and Ellen McCreedy, supported and encouraged me with their friendship. Most importantly, I thank my fam ily. Their love, education, and great example gave me the foundations to successfully undertake this and all challenges in my life. To my sister, nephews, brother in law; and the inspiration of my son, as well as all the people ment ioned here and some others, I am forever thankful.
i TABLE OF CONTENTS LIST OF TABLES.................................................................................................iii LIST OF FIGURES...............................................................................................v LIST OF SYM BOLS............................................................................................ viii ABSTRACT..........................................................................................................ix CHAPTER 1. INTRO DUCTION..........................................................................1 1.1 Purpose of the Study...........................................................................1 1.2 Problem Statement..............................................................................3 1.3 Research Hypothe ses..........................................................................7 CHAPTER 2. LITERATURE REVIEW..............................................................10 2.1 Introdu ction........................................................................................10 2.2 Inorganic Phosphorus........................................................................14 2.3 Organic Phosphor us..........................................................................15 2.4 Oxygen..............................................................................................16 2.5 Submerged Aquat ic Vegetation.........................................................18 2.6 Li ght...................................................................................................23 2.7 Mean Depth and Temperature...........................................................26 2.8 Bacteria and A quatic An imals............................................................27 2.9 Land Use / Land Cove r......................................................................29 2.10 Public Health Implic ations................................................................32 2.11 Summary of Phos phorus Cycl e Revi ew...........................................34 2.12 Modeli ng Revi ew..............................................................................35 2.12.1 Interaction between Sediments and Water Column.........39 2.12.2 Interaction between Sediments and Submerged Aquatic Ve getation...........................................................41 2.12.3 Interaction bet ween Submerged Aquatic Vegetation and Wa ter Column.........................................42 2.12.4 Historical Land Us e and Trend of Population Growth.............................................................................43 2.12.5 Some Parameters Found in Li terature.............................46 2.13 Discu ssion.......................................................................................48 CHAPTER 3. TEMPORAL TRENDS IN LAKE WATER CONCENTRATION OF TOTAL PHOSPHORUS, RATIO OF TOTAL NITROGEN TO TOTAL PHOSPHORUS,
ii AND CHLOROPHYLLFOR LAKES OF DIFFERENT EUTROPHICATION ESTATUS IN TAMPA BAY WATERSHED: 1990-2007.........................................................52 3.1 Introdu ction........................................................................................52 3.2 Objectives and Hypot heses...............................................................53 3.3 Me thods.............................................................................................54 3.3.1 Data Gatheri ng and Sampli ng Met hods.............................54 3.3.2 Statisti cal Anal ysis.............................................................56 3.4 Results and Discu ssion......................................................................62 3.5 Summary...........................................................................................84 CHAPTER 4. IDENTIFICATION OF IMPORTANT VARIABLES AFFECTING WATER QUALITY IN LAKES OF THE TAMPA BAY WA TERSHED.......................................................86 4.1 Introdu ction........................................................................................86 4.2 Me thods.............................................................................................87 4.2.1 Data Sa mpling Pr ogram....................................................87 4.2.2 Statisti cal Anal ysis.............................................................91 4.3 Results and Discu ssion......................................................................91 4.4 Summary...........................................................................................97 CHAPTER 5. EFFECT OF SUBMERGED AQUATIC VEGETATION ON WATER TOTAL PHOSPHORUS CONCENTRATION IN LAKES OF THE TAMPA BAY WATERS HED............................99 5.1 Introdu ction........................................................................................99 5.2 Objectives and Hypot hesis..............................................................101 5.3 Me thods...........................................................................................103 5.3.1 Data Descrip tion..............................................................103 5.3.2 Statisti cal Anal ysis...........................................................103 5.4 Results and Discu ssion....................................................................104 5.5 Summary.........................................................................................112 CHAPTER 6. IMPLICATIONS OF RESE ARCH.............................................113 REFERENC ES.................................................................................................117 APPENDICES..................................................................................................132 Appendix A: General Informa tion about Lakes Examined for Eutrophication Trends .....................................................133 Appendix B: One Time Values of Variables of Lakes Examined for Submerged Vegetationa..............................................135 ABOUT THE AUTH OR...........................................................................E nd Page
iii LIST OF TABLES Table 2.1 Imortant par ameters measured and considered in phosphorus (P) cycling models of shallow aquatic systems.........................................................................................38 Table 2.2 Percent of particulate and dissolved phosphorus content in water column in shallow aquatic systems.....................39 Table 2.3 Standard rates in phosphorus cycling of shallow aquatic syst ems............................................................................47 Table 3.1 Summary statisti cs for annual TP concentration in g L-1 collected by LAKEWATCH fr om 10 oligotrophic and mesotrophic lakes in two counties from 1990 through 2007..............................................................................................64 Table 3.2 Summary statisti cs for annual TP concentration in g L-1 collected by LAKEWATCH from 6 hypereutrophic lakes in three counties from 1990 th rough 2007........................................67 Table 3.3 Summary statisti cs for the annual rati o of TN:TP ratio collected by LAKEWATCH fr om 10 oligotrophic and mesotrophic lakes in two counties from 1990 through 2007..............................................................................................70 Table 3.4 Summary statisti cs for annual TN:TP ratio collected by LAKEWATCH from 5 hypereutr ophic lakes in three counties from 1990 thr ough 2007..................................................73 Table 3.5 Summary statistics for chlorophyllin g L-1 collected by LAKEWATCH from 8 oligotrophic and mesotrophic lakes in two counties from 1990 th rough 2007..............................74 Table 3.6 Summary statistics for annual Chlorophillconcentration in g L-1 collected by LAKEWATCH from 5 hypereutrophic lakes in three counties from 1990 through 2007.................................................................................75
ivTable 4.1 Summary statis tics of Hillsborough lakes examined for association between lake vari ables...............................................96 Table 4.2 Matrix table s howing the analysis of correlation between lake variables. Values in dark correspond to resulting significant associations bet ween trophic state parameters (TN, TP, chlorophyll) and theirs factors controlling for water quality at 95% conf idenc e....................................................96 Table 5.1 Summary table fo r PVI values of lakes with macrophytedominance and phytoplank ton-domi nance..................................105 Table 5.2 Summary table for PAC values of lakes macrophytedominance and phytoplank ton-domi nance..................................105 Table 5.3 Summary table for TP values (g p L-1) for lakes with macrophyte-dominance and phy toplankton-dom inance..............107 Table 5.4 Summary table for chlorophyllvalues (g L-1) for lakes with macrophyte-dominance and phytoplanktondominance..................................................................................107 Table 5.5 Summa ry table for TN (mg L-1) values of lakes with macrophyte-dominance and phy toplankton-dom inance..............108
v LIST OF FIGURES Figure 1.1 Map of the Tampa Bay watershed and major bay segments........................................................................................2 Figure 1.2 Land Use in Tampa Bay watershed (Tampa Bay Estuary Program, 1999)...............................................................................6 Figure 2.1 Total phosphorus cycle in shallow aquatic systems.......................13 Figure 3.1 Trophic State Cla ssification System (Forsberg and Ryding, 1980)................................................................................60 Figure 3.2 Lakes analyzed for tr ends in trophic state variables during the time peri od 1990 to 2007........................................................61 Figure 3.3 Plot of TP concentration for 10 oligotrophic and mesotrophic lakes.........................................................................65 Figure 3.4 Change in TP concentration in 10 oligotrophic and mesotrophic lakes betw een periods 1 and 2.................................66 Figure 3.5 Plot of TP conc entration for 6 hyper eutrophic lakes......................68 Figure 3.6 Change in TP concentration in 6 hipereutrophic lakes between periods 1 and 2...............................................................68 Figure 3.7 Plot of TN:TP ratio for 10 oligotrophic and mesotrophic lakes..............................................................................................71 Figure 3.8 Change in the TN:TP ratio in 10 oligotrophic and mesotrophic lakes betw een periods 1 and 2.................................72 Figure 3.9 Plot of TN:TP ratio for 5 hyper eutrophic lakes...............................73 Figure 3.10 Change in the TN:TP ratio in 5 hypereutrophic lakes between periods 1 and 2...............................................................74 Figure 3.11 Plot of chlorophyllconcentration for 8 oligotrophic and mesotrophic lakes.........................................................................75
vi Figure 3.12 Change in chlorophyllconcentration in 8 oligotrophic and mesotrophic lakes betw een periods 1 and 2..........................76 Figure 3.13 Plot of chlorophyllconcentration for 5 hypereutrophic lakes..............................................................................................77 Figure 3.14 Change in chlorophyllconcentration in 5 hypereutrophic lakes between per iods 1 and 2. .....................................................77 Figure 3.15 Population growth estimates in the Tampa Bay Metropolitan Area fr om 1990 unt il 2 006........................................78 Figure 3.16 Population growth es timates in Hillsborough County from 1990 until 2006..............................................................................78 Figure 3.17 Change in annual averages of lake water TP concentration over time in years in oligotrophic and mesotrophic lakes of t he Tampa Bay waters hed...........................80 Figure 3.18 Change in annual averages of the TN:TP over time in years in oligotrophic and meso trophic lakes of the Tampa Bay wate rshed..............................................................................80 Figure 3.19 Change in annual aver ages of lake water chlorophyllconcentration over time in years in oligotrophic and mesotrophic lakes of t he Tampa Bay waters hed...........................81 Figure 3.20 Change in annual averages of lake waterTP concentration over time in years in hypereutrophic lakes of the Tampa Bay wate rshed........................................................82 Figure 3.21 Change in annual averages of the TN:TP over time in years in hypereutrophic lakes of the Tampa Bay waters hed.....................................................................................82 Figure 3.22 Change in annual aver ages of lake water chlorophyllconcentration over time in years in hypereutrophic lakes of the Tampa Bay wate rshed........................................................83 Figure 4.1 Location of lakes with recent observations on submerged aquatic vegetation (Koeni g and Eilers 2006-2007).......................90 Figure 5.1 TP in macrophyte-domi nated la kes.............................................109
vii Figure 5.2 TP in phy toplankton-domi nated la kes..........................................109 Figure 5.3 Chlorophylla in macrophytedominated lakes.............................109 Figure 5.4 Chlorophylla in Phytoplanktondominated lakes.........................110 Figure 5.5 TN in macrophyte-domi nated la kes.............................................110 Figure 5.6 TN in Ph ytoplankton-domi nated lakes.........................................110 Figure 5.7 Variability of lake water total phosphorus with increment in area covered by veget ation.....................................................111 Figure 5.8 Variability of lake water total phosphorus with increment in volume of the lake infested by vegetat ion...............................111
viii LIST OF SYMBOLS cm centimeter EPA Envir onmental Protection Agency FDEP Florida Departm ent of Environmental Protection ha hectare L liter m meter m3 cubic meter mg milligram g microgram n number of lakes SAV S ubmerged aquatic vegetation PAC Per c ent area covered with vegetation PVI Per cent volume infested with vegetation TBEP Tam pa Bay Estuary Program TN Total nitrogen TN:TP Ratio of total nitrogen to total phosphorus TP Total phosphorus
ix EUTROPHICATION TREND OF LAKES IN THE TAMPA BAY WATERSHED AND THE ROLE OF SUBMERGED AQUA TIC VEGETATION IN BUFFERING PHOSPHORUS CONCENTRATION Max Jacobo Moreno Madrian ABSTRACT Twentieth century human settlement within the Tampa Bay watershed was linked to a dramatic mid-c entury decline in bay water quality and loss of seagrass acreage. Decades of direct and indirect nutrient discharges to the bay from phosphorus mining, fertilizer manufactu ring, and wastewater treatment, as examples, impaired the estuar y. In the past twenty y ears, regional stakeholders have worked to improve the bay water quality by reducing point and non-point source nutrient loading to the bay. Lakes within the Tampa Bay watershed may play an important role in attenuating the flow of nutrients into the bay. This study hypothesized that between 1990 and 2007 lake water concentra tions of total phosphorus (TP) and chlorophyll, as well as the ratio of total ni trogen to total phosphorus (TN:TP), have changed for selected lakes in the Tampa Bay watershed. During this period, the watershed underwent a rapid shift in land use as groves and farms became shopping malls and new homes. A two-way analysis of variance
x (ANOVA) revealed that for 10 lakes clus tered in the northern portion of the Tampa Bay watershed and classified as ol igotrophic or mesotrophic, observed increases in water concentrations of TP and chlorophyllwere statistically significant. For 6 lakes classified as hypereutrophic and scattered across the watershed, observed decreases in water TP concentrations were statistically significant, while chlorophyllconcentrations did not change. For both groups of lakes, the TN:TP ratio declined signi ficantly; however, oligotrophic and mesotrophic lakes were phosphorus-lim ited but hypereutrophic lakes were nitrogen-limited, based on this ratio. A second hypothesis of this study wa s that lake water concentrations of TP, total nitrogen (TN) and chlorophyllwere lower in lakes that had more coverage of submerged aquatic vegetat ion, as vegetation suppresses resuspension of sediments and is a reservoir for nitrogen and phosphorus and a surface for biofilms. The results of a one-way ANOVA showed that for 34 lakes within the Tampa Bay watershed, lakes with a greater than 20 percent volume infested by macrophytes (PVI), water concentrations of TP and chlorophyllbut not TN were statistically lower t han for lakes with a less than 20 PVI.
1 CHAPTER 1. INTRODUCTION 1.1 Purpose of the Study Twentieth century human settlement within the Tampa Bay watershed was linked to a dramatic mid-century decline in bay water quality and loss of seagrass acreage. Decades of direct and indirect nutrient discharges to the bay from phosphorus mining, fertilizer manuf acturing, and wastewat er treatment, as examples, impaired the estuar y. In the past twenty y ears, regional stakeholders have worked to improve the bay water quality by reducing point and non-point source nutrient loading to the bay (T BEP, 2006). Lakes within the Tampa Bay watershed may play an important role in a ttenuating the flow of nutrients into the bay, as these lakes ultimately discharge to the bay via canals, springs, creeks, streams, and rivers. Analyses of temporal trends and a possible role for submerged aquatic vegetation in reducing eutrophication benef it those involved in lake water management within the Tampa Bay watershed (Figure 1.1).
2Figure 1.1 Map of the Tampa Bay watershed and major bay segments (TBEP, 2006)
3 1.2 Problem Statement Excessive nutrient loading to surface waters threatens environmental and human health (FDEP, 2004). Nu trient enrichment of su rface wtaers can cause blooms of algae (Canfield, 1983) wit h the associated stench and stagnation (McCarthy, 2000). Sustained algal blooms can lower the water column dissolved oxygen concentration and light transparency, conditions which lead to a loss of biodiversity and an attending shift in dominant species (Paerl, 1988). Some species of blue-green algae (cyanobacteria) produce potent toxins. These toxins may enter the human body through ingestion inhalation, or dermal contact with water or with fish from water pollut ed with cyanobacteria (Fleming et al., 2002; Karjalainen et al., 2007). Thus, eutrophica tion may impair a lake to the extent that swimming or fishing is not advised (Hansson et al., 1999). The Florida Department of Environm ental Protection (FDEP, 1996; 2000; 2006) reports an increasing percentage of Florida lakes with a degrading trend in eutrophication status. Urban land use has been suggested as the category with the higher contribution of phosphorus to rece iving waters in Florida, followed by agricultural and forested land use (Reddy et al., 1999). For a lake, water concentrations of nitrogen and phosphorus are determined by a combination of (net) external loading and internal cycling (James and Bierman Jr., 1995; Kittiwani ch et al., 2006; Serpa et al., 2007). Phosphorus is most commonly the lim iting nutrient in freshwater systems (ChunLei et al., 2004; Schauser et al., 2004; Schindler, 1977; Scinto and Reddy,
4 2003; Zhou et al., 2001). Factors that control the cycling of phosphorus within a lake are discussed in Chapter 2. Land use (Figure 1.2) affect s the external loading of nutrients to a lake both through runoff (Johnes et al., 1996; Reddy et al., 1999; Soranno et al., 1996) and atmospheric deposition (Poor et al., 2005). The eastern part of the Tampa Bay watershed is within a phosphorus mining region known as Bone Valley (Brown, 2005), and lakes in that area may be influenced by naturally-occurring phosphorus minerals or by phosphorus mining and fertilizer manufacturing activities. Much of the eastern portion of the watershed is agricultural, and cattle ranches, citrus groves, strawberry fa rms, and landscape pl ant nurseries are common sights. Applications of fertilizers or pesticides and seepage from piles or ponds of animal wastes are sources of ni trogen and phosphorus from agricultural land use (Arbukle and Down ing, 2001; Bennett et al ., 2001; Carpenter, 2005; Omernik, 1976). The northeastern sector of the watershed includes forested wetlands that drain into the Hillsborough River, which is the main source of drinking water for the City of Tam pa (Schmidt and Luther, 2002; Xiana and Craneb, 2005). To a lesser ex tent, these forested wetl ands contribute nutrients from natural flora and fauna to its su rface waters (Reddy et al., 1999). Closer to the bay the land use is predominantly urban or suburban, with heavy industry at the confluence of majo r rivers and the bay (Xiana and Craneb, 2005). The four-county Tampa Bay Metropolit an Area, which is host to the Cities of Clearwater, St. Petersburg, and Ta mpa, has a population of 2.7 million residents. Population growth was 30. 45% between 1990 and 2006 (Hilssborough
5 County, 2007; US Bureau of the Cens us 2000, 2007), and wa s higher than the 20.34% of the total United States population for the same time period (Hilssborough County, 2007; US Bureau of the Census, 2006). The shift from rural to suburban or urban land use means more impervious cover and thus increased rainfall run-off (Xiana and Cr aneb, 2005). Run-off may contain nitrate or phosphate from inorganic fertilize rs, or organic forms of nitrogen and phosphorus from leaves, insect debris, and animal excreta, as examples (Johnes et al., 1996; Reddy et al., 1999; Soranno et al., 1996). Atmospheric emissions of nitrogen oxide and ammonia from electrical utility, industrial, and transportation (including motor vehicle) sectors deposit to lake surfaces or adjacent drainage basins (Poor et al., 2005). Non-point source pollution from runoff or atmospheric deposition may increase with population density (Smith et al., 2003). Septic tanks for household sewage tr eatment were typical in twentieth century development of Florida lakefront property, and many if not most of the homes bordering the lakes considered in this study are still on septic systems (Schmidt and Luther, 2002). Much of the Tampa Bay watershed is underlain by karst geology (van Beynen et al., 2007), which in some plac es facilitate groundwater transport and for a few lakes provides a direct connecti on to the cleaner water of the Floridian Aquifer (Cheng and Kindinger, 2004).
6Figure 1.2 Land Use in Tampa Bay watershed (Tampa Bay Estuary Program, 1999). Lakes analyzed in this study are located in areas with residential, commercial and industrial, and agricultural use mainly.
7 1.3 Research Hypotheses The research hypotheses were: Average lake water concentration s of total phosphorus (TP), the ratio of total nitrogen to total phosphorus (TN:TP), and phytoplankton as measured by chlorophyllin selected lakes of the Tampa Bay watershed changed significantly between 1990 and 2007; Lakes with a greater abundance of submerged aquatic vegetation have significantly lower water co ncentrations of total phosphorus, total nitrogen (TN), and chlorophyll. Eutrophication (primary productivity of nat ural waters) is traditionally measured based on the concentration of total phosphor us (TP) and total nitrogen (TN) as well as the concentration of phytopl ankton (Canfield et al., 1985; Dillon and Rigler, 1974). For accuracy and functionalit y, phytoplankton is estimated by the measurement of chlorophyllin lake water (Canfield et al., 1985; Dillon and Rigler, 1974). Consequently, lake water conc entration of TP, TN, and chlorophyllwere water quality parameters used for this analysis. Special attention was given to lake water TP concentration and the metabolism of phosphorus in shallow lakes since this is the most common nutrient limiting phytoplanktonic productivity in freshwater ecosystems (C hunLei et al., 2004; Schauser et al., 2004; Schindler, 1977; Scinto and Reddy, 2003; Zhou et al., 2001). This dissertation has been divided in complementary subtopics, each one covered in an independent c hapter with specific objectives and results as follows:
8 Chapter 2 Review of processes that dete rmine the primary productivity of aquatic systems; Definition of external factor s that influence the processes involved in the phosphorus cycle in aquatic systems; and Review of models of phosphorus fate and transport in aquatic systems. Chapter 3 Analysis of temporal trends of water concentration of total phosphorus (TP), ratio of water c oncentration of total nitrogen to water concentration of total phosphorus (TN:TP), and water concentration of chlorophyllin selected lakes of the Tampa Bay watershed; Discussion of differences in tr ends of lake water concentration of TP, TN:TP ratio, and chlorophyllbetween lakes; and Discussion of factors influenci ng the change in the lake water concentration of TP, TN:TP ratio, and chlorophyll. Chapter 4 Analysis of correlation between measures of submerged aquatic vegetation (SAV), lake water c oncentrations of TP, TN, and chlorophyll, and lake depth, volume, and area; and Discussion of factors influencing lake water concentration of TP, TN, and chlorophyll;
9 Chapter 5 Analysis of association between submerged aquatic vegetation and lake water concentrations of TP, TN, and chlorophyll. Chapter 6 Implications of this research.
10 CHAPTER 2. LITERATURE REVIEW 2.1 Introduction Phosphorus is a fundamental component of life. Its presence in the adenosine diand triphosphorus molecule make possible the conversion of energy from sunlight into chemical ener gy in plants, and fuels the indispensable reactions of photosynthesis and resp iration (Chameides and Perdue, 1997). These processes do not just support the entire ecological web by providing energy to fuel all the metabolic trans formations in living organisms but also determine the flow of phosphorus betw een biosphere and mineral reservoirs. Essentially all the phosphorus significantly present in nature is in the form of orthophosphates (+5 oxidation state) since it is the only form stable in aqueous solution (Chameides and Perdue, 1997). Unlik e other essential elements for life like nitrogen and carbon, phosphorus does not have a stable gaseous form of significance (Chameides and Perdue, 1997; Lahm, 2008; Schlesinger, 1991) and its occurrence in the atmosphere is generally limited to just minor amounts dissolved in moisture or contained in suspended dust particles (Graham and Duce, 1979). Its presence in nature is more prevalent in the form of insoluble compounds since its cycling depends mostly on slow geologic processes such as weathering of apatite and ot her calcium phosphorus minerals (Schlesinger,
11 1991). Although plant roots and mycorrhizae ma y accelerate this process, their effect is not enough to consider a signific ant role of microorganisms in increasing phosphorus bioavailability (Schlesinger, 199 1). Therefore, it is frequently a limiting nutrient for photosynthetic producti vity, especially in aquatic ecosystems (Chameides and Perdue, 1997) and even more so in freshwater ecosystems (ChunLei et al., 2004; Schauser et al., 2004; Schindler, 1977; Scinto and Reddy, 2003; Zhou et al., 2001). Although this limitation slows down natural productivity, the increasing discharge of phosphorus into natural waterwa ys as a result of human activities is accelerating the availability of this nutrient (Rast and Thorton, 1996). Hence, a basic, but clear, understanding of the phos phorus cycle in lakes is of vital importance for environmental management plans addressing the problem of eutrophication. In aquatic ecosystems, phytoplankton growth removes soluble phosphorus from the water column while phytoplankton decomposition releases it back into the water and sediments (Fis her et al., 1982). This classical view of decomposition of organic matter and cons equently regeneration of mineral nutrients to support primary production at the base of the food chain is correct but too simplistic (Sundby et al., 1992). Phosphorus in aquatic ecosystems is pr esent mostly as a particle but also in dissolved form (Schnoor, 1996; Wang and Mitsh, 2000), and each form behaves differently in the phosphorus cycle. Particle phosphorus can settle by gravity while dissolved phosphorus can be a ssimilated by bacteria and plants.
12 Considering total phosphorus rather than differentiating phosphorus forms has been a more functional expression for modeling purposes because it simplifies the process (Wang and Mitsh, 2000). Si nce it is impossible to completely describe the phosphorus cycle, a basic descr iption of the most important factors necessary for a reasonable understanding of the phosphorus cycle in shallow lakes follows. Figures 2.1 provide a simple visual description of some of the main reservoirs and flows of total phosphorus cyc ling in shallow freshwater systems. Although not included in the figure, external inputs and outputs are determinant in the phosphorus budget of the system. The most important external contributions of phosphorus to the system are orig inated from non-point sources such as agriculture and urban activi ty (Carpenter et al., 1998; Howarth et al., 1996). Point sources are usually less important, although may be also significant (Carpenter, 2005; Cowen and Lee, 1976). In most cases, groundwater flow and atmospheric deposit ion are not important sources of phosphorus because it is normally not mobile in soils (Reddy et al., 1999) and its content in dust and suspended solids in the air is very low (Graham and Duce, 1979). This may not be the case for the lakes examined in this study for the following reasons: the central and nort hern part of Hillsborough County has a karst formation of sandy soil (van Beynen et al., 2007) that may facilitate the phosphorus leaching, and atmospheric depos ition is actually a significant contributor of phosphorus to Flori da lakes (Reddy et al., 1999).
13 Figure 2.1 Total phosphorus cycle in shallow aquatic systems. The connections in this figure represent the flows of phosphorus between the different components. The arrow heads point to the component receiving phosphorus. Those arrows with heads in both directions represent bi-flows of phosphorus. External inputs and outputs are not included in the diagram for visual clarity, but are important determinants in the phosphorus budget of the system. TP: total phosphorus. SAV: submerged aquatic vegetation. EAV: emergent aquatic vegetation. TP in SAV TP in Water TP in Periphyton TP in phytoplankton TP in Sediments Sediments Water TP in EAV Air
14 2.2 Inorganic Phosphorus Organic and inorganic phosphorus compounds from the watershed are carried with runoff and discharged into the lake water. Eventually they settle into the lake sediments. Once in the sedi ments, persistent co mpounds of inorganic phosphorus stay buried in their original fo rm while labile forms are dissolved into the sediment pore water. From the sediment pore water, phosphorus is released as phosphorus back into the water column, re-precipitated, or adsorbed by other compounds within the sediments (Sundby et al., 1992). Under oxic conditions, binding of inorganic phosphorus to iron in the sediments has an important effect on the phosphorus mass balance in aquatic systems (Krom and Berner, 1980; 1981). The oxidized surface layer in sediments decreases the flux of dissolved inorganic phosphorus from the sediment to the water column by providing iron oxide that binds to the phosphorus, therefore trapping phosphorus (Mortimer, 1971). H ence, as the oxygen concentration decreases with depth in the sediments, the concentration of sequestered phosphorus decreases with depth as well (Sundby et al., 1992). This process of phosphorus sequestration has been defined in two steps: rapid adsorption on surfaces and then slow diffusion into the par ticles of iron oxide (Barrow, 1983). It has been suggested however, that particles at depths equal or greater than 10 cm may still maintain some capacity to retain phosphorus, despite the anoxic environment (Silverberg et al., 1987).
15 2.3 Organic Phosphorus Usually most of the phosphorus present in the water column and sediment is organic (Rigler, 1956). Bacteria and fungi decompose plant and animal tissue into more simple organic matter (Fenchel 1970). The resulting organic matter is further transformed by heter otrophic bacteria into bacterial protoplasm. Bacteria are in turn consumed by protozoan grazer s, resulting in regeneration of inorganic phosphorus. According to Johannes (1965), bacteria alone release little phosphorus, instead they consume it, so bacterial grazers are needed for mineralization of this ba cterial phosphorus. Barsdate et al. (1974), however, suggested based on laboratory experiments t hat little phosphorus from bacteria pass through grazers before is released into solution. This author explained that the reason for an increased water phosphorus concentration with a high presence of grazers is due to the fact that the physiolog y of bacterial population is changed by grazing pressure by selecti on of rapidly growing forms of bacteria. This change caused by grazing results in more rapid bacterial assimilation of organic phosphorus and subsequent faster regeneratio n of inorganic phosphorus, which is the form assimilat ed by phytoplankton (Barsdate et al., 1974). Once organic phosphorus has been miner alized, it undergoes the same pathways as deposited inorganic phosphorus. A portion of the mineralized phosphorus is concentrated as phosphorus into the pore water where it is maintained in equilibrium with the portion adsorbed to the surface sites. Part of the phosphorus adsorbed to particles is then diffused to the interior of the iron
16 oxides and stays sequestered. As disso lved phosphorus is released from the pore water back to the water column, it is also replaced by surface adsorbed phosphorus to maintain the equilibrium concentration (Sundby et al., 1992). 2.4 Oxygen Changes in the oxygen concentration in the overlying water can cause drastic changes in the phosphorus flux from sediments, with an aerobic conditions increasing phosphorus flux to the overlyi ng water (Moore et al., 1998; Moore et al., 1991), thereby making it available to phytoplankton and ev entually promoting algal blooms (ChunLei et al., 2004). As phyto plankton from a bloom sink to the sediments, decomposer bacteria consume the decaying material (Fenchel, 1970) and the remaining dissolved oxygen, further exacerbating the ox ygen depletion. In opposite circumstances, high photosynt hetic activity of benthic microalgae decrease phosphorus release by increasi ng dissolved oxygen levels (Spears et al., 2008). Additionally photosynthesis can lead to elevation in pH, and high pH under calcareous conditions favors co-precipitation of phosphorus with calcium carbonate (Dierberg et al., 2002; Spears et al., 2008). It has also been suggested that an elevated pH can affect ion exchange processes that decrease the capacity of iron and aluminum co mpounds to bind phosphorus, resulting in release of phosphorus from sediments (Bostr om et al., 1988; Zhou et al., 2001). The flux of phosphorus from sedim ents to water column is regulated primarily by redox reactions involving ir on and aluminum in sediments, and then by the gradients in the concentrati on of phosphorus between pore water and
17 overlying water (Moore et al., 1998). Levels of nitrate and sulfate have been inversely correlated to the flux of phosphor us from sediments to overlying water (Fisher and Reddy, 2001). This might be due to the oxygen content in these compounds that can raise the overall oxygen level in the media hence influencing redox reaction in phosphorus chemistry. By contrast Caraco et al. (1989) suggest greater release of phosphorus fr om sediments at higher sulfate concentrations. In lakes with a thermocline, the rel ease of phosphorus from sediments is typically controlled by the oxygen concentration in the hypolimnion, and decomposition of phytoplankton sinking into this layer is a significant cause of oxygen depletion (Genkai-K ato and Carpenter, 2005). Deeper in the sediment column, even below aerobic overlying wate r, conditions are increasingly anoxic, facilitating the release of phosphorus fr om iron and aluminum bound phosphorus undergoing reduction (Moore et al., 1991; Sundby et al., 1992). This newly regenerated phosphorus can migrate upward to participate in exchange reactions between adsorption sites and pore water, or can be released into the overlying water (Sundby et al., 1992). Additionally to the reduc tion of iron and aluminum, part of the increased phosphorus release from sediments under anoxic conditions may be also explained by metabolic changes in microbi al population that cause release of phosphorus from the cells (Bostrom et al., 1988; Gatcher et al., 1988). Complexed phosphorus and iron (II) is re leased into water solution when the cells, under anoxic conditions and in the absenc e of nitrates, use iron (III) as an
18 alternative electron acceptor (Jones et al., 1983). Moreover, microbial growth yield is typically low u nder anoxic conditions, hence the release of phosphorus previously bound to carbon in bacteria and cyanobacteria is higher (Bostrom et al., 1988). 2.5 Submerged Aquatic Vegetation Aquatic macrophytes in general (both emergent and submerged) can absorb nutrients both from the water and fr om the sediments depending upon the species and relative nutrient concentrations in water and sediments (Denny, 1972; Graneli and Solander, 1988; James et al., 2006; Rattray et al., 1991). Emergent macrophytes take phosphorus from the sediments under all conditions while submerged aquatic vegetation ta kes phosphorus from both water and sediments although the latter pathway is more prevalent under normal conditions as well as under conditions of high water phosphorus concentration (Graneli and Solander, 1988). Barko and Smart (1980), Barsdate et al. (1974) and Mcroy (1972) indicated that aquatic macrophytes can pump nutrients from sediments to the water column making it available to phytoplankton. Wang and Mitsh (2000) included this effect in a phosphorus cycling model. Graneli and Solander (1988) suggested that both types of aquatic v egetation, submerged and emergent, can release minimal or important quantitie s of phosphorus to the water column depending on the vegetative st age. Growing vegetation would release minimal
19 quantities of phosphorus via plant mate rial, and decaying vegetation would release considerable am ounts. According to Gu mbricht (1993), the main phosphorus removal mechanism in desi gned wetlands would be nutrient uptake by submerged aquatic vegetation followed by harvesting. Other theories suggest that aquatic macrophytes should not be removed because they decrease recycling of phosphorus from sediments back into the overlying water by suppressing resuspension of sediments. Th is principle was incl uded in prediction models developed by Genkai-K ato and Carpenter (2005). Inhibition of sediment resuspension is a simple mechanism that explains at least part of the effect of submerged aquatic vegetation in the suppression of phosphorus recycling (Hamilton and Mitc hell, 1996; Scheffer, 2004). Emergent and submerged aquatic vegetation neutralize the effect exerted by waves and wind in causing vertical mixing of the ov erlying water in shallow lakes (Bachmann et al., 2004; Hamilton and Mitchell, 1996). Additionally a possible intense photosynthesis caused by aquatic macrophytes and associated periphyton can increase the pH and lead to coprecip itation of phosphorus with calcium carbonate under alkaline conditions (Dierberg et al., 2002). In contrast, elevated pH in the interface water sediments re sulting from photosynthesis may cause release of phosphorus from iron and aluminum compounds mostly because orthophosphate is replaced by hydroxi de ions in ligand exchange reactions (Bostrom et al., 1988). It has been expl ained that aquatic vegetation in general affects the water chemistry by regulating oxygen and pH therefore influencing the phosphorus cycle in lakes (Graneli and Solander, 1988).
20 Emergent and submerged aquatic vegetati on provide substrate surfaces for epiphytes and periphyton to grow (C attaneo and Kalff, 1980; Dierberg et al., 2002), which can offer another important regulatory impact on the phosphorus flux. Periphytic algae and other epiphyte s growing on the surface of submerged aquatic vegetation remove phosphorus direct ly from the water column (Dierberg et al., 2002; Scinto and Reddy, 2003). S ubmerged aquatic vegetation can also directly uptake part of its required nutrients from the water column in addition to sediments (Graneli and Solander, 1988). This absorption and translocation of phosphorus through out the entire plant mate rial is another impor tant part of the lake water phosphorus cycle (Barko and Smart, 1980). The above-mentioned mechanisms refer to the availability of phosphorus in the water column. Yet there are also other ways by which submerged aquatic vegetation influence the growth of phytopl ankton. Macrophytes can indirectly suppress phytoplankton growth by providing shelter from fish to the zooplankton that graze on phytoplankton (Scheffer, 2004; Scheffer et al., 2001). Zooplankton that graze on bacteria, however, may also accelerate the mineralization process that makes phosphorus available in the water column (Barsdate et al., 1974). Aquatic vegetation, in general, provides a refuge against cladocera (microscopic order of crustacean and part of zooplank ton population), which are the most efficient grazers of bacteria and a favori te prey for fish (Moss, 1990) In a laboratory experiment, Rigler (1956) found that some species of bacteria that utilize inorganic phosphorus grow well in suspension, but others grow well on the
21 walls of storage vessels. Hence, in a natural environment, macrophytes might provide surfaces for phosphorus -consuming bacteria to grow. Independent of the mechani sms utilized by submerged aquatic vegetation for regulation of phosphorus levels, there is extensive literature demonstrating the phosphorus removal capacity of submer ged aquatic vegetation in constructed wetlands and lakes (Dierberg et al., 2002; Gu et al., 2001; Gumbricht, 1993; Knight et al., 2003). Gumbricht (1993) stated that the phosphorus absorption rate of submerged aquatic vegetati on is proportional to the phosphorus concentration within the surrounding water. These results are consistent with experiments done by James et al. (2006) showing a direct relationship between nutrient concentration in submerged veget ation tissue and water column nutrient concentration. This is of special in terest regarding the potential use of submerged aquatic vegetation as a treatm ent method to remove nutrients in nutrient-rich water. Although the nutrient removal capaci ty of submerged aquatic vegetation and mechanisms through which submerged aquatic vegetation may regulate the phosphorus cycle in lakes has been doc umented, the association between nutrient concentration and submerged aqua tic vegetation biomass in Florida lakes has not been well established. It has been theorized based on correlations found in some studies in Florida that submerged aquatic vegetation would cause lower water nutrient concentrations (Bac hmann et al., 2002; Bachmann et al., 2004), except at high nutrient levels, when the influence would work in the opposite way. Other authors indicate that elevated nutrient concentrations in lake
22 water would cause the absence of s ubmerged aquatic vegetation because of light attenuation (Duarte, 1995; Graneli and Solander, 1988). A study conducted in Qubec, Canada, reported that slope of the littoral zone is a more important determinant of the variability in subm erged aquatic vegetation biomass (Duarte and Kalff, 1986). In summary, extensive submerged aquatic vegetation may lead to low nitrogen and phosphorus concentrations in the water column, which in turn leads to low phytoplankton growth. These two associations may explain why submerged aquatic vegetat ion has been reported as inversely influencing phytoplankton (as estimated by chlorophyll) under Florida conditions (Canfield and Hoyer, 1992). This would be an indi rect connection between submerged aquatic vegetation and phytoplankton by me ans of the concentration of total phosphorus in the water column (see Figur e 2.1). The authors indicated that a percentage area cover (PAC) of submer ged aquatic vegetation greater or equal to approximately 30% of lake area is required for noticeable reductions in phytoplankton biomass. They found that a small amount of aquatic vegetation does not play an important role in phy toplankton biomass reduction. A study conducted in New Zealand lakes, however provided evidence that submerged aquatic vegetation dominates phytoplankton in shallow eutrophic lakes (Hamilton and Mitchell, 1996), and the sugges ted mechanism of control was the stabilization of lake sediments and the i nhibition of sediment resuspension. In literature reported on Florida lakes, the association between submerged aquatic vegetation and nutrients is st ill unclear, however, a strong direct
23 association has been reported between water nutrient concentration and phytoplankton concentration, as measured by chlorophyll(Bachmann et al., 2002; Brown et al., 2000; Canfield et al., 1984). Drastic increases in lake water c oncentration of nutrients and turbidity have been reported when submerged aquatic vegetation was removed by herbicide treatment (O'Dell et al., 1995) or by hurricanes (Bachmann et al., 1999). Similarly, lakes have been reported to switch from turbid to clear water state when planktivorous fish were remov ed and submerged aquatic vegetation increased (Ozimek et al., 1990). 2.6 Light Light can affect the phosphorus cycle indirectly through the presence of submerged aquatic vegetation, phytoplankto n, turbidity, and benthic microalgae. Light can influence water phosphorus concentration through phytoplankton (Philips et al., 1997), by favoring gr owth of this microscopic algae and cyanobacteria, consequently removing phosphorus from solution and incorporating it into easily sedim ented biomass (Krivtsov et al., 2000). Likewise the intensity of light r eaching the sediments determines the distribution and abundance of s ubmerged aquatic vegetation (Hoyer et al., 2004). There is some discrepancy in the literat ure regarding the feedback that nutrient loading may have on submerged aquatic vegetation through shading by
24 phytoplankton. Spence (1982) found lig ht availability as an important consequence of excess nutrient loading and one of the major factors determining the submerged aquatic vegetati on distribution. The increase in phytoplankton as a result of larger phosphorus inputs would block light from reaching the submerged aquatic vegetation, reducing their distribution to just the very shallow zone. Subsequently, their capacity to restrain phosphorus recycling back into the water column would also be reduced (Genkai-Kato and Carpenter, 2005). Light penetration to the sediments also allows photosynthetic activity of benthic microalgae. This regulates the fl ux of phosphorus from sediments to water by affecting the oxygen concent ration and pH level and thus phosphorus sequestration in sediments (Spears et al., 2008). It has been speculated that periphy tic algae attached to the surface of submerged aquatic vegetation might be the cause of the shading effect responsible for loss of submerged macrophytes under conditions of high nutrients loads (Philips et al., 1978) Results of Bachmann et al. (2002) suggested that this is unlik ely under conditions of Flor ida lakes, because they did not find an association between increas ing nutrient loading and periphyton abundance in their study. They explain this by suggesting that the shading effect caused by phytoplankton as a result of increasing nutrient concentration keeps periphytic algae from receiving enough light for photosynthesis. Periphytic algae are attached to macrophyte surface and therefore unable to move toward the source of light, while phytoplanktonic alga e can move in the water column toward
25 the source of light, blocking it from reaching the periphyton (Bachmann et al., 2002). In this case, if periphyton compet es with submerged aquatic vegetation for light and periphyton is reduced with increased nutrient loading because is shaded by phytoplankton, then a high nutrien t concentration in the water column would add shading from the phytoplank ton but reduce shading from the periphyton to submerg ed aquatic vegetation. Light penetration is dependent in great measure on turbidity. Scheffers basic model for alternative stable stat es in shallow lakes (Scheffer, 2004) assumes that an increase in nutrient leve ls causes an increase in turbidity, turbidity is reduced by submerged vegetation, but submerged vegetation disappears under conditions of extreme tu rbidity. Likewise Bachmann et al. (2002) indicate that submerged aquatic vegetation tolerate increasing water nutrient concentrations (which is one of the factors leading to turbidity) below excessive levels. Thus, as long as the nutrient concentration is not extreme, submerged aquatic vegetation amount determines nutrient concentrations in lake water (Bachmann et al., 2002; Bachmann et al., 2004). Silica can obstruct the effect of li ght by triggering the growth of diatoms (Krivtsov et al., 2000). Diatoms (phytoplanktonic algae) grow in abundance under high concentrations of silica, thereby shading submerged aquatic vegetation and periphyton, and taking more phosphorus from the water column, making it less available for periphyton.
26 2.7 Mean Depth and Temperature Temperature affects lake water phos phorus cycling primarily through its effect on biological activity (Bostrom et al., 1988). Increased temperature favors microbial activity, which in turn decr eases oxygen concentrations and pH via microbial consumption of oxygen. As prev iously described, low oxygen levels in sediments cause the release of phosphorus bound to iron and aluminum complexes from sediments into soluti on (Moore et al., 1991; Richardson, 1985; Sundby et al., 1992). High temperature coupl ed with high pH can also have the opposite effect by removing phosphorus fr om solution via co-precipitation with calcium carbonate (Bostrom et al., 1988; Dierberg et al., 2002). In a broad sense, temperature influences the phosphorus cyc ling in aquatic systems by affecting the rate at which chemical processe s take place (Simpson and Eaton, 1986). Empirically-derived models developed by Genkai-Kato and Carpenter (2005) indicated that the reversibility of conditions of high concentration of phytoplankton back to a clear water stat e was affected considerably by mean depth, temperature, and presence of s ubmerged aquatic vegetation. In shallow lakes (<1.9 m) with submerged aquatic vegetation, phosphorus recycling from sediments decreased as depth decreased bec ause of more vegetation at shallow depths, increasing the likeli hood of restoration back to the clear water state for shallow lakes. In the absence of submerged aquatic vegetation, however, shallow lakes were more vulnerable to eutrophication and less lik ely to return to clear water state because phosphorus re cycling increased wit h temperature and with decreased depth. Submerged aquatic vegetation was not found to have an
27 effect at depths greater than 10 m. Dilution of phosphorus in the hypolimnion and reduction of temperature are the possi ble factors suppressing recycling of phosphorus at depths greater than 28 m. An interm ediate mean depth (1.9-28 m) is most resistant to reversing from eutrophication to the clear water state because these lakes were too deep to be affected by submerged aquatic vegetation and too shallow for phosphorus dilution in the hypolimnion. 2.8 Bacteria and Aquatic Animals An increase in microbial activity l eads to increase in oxygen and nitrate consumption. As oxygen concentration dec reases, the capacity of complexes of iron, aluminum, and manganese to reta in phosphorus decreases as well (Bostrom et al., 1988). Anaerobic condition s cause changes in the metabolism of bacteria and cyanobacteria that increases phosphorus release from the cells (Gatcher et al., 1988). In addition, the biom ass of microbial population is reduced under anaerobic conditions hence less retent ion of phosphorus in microbial mass means more phosphorus is available in solution (Bostrom et al., 1988). It is not clear the role of bacteri a in the phosphorus cycling but most literature seems to agree t hat in general they do not ac tively excrete phosphorus, and demand more phosphorus from the water column than they release (Andersson et al., 1988; Johannes, 1968). The release of phosphorus caused by bacteria (Azam et al., 1999) is actually presumed to be an indirect effect through zooplankton excretion after bacteria is consumed by zooplankton (Ammerman
28 and Azam, 1985) or when bacteria die or are infected by viruses. Other authors suggested that although bacterial grazers in fact stimulate the release of phosphorus from bacterial biomass into t he water column, this happen because the grazing pressure makes faster and more efficient the phosphorus release process in bacteria but not because phosphorus actually pass through the grazers in significant amount s (Barsdate et al., 1974). In the absence of a large population of bacterial grazers, heterotrophic bacteria utilize a greater amount of inorganic phosphor us in the process of organic matter decomposition (Johannes, 1965). Since most organic matter comes from carbohydrates, which are poor in essential nutrients, bacteria use inorganic phosphorus from free water as an extra source of phosphorus for growth (Fenchel, 1970). This would r educe the amount of available phosphorus in the water column under conditions not limited by carbon, since bacteria can successfully compete with phytoplankt on for uptake of phosphorus due to a faster growth rate (Rhee, 1972). Macroinvertebrates affect phosphorus cycling in many ways. They promote the rate of phosphorus miner alization (Hansen et al., 1998) and increase phosphorus releas e through their digestive and excretory processes (Gardner et al., 1981). Biot urbation generated by macroi nvertebrates in the sediments cause water movement, which fa vors the diffusion of phosphorus into the overlying water. Burrows made by so me animals further increase the surface of contact between interstitial and over lying water (Bostrom et al., 1988), which enhances the exposure of sediments to oxygen, thereby augment ing retention of
29 phosphorus by iron oxide, but also fa cilitates the transport of phosphorus contained in the pore water out to the overlying water (Hansen et al., 1998). Burrowing animals also transport ox idized and reduced compounds back and forward between the reduced and oxidized zone, promoting redox reactions (Canfield et al., 1993). Mussels and other macroinvertebrates also have a strong regulatory impact on phosphorus cycling by regulatin g the population of phytoplankton, zooplankton, and bacteria (Barsdate et al., 1974; Berman and Richman, 1974; Johannes, 1968; Scheffer, 2004). Fish are especially important since they impose a direct and indirect effect on the water column phosphorus in aquatic systems. They exert a predatory pr essure on bacteria, zooplankton, phytoplankton and benthic invertebrates, wh ich indirectly affect the flux of phosphorus (Andersson et al., 1988; Kitchell et al., 1979; Scheffer, 2004). Fish also directly affect phosphorus cycling by increasing resuspension of sediments in search for benthic food (Scheffer et al., 2001). The processes by which fish and invertebrates influence the phosphor us cycle are strongly related to vegetative cycles and are more intens e during high temperature periods (Andersson et al., 1988). 2.9 Land Use / Land Cover Conditions in the watershed influ ences lake water quality through the inputs of materials and nutrients via r unoff (Reddy et al., 1999; Soranno et al.,
30 1996). A study conducted by Smith et al. (2 003) reported that runoff, area of the watershed, and population, were the three va riables more strongly associated to loading of dissolved nitrogen and phosphorus into receiving waters. Trends in land use, agriculture, human population (Jo hnes et al., 1996; Smith et al., 2003) as well as mining and forestry practice s in watersheds have increased the transport and discharge of phosphorus to water bodies (U. S. Environmental Protection Agency 1990 as quoted by Sor anno et al., 1996). Contributions of phosphorus to lakes can come from point sources such as waste water treatment plants, as well as from non-point source s such as runoff from agricultural and urban lands (Carpenter, 2005; Cowen and Lee, 1976). Non-point sources resulting mainly from agricultural and urban activity are the most important sources of phosphorus to natural bodies of water in the United States and other developed countries (Carpenter et al., 1998; Howarth et al., 1996). According to Reddy et al. (1999), urban land use is the category with the higher contribution of phosphorus to rece iving waters, followed by agricultural and forested land use. This is explained by the greater fraction of impervious layer in urban areas, which r educes infiltration and increases flow of runoff. This subsequently results in greater phos phorus export. Groundwater flow is commonly not an important passage of phos phorus from the watershed to lakes because of the generally low mobility of phosphorus in the soil. The high porosity of the karst formation in northeastern and central Hillsborough County (van Beynen et al., 2007) might prove the exception.
31 In agricultural land use, phosphorus su rplus from excessive fertilization and manure production is accumulated in the soil further finding its way with the runoff into natural water bodies (Bennett et al., 2001; Carpenter 2005). A study conducted in agricultural areas of the eas tern United States reported that the magnitude of runoff nutri ent concentration is usually relative to the percentage of the watershed covered with cultivated land (Omernik, 1976). Other study conducted in the highly agricultural r egion of Iowa found that watersheds dominated by animal agriculture constitute the main source of phosphorus into natural waters, while intensive row cr op agriculture-dominated watersheds are larger sources of nitrogen (Arbukle and Downing, 2001). Reddy et al. (1999) says that atmospheric phosphorus depositio n (especially dry deposition) is a significant and local contributor of phos phorus to Florida la kes and that the loading is higher for agricultural lands as compared with forested watersheds. In Florida, phosphorus mining is an im portant economical activity but has heavily impacted waters that reach Tampa Bay (Baskaran and Swarzenski, 2007). The area in the eastern part of the Tampa Bay watershed extends into Bone Valley, which has been mined for phos phorus since the late 1800s. That area is characterized by an abundance of wetlands and its natural ecology that has been disrupted by the phosphorus mi ning. After 20 years of active restoration conducted in the region, however, has water quality improvements have been seen (Brown, 2005). Phosphorus removal by wetland restoration and treatment of point sources can signific antly reduce the frequency of algal blooms (Billen and Garnier, 1997). Land use and its corresponding contributions of
32 phosphorus to lakes must be considered in phosphorus budget calculations for planning lake management options (Cowen and Lee, 1976). 2.10 Public Health Implications Beside the widely known adverse e ffect in the ecosystem: light attenuation, odors, lowering of dissolved oxygen, and fish kills (Paerl, 1988); eutrophication represents also a seri ous threat to aesthetics and most importantly to public health (FDEP, 2004). A study at the University of Colorado concluded that nutrient enrichment in lakes and ponds, cause conditions favourable to abundance of trematode parasites; and also suggested that favour conditions for mosquito vectors of ma laria, cholera causing bacteria, and swimmers itch (Johnson et al., 1999). There are species of phytoplanktonic al gae, that can be toxic and that are triggered by surface water enrichment. Abundant productivities of any toxic algae are better known as harmful al gal blooms (HABs) and are a growing concern for public health in Florida bec ause their potential effect in surface drinking water resources and recreationa l sites (FDEP, 2004). Cyanobacterial species are a phylum of phytoplankton wit h a nitrogen-fixing capability. They would be favored over other species of phytoplankton by increasing conditions of relative limited nitrogen and abundant phosp horus. As a consequence, low ratios of total nitrogen to total phosphorus (TN:TP) could lead to increase of cyanobacterial water concentration (H ecky and Kilham, 1988; Levich, 1996;
33 Levich and Bulgakov, 1992). This can be a concern to public health because certain species of cyanobacteria produce cy anotoxins that cause oxidative stress in affected cells and may promote tumo rs in the nervous, hepatic, and dermatologic systems (Fleming et al., 2002; Karjalainen et al., 2007). These toxins may enter the human body thr ough consumption of drinking water, inhalation of aerosolized toxins, and by contact with water or fish from water polluted with cyanobacteria (Fleming et al ., 2002; Karjalainen et al., 2007). The relative abundance of cyanobacteria over other phytoplanktonic species causes interferences in the food-web (Levich, 1996). They are not consumed by many species, and this lack of predatory pressure further helps them to dominate over competitor specie s that are producers in the food chains (Levich, 1996). The cyanobacteria produced toxin can be transferred through the food chain through consumption of some gr azers, affecting the upper levels of the trophic web and even humans (USEPA, 1997). In small amounts cyanobacteria may complement the diet of tolerant grazers, but when ingested in excess the cyanotoxins may decrease the eggs production of some species of zooplankton and reduce feeding and growth rate s of fish larvae (Karjalainen et al., 2007). The process of detoxicat ion for the possible consumers of cyanobacteria imply a metabolic cost, resulting in a decreased growth and condition and subsequently in their suscept ibility to be predated for other higher consumers (Karjalainen et al., 2007). Some literature suggest that harvest ing of submerged aquatic vegetation may increase the risk of cyanobacterial bl ooms (Scheffer, 2004) but other found
34 that, at least under conditions of lo w nutrient levels, cyanobacteria did not increase as a result of harvesting subm erged aquatic vegetation (Morris et al., 2006). 2.11 Summary of Phosphorus Cycle Review Among the most important processe s regulating phosphorus cycling in shallow freshwater lakes are sedimentat ion, plant nutrient uptake, and regulation of ion exchange processes via disso lved oxygen and pH. Emergent and submerged aquatic vegetati on enhance sedimentation ma inly by suppressing water turbulence and thus favouring c onditions for particle settling, and by facilitating phosphorus co-precipitation with calcium complexes by raising pH or binding with iron and aluminium by altering water column oxygen due to photosynthesis. Both types of aquatic veget ation provide surface of substrate for periphytic algae, an algae that relies mostly on phosphorus dissolved in the water column. Submerged aquatic vegetati on assimilates nutrients from either or both media, water and/or sediments, dependi ng upon the relative availability of nutrients in each media. Emergent aquat ic vegetation relies on sediments for a supply of nutrients. Under conditions of high lake water phosphorus concentration, submerged aquatic vegetat ion would play an important role accumulating in its biomass phosphorus t hat has been directly removed from the water column. Under conditions of high concentration of phosphorus in the
35 sediments, like those in lakes with a long history of phosphorus loading, emergent aquatic vegetation would play a major role in removing phosphorus from the sediments. Th is phosphorus would eventually return to the water column as the vegetative decay. Nutrient uptake from the water column or from the sediments would be a key diff erence between submerged and emergent aquatic vegetation that may determine t he importance of these two types of aquatic vegetation in regard to lake management plans directed toward improvement of wa ter quality. Additionally, external inputs of phos phorus are determinant regulators of the phosphorus status in lakes. T hese depend on land use conditions on the surrounding watershed. Urban and agricultural land uses are usually the most important contributors of phosphorus to lakes, mainly through runoff, although highly porous soil profile might facilit ate underground transport of phosphorus. Atmospheric deposition of phosphorus-lade n dust may be an important input of phosphorus to Florida lakes. Lake water increase in phosphorus concentration can lead to nitrogen limiting conditions that favour the increase in abundance of toxic cyanobacteria with the consequent adverse effe cts in public health. 2.12 Modeling Review Information on the dynamics and quantity of phosphorus is critical in the assessment of water quality (Komatsu et al., 2006) since availability of this
36 element is a limiting factor in the production of phytopl ankton in most freshwater systems (Schauser et al., 2004; Schindle r, 1977; Scinto and Reddy, 2003; Zhou et al., 2001). Some components within the aquatic ec osystem are naturally placed as key factors to buffer drastic fluctuations in phosphorus levels and consequently in algal productivity. Submer ged aquatic vegetation and associated periphyton have been identified as potential controllers of lake water phosphorus concentration and water quality (Bachmann et al., 2002; Ba chmann et al., 2004; Dierberg et al., 2002; Scheffer, 2004). Results presented in chapters 4 and 5 show inverse correlations between submerged aquatic vegetation and water phosphorus concentrations even higher than those r eported on the literature, and support the theory that submerged aquatic vegetation plays an impor tant role as a regulator of water phosphorus levels and water qua lity in general and as such, needs to be included in a lake water quality model. Since early in the 1970s, environm ental managers started using models as a tool for analysis and formulation of plans (Komatsu et al., 2006). Today, models are obligatory tools to solve envir onmental problems. In eutrophication, simple empirical/regression models have been designed to estimate chlorophyllbased on a known total phosphorus (TP) concentration (Bachmann et al., 2002; Canfield and Hoyer, 1992; Canfield et al., 19 84; Canfield, 1983). Regression models, however, do not accura tely account for the non-linearity of the flows between the components of ecolog ical systems. In contrast, dynamical eutrophication models better predict the re servoirs reaction to nutrient inputs
37 (Komatsu et al., 2006), since they in corporate a mechanistic approach that includes time-dependent non-linear closed-loop interactions between determinant components of t he system (Schnoor, 1996). Many dynamical models that simu late phosphorus cycling in aquatic systems have been published. These model s represent all scales from lakes (Everett et al., 2007; James and Bierm an Jr., 1995; Komatsu et al., 2006; Schauser et al., 2004; Schauser et al., 2006; Spears et al., 2008; Zhou et al., 2001), estuaries (Doering et al., 1995; Ki ttiwanich et al., 2006; Serpa et al., 2007), wetlands (Lantzke et al., 1999; Ric hardson et al., 2005; Wang and Mitsh, 2000) to global cycles (Chameides and Perdue, 1997); and different levels of complexity from three co mpartments or reservoirs (H arte, 1988; Lahm, 2008) to more than ten (Jorgensen, 2003; Kittiwanich et al., 2006; Tett and Wilson, 1999). Complicated models with a large num ber of reservoirs and factors considered might be too specific to be applied to more general circumstances. As it is the case with any of the co mponents of aquatic ec osystems, given the complexity of interactions between factors affecting phosphorus cycling in a shallow lake, it is impossible to know and re alistically consider a ll of them (Carr et al., 1997). It is therefore important to def ine what the most important factors, parameters, speciation, and mechanisms involved in water phosphorus dynamics are, in order to limit the complexity of the model. So me important parameters of phosphorus dynamics of shallow aquatic lakes assuming conditions of a closed system are given in Table 2.1.
38Table 2.1 Imortant parameters measured and considered in phosphorus (P) cycling models of shallow aquatic systems. Description Type of aquatic system Value in terms of phosphorus (P) Source P content in peryphiton Wetlands 0.10 0.29 mg g-1 dry weight Scinto and reddy, 2003 P content in submerged aquatic vegetation Lakes 1.41 mg g-1 dry weight Bachmann et al., 2002 P content in lake pore water Lakes 6 mg L-1 Moore et al., 1991 P content in lake pore water Lakes 0.1 > 1 mg L-1 Moore et al., 1998 P content in lake sediment Lakes 0.54-3.84 mg g-1 dry weight Perkins and Underwood, 2000 P content in wetland sediment Wetlands 0.28 0.37 mg g-1 Wang et al., 2006 For instance, phosphorus modeling would ideally consider separate forms of phosphorus, particulate and dissolv ed, as only dissolved forms can be assimilated by primary productivity. Ki netic rates for both forms of phosphorus might not be available (Wang and Mitsh, 2000). Additionally, particulate phosphorus represents the most of t he phosphorus present in aquatic systems (Table 2.2), therefore it is practical fo r modeling efforts to use total phosphorus (TP) as an alternative for differentia ting the two forms of phosphorus (Wang and Mitsh, 2000). For costly decisions rela ted to water quality management, however, a more detailed approach to the problem may be required (Schnoor, 1996).
39Table 2.2 Percent of particulate and dissolved phosphorus content in water column in shallow aquatic systems. Source Percentage Particulate Percentage Dissolved Meybeck (1982) in Wang and Mitsch (2000)95 5 Wang and Mitsch (2000) >75 <25 Schnoor (1996) 70 30 Perkins and Underwo od (2000) 84-85 16-15 Whenever a required level of detail cannot be achieved, assumptions and inputs from other similar conditions might be needed and expected to offer helpful hints into phosphorus dynamics of aquatic systems. Efforts then should attempt to synthesize scientific informa tion from literature into the missing elements in the process of m odel construction. Sections from 2.1 to 2.10 of this chapter describe important factors influencing the me tabolism of phosphorus in shallow lakes. In the pres ent section of this literatur e review, basic processes of phosphorus retention and release in shal low aquatic systems are examined with a goal to gain insight regarding phosphorus cycling and the ro le of submerged aquatic vegetation in lakes. Such pr ocesses will be discussed based on their role in potential basic components or re servoirs of a model that has not been proposed but is theorized in this document. 2.12.1 Interaction between Sediments and Water Column It is well known that the concentration of phosphorus in aquatic systems with an active inflow and outflow of water depends directly on the phosphorus concentration in the inflow (Richard son, 1996; Schnoor, 1996), and inversely on
40 the hydraulic detention time and the s edimentation rate (Schnoor, 1996). In addition to better conditions for sedimentation, which is the main factor responsible for retenti on of phosphorus (Schnoor, 1996; Wang and Mitsh, 2000), a large hydraulic detention time, also increases the opportunity of primary productivity for a higher nutrient uptake (Wang and Mitsh, 2000). The overall cycling of phosphorus in aquatic systems is influenced in great measure by sediments, acting either as a sink or source of phosphorus (Bostrom et al., 1988; Fisher and Reddy, 2001). So me of the phosphorus that settles down to the sediments is recycled back into the lake water column, especially at higher temperature (Genkai-Kato and Ca rpenter, 2005) and anoxic conditions (Schnoor, 1996). Some recycled phosphorus from sediments to water has been mentioned in modeling literat ure not specifying the pat hway (Schnoor, 1996) or has being modeled through resuspensio n caused by benthic organisms and storms; or through emergent macrophytes that pump phosphorus from sediments up to water through litter pathways (Wang and Mitsh, 2000). The latter process happens because emergent macrophytes rely exclusively on sediments for nutrients up-take (Grane li and Solander, 1988). In general, sediments play a key role as potential source of phosphorus into the overlying water (Zhou et al., 2001) and consequently in the recovery of deteriorated aquatic systems (Komatsu et al., 2006). That explains why sediments have been inclu ded in many published models of phosphorus cycling in aquatic systems (Carpenter, 2005; Kitt iwanich et al., 2006; Komatsu et al.,
41 2006; Schauser et al., 2004; Sc hauser et al., 2006; Serpa et al., 2007; Spears et al., 2008; Wang and Mitsh, 2000; Zhou et al., 2001). Concomitant to the importance of sedi ments in the lake water phosphorus cycle are the concentration of oxygen, iron, and aluminum (Moore et al., 1991; Richardson, 1985; Sundby et al., 1992) in the sediments, as these metals determine the capability of sediments to sequester phosphorus. Therefore, phosphorus release rates from sediment s into the water clearly need to be considered in model building. 2.12.2 Interaction between Sediment s and Submerged Aquatic Vegetation Recycling of phosphorus from sediments to the water column represents a critical factor for phosphorus budget calcul ations (Reddy et al., 1999). Wang and Mitsh (2000) included em ergent macrophytes as a pathway for returning phosphorus from sediments back into the water column. The authors calculated that harvesting of macrophytes could re move phosphorus from the system at a rate of 0.32 1.6 g m-2 year-1 depending on the biomass of the macrophytes. Submerged macrophytes have also been modeled as suppressing phosphorus release from sediments to the water co lumn (Genkai-Kato and Carpenter, 2005; Hamilton and Mitchell, 1996; Scheffer, 2004). Literature indicates that for emergent vegetation, absorption and translocation of sediment phosphorus to plant material has important effect on the phosphorus cycle of lacustrine systems (Barko and Smart, 1980; Barsdate et
42 al., 1974, as quoted in Barsdate et al., 1974; Mcroy et al., 1972). While submerged aquatic vegetation influences ph osphorus concentration in lake water through its effect in decr easing sediments resuspension (Bachmann et al., 2002; Bachmann et al., 2004; Scheffer, 2004). Models that intended to determine the concentration of suspended sediments in th e lake water column based on the stress induced by waves have resulted in error because submerged aquatic vegetation was not considered (Hamilton and Mitchell, 1996). Both types of aquatic vegetation have different effects in the phosphorus cycle based, among other reasons, on their s ource of nutrients. These sources of nutrients are sediments for emergent macrophytes but both, sediments and water, for submerged macrophytes (Grane li and Solander, 1988). In the model formulated by Genkai-Kato and Carpent er (2005), shallow depths favor the capacity of submerged macrophytes to s uppress phosphorus recycling in lakes. A shallow depth was assumed in the m odel by Wang and Mitsh (2000) since it was applied to a wetland. 2.12.3 Interaction between Submer ged Aquatic Vegetation and Water Column Other pathway for submerged aquatic vegetation effect in water phosphorus cycling is by increasing nutri ent-up take directly (Graneli and Solander, 1988) and through perip hyton and other epiphytes (Dierberg et al., 2002; Scinto and Reddy, 2003). Except for few dynamical models (Everett et al.,
43 2007; Genkai-Kato and Carpent er, 2005) and a regression model (Canfield et al., 1984) submerged aquatic vegetation has not often been taken into account in modeling for water quality. Based on the results seen in chapters 4 and 5, including submerged aquatic vegetation ma y improve prediction of behaviors in shallow lakes. Models that do not include the submerged aquatic vegetation component but other types of biomass hav e calculated the rate of phosphorus up-take in proportion to the net growth of biomass (Chameides and Perdue, 1997; Wang and Mitsh, 2000), and this same principle may be applied to potential models that include submerged aquatic vegetation. For simplicity, submerged aquatic vegetation might be considered as a single factor, or alternately, as affected by light, temper ature, nutrients, carbon, water velocity and so on (Carr et al., 1997). 2.12.4 Historical Land Use and Trend of Population Growth So far the components considered in this chapter have been about the minimum required for estimating water phosphorus concentration under conditions of a closed system or assuming known inputs and outputs of phosphorus in an open system. In a more r ealistic approach, however, inputs of phosphorus to the system are not known and also need to be estimated. These inputs come with the inflow loads most ly from streams and runoff, are dependent on land use (Reddy et al., 1999), and can be predicted with the use of models (Omernik, 1976). The inclusion of exte rnal components in phosphorus cycling
44 can be complex and require a large amount of data (Soranno et al., 1996), increasing uncertainties and the possib ility of error for the overall model. Research and careful modeling, however can account for this downside and improve the accuracy of the overall estimations. Models for use in conservation and water quality management have predicted increase in the inputs of phosphoru s based on historical trends of land use and human population (Johnes et al ., 1996). Predictions made based upon current increase of urban land cover on Lake Mendota watershed estimated slight increases in annual phosphorus loading but still enough for significant effects on eutrophication. If the entire watershed were urbanized (Soranno et al., 1996), the phosphorus loading would doubl e and the effects in water quality would be severe. Accurate consideration of the trends of phosphorus input into the system will determine the likelihood of the trends of eutrophication or its reversibility. Lakes with long history of heavy inputs of phosphorus would be more difficult to recover or may not be able to recover (Carpenter et al., 1999). This is because the phosphorus that enters into the system is mostly accumulated in the sediments and the biomass fr om where it can be recycled to the water column at a faster rate than what c an be lost from the system (Carpenter, 2005). This recycling of phosphorus can continue for a long time after external inputs have been decreased (Carpenter, 2005). Failure in reaching a goal level for lake water phosphorus concentration projected with a residence time model has been attributed to a recycling of phosphorus from the sediments (Larsen et al., 1979).
45 For this reason internal loadings are more important for predict ions than external loadings unless sediments are removed from the system (Reddy et al., 1999). Yet, probably even more important than re cycling from the sediments is a slow and constant flux of phosphorus from the watershed soil (Carpenter, 2005). This happens as a consequence of accumulation of phosphorus in the soil and is more common in agricultural areas as a consequence of over-fertilization (Bennett et al., 2001). Accord ing to model estimates, when phosphorus inputs are decreased, recovery of eutrophic la kes can be fast if phosphorus recycling from sediments and flux from soils are sl ow, but this recovery can take hundred of years if these two processe s are fast, (Carpenter, 2005). Florida lakes present especial difficul ties for prediction models (Reddy et al., 1999). The flat and low landscape of Flor ida make it difficult for watershed boundaries to be delineated, hence calculations of non-point inputs are difficult. The sandy limestone foundation of the peninsula cause seepage lakes where no surface inflow or outflow can be easily ident ified, complicating the calculations of inputs and outputs of phosphorus. Addition ally, atmospheric inputs of dry phosphorus need to be included in the calc ulations since this represents a significant external source according to what was reported for Reddy et al. (1999).
46 2.12.5 Some Parameters Found in Literature Hence, submerged aquatic vegetat ion should be another important reservoir to be considered in modeling of phosphorus cycling in shallow aquatic systems. Table 2.3 shows some rates fo r important fluxes between water and sediments, periphyton, and phytoplankt on compiled from lit erature. These standard rates can be useful for modelin g efforts addressing the metabolism of phosphorus in shallow lakes. Notice that no rates are included for submerged aquatic vegetation, which reflect the insu fficient inclusion of this component in modeling studies.
47Table 2.3 Standard rates in phosphorus cycling of shallow aquatic systems. Description Type of aquatic system Value in g P m-2 year-1 M:measured C:calculated Source Phytoplankton and periphyton up-take Wetlands 0.12 0.22 C Wang and Mitsch, 2000 Sedimentation rate Wetlands 0.62 1.08 C Wang and Mitsch, 2000 Sedimentation rate Lakes 3.17 M Schauser et al., 2004 Flux by leaching and decomposition of bottom Wetlands 0.20 0.66 C Wang and Mitsch, 2000 Flux from sediments by resuspension Wetlands 0.27 2.47 C Wang and Mitsch, 2000 Flux from sediments by resuspension Lakes 1.46 25.55 M Sondergaard et al., 2004 Flux from sediments Lakes 2.11 C Schauser et al., 2004 Flux from sediments Wetlands 2.37 M Fisher and Reddy, 2001 Flux from sediments Lakes 0.99 M Moore et al., 1991 Flux from sediments Lakes 0.36 M Moore et al., 1998 Flux from sediments Wetlands 3.41 M Lai and Lam, 2008 Flux from sediments Lakes 0.38 M Ogburn (1984) in Reddy et al. (1999) Atmospheric deposition lakes 0.044 0.058 M Ogburn (1984) in Reddy et al. (1999) Sediments accumulation rates Wetlands 6 to 29 mm year-1 C Wang and Mitsch, 2000
48 2.13 Discussion Considering the inverse relationship observed in Chapters 4 and 5 between the prevalence of submerged aqua tic vegetation and the concentration of TP in lake water, it seems more lik ely that the process modelled by GenkaiKato and Carpenter (2005) in which s ubmerged aquatic vegetation function as a suppressor for recycling of phosphorus from sediment back to overlying water is more prevalent than the possible pumping effect from sediments to water modelled by Wang and Mitsh (2000). Th is also agrees with the water quality model of James and Bierman (1995), wh ich calculated that the amount of phosphorus removed by sedimentation fr om water solution in Lake Okeechobee exceeded the net flux of phosphorus from se diments into the overlying water. Table 2.3, however, shows similar rate s reported for phosphor us sedimentation and phosphorus flux from sediments to water. The inverse association between lake water TP concentration and depth also is consistent with the effect of temperature modeled by Genkai-Kato and Carpenter (2005). According to this the hi gher temperature associated to shallow depth favour the chemical reactions lead ing to release of phosphorus from the sediments into the overlying water. This would, subsequently increase the concentration of phosphorus in solution. Unfortunately limit ed information was found about models including the proce sses of phosphorus uptake from sediments and water to submerged aquatic vegetation that could be used to interpret the associations found. It is, therefore, difficult to test for a relationship between submerged aquatic vegeta tion and lake water phosphorus
49 concentration due to the lack of observation al or calculated data for up-take rates from sediments and water to submerged aquatic vegetation as well as release rates from submerged aquatic vegetation to water column. In addition, another difficulty is the lack of literary sources that include all the rates for the three proposed reservoirs in the same system. A simplistic model suggested here to estimate the status of TP concentration in lake water of urban la kes based on the status of submerged aquatic vegetation should be composed by three reservoirs: TP contained in sediments, TP contained in water column, and TP contained in submerged aquatic vegetation. This model would te st assumptions made in literature and results obtained in Chapter 4 and 5 about association and possible causation between submerged aquatic vegetation and c oncentration of phosphorus in lake water. Since phosphorus does not have a stable gaseous form of significance (Chameides and Perdue, 1997; Lahm, 2008; Schlesinger, 1991), this model would not include any reservoir or i nput for phosphorus in gaseous form nor consider any output from the system into th e atmosphere. It is important to make the observation, however, that in lakes were other inputs of phosohorus are not large enough, then phosphorus atmospher ic deposition may be considered significant. Since lakes are open systems with a strong dependence from external inputs and outputs, these need to be included as flows in and out of the system. Among the possible external sources of phos phorus in the lakes with lower levels of total water phosphorus concentrati on may be runoff from urban areas.
50 Although phosphorus is known for not been soluble and readily transported with water, the fact that t he underground soil profile surround ing these lakes is sandy karst (van Beynen et al., 2007) suggests the possibility that leaching of some forms of soluble phosphorus thr ough the underground prof ile may be another external source of phosphorus to the lakes. Note that some of the lakes with higher concentrations of total phosphorus in the water column are located toward the eastern part of the Tampa Bay wate rshed, which also is the part of the watershed that is contained within the ar ea of the Southern Bone Valley (Brown, 2005). Landscape alterations, runoff, groundwat er flow of soluble phosphorus, or discharges from mining activities may explain their high phosphorus levels. Naturally occurring phosphorus, suburban and agricultural land runoff may be the main sources of phosphorus for Lake Thon otosassa, which is the lake with the highest level in water column phosphorus concentration among those examined here. The slightly decreasing trend in to tal phosphorus concentration in the eutrophic and hypereutrophic la kes may be indicative that loading of phosphorus to these lakes was not excessive to t he point that eutrophication could not be reversed. Otherwise due to excessive phosphorus loading, the labels would have produced such accumulation of phosphorus in the system that recycling from sediments and flux from soils were faster than phosphorus loss from the system (Carpenter, 2005; Larsen et al., 1979). Special consideration must be given to the amount of phosphorus loading ov er time since it would determine in great measure the possibility of reco vering (Bennett et al., 2001; Carpenter,
51 2005) and also to the land use in the watershed since it would determine the loading (Johnes et al., 1996; Soranno et al., 1996). A good alternative tool to construct th is model is Stella software, an iconographic computational platform helpf ul to visualise and analyze equations and processes (Costanza and Voinov, 2001), that has been commonly applied for biogeochemical modeling in aquatic systems (Carpenter, 2005; Jorgensen, 2003; Jorgensen et al., 2002; Krivtsov et al., 2000; Tett and Wilson, 1999).
52 CHAPTER 3. TEMPORAL TRENDS IN LAKE WA TER CONCENTRATION OF TOTAL PHOSPHORUS, RATIO OF TOTAL NI TROGEN TO TOTAL PHOSPHORUS, AND CHLOROPHYLLFOR LAKES OF DIFFERENT EUTROPHICATION STATUS IN TAMPA BAY WATERSHED: 1990-2007 3.1 Introduction This chapter provides an analysi s of the nutrient and chlorophyllconcentration trend behavior in a group of lakes of the Tampa Bay watershed, which may prove to be useful indicators of overall watershed trends. These findings will present evidence to further corroborate or contr adict the theory of cultural eutrophication associated wit h watershed development (Smith et al., 2003). On a broader scale, the knowledge of a trend, if one exists, would assist surface water managers and community stak eholders in their efforts to create sustainable development in t he Tampa Bay watershed. The 5,7000 ha Tampa Bay, waters hed lies within the Counties of Hillsborough, Pinellas, and Manatee and extends to parts of Sarasota, Pasco, and Polk Counties. Between 2001 and 203 0, the population within the Tampa Bay watershed is expected to increase by an estimated two million people and 940,000 jobs are projected to be created dur ing the same time period (Tampa
53 Bay Regional Planning Council, 2007). Conc ern over the impact of this regions fast growth on the cultural eutrophication of natural lakes located in the Tampa Bay watershed relates to not only the bodies of water within the watershed, but more broadly, to the possible effects of ecosystem flux on the receiving bay. 3.2 Objectives and Hypotheses Population growth and waters hed development have often been associated with increased nut rient concentration in natur ally occurring bodies of water (Rast and Thorton, 1996; Smith et al ., 2003). Terrell et al. (2000) however, did not find a trend of increasing water concentration of total phosphorus (TP), total nitrogen (TN), and chlorophyllacross 127 Florida lakes during a time period of growing populati on between 1967 and 1997. R eports 305 (b) from the Florida Department of Environmental Pr otection (FDEP), however, have reported over time a decreasing percentage of la kes with a stable water quality. The percentage of lakes with stable levels of eutrophication estate parameters (TP, TN, and chlorophyll) were 71, 58, and 41% for t he years 1996, 2000, and 2006, respectively. The percentage of lakes reporting a degrading trend in eutrophication parameters (increases in TP, TN, and chlorophyll) has increased: 9, 22, and 33% for the year s 1996, 2000, and 2006, respectively. The percentage of lakes showing an improv ing trend in eutrophication parameters (decrease in TP, TN, and chlorophyll) has also increased, but to a lesser extent: 20, 20, and 26% for the year s 1996, 2000, and 2006, respectively. The
54 number of lakes assessed by the FD EP for the above mentioned reports was 627, 541, and 358, respectively (FDEP, 1996; 2000; 2006). With those different findings as a background, the follo wing objective and hypothesis are addressed in this chapter for lakes in the Tampa Bay watershed: Objective: To determine if there is a change in lake water concentration of eutrophication-rela ted parameters netween 1990 and 2007. Null Hypothesis (Ho) : Lake water concentration of total phosphorus (TP), ratio of total nitrogen to total phosphorus (TN:TP), and chlorophylldid not change between 1990 to 2007. Alternate Hypothesis (Ha) : Lake water concentration of total phosphorus (TP), ratio of total nitrogen to total phosphorus (TN:TP), and chlorophylldid change between 1990 to 2007. 3.3 Methods 3.3.1 Data Gathering and Sampling Methods To examine the temporal variability of water chem istry, existing record data for lake water concentrations of TN, TP, and chlorophyllon 16 lakes located in the Tampa Bay watershed were compiled. Out of 649 lakes located in Hillsborough, Pinellas, Manat ee, and Polk Counties, for which information is provided by the Water Atlas (2008), web site of the Florida Center for Community Design and Research at the University of South Florida, only 16 lakes met both
55 inclusion criteria: containment wit hin the Tampa Bay watershed and data availability for at least 75% of the 18 year study period (1990 to 2007). The locations of the lakes chosen for this analysis are represented graphically in Figure 3.1. For more info rmation about the lakes, refer to Appendix A. All of the data analyzed in this study were obtained from the Water Atlas (2008). Most of the water samples we re collected by citizen volunteers sponsored by the water quality monito ring program, LAKEWATCH, and these samples were analyzed in the laborator y of the Department of Fisheries and Aquatic Sciences at the University of Fl orida. The methods used to collect the data are described in Brown et al. ( 1998). TP was determined by oxygenating phosphorus with potassium persulfate (Me nzel and Corwin, 1965) and measuring the liberated phosphorus with the colorime tric technique of Murphy and Riley (1962) as cited in Menzel and Corwin (1965). TN was determined by a persulfate oxidation techniqu e (D'Elia et al., 1977) follo wed with nitrate-nitrogen determination by ultraviolet derivative s pectroscopy of second order (Bachmann and Canfield, 1996; Simal et al., 1985; Wollin, 1987). Determination of chlorophyllwas done by extracting the pigment with ethanol (Sartory and Grobbelar, 1984) and then measuring it with spectrophotometry following the Standard Method (SM) 10200 H me thod (APHA, 1989; 1998). Water samples from Lake Thonotosassa were collected and analyzed by the Environmental Protection Commi ssion of Hillsborough County using a combination of EPA and APHA Standard Methods. TP was determined by EPA 365.4; and TN was the sum of Total Kjel dahl Nitrogen (TKN) a nd nitrate/nitrite
56 nitrogen, where Total Kjeldahl Nitr ogen (TKN) was determined by EPA 351.2 while SM 4500 NO3 F (APHA, 1989; 1998) wa s used for nitrate/nitrite nitrogen. Chlorophyllwas determined by SM 10200 H (APHA, 1989; 1998). Samples from Behula, Bonnet, and Hunter Lakes were collected by the City of Lakeland Division of Lakes and St ormwater and analyzed by the City of Lakeland Wastewater Laboratory. The method used for TP analysis was EPA 365.4; and for TN analysis the methods were EPA 353.2 for nitrate and PAI DK03 (method approved by the Environment al Protection Agency) for Total Kjeldahl Nitrogen (TKN). For chlorophyll, a modification of Standard Methods 10200ha (APHA, 1998) was used. Data for Ward Lake came from STORET, a computerized environmental database of United States Geological Survey, which was also was made available by Water Atlas (2008). 3.3.2 Statistical Analysis The total data set of values for TP, TN:TP, and chlorophyllfrom each lake were used over the 18-year time period. Values for TN were analyzed relative to TP. The importance of lake water TN concentration as a limiting factor depends on its abundance relative to that of lake water TP. Therefore, a TN:TP ratio is more meaningful in regard to the potential for eutrophication and likelihood of algal abundance. According to the nutrient limitation criteria based on Br ezonick (1984), ratios greater than 30 correspond to phosphorus-limited lakes while ratios less than 10 correspond to
57 nitrogen-limited lakes. Nutri ent limitation in lakes is balanced (both nutrients are limiting) if TN:TP ratio is between 10 and 30. The 16 lakes studied to assess the overall trend of TP, TN:TP, and chlorophyllwere sorted in two groups depending on the overall mean of lake water TP concentration for each lake during the study period. The grouping criteria followed the Trophic State Classifi cation System of Forsberg and Ryding (1980) (Fig 3.1). This system suggests uses for surface waters based on the water nutrient concentration and not necessarily implying an adverse effect; however, as discussed in Chapter 1, elevated nutrient levels have been associated with adverse human and environ mental health consequences. The Florida Department of Envir onmental Protection Agency (FDEP) among other parameters considers surface waters as good when TP concentrations are between 0 to 64 mg L-1, fair if TP concentrations range from 65 to 112 mg L-1, and poor for TP concentrations between 112 and 567 mg L-1. The first group of lakes was co mprised of 10 lakes exhibiting both oligotrophic and mesotrophic conditions (less than 25 g TP L-1), while the second group comprised 6 hypereutrophic lakes (more than 100 g TP L-1). The first group were located in the northwest ern corner of Hills borough County, an area that drains into Old Tampa Bay according to Lewis and Estevez (1988) (Figure 1.1). Lakes in the second group we re found in more distant parts of the Tampa Bay watershed, for example, in Hillsborough, Pinellas, Polk, and Manatee Counties. Figure 3.2 shows all the lakes included in this analysis.
58 Since the objective of this research was to characterize patterns of response and change in a variable (TP, TN:TP, or chlorophyll) over time, this analysis falls under the definition of longit udinal research (Ware, 1985). As it is usual in most longitudinal studies, this one violates some assumptions required for standard regression analysis. Measur ements taken over time were not independent, residuals were not normally di stributed, and data were not collected at a constant set of time points and many values are missing: characteristics that make standard regression analysis inapplicab le (Lin and Ying, 2003; Ware, 1985; Zeguer et al., 1988). For each parameter in each trophic group of lakes, however, least squares linear regressions were plotted as a preliminary visual assessment of the overall 18-year trends. Points that were 5 times greater or smaller than the mean and that individually influenced the mean value of the eutrophication parameter were removed as high influence points. The to tal numbers of points removed by this method were 7 out of 2101 for TP and 2 out of 1707 for chlorophyllA randomized complete block design (Ott, 1993) was used for this analysis. Period was the treatment. To a ccount for lake e ffect, the data was blocked by lake. The response was lake water concentration of TP, TN:TP ratio, and chlorophyll, and the experimental unit was lake water in Tampa Bay watershed. It was an assumption that samples were collected at random (randomly distributed in time). The null hypot heses for each of the sets are given below: 1. The means of the periods are equal.
59 2. The means of t he lakes are equal. 3. There is no interaction between periods and lakes. Lakes were first separated into lo w and high eutrophication groups based on TP concentration, as previously discussed. A two-way analysis of variance (A NOVA) was conducted with SYSTAT to test the research hypotheses. The fi rst factor was period. The data set for each trophic group was divided in two peri ods: Period 1 extended from 1990 until 1998 and Period 2 from 1999 until 2007. The second factor was lake (lake drainage basin). ANOVA is a tool typically applied to controlled experiments rather than to "natural" experiments; howev er, ANOVA has been used in natural studies to test for changes in TP (Smith and McCormick, 2001) and mercury (Babiarz et al., 1998) concentration over time or space. T-test considering unequal variance s was applied to detect differences between periods within each lake. A significant level of = 0.05 was used to perform all the statistical tests. Annual averages of TP, TN:TP ratio, and chlorophyllconcentrations across all lakes were plotted against t he corresponding year to easily visualize cycling in the data. Regression coefficient s that represent t he population of data rather than the original specific data are called population-averaged as described by Zeguer et al. (1988). A
60 Annual estimates of population data for the Tampa Bay Metropolitan Area and Hillsborough County were tabulated by year to appreciate the annual growth in population. Figure 3.1 Trophic State Classification System (Forsberg and Ryding, 1980). Concentrations of constituents for each trophic state, and typical uses of waterbodies. http://www.hillsborough.wateratlas.usf.edu
61Figure 3.2 Lakes analyzed for trends in trophic state variables during the time period 1990 to 2007. Lakes are shown in red. Lakes within the circle are oligotrophic and mesotrophic. ! ! ! ! ! ! ! ! ! ! Legend!Lakes Tampa Bay and Coastal Areas Alafia River Basin Hillsborough River Basin Little Manatee River Basin Manatee River Basin Pasco County Pinellas County Hillsborough County Polk County Sarasota County Manatee CountyMap Created by Pete Reehling University of South Florida
62 3.4 Results and Discussion Results from the two-way ANOVA showed a significant difference between means of lakes in all the groups for all the trophic state variables considered ( p <0.0001). Interactions were found between periods and lakes in all the groups for all the trophic state variables considered ( p <0.003). Over the 18year study period, over all average annual TP concentrations for the ten oligotrophic and mesotrophic lake s ranged from 10.47 to 19.09 g L-1. Individual lake values ranged from 2.00 to 59.00 g L-1 (Table 3.1). Many values in the upper annual range exceeded the limit of 25 g L-1 of TP used as trophic class separation. However, the means for eac h one of these lakes remained within the 25 g L-1 limit according to the criteria. The overall annual average of TP concentration increased from 11.71 g L-1 in 1990 to 15.00 g L-1 in 2007 (Table 3.1). The best-fit line for a plot of original data suggested a weak correlation ( r = 0.3) and a positive slope (F igure 3.3). A significant difference was found with twoway ANOVA between the means of the first and second periods ( p < 0.0001). The t-tests performed on TP concentration for each lake between Periods 1 and 2 showed a statistically significant difference ( p 0.05) in all the lakes of this group except for two. The fact that each lake did not increase at the same rate between period 1 and 2 was evidence of si gnificant interaction (Figure 3.4). The average annual TP concentration for the group of hypereutrophic lakes ranged from 456.47 g L-1 in 1997 to 158.29 g L-1 in 2007 (Table 3.2). The individual values rang ed from 10 to 2300 g L-1. The best-fit line for a plot of original data showed a low coefficient of determination and a slightly decreasing
63 slope (Figure 3.5). Two-way ANOVA show ed significant differences between the two periods ( p = 0.03). Such results were the opposite of the increasing trend suggested by the group of oligotrophi c and mesotrophic lakes for the same parameter and time period. The t-tests performed for each lake of this group for TP concentration found no statistical signif icant difference between the first and second period in two lakes ( p > 0.05), and a significant increase in one lake ( p 0.05). Since the effects of period on lake water TP concentration did not remain the same for different lakes then t here was interaction (Figure 3.6). For the group of hypereutrophic lakes, reasons for changes in lake water TP concentration may be related to implem entation of lake management plans, lake restoration, and storm water treatm ent projects (City of Lakeland, 2001; Southwest Florida Water Management District, 2003). The plan of water improvement for Lake Thonotosassa, for in stance, had a reduction in phosphorus loadings from point sources since the Sno-Man seafood processing plant closed in 1992 and the Plant City Wastewater Tr etament Plant ceased discharges in 1997. In regards to non-point sources, a 21-ha marsh was constructed to intercept water from Baker Creek before ent ering the lake. A strategy to control exotic submerged and floating plants, such as water hyacinth ( Eichhornia crassipes ) and water lettuce (Pistia stratiotes, has been implemented to allow penetration of light for s ubmerged aquatic vegetation s pecies. Such programs for lake improvement may have given priori ty to those lakes with the worse water quality in order to maximize the efficien cy of limited funding. The reduction in
64 lake water TP concentration observed in these highly eutrophied lakes demonstrate the efficacy of lake management plans. Differences in the amplitude of TP concentrations between oligotrophic and mesotrophic lakes and hypereutrophic lakes were seen. Standard deviation (SD) in TP concentrations for oligotr ophic and mesotrophic lakes ranged from 3.77 to 10.09 g L-1 (Table 3.1) and coefficient of variation (CV) from 27% to 60%, while for hypereutrophic lakes SD w ent from 72.35 to 482.01 (Table 3.2) and CV from 45% to 105%. Table 3.1 Summary statistics for annual TP concentration in g L-1 collected by LAKEWATCH from 10 oligotrophic and mesotrophic lakes in two counties from 1990 through 2007 Year Annual Average (g L-1) SD (g L-1) Minimum (g L-1) Maximum (g L-1) n (Lakes) n (Samples) 1990 11.71 3.71 5.00 23.00 7 24 1991 14.14 4.91 7.00 29.00 8 68 1992 12.66 5.02 5.00 25.00 10 88 1993 10.51 4.24 4.00 21.00 9 92 1994 10.47 5.05 2.00 25.00 8 73 1995 10.64 5.55 3.00 27.00 9 72 1996 10.63 4.48 3.00 25.00 9 69 1997 14.16 3.39 6.00 24.00 10 74 1998 16.17 7.02 7.00 48.00 9 68 1999 13.96 3.77 5.00 21.00 8 70 2000 13.97 3.43 9.00 29.00 9 73 2001 13.52 4.82 6.00 34.00 9 78 2002 14.12 7.28 4.00 43.00 10 91 2003 17.40 10.59 5.00 59.00 10 86 2004 19.09 6.90 7.00 35.00 10 74 2005 18.93 7.83 8.00 36.00 9 59 2006 17.30 7.29 7.00 34.00 8 41 2007 15.00 5.90 6.00 29.00 6 32
65 Figure 3.3 Plot of TP concentration for 10 oligotrophic and mesotrophic lakes. y = 0.0012x 28.603 r = 0.310 10 20 30 40 50 60 70Jan-90 Jan-91 Jan-92 Dec-92 Dec-93 Dec-94 Dec-95 Dec-96 Dec-97 Dec-98 Dec-99 Dec-00 Dec-01 Dec-02 Dec-03 Dec-04 Dec-05 Dec-06Month-YearTP g (L-1) As mentioned earlier, analysis of TN is expressed here as a ratio to TP since phosphorus is the nutrient limiting nat ural productivity in most of these lakes and because cyanobacteria abundance is a concern under conditions of high lake water TP concentration relative to lake water TN concentration. The overall group of 10 lakes, oligotrophi c and mesotrophic, presented annual mean TN:TP ratio slightly decreasing from 55. 35 in 1990 to 44.14 in 2007 (Table 3.3) and ranged from 35.82 to 55.35 (Table 3.3). Individual values ranged from 12.41 to 132.00, the annual SD r anged from 10.04 to 18.14 and CV from 26% to 33%. A plot of TN:TP ratios for individual va lues shows a decreasing line with a weak correlation ( r = 0.26, Figure 3.7) for this group of lakes.
66Figure 3.4 Change in TP concentration in 10 oligotrophic and mesotrophic lakes between Periods 1 and 2. Least Squares MeansArmistead 1 2 PERIOD 2 12 22 32T P L O W Calm Lake 1 2 PERIOD 2 12 22 32T P L O W Carroll 1 2 PERIOD 2 12 22 32T P L O W Crenshaw 1 2 PERIOD 2 12 22 32T P L O W Deer 1 2 PERIOD 2 12 22 32T P L O W Hiawatha 1 2 PERIOD 2 12 22 32T P L O W Juanita 1 2 PERIOD 2 12 22 32T P L O W Keystone 1 2 PERIOD 2 12 22 32T P L O W Magdalene 1 2 PERIOD 2 12 22 32T P L O W Sunset 1 2 PERIOD 2 12 22 32T P L O W Two-way ANOVA confirmed this trend by finding a significant difference between the first and second periods, being lower in the second ( p < 0.001). Even though all the 10 lakes showed a decreas e in TN:TP ratio from the first to the second period, when tested individual ly by t-test, this change was not statistically significant (p > 0.05) for four lakes. The effect of period on TN:TP ratio was not consistent between lakes in this group (Figure 3.8), indicating interaction between period and lake. The overall decreasing trend in TN:TP ratio might indicate that TP concentration has increased at a faster pace than TN
67 concentration. As the plot of ratios s hows, all lakes in this group are included within the phosphorus lim itation criteria of Brezonick (1984), that is, all ratios 30 (Figure 3.7). However, if the downward trend continues, it would result in a gradual tendency from being phosphorus -limited toward being phosphorusand nitrogen-limited. Table 3.2 Summary statistics for annual TP concentration in g L-1 collected by LAKEWATCH from 6 hypereutrophic lakes in three counties from 1990 through 2007 Year Annual Average (g L-1) SD (g L-1) Minimum (g L-1) Maximum (g L-1) n (Lakes) n (Samples) 1990 456.47 482.01 10.00 2300.00 6 53 1991 311.98 251.78 57.00 890.00 6 53 1992 463.32 449.27 67.00 2290.00 5 28 1993 247.03 155.21 51.00 680.00 6 37 1994 254.53 157.27 64.00 1010.00 6 40 1995 287.53 121.68 50.00 608.00 6 60 1996 263.10 121.50 50.00 504.00 6 61 1997 298.72 251.06 20.00 1800.00 6 58 1998 375.88 328.34 60.00 2000.00 6 57 1999 306.90 277.60 30.00 1940.00 6 58 2000 284.15 199.55 90.00 1050.00 6 55 2001 260.39 109.76 80.00 508.00 6 57 2002 228.64 90.34 68.00 483.00 6 42 2003 300.49 227.86 39.00 1620.00 6 53 2004 304.04 201.79 80.00 1030.00 6 52 2005 250.98 144.73 54.00 727.00 6 49 2006 191.78 120.64 47.00 638.00 6 36 2007 158.29 72.35 51.00 296.00 5 17
68 Figure 3.5 Plot of TP concentration for 6 hypereutrophic lakes. y = -0.0203x + 1026 r = 0.150 500 1000 1500 2000 2500Jan-90 Jan-91 Jan-92 Dec-92 Dec-93 Dec-94 Dec-95 Dec-96 Dec-97 Dec-98 Dec-99 Dec-00 Dec-01 Dec-02 Dec-03 Dec-04 Dec-05 Dec-06Month-YearTP g (L-1) Figure 3.6 Change in TP concentration in 6 hypereutrophic lakes between Periods 1 and 2. Least Squares MeansAlligator 1 2 PERIOD 49.0 144.6 240.2 335.8 431.4 527.0T P H I G H Beulah 1 2 PERIOD 49.0 144.6 240.2 335.8 431.4 527.0T P H I G H Bonnet 1 2 PERIOD 49.0 144.6 240.2 335.8 431.4 527.0T P H I G H Hunter 1 2 PERIOD 49.0 144.6 240.2 335.8 431.4 527.0T P H I G H Thonotossasa 1 2 PERIOD 49.0 144.6 240.2 335.8 431.4 527.0T P H I G H Ward Lake 1 2 PERIOD 49.0 144.6 240.2 335.8 431.4 527.0T P H I G H
69 Low TN:TP ratios could lead to increase of cyanobacterial water concentration (Hecky and Kilham, 1988; Levich, 1996; Levich and Bulgakov, 1992). This group of organisms produc es potent toxins that have been associated to harmful algal blooms (USEPA, 1997) and possibly an increased risk of primary hepatocellular carc inoma (Fleming et al., 2002). As a consequence, control of nitrogen would be increasingly more important in the formulation of environmental management plans for these lakes. Only 5 out of the initial 6 hypereut rophic lakes were examined for TN:TP ratio because no TN data were available for one of them. Data for this ratio plotted for this group of lakes ( r = 0.15, Figure 3.9), show a slightly decreasing slope within the nitrogen limitation zone. This regression line suggests a significant ( p < 0.01) weak negative tendency over time which was confirmed by two-way ANOVA ( p < 0.001). Individual t-tests, however, detected a statistically significant change of increase ( p 0.05) in only two of the lakes. There was interaction for this group too because each lake did not increase in a consistent way between Period 1 and Period 2 (Figure 3.10). For hypereutrophic lakes, TN:TP ratios ranged from 0.41 to 53.92 (Table 3.4) and overall annual average ratios ranged from 4.66 to 15.89 (Table 3.4). These ratios are at the interf ace between balanced (both nitrogenand phosphorus -imited) and nitrogen-limited (T N:TP = 10) primary productivity. The lower TN:TP ratio found in hypereutrophic lakes as compared to oligotrophic and mesotrophic lakes suggested different cr iteria for lake management options
70 between these two groups of lakes. The SD of TN:TP ratios ranged from 2.7 to 10.74 and CV from 47% to 101%. Table 3.3 Summary statistics for the a nnual ratio of TN:TP ratio collected by LAKEWATCH from 10 oligotrophic and mesotrophic lakes in two counties from 1990 through 2007 Year Annual Average SD Minimum Maximum n (Lakes) n (Samples) 1990 55.35 14.69 30.77 101.43 6 24 1991 45.82 10.90 20.00 78.57 8 68 1992 47.22 14.87 21.43 111.25 10 88 1993 53.53 18.14 21.43 132.00 9 92 1994 55.13 16.36 28.24 111.67 8 73 1995 52.90 17.92 20.00 120.00 9 72 1996 49.60 13.49 23.60 90.00 9 69 1997 40.43 11.69 19.00 79.00 10 74 1998 40.10 12.84 15.56 71.43 9 68 1999 43.13 16.40 21.90 110.00 8 70 2000 42.78 10.70 23.50 67.78 9 73 2001 44.52 13.67 16.84 90.00 9 78 2002 43.18 15.73 19.23 112.50 10 91 2003 40.29 14.36 15.93 95.00 10 86 2004 35.82 11.47 12.41 78.00 10 75 2005 37.96 10.04 21.18 67.00 9 59 2006 41.99 12.89 15.42 75.56 8 42 2007 44.14 17.57 26.67 116.67 6 32 The temporal distribution of dat a points plotted for chlorophyllconcentrations from oligotrophic and meso trophic lakes indicated an increasing trend (Figure 3.11, r = 0.23). These results were consistent with the two-way ANOVA ( p < 0.001). The t-te sts of chlorophyllconcentrations for each lake revealed that concentrations in 3 of t he 8 lakes showed no significant change between Period 1 and Period 2 ( p > 0.05) and confirmed an interaction between periods and lakes (Figure 3.12).
71 The study period in this group of lakes started with an overall annual average of 6.52 g L-1 in 1990 and ended with an average of 5.27 g L-1 in 2007 (Table 3.5). Averages values ranged from 4.38 to 13.80 g L-1 and lake values ranged from 1.00 to 48.00 g L-1 (Table 3.5) with SD rangi ng from 3.17 in 2007 to 11.05 in 2003 (Table 3.5) and CV from 60% to 101% for the same years. Figure 3.7 Plot of TN:TP ratio fo r 10 oligotrophic and mesotrophic lakes. y = -0.0023x + 129.2 r = 0.260 20 40 60 80 100 120 140Jan-90 Jan-91 Jan-92 Dec-92 Dec-93 Dec-94 Dec-95 Dec-96 Dec-97 Dec-98 Dec-99 Dec-00 Dec-01 Dec-02 Dec-03 Dec-04 Dec-05 Dec-06Month-YearTN : TP ratio For hypereutrophic lakes, no overall trend for lake water chlorophyllconcentrations was apparent from the visual analysis ( r = 0.04, Figure 3.13). Preliminary results from r egression analysis were confirmed by two-way ANOVA, which found no significant difference betwe en the first and second periods in the concentration of chlorophyllin the lake water of these hypereutrophic lakes ( p = 0.717). The t-tests conduc ted on individual lakes sh owed that 3 lakes had a statistically significant increase ( p 0,5) in chlorophyllconcentrations one lake showed a significant decrease, and one more showed no significant change. The
72 fact that each lake did not behave in t he same way between periods 1 and 2 was evidence of interaction (Figure 3.14). Lake chlorophyllconcentrations had a wide range from 1.00 to 643.50 g L-1, and a much higher overall SD (Table 3.6) from 44.05 to 150.53 and CV from 69 to 128% as compared with the less eutrophied lakes. The overall annual averages ranged from 47.59 in 1990 to 116.70 g L-1 in 2007 (Table 3.6). Figure 3.8 Change in the TN:TP ratio in 10 oligotrophic and mesotrophic lakes between Periods 1 and 2. Least Squares MeansArmistead 1 2 PERIOD 22.0 31.2 40.4 49.6 58.8 68.0T N T P L O W Calm Lake 1 2 PERIOD 22.0 31.2 40.4 49.6 58.8 68.0T N T P L O W Carroll 1 2 PERIOD 22.0 31.2 40.4 49.6 58.8 68.0T N T P L O W Crenshaw 1 2 PERIOD 22.0 31.2 40.4 49.6 58.8 68.0T N T P L O W Deer 1 2 PERIOD 22.0 31.2 40.4 49.6 58.8 68.0T N T P L O W Hiawatha 1 2 PERIOD 22.0 31.2 40.4 49.6 58.8 68.0T N T P L O W Juanita 1 2 PERIOD 22.0 31.2 40.4 49.6 58.8 68.0T N T P L O W Keystone 1 2 PERIOD 22.0 31.2 40.4 49.6 58.8 68.0T N T P L O W Magdalene 1 2 PERIOD 22.0 31.2 40.4 49.6 58.8 68.0T N T P L O W Sunset 1 2 PERIOD 22.0 31.2 40.4 49.6 58.8 68.0T N T P L O W
73Table 3.4 Summary statistics for annual TN :TP ratio collected by LAKEWATCH from 5 hypereutrophic lakes in three counties from 1990 through 2007 Year Annual Average SD Minimum Maximum n (Lakes) n (Samples) 1990 13.15 9.20 0.84 37.14 5 48 1991 12.83 9.33 1.96 38.43 5 53 1992 10.54 10.74 0.42 51.64 4 24 1993 13.19 9.02 2.18 35.29 5 25 1994 9.07 5.17 3.09 24.06 5 27 1995 5.67 4.24 0.41 18.47 5 48 1996 6.66 5.54 1.88 25.96 5 48 1997 7.02 5.56 1.06 23.53 5 46 1998 4.66 2.69 0.35 12.10 5 46 1999 5.89 2.95 0.38 13.15 5 41 2000 9.17 8.86 1.13 48.95 5 40 2001 8.24 5.73 0.72 34.58 5 40 2002 9.21 4.52 2.08 20.39 5 30 2003 7.01 8.05 0.55 53.92 5 40 2004 5.70 2.70 1.66 13.28 5 29 2005 8.90 4.48 2.95 19.87 5 27 2006 11.33 4.81 4.29 21.00 5 22 2007 15.89 7.95 10.48 33.04 4 8 Figure 3.9 Plot of TN:TP ratio for 5 hypereutrophic lakes. y = -0.0006x + 30.866 r = 0.150 10 20 30 40 50 60Mar-90 Mar-91 Feb-92 Feb-93 Feb-94 Feb-95 Feb-96 Feb-97 Feb-98 Feb-99 Feb-00 Feb-01 Feb-02 Feb-03 Feb-04 Feb-05 Feb-06 Feb-07Month-YearTN:TP ratio
74 Figure 3.10 Change in the TN:TP ratio in 5 hypereutrophic lakes between Periods 1 and 2. Least Squares MeansBeulah 1 2 PERIOD 0.0 5.5 11.0 16.5 22.0T N T P H I G H Bonnet 1 2 PERIOD 0.0 5.5 11.0 16.5 22.0T N T P H I G H Hunter 1 2 PERIOD 0.0 5.5 11.0 16.5 22.0T N T P H I G H Thonotossasa 1 2 PERIOD 0.0 5.5 11.0 16.5 22.0T N T P H I G H Ward Lake 1 2 PERIOD 0.0 5.5 11.0 16.5 22.0T N T P H I G H Table 3.5 Summary statistics for chlorophyllin g L-1 collected by LAKEWATCH from 8 oligotrophic and mesotrophic lakes in two counties from 1990 through 2007 Year Annual Average (g L-1) SD (g L-1) Minimum (g L-1) Maximum (g L-1) n (Lakes) n (Samples) 1990 6.52 5.38 2.00 22.00 6 25 1991 8.79 7.36 1.00 34.00 7 71 1992 5.18 3.93 1.00 17.00 8 77 1993 3.83 2.67 1.00 11.00 7 71 1994 5.32 4.50 1.00 20.00 7 59 1995 5.32 5.51 1.00 23.00 7 57 1996 4.38 3.60 1.00 18.00 7 55 1997 6.69 3.84 1.00 18.00 7 64 1998 8.00 4.83 2.00 24.00 7 55 1999 6.76 3.96 1.00 20.00 6 62 2000 6.31 4.54 2.00 25.00 8 64 2001 5.73 4.63 1.00 24.00 8 73 2002 6.84 6.70 1.00 30.00 8 81 2003 10.85 11.05 1.00 48.00 8 73 2004 9.87 7.03 2.00 36.00 8 69 2005 13.80 10.98 2.00 41.00 7 55 2006 11.42 8.78 2.00 39.00 6 38 2007 5.27 3.17 2.00 11.00 5 26
75Figure 3.11 Plot of chlorophyllconcentration for 8 oligotrophic and mesotrophic lakes. y = 0.0009x 25.104 r = 0.230 10 20 30 40 50 60Mar-90 Feb-91 Jan-92 Jan-93 Dec-93 Dec-94 Nov-95 Nov-96 Oct-97 Oct-98 Sep-99 Sep-00 Aug-01 Aug-02 Jul-03 Jul-04 Jun-05 Jun-06 May-07Month-YearChlorophyll(g L-1) Table 3.6 Summary statistics for annual Chlorophillconcentration in g L-1 collected by LAKEWATCH from 5 hypereutrophic lakes in three counties from 1990 through 2007 Year Annual Average (g L-1) SD (g L-1) Minimum (g L-1) Maximum (g L-1) n (Lakes) n (Samples) 1990 116.70 150.53 1.00 643.50 4 46 1991 80.59 73.33 20.30 447.40 5 48 1992 90.52 76.80 5.13 338.16 5 28 1993 52.42 48.26 3.02 185.76 5 36 1994 54.61 59.57 3.23 236.20 5 37 1995 63.72 44.05 5.10 151.41 5 36 1996 79.27 76.23 1.70 279.50 5 36 1997 60.85 62.55 0.50 241.00 5 35 1998 58.88 66.84 1.00 325.00 5 35 1999 70.78 55.79 1.70 273.79 5 37 2000 92.78 78.08 10.40 263.68 5 35 2001 73.92 60.60 3.30 213.33 5 36 2002 72.41 56.16 7.80 182.50 5 36 2003 84.00 54.55 12.00 194.10 5 32 2004 83.04 71.82 8.40 231.80 5 33 2005 88.56 73.22 2.40 366.80 5 41 2006 68.01 85.51 4.40 383.00 5 26 2007 47.59 47.58 6.70 139.60 4 15
76Figure 3.12 Change in chlorophyllconcentration in 8 oligotrophic and mesotrophic lakes between Periods 1 and 2. Least Squares MeansArmistead 1 2 PERIOD 0 4 8 12 16C L A L O W Calm Lake 1 2 PERIOD 0 4 8 12 16C L A L O W Crenshaw 1 2 PERIOD 0 4 8 12 16C L A L O W Hiawatha 1 2 PERIOD 0 4 8 12 16C L A L O W Juanita 1 2 PERIOD 0 4 8 12 16C L A L O W Keystone 1 2 PERIOD 0 4 8 12 16C L A L O W Magdalene 1 2 PERIOD 0 4 8 12 16C L A L O W Sunset 1 2 PERIOD 0 4 8 12 16C L A L O W Lake water concentrations of TP, TN:TP ratios, and chlorophyllwere averaged and plotted by year for the interval 1990-2006. These results were consistent with those from analysis done upon the entire cloud of single data points in terms of the direction and signifi cance of the association. An additional benefit of using the averages instead of the original data is a more clear visualization of a possible cycling effect for all the variables in both groups of lakes.
77Figure 3.13 Plot of chlorophyllconcentration for 5 hypereutrophic lakes. y = -0.0017x + 135.34 r = 0.040 100 200 300 400 500 600 700Jan-90 Jan-91 Jan-92 Jan-93 Jan-94 Jan-95 Jan-96 Jan-97 Jan-98 Jan-99 Jan-00 Jan-01 Jan-02 Jan-03 Jan-04 Jan-05 Jan-06 Jan-07Month-YearChlorophyll(mg L-1) Figure 3.14 Change in chlorophyllconcentration in 5 hypereutrophic lakes between periods 1 and 2. Least Squares MeansAlligator 1 2 PERIOD -6 56 118 180C L A H I G H Beulah 1 2 PERIOD -6 56 118 180C L A H I G H Bonnet 1 2 PERIOD -6 56 118 180C L A H I G H Hunter 1 2 PERIOD -6 56 118 180C L A H I G H Thonotosassa 1 2 PERIOD -6 56 118 180C L A H I G H
78 Plots of annually averaged values against time showed a strong direct increase of TP and chlorophyll( r = 0.78 and 0.68 respectively, p <0.01 and p = 0.02, respectively, Figures 3.17 and 3.19). Figure 3.15 Population growth estimates in the Tampa Bay Metropolitan Area from 1990 until 2006 (Hilssborough County, 2007; US Bureau of the Census 2000, 2007). 0 500,000 1,000,000 1,500,000 2,000,000 2,500,000 3,000,0001990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 Population in millions Figure 3.16 Population growth estimates in Hillsborough County from 1990 until 2006 (Hilssborough County, 2007; US Bureau of the Census 2000, 2007). 0 200,000 400,000 600,000 800,000 1,000,000 1,200,000 1,400,0001990 1992 1994 1996 1998 2000 2002 2004 2006 Population in m illions Rising lake water TP and chlorophyllconcentrations in northeastern Hillsborough County appear to trend al ong with County population growth and
79 may be evidence of an urban signature in lake water quality. Figures 3.15 and 3.16 show a continuous increase in population for the Tampa Bay Metropolitan Area (TBMA) and Hillsborough County: a 30.5% and 38.8% increase in the estimated populations of Tampa Bay Me tropolitan Area and Hillsborough County, respectively, for the time period between 1990 and 2006 (Hilssborough County, 2007; US Bureau of the Census 2000, 2007) Figure 1.2 shows that residential and commercial uses are prevalent in the area. This might also have some association with the results. The increase in lake water TP and chlorophyllconcentrations in northeastern Hillsborough County may be re lated the characteristic karst formation of the area (van Beynen et al ., 2007), which may help the transport of nutrients along with the groundwater flow fr om urban sources to the lakes or for some lakes as a consequence of the dr ought and over-pumping of the Floridian Aquifer, and thus loss of a relatively cl ean water supply. This study, however, cannot conclude the increase in TP concentrations in oligotrophic and mesotrophic lakes is due to new phosphorus been discharged into the lakes or to old phosphorus recycled from the sediments. The historically-elevated lake water TP concentrations in lakes located in the eastern part of the Tampa Bay wate rshed may be explained in part by the mining activity in the Bone Valley. This economical acti vity has been operating in the area since the late 1800s (Brown, 2005) (Figure 1.2)
80Figure 3.17 Change in annual averages of lake water TP concentration over time in years in oligotrophic and mesotrophic lakes of the Tampa Bay watershed. r = 0.780 5 10 15 20 25 199019921994199619982000200220042006YearTP ( g L-1) Figure 3.18 Change in annual averages of the TN:TP over time in years in oligotrophic and mesotrophic lakes of the Tampa Bay watershed. r = -0.770 10 20 30 40 50 60 199019921994199619982000200220042006YearTN:TP ratio
81Figure 3.19 Change in annual averages of lake water chlorophyllconcentration over time in years in oligotrophic and mesotrophic lakes of the Tampa Bay watershed. r = 0.680 2 4 6 8 10 12 14 16 199019921994199619982000200220042006YearChlorophyll( g L-1) The plot of the TN:TP for oligotrophic and mesotrophic lakes suggests a strong decline with time ( r = -0.77, p < 0.01, Figure 3.18). This is again consistent with the result from previ ous analyses, which was explained by a greater trend of increasing lake water TP concentration as compared with TN concentration along the time per iod studied (Figure 3.7).
82Figure 3.20 Change in annual averages of lake waterTP concentration over time in years in hypereutrophic lakes of the Tampa Bay watershed. r = -0.580 50 100 150 200 250 300 350 400 450 500 199019921994199619982000200220042006YearsTP ( g L-1) Figure 3.21 Change in annual averages of the TN:TP over time in years in hypereutrophic lakes of the Tampa Bay watershed. r = -0.410 2 4 6 8 10 12 14 199019921994199619982000200220042006YearTN:TP ratio
83Figure 3.22 Change in annual averages of lake water chlorophyllconcentration over time in years in hypereutrophic lakes of the Tampa Bay watershed. r = -0.060 20 40 60 80 100 120 140 199019921994199619982000200220042006YearChlorophyll(g L-1) In the group of hypereutrophic la kes, annually averaged lake water concentrations of TP showed similar t endencies as with linear regressions with the individual data points and ANOVA.. This curve suggested a significant medium inverse associations ( r = -0.58, p < 0.01 for total phosphorus, Figure 3.20). Unlike results from the analysis on the individual data points, the analysis of annual average TN:TP ratios indicated a medium association ( r =-0.41) but it was not significant ( p > 0.05, Figure 3.21). Re sults for chlorophylloncentrations against years were consistent with prev ious analyses and showed no significant correlation ( r = -0.06, Figure 3.22).
84 3.5 Summary The results from this chapter indica te that 16 lakes in the Tampa Bay watershed had a small but significant change in the concentration of trophic state variables between 1990 and 2007. There was a different behavior in water quality trends depending upon the degree of lake water eutrophicatuion. Concentrations of TP and chlorophyllincreased during the 18-years study period for oligotrophic and mesotrophic lakes. The TN:TP ratio showed that the study group of oligotr ophic and mesotrophic lakes is phosphorus-limited. This ratio showed a significant decline and may reflect that lake water TP concentrations are increasing at a faster pace than lake water TN concentrations. As expected from the increase in TP and the phosphorus limitation of these lakes, an analysi s of the chlorophyllconcentration in lake water also suggested a slight increase over time. One likely explanation for the increase in TP and chlorophyll(and TN, results not included) is that population growth in Hillsborough County has increased the phos phorus and nitrogen loading to these lakes. For example, increased fertilizer us e in lawns and/or higher run-off rates would likely result I elevated chlorophyllconcentration. An important factor that may be facilitating this process is the sandy karst characteristic of the soil profile, which facilitates the leaching of nutri ents into the lakes. These possible explanations for phosphorus increase ov er time, however, are just suggestions since there is not conclusive evid ence showing if the extra phosphorus responsible for the TP incr ease in the lake water co lumn is recycled from the sediments or if it is a new external input to the system. Whatever is the case, the
85 fact is that these results obtained fr om this group of 10 oligotrophic and mesotrophic lakes meet the alternativ e hypotheses forTP, TN:TP, and chlorophyllconcentrations. Hypereutrophic lakes presented a diffe rent behavior as compared to oligotrophic or mesotrophic lakes. This second group of lakes had a decreasing trend in water concentration of TP during t he 18-year study perio d. The ratio of TN:TP showed that these highly eut rophied lakes were nitrogen-limited. Chlorophylldid not show indications of any trend over time at all in these hypereutrophic lakes, and are more in a ccordance with what was indicated by Terrell et al., (2000), in a bigger sample of lakes. Decrease in lake water TP concentration in this second group of lakes may be explained by the implementation of lake management plans (City of Lakeland, 2001; Southwest Florida Wate r Management District, 2003) that may have prioritized highly eutrophied lakes ov er less eutrophied ones. The fact that both groups of lakes resulted having a decre asing trend in the TN:TP ratio raise concerns for a shift in phytoplankton co mposition to more noxious species.
86 CHAPTER 4. IDENTIFICATION OF IMPORTANT VARIABLES AFFECTING WATER QUALITY IN LAKES OF THE TAMPA BAY WATERSHED 4.1 Introduction Aquatic vegetation can contain import ant proportion of the total nutrient content of the lake, it is therefore a key element to consider when assessing the potential concentration of nutrients in t he lake water column (Canfield et al., 1983). As it is known based on the liter ature reviewed in Chapter 2 (Bachmann et al., 2002; Dierberg et al., 2002; Hamilton and Mitchell, 1996), submerged rather than emergent aquatic vegetation has a greater potential to be associated to low levels of TP, TN, and chlorophyllin lake water, hence it is one of the variables examined in this chapter for possible association with eutrophication status, along with lake water total phosphorus (TP), total nitrogen (TN) concentration, and lake area, depth, and volume. It has been very well documented the strong and clear direct association between phytoplankton as measured by chlorophylland nutrients dissolved in lake in Florida (Bachmann et al., 2002; Brow n et al., 2000; Canfield et al., 1984). Although less conclusive, there has been al so documentation reporting that large amounts of submerged aquatic macrophy tes have some association with
87 reduced productivity of lake phytoplankton (Bachmann et al., 2002; Canfield and Hoyer, 1992; Canfield et al., 1984; Landers, 1982). In this chapter a 34-lake database that includes information on the variables submerged aquatic vegetation (two forms); lake water concentration of TP, TN, and chlorophyll; and area, depth and volume is introduced and described. Correlation between these lake variables are examined and compared with results from previous studi es. Some of the species of submerged aquatic vegetation more abundant in the lakes studied were: Vallisneria Americana Algal ssp ., Hydrilla verticilata Egeria densa, and Potamogeton spp., among others. 4.2 Methods 4.2.1 Data Sampling Program The 34 urban and suburban lakes examined in this chapter are distributed over an area with mixed land use: residentia l, recreational, and agricultural, in the northern and eastern portions of Hillsborough County (Figure 4.1). The lakes are distributed over four subbasins within the Tampa Bay watershed; Sweetwater Creek, Rocky / Brushy Creek, Brooker Creek and Curiosity Creek. The analysis was carried out by using existing data from two different sources. Data on lake water TP, TN, and chlorophyllconcentrations, as well as data on submerged aquatic vegetation, was collected by the Florida Center for Community Design and Research at the University of S outh (Koenig and Eilers, 2006-2007). Data
88 availability was the determining factor fo r inclusion of the 34 lakes in this analysis. Values on general lake variabl es: lake surface area, mean depth, and lake volume, were collected and reported by volunteer citizens from the Florida LAKEWATCH water quality monitoring program and were obtained as part of the Hillsborough County samp ling program (2008). Samples taken to determine TP, TN, and chlorophyllwere analyzed by the Hillsborough County Environmental Protection Commission laboratory (Chapter 3). Submerged aquatic vegetation is expressed as percentage of area covered with vegetation (PAC ) and percentage of volume of the lake infested with vegetation (PVI). Variables for each lake correspond to only one measurement done in 2006 or 2007 (depending on the lake). These data proceed from measurements carried out at o ne point in time. They are, however, useful to examine potential long term associations between the mentioned variables because each particular lake water variable depends on a historical trend, and no variable changed suddenly an d independently from their past conditions. According to Griffin (2008) data on PAC were determined by first selecting 100 randomly ordered bathymetr ic points, then reaching the points by boat to determine the presence of submerged aquatic vegetation by the soft return data obtained with a fathometer and expressing t he results in terms of percentage. PVI was calculated by measur ing the depth of the soft re turns (top of vegetation) and the depth of the hard retu rns (lake bottom) using th e bathymetric trace for each point as shown in Equation (1):
89 Equation (1): 100 ) (100 0 Depth Lake Vegetation of Depth Depth Lake = PVI For points where no vegetation exists, t he numerator is zero and that point is counted as zero.
90Figure 4.1 Location of lakes with recent observations on submerged aquatic vegetation (Koenig and Eilers, 2006-2007). ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! Pasco County Pinellas County Hillsborough County Lake Locations (shown in Red)Map Created by Pete Reehling University of South Florida
91 4.2.2 Statistical Analysis Analyses of correlation were used to determine significant associations between the eutrophic state variables lake water TP, TN, and chlorophyllconcentration and variables of submerged aquatic vegetation, lake area, depth, and volume. This analysis was conducted using the formula for the Pearson correlation in Excel shown in Equation (2), where X and Y are the lake variables, X bar and Y bar are the respective means, and SDx and SDy are the respective standard deviations. Equation (2): y x n i i iSD SD n Y Y X X r11 The variables TP and TN were taken as indicators of eutrophication (dependent variables) but also as factors associated to eutrophication (independent variables) when eutrophicati on was indicated by chlorophyll. A correlation > 0.30 was statistically signifi cant at a 95% of confidence level. 4.3 Results and Discussion Values in Table 4.1 present summar y statistics for physical and chemical variables analyzed in this study. Va riables for each lake correspond to one measurement done in 2006 or 2007. The lakes had an average mean depth of
92 2.51 m, ranging from 0.91 to 3.96 m therefore appr oaching the concept of shallow lakes used by Scheffer (2004) according to which lakes depth less than 3 m were considered shallow. Lake su rface area ranged from 1.21 to 174.43 ha with a mean of 25.42 ha. T he average volume was 713,300 m3 with a minimum of 2,087 and maximum of 5, 714,000 m3. According to the Trophic State Classifi cation System of System Forsberg and Ryding (1980, Figure 3. 1) the lakes ranged from mesotrophic to eutrophic with average values of 25.18 g L-1, 0.80 mg L-1, and 7.90 g L-1 for TP, TN, and chlorophyllrespectively. Ranges for these va riables were: 3.00 to 50.00 g L-1 for TP, 0.37 to 1.27 mg L-1 for TN, and 1.20 to 44.61 g L-1 for chlorophyll. The average percent of study lake area co vered by submerged aquatic vegetation (PAC) was 38.78% with a minimum and maximum of 2.00 and 85.00% respectively. On average the volume of the study lakes occupied by submerged aquatic vegetation (PVI) was 16.17 % and ranged between 0.52% and 47.00% (Table 4.1). The analysis of correlation between all the variables examined in this study (Table 4.2) presented significant inverse correlation between chlorophylland both measures of s ubmerged aquatic vegetat ion: PAC and PVI, with r = 0.72 and -0.60, respectively. TP, r = 0.46; and TN, r = 0.41. TP was also inversely correlated with both form s of submerged aquatic vegetation, r = -0.52 and -0.41 for PAC and PVI, respective ly, and additionally with mean depth, r = 0.36. TN was not significantly correlated (p > 0.05) with any variable other than chlorophyll. The fact that chlorophyllpresented a higher correlation with TP
93 as compared to TN may reflect the phosph orus limitation of most of the lakes studied. The greater negative correla tion found between chlorophylland both measures of submerged aquatic veget ation as compared with the positive correlation between chlorophylland TP and TN may suggest that submerged aquatic vegetation rather than TP and TN mi ght be the factor most associated to phytoplankton. If this is the case, th is result suggests that submerged aquatic vegetation may be associated with chlorophyllalso through some other additional way not involving TP and TN. Batchman et al. (2002), however, reported a smaller negative co rrelation between chlorophylland submerged aquatic vegetation ( r = -0.29) as compared to those between chlorophylland TP ( r = 0.82), and chlorophylland TN ( r = 0.70). Yet the authors consider that under conditions where water nutrient conc entration is not excessively elevated, submerged aquatic vegetation reduce nut rients and consequently phytoplankton concentration as measured by chlorophyll, rather than dissolved nutrients reducing submerged aquatic vegetation gr owth. Other authors also suggest association between chlorophyllwith TP and TN (Brown et al., 2000; Canfield et al., 1984). A suggested inverse relationship bet ween submerged aquatic vegetation and chlorophyllin the water column might be explained by the effect possibly caused by submerged aquatic vegetation in reducing concentration of limiting nutrients in the water column, and t he subsequent reduced availability of phytoplankton (Bachmann et al., 2004; Sc heffer, 2004); and by provision of
94 shelters from predatory fish for zooplank ton that directly prey on phytoplankton (Scheffer, 2004). In general, the mechanism by whic h submerged aquatic vegetation reduce nutrient concentration has been described in part by the attenuation of water turbulence, which results in less resuspension and recycling of nutrients back into the water column (Bachmann et al., 2004; Hamilton and Mitchell, 1996; Scheffer, 2004). Other contributing mechani sm may be the provision of substrate surface for periphyton that up-take nutrients from t he water column (Bachmann et al., 2004; Cattaneo and Kalff, 1980); Up-t ake of nutrients from the water column by submerged aquatic vegetat ion directly (Denny, 1972; Graneli and Solander, 1988); and by influencing ion ex change reactions via regulation of dissolved oxygen and pH (Graneli and Sola nder, 1988). All these reasons may help explain the inverse relationship found between submerged aquatic vegetation and TP, and consequent ly also with chlorophyll. The present study found mean depth to be directly correlated with both forms of submerged aquatic vegetation, r = 0.34 and 0.31 for PAC and PVI respectively. This may seems to be unexpected since the required light penetration for photosynthesis is r educed with depth, however, analysis of regression between submerged aquatic vegetation abundance and depths (results not presented) showed an increas e in submerged aquatic vegetation up to a depth of approximately 2. 5 m and then a decrease. The fact that most of the lakes studied were shallower than that dept h, explains why the overall results showed a direct correlation between both variables. Other additional possible
95 explanation is that shallow depths mi ght favor dominance of emergent aquatic vegetation over submerged aqua tic vegetation. As also expected, both mean depth and area were correlated with volume r = 0.38 and 0.98, respectively. Since PAC and PVI are both expressions of the same variable submerged aquatic vegetation, they were consequently highly correlated, r = 0.91. Association of these two related sub va riables has been reported in the literature (Canfield and Hoyer, 1992; Canfield et al., 1984). The inverse characteristics of the relationship between TP and mean depth may be due to the great er distance between the source of resuspended phosphorus in the bottom sediments and the superior layers of the water column. Stronger turbulence would be required to resuspend phosphorus through the entire water column. Another possible reas on is the obvious direct correlation between depth and volume. An increas ing depth would be associated with a greater volume of water and consequent ly greater dilution of phosphorus. An analysis of correlation of the ei ght variables of shallow lakes grouped by groups would result as follow: (1 ) submerged aquatic vegetation variables, PAC and PVI; (2) eutrophication variables, TP, TN, and chlorophyll; and (3) lake size variables, area, depth, and volume. Lakes with more submerged aquatic vegetation have less eutr ophication especially chlorophyllconcentration followed by TP conc entration. Bigger lakes have less eutrophication and depth is t he most important. Bigger lakes tend to have more submerged aquatic vegetati on (for shallow lakes).
96Table 4.1 Summary statistics of Hillsbor ough lakes examined for association between lake variables. n Median Mean SD MinimumMaximum PAC 34 38.0038.7826.752.0085.00 PVI 34 12.4616.1712.580.5247.00 Volume (m3) 34 381700713300105000020875714000 Area (ha) 34 15.9925.4232.851.21174.43 Mean Depth (m) 34 2.592.510.770.913.96 TN (g L-1) 34 0.860.800.240.371.27 TP (g L-1) 34 24.5025.1811.393.0050.00 Chlorophyll(g L-1) 33 6.607.905.191.2021.70 Table 4.2 Matrix table showing the analysis of correlation between lake variables. Values in dark correspond to resulting significant associations between trophic state parameters (TN, TP, chlorophyll) and theirs factors controlling for water quality at 95% confidence. PAC PVI Volume (m3) Area (ha) Depth (m) TN (mg/L) TP (g L-2) Chla (g L-2) PAC 1.00 PVI 0.91 1.00 Volume (m3) 0.24 0.24 1.00 Area (ha) 0.22 0.23 0.98 1.00 Depth (m) 0.34 0.31 0.38 0.28 1.00 TN (mg L-2) -0.27 -0.21 0.11 0.10 -0.27 1.00 TP (g L-2) -0.52 -0.41 -0.13 -0.14 -0.36 0.26 1.00 Chla (g L-2) -0.72 -0.60 -0.29 -0.27-0.27 0.41 0.46 1.00
97 4.4 Summary In summary, among the eight lake indicators (TP, TN, chlorophyll, PAC, PVI, mean depth, lake surface area, and lake volume) examined in this chapter, the strongest statistically significant association was found to be the inverse relationship between chlorophylland submerged aquatic vegetation as represented by PAC and PVI. The second strongest statistically significant association was the relationship between chlorophylland lake water TP concentration. Submerged aquatic vegetation was not significantly associated to lake water TN concentration. In gener al, more submerged aquatic vegetation is associated with less eutr ophication. The higher correlation of chlorophyllwith abundance of submerged aquatic vegetation as compared to water total phosphorus was a finding not expected and may indicate that submer ged aquatic vegetation is associated to chlorophyllalso through some other way that does not involve nutrients. This is consistent with the idea that phytopl ankton is affected by submerged aquatic vegetation through both effects on TP conc entration and through other effect s, for example, as a shelter for phytoplankt on-grazing zooplankton. Batchman et al. (2002) and Batchman et al. (2004) although suggested a weak inverse relationship between both parameters, i ndicated that aquatic macrophytes did not significantly affect the phosphorus ve rsus chlorophyll relationships in their studies.
98 In general, for shallow lakes, bigger lakes had more submerged aquatic vegetation, Bigger lakes were less eutr ophied and the largest correlation was between depth and lake water TP concentration.
99 CHAPTER 5. EFFECT OF SUBMERGED AQUATIC VEGETATION ON WATER TOTAL PHOSPHORUS CONCENTRATION IN LAKES OF THE TAMPA BAY WATERSHED 5.1 Introduction Among the different types of aquatic vegetation, submerged aquatic vegetation has been found to play an import ant role in regulation of nutrient concentrations and subsequently lake phytoplankton (Bachmann et al., 2004; Brenner et al., 1999; Jeppesen et al., 1997; Knight et al., 2003), and also in a more direct way, probably shelter for grazers (Scheffer, 2004). The nature and extent of these relationships, howe ver, still remain vague (Bachmann et al., 2002). It has been speculated that at hi gh levels of nutrient concentrations, nutrients may control submerged aquatic vegetation while in waters with more moderated and lower nutrient concentrati ons, nutrients may be controlled and further reduced by aquatic macrophytes es pecially submerged aquatic vegetation (Bachmann et al., 2002; Bachmann et al., 2004). Studies conducted in other geographic areas seem to be even less unifying in terms of clarifying the nature of this relationship. Nutrient levels in water have been found to trigger growth of submerged aquatic vegetation
100 (Ozimek, 1978, as quoted by Duarte and Ka lff, 1986), not to cause a clear effect (Carpenter and Adams, 1979), and to dec rease prevalence of submerged aquatic vegetation (Duarte, 1995) especially at a large increase in phosphorus level (Graneli and Solander, 1988). It has been reported in liter ature that lakes have changed from being in a clear water state to turbid water state (with a higher concentration of nutrients and suspended solids), when submerged aquatic vegetation was removed by herbici de treatment (O'Dell et al., 1995), or by hurricanes (Bachmann et al., 1999). Li kewise, lakes have been reported to switch from a turbid to clear water state when planktivorous fish were removed and submerged aquatic vegetation incr eased (Ozimek et al., 1990). It was shown in Chapter 4 that s ubmerged aquatic vegetation, and total phosphorus (TP) to a lesser extent, we re the most significant variables associated with phytoplankton productivity. C onsequently it is important to further examine the relationship between these tw o variables. In this chapter the same data analyzed in Chapter 4 from a group of 34 lakes in Hillsborough County was examined for a possible association betwe en the percentage of volume and area of the lake that is occupied by su bmerged aquatic vegetation (PVI and PAC) versus the concentration of total phosphorus (TP), total nitrogen (TN), and chlorophyllin lake water. The difference is that this analysis only covers submerged aquatic vegetation variables and eutrophication variables (not size variables) because these were the more st rongly correlated in Chapter 4. Also examined was the association betw een submerged aquatic vegetation density
101 and inter-annual variability of lake wa ter TP concentration in the mentioned parameters in a subs ample of 24 lakes. A strong link between water nutrient s concentration and phytoplankton biomass has been reported in studies condu cted in Florida lakes (Bachmann et al., 2002; Canfield, 1983). The relations hip between nutrients and chlorophyllwith submerged aquatic vegetation in lake water has shown to be more complex and difficult to clarify. Some author s, however, have already approached some measure of relationship between nutri ents and submerged aquatic vegetation in Florida lakes (Bachmann et al., 2002) and other geographical areas (Cattaneo and Kalff, 1980; Duarte, 1995; Graneli and Solander, 1988; Scheffer, 2004), and the speculated mechanisms (sediments stabiliz ation, plant up-take, precipitation, redox reactions, and shelters) by which they may be related (Bachmann et al., 2004; Cattaneo and Kalff, 1980; Duarte 1995; Graneli and Solander, 1988; Scheffer, 2004). Other authors have repor ted information about the relationship between phytoplankton and submerged aquatic vegetation (Canfield and Hoyer, 1992; Canfield et al., 1984). Still in general, much needs to be done to confirm the relationship, if one exists, between eutrophication variables and submerged aquatic vegetation. 5.2 Objectives and Hypothesis The analysis conducted in this chapter was intended to provide evidence regarding the relationship between submerged aquatic vegetation and TP
102 concentrations. As a complementary analysis, submerged aquatic vegetation was examined in relation to chlorophylland TN concentrations, as was any influence of submerged aquatic vegetation in the inter-annual variability of water TP concentration. The results could be applied to develop best management practices to control cultural eutr ophication associated with watershed development. The following objectives and hypothesis are addressed in this chapter: Objective: To provide evidenc e that the presence of submerged aquatic vegetation plays a significant role in lowering total phosphorus (TP) concentration in lake water in lakes of Hillsborough County. Hypothesis 1: Lakes with little vegetation have higher TP, TN, and chlorophyllconcentrations. Objective: To determine if the presence of submerged aquatic vegetation in urban and suburban lakes of the Tampa Bay watershed is strongly and inversely associated to the inter-annual fluctuation in lake water TP concentration.
103 5.3 Methods 5.3.1 Data Description Methods are described in Chapter 4. 5.3.2 Statistical Analysis In order to detect any possible effect on lake water TP, TN, and chlorophyll concentration presumably due to the presence of submerged aquatic vegetation, the lakes were separated in two groups, those with a high presence of submerged aquatic vegetation versus those with a low presence. The separation criteria used to delineate a difference between the two groups was defined by Bachmann et al. (2002). A ccording to this, high presence of submerged aquatic vegetation is cons idered as macrophyte-dominated and included those lakes with PVI>80 whil e low presence of submerged aquatic vegetation corresponded to phytoplanktondominated, which are those lakes with PVI<20. Since none of the 34 lakes in th is study had a PVI greater than 80, a PVI of 20 was considered as the cut-poi nt for separation of the two groups. Hence for the purpose of this study, la kes with a PVI greater than 20 were considered macrophyte-dominated whil e lakes with a PVI less than 20 were considered phytoplankton-dom inated. One-way ANOVA was used to determine if the means of both groups of lakes were significantly different for each one of the three variables consi dered (TP, TN, and chlorophyll). Bachmann et al.
104 (2002) conducted similar analysis using one-way ANOVA to determine differences between the means of two lake groups for eutrophica tion variables. A t-test assuming unequal variances was addi tionally used to confirm the results. Both tests were performed at the 95% confidence level ( = 0.05). Graphics with frequency distributions were made for vis ual comparison between the two groups of lakes for each one of the three trophic state variables examined. A linear regression was used to represent a possible relationship between variability of lake water TP concentr ation and submerged aquatic vegetation dominance in 24 lakes. The variability of TP concentration was indicated by expressing the coefficient of variation (CV) of this eutrophication variable for the available values from 2005 unt il present in terms of the mean of this variable for the same time period. The year 2005 as a cut-off point for inclusion of values of eutrophication was chosen arbitrarily int ending to have a short period but long enough to show variability. A long period of time for observations in eutrophication variables would have increased the uncertainty for the corresponding values in submerged aquatic vegetation, which are unknown. Submerged aquatic vegetati on dominance was indicated by one-time value of both PAC and PVI since this was the onl y available value for this parameter. 5.4 Results and Discussion Out of the overall group of 34 lakes, 9 were classified as macrophytedominated. The average percent volume in fested (PVI) in this group was 34.15
105 with a minimum and maximum of 23.00 and 47.00 respectively; 9.69 was the average PVI for the 25 lakes compos ing the phytoplankt on-dominated group, with a minimum and maximum of 0.52 and 19.00 respectively (Table 5.1). The average percent of area covered with s ubmerged aquatic vegetation (PAC) in the macrophyte-dominated group was 74.44 wit h a minimum and maximum of 63.00 and 85.00 respectively. The average for t he same variable in the phytoplanktondominated group was 25.93 and ranged bet ween 2.00 and 56.00 (Table 5.2). Table 5.1 Summary table for PVI values of lakes with macrophyte-dominance and phytoplankton-dominance. Lake dominance n MedianMeanSDMinimum Maximum Macrophyte-dominated 9 33 34.1 8.423 47 Phytoplankton-dominated 2510 9.6 5.30.52 19 Table 5.2 Summary table for PAC values of lakes macrophyte-dominance and phytoplankton-dominance. Lake dominance n MedianMeanSD Minimum Maximum Macrophyte-dominated 9 76 74.4 7.6 63 85 Phytoplankton-dominated 2530 25.9 17.72 56 The TP concentration in the group of lakes with macrophyte-dominance ranged from 3 to 31 g L-1 with an average of 18.44 g L-1, while the group with phytoplankton-dominance showed a range from 10 to 50 g L-1 with an average of 27.6 g L-1 (Table 5.3). According to the Tr ophic State Classification System of Forsberg and Ryding (1980, Figure 3.1) so me lakes from both groups overlap under the classification of oligotrophic ([TP] <15 g L-1), mesotrophic (15< [TP] <25 g L-1), and eutrophic ([TP] >25 g L-1). No lake in either group fell under the classification of hy pereutrophic ([TP] >100 g L-1). The range in macrophytedominated lakes for chlorophyllwent from 1.2 to 6.9 g L-1 with an average of
106 3.60 g L-1, while 3.6 to 44.61 g L-1 with average of 10.91 g L-1 was the range for the same variable in phytoplankt on-dominated lakes (Table 5.4). These ranges show substantial overlap for both va riables in both lake groups, which do not guarantee accurate prediction of water quality based only on submerged aquatic vegetation prevalence. Despite the overlap, however, the results of a ttest indicated a significant difference between both groups of lakes for TP and chlorophyll( p = 0.0002 and 6.4 x 10-9 respectively). This difference was confirmed in one-way ANOVA for the same parameters ( p = 0.036 and 0.017 respectively). TP and chlorophyllvalues were significantly higher for lakes with a PVI lower than 20 (phytoplanktondominated) as compared to those with a PVI higher than 20 (macrophyte-dominated). This not just support results of significant association of submerged aquatic vegetation with TP and chlorophylldescribed in Chapter 4 but also indicates a str ong relationship between submerged aquatic vegetation and TP and chlorophyll. Results are additionally supported by literature that indicate the nutrient removal capacity of submerged aquatic vegetation from water column (Dierberg et al., 2002; Gu et al., 2001; Knight et al., 2003). Some possible mechanisms by which submerged aquatic vegetation reduces nutrient concentrations in the wa ter column are described in Chapter 2 and more briefly in Chapter 4.
107 Table 5.3 Summary table for TP values (g p L-1) for lakes with macrophyte-dominance and phytoplankton-dominance. Lake dominance n MedianMeanSD Minimum Maximum Macrophyte-dominated 9 19 18.448.99 3 31 Phytoplanktondominated 2526 27.6011.3510 50 Table 5.4 Summary table for chlorophyllvalues (g L-1) for lakes with macrophytedominance and phytoplankton-dominance. Lake dominance n MedianMeanSD Minimum Maximum Macrophyte-dominated 9 3.80 3.60 1.821.20 6.90 Phytoplankton-dominated 258.80 10.918.643.60 44.61 Analysis of frequencies in both group of lakes for TP (Figures 5.2 and 5.3) and chlorophyllconcentrations (Figures 5.4 and 5. 5) showed that more lakes in the group with macrophyte-dominance were to ward the upper limit of values in both parameters (skewed to the left). And more lakes in the phytoplanktondominated group were toward the lower limit of the range for this group (skewed to the right). There was no significant difference found between the group of lakes with macrophyte-dominance and those with phytoplankton-dominance regarding values in the concentration of TN ( ttest, p<0.071; ANOVA, p<0.326; Figures 5.6 and 5.7). This result matches the lack of association detected between TN and submerged aquatic vegetation in the anal ysis of regression and correlation conducted in the previous chapter, but di ffers with results of Batchman et al. (2002) and Batchman et al. (2004) that s uggested such an association. Average lake water TN concentration in macrophye-dominated lakes was 0.73 mg L-1 with
108 a minimum and maximum of 0.37 and 1.27 mg L-1 respectively. The range for the same parameter in phytoplankton-dom inated lakes went from 0.41 to 1.13 mg L-1 with an average value of 0.82 mg L-1. As expected from the t-test and ANOVA analysis, the overl ap for both groups of lakes regarding this parameter was much greater; in fact the range of the phytoplankton-dominated lakes was totally included within the range of macrophye-dominated lakes. Table 5.5 Summary table for TN (mg L-1) values of lakes with macrophyte-dominance and phytoplankton-dominance. Lake dominance n MedianMeanSD Minimum Maximum Macrophyte-dominated 9 0.60 0.73 0.340.37 1.26 Phytoplankton-dominated 250.87 0.82 0.190.41 1.13 As it can be seen from the analysis of frequencies (Figures 5.6 and 5.7), the distribution of frequencies of lake water TN concentration were opposite of those showed by TP and chlorophyll. Most of the macrophyte-dominated lakes were in the lower values of nitrogen c oncentration of the scale (skewed to the right). Most of the lake s in the phytoplankton-domi nated group were toward the higher concentration of the range (skewed to the left). These results seems to support the hypothesis that submerged aquatic vegetation is associated with lower concentrations of nitrogen in the water column, however, the t-test and ANOVA analysis proved different.
109Figure 5.1 TP in macrophyte-dominated lakes. 0 1 2 3 4 5 312.3333333321.66666667MoreTotal phosphorus ( g L-1)Number of lakes Figure 5.2 TP in phytoplankton-dominated lakes. 0 5 10 1018263442MoreTotal phosphorus ( g L-1)Number of lakes Figure 5.3 Chlorophylla in macrophyte-dominated lakes. 0 1 2 3 4 51.23.15MoreChlorophyll(mg L-1)Number of lakes
110Figure 5.4 Chlorophylla in Phytoplankton-dominated lakes. 0 5 10 15 203.611.80220.00428.20636.408MoreChlorophyll(mg L-1)Number of lakes Figure 5.5 TN in macrophyte-dominated lakes. 0 1 2 3 4 5 6 0.370.6693333330.968666667MoreTotal nitrogen (mg L-1)Number of lakes Figure 5.6 TN in Phytoplankton-dominated lakes. 0 2 4 6 8 10 12 0.410.5540.6980.8420.986MoreTotal nitrogen (mg L-1)Number of lakes Contrary to expected, the linear regression analysis for a possible relationship between the variability of lake water TP concentration (as expressed by CV) and submerged aquatic vegetation di d not show any association for any of the two measures of submerged aquatic vegetation: PAC and PVI (Figures 5.8
111 and 5.9). This result, however, is not conc lusive because of limitations caused by unavailability of repeated measures of submerged aquatic vegetation dominance for the lakes studied. Figure 5.7 Variability of lake water total phosphorus with increment in area covered by vegetation. y = 5E-06x + 0.2039 r = 2.64 x 10-30 0.05 0.1 0.15 0.2 0.25 0.3 0.35 020406080100 Per cent area coveredCoeficient of variation Figure 5.8 Variability of lake water total phosphorus with increment in volume of the lake infested by vegetation. y = -0.0003x + 0.2084 r = 0.07 0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 01020304050 Per cent volume infestedCoeficient of variation
112 5.5 Summary The analysis of t-test and one way ANOVA conducted in this chapter indicated that lakes dominated by macrophytes show a significantly lower lake water concentration of TP and chlorophyllas compared to those lakes dominated by phytoplankton. The same type of analysis did not show a significant difference between both groups of lakes in regard to TN concentration. This result confirms the str ong correlation between submerged aquatic vegetation and lake water chlorophylland lake water TP concentration discussed in Chapter 4. Furthermore, these results strongly suggest that at least under TP levels <50 g L-1, the association of submerged aquatic vegetation with concentrations of TP and chlorophylldid not extend to lake water TN concentrations. This analysis did not show a buffering effect of submerged aquatic vegetation in the inter-annual variability of lake water TP concentration in urban lakes.
113 CHAPTER 6. IMPLICATIONS OF RESEARCH Eutrophication of lakes located in the Tampa Bay watershed show significant trends between 1990 and 2007. There are three main findings regarding these trends. First, the c oncentration of phosphorus (as TP) and phytoplankton (as measured by chlorophyll) increased over time for lakes classified as oligotrophic or meso trophic. Second, in hypereutropic lakes phosphorous concentrations decreased with time and no significant trend was seen for chlorophyll. Third, the ratio of nitrogen to phosphorus (TN:TP) declined for lakes with both low and high leve ls of eutrophication. Historical and recent human settlement patterns and popula tion growth in Hillsborough County, coupled with a karst geology, may have contributed to the observed increase in lake water phosphorus concentrations for oligotrophic and mesotrophic lakes. Lake management plans that have includ ed reducing point and non-point source nutrient flows may be responsible for the declining water phosphorus concentrations in hypereutrophic lakes. For many of the oligotrophic and meso trophic lakes of the region a trend line suggests that in ~20 years primar y productivity may be nitrogen-limited, but for hypereutrophic lakes, pr imary productivity is alr eady nitrogen-limited, as suggested by TN: TP ratios 10. Nitrogen-limit ed primary productivity may have
114 undesirable conseuences related to t he increase in cyanobacterial populations (Hecky and Kilham, 1988; Levich, 1996; Levich and Bulgakov, 1992), and thus a threat thread to public health due to ef fects caused by toxins produced by this blue-green algae (Fleming et al., 2002; Karjalainen et al., 2007). As population growth and develop ment seem inevitable and the underlined karst formation is a permanent co ndition (in the case those factors play in fact a role in the eutrophicati on of these lakes), this study explored submerged aquatic vegetation as a possibl e factor (based on t he literature read) in controlling eutrophication for lake s in the Tampa Bay watershed. Among a group of lakes composed mostly of thos e in the low eutr ophication subgroup, submerged aquatic vegetati on was found to be the most significant factor associated to eutrophication. The strongest variables associated to eutrophication as estimated by wa ter concentration of chlorophyll, were (in order of significance), percentage of lake area covered with submerged aquatic vegetation (PAC), percentage of lake volume occupied by submerged aquatic vegetation (PVI), and concent ration of phosphorus and nitrogen in lake water. Submerged aquatic vegetat ion (both expressions) also had the strongest association with lake water phosphorus co ncentration, followed by mean depth. When water nitrogen concentration was examined as dependent variable, there were no other variables signifi cantly associated with it. Hypothesis-testing revealed that phosphorus and chlorophyllconcentrations were significantly hi gher for lakes with low coverage of
115 submerged aquatic vegetation than for la kes with high coverage. These results support the theory that submerged aquatic v egetation is a strong candidate to be considered a controlling fact or or at least a strong indicator of water quality. Measuring the relationship betwe en submerged aquatic vegetation and phosphorus levels, may in fact, be indirect ly measuring the relationship between submerged aquatic vegetati on and phytoplankton productivity. This is because phytoplankton productivity is mostly limit ed by water phosphorus concentration in this group of lakes. Hence the mec hanisms discussed here as means by which submerged aquatic vegetation may infl uence water phosphorus concentration are indirectly those by which su bmerged aquatic vegetation may influence phytoplankton productivity as well. Thes e mechanisms are: sedimentation of suspended total phosphorus, direct phosphor us uptake from the water column, provision of surfaces for periphyton and bacteria, and influences on ion exchange reactions via regulation of dissolved oxygen concentration and pH levels. Phytoplankton productivity (as measured by chlorophyll), however, was found to be more strongly associated with submerged aquatic vegetation than with total phosphorus concentration. This may indicate that in addition to the indirect mechanisms mentioned above, submerged aquatic vegetation may exert also a more direct effect in phytoplankt on productivity, for example, by sheltering zooplankton from fish, which increases z ooplankton predation of phytoplankton. Eutrophication and change of nutrient ra tios in freshwater urban and suburban lakes may represent a threat to public health by prom oting productivity
116 of toxic algae. Oral, resp iratory, and cutaneous expos ure to toxins released by toxic algae can result in diseases of the nervous, gastrointestinal, and hepatic systems. Furthermore, alterations of environmental aesthetics and ecosystem balance are additional ways through which eutrophication can impact the human wellbeing, from a more holisti c view of public health. A growing body of research is st ill needed in order to determine key variables or factors that might be manipulated to control lake water eutrophication. Results of such research and monitoring programs will contribute to the formulation of effective management plans for sustainable conditions of human wellbeing.
117 REFERENCES Ammerman, J. W.; Azam, F. Bacterial 5Nucleotidase in Aquat ic Ecosystems: A Novel Mechanism of Phosphorus Regeneration. Science 1985, 15, (227), 1338-1340. Andersson, G.; Graneli, W.; Stenson, J. The influence of animals on phosphorus cycling in lake ecosystems. Hydrobiologia 1988, 170, (267-284). APHA, Standard Methods for the Exami nation of Water and Wastewater American Public Health Associat ion: Washington, D. C., 1989. APHA, Standard Methods for the Exami nation of Water and Wastewater American Public Health Associat ion: Washington, D. C., 1998. Arbukle, K. E.; Downing, J. A. The infl uence of watershed land use on lake N;P in a predominantly agricultural lanscape. Limnology and Oceanography 2001, 46, (4), 970-975. Azam, F.; Fonda, U. S.; F unari, E. Significance of bacteria in the mucilage phenomenon in the norther n Adriatic Sea. Ann. Ist. Super Sanita. 1999, 35, (3), 411-419. Babiarz, C. L.; Benoit, J. M. ; Shafer, M. M.; Andren, A. W.; Hurley, J. P.; Webb, D. A. Seasonal influences on parti tioning and transport of total and methylmercury in rivers from contrasting watersheds. Biogeochemistry 1998, 41, 237-257. Bachmann, R. W.; Canfield, D. E. J. Us e of an alternative method for monitoring total nitrogen concentrations in Florida lakes. Hydrobiologia 1996, 323, (1), 1-8. Bachmann, R. W.; Horsburgh, C. A.; Hoyer, M. V.; Mataraza, L. K.; Canfield, D. E. J. Relations between trophic stat e indicators and plant biomass in Florida lakes. Hydrobiologia 2002, 470, 219-234. Bachmann, R. W.; Hoyer, M. V.; Canfield, D. E., Jr. Aquatic plants and nutrients in Florida lakes. Aquatics 2004, 26, (3), 4-10.
118 Bachmann, R. W.; Hoyer, M. V.; Canfie ld, D. E. J. The restoration of Lake Apopka in relation to alternative stable states Hydrobiologia 1999, 394, (0), 219-232. Barko, J. W.; Smart, R. M. Mobilization of Sedi ment phosphorus by submerged freshwater macrophytes. Freshwater Biology 1980, 10, 229-238. Barrow, N. J. A mechanistic model for de scribing the sorption and desorption of phosphate by soil. J. Soil Sci. 1983, 34, 733-750. Barsdate, R. J.; Prentki, R. T.; F enchel, T. Phosphorus cycle of model ecosystems: Significance for the decom poser food chains and effect of bacterial grazers. Oikos 1974, 25, (3), 239-251. Baskaran, M.; Swarzenski, P. W. Seasonal variations on the residence times and partitioning of short-lived radionu clides (234Th, 7Be and 210Pb) and depositional fluxes of 7Be and 210 Pb in Tampa Bay, Florida Marine Chemistry 2007, 104, (1-2), 27-42. Bennett, E. M.; Carpenter, S. R.; Cara co, N. F. Human impact on erodable phosphorus and eutrophication: a global perspective. Bioscience 2001, 51, 227. Berman, M.; Richman, S. The feeding behavior of Daphnia pulex from Lake Winnebago, Wisconsin. Limnol. Oceanogr. 1974, 19, 105-109. Billen, G.; Garnier, J. The Phison River plume: coastal eutrophication in response to changes in land use and water management in the watershed. Aquatic Microbial Ecology 1997, 13, 3-17. Bostrom, B.; Andersson, J. M.; Flei scher, S.; Jansson, M. Exchange of phosphorus across the sediemnt-water interface. Hydrobilologia 1988, 170, 229-244. Brenner, M.; Whitmore, T. J.; Lasi, M. A.; Cable, J. E. ; Cable, P. H. A multi-proxy trophic state reconstruction for sha llow Orange Lake, Florida, USA: possible influence of macrophytes on limnetic nutrient concentrations. Journal of Paleolimnology 1999, 21, 215-233. Brezonik, P. L. Trophic state indices: Rationa le for multivariate approaches ; EPA 440/5-84-001; U. S. Environmental Pr otection Agency: 1984; pp 441-445. Brown, C. D.; Canfield, D. E. J.; Ba chman, R. W.; Hoyer, M. V. Seasonal patterns of chlorophyll, nutrient concentrations and secchi disk
119 transparency in Florida lakes. J. Lake Reserv. Managem. 1998, 14, (1), 60-76. Brown, C. D.; Hoyer, M. V.; Bachmann, R. W.; Canfield, D. E., Jr. Nutrientchlorophyll relationships: an evaluati on of empirical nutrient-chlorophyll models using Florida and nor th-temperate lake data. Can. J. Fish. Aquat. Sci. 2000, 57, 1574-1583. Brown, M. T. Landscape rest oration following phosphate mining: 30 years of coevolution of science, industry and regulation. Ecolgical Engineering 2005, 24, 309-329. Canfield, D. E., Jr.; Hoyer, M. V. Aquatic macrophytes and their relation to the limnology of Florida lakes ; Department of Fisheries and Aquaculture Center for Aquatic Plants, University of Florida: Gainesville, FL, March 1992, 1992; p 599. Canfield, D. E., Jr.,; Shireman, J. V.; Colle, J. V.; Haller, W. T.; Watkins, C. E., II. Prediction of chlorophyllconcentrations in Florida lakes: importance of aquatic macrophytes. Can. J. Fish. Aquat. Sci. 1984, 41, (3), 497-501. Canfield, D. E. J. Pr ediction of chlorophyllconcentrations in Florida lakes: the importance of phosphorus and nitrogen. Journal of the American Water Resources 1983, 19, (2), 255-262. Canfield, D. E. J.; Langeland, K. A.; Linda, S. B.; Halle r, W. T. Relation between water transparency and maximum depth of macrophyte colonization in lakes. J. Aquat. Plant Manage. 1985, 23, 25-28. Canfield, D. E. J.; Langeland, K. A.; Maceina, M. J.; Ha ller, W. T.; Shireman, J. V.; Jones, J. R. Trophic state cla ssification of Lakes with Aquatic Macrophytes. Can. J. Fish. Aquat. Sci. 1983, 40, 1713-1718. Canfield, D. E. J.; Tham drup, B.; Hansen, J. W. The anaherobic degradation of organic matter in danish coastal sediments. Geochim. Cosmochim. Acta 1993, 57, 3867-3883. Caraco, N. F.; Cole, J. J.; Likens, G. E. Evidence for sulphate-controlled phosphorus release from sedi ments of aquatic systems. Nature 1989, 341, 316-318. Carpenter, S. R. Eutrophication of aquatic ecosystems: Bistability and soil phosphorus. Environmental Sciences 2005, 102, (29), 10002-10005.
120 Carpenter, S. R.; Adams, M. S. Effects of nutri ents and temperature on the decomposition of Myriophyllum spicatum L. in a hard-water eutrophic lake. Limnol. Oceanogr. 1979, 24, (3), 520-528. Carpenter, S. R.; Caraco, N. F.; Correll, D. L.; Howarth, R. W.; Sharpley, A. N.; Smith, V. H. Nonpoint Pollution of Surface Waters with Phosphorus and Nitrogen. Ecological Applications 1998, 8, (3), 559-568. Carpenter, S. R.; Ludwig, D.; Brock, W. A. Managem ent of eutrophication for lakes subject to potentiall y irreversible change. Ecological Applications 1999, 9, (3), 751-771. Carr, G. M.; Duthie, H. C.; Taylor, W. D. Models of aquatic plant productivity: a review of the factors that influence growth. Aquatic Botany 1997, 59, 195215. Cattaneo, A.; Kalff, J. The relative cont ribution of aquatic macrophytes and their epiphytes to the production of macrophyte beds. Limnol. Oceanogr. 1980, 25, (280-289). Chameides, W. L.; Perdue, E. M., Biogeochemical Sycles Oxford University Press: New York, 1997; p 224. Cheng, A. H. D.; Kindinger, J. L., Coasta Aquifer Management monitoring, modeling, and case studies Lewis Publishers: Boca Raton, London, New York, Washington D. C., 2004; p 296. ChunLei, S.; XiuYun, C.; BingQin, L.; YiYong, Z. A relation between algae blooms and benthic phosphor us status in ponds. Acta Hidrob. Sinica 2004, 28, (1), 7-12. City of Lakeland, P. W. D., Lakes & Stormwater Division City of Lakeland, Lakes Water Quality Report 1988-2000 ; City of Lakeland, Lakes & Stormwater Division Lakeland, FL, 2001; p 37. Costanza, R.; Voinov, A. Modelling ec ological and economical systems with Stella: part III. Ecological Modelling 2001, 143, (1-2), 1-7. Cowen, W. F.; Lee, G. F. Phosphorus av ailability in particulate materials transported by urban runoff. Journal Water Pollution Control 1976, 48, (3), 580-591. D'Elia, C. F.; Steudler, P. A.; Corwin, N. Determination of total nitrogen in aqueous samples using persulfate digestion. Limnol. Oceanogr. 1977, 22, 760-764.
121 Denny, P. Sites of nutrients absor ption in aquatic macrophytes. J. Ecol. 1972, 60, (3), 819-829. Dierberg, F. E.; DeBusk, T. A.; Jackson, S. D.; Chimney, M. J.; Pietro, K. Submerged aquatic vegetation-based treatment wetlands for removing phosphorus from agricultural runoff: response to hydraulic and nutrient loading. Water Res. 2002, 36, (6), 1409-1422. Dillon, P. J.; Rigler, R. H. The phosphorus-chlorophyll relationship in lakes. Limnol. Oceanogr. 1974, 19, 767-773. Doering, P. H.; Oviatt, C. A.; Nowicki, B. L.; Klos, E. G.; Reed, L. W. Phosphorus and nitrogen limitation of primary production in a simulated estuarine gradient. Marine Ecology Progress Series 1995, 124, 271-287. Duarte, C. M. Submerged aquatic vegetati on in relation to different nutrients regimes. Ophelia 1995, 41, 87-112. Duarte, C. M.; Kalff, J. Littoral slope as a predictor of the maximum biomass of submerged macrophytes communities. Limnol. Oceanogr. 1986, 31, (5), 1072-1080. Everett, J. D.; Baird, M. E.; Suthers, I. M. Nutrient and pl ankton dynamics in an intermittently closed/open lagoon, Smit h Lake, South-eastern Australia: An ecological model. Estuarine, Coastal and Shelf Science 2007, 72, 690702. FDEP 1996 Water quality assessment for the state of Florida 305 (b) report; Florida Department of Environmental Protection: Tallahassee, FL, 1996. FDEP 2000 Florida water quality assessment 305 (b) report; Florida Department of Environmental Protecti on: Tallahassee, FL, 2000. FDEP 2006 Integrated water quality assessment for Florida ; 305 (b) report; Florida Department of Environm ental Protection: Tallahassee, FL, 2006. FDEP Integrated water quality assessemnt fo r Florida: 2004 305 (b) report and 303 (d) list update ; Florida Department of En vironmental Protection: Tallahassee, FL, July 29, 2004. Fenchel, T. Studies on the decompositi on of organic detritus derived from the turtle grass Thalassia testudinum. Limnol. Oceanogr. 1970, 15, 14-20.
122 Fisher, M. M.; Reddy, K. R. Phosphorus Flux from Wetland Soils Affected by Long-Term Nutrient Loading Journal of Environmental Quality 2001, 30, 261-271. Fisher, T. R.; Carlson, P. R.; Barber, R. T. Sediment nutr ient regeneration in three North Carolina estuaries. Estuarine Coastal Shelf Sci. 1982, 14, 101-116. Fleming, L. E.; Rivero, C.; Burns, J.; Williams, C.; Bean, J. A.; Shea, K. A.; Stinn, J. Blue green algal (cyanobacterial) to xins, surface drinking water, and liver cancer in Florida. Harmful Algae 2002, 1, 157-168. Forsberg, C.; Ryding, S. O. Eutrophica tion parameters and tr ophic state indices in 30 Swedish waste-receiving lakes. Archiv fur Hydrobiol. 1980, 89, (1/2), 189-207. Gardner, W. S.; Nalepa, T. F.; Quickley, M. A.; Na kczyk, J. M. Release of phosphorus by certain vertebrates. Can. J. Fish. Aquat. Sci. 1981, 38, 978-981. Gatcher, R.; Meyer, J. S.; Mares, A. Contribution of bacteria to release and fixation of phosphorus in lake sediemnts. Limnol. Oceanogr. 1988, 33, 1542-1558. Genkai-Kato, M.; Carpenter, S. R. Eutrophication due to phosphate recycling in relation to lake morphometry, temperature, and macrophytes. Ecology 2005, 86, (1), 210-219. Graham, W. F.; Duce, R. A. Atmospher ic pathways of the phosphorus cycle. Geochemica et Cosmochimica Acta 1979, 43, 1195-1208. Graneli, W.; Solander, D. Influence of aquatic macrophytes on phosphorus cycling in lakes. Hydrobiologia 1988, 170, (1), 245-266. Griffin, M., Description of methodology to determine percentage of area and volume of Submerged Aquatic Vegetation. In Tampa, 2008. Gu, B.; De Busk, T.; Dierberg, F. E.; Ch imney, M.; Pietro, K. Phosphorus removal from Everglades agricultural ar ea runoff by submerged aquatic vegetation/limerock treatment technol ogy: an overview of research. Water Sci. Technol. 2001, 44, 101-108. Gumbricht, T. Nutrient removal proce sses in freshwater submersed macrophyte systems Ecological Engineering 1993, 2, (1), 1-30.
123 Hamilton, D. P.; Mitchell, S. F. An empirical model fo r sediment resuspension in shallow lakes. Hydrobiologia 1996, 317, 209-220. Hansen, K.; Mouridsen, S.; Kr istensen, E. The impact of Chironomus plumosus larvae on organic matter decay and nutri ent (n. p.) exchange in a shallow eutrophic lake sediment following a phytoplankton sedimentation. Hydrobiologia 1998, 364, 65-74. Hansson, L.-A.; Enell, M.; Bergman, E. Lake Ringsjn: its catchment area, its history and its importance. Hydrobiologia 1999, 404, (0), 1-7. Harte, J., Consider a Spherical Cow, A course in Environmental Problem Solving University Science Books: Mill Valley, California, 1988; p 283. Hecky, R. E.; Kilham, P. Nu trient limitations of phyt oplankton in freshwater and marine environments: A review of re cent evidence on the effects or enrichment. Limnol. Oceanogr. 1988, 33, (4), 796-822. Hilssborough County. http://www.hillsboroughcounty.org/ managementbudget/communitystatistic s/ (accessed Jan 2008). Howarth, R. W.; Billen, G.; Swaney, D.; Townsend, A.; Jaworsky, N.; Lajtha, K.; Downing, J. A.; Elmgren, R.; Caraco N. F.; Jordan, T.; Berendse, F.; Freney, J.; Kudeyarov, V.; Murdoch, P.; Zhao-Liang, Z. Regional nitrogen budgets and riverine N & P fluxes for the drainage to the North Atlantic Ocean: Natural and human influences. Biogeochemistry 1996, 35, 75-139. Hoyer, M. V.; Frazer, T. K.; Notestein, S. K.; Canf ield, D. E. J. Vegetative characteristics of three lo w-lying Florida coastal rivers in relation to flow, light, salinity and nutrients. Hydrobiologia 2004, 528, 31-43. James, C. S.; Eaton, J. W.; Hardwick, K. Responses of three invasive aquatic macrophytes to nutrient enrichment do not explain their observed field displacements. Aquat. Bot. 2006, 84, (4), 347-353. James, T. R.; Bierman Jr., V. J. A pr eliminary modeling analysi s of water quality in lake Okeechobee, Florida. Wat. Res. 1995, 29, (12), 2755-2766. Jeppesen, E.; Jensen, J. P.; Sondergaard, M.; Lauridsen, T.; Pedersen, L. J.; Jensen, L. Top-down control in freshw ater lakes: the role of nutrients state, submerged macrophytes and water depth. Hydrobiologia 1997, 342/343, 151-164.
124 Johannes, R. E. Influenze of marine protozoa on nutrients regeneration. Limnol. Oceanogr. 1965, 10, 203-213. Johannes, R. E. Nutrient regeneration in lakes and oceans. Adv. Microbiol. 1968, 1, (203-213). Johnes, P.; Moss, B.; Philips, G. The det ermination of total nitrogen and total phosphorus concentrations in freshwat ers from land use, stock headage and population data: testing of a model for use in conservation and water quality management. Freshwater Biology 1996, 36, (2), 451-473. Johnson, P. T. J.; Lunde, K. B.; Ritchie E. G.; Launer, A. E. The Effect of Trematode Infection on Amphibian Limb Development and Survivorship. Science 1999, 284, (5415), 802-804. Jones, J.; Gardener, S.; Simon, B. Bacterial reduction of ferric iron in a stratified eutrophic lake. J. Gen. Microbiol. 1983, 129, (1), 131-139. Jorgensen, S. E. The applic ation of models to find the relevance of residence time in lake and reservoir management. J. Limnol. 2003, 62, 16-20. Jorgensen, S. E.; Verdonschot, P.; Lek, S. Explanation of the observed structure of functional feeding groups of aquat ic macro-invertebrates by an ecological model and the ma ximum exegy principle. Ecological Modelling 2002, 158, 223-221. Karjalainen, M.; Engstrm-Ost, J.; Korp inen, S.; Peltonen, H.; Pkknen, J. P.; Rnkknen, S.; Suikkanen, S.; Viitasal o, M. Ecosystem consequences of cyanobacteria in the northern Baltic Sea. Ambio 2007, 36, (2-3), 195-202. Kitchell, j. f.; O'Neil, R. V .; Webb, D.; Galepp, G. W.; Bart ell, S. M.; Koonce, J. F.; Ausmus, B. S. Consumer r egulation of nutrient cycling. BioScience 1979, 29, 28-34. Kittiwanich, J.; Yamamoto, T.; Kawaguc hi, O.; Hashimoto, T. Analysis of phosphorus and nitrogen cycling in the estuarine ecosystem of Hiroshima Bay by a pelagic and benthic coupled model. Estuarine, Coastal and Shelf Science 2006, 75, 189-204. Knight, R. L.; Gu, B.; Clarke, R. A.; Newman, J. M. Long-term phosphorus removal in Florida aquatic systems dominated by submerged aquatic vegetation. Ecol. Eng. 2003, 20, 45-63. Koenig, S.; Eilers, D. Lake Assessment Report for Lak es in Hillsborough County Florida ; University of South Florida C enter for Comunity Design and
125 Research, and Hillsborough County Stormwater Management Section: Tampa, 2006-2007, 2006-2007. Komatsu, E.; Fukushima, T.; Shiraishi, H. Modelling of P-dynamics and algal growth in a stratified reservoir-m echanisms of P-cycle in water and interaction between overlaying water and sediment. Ecological Modelling 2006, 197, 331-349. Krivtsov, V.; Bellinger, E.; Sigee, D.; Cor liss, J. Interrelations between Si and P biogeochemical cycles a new appr oach to the solution of the eutrophication problem. Hydrological Processes 2000, 14, 283-295. Krom, M. D.; Berner, R. A. Adsorption of phosphate in anoxic marine sediments. Limnol. Oceanogr. 1980, 25, 797-806. Krom, M. D.; Berner, R. A. The diagenesis of phosphorus in a nearshore marine sediment. Geochim. Cosmochim. Acta 1981, 45, (207-216). Lahm, T. Computational Science Acro ss the Curriculum Project Capital University Phosphorus Cycling Model. http://www.capital.edu/68/Artsand-Sciences/Computational-S tudies/10649.pdf (May 9th, 2008). Landers, D. H. effects of naturally senescing aquatic macrophytes on nutrient chemistry and chlorophyll-a of surrounding waters. Limnol. Oceanogr. 1982, 27, (3), 428-439. Lantzke, I. R.; Mitchell, D. S.; Heritage, A. D.; Sharma, K. P. A model of factors controlling orthophosphate removal in planted vertical flow wetlands. Ecological Engineering 1999, 12, 93-105. Larsen, D.; Van Sickle, J.; Malueg, K.; Smith, P. The Effect of Wastewater Phosphorus Removal on Shagawa Lake, Minnesota: Phosphorus Supplies, Lake Phosphorus and Chlorophyll A Water Research 1979, 13, (12), 1259-1272. Levich, A. P. The role of nitrogen-phosphorus ratio in selecting for dominance of phytoplankton by cyanobacteria or green algae and its application to resrvoir managemnet. Journal of Aquaticystem Stress and Recovery 1996, 5, (1), 55-61. Levich, A. P.; Bulgakov, N. G. Regulatio n of species and size composition in phytoplankton communities in situ by N:P ratio. Russian Journal of Aquatic Ecology 1992, 1, (2), 149-159.
126 Lewis, R. R. I.; Estevez, E. D. The Ecology of Tampa Bay, Florida: An estuarine profile ; BR-85(7.18); U. S. Department of the Interior, Fish and Wildlife Service Washington D. C., 1988; p 132. Lin, D. Y.; Ying, Z. Semi parametric regression analysis of longitudinal data with informative drop-outs. Biostatistics 2003, 4, (3), 385-395. McCarthy, T., Review: [untitled] Pub lished by: The President and Fellows of Harvard College In The Business History Review 2000; Vol. 74, pp 734736 McRoy, C. P.; Barsdate, R. J.; Neber t, M. Phosphorus cycling in aneelgrass ( Zostera marina L .) ecosystem. Limnol. Oceanogr. 1972, 17, 58-67. Menzel, D. W.; Corwin, N. The measurement of total phosphorus in seawater based on th eliberation of organically bound fractions by persulfate oxidation. Limnol. Oceanogr. 1965, 10, 280-282. Moore, P. A.; Reddy, K. R.; Fisher, M. M. Phosphorus flux between sediemnt and overlaying water in Lake Okeechobee, Florida : Spatial and temporal variations. J. Environ. Qual. 1998, 27, (6), 1428-1439. Moore, P. A.; Reddy, K. R.; Graetz, D. A. Phosphorus geochemistry in the sediment-water column of a hypereutrophic lake. Journal of Environmental Quality 1991, 20, (4), 869-875. Morris, K.; Bailey, P. C. E .; Boon, P. I.; Hughes, L. Ef fects of plant harvesting and nutrient enrichment on phytoplankton co mmunity structure in a shallow urban lake. Hydrobilogia 2006, 571, 77-91. Mortimer, D. H. Chemical exchanges bet ween sediemnts and water in the Great Lakes-speculations on probabl e regulatory mechanisms. Limnol. Oceanogr. 1971, 16, 387-404. Moss, B. Engineering and biological appr oaches to the restoration from eutrophication of shallow lakes in wh ich aquatic plant communities are important components. Hydrobilogia 1990, 200-201, (1), 367-377. O'Dell, K. M.; VanArman, J.; Welch, B. H. ; Hill, S., D. Changes in water chemistry in a macrophyte-dominated lake bef ore and after herbicide treatment. Lake and Reserv. Manage. 1995, 11, (4), 311-316. Omernik, J. M. The Influence of Land Use on Stream Nutrient Levels ; Report EPA 600/3-76-014; United States Envi ronmental Protection Agency: January 1976, 1976; p 114
127 Ott, L. R., An introduction to statistical methods and data analysis. Duxbury Press. : Belmont, California,, 1993; p 1182. Ozimek, T.; Gulati, R. D.; van Donk E. Can macrophytes be useful in biomanipulation of lakes? T he Lake Zwemlust example Hydrobiologia 1990, 200-201, (1), 399-407. Paerl, H. W. Nuisance Phytoplankton Bloom s in Coastal, Estuarine, and Inland Waters. Limnology and Oceanography, 1988, 33, (4, part 2), 823-847 Perkins, R. G.; Underwood, G. J. C. Gradients of chlorophyll a and water chemistry along an eutrophic reservoir with determination of the limiting nutrient by in situ nutrient addition Water Research 2000, 34, (3), 713-724 Philips, E. J.; Cichra, M.; Havens, C. ; Hanlon, C.; Badylak, S.; Reuter, B.; Randall, M.; Hansen, P. Relation ships between phytoplankton dynamics and the availability of light and nutri ents in a shallow sub-tropical lake. J. Plank. Res. 1997, 19, (3), 319-342. Philips, G. L.; Eminison, D.; Moss, B. A mechanism to account for macrophyte decline in progressively eut rophicated freshwaters. Aquatic Botany 1978, 4, (103-126). Poor, N.; Pollman, C.; Tate, P.; Begum, M.; Evans, M.; Campbell, S. Nature and magnitude of atmospheric fluxes of total inorganic nitrogen and other inorganic species to the Tampa Bay watershed, FL, USA. Water, Air, & Soil Pollution 2005, 170, (1-4 ), 267-283. Rast, W.; Thorton, J. Trends in eu trophication research and control. Hydro. Proc. 1996, 10, 295-313. Rattray, M. R.; Williams, C. H.; Brown, J. M. A. Sedi ment and water as sources of nitrogen and phosphorus for submer ged rooted aquatic macrophytes. Aquat. Bot. 1991, 40, 225-237. Reddy, K. R.; O'Connor, G. A.; Schelske, C. L., Phosphorus Biogeochemistry in Subtropical Ecosystems CRS Press LLC: Boca Raton, Florida, 1999; p 707. Rhee, G. Competition between an alga and an aquatic bacterium for phosphate. Limnol. Oceanogr. 1972, 17, (4), 505-514. Richardson, C. J. Mechanisms controlling retention capacity in freshwater wetlands. Science 1985, 228, 1424-1427.
128 Richardson, C. J. Predictive models for phosphorus retention in wetlands. Wetlands Ecology and Management 1996, 4, (3), 159-175. Richardson, C. J.; Qian, S.; Craft, C. B.; Qualls, R. G. Predictive models for phosphorus retention in wetlands. Wetlands Ecology and Management 2005, 4, 159-175. Rigler, F. H. A tracer study of the phosphorus cycle in lake water. Ecology 1956, 37, 550-562. Sartory, D. P.; Grobbelar, J. U. Extraction of chlorophyllfrom freshwater phytoplankton for spectrophotometric analysis. Hydrobiologia 1984, 114, (3), 177-187. Schauser, I.; Hupfer, M.; Bruggeman, R. SPIEL a model for phosphorus diagenesis and its application to lake restoration. Ecological Modelling 2004, 176, 389-407. Schauser, I.; Hupfer, M.; Br uggemann, R. Sensitivity a nalysis with a phosphorus diagenesis model (SPIEL). Ecological Modelling 2006, 190, 87-98. Scheffer, M., Ecology of Shallow Lakes Kluwer Academic Publishers: Norwell, MA, 2004; p 357. Scheffer, M.; Carpenter, S. R.; Foley, J.; Folke, C.; Walker, B. Catastrophic shifts in ecosystems. Nature 2001, 413, 591-596. Schindler, D. W. Evolution of phosphorus limitation in Lakes. Science 1977, 195, 260-262. Schlesinger, W. H., Biogeochemistry Academic Press, Inc.: San Diego, California, 1991; p 443. Schmidt, N.; Luther, M. E. ENSO Impacts on Salinity in Tampa Bay, Florida. Estuaries 2002, 25, (5), 976-984. Schnoor, J. L., Environmental Modeling Fate and Transport of Polutants in Water, Air, and Soil John Wiley & Sons, Inc.: Iowa City, Iowa, 1996; p 676. Scinto, L. J.; Reddy, K. R. Biotic and abiotic uptake of phosphorus by periphyton in a subtropical freshwater wetland. Aquatic Botany 2003, 77, 203-222.
129 Serpa, D.; Falcao, M.; Vale, C. Eval uation of ammonium and phosphate release from intertidal and subtidal sediment s of a shallow co astal lagoon (Ria Formosa Portugal): a modelling approach. Biogeochemistry 2007, 82, 291-304. Silverberg, N.; Barker, J.; Edenborn, H. ; Sundby, B. Oxygen profiles and organic carbon fluxes in Laurentian Trough sediments J. Sea Res. 1987, 21, 95105. Simal, J.; Lage, M. A.; Ig lesias, I. Second derivative ultraviolet spectroscopy and sulfamic acid method for determination of nitrates in water. JANCA 1985, 68, (5), 962-964. Simpson, P. S.; Eaton, J. W. Comparative studies of the photosynthesis of the submerged macrophyte Elodea c anadensis and the filamentous algae Cladophora glomerata and Spirogyra sp. Aquatic Botany 1986, 24, (1), 112. Smith, E. P.; McCormick, P. V. Lon-te rm relationship between phosphorus inputs and wetland phosphorus concentration in a northern Everglades marsh. Environmental Monitoring and Assessment 2001, 68, 153-176. Smith, S. V.; Swaney, D. P.; Talaue-McManus L.; Bartley, J. D.; Sandhei, P. T.; McLaughlin, C. J.; Dupra, V. C.; Cr ossland, C. J.; Buddemeier, R. W.; Maxwell, B. A.; Wulff, F. Humans hydrology, and the distribution of inorganic nutrient loading to the ocean Bioscience 2003, 53, (3), 235-245. Soranno, P. A.; Hubler, S. L.; Carpenter S. R. Phosphorus loads to surface waters: a simple model to account for spatial pattern of land use. Ecological Applications 1996, 6, (3), 865-878. Southwest Florida Water Management District Lake Thonotosassa Surface Water Improvement and Management (SWIM) Plan ; Southwest Florida Water Management District: Plant City, February 2003, 2003; p 25. Spears, B. M.; Carvalho, L. ; Perkins, R.; Paterson, D. M. Effects of light on sediment nutrient flux and water column nutrient stoichiometry in a shallow lake. Water Research 2008, 42, 977-986. Spence, D. H. N. The zonation of plants in freshwater lakes. Advan. Ecol. Res. 1982, 12, 37-125. Sundby, B.; Gobeil, C.; Silverberg, N.; Mucci, A. The phosphorus cycle in coastal marine sediments. Limnol. Oceanogr. 1992, 37, (6), 1129-1145.
130 Tampa Bay Regional Planning Council. http://www.tbrpc.org/resource_cente r/publications.shtml (accessed Jan 2008). TBEP, Charting the Course: The Comprhen sive Conservation and Management Plan for Tampa Bay Tampa Bay Estuary Progr am: St. Petersburg, FL, 2006. Terrell, J. B.; Watson, D. L. ; Hoyer, M. V.; Allen, M. S.; Canfield, D. E. J. Temporal water chemistry trends (1967-1997) for a sample (127) of Florida waterbodies. Lake and Reserv. Manage. 2000, 16, (3), 177-194. Tett, P.; Wilson, H. From biogeochemic al to ecological models of marine microplankton. Journal of Marine Systems 1999, 25, 431-446. US Bureau of the Census 2000. http://www.census.gov/Press-Release/ (accessed Jan 2008). US Bureau of the Census. http://factfinder.census.gov/ USEPA Harmful Algal Blooms. STAR REPORT ; EPA/600/F-97/025; United States Environmental Protection Agency: Dec. 1997, 1997; p 6. van Beynen, P.; Feliciano, N.; North, L. ; Townsend, K. Application of a karst disturbance index in Hills borough County, Florida. Environmental Assessment 2007, 39, 261-277. Wang, N.; Mitsh, W. J. A detailed eco system model of phosphorus dynamics in created riaprian wetlands. Ecological Modeling 2000, 126, 101-130. Ware, J. H. Linear Models for the analysis of longitudinal studies. The American Statistician 1985, 39, (2), 95-101. Water Atlas. http://www.wateratlas.usf.edu/ (accessed Jan 2008). Wollin, K. M. Nitrate det remination in surface waters as an example of the application of UV derivative spectr oscopy to environmental analysis. Acta Hydrochim. Hydrobiol. 1987, 15, 459-469. Xiana, G.; Craneb, M. Assessments of urban growth in the Tampa Bay watershed using remote sensing data Remote Sensing of Environment 2005, 97, (2), 203-215. Zeguer, S. L.; Liang, K.; Albert, P. S. Models for longitudinal data: a generalized estimating equation approach. Biometrics 1988, 44, 1049-1060.
131 Zhou, Q.; Gibson, C. E.; Zhu, Y. Eval uation of phosphorus bioavailability in sediments of three contrasting lakes in China and UK. Chemosphere 2001, 42, 221-225.
133 Appendix A: General Information about Lakes Examined for Eutrophication Trends Hillsborough County Lakes Area (ha) Mean Depth (m) Volume (m3) LatitudeLongitude Watershed Thonotosassa 344 2.4 N/A 28 03 39 -82 16 39 Pemberton Creek Keystone 175 3.3 571417528 07 59 -82 35 24 Brooker Creek Magdalene 83 2.4 238571228 04 55 -82 28 55 Sweet Water Creek Carroll 82 2.4 204418328 03 04 -82 29 15 Sweet Water Creek Hiawatha 55 3.3 187359528 10 10 -82 34 54 Anclote river Calm 47 2.7 147784928 08 32 -82 34 54 Brooker Creek Armistead 14 2.7 347739 28 06 04 -80 33 35 Rocky/Brushy Creek Deer 14 3.6 501064 28 10 04 -82 27 45 Rocky/Brushy Creek Sunset 13 2.4 352140 28 08 06 -82 37 32 Brooker Creek Keene 13 2.7 317223 28 08 40 -82 26 53 Cypres Creek Crenshaw 12 1.5 42304 28 07 33 -82 29 45 rocky/Brushy Creek Juanita 10 2.7 246887 28 07 03 -82 35 20 Brooker Creek Dead Lady 1.2 0.9 2087 28 09 18 -82 34 14 Brooker Creek
134 Appendix A: (Continued) Pinellas County Lakes Area (ha) Mean Depth (m) Volume (m3) Latitude Longitude Watershed Alligator 32 N/A N/A 27 58 55 -82 41 50 Old Tampa Bay Chautauqua 22 N/A N/A 28 00 15 -82 43 21 Old Tampa Bay Manatee County Lakes Area (ha) Mean Depth (m) Volume (m3) Latitude Longitude Watershed Ward Lake 103 N/A N/A 27 25 40 -82 29 09 Manatee River Polk County Lakes Area (ha) Mean Depth (m) Volume (m3) Latitude Longitude Watershed haunter 87 N/A N/A 28 01 58 -81 57 57 Hillsborough River Bonnet 32 N/A N/A 28 02 51 -81 58 36 Hillsborough River Beulah 7 N/A N/A 28 02 26 -81 58 06 Hillsborough River
135 Appendix B: One Time Values of Variables of Lakes Examined for Submerged Vegetation Lake PAC % PIV % TP (g L-1) TN (mg L-1 Chlorophyll(g L-1) Area (ha) Mean Depth (m) Volume (M3) Alice 85 41 19 0.371.20 37.23 2.74 941847 Carroll 85 35 23 0.451.40 81.75 2.44 2044183 White Trout 77 44 14 0.603.20 30.35 3.35 1011391 Reinheimer 77 25 12 1.203.80 8.09 1.83 236600 Magdalene 76 47 14 1.073.80 83.37 2.44 2385713 Eckles 71 27.3 30 1.275.40 11.33 2.13 256854 Mound 69 32 20 0.522.50 30.35 3.96 1280673 Raleigh 67 23 3 0.606.90 9.71 2.74 254667 George 63 33 31 0.504.20 10.93 3.66 378654 Cypress 56 16 10 0.544.40 6.48 3.66 225021 Round 56 17.1 21 0.453.60 4.05 2.74 99847 Horse 46 19 21 0.893.80 10.93 2.13 146282 Rogers 44 13 17 0.9514.40 38.04 2.44 746805 Pine 44 17.9 40 0.998.60 3.24 2.44 555292 Noreast 40 14.1 27 0.729.70 3.24 1.52 87071 Calm 39 9 22 0.414.00 46.54 3.35 1477849 Island Ford 38 12 25 0.878.80 36.02 3.05 1131957 Keystone 38 12 25 1.133.70 174.43 3.35 5714176 Crescent 35 10 35 0.94 18.21 2.74 553353 Dead Lady 34 12.9 50 0.946.60 1.21 0.91 2087 Elizabeth 30 10 24 0.866.30 7.69 3.66 272512 Taylor 2 30 13 12 0.646.00 19.02 2.74 543649 Rainbow 26 9 10 0.778.40 19.02 2.74 544936 Juanita 21 10 10 0.9911.40 9.71 2.74 246887 Crenshaw 20 8.2 22 0.8311.40 12.14 1.52 42304 Church 15 4.6 26 0.525.90 25.09 1.22 138039 Cedar East 8 4.6 33 0.616.60 1.21 1.52 33639 Armistead 7 12 45 1.0921.30 13.76 2.74 347739 Rock 6 4 35 0.9121.70 21.45 2.13 431011 Brant 5 0.5 35 0.939.00 22.26 1.83 384649 Saddleback 3.5 9 27 1.0811.40 12.55 1.52 226065 Cedar West 3 2.4 41 0.7817.30 2.02 1.83 18916 Pretty 2 1 33 0.9511.90 32.78 3.35 1068395 Josephine 2 1 44 0.8512.00 20.24 2.13 422013
136 Appendix B: (Continued) Lake Latitude LongitudeWatershed Alice 2807 -82 36 14 Brooker Creek Carroll 28 03 04 -82 29 15 Sweetwater Creek White Trout 28 02 21 -82 29 46 Sweetwater Creek Reinheimer 28 07 48 -82 29 12 Rocky/Brushy Creek Magdalene 28 04 55 -82 28 55 Sweetwater Creek Eckles 28 03 19 -82 28 19 City of Tampa Mound 28 08 51 -82 34 19 Brooker Creek Raleigh 28 06 21 -82 35 02 Brooker Creek George 28 04 07 -82 29 14 Sweetwater Creek Cypress 28 07 32 -82 33 52 Rocky/Brushy Creek Round 28 07 14 -82 30 00 Rocky/Brushy Creek Horse 28 06 38 -82 34 44 Brooker Creek Rogers 28 06 32 -82 35 19 Brooker Creek Pine 28 03 38 -82 28 20 Curiosity Creek Noreast 28 03 45 -82 28 07 Curiosity Creek Calm 28 08 32 -82 34 54 Brooker Creek Island Ford 28 09 08 -82 35 56 Brooker Creek Keystone 28 07 59 -82 35 24 Brooker Creek Crescent 28 09 29 -82 35 31 Brooker Creek Dead Lady 28 09 18 -82 34 14 Brooker Creek
137 Appendix B: (Continued) Lake Latitude Longitude Watershed Elizabeth 28 09 26 -82 34 24 Brooker Creek Taylor 2 28 08 12 -82 36 43 Brooker Creek Rainbow 28 07 00 -82 35 46 Brooker Creek Juanita 28 07 03 -82 35 20 Brooker Creek Crenshaw 28 07 33 -82 29 45 Rocky/Brushy Creek Church 28 06 11 -82 35 58 Brooker Creek Cedar East 28 03 56 -82 28 13 Curiosity Creek Armistead 28 06 04 -82 33 35 Rocky/Brushy Creek Rock 28 06 48 -82 33 24 Rocky/Brushy Creek Brant 28 07 35 -82 28 20 Rocky/Brushy Creek Saddleback 28 07 13 -82 29 41 Rocky/Brushy Creek Cedar West 28 03 55 -82 28 21 Curiosity Creek Pretty 28 06 27 -82 34 04 Rocky/Brushy Creek Josephine 28 06 35 -82 33 43 Rocky/Brushy Creek
ABOUT THE AUTHOR Max Jacobo Moreno Madrian received his Bachelors degree in Agriculture Science in 1993 from the Universidad Nacional de Colombia, in Colombia. He received his Masters degree in Environmental Management for Sustainable Development wit h Emphasis on Coastal Zones from the Pontificia Universidad Javeriana, in Colombia. During his doctoral studies at Univ ersity of South Florida, he has presented the results of his research at three regional meeti ngs of the Air and Waste Management Association and at one meeting of the Im plementation and Integration Workshop of the Gu lf of Mexico Alliance. He received two awards for his research studies.