<%BANNER%>

Factors Influencing Partial Pressure Carbon Dioxide Levels in Florida Lakes: pH, Aquatic Macrophytes, and Geology


PAGE 1

FACTORS INFLUENCING PARTIAL PRESSURE CARBON DIOXIDE LEVELS IN FLORIDA LAKES: pH, AQUATIC MACROPHYTES, AND GEOLOGY By JENNEY L. KELLOGG A THESIS PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE UNIVERSITY OF FLORIDA 2007 1

PAGE 2

2007 Jenney L. Kellogg 2

PAGE 3

ACKNOWLEDGMENTS Gratitude is expressed to the many people who helped me carry out this project. Special thanks are given to Dr. Carlos Duarte and Dr. Yves Prairie who helped to initiate my project and begin data analysis. Thanks to Dr. Roger Bachmann who helped me with my project in ways too numerous to list. The staff of Florida LAKEWATCH provided guidance on sampling protocol and equipment and helped with the chemical analysis of my samples. Thanks go to Francesco Lazzarino for his help with data entry and analysis and expertise on programming languages. Gratitude is expressed to Christy Horsburgh for her continued support and advice. Thanks are given to Dr. Charles E. Cichra for encouraging my professional growth. Thanks are given to Mark Hoyer for his support and guidance on this project. Gratitude is expressed to my committee members for helping to oversee my project. Finally, thanks are given to Dr. Daniel E. Canfield, Jr., my mentor, advisor and chair for his patience, ingenuity and guidance throughout this project. 3

PAGE 4

TABLE OF CONTENTS page ACKNOWLEDGMENTS ...............................................................................................................3 LIST OF TABLES ...........................................................................................................................5 LIST OF FIGURES .........................................................................................................................6 ABSTRACT .....................................................................................................................................7 CHAPTER 1 INTRODUCTION....................................................................................................................9 2 MATERIALS AND METHODS...........................................................................................14 3 RESULTS AND DISCUSSION.............................................................................................22 Partial Pressure CO 2 Values from Historical Lake Data........................................................22 Limnological Variables..........................................................................................................23 Effect of Macrophyte Presence on pCO 2 Levels in a Florida Lake........................................25 Lake Metabolism in Relationship to pCO 2 Levels in a Florida Lake.....................................27 Heterotrophy in Lakes............................................................................................................28 4 CONCLUSIONS....................................................................................................................46 WORKS CITED............................................................................................................................49 BIOGRAPHICAL SKETCH.........................................................................................................53 4

PAGE 5

LIST OF TABLES Table page 3-1 Comparison of various limnological and physical parameters measured in the 627-lake study to Coles 1994 study.........................................................................................30 3-2 Mean, standard error and range of the variables measured in 627 lakes...........................31 3-3 Correlation coefficients for all variables in relation to pCO 2 with N values....................34 3-4 Average pCO2 and pH levels for each Florida Region.....................................................40 3-5 Average of nutrient and chlorophyll concentrations by month and station in Orange Lake, Florida......................................................................................................................44 3-6 Average carbon estimate variables and wind by month and station in Orange Lake, Florida................................................................................................................................44 3-7 Average pCO 2 by month and station in Orange Lake, Florida..........................................45 3-8 Aquatic plant data collected on July 7, 2005. Source: Florida LAKEWATCH 2005.....45 3-9 Percent coverage of macrophytes in Orange Lake and the estimated pCO 2 for that plant coverage value..........................................................................................................45 3-10 Net ecosystem production values for each station by month with an average for all samples at each station in Orange Lake, Florida...............................................................45 5

PAGE 6

LIST OF FIGURES Figure page 1-1 The relationship of free carbon dioxide and pH taken from Juday et al. (1935). The dotted line horizontal line indicates atmospheric level of carbon dioxide. Free CO 2 is represented by the dashed line Calculated CO 2 ..............................................................12 1-2 Relation between pH and the relative proportions of inorganic carbon species modified from Wetzel 2001...............................................................................................13 2-1 Satellite photo of Orange Lake, Florida with sampling sites indicated by diamonds........21 3-1 Locations of 627 lakes sampled in the state of Florida......................................................32 3-2 Relationship of pH to pCO 2 (atm) values for the 627 lake study. Lake values are represented by open cirlces, with modeled pH changes shown in black. The black line indicates atmospheric pCO 2 level...............................................................................33 3-3 Relationship of pH to pCO 2 (atm) values for the 627 lake study....................................34 3-4 Relationship of TP (g/L) to pCO 2 (atm) values for the 627 lake study........................35 3-5 Relationship of TN (g/L) to pCO 2 (atm) values for the 627 lake study.......................36 3-6 Relationship of Chlorophyll (g/L) to pCO 2 (atm) values for the 627 lake study.........37 3-7 Relationship of color (Pt-Co Units) to pCO 2 (atm) values for the 627 lake study.........38 3-8 Relationship of specific conductance (S/cm @ 25C) to pCO 2 (atm) values for the 627 lake study..............................................................................................................39 3-9 Average pH and pCO 2 values by region. Regions move from northwest of Florida to south Florida in a southeast direction................................................................................41 3-10 Monthly pCO 2 averages by station for Orange Lake.........................................................42 3-11 Partial pressure carbon dioxide values in relationship to % oxygen saturation for 284 historical lakes. The vertical line represents 100% oxygen saturation. The horizontal line represents atmospheric level of pCO 2 ........................................................43 6

PAGE 7

Abstract of Thesis Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Master of Science FACTORS INFLUENCING PARTIAL PRESSURE CARBON DIOXIDE LEVELS IN FLORIDA LAKES: pH, AQUATIC MACROPHYTES, AND GEOLOGY. By Jenney L. Kellogg May 2007 Chair: Daniel E. Canfield Jr. Major: Fisheries and Aquatic Sciences Physical, chemical, and biological factors associated with the supersaturation of CO 2 in aquatic ecosystems are not clearly evident. The goal of our study was to evaluate partial pressure carbon dioxide (pCO 2 ) levels in a set of Florida Lakes and to investigate if any limnological factors are correlated with levels of pCO 2 The pCO 2 levels were calculated for 627 Florida Lakes using pH, specific conductance, alkalinity and water temperature. Eighty percent of lakes were found to be supersaturated with carbon dioxide. Trophic state indicators such as total phosphorus (TP), total nitrogen (TN) and chlorophyll (Chl) were all weakly correlated with pCO 2 concentrations (TP, r = -0.30 TN, r = -0.35 Chl r = -0.48). The limnological parameter pH had a highly significant positive correlation with pCO 2 concentration (r = 0.91). When modeling changes in pCO 2 concentrations, pH seems to be the regulating factor in pCO 2 levels in Florida lakes. Plants remove carbon dioxide from the atmosphere by carbon assimilation, and produce carbon dioxide during cellular respiration processes. Photosynthesis and respiration directly influence the pH of the water. The effect of aquatic macrophyte presence on pCO 2 levels was examined using a summer study (June-August 2005) that examined pCO 2 levels in both the plant and open water areas in a hypereutrophic Florida lake. The presence of plants in a 7

PAGE 8

hypereutrophic Florida Lake significantly increased pCO 2 measurements. The pH was found to be lower in plant stations in comparison to open water stations. The same study was used to evaluate how the presence of plants affected lake metabolism using the diurnal curve method. Presence of plants significantly decreased the net ecosystem production estimates based on dissolved oxygen and wind data. All stations on the study lake (both plant and open water) were heterotrophic over the duration of the study. Saturation of carbon dioxide in an aquatic ecosystem is often used an in indicator of heterotrophy at the ecosystem scale. To investigate this statement, dissolved oxygen data on a group of 284 Florida Lakes were used to examine the relationship between pCO 2 and dissolved oxygen saturations. Several lakes exhibited both supersaturation of oxygen and carbon dioxide simultaneously. Therefore, dissolved oxygen was determined to be an inadequate predictor of lake heterotrophy in 25% of 284 Florida lakes. The limnological parameter pH is often called the master variable in determining the aqueous geochemistry of an aquatic system. Many attempts have been made to quantify and rank variables of significance to predict mean values of lake pH. The pH value of a water body is directly influenced by geologic factors. When investigating the relationship between pCO 2 levels and pH across the state of Florida, a geologically regulated trend was found. As pH values increased in a southeastern direction across the state, pCO 2 levels decreased. Changes in pH values between regions were reflected in changes in pCO 2 levels between the same regions. Changes in regional geology may explain variation between lakes in reference to pCO 2 levels. 8

PAGE 9

CHAPTER 1 INTRODUCTION The limnological parameter pH is often considered a master variable in determining the biological, chemical and physical processes of an aquatic system. Juday et al. (1935) examined relationships between pH levels and carbon dioxide levels in several lakes in Wisconsin. Juday et al. (1935) calculated pH values from free carbon dioxide values, and compared these calculated values to directly measured values. They found no significant difference (Figure 1-1). Juday et al. (1935) also found that lakes did not exhibit carbon dioxide levels exceeding the atmospheric level of carbon dioxide above a pH of 8. The relationship between pH and the relative proportions of inorganic carbon species (Figure 1-2) of CO 2 (+H 2 CO 3 ), HCO 3 and CO 3 = in solution is clearly recognized (Wetzel 1983). At a pH of above 8, the proportion of carbon in the form of free carbon dioxide is very small (Figure 1-2). In the past decade, many researchers have supported the idea that organic matter located within lake basins can either be mineralized or lost as CO 2 to the atmosphere (Cole et al. 1994, 2000; Richey et al. 2002; Sobek et al. 2003; Striegl et al. 2001). When the level of carbon dioxide within a water body exceeds the amount of carbon dioxide present in the atmosphere (supersaturation), the water body becomes a carbon source (Kling et al. 1991; Hesslein et al. 1991; Cole et al. 1994, 1998). Cole et al. (1994) conducted a study of carbon dioxide levels in 1835 globally distributed lakes. Cole used both direct and indirect measurements of CO 2 and found that 87% of lakes were supersaturated with CO 2 Cole found a mean partial CO 2 pressure for all lakes of 1036 micro atmospheres. The atmospheric level of carbon dioxide is about 370 micro atmospheres. Physical and chemical and biological factors associated with the supersaturation of CO 2 in aquatic ecosystems are not clearly evident (Cole 1994; Kling 1991). There has been much 9

PAGE 10

investigation into the role of allochthonous carbon and its effect on lake metabolism (Prairie et al. 2002). The role of the lake as a CO 2 producer may vary with trophic state and other limnological factors affecting CO 2 including chemical composition, lake morphometry, and surrounding terrestrial ecosystems (Hanson 2003). It also has not yet been investigated whether the presence of aquatic macrophytes would have an effect on pCO 2 levels in a freshwater lake. Aquatic macrophytes are prevalent in many Florida lakes and play and important role in the structure and function of lakes (Brenner et al. 1991). Macrophytes are a major component of total biomass and primary productivity (Canfield et al. 1983). Aquatic macrophytes are an integral part of lake metabolism. In aquatic ecosystems, metabolism directly affects concentrations of dissolved inorganic carbon (DIC) and dissolved oxygen (DO) concentrations (Hanson 2006). Aquatic macrophytes produce carbon dioxide during cellular respiration processes (C 6 H 12 O 6 + 6 O 2 6 CO 2 + 6 H 2 O + energy) and also excrete dissolved organic compounds which are acted on by microorganisms and are oxidized into carbon dioxide and returned to the atmosphere (Wetzel 1969). Understanding the carbon balance of a lake helps us to understand both the biological activity occurring within a lake and how the surrounding watershed may be affecting the lake. Whole-ecosystem metabolism integrates the net effect of all primary producers and all consumers on the carbon cycle (Hanson 2006). Ecosystem production may be negative or positive (Howarth et al. 1996). When ecosystem production is positive, the system is considered autotrophic, and the water body can store organic carbon (Schindler et al. 1972). When ecosystem production is negative, the system is considered heterotrophic, and respires more carbon than what was produced within the system (Cole 2000). There is much debate regarding 10

PAGE 11

the metabolic balance between heterotorphy and autotrophy in lakes in general (Carignan et al. 2000; del Giorgio and Peters 1994). Given the complexities of dissolved carbon dioxide dynamics in freshwater lakes, an investigation of the physical, chemical and biological factors associated with changes in partial pressure carbon dioxide levels in Florida lakes is needed. My objectives for this study are: Objective 1: To evaluate partial pressure carbon dioxide (pCO 2 ) levels in a set of Florida Lakes; Objective 2: To investigate if any limnological factors are correlated with levels of pCO 2 in a set of Florida Lakes; Objective 3: To investigate the effect of aquatic macrophyte presence on pCO 2 levels in a Florida Lake; Objective 4: Evaluate lake metabolism and its relationship to pCO 2 levels in a Florida Lake; 11

PAGE 12

Figure 1-1. Relationship of free carbon dioxide and pH taken from Juday et al. (1935). Dotted horizontal line indicates atmospheric level of carbon dioxide. Dashed line represents Calculated CO 2 . 12

PAGE 13

Figure 1-2. Relationship between pH and the relative proportions of inorganic carbon species. Adapted from Wetzel R.G. 2001. Limnology. Lake and River Ecosystems. Academic Press, San Diego. 13

PAGE 14

CHAPTER 2 MATERIALS AND METHODS Published water chemistry information was obtained for 627 lakes that were sampled between one and 48 times over a 17-year period of record (Canfield 1981; Canfield and Hoyer 1992; Florida LAKEWATCH 1993, 1996; Greis 1995). Seventy-five percent of the lakes were sampled four or less times, with the remaining 25% being sampled between five and 48 times. One hundred and fifty-nine lakes were sampled 10 or more times. Carbon dioxide concentrations were computed using averaged pH, total alkalinity, and specific conductance values from the data set for each lake using the approach of Cole et al. (1994) and Prairie et al. (2002). Lakes with multiple samples per lake (more than one station) were averaged, followed by an average of the months sampled. A yearly average was taken from the averaged months. Finally all years for each lake were averaged to attain a grand average for each lake. Ionic strength (I) was calculated (Equation 2-1) from specific conductivity (S/cm @ 25 C) values represented by SC (Cole et al. 1994; Prairie et al. 2002) as follows: I = 0.000025 7 SC (2-1) This equation is based on a relation between ionic strength and total dissolved solids, taken from Sawyer and McCarty (1967), which accounts for the first constant (0.000025), and an empirical relation between total dissolved solids and specific conductance, taken from Hem (1992), which accounts for the second constant (7). Equilibrium disassociation constants of carbonic acid to hydrogen atoms and bicarbonates, and the disassociations of carbonates to bicarbonates were calculated (K 1 and K 2 ) where concentrations are indicated by square brackets [ ], activities are denoted by braces { }. These calculations are shown below in Equations 2-2 and 2-3 (Stumm and Morgan 1995). 14

PAGE 15

K 1 = {H + } [HCO 3 ] (2-2) [H 2 CO 3 ] K 2 = {H + } [CO 3 2] [HCO 3 ] (2-3) These values were then used to calculate the fraction of free CO 2 (a 0 ), bicarbonate (a 1 ), and carbonate (a 2 ) (Equations 2-4, 2-5, 2-6 respectively). a 0 = ___________1_____________ 1+10 (K1+pH) + 10 (-(K1+K2)+2*pH) (2-4) a 1 = ________1____________ 1+10(-pH+K1)+ 10(-K2+pH) (2-5) a 2 = ____________1____________ 10(-2*pH+(K1+K2))+10(-pH+K2)+1 (2-6) These three values (a 0 a 1 and a 2 ) together are the total amount of inorganic carbon compounds in the system (Stumm and Morgan 1995). Total carbon in the system (DIC, M) is expressed as a concentration and is derived using the three carbon components (a 0 a 1 and a 2 ) along with alkalinity (meq/L) in Equation 2-7 below. DIC (M)= 105* (Alkalinity(meq/L)*0.000001-10(-14+pH)+10(-pH)) (a 1 +2*a 2 ) (2-7) Using Henrys Law, the solubility of carbon dioxide in water (KH) is determined which varies with temperature (Equation 2-8). KH=1.11+0.016 Temperature(C) 0.00007 Temperature(C)2 (2-8) The concentration of free CO 2 in the system is expressed in micromoles (Equation 2-9): CO 2 (M) = DIC(M) a o (2-9) Finally, CO 2 is expressed as a partial pressure in microatmospheres (Equation 2-10): 15

PAGE 16

pCO 2 (atm)= CO 2 M (2-10) 10(-KH) An average yearly temperature value of 21 C was used for solubility calculations for the 627 lake study due to lack of available water temperature data for the individual lakes. All water samples were collected during daylight hours from the surface (0.5 m) at multiple (1-4) open-water sites within each lake. Water clarity was measured by use of a Secchi disc. Water samples were collected in acid-cleaned, triple rinsed Nalgene bottles, and placed on ice. In the early studies, unfrozen water samples were analyzed for total nitrogen (TN) and total phosphorus (TP). In later studies, portions of the collected water were subsequently frozen and used for later analysis of TN and TP. At the laboratory, a pH meter (an Orion Model 601A pH meter or an Accumet model 10 pH meter) calibrated with pH 4.0 and pH 7.0 buffers was used to measure pH. Total alkalinity (mg/L as CaCO 3 ) was determined by titration with 0.02 N sulfuric acid (APHA 1085). Specific conductance (S/cm @ 25 C) was measured using a Yellow Springs Instrument Model 21 conductivity bridge or a Yellow Spring Instrument Model 35 conductance meter. TP concentrations (g/L) were determined using the procedures of Murphy and Riley (1962) with a persulfate digestion (Menzel and Corwin 1965). For some samples, total nitrogen was measured using USEPA methods involving sum of nitrate-nitrogen and Kjeldahe-nitrogen values (USEPA 1979). For other samples, TN concentrations (g/L) were determined by oxidizing water samples with persulfate and determining nitrate-nitrogen with second derivative spectroscopy (DElia et al. 1977; Simal et al. 1985; Wollin 1987). Bachmann et al. (1996) found this method to be suitable for samples frozen for 90 days. Chlorophyll a concentrations (g/L) were determined by filtering water through a Gelman type A-E glass fiber filter and using the method of Yentsch and Menzel (1963) with the equations of Parsons and Strickland (1963). Prior to 1994, chlorophyll concentrations (g/L) were determined via 16

PAGE 17

pigment extraction with aqueous acetone followed by measurement of optical density with a spectrophotometer (Method 10200 H; APHA 1989). After 1994, chlorophyll concentrations (g/L) were determined spectrophotometrically (Method 10200 H; APHA 1992) following pigment extraction with ethanol (Sartory and Grobbelaar 1984). Prior to 1993, color (Pt-Co units) was determined by using the platinum-cobalt method and matched Nessler tubes (APHA 1985). After 1993, color was determined by spectroscopy (Bowling et al. 1986). Calcium, magnesium and potassium concentrations (mg/L) were determined by atomic absorption spectrophotometry (APHA 1985). For some samples (Canfield 1981; Canfield and Hoyer 1992; Florida LAKEWATCH 1993,1996; Greis 1995), calcium concentrations were determined from calcium hardness concentrations. Magnesium concentrations were then estimated from the difference between total hardness and calcium hardness concentrations (Hach Chemical Company 1992). Potassium, along with sodium, concentrations were determined by flame photometry (Methods 3500-Na D and 3500-K D, APHA 1989). Total iron concentrations were determined using the Ferrozine method (Method 8147, Hach Chemical Company 1992). Chloride concentrations (mg/L) were measured by titration with 0.0141 N mercuric nitrate and using diphenylcarbazone for determining endpoints (Hach Chemical Company 1975). Sulfate concentrations were determined using a turbidimetric method with SulfaVer sulfate reagent (Method 4500-SO4 E, APHA 1992). Data on surface oxygen saturation were obtained from Florida Lakewatch for 284 lakes within the 627 lake data set (Florida LAKEWATCH unpublished data). Measured oxygen percent saturation values were compared to calculated pCO 2 levels for 284 Florida lakes. Orange Lake, an approximately 5,250-hectare lake located in southeastern Alachua County, Florida was the study site for the macrophyte and lake metabolism portion of this study. 17

PAGE 18

Sampling of Orange Lake was conducted during the months of June, July and August 2005. One 24-hour sampling period was carried out each month. A total of four stations were chosen at Orange Lake (Figure 2-1). These stations included three stations established within aquatic macrophyte beds, and one open-water (no aquatic macrophytes) station. Each month, one water sample was taken at each station in 250-ml, acid-cleaned, triple-rinsed Nalgene bottles at a depth of 0.5 m for nutrient analysis. These samples were subsequently frozen and analyzed at the end of the summer season. This sample was used to determine total phosphorus (g/L) and total nitrogen (g/L) concentrations. Chlorophyll samples were filtered in the field and filters were subsequently frozen. At the end of the sampling term (Summer 2005) chlorophyll concentrations (g/L) were determined spectrophotometrically (Method 10200 H; APHA 1992) following pigment extraction with ethanol (Sartory and Grobbelaar 1984). Measurements were taken at every station every three hours for a 24 hr period, yielding eight sampling sets per day. Specific conductance (S/cm), dissolved oxygen (mg/L), pH, and water temperature (C) were taken every 0.5 m from surface to bottom with a Yellow Springs Instrument Model 560 multi-probe handheld system. A Simeral model DIC anemometer was used to measure wind speed (m/sec) at each sampling period. A 250-ml water sample was taken at 0.5 m depth in a 250 mL acid-washed dark Nalgene bottle at each station every three hrs for laboratory analysis of total alkalinity (mg/L as CaCO 3 ). These samples were kept on ice and analyzed directly after the 24 hr sampling was completed. Total alkalinity (mg/L as CaCO 3 ) was determined by titration with 0.02 N sulfuric acid (Method 2320 B; APHA 1992). All recorded values were converted to meq/L. To assess lake metabolism, the diurnal curve method was used to estimate gross primary production and respiration within the lake (Odum and Hoskin 1958; Bachmann et al. 2000). At 18

PAGE 19

each sampling site, the average water column amount of DO (mg/L) was calculated by multiplying the average oxygen concentration () of the water column by the depth in meters (Z) of the water column (Equation 2-11). Average Water Column DO= (water column DO) Z (2-11) The uncorrected rate of oxygen change (g/m2/hr) was the average water column amount of DO at the end of the sampling interval minus the average water column amount of DO at the beginning of the interval divided by the length of the interval in hours (Equation 2-12). Uncorrected O 2 Change= Integral O 2 End-Integral O 2 Begininng Interval Length (2-12) The gas transfer coefficient (diffusion coefficient DC) across the water surface for station 4 (the open-water station) was calculated from the wind velocity (m/sec) represented by W and water temperature (C), represented by T using the equations of Hartman and Hammonds (1985) (Equation 2-13). DC = 34.6 (1+(T-20)+0.026 *(0.00002060.5) W1.5 (2-13) For the plant stations (Stations 1,2, and 3), the equations of Hartman and Hammonds (1985) were used with an alteration made in wind speed to adjust for the fact that wind had limited interaction with the water surface due to plant cover. For plant stations 1,2, and 3, wind speeds were reduced to 0.1 m/sec. The oxygen saturation deficit (OSD) was found by subtracting the saturation concentration (mg/L) from the surface concentration (Equation 2-14) of dissolved oxygen (mg/L). OSD = Surface DO Saturation O 2 (2-14) This was multiplied by the gas transfer coefficient to find the rate that oxygen was entering or leaving the lake surface during each sampling interval (g/m 2 ) (Equation 2-15). Diffusion = Average (OSD of Interval) DC/24 (2-15) 19

PAGE 20

The corrected oxygen change was the sum of the uncorrected oxygen change and the oxygen flux due to diffusion (Equation 2-16). Corrected Oxygen Change= Uncorrected Oxygen Change + Diffusion (2-16) The sum of the oxygen changes over the 24 hr period was the net oxygen production for the day in units of g O 2 m-2 d-1. To convert to units of g C m-2 d-1, the g O2 m-2 d-1 was multiplied by 0.375 and divided by the photosynthetic quotient (PQ) of 1.2 (Wetzel and Likens, 1991). Respiration was estimated as the average of the corrected oxygen changes during intervals between sunset and sunrise. This was converted to g C m-2 d-1 by multiplying by 0.375 and a respiratory quotient (RQ) of 1.0 (Wetzel and Likens, 1991). Gross production was the sum of net production and respiration. Data were analyzed using JMP 4.0 statistical software (SAS Institute Inc.). All trophic state parameters and ions were logarithmically transformed to attain normal distribution. All variables were correlated with logarithmically transformed values for pCO 2 An value of 0.05 was used for all statistical tests. 20

PAGE 21

1 23 4 Figure 2-1. Satellite photo of Orange Lake, Florida with sampling sites indicated by diamonds. 21

PAGE 22

CHAPTER 3 RESULTS AND DISCUSSION Partial Pressure CO 2 Values from Historical Lake Data A total of 4,730 water samples were collected from 627 lakes located throughout Florida (Figure 3-1). The calculated partial pressure carbon dioxide level (pCO 2 ) values for the lakes exhibited a large range (below -0.0003 to 915,000 atm). The average calculated pCO 2 was 13,000 atm. Twenty-five percent of Florida lakes however, ranged from below zero (meaning they take in carbon dioxide from the atmosphere) to 800 atm with a median value of 400 atm. The next quartile of lakes ranged from 800 atm to 1500 atm with a median value of 1000 atm. Coles mean pCO 2 of 1036 atm was within this range. The third quartile ranged from 1500 atm to 3000 atm with a median value of 2000 atm. The fourth quartile of lakes ranged from 3000 atm to 916, 000 atm with a median value of 460,000 atm. When calculated pCO 2 values are very high, some would suggest there may be an allochthonous carbon supply to the lake through surface water input (Jonosson 2001; Duarte 2005). The geometric mean for the Florida lakes is 2,000 atm, which is nearly identical (1036 atm) as Coles average pCO 2 for 1835 lakes (Table 3-1). The geometric mean is a more appropriate comparison to Coles findings because it better represents the central tendency of the Florida data set. Coles values for pCO 2 ranged from 1 to 20,249 atm. Florida lakes exhibited similar ranges in pH, temperature, specific conductance, and alkalinity to Coles data set (Table 3-1), but only 11% of the Florida lakes fell above Coles maximum of 20, 249 atm. To model the effect of changes in pH, all pCO 2 calculation variables were held at a constant average (determined by mean values for the 627 lakes), while pH was changed in increments of one unit from 3 to 12 (Figure 3-2). Also, pH values for all 627 lakes were raised one unit of pH, and lowered one unit of pH, and pCO 2 was calculated for all lakes with both 22

PAGE 23

changes. When pH was changed by one unit, there were significant differences found between the calculated pCO 2 values and the original pCO 2 values. Partial pressure carbon dioxide values were significantly higher (p < 0.0001) when pH was decreased by one unit (p < 0.0001). To model changes in temperature, the value of 21 C was raised to 31 C and lowered to 11 C and pCO 2 values were subsequently recalculated. No significant difference was found between the three (11 C, 21 C, 31 C) pCO 2 values calculated for the lakes using the alteration in average yearly temperature (p > 0.925). To model changes in specific conductance, all pCO 2 calculation variables were held at a constant average (determined by mean values for the 627 lakes), while specific conductance values were changed from 50 S/cm to 400 S/cm. Partial pressure carbon dioxide values were calculated over a range in pH values from 3 to 12. No significant difference was found between varying values for specific conductance (P>0.925). My calculations indicate 80% of the Florida lakes are supersaturated with carbon dioxide. Supersaturation is this instance is defined as any value exceeding the atmospheric level of carbon dioxide (370 atm) by more than 10%. This is similar to Coles finding that 87% of his study lakes were supersaturated with carbon dioxide. Limnological Variables The lakes, as a group (Table 3-2), had an average pH of 6.7, but individual average lake pH values ranged from 3.9 to 11.7. Total alkalinity (meq/L) ranged from 0 to 467,000 meq/L with an average alkalinity of 500 meq/L. Lakes ranged from oligotrophic (TP < 15g/L, TN < 400 g/L, Chlorophyll a < 3 g/L) to hypereutrophic (TP > 100 g/L, TN > 1500 g/L, chlorophyll a > 40 g/L) (Forsberg et al. 1980) with an average TP of 46 g/L, an average TN of 860 g/L, and an average chlorophyll of 17 g/L. Water clarity (Secchi depth) ranged from < 0.1 m to 7.2 m with an average Secchi depth of 1.7 m. Color (Pt-Co units) ranged from 0 (very clear) to almost 700 (black water) with an average color value of the lakes being 52 Pt-Co units. 23

PAGE 24

There exists a significant relationship (r = 0.91 Figure 3-3) between pH and pCO 2 as expected from the findings of Juday et al. (1935). If you examine Figure 3-3, and Figure 1-2 from Juday and others, both graphs do not show supersaturation carbon dioxide levels in lake waters when the pH of water was >8. Both the findings in Wisconsin lakes and Florida lakes follow the relationship between pH and the relative proportions of inorganic carbon species (Figure 1-2) of CO 2 (+H 2 CO 3 ), HCO 3 -, and CO 3 2in solution (Wetzel 1983). The data suggests that the availability of free CO 2 in an aquatic system is regulated by the pH of the system itself. Using pairwise correlation analysis, significant correlations were found for each of the limnological variables in relation to pCO 2 (Table 3-3). Trophic state indicators (i.e., TP, TN, Chlorophyll ) were negatively, but weakly (TP, r = -0.30; TN, r = -0.35; Chl, r = -0.48) correlated to pCO 2 ( Figures 3-4, 3-5, and 3-6 respectively). With dense populations of phytoplankton present, the surrounding natural water becomes depleted of carbon dioxide due to photosynthesis (Talling 1976). So one would expect a negative relationship between chlorophyll and pCO 2 Secchi disc measurements were positively, but very weakly correlated (r = -0.10). It was surprising that the correlation of pCO 2 with color (r = 0.09) was not stronger (Figure 3-7). Increases in color can often be attributed to the loading of humic compounds from terrestrial inputs (Duarte 2005). These compounds, rich in dissolved organic carbon (DOC) can convert from DOC to CO 2 (Reche et al. 2002). When DOC undergoes bacterial degradation upon entry to the aquatic system, there is a potential to raise aquatic respiration beyond the limits imposed by aquatic photosynthesis (Cole 1999; Scully et al. 2003; Duarte et al. 2005). If DOC from the watershed was a large contributing factor to increases levels of pCO 2 in Florida lakes, there should be a greater correlation between the two variables. Specific conductance was weakly correlated (r = -0.46) with pCO 2 levels (Figure 3-8). 24

PAGE 25

Canfield et al. (1988), found based on a survey of 165 lakes within Florida that the mineral composition of the lakes was strongly related to Floridas geologic and physiographic development. The study (Canfield 1988) was a limnological survey of 165 lakes within Florida. They found, with the exception of iron and color, pH values and chemical concentrations generally increased as one moved from northwest to southeast. If this trend exists for pH values in Florida lakes, then partial pressure carbon dioxide levels should therefore decrease as one moves from northwest to southeast. This pattern does indeed exist. Using the lake regions established in Griffiths et al. (1997), Table 3-4 shows the average pH and pCO 2 values by region. As the regions progress from the northwest to south Florida, the pCO 2 levels show a general decrease as pH increases (Figure 3-9). Effect of Macrophyte Presence on pCO 2 levels in a Florida Lake Samples from Orange Lake were taken in the summer of 2005. Samples had an average total phosphorus of 230 g/L from all three sampling periods (35). Sample TP concentrations ranged from 190 g/L to 650 g/L. Total nitrogen averaged 2306 g/L with a range of 1600 g/L to 6100 g/L. Chlorophyll averaged 60 g/L, with a range of 10 g/L to 100 g/L. Based on the Forsberg and Ryding trophic scale (1980), this lake is hypereutrophic. Total alkalinity (meq/L) averaged 360, with a range of 300 meq/L to 420 meq/L. Measured water temperature averaged 29.6 C with a range of 27.8 C to 31.2 C. Specific conductance averaged 86 S/cm over the three sampling months. Specific conductance ranged from 81 to 92 S/cm. The pH of Orange Lake averaged 5.9 over the 3 monthly sampling periods, with a range of 4.9 to 6.8 (Table 3-6). The calculated partial pressure carbon dioxide (pCO 2 ) averaged 21,000 atm over the 3 sampling periods, with a range of 10,000 to 48,000 atm. Based on these calculations, Orange Lake was a continual source (pCO 2 > 370 atm) of carbon dioxide to the atmosphere over every sampling period and at each station (Table 3-7). 25

PAGE 26

In June, pCO 2 values ranged from 11,000 atm to 17,000 atm (range includes stations 1,2,3 and 4). In July, pCO 2 values ranged from 24,000 atm to 48,000 atm and in August, pCO 2 values ranged from 10,000 atm to 18,000 atm. In the month of July, large areas of Orange Lake were subject to a mechanical aquatic plant maintenance technique known as cookie cutting. This process cuts large segments of floating and rooted vegetation. It causes water agitation and sediment resuspension. This process may be the reason July pCO 2 values were greater than both June and August. The range for Orange lake pCO 2 values (10,000 atm to 48,000 atm) falls within the top 25th percentile of values exhibited in the 627 study. In the 627-lake study, Orange Lake had an average pCO 2 value of 1449 atm. This difference can be explained by the difference in pH values from the historical data collection to the present. In 1987, Orange Lake was sampled in 5 different months at 3 stations (Canfield and Hoyer, 1992). The average pH over that time period was 7.6. Referring to modeled pH and pCO 2 changes in Figure 3-3, a pCO 2 value of about 1000 would be expected for a pH of 7.6. In the data collected on Orange Lake in 2006, the average pH value was 5.9. A pH value of 5.9 corresponds to a pCO 2 level of about 30,000 according to the modeled values in Figure 3-3. The change in pH might be the cause of the difference in pCO 2 values from the historical data and present data. In each month, Stations 1, 2, and 3 (the macrophyte stations) had significantly higher pCO 2 levels than the open water station 4 (Table 3-7, Figure 3-10). This result was anticipated for three possible reasons: (1) With the excessive growth of macrophytes on the shorelines in such shallow depths, I would expect a higher content of organic matter from the breakdown of decaying plant matter by bacteria. This process releases carbon into the water column. 26

PAGE 27

(2) Another factor that may increase shoreline levels of pCO 2 is terrestrial runoff entering the system. This runoff may contain large amounts of humic compounds. These compounds, rich in dissolved organic carbon (DOC) can photomineralize convert from dissolve organic carbon (DOC) to carbon dioxide (CO 2 ) (Reche et al. 2002). (3) The average pH of stations 1,2 and 3 (the plant stations) was lower than station 4 (open water station) over the 3 month period. A plant survey was conducted by Florida LAKEWATCH in July 2005 on Orange Lake. The percent volume infested with aquatic macrophytes (PVI) and the percent area covered by macrophytes (PAC) were determined. Lakewatch found that 58% of the surface area of Orange Lake in July 2005 was covered with aquatic macrophytes (Table 3-8). They also found that 14% of the lakes volume was filled with aquatic vegetation. Finally, LAKEWATCH calculated a floating-leaved zone of 380 m. Floating and emergent plants extended from the shoreline 380 m into the open water area of the lake. The average pCO 2 value for the plant stations (1, 2, and 3) was 24,000 atm, while the average pCO 2 value for the open water station (4) was 17,000 atm. This represents more then half of the lake, having a significantly higher (p= 0.002) pCO 2 level then the rest of the lake. When making estimations using data obtained on Orange Lake in Summer 2005, a change from 0% to 100% of macrophyte coverage in the lake increased average pCO 2 levels 25% (see Table 3-9). With large changes in plant coverage a noticeable change is pCO 2 may be apparent. However, small-scale changes would result in trivial changes in pCO 2 levels. This suggests that aquatic macrophytes as a whole may not be as important as I had thought to pCO 2 levels in the lake. Lake Metabolism in Relationship to pCO 2 Levels in a Florida Lake The calculated net ecosystem production (NEP) for each station by month is shown in Table 3-10. Station 1 values ranged from -1.6 to -3.1 mg C/m2 over the three months with an 27

PAGE 28

average -2.4 mg C/m2. Station 2 values ranged from -2.1 to -3.2 mg C/m2 with an average value of -2.5 mg C/m2. Station 3 values ranged from -5.1 to -6.2 mg C/m2 with an average of -5.6 mg C/m2. Station 4 values ranged from -5.2 to -10.7 mg C/m2 with an average value -8.6 mg C/m2. There was no significant difference in net ecosystem production values between the 3 months (p= 0.666 at = 0.05). There is a significant difference between the 4 stations (p= 0.003 at = 0.05). The open-water had a significantly lower NEP than the macrophyte stations. The data indicate that plants stations are significantly less heterotrophic than the open water stations in Orange Lake. This may be due to the different impact wind has the surface of the water at the various stations. The plant stations were located in the littoral areas where the sampling equipment was in a bed of plants. Lack of wind interaction and a greater potential for higher photosynthesis rates would make these stations less heterotrophic. The data indicate that this lake remains in a heterotrophic state for at least some periods during the summer months. Heterotrophy in Lakes Some scientists contend that supersaturation of carbon dioxide indicates heterotrophy at the ecosystem level (Raymond et al 2000). In contrast, Bachmann et al. (2000) concluded on the basis of light and dark bottle oxygen measurements and diel oxygen measurements that Lake Apopka, a hypereutrophic lake in central Florida, was hetorotrophic (Bachmann et al.2000). The data on Lake Apopka from the 627-lake study indicate that Lake Apopka is not supersaturated with respect to the atmosphere (pCO 2 of 78.3 atm). This is a heterotrophic lake that does not exhibit supersaturation levels of carbon dioxide with respect to the atmosphere. The pH of Lake Apopka is approximately 8, which is in accord with the findings of Juday et al. (1935) and the 627-lake study. In this case, the relationship between pH and the relative proportions of inorganic carbon species is demonstrated in an aquatic system. At a pH of 8, the fraction of dissolved inorganic carbon expressed as free CO 2 is very small (see Figure 1-2). 28

PAGE 29

29 Out of the 627 historical lakes, oxygen data were available for 284 lakes (Florida LAKEWATCH, unpublished data). There was a negative relationship between pCO 2 and percent oxygen saturation (Figure 3-11). This gr oup of lakes can be put in to 4 categories. In Figure 3-11, the 4 quadrants of the graph represen t these categories. In quadrant 1, lakes are supersaturated with carbon dioxide (pCO 2 >370 atm), and undersaturated with oxygen (<100%). Quadrant 2 represents lakes that are supersatur ed with carbon dioxide, an d supersaturated with oxygen (>100%). Quadrant 3 represents lakes th at are below saturation with respect to carbon dioxide and oxygen. Quadrant 4 repr esents lakes that are below sa turation with respect to carbon dioxide, and supersaturated with oxygen. If a de termination of heterotrophy is made on a basis of carbon dioxide supersaturation alone, 25% of th ese lakes could be mistakenly identified as heterotrophic. Twenty-four percent of the 284 lake s fall into quadrant 2. These lakes are supersaturated in both carbon dioxide and oxyge n. Heterotrophy is generally indicated by supersaturation of carbon dioxide. In this case, it is unclear if th ere is more respiration (indicated by CO 2 concentrations) than photos ynthesis (indicated by O 2 ) because both variables are in a state of supersaturation.

PAGE 30

30 Table 3-1. Comparison of various limnological and physical parameters measured in the 627-lake study to Coles 1994 study. Direct pCO2 ParameterMeasurementsLake in autumnFull season cyclesLakes in summerAfrican lakesFlorida LakesLakes (n)371612696059627Samples (n)39016122395179794,730pH (range)4.7-9.53.8-9.44.2-9.84.9-9.36.0-9.93.9-11.7DIC (M) (range)4.9-2,50013.3-4,0776.6-4,80011-3,57843-145,790-2053.7-35,286.6pCO2 (atm) (mean)801103110646802296*2000pCO2 (atm) (range)107-4,12820-9,7891-7,8455-8,99132-20,249-0.0003-9,000Basis of pCO2 estimateDirect measurementpH, DICpH, DIC, pH, ANCpH, DICpH, ANCpH, ANCSampling intensityWeekly to quarterlyEach once inVaried but more than Summer only One to three timesMonthly to quarterlyAutumn10 samples per yearSourceCole 1994Cole 1994Cole 1994Cole 1994Cole 1994This Study*Geometric Mean

PAGE 31

Table 3-2. Mean, standard error and range of the variables measured in 627 lakes. Variable Mean SE Min Max pH 6.7 0.048 3.9 11.7 Total Alkalinity (meq/L) 500 25.15 0 467000 Temp (C) 21 0 21 21 Specific Conductance (S/cm @ 25 C) 200 19.3 11 7200 atmospheric pCO 2 (atm) 370 0 370 370 pCO 2 atm 13000 2300 -0.0003 916000 Total Phosphorus (g/L) 46 4.22 1 1221 Total Nitrogen (g/L) 860 26.1 23 5600 Chlorophyll a (g/L) 17 1.25 0.47 300 Color (Pt-Co units) 52 3.08 0 690 Secchi (m) 1.7 1.29 0 7.3 Cloride (mg/L) 33 5.58 1.7 2200 Silicon (mg/L) 1.1 0.07 0 12.1 Sulfate (mg/L) 15 0.73 0 186 Calcium (mg/L) 11 0.49 0.19 94 Magnesium (mg/L) 8 1.25 0.16 600 Sodium (mg/L) 15 2.7 1 1160 Potassium (mg/L) 3 0.17 0 50 Iron (mg/L) 0.1 0.01 0 2.3 31

PAGE 32

Figure 3-1. Locations of 627 lakes sampled in the state of Florida. 32

PAGE 33

Figure 3-2. Relationship of pH to pCO 2 (atm) values for the 627 lake study. Lake values are represented by open cirlces, with modeled pH changes shown in black. The black line indicates atmospheric pCO 2 level. 33

PAGE 34

Figure 3-3. Relationship of pH to pCO 2 (atm) values for the 627 lake study. Table 3-3. Correlation coefficients for all variables in relation to pCO 2 with N values. Variable Correlation CoefficientNP ValueTotal Phosphorus (g/L)-0.306610Total Nitrogen (g/L)-0.356580Chlorophyll a (g/L)-0.486460.0184Color (Pt-Co units)0.096360Secchi (m)0.194790Conductivity (S/cm @ 25C)-0.466630.0001Cloride (mg/L)-0.356590Silicon (mg/L)-0.115100Sulfate (mg/l)-0.276020Calcium (mg/L)-0.596090Magnesium (mg/L)-0.446090Sodium (mg/L)-0.366120Potassium (mg/L)-0.506100Iron (mg/L)0.294690.0123 34

PAGE 35

Figure 3-4. Relationship of TP (g/L) to pCO 2 (atm) values for the 627 lake study. 35

PAGE 36

Figure 3-5. Relationship of TN (g/L) to pCO 2 (atm) values for the 627 lake study. 36

PAGE 37

Figure 3-6. Relationship of Chlorophyll (g/L) to pCO 2 (atm) values for the 627 lake study. 37

PAGE 38

Figure 3-7. Relationship of color (Pt-Co Units) to pCO 2 (atm) values for the 627 lake study. 38

PAGE 39

Figure 3-8. Relationship of specific conductance (S/cm @ 25C) to pCO 2 (atm) values for the 627 lake study 39

PAGE 40

Table 3-4. Average pCO2 and pH levels for each Florida Region. Region pCO 2 pH 65-01 47000 6.5 65-02 6000 6.2 65-03 11000 5.5 65-04 5000 6.7 65-05 10000 5.7 65-06 5000 6.8 75-01 52000 5.9 75-02 315000 4.8 75-03 4000 6.2 75-04 10000 5.9 75-05 6000 5.8 75-06 6100 7.6 75-07 1000 7.0 75-08 2000 7.7 75-09 30000 5.2 75-10 14000 6.7 75-11 5000 6.8 75-12 2000 7.4 75-13 22000 6.9 75-14 2000 6.6 75-15 3000 7.4 75-16 10000 7.1 75-17 1000 7.2 75-18 3000 7.2 75-19 4000 6.4 75-20 1000 7.0 75-21 3000 7.7 75-22 2000 7.7 75-23 8000 6.5 75-24 2000 7.3 75-25 300 8.1 75-27 5000 6.2 75-28 1000 7.9 75-30 1000 8.5 75-31 3000 7.7 75-32 3000 7.5 75-33 2000 7.0 75-34 4000 6.7 75-35 1000 7.6 75-36 17000 6.9 75-37 300 8.5 76-02 1000 8.2 76-03 8000 7.3 40

PAGE 41

Figure 3-9. Average pH and pCO 2 values by region. Regions move from northwest of Florida to south Florida in a southeast direction. 41

PAGE 42

Figure 3-10. Monthly pCO 2 averages by station for Orange Lake. 42

PAGE 43

Figure 3-11. Partial pressure carbon dioxide values in relationship to % oxygen saturation for 284 historical lakes. The vertical line represents 100% oxygen saturation. The horizontal line represents atmospheric level of pCO 2 43

PAGE 44

Table 3-5. Average of nutrient and chlorophyll concentrations by month and station in Orange Lake, Florida. MonthStationTotal PhosphorusTotal NitrogenChlorphyll(g/L)(g/L)(g/L)6124020005062190180030632902000806422018002071230170030723603300100733002400507421016001081260170060823503000908322018002084240170030 Table 3-6. Average carbon estimate variables and wind by month and station in Orange Lake, Florida. MonthStationTotal AlkalinityTemp Specific Conductance pHWind (meq/L)(C)(S/cm @ 25 C)(m/sec)6139028.2896.156234029.4936.076338027.9886.276442028.3926.287141029.5875.677230029.9825.857337030.7825.7127437030.1815.798135030.2906.078234031.2826.278335030.0846.048437030.0866.36 44

PAGE 45

Table 3-7. Average pCO 2 by month and station in Orange Lake, Florida. Month Station pCO2 (atm)611700062150006313000641100071480007224000734200074310008118000821100083150008410000 Table 3-8. Aquatic plant data collected on July 7, 2005. Source: Florida LAKEWATCH 2005. VariableValue% area covered with aquatic vegetation (PAC %)58% of lake's total volume filled with vegetation (PVI %)13.5Average emergent plant biomass (kg wet wt/m2)3.2Average floating-leaved biomass (kg wet wt/m2)4.5Average submersed plant biomass (kg wet wt/m2)5.6Average width of emergent and floating-leaved zone (m)380Average lake depth (m)3 Table 3-9. Percent coverage of macrophytes in Orange Lake and the estimated pCO 2 for that plant coverage value. % CoverEstimated pCO201800010180002019000302000040200005021000602100070220008023000902300010024000 Table 3-10. Net ecosystem production values for each station by month with an average for all samples at each station in Orange Lake, Florida. Net Ecosystem Production S tation # JuneJulyAugustAverage1-1.6-2.6-3.1-2.42-2.1-2.25-3.2-2.53-5.6-5.1-6.2-5.64-5.2-10.7-9.8-8.6 45

PAGE 46

CHAPTER 4 CONCLUSIONS From a 627-lake study, a large numbers (80% of those sampled) of Florida lakes are potential sources of carbon to the atmosphere. Partial pressure carbon dioxide levels in Orange Lake even at their lowest calculated atm were more than 26 times than that of the overlying atmosphere. It would be difficult to argue that this lake is not a carbon source to the atmosphere. The data set, however, is only representative of the summer months. Full seasonal data may not be necessary to accurately estimate pCO 2 levels based on the temperature modeling in the 627 lake study. Certain meteorological events such as hurricanes may cause significant short term increases to pCO 2 depending on runoff volume and its effect on alkalinity values, and also the amount of sediment resuspension. Trophic state indicators, alkalinity, and specific conductance were all weakly correlated with pCO 2 levels. The 627-lake study (Canfield 1981; Canfield and Hoyer 1992; Florida LAKEWATCH 1993, 1996; Greis 1995) supports the conclusion that levels of partial pressure carbon dioxide in lakes are controlled by pH levels in the lake. Much of the literature to date has focused on the effects of DOC from the watershed and its relationship to supersaturation (Dillon et al. 1997, Frankignoulle et al. 1998). The data from the 627-lake study (based on color data) did not support the higher correlations with dissolved organic carbon (DOC) found in the literature. The role of aquatic macrophyte coverage and its potential effects on carbon dioxide concentrations in surface waters may need further investigation. When making estimations using data obtained on Orange Lake in Summer 2005, a change from 0% to 100% of macrophyte coverage in the lake increased average pCO 2 levels 25% (see Table 3-9). These findings suggest that small-scale changes in aquatic macrophyte coverage would not produce significantly higher 46

PAGE 47

changes in pCO 2 levels in a Florida Lake. These findings may change if lakes of different trophic states are examined. Data indicate that Orange Lake is a supersaturated with carbon dioxide and seems to be a continual source of CO 2 to the atmosphere. Partial pressure carbon dioxide levels were significantly higher in plant zones. However, net ecosystem production values were lower in plant areas than in the open water area. Both areas were heterotrophic (NEP was negative). This could be another indicator that plants themselves may not be a causal factor in increases pCO 2 levels. The location of the plants in the littoral zone and their effect on pH, sediment resuspension and nutrient cycling may be more important factors. Oxygen saturation data indicates that supersaturation of carbon dioxide alone should not be used as measure of heterotropy at an ecosystem scale. There were several Florida lakes that were supersatured with both oxygen and carbon dioxide. The limnological parameter pH is a key parameter that influences many biological and chemical factors including species distribution, multiple equilibrium expressions, contaminant metal ion concentrations, naturally occurring organic and inorganic species and the dissociation of carbonic acid to name a few (Wetzel 2001). Many attempts have been made to quantify and rank variables of significance to predict mean values of lake pH. Several catchment and lake morphometric parameters have been investiged. The data from the 627 lake study followed a geologically regulated trend in pH as found by Canfield et al. 1988. The trend in pH corresponded with an expected trend in pCO 2 values by lake region. This further supports that pH is the regulating factor in indirectly measured pCO 2 values in Florida freshwater lakes. When looking for causal factors to explain variability in pCO 2 levels between aquatic systems, 47

PAGE 48

one should examine the possible influence of regional geology underlying the aquatic systems in question. 48

PAGE 49

WORKS CITED American Public Health Association (APHA). 1985,1989,1992. Standard Methods for the examination of water and wastewater, 16th, 17th, 18th Editions, Washington D.C. Bachmann, R., M. V. Hoyer, and D.E. Canfield, Jr. 2000. Internal heterotrophy following the switch from macrophytes to algae in Lake Apopka, Florida. Hyrdrobiologia 418:217-227. Bachmann, R. and D.E. Canfield Jr. 1996. Use of an alternative method for monitoring total nitrogen concentrations in Florida lakes. Hydrobiologia 323:1-8. Bowling, L, M. Steane, P.Tyler. 1986. Spectral Distribution and Attenuation of Underwater Irradiance in Tasmanian Inland Waters. Freshwater Biology FWBLAB 16:313-335. Brenner, M., M.W. Binford and E.S. Deevey. 1990. Lakes. Pages 363-391 in R.L. Myers and J.J. Ewel (eds.), Ecosystems of Florida. University of Central Florida Press, Orlando. Canfield, D.E. Jr., and R. Bachmann. 1981. Prediction of total phosphorus concentrations, chlorophyll a, and Secchi depths in natural and artificial lakes. Canadian Journal of Fisheries and Aquatic Sciences 38:414-423. Canfield, D.E. Jr., K. Langeland, M.J. Maceina, W.T. Haller, J.V. Shireman, and J.R. Jones. 1983. Trophic state classification of lakes with aquatic macrophytes. Canadian Journal of Fisheries and Aquatic Sciences 40:1713-1718. Canfield, D.E. Jr., and M.V. Hoyer. 1988. Regional geology and the chemical and trophic state characteristics of Florida lakes. Lake and Reservoir Management 4:21-31. Canfield, D.E., Jr., and M.V. Hoyer. 1992. Aquatic macrophytes and their relation to the limnology of Florida lakes. Final Report submitted to the Bureau of Aquatic Plant Management, Florida Department of Natural Resources, Tallahassee, FL. Carignan R, D. Planas, C. Vis. 2000. Planktonic production and respiration in oligotrophic Shield lakes. Limnology Oceanography 45:189. Cole, J.J., M.L. Pace, S.R. Carpenter, and J.F. Kitchell. 2000. Persistence of net heterotrophy in lakes during nutrient addition and food web manipulations. Limnology Oceanography 45, 1718. Cole J.J. 1999. Aquatic microbiology for ecosystem scientists: new and recycled paradigms in ecological microbiology. Ecosystems 2:2152. Cole J.J., and N.F. Caraco. 1998. Atmospheric exchange of carbon dioxide in a low-wind oligotrophic lake measured by the addition of SF6. Limnology and Oceanography 43:647656. Cole, J. J., N. F. Caraco, G. W. Kling, and T. W. Kratz. 1994. Carbon dioxide supersaturation in the surface waters of lakes. Science 265:1568-1570. 49

PAGE 50

del Giorgio PA, and R.H. Peters. 1994. Patterns in planktonic P:R ratios in lakes: Influence of lake trophy and dissolved organic carbon. Limnology Oceanography 39:772. D'Elia C.F., P.A, Steudler, and N. Corwin. 1977. Determination of total nitrogen in aqueous samples using persulfate digestion. Limnology Oceanography 22:760-764. Duarte, C., and Y. Prairie. 2005. Prevalence of Heterotrophy and Atmospheric CO2 Emissions from Aquatic Ecosystems. Ecosystems 8:862-870. Dillon P.J., and L.A. Molot. 1997. Dissolved organic and inorganic carbon mass balances in central Ontario lakes. Biogeochemistry 36:29. Florida LAKEWATCH. 2002. Florida LAKEWATCH Annual Data Summaries for 1986 through 2001. Department of Fisheries and Aquatic Sciences, University of Florida/Institute of Food and Agricultural Sciences. Library, University of Florida. Gainesville, Florida. Forsberg, C., S. Ryding. 1980. Eutrophication Parameters and Trophic State Indices in 30 Swedish Waste-Receiving Lakes. Archiv fur Hydrobiologie 89:189-207. Frankignoulle M., G. Abril, A. Borges, I. Bourge, C. Canon, B. Delille, E. Libert, and J.M. Thate. 1998. Carbon dioxide emisin from European estuaries. Science 282:434. Greis, J. G. 1985. A characterization of 60 Ocala National Forest Lakes. Report by John G. Greis, Hydrologist. National Forests in Florida. Tallahassee, FL. Griffith, G.E., D.E. Canfield, Jr., C.A. Horsburgh, and J.M. Omernik. 1997. Lake Regions of Florida. EPA/R-97/127. U.S. Environmental Protection Agency, National Health and Environmental Effects Research Laboratory, Corvallis, OR. 89p. Hach Chemical Company. 1975, 1992. Procedures, chemical lists and glassware for water and wastewater analysis. HACH Chem. Co., Ames, Iowa. Hanson, P., R. Stephen, D. E. Carpenter, E. Armstrong, H. Stanley, and T. K. Kratz. 2006. Lake dissolved inorganic carbon and dissolved oxygen: changing drivers from days to decades. Ecological Monographs 76:343-363. Hanson, P.C., Bade, D. L., Carpenter, S. R., and T. K. Kratz. 2003. Lake metabolism: Relationships with dissolved organic carbon and phosphorus. Limnol. Oceanogr. 48: 1112-1119. Harrison WG, J. Arstegui, E.J.H. Head, W.K.W. Li, A.R. Longhurst, and D.D. Sameoto. 2001. Basin-scale variability in plankton biomass and community metabolism in the subtropical North Atlantic Ocean. Deep-Sea Res II 48:2241. Hartman, B. and D. E. Hammond. 1985. Gas exchange in San Francisco Bay. Hydrobiologia 129:59-68. 50

PAGE 51

Hem, J.D. 1992. Study and interpretation of chemical characteristics of natural water: U.S. Geological Survey Water-Supply Paper 2254, 263 p. Hesslein R.H., J.W.M. Rudd, C. Kelly, P. Ramlal, and K.A. Hallard. 1991. Carbon dioxide pressure in surface waters of Canadian lakes. In: Wilhelms SC & Gulliver JS (Eds) Air-Water Mass Transfer, Second International Symposium on Gas Transfer at Water Surfaces (pp 413). Am. Soc. Civil Eng., New York, NY. Howarth, R. W., R. Schneider, and D. Swaney. 1996. Metabolism and organic carbon fluxes in the tidal, freshwater Hudson River. Estuaries 19: 848-865. Jonsson, A., M. Meili, A. Bergstram, and M. Jansson. 2001. Whole-lake mineralization of allochthonous and autochthonous organic carbon in a large humic lake. Limnology Oceanography 46: 1691-1700. Juday, C., E. Birge, and W. W. Meloche. 1935. The carbon dioxide and hydrogen ion content of the lake waters of northeastern Wisconsin. Wisconsin Academy of Sciences, Arts, and Letters. Kling G.W., G.W. Kipphut, and M.C. Miller. 1991. Arctic lakes and streams as gas conduits to the atmosphere: implications for tundra carbon budgets. Science 251:298. Menzel, D. W. and N. Corwin. 1965. The measurement of total phosphorus in seawater based on the liberation of organically bound fractions by persulfate oxidation. Limnology and Oceanography 10: 280-282. Odum, H. T. and C. M. Hoskin. 1958. Comparative studies on the metabolism of marine waters. Institute of Marine Science, University of Texas 5:16-46. Parsons, T.T. and J.D.H. Strickland. 1963. Discussion of spectrophotometric determination of marine-plant pigments, with revised equations for ascertaining chlorophylls and carotenoids. Journal of Marine Research 21: 155-163. Prairie Y.T., D.F. Bird, J.J. Cole. 2002. The summer metabolic balance in the epilimnion of southeastern Quebec lakes. Limnology and Oceanography 47:316. Raymond, P., J. Bauer, and J. Cole. 2000. Atmospheric CO 2 Evasion, Dissolved Inorganic Carbon Production, and Net Heterotrophy in the York River Estuary. Limnology and Oceanography 45:1707-1717. Reche, I., and M.L. Pace. 2002. Linking dynamics of dissolved organic carbon in a forested lake with environmental factors. Biogeochemistry 61:21-36. Sartory, D. P., and J. U. Grobbelaar. 1984. Extraction of chlorophyll a from freshwater phytoplankton for spectrophotometric analysis. Hydrobiologia 114: 177-187. Sawyer, C.N. and P.L. McCarty. 1967. Chemistry for sanitary engineers, 2nd ed.: McGraw-Hill, St. Louis, MO. 51

PAGE 52

Scully N.M., W.J. Cooper, and L.J. Tranvik. 2003. Photochemical effects on microbial activity in natural waters: the interaction of reactive oxygen species and dissolved organic matter. FEMS Microbial Ecology 46:353. Simal, J. 1985. Second derivative ultraviolet spectroscopy and sulfumic acid method for determination of nitrates in water. Journal of Analytical Chemistry 68:962-964. Sobek S, G. Algesten, A.K. Bergstrn, M. Jansson, and L.J. Tranvik. 2003. The catchment and climate regulation of pCO2 in boreal lakes. Global Change Biology 9:630. StatView, Version 5. SAS Institute Inc., Cary, NC. Striegl R.G., P. Kortelainen, and J.P. Chanton. 2001. Carbon dioxide partial pressure and 13C content of north temperate and boreal lakes at spring ice melt. Limnology and Oceanography 46:941. Stumm, W. and J.J. Morgan. 1981. Aquatic chemistry, 2nd ed.: New York, John Wiley & Sons, 780 p. Stumm, W., and J.J. Morgan. 1995. Aquatic Chemistry: Chemical Equilibria and Rates in Natural Waters. Wiley. NY. Talling J. F. 1976. The depletion of carbon dioxide from lake water by phytoplankton. Journal of Ecology 64:79-121. USEPA. 1979. Methods for chemical analysis of waters and wastes. U. S. Environmental Protection Agency. Washington D. C. Wetzel R.G. 2001. Limnology. Lake and River Ecosystems. Academic Press, San Diego. Wetzel, R.G. and G. E. Likens. 1991. Limnological analyses. Springer-Verlag, New York: 391 pp. Wollin, K. M. 1987. Nitrate determination in surface waters as an example of the application of UV derivative spectrometry to environmental analysis. Acta Hydrochemica Hydrobiologia 15:459-469 (Ger.). Yentsch, C. S. and D. W. Menzel. 1963. A method for the determination of phytoplankton chlorophyll and phaeophytin by fluorescence. Deep Sea Res. 10: 221. 52

PAGE 53

BIOGRAPHICAL SKETCH Jenney Kellogg was born in Bath, NY in 1981. She grew up in the Finger Lakes area of NY surrounded by vineyards and wineries. She has always loved the outdoors. She attended Hammondsport Central High School and graduated in 2000 ranked 4th in her class. She attended Niagara University from 2000-2004 and received a Bachelor of Science degree with Honors in Biology. Her undergraduate thesis work dealt with oxygen depletion in the central basin of Lake Erie. 53


Permanent Link: http://ufdc.ufl.edu/UFE0020780/00001

Material Information

Title: Factors Influencing Partial Pressure Carbon Dioxide Levels in Florida Lakes: pH, Aquatic Macrophytes, and Geology
Physical Description: Mixed Material
Copyright Date: 2008

Record Information

Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
System ID: UFE0020780:00001

Permanent Link: http://ufdc.ufl.edu/UFE0020780/00001

Material Information

Title: Factors Influencing Partial Pressure Carbon Dioxide Levels in Florida Lakes: pH, Aquatic Macrophytes, and Geology
Physical Description: Mixed Material
Copyright Date: 2008

Record Information

Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
System ID: UFE0020780:00001


This item has the following downloads:


Full Text





FACTORS INFLUENCING PARTIAL PRESSURE CARBON DIOXIDE LEVELS IN
FLORIDA LAKES: pH, AQUATIC MACROPHYTES, AND GEOLOGY





















By

JENNEY L. KELLOGG


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

UNIVERSITY OF FLORIDA

2007

































O 2007 Jenney L. Kellogg









ACKNOWLEDGMENTS

Gratitude is expressed to the many people who helped me carry out this project.

Special thanks are given to Dr. Carlos Duarte and Dr. Yves Prairie who helped to initiate

my project and begin data analysis. Thanks to Dr. Roger Bachmann who helped me with

my proj ect in ways too numerous to list. The staff of Florida LAKEWATCH provided

guidance on sampling protocol and equipment and helped with the chemical analysis of my

samples. Thanks go to Francesco Lazzarino for his help with data entry and analysis and

expertise on programming languages. Gratitude is expressed to Christy Horsburgh for her

continued support and advice. Thanks are given to Dr. Charles E. Cichra for encouraging

my professional growth. Thanks are given to Mark Hoyer for his support and guidance on

this proj ect. Gratitude is expressed to my committee members for helping to oversee my

project. Finally, thanks are given to Dr. Daniel E. Canfield, Jr., my mentor, advisor and

chair for his patience, ingenuity and guidance throughout this project.












TABLE OF CONTENTS


page

ACKNOWLEDGMENT S .........__.. ..... .__. ...............3....


LIST OF TABLES ........._.___..... .__. ...............5....


LI ST OF FIGURE S .............. ...............6.....


AB S TRAC T ......_ ................. ............_........7


CHAPTER


1 INTRODUCTION ................. ...............9.......... ......


2 MATERIALS AND METHODS .............. ...............14....


3 RE SULT S AND DI SCU SSION ............... ...............2


Partial Pressure CO2 Values from Historical Lake Data .............. ...............22....
Limnological Variables .............. ..... ... .. ......... ...............2
Effect of Macrophyte Presence on pCO2 Levels in a Florida Lake ................. ................ ..25
Lake Metabolism in Relationship to pCO2 Levels in a Florida Lake ................. ................ .27
Heterotrophy in Lakes .............. ...............28....

4 CONCLUSIONS .............. ...............46....


WORKS CITED .............. ...............49....


BIOGRAPHICAL SKETCH .............. ...............53....










LIST OF TABLES


Table page

3-1 Comparison of various limnological and physical parameters measured in the 627-
lake study to Cole's 1994 study. .............. ...............30....

3-2 Mean, standard error and range of the variables measured in 627 lakes. ................... .......3 1

3-3 Correlation coefficients for all variables in relation to pCO2, with N values. .................. .34

3-4 Average pCO2 and pH levels for each Florida Region. ............. .....................4

3-5 Average of nutrient and chlorophyll concentrations by month and station in Orange
Lake, Florida. .............. ...............44....

3-6 Average carbon estimate variables and wind by month and station in Orange Lake,
Florida. .............. ...............44....

3-7 Average pCO2 by month and station in Orange Lake, Florida ................. ................ ...45

3-8 Aquatic plant data collected on July 7, 2005. Source: Florida LAKEWATCH 2005.....45

3-9 Percent coverage of macrophytes in Orange Lake and the estimated pCO2 for that
plant coverage value. ............. ...............45.....

3-10 Net ecosystem production values for each station by month with an average for all
samples at each station in Orange Lake, Florida. ............. ...............45.....










LIST OF FIGURES


Figure page

1-1 The relationship of free carbon dioxide and pH taken from Juday et al. (1935). The
dotted line horizontal line indicates atmospheric level of carbon dioxide. Free CO2 is
represented by the dashed line "Calculated CO2."............ ...............12..

1-2 Relation between pH and the relative proportions of inorganic carbon species
modified from Wetzel 2001 ................ ...............13........... ...

2-1 Satellite photo of Orange Lake, Florida with sampling sites indicated by diamonds........21

3-1 Locations of 627 lakes sampled in the state of Florida ......... ................. ...............32

3-2 Relationship of pH to pCO2 (Clatm) values for the 627 lake study. Lake values are
represented by open cirlces, with modeled pH changes shown in black. The black
line indicates atmospheric pCO2 level .......... ................ ...............33.....

3-3 Relationship of pH to pCO2 (Clatm) values for the 627 lake study ................ ................34

3-4 Relationship of TP (Clg/L) to pCO2 (Clatm) values for the 627 lake study ................... .....3 5

3-5 Relationship of TN (Clg/L) to pCO2 (Clatm) values for the 627 lake study. ................... ...36

3-6 Relationship of Chlorophyll (Clg/L) to pCO2 (Clatm) values for the 627 lake study.........37

3-7 Relationship of color (Pt-Co Units) to pCO2 (Clatm) values for the 627 lake study.........38

3-8 Relationship of specific conductance (CIS/cm @ 250C) to pCO2 (Clatm) values for
the 627 lake study .............. ...............39....

3-9 Average pH and pCO2 ValUeS by region. Regions move from northwest of Florida to
south Florida in a southeast direction. ............. ...............41.....

3-10 Monthly pCO2 averages by station for Orange Lake ................. ................ ........ .42

3-11 Partial pressure carbon dioxide values in relationship to % oxygen saturation for 284
historical lakes. The vertical line represents 100% oxygen saturation. The
horizontal line represents atmospheric level of pCO2................ ...............43









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

FACTORS INFLUENCING PARTIAL PRESSURE CARBON DIOXIDE LEVELS IN
FLORIDA LAKES: pH, AQUATIC MACROPHYTES, AND GEOLOGY.

By

Jenney L. Kellogg

May 2007

Chair: Daniel E. Canfield Jr.
Major: Fisheries and Aquatic Sciences

Physical, chemical, and biological factors associated with the supersaturation of CO2 in

aquatic ecosystems are not clearly evident. The goal of our study was to evaluate partial

pressure carbon dioxide (pCO2) lCVOIS in a set of Florida Lakes and to investigate if any

limnological factors are correlated with levels of pCO2. The pCO2 leVOIS were calculated for 627

Florida Lakes using pH, specific conductance, alkalinity and water temperature. Eighty percent

of lakes were found to be supersaturated with carbon dioxide. Trophic state indicators such as

total phosphorus (TP), total nitrogen (TN) and chlorophyll (Chl) were all weakly correlated with

pCO2 COncentrations (TP, r = -0.30 TN, r = -0.35 Chl r = -0.48). The limnological parameter pH

had a highly significant positive correlation with pCO2 COncentration (r = 0.91). When modeling

changes in pCO2 COncentrations, pH seems to be the regulating factor in pCO2 leVOIS in Florida

lakes.

Plants remove carbon dioxide from the atmosphere by carbon assimilation, and produce

carbon dioxide during cellular respiration processes. Photosynthesis and respiration directly

influence the pH of the water. The effect of aquatic macrophyte presence on pCO2 leVOIS was

examined using a summer study (June-August 2005) that examined pCO2 leVOIS in both the plant

and open water areas in a hypereutrophic Florida lake. The presence of plants in a










hypereutrophic Florida Lake significantly increased pCO2 meaSurements. The pH was found to

be lower in plant stations in comparison to open water stations. The same study was used to

evaluate how the presence of plants affected lake metabolism using the diurnal curve method.

Presence of plants significantly decreased the net ecosystem production estimates based on

dissolved oxygen and wind data. All stations on the study lake (both plant and open water) were

heterotrophic over the duration of the study.

Saturation of carbon dioxide in an aquatic ecosystem is often used an in indicator of

heterotrophy at the ecosystem scale. To investigate this statement, dissolved oxygen data on a

group of 284 Florida Lakes were used to examine the relationship between pCO2 and dissolved

oxygen saturations. Several lakes exhibited both supersaturation of oxygen and carbon dioxide

simultaneously. Therefore, dissolved oxygen was determined to be an inadequate predictor of

lake heterotrophy in 25% of 284 Florida lakes.

The limnological parameter pH is often called the master variable in determining the

aqueous geochemistry of an aquatic system. Many attempts have been made to quantify and

rank variables of significance to predict mean values of lake pH. The pH value of a water body

is directly influenced by geologic factors. When investigating the relationship between pCO2

levels and pH across the state of Florida, a geologically regulated trend was found. As pH values

increased in a southeastern direction across the state, pCO2 leVOIS decreased. Changes in pH

values between regions were reflected in changes in pCO2 leVOIS between the same regions.

Changes in regional geology may explain variation between lakes in reference to pCO2 leVOIS.









CHAPTER 1
INTTRODUCTION

The limnological parameter pH is often considered a master variable in determining the

biological, chemical and physical processes of an aquatic system. Juday et al. (1935) examined

relationships between pH levels and carbon dioxide levels in several lakes in Wisconsin. Juday

et al. (1935) calculated pH values from free carbon dioxide values, and compared these

calculated values to directly measured values. They found no significant difference (Figure 1-1).

Juday et al. (1935) also found that lakes did not exhibit carbon dioxide levels exceeding the

atmospheric level of carbon dioxide above a pH of 8. The relationship between pH and the

relative proportions of inorganic carbon species (Figure 1-2) of CO2 (+H2CO3), HCO3-, and CO3

in solution is clearly recognized (Wetzel 1983). At a pH of above 8, the proportion of carbon in

the form of free carbon dioxide is very small (Figure 1-2).

In the past decade, many researchers have supported the idea that organic matter located

within lake basins can either be mineralized or lost as CO2 to the atmosphere (Cole et al. 1994,

2000; Richey et al. 2002; Sobek et al. 2003; Striegl et al. 2001). When the level of carbon

dioxide within a water body exceeds the amount of carbon dioxide present in the atmosphere

(supersaturation), the water body becomes a carbon source (Kling et al. 1991; Hesslein et al.

1991; Cole et al. 1994, 1998).

Cole et al. (1994) conducted a study of carbon dioxide levels in 1835 globally distributed

lakes. Cole used both direct and indirect measurements of CO2 and found that 87% of lakes

were supersaturated with CO2. COle found a mean partial CO2 preSsure for all lakes of 1036

micro atmospheres. The atmospheric level of carbon dioxide is about 370 micro atmospheres.

Physical and chemical and biological factors associated with the supersaturation of CO2 in

aquatic ecosystems are not clearly evident (Cole 1994; Kling 1991). There has been much









investigation into the role of allochthonous carbon and its effect on lake metabolism (Prairie et

al. 2002). The role of the lake as a CO2 prOducer may vary with trophic state and other

limnological factors affecting CO2, including chemical composition, lake morphometry, and

surrounding terrestrial ecosystems (Hanson 2003). It also has not yet been investigated whether

the presence of aquatic macrophytes would have an effect on pCO2 leVOIS in a freshwater lake.

Aquatic macrophytes are prevalent in many Florida lakes and play and important role in

the structure and function of lakes (Brenner et al. 1991). Macrophytes are a major component of

total biomass and primary productivity (Canfield et al. 1983). Aquatic macrophytes are an

integral part of lake metabolism. In aquatic ecosystems, metabolism directly affects

concentrations of dissolved inorganic carbon (DIC) and dissolved oxygen (DO) concentrations

(Hanson 2006). Aquatic macrophytes produce carbon dioxide during cellular respiration

processes (C6H12O6 + 6 Oz 6 COz + 6 HzO + energy) and also excrete dissolved organic

compounds which are acted on by microorganisms and are oxidized into carbon dioxide and

returned to the atmosphere (Wetzel 1969).

Understanding the carbon balance of a lake helps us to understand both the biological

activity occurring within a lake and how the surrounding watershed may be affecting the lake.

Whole-ecosystem metabolism integrates the net effect of all primary producers and all

consumers on the carbon cycle (Hanson 2006). Ecosystem production may be negative or

positive (Howarth et al. 1996). When ecosystem production is positive, the system is considered

autotrophic, and the water body can store organic carbon (Schindler et al. 1972). When

ecosystem production is negative, the system is considered heterotrophic, and respires more

carbon than what was produced within the system (Cole 2000). There is much debate regarding









the metabolic balance between heterotorphy and autotrophy in lakes in general (Carignan et al.

2000; del Giorgio and Peters 1994).

Given the complexities of dissolved carbon dioxide dynamics in freshwater lakes, an

investigation of the physical, chemical and biological factors associated with changes in partial

pressure carbon dioxide levels in Florida lakes is needed. My obj ectives for this study are:

* Objective 1: To evaluate partial pressure carbon dioxide (pCO2) lCVOIS in a set of Florida
Lakes;

* Objective 2: To investigate if any limnological factors are correlated with levels of pCO2
in a set of Florida Lakes;

* Objective 3: To investigate the effect of aquatic macrophyte presence on pCO2 1OVOIS in a
Florida Lake;

* Objective 4: Evaluate lake metabolism and its relationship to pCO2 1OVOIS in a Florida
Lake;











































Figure 1-1. Relationship of free carbon dioxide and pH taken from Juday et al. (1935). Dotted
horizontal line indicates atmospheric level of carbon dioxide. Dashed line represents
"Calculated CO2-"






















4 I 1 1
Pp







Prss SanDigo









CHAPTER 2
MATERIALS AND METHODS

Published water chemistry information was obtained for 627 lakes that were sampled

between one and 48 times over a 17-year period of record (Canfield 1981; Canfield and Hoyer

1992; Florida LAKEWATCH 1993, 1996; Greis 1995). Seventy-Hyve percent of the lakes were

sampled four or less times, with the remaining 25% being sampled between Hyve and 48 times.

One hundred and fifty-nine lakes were sampled 10 or more times.

Carbon dioxide concentrations were computed using averaged pH, total alkalinity, and

specific conductance values from the data set for each lake using the approach of Cole et al.

(1994) and Prairie et al. (2002). Lakes with multiple samples per lake (more than one station)

were averaged, followed by an average of the months sampled. A yearly average was taken from

the averaged months. Finally all years for each lake were averaged to attain a grand average for

each lake. Ionic strength (I) was calculated (Equation 2-1) from specific conductivity (CIS/cm @

25 C) values represented by "SC" (Cole et al. 1994; Prairie et al. 2002) as follows:

I = 0.000025 7 SC (2-1)

This equation is based on a relation between ionic strength and total dissolved solids, taken from

Sawyer and McCarty (1967), which accounts for the first constant (0.000025), and an empirical

relation between total dissolved solids and specific conductance, taken from Hem (1992), which

accounts for the second constant (7).

Equilibrium disassociation constants of carbonic acid to hydrogen atoms and bicarbonates,

and the disassociations of carbonates to bicarbonates were calculated (K1 and K2) where

concentrations are indicated by square brackets [ ], activities are denoted by braces { }. These

calculations are shown below in Equations 2-2 and 2-3 (Stumm and Morgan 1995).










K1=CHIF [HCO, J C (2-2)
[H2CO3*


K2=~' c,~
[HCO3 ] (2-3)

These values were then used to calculate the fraction of free CO2 ta0), bicarbonate (al), and

carbonate (a2) (Equations 2-4, 2-5, 2-6 respectively).

ao= 1
1+10(Kl+pH) + 10(-(Kl+K2)+2*pH) (2-4)

al= 1
1+10(-pH+Kl)+ 10(-K2+pH) (2-5)

a2-
10(-2*pH+(Kl+K2))+10O(-pH+K2)+1 (2-6)


These three values (ao, al, and a2) together are the total amount of inorganic carbon compounds

in the system (Stumm and Morgan 1995). Total carbon in the system (DIC, CIM) is expressed as

a concentration and is derived using the three carbon components (ao, al, and a2) alOng with

alkalinity (meq/L) in Equation 2-7 below.

DIC (CIM)= 105 *(Alkalinitvymmeq/L *0.000001-10(-1 4+pH)+10O(-pH))
(a1+2*a2) (2-7)


Using Henry's Law, the solubility of carbon dioxide in water (KH) is determined which varies

with temperature (Equation 2-8).

KH=1.11+0.016 Temperature(C) 0.00007 Temperature(C)2 (2-8)

The concentration of free CO2 in the system is expressed in micromoles (Equation 2-9):

CO2 (IM) = DIC(CIM) ao (2-9)

Finally, CO2 is expressed as a partial pressure in microatmospheres (Equation 2-10):










p7CO2 (patm)= COg (2-10)
10(-KH
An average yearly temperature value of 21 C was used for solubility calculations for the

627 lake study due to lack of available water temperature data for the individual lakes.

All water samples were collected during daylight hours from the surface (0.5 m) at

multiple (1-4) open-water sites within each lake. Water clarity was measured by use of a Secchi

disc. Water samples were collected in acid-cleaned, triple rinsed Nalgene bottles, and placed on

ice. In the early studies, unfrozen water samples were analyzed for total nitrogen (TN) and total

phosphorus (TP). In later studies, portions of the collected water were subsequently frozen and

used for later analysis of TN and TP. At the laboratory, a pH meter (an Orion Model 601A pH

meter or an Accumet model 10 pH meter) calibrated with pH 4.0 and pH 7.0 buffers was used to

measure pH. Total alkalinity (mg/L as CaCO3) WAS determined by titration with 0.02 N sulfuric

acid (APHA 1085). Specific conductance (CIS/cm @ 25 C) was measured using a Yellow

Springs Instrument Model 21 conductivity bridge or a Yellow Spring Instrument Model 35

conductance meter. TP concentrations (Clg/L) were determined using the procedures of Murphy

and Riley (1962) with a persulfate digestion (Menzel and Corwin 1965). For some samples, total

nitrogen was measured using USEPA methods involving sum of nitrate-nitrogen and Kj eldahe-

nitrogen values (USEPA 1979). For other samples, TN concentrations (Clg/L) were determined

by oxidizing water samples with persulfate and determining nitrate-nitrogen with second

derivative spectroscopy (D'Elia et al. 1977; Simal et al. 1985; Wollin 1987). Bachmann et al.

(1996) found this method to be suitable for samples frozen for 90 days. Chlorophyll a

concentrations (Clg/L) were determined by filtering water through a Gelman type A-E glass fiber

filter and using the method of Yentsch and Menzel (1963) with the equations of Parsons and

Strickland (1963). Prior to 1994, chlorophyll concentrations (Clg/L) were determined via









pigment extraction with aqueous acetone followed by measurement of optical density with a

spectrophotometer (Method 10200 H; APHA 1989). After 1994, chlorophyll concentrations

(Clg/L) were determined spectrophotometrically (Method 10200 H; APHA 1992) following

pigment extraction with ethanol (Sartory and Grobbelaar 1984). Prior to 1993, color (Pt-Co

units) was determined by using the platinum-cobalt method and matched Nessler tubes (APHA

1985). After 1993, color was determined by spectroscopy (Bowling et al. 1986). Calcium,

magnesium and potassium concentrations (mg/L) were determined by atomic absorption

spectrophotometry (APHA 1985). For some samples (Canfield 1981; Canfield and Hoyer 1992;

Florida LAKEWATCH 1993,1996; Greis 1995), calcium concentrations were determined from

calcium hardness concentrations. Magnesium concentrations were then estimated from the

difference between total hardness and calcium hardness concentrations (Hach Chemical

Company 1992). Potassium, along with sodium, concentrations were determined by flame

photometry (Methods 3500-Na D and 3500-K D, APHA 1989). Total iron concentrations were

determined using the Ferrozine method (Method 8147, Hach Chemical Company 1992).

Chloride concentrations (mg/L) were measured by titration with 0.0141 N mercuric nitrate and

using diphenylcarbazone for determining endpoints (Hach Chemical Company 1975). Sulfate

concentrations were determined using a turbidimetric method with SulfaVer sulfate reagent

(Method 4500-SO4 E, APHA 1992).

Data on surface oxygen saturation were obtained from Florida Lakewatch for 284 lakes

within the 627 lake data set (Florida LAKEWATCH unpublished data). Measured oxygen

percent saturation values were compared to calculated pCO2 leVOIS for 284 Florida lakes.

Orange Lake, an approximately 5,250-hectare lake located in southeastern Alachua

County, Florida was the study site for the macrophyte and lake metabolism portion of this study.









Sampling of Orange Lake was conducted during the months of June, July and August 2005. One

24-hour sampling period was carried out each month. A total of four stations were chosen at

Orange Lake (Figure 2-1). These stations included three stations established within aquatic

macrophyte beds, and one open-water (no aquatic macrophytes) station. Each month, one water

sample was taken at each station in 250-ml, acid-cleaned, triple-rinsed Nalgene bottles at a depth

of 0.5 m for nutrient analysis. These samples were subsequently frozen and analyzed at the end

of the summer season. This sample was used to determine total phosphorus (Clg/L) and total

nitrogen (Clg/L) concentrations. Chlorophyll samples were filtered in the field and filters were

subsequently frozen. At the end of the sampling term (Summer 2005) chlorophyll concentrations

(Clg/L) were determined spectrophotometrically (Method 10200 H; APHA 1992) following

pigment extraction with ethanol (Sartory and Grobbelaar 1984).

Measurements were taken at every station every three hours for a 24 hr period, yielding

eight sampling "sets" per day. Specific conductance (CIS/cm), dissolved oxygen (mg/L), pH, and

water temperature (C) were taken every 0.5 m from surface to bottom with a Yellow Springs

Instrument Model 560 multi-probe handheld system. A Simeral model DIC anemometer was

used to measure wind speed (m/sec) at each sampling period. A 250-ml water sample was taken

at 0.5 m depth in a 250 mL acid-washed dark Nalgene bottle at each station every three hrs for

laboratory analysis of total alkalinity (mg/L as CaCO3). These samples were kept on ice and

analyzed directly after the 24 hr sampling was completed. Total alkalinity (mg/L as CaCO3) WAS

determined by titration with 0.02 N sulfuric acid (Method 2320 B; APHA 1992). All recorded

values were converted to meq/L.

To assess lake metabolism, the diurnal curve method was used to estimate gross primary

production and respiration within the lake (Odum and Hoskin 1958; Bachmann et al. 2000). At









each sampling site, the average water column amount of DO (mg/L) was calculated by

multiplying the average oxygen concentration (X) of the water column by the depth in meters (Z)

of the water column (Equation 2-1 1).

Average Water Column DO= X (water column DO) Z (2-11)

The uncorrected rate of oxygen change (g/m2/hr) was the average water column amount of

DO at the end of the sampling interval minus the average water column amount of DO at the

beginning of the interval divided by the length of the interval in hours (Equation 2-12).

Uncorrected 02 Change= Intenral Oz End-Intenral Oz Benininnn
Interval Length (2-12)

The gas transfer coefficient (diffusion coefficient DC) across the water surface for station 4 (the

open-water station) was calculated from the wind velocity (m/sec) represented by "W" and water

temperature (C), represented by "T" using the equations of Hartman and Hammonds (1985)

(Equation 2-13).

DC = 34.6 (1+(T-20)+0.026 *(0.00002060.5) W1. (2-13)

For the plant stations (Stations 1,2, and 3), the equations of Hartman and Hammonds

(1985) were used with an alteration made in wind speed to adjust for the fact that wind had

limited interaction with the water surface due to plant cover. For plant stations 1,2, and 3, wind

speeds were reduced to 0.1 m/sec.

The oxygen saturation deficit (OSD) was found by subtracting the saturation concentration

(mg/L) from the surface concentration (Equation 2-14) of dissolved oxygen (mg/L).

OSD = Surface DO Saturation Ol (2-14)

This was multiplied by the gas transfer coefficient to find the rate that oxygen was entering

or leaving the lake surface during each sampling interval (g/m2) (Equation 2-15).

Diffusion = Average (OSD of Interval) DC/24 (2-15)









The corrected oxygen change was the sum of the uncorrected oxygen change and the

oxygen flux due to diffusion (Equation 2-16).

Corrected Oxygen Change= E Uncorrected Oxygen Change + Diffusion (2-16)

The sum of the oxygen changes over the 24 hr period was the net oxygen production for

the day in units of g 02 m-2 d- To convert to units of g C m-2 d- the g 02 m-2 d-l was

multiplied by 0.375 and divided by the photosynthetic quotient (PQ) of 1.2 (Wetzel and Likens,

1991).

Respiration was estimated as the average of the corrected oxygen changes during intervals

between sunset and sunrise. This was converted to g C m-2 d-l by multiplying by 0.375 and a

respiratory quotient (RQ) of 1.0 (Wetzel and Likens, 1991). Gross production was the sum of

net production and respiration.

Data were analyzed using JMP 4.0 statistical software (SAS Institute Inc.). All trophic

state parameters and ions were logarithmically transformed to attain normal distribution. All

variables were correlated with logarithmically transformed values for pCO2. An a value of 0.05

was used for all statistical tests.




































Figure 2-1. Satellite photo of Orange Lake, Florida with sampling sites indicated by diamonds.









CHAPTER 3
RESULTS AND DISCUSSION

Partial Pressure CO2 Values from Historical Lake Data

A total of 4,730 water samples were collected from 627 lakes located throughout Florida

(Figure 3-1). The calculated partial pressure carbon dioxide level (pCO2) ValUeS for the lakes

exhibited a large range (below -0.0003 to 915,000 Cpatm). The average calculated pCO2 WAS

13,000 Cpatm. Twenty-five percent of Florida lakes however, ranged from below zero (meaning

they take in carbon dioxide from the atmosphere) to 800 Clatm with a median value of 400 Clatm.

The next quartile of lakes ranged from 800 Clatm to 1500 Clatm with a median value of 1000

Clatm. Cole's mean pCO2 Of 1036 Clatm was within this range. The third quartile ranged from

1500 Clatm to 3000 Clatm with a median value of 2000 Clatm. The fourth quartile of lakes ranged

from 3000 Clatm to 916, 000 Clatm with a median value of 460,000 Clatm.

When calculated pCO2 ValUeS are very high, some would suggest there may be an

allochthonous carbon supply to the lake through surface water input (Jonosson 2001; Duarte

2005). The geometric mean for the Florida lakes is 2,000 Cpatm, which is nearly identical (1036

Cpatm) as Cole's average pCO2 for 1835 lakes (Table 3-1). The geometric mean is a more

appropriate comparison to Cole' s findings because it better represents the central tendency of the

Florida data set. Cole's values for pCOz ranged from 1 to 20,249 Clatm. Florida lakes exhibited

similar ranges in pH, temperature, specific conductance, and alkalinity to Cole's data set (Table

3-1), but only 1 1% of the Florida lakes fell above Cole' s maximum of 20, 249 Clatm.

To model the effect of changes in pH, all pCO2 calculation variables were held at a

constant average (determined by mean values for the 627 lakes), while pH was changed in

increments of one unit from 3 to 12 (Figure 3-2). Also, pH values for all 627 lakes were raised

one unit of pH, and lowered one unit of pH, and pCO2 WAS calculated for all lakes with both









changes. When pH was changed by one unit, there were significant differences found between

the calculated pCO2 ValUeS and the original pCO2 ValUeS. Partial pressure carbon dioxide values

were significantly higher (p < 0.0001) when pH was decreased by one unit (p < 0.0001).

To model changes in temperature, the value of 21 C was raised to 3 1 C and lowered to 1 1

C and pCO2 ValUeS were subsequently recalculated. No significant difference was found

between the three (11 C, 21 C, 31 C) pCO2 ValUeS calculated for the lakes using the alteration in

average yearly temperature (p > 0.925). To model changes in specific conductance, all pCO2

calculation variables were held at a constant average (determined by mean values for the 627

lakes), while specific conductance values were changed from 50 CIS/cm to 400 CIS/cm. Partial

pressure carbon dioxide values were calculated over a range in pH values from 3 to 12. No

significant difference was found between varying values for specific conductance (P>0.925).

My calculations indicate 80% of the Florida lakes are supersaturated with carbon dioxide.

Supersaturation is this instance is defined as any value exceeding the atmospheric level of carbon

dioxide (370 Clatm) by more than 10%. This is similar to Cole's finding that 87% of his study

lakes were supersaturated with carbon dioxide.

Limnological Variables

The lakes, as a group (Table 3-2), had an average pH of 6.7, but individual average lake

pH values ranged from 3.9 to 11.7. Total alkalinity (meq/L) ranged from 0 to 467,000 meq/L

with an average alkalinity of 500 meq/L. Lakes ranged from oligotrophic (TP < 15Clg/L, TN <

400 Clg/L, Chlorophyll a < 3 Clg/L) to hypereutrophic (TP > 100 Clg/L, TN > 1500 Clg/L,

chlorophyll a > 40 Clg/L) (Forsberg et al. 1980) with an average TP of 46 Clg/L, an average TN of

860 Clg/L, and an average chlorophyll of 17 Clg/L. Water clarity (Secchi depth) ranged from <

0.1 m to 7.2 m with an average Secchi depth of 1.7 m. Color (Pt-Co units) ranged from 0 (very

clear) to almost 700 (black water) with an average color value of the lakes being 52 Pt-Co units.









There exists a significant relationship (r = 0.91 Figure 3-3) between pH and pCO2 aS

expected from the findings of Juday et al. (1935). If you examine Figure 3-3, and Figure 1-2

from Juday and others, both graphs do not show supersaturation carbon dioxide levels in lake

waters when the pH of water was >8. Both the Eindings in Wisconsin lakes and Florida lakes

follow the relationship between pH and the relative proportions of inorganic carbon species

(Figure 1-2) of CO2 (+H2CO3), HCO3-, and CO32- in Solution (Wetzel 1983). The data suggests

that the availability of free CO2 in an aquatic system is regulated by the pH of the system itself.

Using pairwise correlation analysis, significant correlations were found for each of the

limnological variables in relation to pCO2 (Table 3-3). Trophic state indicators (i.e., TP, TN,

Chlorophyll ) were negatively, but weakly (TP, r = -0.30; TN, r = -0.35; Chl, r = -0.48)

correlated to pCO2 ( Figures 3-4, 3-5, and 3-6 respectively). With dense populations of

phytoplankton present, the surrounding natural water becomes depleted of carbon dioxide due to

photosynthesis (Talling 1976). So one would expect a negative relationship between chlorophyll

and pCO2. Secchi disc measurements were positively, but very weakly correlated (r = -0.10).

It was surprising that the correlation of pCO2 with color (r = 0.09) was not stronger (Figure

3-7). Increases in color can often be attributed to the loading of humic compounds from

terrestrial inputs (Duarte 2005). These compounds, rich in dissolved organic carbon (DOC) can

convert from DOC to CO2 (Reche et al. 2002). When DOC undergoes bacterial degradation

upon entry to the aquatic system, there is a potential to raise aquatic respiration beyond the limits

imposed by aquatic photosynthesis (Cole 1999; Scully et al. 2003; Duarte et al. 2005). If DOC

from the watershed was a large contributing factor to increases levels of pCO2 in Florida lakes,

there should be a greater correlation between the two variables. Specific conductance was

weakly correlated (r = -0.46) with pCO2 leVOIS (Figure 3-8).









Canfield et al. (1988), found based on a survey of 165 lakes within Florida that the mineral

composition of the lakes was strongly related to Florida' s geologic and physiographic

development. The study (Canfield 1988) was a limnological survey of 165 lakes within

Florida. They found, with the exception of iron and color, pH values and chemical

concentrations generally increased as one moved from northwest to southeast. If this trend exists

for pH values in Florida lakes, then partial pressure carbon dioxide levels should therefore

decrease as one moves from northwest to southeast. This pattern does indeed exist. Using the

lake regions established in Griffiths et al. (1997), Table 3-4 shows the average pH and pCO2

values by region. As the regions progress from the northwest to south Florida, the pCO2 leVOIS

show a general decrease as pH increases (Figure 3-9).

Effect of Macrophyte Presence on pCO2 leVelS in a Florida Lake

Samples from Orange Lake were taken in the summer of 2005. Samples had an average

total phosphorus of 230 Clg/L from all three sampling periods (3- 5). Sample TP concentrations

ranged from 190 Clg/L to 650 Clg/L. Total nitrogen averaged 2306 Clg/L with a range of 1600

Clg/L to 6100 Clg/L. Chlorophyll averaged 60 Clg/L, with a range of 10 Clg/L to 100 Clg/L. Based

on the Forsberg and Ryding trophic scale (1980), this lake is hypereutrophic.

Total alkalinity (meq/L) averaged 360, with a range of 300 meq/L to 420 meq/L.

Measured water temperature averaged 29.6 C with a range of 27.8 C to 3 1.2 C. Specific

conductance averaged 86 CIS/cm over the three sampling months. Specific conductance ranged

from 81 to 92 CIS/cm. The pH of Orange Lake averaged 5.9 over the 3 monthly sampling

periods, with a range of 4.9 to 6.8 (Table 3-6). The calculated partial pressure carbon dioxide

(pCO2) averaged 21,000 Clatm over the 3 sampling periods, with a range of 10,000 to 48,000

Clatm. Based on these calculations, Orange Lake was a continual source (pCO2 > 370 Clatm) of

carbon dioxide to the atmosphere over every sampling period and at each station (Table 3-7).









In June, pCO2 ValUeS ranged from 11,000 Clatm to 17,000 Clatm (range includes stations

1,2,3 and 4). In July, pCO2 ValUeS ranged from 24,000 Clatm to 48,000 Clatm and in August,

pCO2 ValUeS ranged from 10,000 Clatm to 18,000 Clatm. In the month of July, large areas of

Orange Lake were subj ect to a mechanical aquatic plant maintenance technique known as

"cookie cutting." This process cuts large segments of floating and rooted vegetation. It causes

water agitation and sediment resuspension. This process may be the reason July pCO2 ValUeS

were greater than both June and August. The range for Orange lake pCO2 ValUeS (10,000 Clatm

to 48,000 Clatm) falls within the top 25th percentile of values exhibited in the 627 study. In the

627-lake study, Orange Lake had an average pCO2 Value Of 1449 Clatm. This difference can be

explained by the difference in pH values from the historical data collection to the present.

In 1987, Orange Lake was sampled in 5 different months at 3 stations (Canfield and Hoyer,

1992). The average pH over that time period was 7.6. Referring to modeled pH and pCO2

changes in Figure 3-3, a pCO2 Value Of about 1000 would be expected for a pH of 7.6. In the

data collected on Orange Lake in 2006, the average pH value was 5.9. A pH value of 5.9

corresponds to a pCO2 leVel Of about 30,000 according to the modeled values in Figure 3-3. The

change in pH might be the cause of the difference in pCO2 ValUeS from the historical data and

present data.

In each month, Stations 1, 2, and 3 (the macrophyte stations) had significantly higher

pCO2 leVOIS than the open water station 4 (Table 3-7, Figure 3-10). This result was anticipated

for three possible reasons:

(1) With the excessive growth of macrophytes on the shorelines in such shallow depths, I would

expect a higher content of organic matter from the breakdown of decaying plant matter by

bacteria. This process releases carbon into the water column.









(2) Another factor that may increase shoreline levels of pCO2 is terrestrial runoff entering the

system. This runoff may contain large amounts of humic compounds. These compounds, rich in

dissolved organic carbon (DOC) can photomineralize convert from dissolve organic carbon

(DOC) to carbon dioxide (CO2) (Reche et al. 2002).

(3) The average pH of stations 1,2 and 3 (the plant stations) was lower than station 4 (open water

station) over the 3 month period.

A plant survey was conducted by Florida LAKEWATCH in July 2005 on Orange Lake.

The percent volume infested with aquatic macrophytes (PVI) and the percent area covered by

macrophytes (PAC) were determined. Lakewatch found that 58% of the surface area of Orange

Lake in July 2005 was covered with aquatic macrophytes (Table 3-8). They also found that

14% of the lake's volume was filled with aquatic vegetation. Finally, LAKEWATCH calculated

a floating-leaved zone of 380 m. Floating and emergent plants extended from the shoreline 380

m into the open water area of the lake. The average pCO2 ValUe for the plant stations (1, 2, and

3) was 24,000 Clatm, while the average pCO2 Value for the open water station (4) was 17,000

Clatm. This represents more then half of the lake, having a significantly higher (p= 0.002) pCO2

level then the rest of the lake. When making estimations using data obtained on Orange Lake in

Summer 2005, a change from 0% to 100% of macrophyte coverage in the lake increased average

pCO2 leVOIS 25% (see Table 3-9). With large changes in plant coverage a noticeable change is

pCO2 may be apparent. However, small-scale changes would result in trivial changes in pCO2

levels. This suggests that aquatic macrophytes as a whole may not be as important as I had

thought to pCO2 leVOIS in the lake.

Lake Metabolism in Relationship to pCO2 Levels in a Florida Lake

The calculated net ecosystem production (NEP) for each station by month is shown in

Table 3-10. Station 1 values ranged from -1.6 to -3.1 mg C/m2 Over the three months with an









average -2.4 mg C/m2. Station 2 values ranged from -2.1 to -3.2 mg C/m2 with an average value

of -2.5 mg C/m2. Station 3 values ranged from -5.1 to -6.2 mg C/m2 with an average of -5.6 mg

C/m2. Station 4 values ranged from -5.2 to -10.7 mg C/m2 with an average value -8.6 mg C/m2

There was no significant difference in net ecosystem production values between the 3 months

(p= 0.666 at a = 0.05). There is a significant difference between the 4 stations (p= 0.003 at a =

0.05). The open-water had a significantly lower NEP than the macrophyte stations.

The data indicate that plants stations are significantly less heterotrophic than the open

water stations in Orange Lake. This may be due to the different impact wind has the surface of

the water at the various stations. The plant stations were located in the littoral areas where the

sampling equipment was in a bed of plants. Lack of wind interaction and a greater potential for

higher photosynthesis rates would make these stations less heterotrophic. The data indicate that

this lake remains in a heterotrophic state for at least some periods during the summer months.

Heterotrophy in Lakes

Some scientists contend that supersaturation of carbon dioxide indicates heterotrophy at

the ecosystem level (Raymond et al 2000). In contrast, Bachmann et al. (2000) concluded on the

basis of light and dark bottle oxygen measurements and diel oxygen measurements that Lake

Apopka, a hypereutrophic lake in central Florida, was hetorotrophic (Bachmann et al.2000). The

data on Lake Apopka from the 627-lake study indicate that Lake Apopka is not supersaturated

with respect to the atmosphere (pCO2 Of 78.3 Clatm). This is a heterotrophic lake that does not

exhibit supersaturation levels of carbon dioxide with respect to the atmosphere. The pH of Lake

Apopka is approximately 8, which is in accord with the Eindings of Juday et al. (1935) and the

627-lake study. In this case, the relationship between pH and the relative proportions of

inorganic carbon species is demonstrated in an aquatic system. At a pH of 8, the fraction of

dissolved inorganic carbon expressed as free CO2 is Very small (see Figure 1-2).









Out of the 627 historical lakes, oxygen data were available for 284 lakes (Florida

LAKEWATCH, unpublished data). There was a negative relationship between pCO2 and

percent oxygen saturation (Figure 3-11). This group of lakes can be put into 4 categories. In

Figure 3-11, the 4 quadrants of the graph represent these categories. In quadrant 1, lakes are

supersaturated with carbon dioxide (pCO2>370 Clatm), and undersaturated with oxygen (<100%).

Quadrant 2 represents lakes that are supersatured with carbon dioxide, and supersaturated with

oxygen (>100%). Quadrant 3 represents lakes that are below saturation with respect to carbon

dioxide and oxygen. Quadrant 4 represents lakes that are below saturation with respect to carbon

dioxide, and supersaturated with oxygen. If a determination of heterotrophy is made on a basis

of carbon dioxide supersaturation alone, 25% of these lakes could be mistakenly identified as

heterotrophic. Twenty-four percent of the 284 lakes fall into quadrant 2. These lakes are

supersaturated in both carbon dioxide and oxygen. Heterotrophy is generally indicated by

supersaturation of carbon dioxide. In this case, it is unclear if there is more respiration (indicated

by CO2 COncentrations) than photosynthesis (indicated by Oz) because both variables are in a

state of supersaturation.













Table 3-1. Comparison of various limnological and physical parameters measured in the 627-lake study to Cole's 1994 study.
Direct p CO2
Parameter Measurements Lake in autumn Full season cycles Lakes in summer African lakes Florida Lakes
Lakes (n) 37 1612 69 60 59 627
Samples (n) 390 1612 2395 179 79 4,730
pH (range) 4.7-9.5 3.8-9.4 4.2-9.8 4.9-9.3 6.0-9.9 3.9-11.7
DIC (p-M) (range) 4.9-2,500 13.3-4,077 6.6-4,800 11-3,578 43-145,790 -2053.7-35,286.6
p CO2 (platm) (mean) 801 1031 1064 680 2296 *2000
p CO2 (platm) (range) 107-4,128 20-9,789 1-7,845 5-8,991 32-20,249 -0.0003-9,000
Basis of p CO2 estimate Direct measurement pH, DIC pH, DIC, pH, ANC pH, DIC pH, ANC pH, ANC
Sampling intensity Weekly to quarterly Each once in Varied but more than Summer only One to three times Monthly to quarterly
Autumn 10 samples per year

Source Cole 1994 Cole 1994 Cole 1994 Cole 1994 Cole 1994 This Study
*Geometric Mean












Table 3-2. Mean, standard error and range of the variables measured in 627 lakes.
Variable Mean SE Min Max
pH 6.7 0.048 3.9 11.7
Total Alkalinity (meq/L) 500 25.15 0 467000
Temp (C) 21 0 21 21
Specific Conductance (pLS/cm a 25 C) 200 19.3 11 7200
atmospheric pCO2 (pLatm) 370 0 370 370
pCO2 pLatm 13000 2300 -0.0003 916000
Total Phosphorus (pLg/L) 46 4.22 1 1221
Total Nitrogen (pLg/L) 860 26.1 23 5600
Chlorophyll a (pLg/L) 17 1.25 0.47 300
Color (Pt-Co units) 52 3.08 0 690
Secchi (m) 1.7 1.29 0 7.3
Cloride (mg/L) 33 5.58 1.7 2200
Silicon (mg/L) 1.1 0.07 0 12.1
Sulfate (mg/L) 15 0.73 0 186
Calcium (mg/L) 11 0.49 0.19 94
Magnesium (mg/L) 8 1.25 0.16 600
Sodium (mg/L) 15 2.7 1 1160
Potassium (mg/L) 3 0.17 0 50
Iron (mg/L) 0.1 0.01 0 2.3












































-n7 -88 -8 -4 -BS -2

Longitude


Figure 3-1. Locations of 627 lakes sampled in the state of Florida.





































O Lakes pH vs Lakes log pco2
SModel pH vs Model Log10 pCO2
-- Atmospheric pCO,


4 6 8 10


Figure 3-2. Relationship of pH to pCO2 (Clatm) values for the 627 lake study. Lake values are
represented by open cirlces, with modeled pH changes shown in black. The black
line indicates atmospheric pCO2 lCVel.






































I
10 1


Table 3-3. Correlation coefficients for all variables in relation to pCO2, with N values.
Variable Correlation Coefficient N P Value
Total Phosphorus (pLg/L) -0.30 661 0
Total Nitrogen (pLg/L) -0.35 658 0
Chlorophyll a (pLg/L) -0.48 646 0.0184
Color (Pt-Co units) 0.09 636 0
Secchi (m) 0.19 479 0
Conductivity (pLS/cm & 25C) -0.46 663 0.0001
Cloride (mg/L) -0.35 659 0
Silicon (mg/L) -0.11 510 0
Sulfate (mg/1) -0.27 602 0
Calcium (mg/L) -0.59 609 0
Magnesium (mg/L) -0.44 609 0
Sodium (mg/L) -0.36 612 0
Potassium (mg/L) -0.50 610 0
Iron (mg/L) 0.29 469 0.0123


pH-

Figure 3-3. Relationship of pH to pCO2 (Clatm) values for the 627 lake study.


**












107

106 -i eg
t*
106 *


S1041 *

0". *.
O g **


101

100


0.1 1 10 100 -1000 10000

TP (pg/L)

Figure 3-4. Relationship of TP (Clg/L) to pCO2 (Clatm) values for the 627 lake study.











T U'

106 -( e




104 -, *l **
E~ eI




100


O 10 100 100100
TN (pg/L
Figure~~~~~~~~~ 3-.Rltosi fT /)t C2 +t)vle o h 2 aesuy










107





101 a
0 0



100 .~

10-
S0.1 1 10 100 100
Chorpyl pgL





Figue 36. Rlatonsip o Chorohyll (pg/L) topO2(am)vlesfr h 27lk suy




































I


107 -



106 -





10 -






10


**Ir


I, +


0,01 0.1 1 10 100 1000 10000

Color (Pt-Co Units)


Figure 3-7. Relationship of color (Pt-Co Units) to pCO2 (Clatm) values for the 627 lake study.


a
**+


4


r.











107


106 -




104-*


10 -
o I at
O
Q 102-




100

10-
1 10 100 1000 10000

Specific Conductance (CIS/cm @ 25 C)

Figure 3-8. Relationship of specific conductance (CIS/cm @ 250C) to pCO2 (Clatm) values for
the 627 lake study













Table 3-4. Average pCO2 and pH levels for each Florida Region.
Region pCO2 pH
65-01 47000 6.5
65-02 6000 6.2
65-03 11000 5.5
65-04 5000 6.7
65-05 10000 5.7
65-06 5000 6.8
75-01 52000 5.9
75-02 315000 4.8
75-03 4000 6.2
75-04 10000 5.9
75-05 6000 5.8
75-06 6100 7.6
75-07 1000 7.0
75-08 2000 7.7
75-09 30000 5.2
75-10 14000 6.7
75-11 5000 6.8
75-12 2000 7.4
75-13 22000 6.9
75-14 2000 6.6
75-15 3000 7.4
75-16 10000 7.1
75-17 1000 7.2
75-18 3000 7.2
75-19 4000 6.4
75-20 1000 7.0
75-21 3000 7.7
75-22 2000 7.7
75-23 8000 6.5
75-24 2000 7.3
75-25 300 8.1
75-27 5000 6.2
75-28 1000 7.9
75-30 1000 8.5
75-31 3000 7.7
75-32 3000 7.5
75-33 2000 7.0
75-34 4000 6.7
75-35 1000 7.6
75-36 17000 6.9
75-37 300 8.5
76-02 1000 8.2
76-03 8000 7.3
















e *
10$ -1 6

# O V



1@







r pH


Figur 3-9 Avrg HadpO yrgon ein oefo otws f Foiat
sot Flrd inasuhatdrcin













60000


50000


40000

-i 4
30000
O

20000 -


1000 -




June July August

Monthy Averages for Stations 1-4


Figure 3-10. Monthly pCO2 averages by station for Orange Lake.
















106- a




E 10j ....



a 3 4


10'

100 .

10-1
20 40 60 80 100 120 140 160 180 200

Oxygen Percent Saturation
Figure 3-11. Partial pressure carbon dioxide values in relationship to % oxygen saturation for
284 historical lakes. The vertical line represents 100% oxygen saturation. The
horizontal line represents atmospheric level of pCO2.






































(pS/cm & 25 C)
89
93
88
92
87
82
82
81
90
82
84
86


(m/sec)
6.1 5
6.0 7
6.2 7
6.2 8
5.6 7
5.8 5
5.7 12
5.7 9
6.0 7
6.2 7
6.0 4
6.3 6


concentrations by month and station in Orange


Table 3-6. Average carbon estimate variables and wind by month and station in Orange Lake,
Florida.
Month Station Total Alkalinity Temp Specific Conductance pH Wind


Table 3-5. Average of nutrient and chlorophyll
Lake, Florida.
Month Station Total Phosphorus

6 1 240
6 2 190
6 3 290
6 4 220
7 1 230
7 2 360
7 3 300
7 4 210
8 1 260
8 2 350
8 3 220
8 4 240


Total Nitrogen

2000
1800
2000
1800
1700
3300
2400
1600
1700
3000
1800
1700


Chlorphyll

50
30
80
20
30
100
50
10
60
90
20
30


(meq/L)
390
340
380
420
410
300
370
370
350
340
350
370


(C)
28.2
29.4
27.9
28.3
29.5
29.9
30.7
30.1
30.2
31.2
30.0
30.0












Table 3-7. Average pCO2 by month and station in Orange Lake, Florida.
Month Station pCO2 (pLatm)
6 1 17000
6 2 15000
6 3 13000
6 4 11000
7 1 48000
7 2 24000
7 3 42000
7 4 31000
8 1 18000
8 2 11000
8 3 15000
8 4 10000



Table 3-8. Aquatic plant data collected on July 7, 2005. Source: Florida LAKEWATCH 2005.
Variable Value
% area covered with aquatic vegetation (PAC %) 58
% of lake's total volume filled with vegetation (PVI %) 13.5
Average emergent plant biomass (kg wet wt/m2) 3.2
Average floating-leaved biomass (kg wet wt/m2) 4.5
Average submersed plant biomass (kg wet wt/m2) 5.6
Average width of emergent and floating-leaved zone (m) 380
Average lake depth (m) 3

Table 3-9. Percent coverage of macrophytes in Orange Lake and the estimated pCO2 for that
plant coverage value.
% Cover Estimated pCO2
0 18000
10 18000
20 19000
30 20000
40 20000
50 21000
60 21000
70 22000
80 23000
90 23000
100 24000


Table 3-10. Net ecosystem production values for each station by month with an average for all
samples at each station in Orange Lake, Florida.
Net Eco sy stem Production
Station # June July August Average
1 -1.6 -2.6 -3.1 -2.4
2 -2.1 -2.25 -3.2 -2.5
3 -5.6 -5.1 -6.2 -5.6
4 -5.2 -10.7 -9.8 -8.6









CHAPTER 4
CONCLUSIONS

From a 627-lake study, a large numbers (80% of those sampled) of Florida lakes are

potential sources of carbon to the atmosphere. Partial pressure carbon dioxide levels in Orange

Lake even at their lowest calculated Clatm were more than 26 times than that of the overlying

atmosphere. It would be difficult to argue that this lake is not a carbon source to the atmosphere.

The data set, however, is only representative of the summer months. Full seasonal data may not

be necessary to accurately estimate pCO2 leVOIS based on the temperature modeling in the 627

lake study. Certain meteorological events such as hurricanes may cause significant short term

increases to pCO2 depending on runoff volume and its effect on alkalinity values, and also the

amount of sediment resuspension.

Trophic state indicators, alkalinity, and specific conductance were all weakly correlated

with pCO2 leVOIS. The 627-lake study (Canfield 1981; Canfield and Hoyer 1992; Florida

LAKEWATCH 1993, 1996; Greis 1995) supports the conclusion that levels of partial pressure

carbon dioxide in lakes are controlled by pH levels in the lake. Much of the literature to date

has focused on the effects of DOC from the watershed and its relationship to supersaturation

(Dillon et al. 1997, Frankignoulle et al. 1998). The data from the 627-lake study (based on color

data) did not support the higher correlations with dissolved organic carbon (DOC) found in the

literature.

The role of aquatic macrophyte coverage and its potential effects on carbon dioxide

concentrations in surface waters may need further investigation. When making estimations using

data obtained on Orange Lake in Summer 2005, a change from 0% to 100% of macrophyte

coverage in the lake increased average pCO2 leVOIS 25% (see Table 3-9). These findings suggest

that small-scale changes in aquatic macrophyte coverage would not produce significantly higher










changes in pCO2 leVOIS in a Florida Lake. These findings may change if lakes of different

trophic states are examined.

Data indicate that Orange Lake is a supersaturated with carbon dioxide and seems to be a

continual source of CO2 to the atmosphere. Partial pressure carbon dioxide levels were

significantly higher in plant zones. However, net ecosystem production values were lower in

plant areas than in the open water area. Both areas were heterotrophic (NEP was negative). This

could be another indicator that plants themselves may not be a causal factor in increases pCO2

levels. The location of the plants in the littoral zone and their effect on pH, sediment

resuspension and nutrient cycling may be more important factors.

Oxygen saturation data indicates that supersaturation of carbon dioxide alone should not be

used as measure of heterotropy at an ecosystem scale. There were several Florida lakes that

were supersatured with both oxygen and carbon dioxide.

The limnological parameter pH is a key parameter that influences many biological and

chemical factors including species distribution, multiple equilibrium expressions, contaminant

metal ion concentrations, naturally occurring organic and inorganic species and the dissociation

of carbonic acid to name a few (Wetzel 2001). Many attempts have been made to quantify and

rank variables of significance to predict mean values of lake pH. Several catchment and lake

morphometric parameters have been investiged. The data from the 627 lake study followed a

geologically regulated trend in pH as found by Canfield et al. 1988. The trend in pH

corresponded with an expected trend in pCO2 ValUeS by lake region. This further supports that

pH is the regulating factor in indirectly measured pCO2 ValUeS in Florida freshwater lakes.

When looking for causal factors to explain variability in pCO2 leVOIS between aquatic systems,










one should examine the possible influence of regional geology underlying the aquatic systems in

question.










WORKS CITED


American Public Health Association (APHA). 1985,1989, 1992. Standard Methods for the
examination of water and wastewater, 16th, 17th, 18th Editions, Washington D.C.

Bachmann, R., M. V. Hoyer, and D.E. Canfield, Jr. 2000. Internal heterotrophy following the
switch from macrophytes to algae in Lake Apopka, Florida. Hyrdrobiologia 418:217-227.

Bachmann, R. and D.E. Canfield Jr. 1996. Use of an alternative method for monitoring total
nitrogen concentrations in Florida lakes. Hydrobiologia 323:1-8.

Bowling, L, M. Steane, P.Tyler. 1986. Spectral Distribution and Attenuation of Underwater
Irradiance in Tasmanian Inland Waters. Freshwater Biology FWBLAB 16:313-335.

Brenner, M., M.W. Binford and E.S. Deevey. 1990. Lakes. Pages 363-391 in R.L. Myers and J.J.
Ewel (eds.), Ecosystems of Florida. University of Central Florida Press, Orlando.

Canfield, D.E. Jr., and R. Bachmann. 1981. Prediction of total phosphorus concentrations,
chlorophyll a, and Secchi depths in natural and artificial lakes. Canadian Journal of
Fisheries and Aquatic Sciences 38:414-423.

Canfield, D.E. Jr., K. Langeland, M.J. Maceina, W.T. Haller, J.V. Shireman, and J.R. Jones.
1983. Trophic state classification of lakes with aquatic macrophytes. Canadian Journal of
Fisheries and Aquatic Sciences 40:1713-1718.

Canfield, D.E. Jr., and M.V. Hoyer. 1988. Regional geology and the chemical and trophic state
characteristics of Florida lakes. Lake and Reservoir Management 4:21-3 1.

Canfield, D.E., Jr., and M.V. Hoyer. 1992. Aquatic macrophytes and their relation to the
limnology of Florida lakes. Final Report submitted to the Bureau of Aquatic Plant
Management, Florida Department of Natural Resources, Tallahassee, FL.

Carignan R, D. Planas, C. Vis. 2000. Planktonic production and respiration in oligotrophic
Shield lakes. Limnology Oceanography 45:189-99.

Cole, J.J., M.L. Pace, S.R. Carpenter, and J.F. Kitchell. 2000. Persistence of net heterotrophy in
lakes during nutrient addition and food web manipulations. Limnology Oceanography 45,
1718-1730.

Cole J.J. 1999. Aquatic microbiology for ecosystem scientists: new and recycled paradigms in
ecological microbiology. Ecosystems 2:2152-25.

Cole J.J., and N.F. Caraco. 1998. Atmospheric exchange of carbon dioxide in a low-wind
oligotrophic lake measured by the addition of SF6. Limnology and Oceanography 43:647-
656.

Cole, J. J., N. F. Caraco, G. W. Kling, and T. W. Kratz. 1994. Carbon dioxide supersaturation in
the surface waters of lakes. Science 265:1568-1570.










del Giorgio PA, and R.H. Peters. 1994. Patterns in planktonic P:R ratios in lakes: Influence of
lake trophy and dissolved organic carbon. Limnology Oceanography 39:772-87.

D'Elia C.F., P.A, Steudler, and N. Corwin. 1977. Determination of total nitrogen in aqueous
samples using persulfate digestion. Limnology Oceanography 22:760-764.

Duarte, C., and Y. Prairie. 2005. Prevalence of Heterotrophy and Atmospheric CO2 Emissions
from Aquatic Ecosystems. Ecosystems 8:862-870.

Dillon P.J., and L.A. Molot. 1997. Dissolved organic and inorganic carbon mass balances in
central Ontario lakes. Biogeochemistry 36:29-42.

Florida LAKEWATCH. 2002. Florida LAKEWATCH Annual Data Summaries for 1986
through 2001. Department of Fisheries and Aquatic Sciences, University of
Florida/Institute of Food and Agricultural Sciences. Library, University of Florida.
Gainesville, Florida.

Forsberg, C., S. Ryding. 1980. Eutrophication Parameters and Trophic State Indices in 30
Swedish Waste-Receiving Lakes. Archiv fur Hydrobiologie 89:189-207.

Frankignoulle M., G. Abril, A. Borges, I. Bourge, C. Canon, B. Delille, E. Libert, and J.M.
Theate. 1998. Carbon dioxide emision from European estuaries. Science 282:434-8.

Greis, J. G. 1985. A characterization of 60 Ocala National Forest Lakes. Report by John G.
Greis, Hydrologist. National Forests in Florida. Tallahassee, FL.

Griffith, G.E., D.E. Canfield, Jr., C.A. Horsburgh, and J.M. Omernik. 1997. Lake Regions of
Florida. EPA/R-97/127. U.S. Environmental Protection Agency, National Health and
Environmental Effects Research Laboratory, Corvallis, OR. 89p.

Hach Chemical Company. 1975, 1992. Procedures, chemical lists and glassware for water and
wastewater analysis. HACH Chem. Co., Ames, Iowa.

Hanson, P., R. Stephen, D. E. Carpenter, E. Armstrong, H. Stanley, and T. K. Kratz. 2006. Lake
dissolved inorganic carbon and dissolved oxygen: changing drivers from days to decades.
Ecological Monographs 76:343-363.

Hanson, P.C., Bade, D. L., Carpenter, S. R., and T. K. Kratz. 2003. Lake metabolism:
Relationships with dissolved organic carbon and phosphorus. Limnol. Oceanogr. 48: 1112-
1119.

Harrison WG, J. Aristegui, E.J.H. Head, W.K.W. Li, A.R. Longhurst, and D.D. Sameoto. 2001.
Basin-scale variability in plankton biomass and community metabolism in the subtropical
North Atlantic Ocean. Deep-Sea Res II 48:2241-69.

Hartman, B. and D. E. Hammond. 1985. Gas exchange in San Francisco Bay. Hydrobiologia
129:59-68.










Hem, J.D. 1992. Study and interpretation of chemical characteristics of natural water: U. S.
Geological Survey Water-Supply Paper 2254, 263 p.

Hesslein R.H., J.W.M. Rudd, C. Kelly, P. Ramlal, and K.A. Hallard. 1991. Carbon dioxide
pressure in surface waters of Canadian lakes. In: Wilhelms SC & Gulliver JS (Eds) Air-
Water Mass Transfer, Second International Symposium on Gas Transfer at Water Surfaces
(pp 413-431). Am. Soc. Civil Eng., New York, NY.

Howarth, R. W., R. Schneider, and D. Swaney. 1996. Metabolism and organic carbon fluxes in
the tidal, freshwater Hudson River. Estuaries 19: 848-865.

Jonsson, A., M. Meili, A. Bergstram, and M. Jansson. 2001. Whole-lake mineralization of
allochthonous and autochthonous organic carbon in a large humic lake. Limnology
Oceanography 46: 1691-1700.

Juday, C., E. Birge, and W. W. Meloche. 1935. The carbon dioxide and hydrogen ion content of
the lake waters of northeastern Wisconsin. Wisconsin Academy of Sciences, Arts, and
Letters .

Kling G.W., G.W. Kipphut, and M.C. Miller. 1991. Arctic lakes and streams as gas conduits to
the atmosphere: implications for tundra carbon budgets. Science 251:298-301.

Menzel, D. W. and N. Corwin. 1965. The measurement of total phosphorus in seawater based on
the liberation of organically bound fractions by persulfate oxidation. Limnology and
Oceanography 10: 280-282.

Odum, H. T. and C. M. Hoskin. 1958. Comparative studies on the metabolism of marine
waters. Institute of Marine Science, University of Texas 5:16-46.

Parsons, T.T. and J.D.H. Strickland. 1963. Discussion of spectrophotometric determination of
marine-plant pigments, with revised equations for ascertaining chlorophylls and
carotenoids. Journal of Marine Research 21: 155-163.

Prairie Y.T., D.F. Bird, J.J. Cole. 2002. The summer metabolic balance in the epilimnion of
southeastern Quebec lakes. Limnology and Oceanography 47:316-321.

Raymond, P., J. Bauer, and J. Cole. 2000. Atmospheric CO2 Evasion, Dissolved Inorganic
Carbon Production, and Net Heterotrophy in the York River Estuary. Limnology and
Oceanography 45:1707-1717.

Reche, I., and M.L. Pace. 2002. Linking dynamics of dissolved organic carbon in a forested lake
with environmental factors. Biogeochemistry 61:21-36.

Sartory, D. P., and J. U. Grobbelaar. 1984. Extraction of chlorophyll a from freshwater
phytoplankton for spectrophotometric analysis. Hydrobiologia 114: 177-187.

Sawyer, C.N. and P.L. McCarty. 1967. Chemistry for sanitary engineers, 2nd ed.: McGraw-Hill,
St. Louis, MO.










Scully N.M., W.J. Cooper, and L.J. Tranvik. 2003. Photochemical effects on microbial activity
in natural waters: the interaction of reactive oxygen species and dissolved organic matter.
FEMS Microbial Ecology 46:353-7.

Simal, J. 1985. Second derivative ultraviolet spectroscopy and sulfumic acid method for
determination of nitrates in water. Journal of Analytical Chemistry 68:962-964.

Sobek S, G. Algesten, A.K. Bergstroin, M. Jansson, and L.J. Tranvik. 2003. The catchment and
climate regulation of pCO2 in boreal lakes. Global Change Biology 9:630-641.

StatView, Version 5. SAS Institute Inc., Cary, NC.

Striegl R.G., P. Kortelainen, and J.P. Chanton. 2001. Carbon dioxide partial pressure and 13C
content of north temperate and boreal lakes at spring ice melt. Limnology and
Oceanography 46:941-945.

Stumm, W. and J.J. Morgan. 1981. Aquatic chemistry, 2nd ed.: New York, John Wiley & Sons,
780 p.

Stumm, W., and J.J. Morgan. 1995. Aquatic Chemistry: Chemical Equilibria and Rates in
Natural Waters. Wiley. NY.

Tailing J. F. 1976. The depletion of carbon dioxide from lake water by phytoplankton. Journal
of Ecology 64:79-121.

USEPA. 1979. Methods for chemical analysis of waters and wastes. U. S. Environmental
Protection Agency. Washington D. C.

Wetzel R.G. 2001. Limnology. Lake and River Ecosystems. Academic Press, San Diego.

Wetzel, R.G. and G. E. Likens. 1991. Limnological analyses. Springer-Verlag, New York: 391
pp.

Wollin, K. M. 1987. Nitrate determination in surface waters as an example of the application of
UV derivative spectrometry to environmental analysis. Acta Hydrochemica Hydrobiologia
15:459-469 (Ger.).

Yentsch, C. S. and D. W. Menzel. 1963. A method for the determination of phytoplankton
chlorophyll and phaeophytin by fluorescence. Deep Sea Res. 10: 221-231.









BIOGRAPHICAL SKETCH


Jenney Kellogg was born in Bath, NY in 1981. She grew up in the Finger Lakes area of

NY surrounded by vineyards and wineries. She has always loved the outdoors. She attended

Hammondsport Central High School and graduated in 2000 ranked 4th in her class. She attended

Niagara University from 2000-2004 and received a Bachelor of Science degree with Honors in

Biology. Her undergraduate thesis work dealt with oxygen depletion in the central basin of Lake

Erie.