Citation
Trace Metals in Florida Lake Sediments

Material Information

Title:
Trace Metals in Florida Lake Sediments
Creator:
Blair, Susanna Whitman
Place of Publication:
[Gainesville, Fla.]
Publisher:
University of Florida
Publication Date:
Language:
english
Physical Description:
1 online resource (230 p.)

Thesis/Dissertation Information

Degree:
Doctorate ( Ph.D.)
Degree Grantor:
University of Florida
Degree Disciplines:
Geology
Geological Sciences
Committee Chair:
BRENNER,MARK
Committee Co-Chair:
MARTIN,ELLEN ECKELS
Committee Members:
MARTIN,JONATHAN BOWMAN
BONZONGO,JEAN-CLAUDE J
KAMENOV,GEORGE DIMITROV
Graduation Date:
5/3/2014

Subjects

Subjects / Keywords:
Geological Sciences -- Dissertations, Academic -- UF
drought -- lakes -- lead -- metals -- paleolimnology
Little Lake Johnson ( local )
Sediments ( jstor )
Lakes ( jstor )
Groundwater ( jstor )
Genre:
Electronic Thesis or Dissertation
born-digital ( sobekcm )
Geology thesis, Ph.D.

Notes

Abstract:
Despite substantial research on metals in the environment, there remain uncertainties concerning trace metal pollution and associated human health concerns. Although metal pollution is found world-wide, impacts are often site-specific. I investigated the input, fate and transport of trace metals in north Florida lake sediments. To accurately quantify anthropogenic impacts on trace metal inputs to lakes, it is necessary to estimate background or pre-anthropogenic inputs. Sediment cores from Florida Lakes Sheelar, Pebble, and Little Johnson were dated using 210Pb and analyzed for trace metals. Metals V, Cr, Ni, Cu, Zn, Sn, Sb, Bi, and Pb show up-core enrichment beginning ~1900, coinciding with the onset of population growth and development in Florida. Trace metal concentrations measured in these lake sediments provide a baseline (reference) record of pre-anthropogenic metal accumulation and a record of modern, anthropogenically influenced metal deposition this area. Lake sediments are typically sinks for trace metals, if the deposits remain undisturbed and permanently buried. Future climate change scenarios suggest increased duration and frequency of dry events, which may cause stage declines in shallow Florida lakes and lead to physical and/or chemical redistribution of legacy metal pollution. I examined the impact of low water levels on metal transport in lake sediments by measuring metal concentrations along transects of lake sediment exposed following extreme lake level declines in 2012. From the historic lake shore to the center of the lake trace metal concentrations increased up to 3 times. The main mechanism for metal dispersal during low lake levels was focusing of fine-grain material and organic matter toward the lake center. The main source of Pb in these exposed lake sediments was anthropogenic, largely from automobile gas additives, and was associated with a relatively mobile sediment fraction. Finally, I assessed the preservation of trace metal stratigraphy in lake sediments after dry periods. Using metal concentrations and Pb isotope ratios from repeat cores in two shallow lakes, taken before and after low water levels, I determined that the record of metal accumulation is preserved stratigraphically. Metal concentrations in Florida lake sediments have increased over the last century. Climate change scenarios predict more intense rainy periods, alternating with more severe droughts. The combined effects of changing hydrology and relic metal pollution may create future management concerns, including legacy metal pollution in sediments and water, as well as the potential for increased metal-laden dust. ( en )
General Note:
In the series University of Florida Digital Collections.
General Note:
Includes vita.
Bibliography:
Includes bibliographical references.
Source of Description:
Description based on online resource; title from PDF title page.
Source of Description:
This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Thesis:
Thesis (Ph.D.)--University of Florida, 2014.
Local:
Adviser: BRENNER,MARK.
Local:
Co-adviser: MARTIN,ELLEN ECKELS.
Electronic Access:
RESTRICTED TO UF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE UNTIL 2014-11-30
Statement of Responsibility:
by Susanna Whitman Blair.

Record Information

Source Institution:
University of Florida
Holding Location:
University of Florida
Rights Management:
Copyright by Whitman Blair. Permission granted to University of Florida to digitize and display this item for non-profit research and educational purposes. Any reuse of this item in excess of fair use or other copyright exemptions requires permission of the copyright holder.
Embargo Date:
11/30/2014
Resource Identifier:
907294926 ( OCLC )
Classification:
LD1780 2014 ( lcc )

Downloads

This item has the following downloads:


Full Text
xml version 1.0 encoding UTF-8
REPORT xmlns http:www.fcla.edudlsmddaitss xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.fcla.edudlsmddaitssdaitssReport.xsd
INGEST IEID E8BDK54ZJ_RXJ7KU INGEST_TIME 2014-10-03T22:24:12Z PACKAGE UFE0046613_00001
AGREEMENT_INFO ACCOUNT UF PROJECT UFDC
FILES



PAGE 1

TRACE METAL S IN FLORIDA LAKE SEDIMENTS By SUSANNA WHITMAN BLAIR A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2014

PAGE 2

2014 Susanna Whitman Blair

PAGE 3

To Ed, Mary, Blair, and Poppy

PAGE 4

4 ACKNOWLEDGMENTS This project was funded by grants from Drilling Observation and Sampling of the Ea The Gulf Coast Association, Geo logical Society of America University of Florida Land Use and Enviro nmental Change Institute Univers ity of Florida Department of Geological Sciences, University of Florida College of Liberal Arts and Sciences and the NSF IGERT Grant # 0504422 Adaptive Management: Wise use of Water, Wetlands and Watersheds. I would like to thank the Mike Roess Gold Head B ranch State Park staff for their a ssistance in sampling efforts. I would like to thank Jason Curtis for assisting with sampling William Kenney for his help with 210 Pb dating and many helpful conversations and also to Edmond Dunne for help with data analy sis and editing I would like to thank my advisor, Mark Brenner for his support and encouragement, thoughtful contribution s, and also field assistance. I would like to thank my committee members, Ellen Martin, Jonathan Martin, and Jean Claude Bonzongo, for their helpful comments and suggestions and especially George Kamenov for his assistance in laboratory analysis and genuine interest in this project. I w ould like to acknowledge the University of Florida and NSF Science Partners hips in Collaborative Education (SPICE) Fellowship and the Conservation Clin ic at the Levin College of Law for these opportunities and for helping me shape my future career aspirations I would like to acknowledge my dear friends in my IGERT cohort ; Megan, Lisa, Hollie, Sarah, Estelle, Sarah, and Robin You made this journey worthwhile and I am so lucky to have you in my life I thank my mom, Mary Blair, for her unwavering support and for always being proud of me. Finally, I wish these two words were bigger thank you to Ed Dunne. I c His love and support are immeasurable.

PAGE 5

5 TABLE OF CONTENTS page ACKNOWLEDGMENTS ................................ ................................ ................................ .. 4 LIST OF TABLES ................................ ................................ ................................ ............ 8 LIST OF FIGURE S ................................ ................................ ................................ ........ 10 ABSTRACT ................................ ................................ ................................ ................... 13 CHAPTER 1 TRACE METALS IN THE ENVIRONMENT ................................ ............................ 15 1.1 Problem Statement ................................ ................................ ........................... 15 1.2 Trace Metals: General Overview ................................ ................................ ....... 17 1.3 Using Lake Sediments to Measure Trace Metals ................................ ............. 19 1.4 Florida Study Site ................................ ................................ ........................... 20 1.5 Scope of Research ................................ ................................ ........................... 23 2 RECE NT ACCELERATION OF TRACE METAL DEPOSITION IN THREE FLORIDA LAKES ................................ ................................ ................................ ... 26 2.1 Introduction ................................ ................................ ................................ ....... 26 2.2 Materials and Methods ................................ ................................ ...................... 31 2.2.1 Sample Collection ................................ ................................ ................... 31 2.2. 2 Dry Mass, Organic Matter Content and Bulk Density .............................. 31 2.2.3 Geochronology ................................ ................................ ........................ 32 2.2.4 Trace Metal Analysis ................................ ................................ ............... 32 2.2.4 Pb Isotope Analysis ................................ ................................ ................. 33 2.2.5 Data Treatment ................................ ................................ ....................... 33 2.3 Results ................................ ................................ ................................ .............. 35 2.3.1 210 Pb Chronology ................................ ................................ ..................... 35 2.3.2 Physical Properties of the Sediment Cores ................................ ............. 35 2.3.3 Sediment Geochemistry ................................ ................................ .......... 35 2.3.3.1 Lake Sheelar sediment core ................................ .......................... 36 2.3.3.2 Pebble Lake sediment core ................................ ............................ 36 2.3.3.3 Little Lake Johnson sediment core ................................ ................. 37 2.3.4 Normalization and Enrichment Factors (EFs) of Trace Metals ................ 37 2.4 Discussion ................................ ................................ ................................ ........ 40 2. 4.1 Pre Anthropogenic Trace Metal Delivery ................................ ................. 40 2.4.2 Anthropogenic Trace Metal Delivery and Timing ................................ ..... 44 2.4.3 Metal Inventories and Accumulation Rates ................................ ............. 45 2.4.4 Anthropogenic Pb Concentration and Pb Isotopes ................................ .. 48 2.4.5 Anthropogenic Zn ................................ ................................ .................... 51

PAGE 6

6 2.5 Conclusions ................................ ................................ ................................ ...... 53 3 IMPLICATIONS OF TRACE METAL DISTRIBUTIONS IN FLORIDA LAKE SEDIMENTS DURING EXTREME LAKE LEVEL LOWS ................................ ........ 68 3.1 Introduction ................................ ................................ ................................ ....... 68 3.1.1 Climate Change in Florida ................................ ................................ ....... 69 3.1.2 Indirect Effects of Climate Change ................................ .......................... 71 3.1.3 Florida Lakes ................................ ................................ ........................... 72 3.2 Methods ................................ ................................ ................................ ............ 73 3.2.1 Study Sites ................................ ................................ .............................. 73 3.2.2 Sample Collection ................................ ................................ ................... 75 3.2.3 Dry Mass and Organic Matter Content ................................ .................... 76 3.2.4 Grain Size Analysis ................................ ................................ ................. 76 3.2.5 Trace Metal Analysis of Total Digested Sediment ................................ ... 76 3.2.6 Statistical Analysis ................................ ................................ ................... 77 3.3 Results ................................ ................................ ................................ .............. 78 3.3.1 Water Leve ls ................................ ................................ ........................... 78 3.3.2 Sediment Characteristics ................................ ................................ ......... 78 3.3.3 Sediment Geochemistry ................................ ................................ .......... 79 3.3.4 Grain size and Organic Matter ................................ ................................ 81 3.4 Discussion ................................ ................................ ................................ ........ 82 3.4.1 Me tal Variability ................................ ................................ ....................... 82 3.4.2 Sediment Character and Metal Distribution ................................ ............. 85 3.4.3 Organic Matter ................................ ................................ ......................... 86 3.4.4 Grain Size ................................ ................................ ................................ 87 3.4.5 Other Factors Controlling Metal Distribution ................................ ............ 90 3.5 Future Implications ................................ ................................ ............................ 93 3.5.1 Sediment Quality Concerns ................................ ................................ ..... 94 3.5.2 Water Quality Concerns ................................ ................................ .......... 95 3.5.3 Sediment Dust ................................ ................................ ......................... 96 3.6 Conclusions ................................ ................................ ................................ ...... 97 4 LEAD CONCENTRATIONS AND ISOTOPE RATIOS IN FLORIDA LAKE SEDIMENTS EXPOSED DURING LOW LAKE LEVELS: IMPLICATIONS FOR LEGACY PB POLLUTION IN LIGHT OF FUTURE CLIMATE CHANGE SCENARIOS ................................ ................................ ................................ ......... 114 4.1 Introduction ................................ ................................ ................................ ..... 114 4.2 Methods ................................ ................................ ................................ .......... 117 4.2.1 Study Sites ................................ ................................ ............................ 117 4.2.2 Sample Collection ................................ ................................ ................. 118 4.2.3 Total Pb Analysis of Digested Sediment ................................ ................ 120 4.2.4 BCR Sequential Extraction Procedure ................................ ................... 121 4.2.5 Pb Isotope Analysis ................................ ................................ ............... 122 4. 3 Results and Discussion ................................ ................................ ................... 122 4.3.1 Pb Concentrations from Sequential Extractions ................................ .... 126

PAGE 7

7 4.3.2 Pb Isotopes ................................ ................................ ........................... 130 4.4 Conclu sions ................................ ................................ ................................ .... 13 2 5 PRESERVATION OF TRACE METAL STRATIGRAPHY IN FLORIDA LAKE SEDIMENTS AFTER LOW LAKE STAGE EVENTS ................................ ............ 145 5.1 Introduction ................................ ................................ ................................ ..... 145 5.2 Materials and Methods ................................ ................................ .................... 148 5. 2.1 Study Sites ................................ ................................ ............................ 148 5.2.2 Sample Collection ................................ ................................ ................. 149 5.2.3 Dry Mass, Organic Matter Content, and Bulk Density ........................... 150 5.2.4 210 Pb Chronology ................................ ................................ ................... 150 5.2.5 Trace Metal Analysis ................................ ................................ ............. 151 5.2.6 Stable Pb Isotope Analysis ................................ ................................ .... 152 5.2.7 Statistical Analysis and Inventory Calculations ................................ ...... 153 5.3 Results ................................ ................................ ................................ ............ 154 5.3.1 Water Levels ................................ ................................ ......................... 154 5.3.2 21 0 Pb Chronology and Inventory ................................ ............................ 154 5.3.3 Sediment Physical Properties ................................ ................................ 155 5.3.4 Trace Metal Concentrations and Inventories ................................ ......... 156 5.3.5 Stable Pb Isotopes ................................ ................................ ................ 157 5.4 Discussion ................................ ................................ ................................ ...... 157 5.4.1 Lake Harris ................................ ................................ ............................ 157 5.4.2 Newn ans Lake ................................ ................................ ....................... 161 5.5 Conclusions ................................ ................................ ................................ .... 165 6 SUMMARY ................................ ................................ ................................ ........... 185 APPENDIX A CHAPTER 2 SUPPLEMENTARY MATERIAL ................................ ...................... 191 B CHAPTER 4 SUPPLEMENTARY MATERIAL ................................ ...................... 206 LIST OF REFERENCES ................................ ................................ ............................. 210 BIOGRAPH ICAL SKETCH ................................ ................................ .......................... 230

PAGE 8

8 LIST OF TABLES Table page 2 1 Statistical analysis of trace elements in the Sheelar, Pebble and Little Lake Johns o n sediment cores. All concentration values are g/g. .............................. 58 2 2 Inventories g/cm 2 ) of selected anthropogenic metals for the last ~110 years, excess 210 Pb (dpm/cm 2 ) inventories (~1900 to present) and mean annual accumulation rates for metals during the intervals from 1885 (Sheelar) or 1900 (Pebble and Little Johnson) to 1970 and 1970 to 2009 (Sheelar) or 2012 (Pebble and Little Johnson). ................................ .................. 65 3 1 Measured Canadian Certified Reference Materials Project lake sediment standard LKSD 4 concentrations compared with reported values .................... 102 3 2 Summary statistics for water levels from Little Lake Johnson and Lake Geneva ................................ ................................ ................................ ............. 102 3 3 Descriptive statistics for pH, percent water and organic matter, and metal concentrations ( g/g) from all samples and from Little Lake Johnson (n=33) and Lake Geneva (n=36). Descriptive statistics for pH, percent water and organic matter, and metal concentrations ( g/g) from sampling locations at 10 meter intervals along transects from this historic high w ater shoreline (0 m) at Little Lake Johnson and Lake Geneva (n=3). Values are means one standard deviation. ................................ ................................ ........................... 105 3 4 Pearson's co rrelation coefficients among trace elements and percent organic matter from Little Lake Johnson and Lake Geneva sediment transects ........... 110 3 5 size distribution of sediment transects from Little Lake Johnson and Lake Geneva ................................ ................................ ................................ ............. 112 4 1 Modified BCR sequential extraction scheme used for metal speciation ........... 136 4 2 Lead concentration ( g/g) in sediment fractions from BCR sequential extraction and total digestion in sediment samples from Little Lake Johnson and Lake G eneva transects, and LKSD 4 lake sediment standard .................. 137 5 1 Lake Harris and Newnans Lake characteristics ................................ ................ 169 5 2 Summary statistics for water levels from Lake Harris and Newnans Lake ....... 172 5 3 Pb 210 date, mass sedimentation rate, percent organic matter, trace metal concentrations from sediment core intervals from Lake Harris (1999 and 2013) and Newnans Lake (1997 and 2005). Means, standard deviation and z statistic from Wilcoxon rank sum test of means are reported for each core .. 175

PAGE 9

9 5 4 Stable Pb isotope data from sediment cores collected from Lake Harris (1999 and 2013) and Newnans Lake (1997 and 2005) ................................ ............... 183 5 5 Index ratios of sediment character and trace metal concentration inventories from repeat coring events from Lake Harris (1999 and 2013) and Newnans Lake (1997 and 2005) ................................ ................................ ...................... 184

PAGE 10

10 LIST OF FIGURE S Figure page 2 1 M ap of Florida showing the location (star) of Mike Roess Gold H ead Branch State Park ................................ ................................ ................................ ........... 55 2 2 Sediment depth (cm) versus 210 Pb dates (solid lines) and sediment accumulation rate (g cm 2 yr 1 ) versus 210 Pb dates (dashed lines) for the Lake Sheelar, Pebble Lake, and Little Lake Johnson cores ................................ ........ 56 2 3 Percent organic matter versus dept h in sediment cores from Lake Sheelar, Pebble Lake, Little Lake Johnson ................................ ................................ ....... 57 2 4 Concentration profiles of Sc, V, Ni, Cu, Zn, Sr, Z r, Sn, Sb, and Pb in the Lak e Sheelar sediment core ................................ ................................ ........................ 60 2 5 Concentration profiles of Sc, V, Ni, Cu, Zn, Sr, Zr, Sn, Sb, and Pb in the Pebble Lak e sediment core. ................................ ................................ ............... 61 2 6 Concentration profiles of Sc, V, Ni, Cu, Zn, Zr, Sn, Sb, and Pb in the Little Lake Johnson sediment cor e ................................ ................................ .............. 62 2 7 Enrichment factor (EF) profiles for selected metals from the Lake Sheelar, Pebble, and Little Johnson sedim ent cores with selected 210 Pb dates ............... 63 2 8 Sc normalized trace metal concentrations for V, Cu, Zn, Sr, Zr and Pb in the Lake Sheelar, Pebble Lake, and Little Lake Johnson sedi ment cores ................ 64 2 9 Pb concentration and 206 Pb/ 207 Pb isotope ratio versus date in sediment cores from Lake Sheelar, Pe bble Lake, Little Lake Johnson ................................ ........ 67 3 1 Maps of study locations ................................ ................................ ...................... 99 3 2 Lake stage (ft NAVD88) for Lake Geneva and Little Lake Johnson since 1 957 and 1945, respectively ................................ ................................ ...................... 100 3 3 Photographs of Little Lake Johnson (left) and Lake Geneva (right) during sampling ev ents ................................ ................................ ................................ 101 3 4 Cumulative distribution function of water levels from Little Lake Johnson and Lake Geneva from 1957 and 1945 to 2013, respectively ................................ 103 3 5 Percent water by weight and percent organic matter by dry weight in sediments collected from Little Lake Johnson transect and Lake Geneva transects 1 an d 2. ................................ ................................ ............................ 104

PAGE 11

11 3 6 Concentrations ( g/g) of V, Cr, Ni, Cu, Zn, Sn, Pb, Sc and Nd measured in sediments collected along a transect from a) Little Lake Johnsons and b) two transects from Lake Geneva ................................ ................................ ............. 107 3 7 Percent distribution of grain size (>63 m<) and organic matter in sediment from Lit tle Lake Johnson transect and transects 1 and 2 from Lake Geneva ... 109 3 8 Scatter plots of percent organic matter and Vanadium and Zinc concentrations ( g/g) measured from sediments transects from Little Lake Johnson and Lake Geneva ................................ ................................ ............... 113 4 1 Maps of study loca tions. ................................ ................................ .................. 134 4 2 Lake stage (ft NAVD88) for Lake Geneva and Little Lake Johnson since 1957 and 1945, respectively. ................................ ................................ ..................... 135 4 3 Average pH, percent water weight, percent organic matter, and total Pb concentrations (n=3), and measured percent distribution of grain si ze (>63 m<) measured at 10 m intervals along sediment transects from Little Lake John son and Lake Geneva ................................ ................................ ............... 138 4 4 Percent organic matter vs. total Pb concentration ( g/g) in surface sediments from Little Lake Johnso n and Lake Geneva. ................................ .................... 139 4 5 Percent <63 m grain size fraction vs. total Pb concentration ( g/g) in sediments from Little Lake Johnson and Lake Geneva. Combined data from all three lak es (n=23) yield an r value of 0.87 (p<0.01). ................................ .... 140 4 6 Relative abundance of Pb in surface sediment fractions, determined by the BCR sequential extraction procedure. ................................ .............................. 141 4 7 208 Pb/ 204 Pb measured in sequentially extracted sediment fractions from Little Lake J ohnson and Lake Geneva transects from the historic high lake level (0 m) towards the lake center ................................ ................................ ............... 142 4 8 208 Pb/ 204 Pb measured in total digested sediment from FL lake and peat cores over the last century.. ................................ ................................ ....................... 143 4 9 Comparison between 206 Pb/ 204 Pb, 207 Pb/ 204 Pb, a nd 208 Pb/ 204 Pb in leached sediment fractions ................................ ................................ ............................ 144 5 1 Map of Florida showing the locations of Lake Harris and Newnans Lake. ........ 168 5 2 Lake stage (ft NAVD88) for Lake Harris from 19 56 to 2014 ............................. 170 5 3 Lake stage (ft NAVD88) for Newnans Lake from 1936 to 201 4. ....................... 171

PAGE 12

12 5 4 Cumulative distribution functions of water levels from Lake Harris and Ne wnans Lake ................................ ................................ ................................ .. 173 5 5 210 Pb and 226 Ra activities and date versus depth from repeat sediment cores collected at Lake Harris (1999 and 2013) and Newnans Lake (1997 and 2005). ................................ ................................ ................................ ............... 174 5 6 Percent organic matter and trace element concentration profiles measured from Pb 210 dated sediment cores taken in 1999 (black) and 2013 (red) from Lake Harris. ................................ ................................ ................................ ...... 178 5 7 Percent organic matter and trace element concentration profiles measured from 210 Pb dated sediment cores taken in 1997 (black) and 2005 (re d) from Newnans Lake. ................................ ................................ ................................ 179 5 8 Inventories for unsupported 210 Pb, sediment bulk density, organic matter, and trace metals calculated from 1950 to 1999 from sediment cores co llected in 1999 and 2013 from Lake Harris. ................................ ................................ ..... 180 5 9 Inventories for unsupported 210 Pb, sediment bulk density, organic matter, and trace metals calculated from 1951 to 1997 from sediment cores collected in 1997 and 2005 from Newnans Lake ................................ ................................ 181 5 10 Changes in 208 Pb/ 204 Pb ratio with age in sediment cores from Lake Harris (1999 and 2013), Newnans Lake (1997 and 2005), Lake Sheelar (Blair et al. in review), Little Lake Jackson (Escobar et al. 2013), and a peat core from Blue Cypress Marsh (Kamenov et al. 2009). ................................ .................... 182

PAGE 13

13 A bstract of Dissertation Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy TRACE METAL S IN FLORIDA LAKE SEDIMENTS By Susanna Whitman Blair May 2014 Chair: Mark Brenner Major: Geological Sciences Despi te substantial research on metal s in the environment there remain uncertainties concerning trace metal pollution and associated human health concerns Although metal pollution is found world wide, impacts are often site specific. I investigated the input, fate and transport of trace metals in north Florida lake sediments. To accurately quantify anthropogenic impacts on trace metal inputs to lakes, it is necessary to estimate background or pre anthropogenic inputs Sediment cores from Florida Lakes Sheelar, Pebble, and Little Johnson were da ted using 210 Pb and analyzed for trace metals. Metals V, Cr, Ni, Cu, Zn, Sn, Sb, Bi, and Pb show up core enrichment beginning ~1900 coinciding with the onset of population growth and development in Florida. Trace metal concentrations measured in these lak e sediments provide a baseline (reference) record of pre anthropogenic metal accumulation and a record of modern, anthropogenically influenced meta l deposition this area L ake sediments are typically sinks for trace metals, if the deposits remain undistur bed and permanently buried. Future climate change scenarios suggest increased d uration and frequency of dry events which may cause stage declines in shallow Florida lakes and lead to physical and/ or chemical redistribution of legacy metal

PAGE 14

14 pollution. I exa mined the impact of low water levels on metal transport in lake sediments by measuring metal concentrations along transects of lake sediment exposed following extreme lake level declines in 2012 From the historic lake shore to the center of the lake trace metal concentrations increased up to 3 times. T he main mechanism for metal dispersal during low lake levels was focusing of fine grain material and organi c matter toward the lake center. T he main source of Pb in these exposed lake sediments was anthropogenic, largely from automobile gas additives, and was associated with a relatively mobile sediment fraction. Finally, I assessed the preservation of trace metal stratigraphy in lake sediments after dry periods. Usin g metal concentrations and Pb isotope ratios from repeat cores in two shallow lakes, taken before and after low water levels, I determined that the record of metal accumulation is preserved stratigraphically Metal concentrations in Florida l ake sediment s have increased over the last century. Climate change scenarios predict more intense rainy periods alternating with more severe droughts. The com bined effects of changing hydrology and relic metal pollution may create future management concerns includin g legacy metal pollution in sediments and water, as well as the potential for increased metal l aden dust.

PAGE 15

15 CHAPTER 1 TRACE METALS IN THE ENVIRONMENT 1.1 Problem Statement Despite substantial research on major metal accumulation and associated health concerns, there remain uncertainties concerning historic and modern trace metal pollution. In a feature article in the journal Nature Rockstrm et al. (2009) commented on the l ack of information regarding metals in our environment. Since the Industrial Revolution human actions have been the main drivers of global environmental change and the authors argue that we need a framework of boundaries to maintain global safe space biophysical system properties, which often react abruptly, especially around threshold levels. Once threshold levels are surpassed, ecosystems shift into a new state, often with catastr ophic consequences. One of the nine identified planetary boundaries that ha s Although levels were defined for nearly all of the boundaries, those for chemical pollution, inc believe there is not enough information regarding the effects, thresholds, and consequences of chemical pollution to set such boundaries. Similarly Fleishman et al. (2011) defined the 40 top priorities for s cience to inform United States conservation and management p olicy, and include d the determination of the aggregate effects on ecosystems of current use and emerging toxicants, including metals. It has been argued that metal pollution is one of the first types of anthropogenic pollution. Records of atmospheric metal pollution from mining and smelting are preserved in Greenland ice layers and date back to Roman times (500 B.C. to A.D.

PAGE 16

16 300). However, since the Industrial Revolution ( ca. AD 1850), metal production and atmospheric emissions have increased substantially (Nriagu 1996). Metal pollution is now ubiquitous, most commonly via atmospheric circulation and deposition (Dillon and Evans 1982). There is no doubt about the impact that anthropogenic activities have on the cycling of trace metals both regionally and globally. Although some debate continues, there is research that shows positive correlations between metal concentrations and human health problems such as se, autism, neurological impairment, and some cancers (Huang et al. 2006; K idd 2000; Dexter et al. 1991). Historically, lead pollution was limited to gasoline additive and paint, however Richmond Bryant et al (2013) found that blood levels in children in Detroit, MI are a function of current lead levels in atmospheric dust. In India and China air pollution is one of the greatest human health concerns and metals are found to be abundant in the airborn e particulate matter (Cheng 2003; Massey et al. 2013). Additionally, certain metals can bio accumulate in organisms and bio magnify up the food chain. High metal concentrations are shown to decrease immune function, increase parasite burdens, decrease survival of shore birds (Provencher et al. 2014) and increa se reproductive deficits in alligators (Burger et al. 2000). Lake sediments are excellent archives for studying regional and temporal deposition of metals, as these sediments generally remain undisturbed and are datable. Paleolimnological studies allow for the examination of recent human activities on trace metal cy cling, and for the reconstruction of watershed histories including more long term metal variability, necessary to gauge the overall impact of the human influence (Kolak et al 1998; Gelinas et al 2000; Norton et a l 1992; Bindler et al 2011; Liu et al 2 012).

PAGE 17

17 Using paleolimnological techniques, metal concentration measurements, and Pb isotope analysis, my research is designed to contribute new knowledge and understanding of the concentrations and rates of metal accumulation i n northern Florida lake system s. S o that human related impacts c ould be evaluated accurately I quantified background or p re anthropogenic concentrations and used this paleolimnological data to study present day biogeochemical cycling of metals and possible future impacts in light of c hanging hydrologic patterns 1.2 Trace Metals: General Overview In the periodic table of chemical elements, the m ajor, minor and trace elements of interest in environmental studies include; Al, As, Ba, Be, Bi, Ca, Cd, Co, Cr, Cu, Fe, Ga, Ge, Hg, K, Li, Mg Mn, Mo, Na, Nb, Ni, Pb, Re, Sc, Se, Si, Ta, Te, Th, Ti, W, Y, Zn, Zr. Unlike the major elements t race metals are commonly considered depleted in the they tend to be more abundant in igneous and metamorphic rocks than in sedimentary rocks. Generally these metals are measured in the ppm to ppb range. In biological and nutritional studies that t race metals in very small amounts are required by living organisms, particularly mammals; however high doses of many of these same metals are toxic and are considered pollutants (Kabata Pendias 2010). Metals are commonly defined by their physical properties in their elemental state. Metals have a metallic luster, the capacity to lose electrons and form positive ions, and the ability to cond uct heat and electricity (Phipps 1981). However, the term is often used indiscriminately to refer to both the element and compounds (Duffus 2002). For example, throughout this dissertation I will refer to Pb, but I am specifying the Pb compound.

PAGE 18

18 Before an thropogenic use of metals, the presence of specific metals in a sedimentary system was controlled solely by weathering, atmospheric dust, and local lithology (Boyle 2001b; Hofmann et al 2005). Metals weather from source rock and accumulate d in sedimentary systems and are therefore useful for provenance studies (Engstrom and Wright 1984; Bilali et al. 2002). Since the onset of anthropogenic use of mineral resources and fossils fuels, presence or absence of elements no longer reflects solely natural processes Trace metals in the modern environment are sourced from both natural and anthropogenic sources (Kamenov et al. 2009). Natural sources are generally attributed to weathering and erosion of local lithology (Kamenov et al. 2009, Outridge et al. 200 5; Koinig et al. 2003) or input from geologic events (i.e. volcanic debris) (Pearson et al. 2010) or from atmospheric dust (Shotyk et al. 2001). Anthropogenic sources are related to nearby human activities and land uses (i.e. fertilizer run off) or atmosph eric inputs from distant anthropogenic activities. These include airborne particulates from industrial activities such as smelting, incineration, coal combustion, or vehicle emissions. Additionally, land use s such as forest clearing and agriculture can en hance soil erosion, which may contribute to loading of trace metals to lakes (Davis et al 2006; Gelinas and Schmit 1998). As a consequence of some environmental laws and regulations, point source discharge of metals into the environment is limited and mu ch current concern is focused on airborne particulate matter that enters lake systems from non point sources and is much more difficult to control. Airborne metal particulates are a product of a number of human activities from mining and smelting to the widespread use of leaded gasoline. Trace elements are

PAGE 19

19 long range, diffuse urban and/or industrial pollutants (Nriagu, 1996) which can travel long distances before being re deposited (Patterson and Settle 1987; Wang et al. 2010). Trace metal contaminatio n from atmospheric transport has given rise to ubiquitous background contamination and is not limited to urban centers or land adjacent to pollution producing activities (VanMetre et al. 2006) It is measured in remote locations such as high Alpine lakes and Antarctic glaciers (Santos et al. 2004; Camarero et al. 2009; Thevenon et al. 2010). Once deposited airborne particles can be subjected to a range of environmental physicochemical transformations. These can include precipitation or co precipitation, cation exchange, and adsorption to organic matter and/or Fe and Mn oxides (Sigg et al. 1987; Tessier et al. 1996) as well as incorporation by plants and bioaccumulation in consumers (Nagel and Loskill, 1991). Paleolimnological studies have proven effective for documenting historical trace metal trends and have demonstrated human alteration of the trace meta l cycle, including smelting operations during the Roman Empire and the extensive atmospheric emissions during the Industrial Revolution (Nriagu 1996; Shotyk et al. 2001). Lake sediments can be dated with a number of techniques, including 210 Pb, radiogenic element deposition, or historical events, such as volcanic eruptions. If undisturbed, lake sediments represent natural archives for reconstructing past environmental changes, including recent environmental change s caused by anthropogenic influence s 1. 3 Using Lake Sediments to Measure Trace Metals Lake sediment chemistry is a function of a number of factors including : 1) the flux of autochthonous material, 2) the relative proportions of organic and inorganic mat ter and 3) the input of allochthonous atm ospheric material, including pollutants, to the lake

PAGE 20

20 system (Norton et al. 1992). Lake sediments accurately preserve a record of trace metal deposition to lake systems (Siver and Wozniak 2001; Bindler et al. 2008) However, significant research has investigat ed the degree to which early diagen e tic alteration and physical sediment mobility may alter metal profiles in sediments (Winderlund et al. 2002; Schottler and Engstrom 2006; Rydberg et al. 2008). In lake processes must be considered when in vestig ating anthropogenic influence s (Outridge et al. 2005; Augustsson et al. 2010). These processes include post burial remobilization, physical mixing, compaction, bioturbation, trophic state change, and diffusive migration of dissolved metals as a result of r edox related concentration gradients (Boudreau 1999; Boyle 2001b; Pearson et al. 2010). I f paleolimnological studies are to yield reliable histories of metal accumulation, the sediment metal ngr et al. (2007) argue there are two time scales of trace element cycling one fast and temporary on the scale of sea s ons and the other slow and long lived on the scale of years. The scale of the former does not seem to be recorded because it is overr id den by the lat t er. Although remobilization of metals in surficial lake sediment is acknowledged, many studies conclude d that remobilization was not an important problem in most situations and that the metal profiles in lakes sediments faithfully record t he history of trace metal accumulation (Taylor and Davison 1995; Boudreau 1999; Couillard et al. 2008; Percival and Outridge 2013). 1.4 Florida Study S ite In 2003, Florida had the 4th largest population in the U.S., with 17 million inhabitants. In 1900, grew very rapidly since the 1930s and among states in the U.S., Florida experienced

PAGE 21

21 the 3rd fastest growth since the mid 1900s, only after California and Texas apid growth rate contributed to large scale environmental impacts. Florida did not experience the Industrial Revolution as did the northern United States and environmental degradation in Florida has been largely limited to the 20th century. Florida has about 7800 lakes, which cover at least 9270 km 2 about 6% of the disturbance in the last century, especially with respect to increased nutrient loading (Kratzer and Brezonik 1981; Riedinger Whitmore et al. 2005) Few studies, however, have investigated recent changes in trace metal delivery to the water bodies, an alternative form of anthro pogenic influence. Some studies focused on trace metal delivery to estuarine environments (Caccia et al. 2003; MacDonald et al. 1996), and the few lakes that have been investigated also experienced cultural eutrophication or acidification (Norton et al. 1992; Karlen et al. 2009) or were st udied to evaluate the effects of trace metals on biota (Burger et al. 2000; Heinz et al. 2012) Escobar et al. (2013) investigated Pb accumulation in two south central Florida lakes that were highly influenced by l ocal agriculture and industry. In contrast to that study, I focused on small, closed basin lakes in a relatively undisturbed area of north Florida to study the recent history of trace metal deposition. Although not discussed further in this dissertation, m ercury is another important trace metal pollutant and has been the subject of many studies in Florida (Gottgens et al. 1999). One component of my dissertation is an assessment of climate change effects on trace metal cycling in Florida. Most c limate change outcomes in Florida are focused

PAGE 22

22 on the impacts of sea level rise. However, in my study, the important aspect of future climate change scenarios relates to hydrologic cycling, namely the f requency and duration of dry and wet events. The impact of climate c hange on water resources can be characterized by changes in temperature and precipitation. However, the output of most climate models is insufficiently resolved to make accurate predict ions of hydrologic changes for small regions. In a study examining wat er resources in south Florida, Obeysekera et al. (2010) found that many General Circulation Model (GCMs) do not capture the statistical characteristics of regional rainfall because of the large size of grid cells. This is further complicated by the uncerta inty related to changes in the El Nino Southern Oscillation (ENSO), the Atlantic Multidecadal Oscillation (AMO), and the Pacific Decadal Oscillation (PDO), all of which a ffect precipitation on the Florida peninsula (Moses et al. 2012). Using precipitation and temperature data across Florida from the last century, increasing temperatures are fairly constrained, wh ereas predict ed changes with respect to precipitation are highly uncertain (Mulholland et al. 1997; Martinez et al. 2012). In Florida, over the las t century there ha s been a significant decreasing trend in precipitation in October and May, and an increasing trend w as found for the summer wet season (Florida Oceans and Coastal Council 2010; Martinez et al. 2012). P rojected increasing temperature will also increase evapotranspiration (Mulholland et al. 1997). Although there remains some uncertainty, projections suggest there will be increases in storm intensity and clustering in the summer months. This will result in extreme hydrographs, with large peaks in flow coupled with lower baseflows and longer periods of drought (Mulholland et al. 1997).

PAGE 23

23 1.5 Scope of Research The research presented in this dissertation addresses the source, transport and fate of trace metals in Florida lake sediments. F uture management of lake sediment and water quality with respect to metal contamination, will require accurate records of previous metal loading and quantification of potential metal mobility. In these studies I estimated bac kground or pre anthropogenic metal concentrations and quantif ied historic and modern human related metal accumulation rates Although metals are sequestered in lake sediments it is possible that in light of future changes in hydrologic patterns resulting from climate change and present day biogeochemical cycling of pollutants these metal s may become sources of pollution This introductory chapter (Chapter 1) outline s the importance of studying trace metals and why lake sediments serve as effective archiv es that record temporal and spatial changes in metal accumulation. The rest of the dissertation is structured as follows : The first study (Chapter 2) investigates the metal accumulation record in three rural north Florida lakes from before 1900 (i.e. effe ctively pre anthropogenic metal concentrations ) to present day. Using 210 Pb dated sediment cores reference or background pre anthropogenic trace metal concentrations were determined, and changing concentrations through to the present were quantified. The goals were to 1) quantify background or pre anthropogenic metal concentrations, 2) define local reference conditions in atmospheric metal deposition and evaluate changes in metal accumulation from 1900 to present, which covers the per iod of major population growth in Florida, and 3) assess inter lake differences in trace metal histories. Chapter 2 was submitted, reviewed revised and resubmitted to t he Journal of Paleolimnology

PAGE 24

24 Chapters 3 and 4 explore the indirect effects of climate change on metal laden lake sediments. Predictions of future climate change include alteration of hydrologic patterns. In Florida longer wet periods followed by more intense drought periods are expected ( Martinez et al. 2012 ). Because of the shallow nature of many Florida lake s and their connection to groundwater, this will drastically alter water levels in these lakes. Water level fluctuation is natural and even necessary for the survival of some species. Climate driven changes in the frequency, duration, and severity of water level fluctuations, however, will impair ecosystem functioning (Mulholland et al. 1997; Zohary and Ostrovsky 2011) and affect water quality (White et al. 2008; Martinez et al. 2012) particularly in shallow lake systems. Declining water levels are ass ociated with sediment transport and drying, as well as organic matter decomposition, factors that influence the fate of trace metals in the sediments. For this study I collected transect samples across two shallow Florida lakes in which low water level s oc curred durin g the extreme 2012 winter spring dry period I aimed to constrain the distribution of trace metals in exposed lake bottom sediments during the period of historic water level lows, using total metal concentrations, grain size distribution and se diment lithology. In Chapter 3 I investigated the factors that account for the distribution of Pb, Cu, Zn, Ni, Sn, V, Cr, Sc, and Nd concentrations across the lake bottoms. Also, I discussed some of the future implications of subaerial sediment exposure. In Chapter 4 I focused specifically on Pb and utilized sequential leaching techniques and Pb isotopic measurements to further constrain the mechanism for Pb distribution in the exposed lake sediments and the source of Pb present in these sediments. Results from both studies have environmental management implications.

PAGE 25

25 The results of Chapters 3 and 4 demonstrate that changing hydrologic patterns can alter sediment and metal distributions in lake basins. This begs the question of how these hydrologic changes may a ffect paleolimnological investigations of trace metal accumulation. The work presented in Chapter 5 looks to address this question using trace metal measurements from repeat cores collected from two Florida lakes before and after water level fluctuations. Lake Harris was cored in 1999 and 2013 and Newnans Lake was cored in 1997 and 2005 B etween coring events in each lake, there were dra matic decreases in lake stage as a consequence of low rainfal l Metals commonly attributed to anthropogenic activities, V, Ni, Cu, Zn, and Pb, and conservative elements, Sc and Nd, were measured in 210 Pb dated cores. Metal concentration profiles and inventories were compared between the repeat cores to determine the degree of similarity. Additionally, radiogenic Pb isotopes were measured to determine potential Pb mobility. Using these measurements I aimed to deter mine if drying events altered the recorded metal deposition in shallow Florida lakes. Finally, in Chapt er 6 I present my concluding remarks

PAGE 26

26 CHAPTER 2 RECENT ACCELERATION OF TRACE METAL DEPOSITION IN THREE FLORIDA LAKES 1 2.1 Introduction Anthropogenic activities have affected regional and global trace metal cycling, thereby altering biogeochemical cycles and causing environmental deterioration. Evidence from South America, Europe, and Asia shows disruption of trace metal cycles by early hu man activities (e.g. smelting), thousands of years ago (Norton and Kahl 1987; Shotyk et al. 2001; Bindler et al. 2008; Thevenon et al. 2011) Over the last 100 150 years, however, concentrations and accumulation ra tes of trace metals in lake and estuary sediments have increased dramatically (Graney et al. 1995; Nriagu 1996; Liu et al. 2012). Paleolimnological studies have assessed the magnitude and geographic extent of recent human mobilization of trace metals and t heir accumulation in lake sediments (Siver and Wozniak 2001; Mahler et al. 2006; Pearson et al. 2010; Bindler et al. 2011). These modern increases are a consequence of localized industrial and land use practices, and fallout of globally ubiquitous, metal laden atmospheric dust. Effective future management of trace metal impacts requires understanding local natural sources, quantification of pre anthropogenic baseline conditions, and measurement of atmospheric metal deposition in the region. Trace metals in the environment come from natural and anthropogenic sources. Natural sources include weathering and erosion of local rock (Koinig et al. 2003; Outridge et al. 2005) and input from other geologic processes (e.g. volcanic eruptions) This chapter has been accepted with revisions to the Journal of Paleolimnology Revisions were addressed and the article was re submitted. At the time of submission of this dissertation the article was in second review.

PAGE 27

27 (Pearson et al. 2010) Anthropogenic sources include activities such as mining, manufacturing and waste disposal and land uses such as agriculture (Gelinas and Schmit 1998; Davis et al. 2006) Another major source for anthropogenic trace metal enrichment is atmospheric fallout from smelting, incineration, coal combustion, and vehicle emissions. Airborne particulates can travel long distances before being deposited (Wang et al. 2010) Thus, trace metal contamination from atmospheric transport is not limited to urban centers or areas near p ollution sources, but now accounts for pervasive global contamination (VanMetre et al. 2006) Trace metal contamination is even detected in remote locations such as Alpine lakes and Antarctic glaciers (Santos et al. 2005; Camarero et al. 2009) In 2003, Flo rida had the 4 th largest population in the U.S., with 17 million grew very rapidly since the 1930s and among states in the U.S., Florida experienced the 3 rd fastest growt h since the mid 1900s, behind only California and Texas ( Demographia 2013). scale environmental impacts. Many lakes in Florida experienced human disturbance in the last century, especially with respect to in creased nutrient loading (Kratzer and Brezonik 1981; Riedinger Whitmore et al. 2005) Few studies, however, have investigated recent changes in trace metal delivery to th e water bodies, an alternative proxy for anthropogenic influence. Some studies focused on trace metal delivery to estuarine environments (Caccia et al. 2003; MacDonald et al. 1996), and the few lakes that have been investigated also experienced cultural e utrophication or acidification (Norton et al. 1992; Karlen et al. 2009) or were studied to evaluate the effects of trace metals on

PAGE 28

28 biota (Burger et al. 2000; Heinz et al. 2012) Escobar et al. (2013) investigated Pb accumulation in two south central Florida lakes that were highly influenced by local agriculture and industry In contrast to that study, I selected three small, closed basin lakes in a relatively undisturbed area of north Florida to study the recent history of trace metal deposition. Giv en the relatively undisturbed surroundings and considerable distance from major anthropogenic sources, I surmised that these lakes record only regional at mospheric metal deposition. The main objectives were to: 1) define local reference conditions with res pect to trace metal deposition, 2) establish the timing and quantify changes in atmospheric trace metal deposition resulting from human activities and 3) evaluate inter basin differences with respect to the trace metal histories. Study sites: Lakes Sheela r, Pebble, and Little Johnson are located in Mike Roess Gold Head Branch State Par k, north central Florida ( Figure 2 1 ). The park is about 8 km NE of the town of Keystone Heights. All three lakes are located in the Sandhill Lake District. Lakes in this reg ion are underlain by deep quartz sands and in most cases occupy sinkholes formed by the dissolution and collapse of underlying limestone. The drainage basins are small and water enters primarily via direct rainfall and subsurface inflows. During particular ly wet periods, Pebble Lake receives some inflow from an ephemeral stream (SFDEP 2010) and Little Johnson receives inflow from a small groundwater derived stream. All three lakes lack outflows. This area is important for groundwater recharge and characterized by excessively well drained Entisols (Weatherspoon et al. 1989). Lakes in the region have received much attention because of their astatic nature and recent declines in stage. They have not undergone nutrient enrichment.

PAGE 29

29 Lake Sheelar (29 max = 21 m) and has held water continuously since the end of the last glacial period (Watts and Stuiver 1980) The watershed is ~69.5 ha and the lake area is ~7.3 ha (SJRWMD 2013 a ). The north shore possesses inflow groundwater seeps that are visible at low lake levels. The lake is oligotrophic and has relatively stable water levels compared to many other lakes in the area. Recreation at Sheelar is limited to hiking around the lake. its highest lake stage, the lake area is only ~4 ha. Pebble Lake is the most astatic water body in the area, with r ecorded water level decreases of 93 cm in one month, and a nearly 10 m stage range from 1948 to 1956. This high variability is attributed to changes in precipitation and artesian pressure, downward leakage (seepage), and perhaps excessive groundwater extra ction (Deevey 1988; Motz and Heaney 1991) Pebble Lake and is larger and shallo wer than the other two lakes. It is oligotrophic to mesotrophic, with an average maximum depth and area of 4.5 m and 11 ha, respectively. Groundwater fed Gold Head Branch Stream flows into Little Lake Johnson and at high stage the lake connects with larger Lake Johnson. In 1957 a dam was constructed to stabilize water levels for recreational activities, but was removed in 2002 to restore the natural hydrology (Motz and Heaney 1991) Both Little Lake Johnson and Pebble Lake are close to recreat ional centers in Goldhead Branch State Park. Camping, picnic, and parking facilities are only a few hundred m from these water bodies and may be local sources of metals to these two lakes.

PAGE 30

30 Local bedrock consists of Pleistocene age Trail Ridge Sands and the Pliocene age Cypresshead formation. Both formations consist of several hundred feet of semi consolidated marine and non marine deposits of sand, clay, marl, gravel, limestone, dolomite and dolomitic limestone (SFDEP 2010) The Trail Ridge Sand also has economically important concentrations of zircon and staurolite, among other titanium minerals (Pirkle et al. 1984). The study lakes are located in a relatively rural area of north Florida, however there are historic and modern sources of at mospherically derived metals nearby. The closest large city is Jacksonville, Florida, which lies ~65 km to the northeast. Gainesville, Florida is a smaller city, ~40 km to the southwest. The rail line that connected Fernandina, north of Jacksonville, to C edar Key, on the Gulf Coast, was completed in 1860 and passed within 10s of km of these lakes. Another rail line, connecting Jacksonville and Orlando (central Florida) was completed ~1900 and also passed fairly close to the study area (Florida Memory 2013) Interstate highways I 95 and I 75 were completed in northern Florida in the 1960s (Historic Florida Mainlines 2013). A few heavy mineral, sand, and clay mining operations near Goldhead Branch State Park began operations as early as 1916 and continue toda y. The heavy mineral mines are wet mills and surficial mining operations that may emit limited dust, consisting primarily of sand and titanium rich minerals (Pirkle et al. 1984). Phosphate mining began in northern Florida in the 1880s, however it was relat ively short lived there, and soon moved to richer deposits in south central Florida (Florida Industrial and Phosphate Research Institute 2013). There is a coal fired power plant ~40 km southeast of the lakes that began operating in the mid 1980s and anoth er in Jacksonville that began

PAGE 31

31 burning oil in the mid 1960s and coal in 2002. Although power plants are important potential sources of metals, they represent relatively recent potential contributors to the iest coal fired power plant came online in 1906 in Tampa. Wind measurements taken at Lake Sheelar for the past 13 years show no dominant wind direction (SJRWMD 2013 a ), so it is difficult to discern the degree to which these sources contributed to fallout in this area. 2.2 Materials and Methods 2.2.1 Sample C ollection I collected sediment cores from deep areas in each of the three lakes. I collected an 80 cm core from Lake Sheelar on 31 August 2009 using a sediment water interface corer (Fisher et al. 1992) I collected cores at both Pebble Lake and Littl e Lake Johnson on 29 March 2012. Both lakes had just gone dry and I obtained cores while standing on the lake bottom. A 134.5 cm core was retrieved from Pebble Lake and an 85 cm core was taken from Little Lake Johnson. I extruded the Sheelar core vertical ly and sampled at 2 cm intervals. I treated the uppermost sediments in Lakes Pebble and Little Johnson similarly, but I sectioned sediments below 30 cm depth in the two cores at 4 cm intervals. 2.2.2 Dry Mass, Organic Matter Content and Bulk D ensity I use d acid cleaned containers and Optima grade reagents for sample preparation and analysis. To determine percent dry mass, I weighed wet samples and then re weighed them after freeze drying. I ground the dry samples to a fine powder. I estimated percent orga nic matter for dry samples by loss on ignition (LOI) at 550 C (Hkanson and Jansson 1983 ) All elements of interest have boiling points >550 C, so

PAGE 32

32 ashing caused no elemental loss. I calculated sediment bulk density (g dry mass cm 3 ) from percent dry mass in wet sediment and the organic/inorganic proportion of dry sediment (Binford 1990) 2.2.3 Geochronology I dat ed cores using 210 Pb and measured radioisotope activities ( 210 Pb, 226 Ra [ 214 Bi], 137 Cs) by gamma counting (Appleby et al. 1986) using EG&G Ortec GWL High Purity Germanium coaxial well detectors attached to a 4096 channel pulse height analyzer (Schelske et al. 1994) Unsupported 210 Pb activity was calculated as the difference between total 210 Pb activity and supported 210 Pb ( 226 Ra) activity at each depth. Dates were determined using the constant rate of supply (CRS) model (Appleby and Oldfield 1978; 1983; Ol dfield and Appleby 1984) 2.2.4 Trace Metal A nalysis I prepared and digested all samples from the three cores for geochemical analysis in a class 1000 clean lab, equipped with class 10 laminar flow hoods, at the Department of Geological Sciences, Univ ersity of Florida. All reagents used for sample preparation were Optima grade. About 0.05 g of sediment was weighed, transferred to acid cleaned Teflon vials, and digested with 1 ml concentrated HF and 2 ml concentrated HNO 3 The samples were placed in an oven for 48 hours at 90 C. The vials were opened and the solution was evaporated to dryness on a hot plate. After evaporation, samples were treated with 2 ml of 6 N HCl and vials were capped and heated on a hot plate overnight to ensure full dissolut ion. The solution was evaporated again to dryness. Five ml of 0.8 N HNO 3 with 100 ppm HF spike with 8 ppb Rh and Re, w ere added to the samples to re dissolve the residue. A small aliquot of the solution was removed and diluted with the same spiked 0.8 N HNO 3 for a final dilution of ~2,000x

PAGE 33

33 for trace metal analyses (Kamenov et al. 2009). Procedural blank was prepared with the samples. Trace element analysis was performed on an Element2 HR ICP MS in medium resolution with Rh and Re used as internal standa rds. Quantification of the results was done with external calibration using USGS rock standards AGV 1, BIR 1, BCR 2. The following elements were analyzed for all samples and standards: Sc, V, Cr, Co, Ni, Cu, Zn, Ga, Ge, Rb, Sr, Y, Zr, Nb, Cd, Sn, Sb, Cs, B a, Hf, Ta, Tl, Pb, Bi, Th and U. Analytic recovery was ensured by including subsamples of USGS certified reference standard AGV 1, with recovery between 95 and 103%. Precision was measured by replicate measurement of 10% of the samples, and was >90% for al l elements. Only elements well above the detection limit and with satisfactory replicate values were included in the interpretation. Metal concentrations are reported as g/g dry sediment. 2.2.4 Pb Isotope A nalysis The remaining sample solution was evapo rated to dryness and dissolved in 1 N HBr and loaded onto columns packed with Dowex 1X 8 resin to separate Pb. The sample was washed 3x with 1 ml of 1N HBr and the Pb fraction was collected in 1 ml of 3N HNO 3 (Kamenov et al. 2009). I determined Pb isotope composition on a Nu Plasma MC ICP MS with Tl normalization following the procedure described in Kamenov et al. (2004). The Pb isotope data are expressed relative to the following values of National Bureau of Standards (now National Institute of Standards and Technology) Standard Reference Material for native Pb NBS 981: 206 Pb/ 204 207 Pb/ 204 208 Pb/ 204 2.2.5 Data T reatment Trace metal concentrations in bulk dry sediment may vary due to changes in metal delivery and shifts in the accumulation of mineral matter. To correct for the

PAGE 34

34 variability in mineral sedimentation, I normalized metal concentrations of interest (M) to a cons ervative element (X) that is thought to be unaffected by anthropogenic inputs and/or recent disturbance of hydrological processes. I assumed that this conservative element represents mineral material originating from local rock or atmospheric soil/rock dus t. Geochemical normalization is often used to distinguish unenriched, historical baseline conditions from those of modern environmental impacts (Wu et al. 2010; Bos et al. 2011) Calculated enrichment factors (EFs) reflect the degree of anthropogenic atmosph eric contribution compared to metals from natural origin. The EF is the concentration ratio of a metal (M) to a conservative element (X) in a recent, enriched sample, divided by the same ratio in a pre industrial, i.e. baseline sample, which lacks enrichme nt. EF = (M/X) sample / (M/X) baseline (1 1 ) Often, upper continental crust (UCC) trace metal concentrations are used for baseline ratios. This, however, may not reflect local bedrock chemistry accurately and is more appropriate for large systems, not small catchments (Reimann and Caritat 2000; N'guessan et al. 2009) Another commonly used baseline value is the ratio in regional bedrock, however, that has not been well established for this region. Some studies de termined the baseline or natural background using ratio values from the base of the core, however this requires cores that extend well beyond the period of human influence. Bindler et al. (2 011) argued that background concentrations in Europe are found in deposits >3,000 years old. Given the recent development in Florida, I established the average baseline concentration for each core using unenriched, pre 1900 conditions, as discussed bel ow.

PAGE 35

35 2.3 Results 2.3.1 210 Pb C hronology In all three cores, 210 Pb and 226 Ra activity profiles showed relatively consistent decreases in unsupported 210 Pb activity with depth, suggesting that the CRS modeled dates are reliable. There was low unsupported 210 Pb activity at a depth of 14 16 cm in the Little Lake Johnson core, and the 12 cm depth was assigned the supported/unsupported 210 Pb boundary. Age depth relations are shown in Figure 2 2. 2.3.2 Physical P roperties of the Sediment C ores Percent organic matt er (LOI 550 C) differed among the cores from the three lakes ( Figure 2 3). In Sheelar sediments, OM content was fairly constant, with an average of 24%. Organic matter in the Pebble core averaged 20% from 4 to 58 cm depth, but below that OM increased to 80% in the peaty, organic rich sediments formed the base of the core. Percent OM in the Little Lake Johnson core averaged 48% in the top 26 cm, but declined down core to only 2%, with the base of this core largely composed of sand. 2.3.3 Sediment G eoche mistry Table 2 1 lists the mean concentration, standard deviation, and maximum value for Sc, V, Cr, Co, Ni, Cu, Zn, Ga, Ge, Rb, Sr, Y, Zr, Nb, Cd, Sn, Sb, Cs, Ba, Hf, Ta, Tl, Pb, Bi, Th and U in modern (~1900 to present) sediments, and the mean and standar d deviation for those elements in baseline intervals (pre 1900, unenriched deposits) of the Sheelar, Pebble, and Little Johnson cores. Baseline values are considered to be largely unaffected by anthropogenic influence. Metal variability is discussed from t he core

PAGE 36

36 bottom to the sediment surface and metals with similar stratigraphic profiles were grouped to display major trends. Complete data are in Appendix A Table A 1. 2.3.3.1 Lake Sheelar sediment core Based on obtained results, metals were subdivided in 3 groups based on concentration trends. The first group of metals, group 1, included V, Cr, Ni, Cu, Zn, Sn, Sb, Pb and Bi ( Figure 2 4). These metals maintained relatively constant concentrations from the core bottom, but displayed increased concentrations between the 48 46 cm and 26 24 cm intervals. Many of these metals increased steadily to the core top. Pb increased and peaked in the 12 10 cm interval, then decreased slightly at the top of the core. Zn concentrations increased gradually towards the co re top with an extreme increase in the surface sample. Group 2 metals included Co, Ga, Ge, Zr, Nb, Cd, Hf, and U. These metals displayed little variability in concentration over the whole length of the core. Ga, Zr, Nb, Hf, U, however, show slight variab ility and trend together. Group 3 metals included Sc, Rb, Sr, Y, Cs, Ba, Ta, Tl, and Th. They showed variable concentrations throughout the core and generally trended together. 2.3.3.2 Pebble Lake sediment core Metal concentrations recorded in the Peb ble Lake core were also divided into three groups. Group 1 consisted of V, Cr, Co, Ni, Ga, Zr, Nb, Sn, Sb, Sc, Hf, Ta, Tl, Pb, Bi, and U ( Figure 2 5). Many of these elements exhibited their lowest concentrations at the base of the core and increased grad ually to higher concentrations ~66 cm depth, but remained relatively constant from 62 58 cm to the 16 14 cm interval. At 16 14 cm the concentrations increased, then peaked between 10 and 6 cm depth, and decreased slightly at the core surface. Group 2 elem ents included Cu, Zn, Cd, and Sb. They

PAGE 37

37 showed a pattern slightly different at the base and surface than did Group 1 trace metals. Concentrations for group 2 were relatively high and then decreased to relatively low and constant values from the 62 to 14 cm and then increased again toward the core top. Group 3 elements included Y, Rb, Th, Cs, Sc, Sr, Ba and showed relatively variable concentrations, varying as much as two fold in concentration from one sample depth to the next, but generally trended together. Concentrations were relatively low at the base of the core and peaked at 66 64 cm and again at 6 4 cm. 2.3.3.3 Little Lake Johnson sediment core I divided trace metal concentrations in the Little Johnson sediment core into two main groups. Group 1 metal s, comprising most of the measured elements, displayed low and fairly constant concentrations in the lower part of the core, 54 to 34 cm, and increased above that depth. The rate of change of concentration differs among the elements, many of which exhibit the steepest increase and peak concentrations in the top 10 cm. Elements in Group 1 included V, Cr, Co, Ni, Cu, Ga, Rb, Sr, Y, Zr, Nb, Cd, Sn, Sb, Cs, Ba, Hf, Tl, Pb, Bi, Th and U ( Figure 2 6). Group 2 elements Sc, Zn, Ge, and Ta displayed fairly constan t concentrations throughout most of the sediment core, but increased from 8 cm to the core top. 2.3.4 Normalization and E nrichme nt Factors (EFs) of Trace M etals Normalization and calculation of EFs required that a conservative element be chosen. This element must reflect geogenic sources, have no anthropogenic source, and be unaltered by post depositional biogeochemical processes. Sources of these elements can in clude weathering of local lithology or longer range transport of other mineral soil dust. Commonly used conservative elements include Al (Tylmann 2005)

PAGE 38

38 Sc (Shotyk et al. 2002; Weiss et al. 1999) and Ti or Zr (Koinig et al. 2003) I selected Sc for both normalization and EF calculations, because the concentrations were fairly constant over the lengths of the cores c ompared to the highly variable concentrations of of zircons in a sample from a given interval). The area is mined for heavy, Zr rich minerals, so it was not surprising to find Zr in the lake sediments. Use of Zr for normalization, however, yields high variability in the record that obscures the changes in concentrations of other metals. For this reason, I selected Sc for normalization. In the Sheelar core, Sc normalized values exhibited similar trends to non normalized profiles, and Zn, Sn, Sb, Pb, and Bi still exhibited enrichment towards the top of the core. In the Pebble core, Sc normalization removed some of the variability seen in many elements, especially those in Group 3 (Y, Rb, Th, Cs, Sr, Ba). Normalized metals V, Cr, Ni, Cu Ba, Zn, Sr, and Cd, had relatively high, in some cases highest values, at the base of the core, then decreased and remained relatively constant toward the core top. I observed enrichment of Cr, Zn, Sn, Sb, Bi, and Pb towards the top of the core. Only Nb, Th, and Ta concentrations were relatively constant throughout the core. In the core from Little Lake Johnson, Sc normalization eliminated the increasing trends from ~38 cm to the core top se en for most elements. Sc normalized metal values increased from the 34 to 22 cm, however most remained relatively constant for the remainder of the core. Values of Zn, Sn, and Pb increased from 10 cm depth to the top of the core. Complete data can be foun d in Appendix A Table A 3. Enrichment factors (EFs) were calculated to determine the degree of anthropogenic contribution of metals. The calculations required defining the baseline

PAGE 39

39 concentration. For this study, baseline metal concentrations were defined as concentrations in sample intervals that pre date metal enrichment, i.e. before 1900. Development in Florida began in earnest after 1900 and population growth, along with industry, saw rapid growth only after the 1930s. Given the limits of 210 Pb dating I cannot pinpoint a common pre 1900 age in each core. I thus defined a unique baseline metal concentration for each core. Establishing baseline conditions in the Sheelar core was fairly straightforward because the relatively low metal concentrations wer e quite constant below 46 cm depth ( Figure 2 4). Element concentrations were averaged for intervals from 46 cm depth to the base of the core. Baseline concentrations in the Pebble and Little Lake Johnson cores were determined differently because metal con centrations in these cores did not reach an asymptotic value at depth. Both cores exhibited large changes in lithology, with deep sediments becoming more organic in Pebble Lake and more quartz rich in Little Lake Johnson. These changes in lithology may d ilute or concentrate metals. The sediment character change was accompanied by changes in metal trends, so my conservative approach was to compare modern metal concentrations to values in deep sediments of similar character to the modern (post 1900) sedime nt. Baseline metal concentrations were determined as averages of values at depths that pre date anthropogenic influence, i.e. before 1900, but are above depths at which there was significant change in sediment physical properties. Baseline concentrations w ere defined as the mean metal concentrations for the interval of 58 24 cm in the Pebble core and 26 10 cm in the Little Lake Johnson core, the shaded regions in Figure s 2 5 and 2 6. Although these baselines may include sediment deposited in the 19 th centu ry, relatively late development in Florida compared with other areas of the

PAGE 40

40 United States, suggests that anthropogenic influence on local or regional metal deposition was minimal at that time. Furthermore, this conservative approach accounts for Florida p re development concentrations and variability in watershed lithology that affects metal concentrations. The baseline means for most elements are comparable among the three lakes, with most variability likely a consequence of in lake characteristics. Strati graphic trends in trace metal EFs ( Figure 2 7; Appendix A Table A 2 ) were different from trends exhibited by concentrations. An EF<1.5 reflects natural variability of sample mineralogical composition (N'guessan et al. 2009) whereas an EF>1.5 can be interpreted as reflecting enrichment. In the Lake Sheelar core, metals that have peak enrichments in Zn, Pb, Sn, and Cd, followed by Sb and Cu. In the Little Lake Johnson core, EFs exhibited greatest enrichment in Pb, Sn, Bi, Zn, and V. 2.4 Discussion 2.4.1 Pre A nthropogenic Trace Metal D elivery Despite my inability to determine accurately the pre anthropogenic time period in each of the cores, the 210 Pb dates, along with historical information on development in Florida, enabled me to estimate trace metal concentrations in sediments that pre date local human disturbance of trace metal cycles. In the Sheelar core, pre anthropogenic deposits are below 46 cm depth, in the Pebble core, below 24 cm, and in the Little Johnson core, below 12 cm. The baseline means (Table 1) are similar among the lakes and slight differences reflect variability in the relative contribution of erosional material and atmospheric inputs to the basins. Pre anthropogenic trace metal delivery sources

PAGE 41

41 include ero sion and atmospheric dust, from local or more distant sources. All three lakes lie in similar geologic terrain of the Trail Ridge and Cypresshead Formations and as a consequence of their small catchment size, local erosion is limited to these sources. Th e influence of morphology and hydrology on recorded metal trends was also evident from these sediment lake records. In the Pebble Lake core, the change in sediment character from peaty, 80% organic matter at the base of the core, to ~20% OM from 54 cm to t he top of the core ( Figure 2 2), suggested that the lake was historically a shallow water, marsh system and over time became a true lake, receiving more mineral material from local weathering processes. The mineral component of ombrotrophic peat bogs or ma rshes comes primarily from atmospheric dust, with little to no input from local soil erosion (Shotyk et al. 2002) By contrast, lakes receive more geogenic material from local erosion. In this core, Zn, Cu, Cd, and Sb exhibited the highest concentrations at the base of the core and decreased to 66 cm ( Figure 2 5). Using trace metals recorded in a peat core, Shotyk et al. (2002) argued that enrichment in these elements reflects increased supp ly of airborne dust or the fine fraction of weathered rock to the bog. At the time when Pebble was a shallow, marshy environment, there was little to no supply of locally sourced, coarse heavy minerals. Instead, mostly fine dust particulates, enriched in C u, Zn, Cd, and Sb, were delivered to the basin. Sc normalization removes the effects of variable bulk density and/or changes in the amount of mineral matter. The Sc normalized Zn, Sr, V, Cu, Ni, Cd, and Sb values were high at the base of the core and decre ased toward the sediment surface, suggesting that the dust was enriched in these elements compared to the local

PAGE 42

42 geogenic sediments that later infilled Pebble Lake ( Figure 2 8). This can be compared to the Zr/Sc trend, which is relatively constant above a d epth of 114 cm. Zr and Sc are both thought to be abundant in the local sediments. The transition of Pebble Lake from a marsh system to a lake is supported by the shifts in Sc and Zr concentrations, as these metals are geogenically sourced and conservative during weathering. Zr, especially, is expected to be highly elevated in the heavy fraction of local sands due to the presence of zircons (Pirkle et al. 1984). The steep increase in Zr, nearly five times the concentration at the base and double the concentr ation in Lake Sheelar, and the increases in many of the other metal concentrations from the base of the core to the 66 64 cm interval, suggest increasing heavy mineral delivery to the lake over that time period. Stream input to Pebble Lake during high stan ds, as well as highly astatic water levels may be mechanisms for increased sediment delivery. Throughout the middle of the core and the transition to a lake system, Sc normalized values of Zn, Sr, V, Cu, Ni, Cd, and Sb decreased and leveled out, possibly a consequence of dilution by increased input of coarser grained, locally sourced geogenic sediment, rather than decreasing dust deposition. Concentrations of metals such as Zr, Sc, Sn, Ta, U and Th increase from the base of the core to 50 cm depth. These metals are most likely from local geogenic sources. In Sheelar, metal concentrations were relatively constant over the time period recorded by the core and suggest a relatively invariable lithology and mechanism of sediment delivery. This is supported b y Sc normalization that only slightly smoothed individual metal variability, but did not alter up core trends ( Figure 2 8). As mentioned, Lake Sheelar is relatively deep, with water supplied by groundwater and precipitation.

PAGE 43

43 Its lack of overland inflow and relatively stable water levels may limit delivery of locally sourced mineral material, compared to Pebble Lake. I assume that the dust that supplied material to the Pebble wetland, and continues to enter the lake, also entered Lake Sheelar. Much of the mineral component of Sheelar was probably sourced from atmospheric dust, having only changed in the recent past as a consequence of anthropogenic influence. These early elemental concentrations represent reference atmospheric metal concentrations for this region. Similar to Pebble, the metal concentrations in the Little Lake Johnson core co vary with sediment composition. In the Little Johnson core, trace metal concentrations increase from the 38 34 cm interval to the core surface. Sc normalization, howe ver, constrains this increase to only 38 to 24 cm, whereas values remain constant to the anthropogenically influenced portion ( Figure s 2 6 and 2 8). The Sc normalized metal profiles were similar to the pattern exhibited by the OM percentage, which changed from <10% below 38 cm, to ~50% from 38 cm to 24 cm, and remained at 50% to the core surface. This transition represents the inception of this lake and consequent increased organic matter production. Little Johnson has an overland input stream during high stands that may be a mechanism for water and sediment delivery. Sheelar and Pebble have small catchments with steep banks, creating relatively deep lake basins compared with shallow, pan shaped basin of Little Johnson. This arguably could limit sediment distribution, however changing water levels most likely increased sediment delivery to the lake basin. This is supported by the Sc normalized Zr, known to be concentrated in local sediments, that was relatively constant at the base of the core, and increa sed over the infilling of the lake, and remained relatively constant once the lake was established

PAGE 44

44 ( Figure 2 8). There does seem to be a slight signature of atmospheric dust in the early Little Johnson sediments, but it was diluted by quartz mineral sedime nt. At the base of the core, Sc normalized Zn values were relatively high (similar to anthropogenically influenced values) and decreased as the other metal values increased. This is consistent with higher Zn values in the other lakes, hypothesized to be from Zn enriched atmospheric dust deposition. 2.4.2 A nthropogenic Trace Metal D elivery and T iming The trace metal trends from each study lake contain evidence of anthropogenic influence. In the Sheelar core, V, Ni, Cu, Zn, Sn, Sb, Bi and Pb concentratio ns increased in the upper section of the core and peak at or near the surface. In Pebble Lake and Little Lake Johnson cores, Sc normalization and EFs reduced up core increases in concentrations of most metals, but showed enrichment in Cr, Zn, Sn, Cd, Sb, Bi, and Pb in Pebble and Pb, Sn, V, Zn, Bi, Cd, and Sb in Little Johnson ( Figure 2 7 and Figure 2 8). All metals that showed enrichment may be attributed to anthropogenic activities, commonly as products of fossil fuel combustion (Nriagu and Pacyna 1988; Kamenov et al. 2009) V, Zn, Pb, Sn, Cd, Sb, and Bi are al l relatively volatile (Toutain et al. 2003) and readily enter the atmosphere during combustion. Copper and Ni are slightly less volatile an d therefore less readily transported in the atmosphere, perhaps explaining their lower degree of enrichment. Coal combustion releases a number of metals, including those enriched in these sediments, but in a recent study it was significantly correlated wit h Sn and Sb (Landing et al. 2010). The increase in Sn and Sb ~1900 may be associated with coal powered rail lines or regional coal fired power plants.

PAGE 45

45 In all three lake cores, the onset of increasing concentrations and EFs for all anthropogenically influen ced metals occurred slightly before 1900. This is in agreement with findings at Blue Cypress Marsh, east central Florida (Kamenov et al. 2009) and in 17 New England lakes, which showed that the first major increase in metals derived largely from atmospheric pollutants (Pb, Zn, Cu, V) occurred between 1870 and 1930 (Davis et al. 2006) Lacking good age constraints on sediments deposited before ~1900, I suggest that anthropogenic influence on the flux of these metal s coincided with the onset of the major human activities in Florida, beginning in the late 19 th century and continuing to present. The state experienced rapid population growth after 1930, followed by increased electrical production in the 1940s. Increas ing concentrations of trace metals toward the tops of the cores are consistent with these activities. 2.4.3 Metal I nventories and Accumulation R ates I hypothesized that because of the close proximity of these lakes to each other, they should display simila r anthropogenic metal accumulation if the metals were sourced only from atmospheric fallout. To test this, I calculated inventories for metals attributable to anthropogenic activities. Only sediment intervals from ~1900, or the oldest 210 Pb date, were us ed, representing ~110 years of metal accumulation. Baseline, mean concentrations (Table 2 1) were subtracted from the metal concentrations for these intervals, to remove the contribution from pre disturbance inputs and control for natural, inter lake vari density (g dry/cm 3 wet) and the interval length (2 cm) for each interval, and values 2 for the last ~110 ye ars (Table 2 2). I found an order of magnitude difference in sediment

PAGE 46

46 bulk densities, a consequence of lower percent dry mass in Sheelar compared to sediments in Pebble and Little Johnson. The inventories of metal accumulation in Lakes Pebble and Little J ohnson are comparable, but higher than the inventories for Lake Sheelar. Natural system variability could account for these discrepancies. For instance, Nort on et al. (1992) measured up to 20% variation in Pb accumulation rates in multiple cores from a single lake in New York State, noting variability in sediment character, e.g. amount of clay or organic matter. Individual watershed characteristics, such as catchment size, slope, soil type, faunal density and type, may also play a role in metal accumulation. During storms and other erosional events, soils and associated metals are moved into the lake from the catchment (Blake et al. 2003; Landre et al. 2010) I speculate that sediment focusing, given the contrasting morphometries of the basins, might be responsible for the differences in metal inventories. But the inventory for atmospherically derived 210 Pb was comparable among the three lakes (Table 2 2) and similar to the global average of 39.3 dpm/c m 2 (Appleby and Oldfield 1983), suggesting that sediment focusing could not account for differences in metal inventories between Sheelar and the other two lakes. Another possibility for the discrepancies among metal inventories is the extreme water fluct uations experienced by Lakes Little Johnson and Pebble. Drying and wetting, even of littoral sediments, could arguably alter metal delivery and accumulation at the lake center, or the ability of the sediments to preserve a continuous, unaltered record. Bo th of these lakes have surface inflows that potentially carry local geogenic material to the lake, along with metals that were atmospherically deposited in the watershed. If

PAGE 47

47 these processes had a significant effect on metal accumulations, however, I would expect to see large inter lake discrepancies in the 210 Pb inventories, but I do not. Water level fluctuations could potentially change redox conditions of the sediments, which in turn could alter element mobility in the sediments and pore waters and cause redistribution of the metals. The onset of increasing metal accumulation ~1900, and subsequent increases, especially in Pb, however, suggest that sediments of all three basins have preserved an accurate record of metal deposition. Additionally, similar E Fs for most of the metals across the three lakes suggest that the lakes have experienced similar changes in fluxes. Alternatively, the higher metal inventories in Lakes Pebble and Little Johnson suggest the possibility of a local metal source for those l akes. Both lakes are very close to the recreational facilities in the park. A large parking lot, camping sites and picnic areas are within a few hundred m of both lakes. Because the lakes are in a state park, far from urban and industrial activities, a st rictly atmospheric source for metals was assumed. The large metal inventories in these two lakes, however, suggest an additional input from local anthropogenic sources, only revealed because I had a sediment core from more remote Lake Sheelar. This suggest s that Lake Sheelar provides a more accurate record of atmospheric trace metal deposition in this region. I calculated the total amounts of metals that had accumulated in the sediments of each lake since ~1900. I also examined the accumulation rates of me tals in each lake over two time intervals. Because of the limited number of samples in the Pebble and Little Johnson cores, I calculated the average mass accumulation rate from ~1900 to 1970 and from 1970 to 2009 (Sheelar) or 2012 (Pebble and Little Johns on) (Table 2 2).

PAGE 48

48 Distinguishing these two time periods enabled us to examine changes in accumulation rate over the past century as population grew and development expanded in Florida. Not surprisingly, accumulation rates for metals in Pebble and Little Jo hnson were greater than those calculated for Lake Sheelar, over the two time periods. Average metal accumulation rates for the period up to 1970, however, were more comparable among the lakes than rates since that date. This suggests that all three lakes r eceived comparable metal inputs through the first two thirds of the century, but as localized recreational activity around Pebble and Little Johnson increased, so too did metal accumulation in those two lakes. Even though the rates for the later part of t he 20 th century are generally twice those for earlier times, they fall within the range of urban and reference lake sediments from across the United States (Mahler et al. 2006). 2.4.4 Anthropogenic Pb Concentration and Pb I sotopes Natural sources, in lake processes and post depositional physico chemical transformations of metals must be considered to constrain anthropogenically sourced deposition (Outridge et al. 2005; Augustsson et al. 2010) Such processes can include redox related concentration gradients, precipitation and co precipitation, cation exchange, and adsorption to organic matter and/or Fe and Mn oxides (Sigg et al. 1987; Tessier et al. 1996; Boyle 2001a; Pearson et al. 2010) as well as incorporation by plants and bioaccumulation in consumers B ecause lead is relatively immobile after sedimentation, it is not significantly influenced by post depositional processes, and is thus an ideal pollution indicator (Ault et al. 1970). Additionally, Pb has four stable isotopes ( 204 Pb, 206 Pb, 207 Pb, 208 Pb) that can be utilized to fingerprint Pb sources (Shirahata et al. 1980; Gallon et al. 2005; Escobar et al. 2013).

PAGE 49

49 In all three lakes there was evidence of anthropogenic influence on Pb accumulation. The initiation of increasing Pb concentrations occurre d before ~1900. This early increase was probably from burning of fossil fuels and smelting, and later increments probably came from gasoline additives, with the use of tetraethyl lead and tetramethyl lead commencing ca. 1923 (Norton et al. 1992; Davis et al. 2006) Initiation of accelerated Pb accumulation occurred between 1800 and 1850 in New England and about 1900 in Florida (Norton et al. 1992) I could not constrain sediment ages before ~1900, but the increase in Pb deposition began somewhat earlier than 1900 in Florida (Kamenov et al. 2009). In all cores, peak Pb concentration occurred below the surface, at 1 0 12 cm depth in Lake Sheelar and 4 6 cm depth in both Pebble and Little Johnson, and coincides with sediments deposited ~1975 (+/ 5) ( 2 9). Similarly, in Blue Cypress Marsh (Kamenov et al. 2009) and Little Lake Jackson, southern Florida (Escobar et al. 2013), Pb concentration peaked in the 1970s ( Figure 2 9). The Sc nor malized Pb concentrations from the lake sediments in this study do exhibit slight decreases in concentration at the surface ( Figure 2 8). This subsurface concentration peak, with slight declines at the surface, is common in Pb studies in the USA, with high er Pb concentrations in the 1970s than the 1990s or later. In Connecticut lakes, peak Pb concentrations occurred from 1971 to 1990 (Siver and Wozniak 2001) Similarly, peak Pb concentrations occurred in the Great Lakes during the early 1970s (Graney et al. 1995). Overall, peak sediment Pb concentrations coincide with peak leaded gasoline use. Subsequent concentration declines are associated with regulations that removed this additive (Shen and Boyle 1987) In some Florida lake cores, peak Pb

PAGE 50

50 concentrations range from 57 to 70 g/g and are within the range measured in other lightly impacted lakes (Mahler et al. 2006). Peak values in cores f rom these three study lakes were low compared to concentrations in other Florida lake cores, perhaps a consequence of the r elatively remote location of th is study and the proximity of high impact land uses in the other studies (Escobar et al. 2013; Sweets et al. 1990) A shift in Pb isotopes from non radiogenic before 1960 to more radiogenic after 1960 was identified in a number of studies ( Figure 2 8, Appendi x A Table A 3 ), and signified the appearance of a new source of anthropogenic Pb. The shift was attributed to the transition from relatively non radiogenic Pb from Idaho ore deposits to more radiogenic Pb from Mississippi Valley Type (MVT) (Graney et al. 1995; Kamenov et al. 2009). Use of MVT ore began in the 1960s and increased through to the 1980s, as reflected by higher radiogenic values. This shift was identified elsewhere in Florida, at Little Lake Jackson and Little Lake Bonnet (Escobar et al. 2013 ) and Blue Cypress Marsh (Kamenov et al. 2009) ( Figure 2 8). The shift is also synchronous with a Pb isotope record for mean atmospheric Pb deposition over the Caribbean (Desefant et al. 2006). In sediments from Lake Sheelar, this same shift is seen in 206 Pb/ 207 Pb and supports the conclusion that the lead deposited at the study site is influenced by the anthropogenic use of lead additive in gasoline. The return of 207 Pb/ 204 Pb to pre anthropogenic values in the topmost intervals reflects the phasing out o f leaded gasoline, however the 208 Pb/ 204 Pb has a distinct value at the core surface, indicating continued presence of anthropogenic Pb at this location. Total Pb concentrations support this interpretation, as they decreased only slightly at the top of the core, but are elevated relative to background levels.

PAGE 51

51 The 206 Pb/ 207 Pb values in the Little Lake Johnson core are fairly constant from 1900 to present, and in Pebble Lake there is only a very slight shift to higher values in the 1970s. This could be a resu lt of post depositional Pb mobilization or sediment mixing. However, neither of these two possibilities is supported by the 210 Pb data. Another explanation is that because of low sampling resolution the sampled intervals failed to capture the shift in isot opic ratios. Both Little Lake Johnson and Pebble have samples that fall above and below the large isotope shift from 1960 to 1980 ( Figure 2 9) and the samples that bound that shift fall within the appropriate range for the 206 Pb/ 207 Pb value for these date s. In Little Johnson there is a slight shift towards lower anthropogenic values ~1940, and in both lakes values increase below 1900, evidence against sediment mixing in this lake. Additionally, in both lakes, the Sc normalized Pb concentrations display inc reasing trends through the 1970s and decreases in the surface sediments, arguing that these sediments were not mobilized since deposition. 2.4.5 Anthropogenic Zn Zinc in atmospheric dust is often associated with fossil fuel combustion, but is also associ ated with vehicle tires, brake pads, and engine oil additives (Zn dialkyldithiophosphate) (Mckenzie et al. 2009; Fujiwara et al. 2011) The Zn concentrations in the Sheelar and Pebble c ores increased dramatically in the surface sample and EFs were 6.8 and 3.5, respectively, more than double the underlying interval. This drastic increase in surficial Zn concentrations was also measured in a peat core from Blue Cypress Marsh, Florida (Kam enov et al. 2009). Commonly with Zn increases, one might expect to see increases in other metals associated with vehicle emissions (Cu, Mn, Mo, Ni, Pb, and Sb). However, in a study by Kreider et al. (2010),

PAGE 52

52 the physical and chemical properties of particle s generated from the interaction between tires and road surfaces were examined and results showed Zn concentrations ranging from 3,000 9,000 ppm, two orders of magnitude greater than other trace metals. This suggests that the drastic increases could be att ributable to anthropogenic activities. Another hypothesis is that the Zn in surface samples may have also been influenced by diagenesis and metal mobility. The sediment water interface is dynamic, with chemical and biological activity. Metals often accu mulate to relatively high concentrations in topmost sediments, and compared to the water column, have a long residence time to undergo reactions (Zhang et al. 1995) Zinc and other chemically similar metals such as Cd, Pb, and Cu, have high affinity for humic and fulvic acids and Fe and Mn oxides (Carignan and Tessier 1985; Bilali et al. 2002) Previous studies investigated elevated Zn concentrations at the anoxic/oxic sediment water inte rface and attributed it to redox reactions. Under reducing conditions, Zn, one of the most mobile heavy metals, is mobilized from decomposing biological material and dissolution of oxides. In pore waters, metals diffuse through the sediment and precipitat ion is controlled by pH and the presence of stable sulfide minerals (Carignan and Tessier 1985; Norton and Kahl 1987; Hamilton Taylor et al. 1995 and 1996; Drever 1997) The sulfur concentration in these lakes is relatively low, averaging 0.25%. Calmano et al. (1993) found that the proportion of Zn, compared to Cu, Cd, and Pb, is lower in the sulfidic fraction of sediment and that Zn has the weakest binding stability, because of competitive adsorption onto surface sites by other metals ( Hamilton Taylor et al. 1997) Ad ditionally, Calmano et al. (1993) found that after repeated anoxic and oxic cycling, the proportion of Zn in the sulfide fraction decreases, but increases in the more easily

PAGE 53

53 reducible fractions such as metal oxides During reducing conditions, Zn will preferentially go into the pore waters that diffuse upwards, thereby increasing the Zn concentrations in the surface sediment. Ultimately I must question whether these sediment cores provided trace metal accumulation records that accurately reflect atmospher ic trace metal deposition to the study lakes. Other studies documented meta l migration within sediments, suggesting that metal profiles do not necessarily reflect metal deposition (Augustsson et al. 2010; Pearson et al. 2010) In contrast, it has been argued that if there is metal migration at the sediment water interface, it does not significantly influence the long term sediment concentration record (Boyle 2001a, 2001b; Sakata 19 85) Although it seems that Zn concentrations in the sediment cores did not accurately reflect metal deposition at the sediment surface, the onset of enrichment in all the lakes, pre 1900, was consistent with the onset of industrialization in Florida, suggesting the trends reflect true metal deposition rates. The Pb concentration isotope records, which correspond well with the timing of leaded gasoline use, also indicated that these sediment records of trace metal accumulation accurately reflect trace metal deposition to the study lakes. 2.5 Conclusions Trace metal profiles in sediment cores from three north Florida Lakes provided information on local erosion and deposition, regional atmospheric trace metal deposition, and anthropogenic influence on trace metal accumulation. At the base of the cores, met al concentrations were controlled by sediment source and lake morphology. Natural sources of metals to these lakes include the local geology and atmospherically transported dust. Many metals do not show enrichment from the bottom

PAGE 54

54 to the top of the cores ( Sc, Li, Co, Ga, Ge Cs, Ba, Hf, Ta, Th, and U), indicating that they were not influenced by human activities. Metals including V, Cr, Ni, Cu, Zn, Sn, Sb, Bi, Pb show increased up core enrichment beginning ~1900, as a consequence of anthropogenic activities Increases in metal concentrations were a consequence of deposition of atmospheric dust containing high metal concentrations and coincide with population growth and development in Florida. One known source of these metals is fossil fuel combustion, recor ded by Pb concentrations and isotope records. Lead profiles in the sediments reflect the increased use and later prohibition of Pb additives in gasoline, with peak concentrations in the mid 1970s and declines by the 1 990s. Pb concentrations in the study lakes were similar to concentrations measured in lightly impacted lakes elsewhere in the United States. Even though these lakes are in very close proximity to one another and are located in a relatively rural environment, localized anthropogenic acti vities may have resulted in increased metal delivery to two of the lakes, Little Johnson and Pebble. I suggest that more isolated Lake Sheelar may better reflect modern reference metal accumulations for this area. Trace metals measured in lake sediment co res from this remote location of north Florida, USA, provide records of; 1) the supply of pre anthropogenic metals, a baseline reference for this location, and 2) a record of modern metal accumulation governed by anthropogenically influenc ed atmospheric du st deposition.

PAGE 55

55 Figure 2 1. Map of Florida showing the location (star) of Mike Roess Gold Head Branch State Park. The detailed map indicates the park boundary (solid line), locations of Lakes Sheelar, Pebble, and Little Johnson, and roads (dashed line)

PAGE 56

56 Figure 2 2. Sediment depth (cm) versus 210 Pb dates (solid lines) and sediment accumulation rate (g cm 2 yr 1 ) versus 210 Pb dates (dashed lines) for the Lake on age calculations

PAGE 57

57 Figure 2 3. Percent organic matter versus depth in sediment cores from Lake Sheelar, Pebble Lake, Little Lake Johnson

PAGE 58

58 Sc V Cr Co Ni Cu Zn Ga Ge Rb Sr Y Zr Lake Sheelar Modern mean* 2.94 16.5 24.5 2.21 11.2 9.69 24.7 7.32 0.87 4.20 29.1 5.19 60.1 Modern S.D. 0.28 0.58 1.16 0.09 0.87 1.14 24.1 0.74 0.14 1.40 3.84 1.31 3.10 Modern max 3.46 17.5 26.9 2.37 12.8 12.0 118 8.94 1.09 6.27 35.7 6.64 65.0 Baseline mean^ 2.43 15.1 20.1 2.14 8.85 8.43 14.3 5.78 1.00 2.55 26.9 3.72 49.5 Reference S.D. 0.27 0.61 1.44 0.12 0.50 0.26 0.20 0.60 0.11 0.73 3.75 1.35 4.52 Pebble Lake Modern mean 2.65 20.3 27.2 1.96 13.8 12.4 33.2 10.8 1.18 7.28 32.6 7.31 82.4 Modern S.D. 0.84 5.12 6.56 0.29 3.13 3.26 18.3 2.94 0.14 4.03 9.31 2.18 8.44 Modern max 3.72 28.2 36.4 2.35 17.9 20.1 77.7 15.1 1.39 12.2 43.5 9.63 98.4 Baseline mean 2.59 17.2 22.9 1.85 10.6 11.2 21.0 8.12 1.03 7.70 32.2 8.64 92.2 Reference S.D. 0.24 1.58 1.92 0.24 0.99 0.86 1.38 0.74 0.04 1.46 2.08 0.59 9.78 Little Lake Johnson Modern mean 1.67 22.2 20.4 0.60 8.92 13.2 28.8 6.99 0.70 6.02 45.2 10.3 58.0 Modern S.D. 0.58 1.16 5.65 0.19 2.52 3.61 13.8 1.81 0.08 1.79 5.88 1.64 15.7 Modern max 2.28 23.6 26.6 0.87 11.8 17.2 45.7 8.93 0.80 8.20 51.9 12.2 77.6 Baseline mean 1.14 15.8 15.2 0.78 6.10 8.73 14.8 4.80 0.62 4.48 32.0 7.40 47.7 Baseline S.D. 0.22 2.09 1.39 0.16 0.53 0.61 0.86 0.76 0.04 0.56 5.19 0.96 5.40 Table 2 1. Statistical analysis of trace elements in the Sheelar, Pebble and Little Lake Johnson sediment co res All concentration values are g/g

PAGE 59

59 Table 2 1. Continued Nb Cd Sn Sb Cs Ba Hf Ta Tl Pb Bi Th U 5.23 0.07 1.30 0.20 0.39 46.4 1.67 0.41 0.08 38.9 0.23 4.45 1.61 0.36 0.01 0.60 0.04 0.18 10.0 0.08 0.12 0.01 11.9 0.04 0.45 0.13 5.84 0.10 2.08 0.24 0.66 58.9 1.81 0.85 0.11 56.9 0.27 5.40 1.79 4.76 0.07 0.40 0.12 0.22 49.6 1.33 0.34 0.07 18.7 0.14 4.04 1.53 0.56 0.01 0.04 0.01 0.09 8.24 0.12 0.08 0.01 1.80 0.01 0.81 0.18 8.74 0.09 1.38 0.21 1.22 76.4 2.34 0.63 0.19 40.8 0.21 5.96 1.57 0.91 0.03 0.65 0.06 0.60 23.2 0.23 0.06 0.06 22.2 0.07 1.60 0.30 10.2 0.13 2.26 0.30 1.97 110 2.72 0.72 0.31 71.1 0.31 8.40 2.01 9.87 0.06 0.81 0.15 1.27 98.3 2.67 0.73 0.17 19.5 0.15 6.47 1.55 0.78 0.01 0.04 0.01 0.19 9.06 0.27 0.08 0.07 1.66 0.03 0.44 0.09 7.17 0.15 1.64 0.22 0.99 98.5 1.70 0.53 0.17 41.0 0.25 5.77 2.25 1.44 0.05 0.98 0.02 0.25 7.45 0.38 0.10 0.04 23.1 0.07 1.23 0.34 8.74 0.21 2.95 0.25 1.28 109 2.18 0.63 0.22 66.3 0.33 7.21 2.64 6.36 0.08 0.68 0.15 0.73 74.5 1.44 0.48 0.10 14.9 0.12 4.68 1.57 0.63 0.01 0.12 0.03 0.10 13.3 0.16 0.05 0.02 2.96 0.02 0.45 0.21 Modern mean is the average concentration of values from intervals spanning from the earliest 210 Pb date (~1900) to the core surface. ^Baseline average is the average concentration of values in intervals below a depth considered influenced by anthropogen ic activities, but above a depth where there are large changes in sediment properties. Baseline values come from the following intervals; Sheelar 80 48 cm, Pebble 58 26 cm, and Little Johnson 26 14 cm. The bolded paired modern and baseline means are signi ficantly different based on the Wilcoxon rank sum test.

PAGE 60

60 Figure 2 4 Concentration profiles of Sc, V, Ni, Cu, Zn, Sr, Zr, Sn, Sb, and Pb in the Lake Sheelar sediment core. The shaded area indicates the segment of the core from which sample s were used to calculate average baseline concentrations. The line at 24 cm depth marks 1885. Note change in scale at 5 g/g

PAGE 61

61 Figure 2 5 Concentration profiles of Sc, V, Ni, Cu, Zn, Sr, Zr, Sn, Sb, and Pb in the Pebble Lake sediment core. The shaded area indicates the segment of the core from which samples were used to calculate average baseline concentrations. The line at 10 cm depth ma rks 1900. Note change in scale at 5 g/g

PAGE 62

62 Figure 2 6 Concentration profiles of Sc, V, Ni, Cu, Zn, Zr, Sn, Sb, and Pb in the Little Lake Johnson sediment core. The shaded area indicates the segment of the core from which samples were used to calculate average baseline concentrations. The line at 10 cm depth marks 1900. Note change in scale at 10 g/g

PAGE 63

63 Figure 2 7 Enrichment factor (EF) profiles for selected metals from the Lake Sheelar, Pebble, and Little Johnson sediment cores wi th selected 210 Pb dates

PAGE 64

64 Figure 2 8 Sc normalized trace metal concentrations for V, Cu, Zn, Sr, Zr and Pb in the Lake Sheelar, Pebble Lake, and Little Lake Johnson sediment cores. The line at 26 cm (Sheelar) and 10 cm depths (Pebble and Little Johnson) represent 210 Pb dates of 1885 and 1900, respectively

PAGE 65

65 Sheelar Pebble 110 yr inventory a Mea n annual accum. rate 1885 1970 b Mean annual accum. rate 1970 2009 110 yr inventory Mean annual accum. rate 1900 1970 Mean annual accum. rate 1970 2012 g/cm 2 g/cm 2 g/cm 2 g/cm 2 g/cm 2 g/cm 2 V 4.03 0.034 0.033 38.2 0.27 0.46 Cr 12.5 0.10 0.093 51.5 0.31 0.71 Ni 7.1 0.048 0.075 29.0 0.17 0.40 Cu 4.01 0.020 0.057 17.9 0.066 0.32 Zn 22.9 0.10 0.31 120 0.52 2.00 Cd 0.02 0.0003 0.00001 0.26 0.001 0.004 Sn 3.07 0.018 0.037 5.62 0.031 0.082 Sb 0.27 0.002 0.002 0.53 0.003 0.007 Pb 64.9 0.39 0.80 200 1.17 2.83 Bi 0.34 0.003 0.003 0.62 0.004 0.008 Excess 210 Pb (dpm/cm 2 ) 34.9 --40.1 --Table 2 2. Inventories g/cm 2 ) of selected anthropogenic metals for the last ~110 years, excess 210 Pb (dpm/cm 2 ) inventories (~1900 to present) and mean annual accumulation rates for metals during the intervals from 1885 (Sheelar) or 1900 (Pebble and Little Johnson) to 1970 and 1970 to 2009 (Sheelar) or 2012 (Pebble and Little Johnson)

PAGE 66

66 Table 2 2. C ontinued Little Johnson Reference accum. rate 1970 1990 c 110 yr inventory Mean annual accum. rate. 1900 1970 Mean annual accum. rate 1970 2012 g/cm 2 g/cm 2 g/cm 2 g/cm 2 V 23.9 0.17 0.29 Cr 27.4 0.11 0.48 0 4.1 Ni 14.4 0.064 0.23 0.2 8 Cu 22.2 0.11 0.35 0 14 Zn 76.6 0.36 1.23 1 14 Cd 0.3 0.001 0.005 0.01 0.16 Sn 5.2 0.022 0.087 Sb 0.29 0.002 0.004 Pb 136 0.66 2.14 0 30 Bi 0.59 0.004 0.008 Excess 210 Pb (dpm/cm 2 ) 44.7 ---a To calculate inventories, baseline, mean metal concentrations (Table 2 1) were first subtracted from (g dry/cm 3 wet) and the interval length (2 cm) for ea ch interval, and values were summed back to ~1900. b To calculate the accumulation rates, inventories for the given sample intervals were divided by the number of years in the interval. In cores for which there was not a 210 Pb date of 1970, samples above and below were used to calculate the concentration at that time, assuming constant accumulation. c Ranges of metal accumulation rates in reference (background) and lightly urbanized lakes (Mahler et al. 2006)

PAGE 67

67 Figure 2 9. Pb concentration a nd 206 Pb/ 207 Pb isotope ratios versus date in sediment cores from Lake Sheelar, Pebble Lake, Little Lake Johnson. 206 Pb/ 207 Pb isotope ratio versus date in sediment cores from Lake Sheelar, Pebble Lake, Little Lake Johnson, Little Lake Jackson, south central FL (Escobar et al. 2013), and Blue Cypress Marsh, southeastern FL (Kamenov et al. 2009)

PAGE 68

68 CHAPTER 3 IMPLICATIONS OF TRACE METAL DISTRIBUTIONS IN FLORIDA LAKE SEDIMENTS DURING EXTREME LAKE LEVEL LOWS 3.1 Introduction S ediments are sink s for trace metals that enter lakes from the atmosphere and in input waters. Accumulating lake sediments typically remain undisturbed and trace metals within them become permanently buried. H owever, declines in lake level may make these lacustrine sediment deposits sources of le gacy metal pollution. This project was designed to constrain th e indirect effect of climate change (drought) on trace metal fate and transport in sediments of shallow Florida lake s Dec lining water levels are associated with sediment transport and drying, as well as organic matter decomposition, factors that influence the fate of trace metals in the sediments. I assessed the effects of the severe 2012 winter spring dry period on two shallow lakes in north Florida, Little Lake Johnson and Lake Geneva ( Figure 3 1). I constrained the distribution of trace metals in exposed lake bottom sediments during the period of historic water level lows u sing total metal concentrations, grain size distribution and sediment lithology R esults from this study have environmental management implications, given predicted climate change scenarios. There has been extensive research on the fate of heavy and trace metals in the environment. Much of this work was prompted by health concerns associated with high concentrations of some metals in the environment (Batterman et al. 2011 ; Hargreaves et al. 2011 ; Luo et al. 2011) In many systems these metals are introduced through atmospheric deposition, making corrective management difficult. There is, however, some evidence for decreasing emission and deposition over the past few decades

PAGE 69

69 (Nriagu 1996; Siver and Wozniak 2001) With respect to undisturbed lake systems however historical ly accumulated metal s still reside in the sediments. This is perhaps best exemplified by Pb for which the peak concentration in lake sediments which coincides with peak Pb use as a gasoline additive in the mid 1970s, is typically several cm below the sediment surface (Shotyk et al. 1998 ; Siver and Wozniak 2001 ; Blair et al. in review). Blair et al (in review) show ed evidence for anthropogenically sourced metal deposition in north Florida. The primary source s of metals are thought to be atmospheric dust deposition from mining operations, power generating stations, and vehic le emissions. Additionally, local activities, including recreation in and around lakes, may be sources of metals. In the study by Blair et al. (in review) sediments from lakes in north Florida showed enrichment of V, Cr, Ni, Cu, Zn, Sn, and Pb above defined background concentrations, compared to conservative elements such as Sc and Nd. These same metals are the focus of this st udy. 3.1.1 Climate Change in Florida The predicted effects of climate change in Florida include increasing temperature, sea level rise, and changing rainfall patterns, all of which will influenc e lake hydrolog y throughout the state. Temperature data thr oughout the 20 th century across Florida show an increase in air temperature of 3 4 o C associated with a doubling of pre industrial atmospheric CO 2 levels (Martinez et al. 2012) Most research on climate change in Florida has focuse d on coastal impacts, especially sea level rise Florida has the longest coastline of any state in the contiguous United States, and because of its low coastal topography even slight increases in sea level will influence large land area s E stimates for future sea level rise range from 85 cm ( IPCC 20 07 ) to

PAGE 70

70 300 400 cm by 2100 ( Obeysekera et al. 2010) Rising sea level will also cause saltwater intrusion into freshwater aquifers in coastal regions (Ferguson and Gleeson 2012) P redictions for changing rainfall patterns in Florida show mixed outcomes because the scales of most climate models are too coarse to predict hydrologic changes for small regions (Mulholland et al. 1997 ; Martinez et al. 2012) In a study examining water resources in south F lorida, Obeysekera et al. (2010) found that many General Circulation Model (GCMs) do not capture the statistical characteristics of regional rainfall because of the poor resolution of grid cells. Model predictions are further complicated by uncertaint ies with respect to future behavior of the El Nino Southern Oscillation (ENSO), the Atlantic Multidecadal Oscillation (AMO), and the Pacific Decadal Oscillation (PDO), all of which a ffect precipitation on the Florida peninsula (Moses et al. 2012) GCMs suggest both lower and higher likelihoods of increasing annual precipitation, but agree that individual rain events will be more intense. Significant decreasing trends in October and May precipitation, and incr easing trends in July through August rainfall were measured in Florida over the last century (Martinez et al. 2012) Projections suggest increase d storm intensity and clustering in the summer months, resulting in extreme hydrographs, with large r peaks in flow but lower basefl ow conditions and longer periods of drought (Mulholland et al. 1997) Most models show that drought periods wil l be longer and more frequent (Yates and Towler, 2011).

PAGE 71

71 3.1.2 Indirect Effects of Climate Change Direct effects of climate change in Florida including increasing temperature, changes in the severity and frequency of droughts and precipitation, and shift s in the timing and character of seasonal precipitation continue to be defined. Little, however, is known about the indirect effects of climate change on humans and the environment. These indirect effects are defined as the interactions between climate c hange variables such as temperature or hydrolog y with ecosystems. Indirect effects also include the consequences of human activities that alter ecosystem characteristics e.g. channelization as a response to climate change (Mulholland et al. 1997) I nteraction s between changing rainfall patterns and pollution can thus be considered indirect effect s of climate change. I ncreased or decreased precipitation can alter the fate and transport of pollutants (Mimikou et al. 2000 ; Batterman et al. 2011; Vorosmarty 2000) The changing frequ ency and duration of precipitation events may mobilize and transport legacy contaminants thereby increasing the delivery of contaminants to water bodies (Kistemann et al. 2002; Hilscherova et al. 2007). The predicted increas ed duration of drought events i n Florida will also affect pollut ant transport. For instance, drying of soils increase s surface dust leading to increase d atmospheric particulates (Hargreaves et al. 2011; Zobeck and V an Pelt 2006) some of which negatively a ffect human health (Smith and Lee 2003; Luo et al. 2011) Greater soil temperature s and lower moisture will speed decomposition causing soil carbon to decrease (Nriagu 1996; Siver and Wozniak 2001; Smith et al. 2005) Organic matter in soils and sediments sequester s sor bed contaminants, such as metals, so a loss of organic carbon will alter cont aminant fate and transport (Shotyk et

PAGE 72

72 al. 1998; Grybos et al. 2007) F uture risks associated with these contaminants may be different from those today, and it is important to assess the implications of climate change with respect to how these chemicals might affect human and environmental health in the near term. 3.1.3 Florida Lakes Florida has about 7800 lakes, which cover at least 9270 km 2 about 6 % of the landscape ( Brenner et al. 1990). Three fourths of the 625 lakes in the Florida Lakes Data Base (FLADAB ) have maximum depths < 5 m ( Brenner et al. 1990). Because of the shallow nature of most Florida lakes extreme dry conditions lower water levels and often expose large areas of lake bed (Schif fer 1999) An estimated 70 % of Florida lakes are seepage lakes, losing water only to evapotranspiration and groundwater, not via drainage in surface out flows (Schiffer 1999; Martinez et al. 2012) The combined effects of low precipitation and intensified groundwater use, and the ever increasing water demand with population growth will further lower water table s causing additional stress on lake systems (Annable et al. 1996; Mulholland et al. 1997; Obeysekera et al. 2010 ; Ferguson and Gleeson 2012) Water level fluctuation is natural and even necessary for the survival of some species Climate driven changes in the frequency, duration, and severity of water level fluctuations, however will impair ecosystem functioning (Mulholland et al. 1997; Zohary and Ostrovsky, 2011) and a ffect water quality (White et al. 2008; Martinez et al. 2012) particularly in shallow lakes systems. Lake Geneva and Little Lake Johnson are strongly influenced by local groundwater levels and during the regional dry period in winter and spring 2012 the

PAGE 73

73 lowest water level s in recorded history were measured in both lakes ( Figure 3 2) which resulted in large expanses of exposed sediment (Figure 3 3). In March 2012 I sampled surface sediment along transects from the lake edge (at high stage ) to ward the lake center Surface s e diments were analyzed for total metal concentrations, grain size distribution and composition t o constrain the factors that influence metal distribution along this gradient. I hypothesize d that during dry periods lower lake levels lead to mobiliz ation of flocculent and fine grain sediments which accumulate in topographic lows and concentrate anthropogenically sourced trace metals The goals of my research were to 1) quantify the distribution of trace metals in exposed lake sediment; 2) determine grain siz e distribution in surficial lake sediment and assess its relation to metal concentrations ; and 3) evaluate implications of dry periods on metal geochemistry in lake sediments. T he rise and fall of lake level and mobility of sediment is a dynamic process T his study focused on metal distribution during historic low water levels. This study explored how drought conditions a ffect metal distribution s in sediments of shallow Florida lakes, and the implications of such distribution in light of anticipated future climate change s 3.2 Methods 3.2.1 Study Sites L akes Geneva and Little Johnson are located in the Sandhill Lake District of north Florida. Lakes in th e region are surrounded by thick quartz sands and in most cases occupy sinkholes formed by dissolution and collapse of underlying limestone. The drainage basins are small and water enters primarily via direct rainfall and subsurface inflow. Th e area is important for groundwater recharge and characterized by

PAGE 74

74 excessively well drained Entisols (Weatherspoon et al. 1989). Local bedrock consists of Pleistocene age Trail Ridge Sands and the Pliocene age Cypresshead formation, composed of several hundred feet of semi consolidated marine and non marine deposits of sand, clay, marl, gravel, limestone, dolomite and dolomitic limestone (FLDEP 2010; Obeysekera et al. 2010) Lakes i n the region have received much attention because of their astatic nature and recent pronounced declines in stage (SJRWMD 2013 b ). adjacent to the town of Keystone Heights and covers approximately 6.6 km 2 with a wat ershed of 117 km 2 The city of Keystone Heights stretches along the north shore and maintains a public beach and park, wh ereas private residences surround the remainder of the lake. This oligotrophic shallow groundwater fed lake has displayed very low wat er levels since the early 1990s leaving many private docks 10s of meters from the water This study focuse d on the norther nmost area of the lake, which has a public access. Because of recent low water levels, this area has been disconnected from th e main water body to the south. I collected sediments from transect 1 on the southwest side and transect 2 on the southeast side of this northern sub basin of the lake ( Figure 3 1). Head Branch State Park, five miles NE of the town of Keystone Heights. It is a relatively small and shallow oligotrophic to mesotrophic lake. It has an average maximum depth of 4.5 m and an area of 0.11 km 2 Groundwater fed Gold Head Branch Stream flows into Little Lake Johnson and at high stage the lake connects with larger Lake Johnson. In 1957 a dam was constructed to stabilize water levels for recreational activities, but it

PAGE 75

75 was removed in 2002 to restore the natural hydrology (Motz and Heaney 1991; Moses et al. 2012) I collected one northeast trending transect starting on the southwest side of the lake ( Figure 3 1). 3.2.2 Sample Collection I used acid cleaned 4x water rinsed 20 ml plastic vials for sample collection and collected surface sediments at Lake Geneva on 3 March 2012 and at Little Lake Johnson on 29 March 2012. Both lakes were at the ir lowest rec orded water levels ( Figure 3 2), with no standing water at Little Lake Johnson and only a few small disconnected pools in the north basin of Lake Geneva. At each lake the approximate location of a previous lake level highstand was determined visual ly The location was identified by the presence of dried, dark organic matter on top of quartz sand. At Lake Geneva I collected samples every 10 m along the two 50 m transects. A long both of these transects, the 50 m sample was collected at the water At each sampling location three samples were collected, each 1 m from each other perpendicular to the transect. Sample A was taken o n the transect line whereas samples B and C were taken 1 m left and right of the transect (downslope), respectively At Little Lake Johnson I collected samples every 10 m over a 100 m transect. Along this transect, three samples were collected at each location as described above. At each sampl ing location the 20 ml vial was pressed into the surface sediment and the to p ~3 cm of sediment w ere collected. Samples were place d in a cooler S ediment pH was measured with in 48 hours of collection using a 1:1 mix of DI water and sediment.

PAGE 76

76 3.2.3 Dry Mass and Organic Matter Content I used acid cleaned containers and Optima grade reagents for sample preparation and analysis. To determine percent dry mass, I weighed wet samples and re weighed them after freeze drying. Dry samples were ground to a fine powder. I estimated percent organi c matter for dry samples by loss on ignition (LOI) at 550 C ( Hka nson and Jansson 1983 ; Martinez et al. 2012) A ll elements of interest have boiling points >550 C, so ashing caused no elemental loss. 3.2.4 Grain Size Analysis One sample from each sampl ing location along all three transects was wet sieved to determine the fraction s of sediment larger and smaller th an 63 m. I treated the < 63 m fraction with hydrogen peroxide to remove organic matter and determine percent clay/silt. The difference between the organic matter fraction estimated by LOI and hydrogren peroxide treatment was <10%. 3.2.5 Trace Metal Anal ysis of Total Digested Sediment I completed preparation and total digestion of all the samples from the three transects for geochemical analysis in a class 1000 clean lab, equipped with class 10 laminar flow hoods, at the Department of Geological Sciences, University of Florida. All reagents used for sample preparation were Optima grade. About 0.05 g of sediment was weighed, transferred to acid cleaned Teflon vials, and digested with 1 ml concentrated HF and 2 ml concentrated HNO 3 S amples were place d in a 90 o C oven for 48 hours. The vials were opened and the solution was evaporated to dryness on a hot plate. After evaporation samples were treated with 2 ml of 6 N HCl and vials were

PAGE 77

77 capped and heated on a hot plate overnight to ensure full dissolution. The solution was evaporated again to dryness. Five ml of 0.8 N HNO 3 with 100 ppm HF spike with 8 ppb Rh and Re was added to the samples to re dissolve the residue. A small aliquot of the solution was removed and diluted with the same spiked 0.8 N HNO 3 f or a final dilution of ~2,000x for trace metal analyses. Trace element analysis was performed on a Thermo Element2 HR ICP MS in medium resolution with Rh and Re used as internal standards. I calibrated results using USGS rock standards AGV 1, BIR, BCR 2 an d a procedural blank following Kamenov et al. (2008) Duplicate samples were analyzed and results showed <10% variability. Meta l concentrations are reported as ppm per gram dry sediment 1 ) Errors for reported concentration values we re between 4 17% based on replicate measurements of Canadian Certified Reference Materials Project lake sediment standard LKSD 4. Table 3 1 sh ows recovery error percentage for each metal reported. 3.2.6 Statistical Analysis All statistical analysis was conducted with JMP Pro Version 11.0. Prior to statistical analysis, I tested data distributions of sediment geochemical and physico chemical characteristics for normality. I used the Shapiro Wilks test to test for normality and found that the majority of distributions were not normal. To test for significant differences between lakes, I used the Wilcoxon text, which is similar to the Kruskal W allis test, which is the non parametric equivalent of a two sample t test. Significance was determined at the p < 0.05, and 0.01 levels. associated significance to test for correlations between sediment characteristics and concentrations.

PAGE 78

78 3.3 Results 3.3.1 Water L evels Lake Geneva generally had higher water levels than Little Lake Johnson before the late 1980s. After that time, however, Little Lake Johnson had greater water elevations. During sediment sampling in March 2012, water levels in both lakes were extremely low (~81 ft NAVD 1988). Relative to the periods of record for Lake Geneva (55 years) and Little Lake Johnson (67 years), this was similar to the minimum historic lake elevation (Table 3 2). Past water levels did not fall this low often and there have been more lake level lows after 1990. These water level minima occurred <2% of the time during the period of record ( Figure 3 4 ), indicat ing that these extreme low water levels are rare. 3.3.2 Sediment Charact eristics Sediment organic matter content was much greater at Little Lake Johnson than Lake Geneva (Wilcoxon test; p<0.01). Median percent organic matter was 33% at Little Lake Johnson, while it was about half that (13%) in Lake Geneva. Percent water weight and organic matter tended to increase along transects from the shoreline (0 m) toward the lake centers ( Figure 3 5 ). Organic matter in Little Lake Johnson increased gradually to about 50 m from the historic shoreline, after which it doubled and remained h igh (45 50%). At Lake Geneva, the pattern was somewhat different. Organic matter content increased gradually with distance from the shoreline and sediments collected farthest from shore had only about 18% organic matter.

PAGE 79

79 Sediment collected in Lake Geneva w as generally wetter than the sediment collected in Lake Johnson. Wetness increased from about 8 to 56% with distance from the shoreline at Lake Geneva, whereas water content of sediments collected in Lake Johnson was relatively uniform, typically <10%. 3.3 .3 Sediment G eochemistry Physico chemical and geochemical characteristics of Little Lake Johnson and Lake Geneva sediments are presented in Table 3 3. Lake Geneva sediments had higher pH, greater percent water and ash content, but much less organic matter than Little Lake Johnson sediments (Wilcoxon test; p<0.01). Median organic matter content of Lake Geneva sediments was 13% (n=36), whereas the median organic matter content of sediments collected in Littl e Lake Johnson was 71% (n=33). Metal concentrations in sediments were quite similar between lakes. The only metals that differed significantly (Wilcoxon test; p<0.05) between lakes were V and Zn. Average pH, percent water and organic matter values for triplicate samples collected at each location along the three transects are shown in Table 3 3 Distance from shore is the distance from the historic shoreline (0 m) along the transect extending into the center of the lake. The pH values of sediments from Little Lake Johnson ranged between 4.73 and 5.30, with no systematic trend. The pH values of sediments collected along transect 1 at Lake Geneva were lowest at the historic high water line (4.92) and increased to 6.09 at 50 m. Sediments from transect 2 showed a similar increasing pH trend moving away from th e old high water line, from 5.25 to 6.04. The average total metal concentrations for sediments collected from the three transects are shown in Table 3 3 and Figure 3 6 All metal concentrations measured in

PAGE 80

80 sediments from Little Lake Johnson were lowest close to the historic shoreline (0 m) and increase d with distance from shore. Metal concentrations generally reach ed greatest values around 60 m from the shore line and remain ed relatively constant to the 100 m sampling location. T here wa s a 2x or greater increase in concentrations of V, Cr, Ni, Cu, Zn, Sn, Sc, and Nd concentrations and a more than 3x increase in mean Pb concentration along the transect. S ediments collected from the two Lake Geneva transects show ed slightly different pat terns from one another and from the Little Lake Johnson transect In transect 1 sediment metal concentrations were lowest closest to shore, increasing with distance from shore, and peak ing at the farthest location. Mean concentrations of Cr, Zn, and Sc increase d 3x, Ni, Cu, and Nd increase d > 4x, and V, Sn, and Pb i ncrease d >5x along the transect. In transect 2 there wa s no systematic increase away from the shore line however for most metals, the lowest value wa s at the historic shoreline and the metal c oncentrations were comparable to the higher concentrations measured in transect 1. I used the t test to assess relationships between metal concentrations along each transect in each lake ( Table 3 4). I found significant co rrelations between all anthropogenically sourced metals (V, Cr, Ni, Cu, Zn, Sn, and Pb) in all three lakes. In general, correlations between anthropogenic metals are stronger than those between anthropogenic and conservative metals. Along Lake Geneva trans ect 2, there was also no significant relationship between Sc and Cu, Nd and Cu, Nd and Pb (r range = 0.52 0.58).

PAGE 81

81 3.3.4 Grain size and Organic Matter In Little Lake Johnson sediments <63 m and organic matter increased with distance from shore, wh ereas the >63 m fraction percent decreased ( Figure 3 7 ) Values for each variable were relatively constant from 60 to 100 m. For example, organic matter content was <20% at the shoreline and increased to about 25% at 40 m, but was about 45% from 60 100 m. In t he sediments from Lake Geneva, organic matter content was much lower. At the old shoreline of transect 1, percent organic matter was <4%. With distance from the edge, however, organic matter and the < 63 m fraction increased, while the > 63 m fraction decr eased by 50% from the historic shoreline (0 m) to the farthest point along the transect. In sediments from Lake Geneva transect 2, organic matter content silt and clay fraction s were smallest closest to the historic shore line but values were relatively c onstant from 10 to 50 m with organic matter ~15%, fine fraction ~35 % and the >63 m fraction ~50%. Correlation coefficients between individual metals and organic matter content in sediments from Little Lake Johnson and transect 1 at Lake Geneva were sign ifi cant, with r values between 0.75 and 0.91 (Table 3 4). There was, however, no significant relationship bet ween organic matter and Sn, Pb, Sc Nd in transect 2 from Lake Geneva (r values between 0.29 and 0.58) I also investigated the relation between gr ain size and trace metal concentrations. Grain size was analy zed on only one sample at each sampling location on each transect, so data from all three transects were pooled (n=23) for this statistical analysis (Table 3 5 ). Correlations between individual trace metals and the <63 m fraction were all positive and significant (r = 0.61 0.92). Correlations

PAGE 82

82 between individual metals and >63 m fraction were negative and significant except for Zn (r = 0.45 ; p<0.01 ) and V (r = 0.44; p<0.01) 3.4 Discussion 3 .4.1 Metal Variability Mean concentrations of Sc and Nd in sediment samples are very similar between the two lakes (Table 3 3). These conservative elements reflect geogenic sources, have no anthropogenic source, and are unaltered by post depositional biog eochemical processes. Sources of these elements can include weathering of local lithology or longer range transport of other mineral soil dust. Sc is a common conservative element (Mulholland et al. 1997; Weiss et al. 1999; Mimikou et al. 2000; Shotyk et al. 2002) and Nd, a rare earth element ( REE ) also behaves conservatively (Kamenov et al. 2009) The two study lakes sit in the same geologic terrain so it was expected that metals derived from the local geology would display similar concentrations. The metals commonly attributed to anthropogenic activity (V, Cr, Ni, Cu, Zn, Sn, and Pb) also display similar mean concentrations in sediments of the two lakes. Most of these metals are enriched in recent deposits of lakes in this region of Florida Blair et al. (in review) calculated enrichment factors above background (pre a nthropogenic) levels in sediment cores from three lakes in Goldhead Branch State Park, including Little Lake Johnson. An EF reflects enrichment above the natural background value and is 09). Peak enrichment factors were generally found within the top 10 cm of the sediment profile, and were V = 2.05, Cr = 1.07, Ni = 1.44, Cu = 1.71, Zn = 6.67, Sn = 3.92, Pb = 2.54. Because recent deposits in the Little Lake Johnson core show anthropogenic enrichment of V, Cu, Zn,

PAGE 83

83 Sn, and Pb and concentrations of these elements in surface deposits from Lakes Geneva and Little Johnson are similar, I suspect that these metals are also enriched in the most recent Lake Geneva deposits. At Little Lake Johnson these anthropogenic metals most likely came from atmospheric deposition, however nearby recreational facilities, including picnic, camping and parking areas, may also be sources (Blair et al. in review) Despite the proximity of down town Keystone Heights, lakeside residences, and a public beach area, concentrations of these anthropogenic metals are not higher in the sediments of Lake Geneva. This may suggest that this lake does not receive a significant amount of metal input from surr ounding land uses, and that atmospheric deposition is the main input source. It is also possible that metal concentrations continue to increase in areas still covered by water. Triplicate samples were collected along each of the three transects, and stand ard deviations of values varied depending on the metal and sample location along the transect (Table 3 3) Along all transects, t he standard deviation was generally higher in samples taken closest to the historic shoreline and decreased towards the lake ce nter. Some variation may simply reflect natural system variability. For instance, Norton et al. (1992) measured up to 20% variation in Pb accumulation rates in multiple cores from a single lake in New York State, noting variability in sediment character s such as the amo unt of clay or organic matter. I noted litholigic variability during sediment sampling along the t ransects. Samples collected at the historic shoreline (0 m) were composed mostly of quartz sand grains, and fine grain material (dark grey to brown, dry brittle clumps) was sporadically present.

PAGE 84

84 Among the three transects from the two lakes, overall trends for the metals varied. Along the transect from Little Lake Johnson, metal concentrations increased from the historic shoreline toward the center of the lake, but remained fairly constant beyond about 60 m. Values from sediments along transect 1 at Lake Ge neva increased rather consistently with distance from the old shoreline. Along transect 2, there was no systematic trend in the metal concentrations. These different patterns may reflect lake specific and even within lake, site specific responses of sedim ents and metals to declining water levels, but may also be a product of sampling strategy. Little Lake Johnson was completely dry at the time of sampling and the transect extended into the lowe st topographic area of the lake bed. At Lake Geneva, some wate r remained in the sampled sub basin and because I did not collect sub aqueous samples, I may not have sampled the full extent of the metal trend. Since the early 2000s, water levels at Lake Geneva have caused the northern part of the lake to become separa ted from the southern basin. Transect 2 was located in an area of the lake that was disconnected in the last 10 years, but does not include the historic lake shoreline. This may account for different trends between transects. Although trends in metal con centration differed among the three lake sediment transects, most metals behaved similarly along each transect. For instance, along the Little Lake Johnson transects, metals in the sediments increased to 60 m and then remained relatively constant thereaft er. Conservative metals Sc and Nd followed the same pattern as the anthropogenic metals. C ompari son of the anthropogenically sourced and conservative elements enables conclusions about the metal distributions along transects

PAGE 85

85 C orrelation s (Table 3 4) show e d significant relationships between all metal concentrations in sediments from Little Lake Johnson and Lake Geneva transect 1 There were weaker relationships between most metals associated with anthropogenic activity (V, Cr, Ni, Cu, Zn, Sn, and Pb) and c onservative Sc and Nd in sediments from all three transects Conservative metals are commonly bound in crystalline minerals and do not undergo diagenesis under natural conditions. Positive correlation between anthropogenic ally sourced and conservative meta ls suggests that the factors controlling metal concentration s in the sediment do not discriminat e between the two metal types Because conservative metals are bound in mineral structures, increasing concentrations along the transects suggest that these minerals are concentrated toward the lake center. L ack of correlation between conservative and anthropogenic metals suggests that factors that control metal distribution s in sediments affect the two metal groups differently. 3.4.2 Sedi ment Character and Metal Distribution C oncentration s and distribution s of trace metals i n wetland s ediments are determined by the interacting processes of sedimentation, adsorption, co precipitation, cation exchange, complexation, microbial activity, tempe rature, and plant activity (Matagi et al. 1998). In this study of lake sediments, plant activity can be excluded as a factor because at the time of sampling there were no plant s grow ing from the sediments. The extent to which the above processes a ffect met al distribution is the result of metal equilibri a with organic matter, clay minerals, and hydrous oxides, which are strongly affected by pH and redox potential (Foster and Charlesworth 1996 ; Matagi et al. 1998; Schulz Zunkel and Krueger 2009) All of these processes are mutua lly

PAGE 86

86 dependent, and it is difficult to determine which reactions are occurring I nvestigati on of each potential factor however, can constrain the mechanism(s) that control metal concentrations and distribution in the sediment. 3.4.3 Organic M atter Anthropogenic metals introduced into the aquatic environment occur in particulate form or are rapidly sorbed to particles They are deposited in sediments or remain as suspended particulate matter depending on the particle size and composition (Regnier an d Wollast 1993; Stecko and Bendell Young 2000). Therefore, sediment characteristics, namely organic matter and grain size distribution, are important factors that influenc e metal distribution. Organic matter plays a large role in trace metal cycling, by 1) complexing with metals and keeping them in solution, or 2) enhancing the association of metals with particulate matter by becoming adsorbed to the particulate surface (Hart 1982 ; Kabata Pendias 2001; Olivie Lauquet et al. 2001). A erobic and an a erobic incubation experiments on wetland soils found organic matter to be the primary driver in trace metal release from soils ( Grybos et al. (200 7) and high metal concentrations were associated with high organic content in s ediment from a river flood plain (Kruger et al. 2005). Sediments from Little Lake Johnson and transect 1 from Lake Geneva show significant positive relationships between al l metals and percent organic matter in the sediments (Table 3 4 ). This suggests metals a re bound to organic matter and that increasing organic matter in sediments nearer the center of the lakes accounts for the higher concentrations of metals. Furthermore, dry conditions closest to the historic high lake level probably caused organic matter degradation. O xidization and mineralization

PAGE 87

87 of the exposed deposits may have resulted in metal mobilization even via aeolian processes decreasing metal concentration s in these sediments (Calmano et al. 2005). In transect 2 from Lake Geneva there was no significant correlation between per cent organic matter and Sn, Pb, Sc, and Nd suggesting that other factors such as chemical variability, controll ed metal distribution. Although sediments from Little Lake Johnson contain more than twice the percent organic matter in sediments from Lake Geneva, the metal concentrations are comparable between the two lakes. This suggests that although there is a correlation with relative a mounts of organic matter present and increasing metal concentrations, higher organic content does not necessarily equate to higher metal concentrations. This is illustrated in Figure 3 8 with V and Zn concentrations and percent organic matter. The percen t organic matter in Little Lake Johnson is greater but the highest metal concentrations measured are slightly less. Concentrations reported in Table 3 3 are ppm per sediment dry weight and were corrected for the amount of organic matter measured using val ues obtained from LOI. Trends of metal concentrations along the transects were examined without this organic matter correction and results were very similar to the corrected values. This suggests that organic matter is not the only variable determinin g met al distributions along the transects. 3.4.4 Grain S ize Gunawardana et al (2014) confirmed that sediment characteristics favorable for metal adsorption, such as specific surface area, organic carbon content, effective cation exchange capacity and clay for ming minerals decrease with an increase in particle

PAGE 88

88 size. Metals have relative potential for sorption onto clay minerals, hydrous oxides, and organic matter surfaces, all of which tend to be found in the smaller grain sizes (Frstner and Patchineelam 1980 ; Calmano et al. 1993 ; Ikem et al. 2003) Generally, sediment of smaller grain size has decreased quartz content and higher clay forming minerals (Gunawardna et al. 2014). Because of large surf ace area surface chemical and mechanic al reactions of clay minerals are very important for collection and transport of inorganic constituents (Horowitz 1985) Not only the quantity of clay, but also the type of clay present controls metal adsorption, as each clay mineral has a specific cation exchange capacity. The correlation coefficient test showed significant relationships between grain size and metal concentrations. C orrelation coefficients were significant be tween the <63 m sediment grain size fraction and the trace metal concentrations T here were significant negative relationships between percent the >63 m grain size and metal concentrations, e xcept for Zn and V One explanation for the lack of relationship between grain size and Zn is the very low concentration of Zn in the local geology (Blair et al. in review). The >63 m grain size fraction is primarily composed of quartz sand. Zinc in this area is primarily attributed to anthropogenic source s and is likely bound to finer grain size s There is a positive correlation between the relative amount of fine grain material and metal concentration ; however the more fine grain material does not equate to higher metal concentrations. Organic matter an d fine particulate matter are often associated with one another S mall particles that make up the suspended matter primarily consist of clay and silica substrates that are coated with metal oxides and organic matter (Hart 1982) Early work

PAGE 89

89 by Tipping (1981) and Hunter (1980) su ggest s that an organic film present on oxide and clay surfaces has carboxylic acid and phenolic functional groups available for binding trace metals. Although I addressed o rganic matter and fine particulate matter separately in this study, they are genera lly associated with one another in lake sediments. Sediments from the Little Lake Johnson transect and transect 1 from Lake Geneva display systematic increas es in trace metal concentrations from the historic shoreline to the center of the lake. There are correlation s between anthropogenic trace metals, conservative trace metals, and both amounts of organic matter and fine grain material. Trace metals bound to flocculent material or organic matter and/or fine grained material, are more easily suspended in t he lake water column (N guyen et al. 2005) and become concentrated as water level fall s C onservative elements Sc and Nd present in lithogenic minerals undergo this same sediment mobili zation and concentrati on process and exhibit the same increasing concentration trends This same concentrating mechanism was seen in a partial drawdown of another shallow northern Florida lake in the spring of 1989. The St. Johns River Water Management District conducted a partial drawdown of Newnans Lake, FL (z max = 3.6 m) to expose organic rich littoral sediment in an effort to improve water quality. It was hypothesized that oxidation and compaction of the exposed littoral sediment would improve the establishment of littoral plant communities and th at resuspended nutrient rich flocculent or ganic matter would be flushed downstream through the outlet stream. R esults of this study showed that large amounts of organic mat ter w ere eroded from the littoral zone in some areas and there

PAGE 90

90 was evidence of material redistribution to the deeper portions of the lake (Gottgens and Crisman, 1992). 3.4.5 Other Factors Controlling Metal Distribution A lthough much of the variability in metal concentrations along the sediment transects can be attributed to the amounts of organic matter and fine grain sediment s, other factors can also control metal distribution s Gradients along the transects in pH and redox potential can affect metal phase s inhibiting or enhancing metal mobility within lake water, pore water and sediments (Salomans and Frstner 1984 ; Calmano et al. 1993) Most trace metals become increasingly mobile under acid ic con ditions and pH is a princip al factor governing metal solub ility (Brallier et al 1996). The pH values of the sediments ranged from 4.92 to 6.09. In sediments from Lake Geneva pH increased from 4.92 to 6.09 (transect 1) and 5.25 to 6.04 (transect 2) from t he historic shoreline to the lake center The decreasing pH in these low buffered oxidized sediments most likely resulted from the oxidation of sulfides (Calmano et al. 1993). Although pH along both transects increase d toward the lake center metal concentrations increased along transect 1 but were relatively constant over transect 2. If pH were a main control of metal concentrations, I would expect metal trends along the two transects to be similar There was no systematic change in pH along the Little Lake Johnson transect (Table 3 1), suggesting that pH had little role in the increasing metal trends a long this transect. Most cationic metals (Pb, Cd, Cu, Zn) become more mobile with decreasing pH (2 8) because the dissolution of metal hydroxi des and carbonates increases and the adsorption at sediment cation exchange surfaces decreases. In other studies a pH < 4.5 was necessary to mobili ze metals within soils (Calmano et al. 1993) A lthough pH can

PAGE 91

91 alter metal solubility it does not act independent ly on metal behavior. C omplex interactions between redox potential organ ic matter content concentrations of other ions (sulfate and chloride) and pH complicate interpretations. Trace metals are primarily distributed over clay minerals and fine grain sediments such as organic carbon, sulfides and/or iron and manganese hydro xides, which are all susceptible to phase changes under variable redox conditions (Hamilton Taylor and Davison 1995) The behavior of trace metals is to a large extent controlled by redox cycles of major elements such as N, S, O, Fe and Mn and the availability of organic matter (Shaw et al. 1990; Olivie Lauquet 2001) A transition from oxidized to reduced conditions can destabilize chemical phases and cause metal mobility within sediments (Calmano et al. 1993) Therefore, a large redox gradient over the length of a transect would result in different metal mobility at each site Redox conditions were not measured, however some inferences can be made regarding the role of redox in this study. All sediments were exposed to atmosphere and samples were limited to 3 cm d epth Although o xidizing conditions prevailed at the sediment surface, reducing conditions c an persist in wet conditions even at shallow depths in the sediment (Reddy and DeLaune 2008). It is likely that sediments closest to the historic high water line were most oxidized and reducing conditions increased with greater distance from shore and increasing water content. Within the measured range of pH, metals commonly complex with Mn and Fe oxides under oxidizing conditions. The hydrous oxides of Fe, Mn, an d Al are important for controlling the retention of metals in soils and individual metal ions have greater affinities to binding with oxides (e.g. Pb>Cu>Zn) (Trivedi and Axe 2000) Under

PAGE 92

92 oxidizing conditions these oxides become stable crystalline structure s immobilizing metals (Alloway 1997). Alternatively, under reducing conditions metals can be released into solution because of the dissolution of Fe and Mn oxides (Grybos et al. 2007) and metals are often present as stable metal sulfides. The redox conditions and individual metal characteristics de termine the rate at which the reactions occur (Trivedi and Axe 20 00). As water levels fell, exposing previously submerged sediments to atmosphere, conditions likely became increasing ly oxidizing. The process of oxidation releases sulfate and associated metals as cations (Chesworth et al. 2006). Nick el Cu, and Zn are considered relatively mobile because of this reaction (Reddy and DeLaune 2008). In studies that have measured metal diffusion between sediments and pore water, metals generally diffuse downward following the redox boundary and/or groundwater interface (Carignan and Tessier 1985) This process may explain the lack of correlation between the concentration of some metals and percent organic matter in transect 2 from Lake Geneva. Both pH and redox gradients play a role in trace metal distribution Nevertheless, predicting the behavior of a specific element under given conditions is not straightforward O ften basin specific conditions, reflecting variations in factors such as availability of competing scavenging phases (oxides, sulfides, and organic matter) and the specific chemical composition of the water and sediments can affect metal distributions (Hamilton Taylor et al. 2005 ; Schulz Zunkel and Krueger 2009) Finally, I consider whether metal trends measured in the sediments would be similar under high water conditions and whether declining water leve ls influence d trace

PAGE 93

93 metal distribution s In all lakes there are post depositional sediment transport processes, such as focusing. Waves, currents, and slope control this process of erosion, transport and deposition, which ultimately moves sediment to the deepest portions of the lake (Lehman 1975; Likens and Davis 1975). In shallow lakes focusing occur s, however sediment resuspension is common as wind driven waves create shear stress at the lakebed and return suspended sediment to the littoral zone (Luett ich et al. 1990 ; Tao et al. 2011) Wind and storm events can move large amounts of suspended sediment s around the basin increasing the conc entration of suspended particulate matter. This would suggest sediment and metal s become widely distributed under higher water levels in shallow lakes and suspended matter is entrained, altering metal distributions along the lake bottom under lower water conditions 3.5 Future I mplications The aim of this study was to examine the distribution of metals and sediment characteristics in bottom deposits of two north Florida lakes during a period of historic low water levels. L ake systems are dynamic and the chemical and hydrologic conditions of these lakes will change constantly and in turn alter metal distributions (Hamilton Taylor et al. 1996 ; Stecho and Bendell Young 2000; Anshumali et al. 2009) Nevertheless, this study suggests several environmental implications and raises several questions for future investigation. The i mplications pertain to th e indirect effect s of climate change on metal distributions in lake sediments particularly under dry conditions

PAGE 94

94 3.5.1 Sediment Quality Concerns It has been established that many trace metals are highly correlated to organic mat ter and fine grain particulates in lake sediments Water level fluctuations enhance sediment erosion and focusing, especially with respect to fine grained particles, fundamentally changing littoral sediment and its biogeochemical characteristics ( Gottgens and Crisman 1992) T his however, is much more evident in deep lakes (Boyle and Birk s 1999 ; Furey et al. 2004) The results of this study show that under extreme water level lows metal laden organic matter and fine grain partic ulates can be concentrated in the topographic lo ws of even shallow lakes. This may have adverse effects on bottom dwelling organisms or those that feed o n suspended sediment. There are also possible human health implications if water levels fall and expos e sediment s with high metal concentration s. Rhodes and Wiley (1993) report ed that declining water levels in the Great Lakes may cause contaminated sediments to be resuspended and/or exposed at the lake margins, representing potential lo ng term environmental remediation problems. In my study, metal concentrations d id not approach levels identified by the Florida Department of Environmental Protection as Soil Cleanup Target Levels in residential areas (V =67 mg/kg; Cr = 210 mg/kg; Ni = 340 mg/kg; Cu = 150 mg/kg; Zn = 26 000 mg/kg; Sn = 47 000 mg/kg; Pb = 400 mg/kg). I n urban lakes or water bodies that receive polluted runoff, it is possible that these concentrations may be measured. Also, even if sediments have been monitored prior to la ke level fall, higher metal concentrations may be measured after levels fall because of focusing of fine particulate matter.

PAGE 95

95 The lakes studied here are hydrologically closed systems, however in lakes with out flows, wind stress can re suspend fine grained particles and their associated metals, enabling them to be transport ed downstream In such lakes, higher lake levels combined with the windy conditions move metals out of the lake basin and into outflow rivers posing potential pollution risks for those downstrea m. 3.5.2 Water Quality Concerns Climate change projections for rainfall variability suggest longer drought periods alternating with periods of heavy precipitation, so long term continued atmospheric exposure of these and other lake sediments in Flor ida is unlikely. It is more likely that in these shallow lakes, this sediment will be mobilized and redistributed over the basin during water level highs (Nguyen et al. 2005) I t is also possible that during water level increases changing water chemistry, including pH and re dox conditions, will change the bioavailability of elements in these sediments. In wetland environments following drought periods stored sul f ur in the soil is oxidized generating SO 4 2 and during rewetting it is mobilized as H 2 SO 4 In lakes this process can cause peak metal concentrations in years with pronounced summer droughts because of metal mobility from soils in the lake catchment (Adkinson et al. 2 008) I ncrease d soil acidity releases sulfur bound metals back into the water column and pore waters. White et al. (2008) investigated the relationship between water level fluctuations and water qualit y in the Great Lakes and Boreal Shield regions and found correlation s between water quality variables (DOC, Ca 2+ conductivity, pH, and SO 4 2 ) and water levels. Variability of these factors can affect metal mobility between sediments and the water column. Again, metal

PAGE 96

96 concentrations measured in this study we re relatively low and do not pose concern, however the same cannot be said for highly contaminated lakes. 3.5.3 Sed iment D ust Dust, as an aerosol, significantly impacts the energy balance on Earth throu gh adsorption and scattering of radiation in the atmosphere a s recognized by the IPCC 4 th Assessment Report (IPCC 2007) which highlighted the net global cooling effect of aerosols. T he role of dust however, extends beyond th e impact on radiation. D ust may affect aqueous productivity, influence nutrient cycling, and alter regional and global climate (Tegen et al. 1996; Chadwick et al. 1999; Kawahata et al. 2000; Maher et al. 2010). Some studies indicate that mineral aerosols affect air quality and human health. This airborne particulate matter can lead to respiratory and cardiopulmonary morbidity and mortality (Pope and Dockery 2006). Aeolian dust is g enerated by human activities such as industrial emissions and by natural phenomena such as wind erosion o f soils (Tanaka and Chiba 2006). S oil derived dust is one the largest contributors to the global aerosol load (Tegen et al. 2002). Both sources can contain metals at levels high enough to be dangerous to both human and ecosystem health (Galloway et al. 1982) Playas and other typically dry areas, such as the southwest US, are primary source areas for soil dust because of sparse vegetation and low soil moisture (Pope and Dockery 2006). D ust emissions are also high in areas where recent geomorphological history has concentrated fine grain material and where there is low surface roughness (Tegen et al. 2002). Dry lake beds are excellent examples of these source areas (Pye 1987; Prospero et al. 2002). Lake bottoms have little to no vegetative cover and low soil surface roughness both limiting factors in dust dispersal

PAGE 97

97 because of the ir flat bottoms. Paleolake beds such as the Bodele depression north of Lake Chad Africa and others in central Australia are large modern sources of dust (Middleton 1984; Livingston and Warren 1996). D eclining water tables a consequence of changing climate patterns and increased groundwater pumping, will increase windblown dust emissions (Pelletier 2006). For exam ple, the dry Owens L ake bed has had the highest long term dust flux of any area in the U S since 1913 when water from the lake was diverted to Los Angeles, California (Reheis et al. 2002). Although Florida will not be an area of high dust emissions compared to the southwest U.S. climate change scenarios may alter the dust flux in the region R egional hydrologic modeling suggests that i f rainfall amount decrease s by 10% and temperature rises by 1.5 o C, there will be a significant reduction in water s upply for south Florida over a 50 year period. Because of decreased water supply and increased demand a 1 m lowering of Lake Okeechobee and 0.3 m mean decrease in water levels in natural areas, including the Everglades region is expected (Ob eysekera et al. 2010) Lakes cover ~6% and wetlands cover ~30% of the landscape in Florida and both ecosystems have accumulated anthropogenically sourced metals over the last 100 years (Kamenov et al. 2009; Escobar et al. 2013; Blair et al. in review). Atmospheric exposure of fine grain organic rich sediments in the systems may be a source for legacy metal pollution. 3.6 C onclusions I measured metal distribution s in transects of lake sediments in two north Florida lakes exposed during extreme dry co nditions. I n two of three transects m etals associated with anthropogenic activities V, Cr, Ni, Cu Zn, Sn, Pb, and conservative elements Sc and Nd exhibited increasing concentrations from deposits near the historic

PAGE 98

98 high water level to sediments near the center of the lake. Some basin specific differences in the metal trends were demonstrated, however significant relationships were determined between metals and some sediment characteristics. Conservative metals are often bound in the crystalline structure of minerals, wh ereas metals sourced from anthropogenic activities are often adsorbed onto sediments in hydrous oxides, organic matter, or sulfates. S ignificant positive relationships were found between anthropogenic and conservative metal concentrations, suggesting that the same mechanism(s) influence dispersal of mineral grains and adsorbed material. Metal concentrations were significantly correlated with percent organic matter and relative amount of fine grain material. I suggest that the main mechanism influencing metal concentrations along the sediment transects, was focusing of both fine grain material and organic matter toward the lake center when water levels fell. Th ese sediment components commonly found as flocculent material are easily suspended in to the water column, and as water levels fell, th ese components were concentrated in exposed deposits It is also possible that metals were mobilized within the sediments because of changing pH, redox conditions, and degradation of organic matter. Climate change scenarios for Florida predict more intense rainy periods alternating with more severe droughts. Falling groundwater combined with increased water need, will likely lead to lower water levels in shallow Florida lakes. Metal concentrations i n l ake and wetland sediments across the world have increased for the last century. The com bined effects of changing hydrology and relic metal pollution may create future management concerns including legacy metal pollution in sediments and water, as well as the potential for increased metal l aden dust.

PAGE 99

99 Figure 3 1. Maps of study locations. a) Map of Florida; star indicates study location. b) Map view of Lake Geneva and Little Lake Johnson. c) Outline of average water levels of Little Lake John son D ashed link indicates location of sampling transect with the measured distance from shoreline (0 100 m) marked d) Outline of average water levels of Lake Geneva with the location of the enlarged northern end of the sampled basin Fine dashed line on enlarged map was lake level in March 2012. Numbered dashed lines indicate location of sampling transect s 1 and 2 with the measured distances from shoreline (0 50 m) marked.

PAGE 100

100 Figure 3 2. Lake stage (ft NAVD88) for Lake Geneva and Little Lake Johnson sinc e 1957 and 1945, respectively. Lake stage measurement s from 2005 to 2013 are shown in the second panel and arrow indicates the date lake se diment samples were collected. Data from SJRWMD.

PAGE 101

101 Figure 3 3. Photographs of Little Lake Johnson (left) and Lake Geneva (right) during sampling events. Photo credit: Mark Brenner

PAGE 102

102 Lake Number of days in period of record Minimum Q1 Mean Median Q3 Maximum -----------------------------------ft NAVD 1988 -----------------------------------Little Lake Johnson 3371 80.6 92.3 93.8 94.0 94.5 103.3 Lake Geneva 7353 81.1 86.6 92.3 88.6 99.5 106.4 Table 3 1. Measured Canadian Certified Reference Materials Project lake sediment standard LKSD 4 concentrations compared with reported values Element Measured g/g Measured SD Reported g/g Reported SD Recovery (%) V 46.0 1.64 49 8 93.9 Cr 31.2 1.93 33 6 94.5 Ni 33.3 1.50 31 5 107 Cu 30.8 1.27 31 4 99.4 Zn 161 14.4 194 19 82.9 Sn 4.81 0.421 5 -96.2 Pb 101 4.49 91 6 111 Sc 6.73 0.548 7 -96.1 Nd 24.0 2.26 ---*n = 10; -= not reported Table 3 2. Summary statistics for water levels from Little Lake Johnson and Lake Geneva

PAGE 103

103 Figure 3 4 Cumulative distribution function of water levels from Little Lake Johnson and Lake Geneva from 1957 and 1945 to 2013, respectively Red line is a distribution of water level data, and the blue line is a normal distribution curve fitted to the water level data.

PAGE 104

104 Figure 3 5 P ercent water by weight and percent or ganic matter by dry weight in sediments collected from Little Lake Johnson transect and Lake Geneva transects 1 and 2. Values are the average of three samples collected and error bars are one standard deviatio n pH Water wei ght (%) Organic matter (%) V Cr Ni Cu

PAGE 105

105 Distance from shore (m) SD SD SD g/g SD g/g SD g/g SD g/g SD Little Lake Johnson all 5.12 0.18 6.35 7.45 33.3 15.6 16.1 5.51 20.4 6.21 9.08 2.69 12.9 4.12 Lake Geneva all 5.53 0.44 29.5 16.7 13.1 4.69 20.8 7.59 22.5 6.56 8.98 2.76 12.2 3.86 Little Lake Johnson 0 5.29 0.15 1.22 0.68 19.9 14.0 9.67 5.56 10.8 4.74 5.73 1.96 7.16 4.02 10 5.30 0.11 0.93 0.39 10.1 5.16 5.15 3.81 10.3 4.29 4.42 1.75 6.21 3.24 20 5.20 0.08 1.64 0.20 20.3 3.89 11.4 4.04 17.1 0.688 6.75 1.76 9.09 1.11 30 4.73 0.16 2.8 0 1.27 20.0 6.03 13.7 4.25 17.2 4.17 7.84 2.03 11.3 2.12 40 5.05 0.20 2.44 1.03 25.9 3.92 17.2 6.32 18.9 8.48 8.43 3.60 12.2 4.77 50 5.24 0.16 4.75 4.01 25.2 3.17 15.4 3.72 19.6 3.64 8.81 1.03 12.4 2.74 60 5.11 0.14 4.52 1.70 47.7 1.39 20.2 1.91 25.2 0.291 11.3 0.426 16.4 0.714 70 4.97 0.11 23.3 5.52 48.8 0.24 21.0 2.30 26.2 0.576 11.8 0.526 17.3 0.900 80 4.95 0.09 4.29 1.10 49.5 0.57 20.6 2.00 28.6 3.58 12.0 0.782 16.5 0.868 90 5.23 0.17 5.33 3.15 50.0 0.16 21.2 1.46 25.9 0.612 11.7 0.280 17.1 0.335 100 5.24 0.15 18.6 9.66 49.4 1.36 21.2 2.13 24.2 2.28 11.2 1.24 16.5 2.60 Lake Gene va 1 0 4.92 0.14 7.66 0.78 3.6 0.27 5.0 2.16 11.9 6.44 3.08 0.369 4.07 1.73 10 4.93 0.11 8.52 0.21 8.3 1.72 19.3 9.75 19.0 6.16 8.62 3.05 11.4 2.72 20 5.09 0.16 20.3 4.28 10.6 1.68 13.9 5.40 17.0 4.65 6.52 1.42 9.00 1.68 30 5.80 0.26 39.4 1.82 11.3 0.42 16.7 4.12 17.1 0.886 7.41 0.638 9.87 0.481 40 5.96 0.14 45.9 1.55 14.8 2.04 23.6 7.90 23.4 4.65 9.92 2.44 13.8 2.59 50 6.09 0.13 57.0 3.69 19.9 2.42 31.9 0.488 36.2 2.38 12.9 0.381 18.3 0.392 Lake Geneva 2 0 5.25 0.06 16.0 4.89 10.0 2.62 14.5 5.99 17.8 6.68 6.97 2.23 9.58 2.41 10 5.25 0.27 16.5 2.55 19.2 5.61 31.7 5.30 30.2 2.57 13.1 1.51 17.8 1.64 20 5.44 0.06 22.9 2.01 14.4 0.60 23.6 3.90 23.7 5.33 10.3 1.93 13.3 3.82 30 5.55 0.09 28.3 4.57 12.7 1.55 22.5 9.83 25.8 9.41 9.73 4.54 13.1 6.91 40 6.01 0.08 42.2 1.56 15.9 1.27 23.6 3.32 22.8 1.51 9.62 0.664 12.7 1.02 50 6.04 0.12 49.4 0.63 16.8 1.27 23.7 1.91 25.5 3.18 9.61 0.144 13.5 0.507 Table 3 3. Descriptive statistics for pH, percent water and organic matter, and metal concentrations ( g/g) from all samples and from Little Lake Johnson (n=33) and Lake Geneva (n=36). Descriptive statistics for pH, percent water and organic matter, and metal con centrations ( g/g) from sampling locations at 10 meter intervals along transects from this historic high water shoreline (0 m) at Little Lake Johnson and Lake Geneva (n=3). Values are means one standard deviation.

PAGE 106

106 Table 3 3. C ontinued Lake Distance from shore (m) Zn Sn Pb Sc Nd g/g SD g/g SD g/g SD g/g SD g/g SD Little Lake Johnson all 33.0 9.10 1.63 0.625 41.0 15.9 2.40 0.870 10. 2 2.92 Lake Geneva all 41.1 11.0 1.40 0.509 40.3 15.5 2.62 0.904 8 .30 2.81 Little Lake Johnson 0 19.0 6.68 0.703 0.502 17.6 13.0 0.788 1.34 6.96 4.64 10 21.6 4.22 0.632 0.426 17.6 8.84 1.87 0.381 4.96 2.23 20 23.9 4.83 1.20 0.277 24.4 4.85 1.34 0.046 7.80 2.60 30 28.6 7.69 1.41 0.228 33.8 6.54 1.97 0.185 8.86 1.40 40 29.5 11.3 1.50 0.894 39.4 19.3 2.43 1.17 10.0 5.09 50 31.3 3.34 1.52 0.396 40.3 11.4 2.11 0.937 9.45 2.43 60 41.4 3.54 2.30 0.057 53.7 5.64 3.13 0.097 12.4 1.02 70 43.0 3.71 2.38 0.082 57.8 7.93 3.43 0.121 14.1 1.32 80 40.2 6.96 1.91 0.363 59.9 2.73 2.86 1.65 11.0 7.62 90 43.0 3.99 2.31 0.053 54.9 6.99 3.26 0.046 13.1 0.147 100 41.0 6.76 2.11 0.392 51.1 2.02 3.26 0.157 13.3 0.061 Lake Geneva 1 0 19.7 2.06 0.394 1.12 9.11 1.12 1.62 0.562 3.41 0.830 10 35.4 12.1 1.10 10.0 33.6 10.0 1.78 0.376 6.72 2.13 20 32.5 9.56 0.872 8.27 30.2 8.27 2.59 0.780 7.30 2.49 30 35.1 8.78 1.32 1.78 30.3 1.78 1.67 0.347 6.60 0.708 40 43.9 12.3 1.57 6.91 44.0 6.91 2.17 0.593 8.11 1.82 50 61.2 5.07 2.25 4.32 70.4 4.32 4.62 0.384 14.7 0.974 Lake Geneva 2 0 34.4 11.7 0.969 13.6 30.5 13.6 2.64 1.15 7.02 3.19 10 56.1 9.20 2.01 1.15 56.1 1.15 2.90 0.987 8.49 4.06 20 42.4 7.93 1.51 17.5 41.8 17.5 2.04 0.539 6.69 1.40 30 47.9 15.7 1.67 30.7 50.2 30.7 3.78 1.16 10.2 5.77 40 41.2 3.84 1.57 6.85 40.9 6.85 2.41 0.513 10.9 2.20 50 42.8 2.85 1.58 2.38 46.9 2.38 3.23 0.449 9.50 0.729

PAGE 107

107 Figure 3 6 Concentrations ( g/g) of V, Cr, Ni, Cu, Zn, Sn, Pb, Sc and Nd measured in sediments collected along a transect from a) Little Lake Johnsons and b) two transects from Lake Geneva

PAGE 108

108 Figure 3 6 Continued

PAGE 109

109 Figure 3 7 Percent distribution of grain size (>63 m<) and organic matter in sediment from Little Lake Johnson transect and trans ects 1 and 2 from Lake Geneva

PAGE 110

110 V Cr Ni Cu Zn Sn Pb Sc Nd Little Lake Johnson Cr 0.93 Ni 0.97 0.97 Cu 0.95 0.97 0.98 Zn 0.93 0.94 0.96 0.95 Sn 0.93 0.95 0.96 0.97 0.95 Pb 0.89 0.96 0.95 0.95 0.89 0.93 Sc 0.84 0.81 0.84 0.88 0.86 0.84 Nd 0.85 0.83 0.84 0.83 0.85 0.87 0.80 0.89 % organic matter 0.89 0.88 0.90 0.90 0.87 0.87 0.88 0.75 0.75 Lake Geneva 1 Cr 0.91 Ni 0.99 0.91 Cu 0.94 0.90 0.97 Zn 0.95 0.93 0.95 0.91 Sn 0.89 0.86 0.90 0.90 0.92 Pb 0.93 0.94 0.95 0.97 0.93 0.90 Sc 0.70 0.87 0.71 0.72 0.80 0.65 0.84 Nd 0.90 0.92 0.90 0.90 0.93 0.89 0.96 0.88 % organic matter 0.84 0.86 0.87 0.91 0.88 0.86 0.93 0.78 0.88 Table 3 4. Pearson's correlation coefficients among trace elements and percent organic matter from Little Lake Johnson and Lake Geneva sediment transects

PAGE 111

111 Table 3 4. Continued V Cr Ni Cu Zn Sn Pb Sc Nd Lake Geneva 2 Cr 0.90 Ni 0.97 0.94 Cu 0.91 0.94 0.97 Zn 0.93 0.95 0.95 0.91 Sn 0.89 0.96 0.94 0.95 0.91 Pb 0.79 0.94 0.88 0.93 0.86 0.95 Sc 0.48 0.69 0.51 0.52* 0.70 0.60 0.68 Nd 0.64 0.63 0.60 0.56* 0.69 0.63 0.58* 0.74 % organic matter 0.81 0.62 0.71 0.65 0.67 0.58* 0.44** 0.29** 0.51* Not signficant at p<0.01, ** Not significant at p<0.05

PAGE 112

112 V Cr Ni Cu Zn Sn Pb Sc Nd % <63 m 0. 61 0.76 0.8 3 0.87 0.6 2 0.92 0.8 5 0.68 0.8 7 % >63 m 0.44* 0.6 2 0.7 3 0.7 9 0.45 0.8 6 0.7 5 0.57 0.8 4 Not significant at p<0.0 1 Table 3 5. correlation coefficients (n=23) among trace elements and grain size distribution of sediment transects from Little Lake Johnson and Lake Geneva

PAGE 113

113 Figure 3 8 Scatter plots of percent organic matter and Vanadium and Zinc concentrations ( g/g) measured from sediments transects from Little Lake Johnson and Lake Genev a

PAGE 114

114 CHAPTER 4 LEAD CONCENTRATIONS AND ISOTOPE RATIOS IN FLORIDA LAKE SEDIMENT S EXPOSED DURING LOW LAKE LEVELS : I MPLICATIONS FOR LEGACY PB POLLUTION IN LIGHT OF FUTURE CLIMATE CHANGE SCENARIOS 4 .1 Introduction Lead is an important environmental contaminant because of its known toxicity to humans and other or ganisms (Forstner and Wittmann 1981; Needleman and Bellinger 1991). Over the past century the major source of Pb in the environment was particulate emissions from vehicles that use d leaded gasoline. Leaded gasoline was introduced in the U S in the 1920s and use increased until it was phase d out by the Clean Air Act in the 1970s. Since th en Pb emissions from vehicles in the U.S. ha ve decreased by ~ 99.8% (USEP A 2000). Other sources of Pb include coal combustion, natura l geologic sources smelting, waste incineration and Pb based paint (Nriagu and Pacyna 1988) Dust from legacy Pb in soils is also recognized as a significant source of modern Pb pollution (Kamenov 2008; Zahran et al. 2013). L ead is commonly introduced into the environment as an atmospheric particulate and transferred from the atmosphere to water bodies, soils and sediments. In aqueous systems, both specific adsorption and ion exchange mechanisms transfer Pb onto suspended particles. Over the past century, Pb has accumulated in sediments to levels up to several times background concentration s (Nriagu 1996 ; Callender and Van Metre 1997; Siver and Wozniak 2001) From concentration data a lone, however, it is difficult to evaluate the relative importance of different Pb sources, discriminate between geogenic background and non geogenic components, as well as determine the potential toxicity of the accumulated Pb. Stable Pb isotope ratios an d sequential extraction methods, however, have the potential to address these questions. Pb isotope ratios can be used

PAGE 115

115 (Shotyk et al 1998 ; Escobar et al. 2013 ). The isotopic signature of Pb in the environment is derived from thre e major sources: natural, gasoline additives, and metal industries/smelters (Abi Ghanem et al. 2009) The isotop ic composition of Pb ore often differs significantly from the lead isotopic composition of local bedrock and soils ( Shirahata et al. 1980 ). Sequential extraction procedures are used to track Pb mobility and potential toxicity. Anthropogenically sourced Pb commonly adsorbs to more labile sediment fractions (carbonates, hydrous metal oxides, and organic matter), and is potentially mobile under changing chemical conditions (Pearson et al. 2010) whereas geogenically sourced Pb is sequestered in the mineral fraction (Tessier et al. 1979) L ake sediments can be sinks for trace metals, as these deposits typically remain undisturbed and permanently buried. Dredging or dam fa ilure/removal, however, can re mobilize these sediments and associated contaminants (Saeki et al. 1992; Juracek and Ziegler 2006) In light of climate change predictions, changing hydrologic patterns may also be a mechanism for either physical or chemical redistribution of leg acy Pb pollution. The frequency, intensity and duration of precipitation events may cause sediments in lake catchments to mobilize, disperse and become sources of legacy metal pollution (Foul ds et al. 2014). Intense dry conditions will lead to decreased wa ter levels in lakes, and thus impact sediment transport, especially in littoral zones (Rhodes and Wiley 1993). Additionally, the consequences of dry conditions, such as sediment drying, organic matter decomposition, and changes in sediment chemistry, will lead to diagenetic changes in sediments and possible mobilization of associated metals (Foster and Charlesworth 1996 ; Matagi et al. 1998; Schulz Zunkel and Krueger 2009)

PAGE 116

116 Predicted e ffec ts of climate change in Florida, especially increasing temperature and changing rainfall patterns, will influence hydrologic patterns across the state. Conflicting outputs of general circulation models suggest both lower and higher future annual precipitation A significant decreasing trend in October and May precipitation a nd an increasing trend in July through August rainfall, were measured in Florida over the last century (Martinez et al. 2012) Projections suggest increased future storm intensity and clustering in the summer months, resulting in extreme hydrographs, with large peaks in flow, but lower baseflow conditions (Mulholland et al. 1997) Most models agree that drought periods will be longer a nd more frequent (Yates and Towler 2011). About three fourths of Florida 7 800 lakes, which overall cover ~6% of the have maximum dept hs <5 m (Brenner et al. 1990). Because of the shallow nature of these lakes, lake level lows stands often have severe effects exposing large areas of lake bed (Schiffer 1999) An estimated 70% o f Florida lakes are seepage lakes, losing water only to evapotranspiration and the groundwater system, not via drainage in surface outflows (Schiffer 1999) The combined effects of decreased precipitation and inten sified groundwater use, and the ever increasing water demand associated with increasing population growth will further lower the water table, causing additional stress on these lake systems (Mulholland et al. 1997; Obeysekera et al. 2010 ; Annable et al. 1996; Ferguson and Gleeson 2012) Lake Ge neva and Little Lake Johnson, two lakes in north Florida (Figure 4 1) are highly influenced by local groundwater levels. During the regional dry period in the winter and spring 2012 these lakes experienced record low water levels ( Figure 4 2 ). Little Lake Johnson was completely dry and an estimated 25% of the previously

PAGE 117

117 inundated lake bottom of Lake Geneva was expose d. T he goal of this study was to investigate the effect of extr eme lake level low stands on the distribution of Pb in exposed lake sediments. Pb concentrations in lake sediments in north Florida are enriched above background levels (Blair et al. in review ), but there is uncertainty concerning the fate of anthropogenic ally sourced Pb when these sediments are exposed during extreme low water levels Specific objectives of this study were to: 1) determine total Pb concentration along transects of exposed lake sediments from the old shoreline toward the basin center ; 2) d etermine the distribution of Pb among sediment fractions us ing sequential extraction methods and; 3) identify the source of anthropogenically sourced Pb in the exposed lake sediment us ing radiogenic Pb isotopes 4.2 Methods 4.2.1 Study S ites Lakes Geneva and Little Johnson are located in an area known as the Sandhill Lake District of north Florida Lakes in the region have received much attention because of their astatic nature and recent pronounced declines in stage (SJRWMD 2013). The l akes are surround ed by deep quartz sands and in most case s occupy sinkholes formed by dissolution and collapse of underlying limestone. The drainage basins are small and water enters primarily via direct rainfall and subsurface inflow Local bedrock consists of Pleistocene age Trail Ridge Sands and the Pliocene age Cypresshead formation, composed of several hundred feet of semi consolidated marine and non marine deposits of sand, clay, marl, gravel, limestone, dolomite and dolomitic limestone (FLDEP 2010)

PAGE 118

118 La ke Geneva the town of Keystone Heights and covers approximately 6.6 km 2 (when full) and has a watershed of about 117 km 2 The city of Keystone Heights stretches along the northern shore and maintains a public beach a nd park, whereas private residences surround the remainder of the lake. This oligotrophic shallow groundwater fed lake has displayed very low water levels since the early 1990s leaving many private docks tens of meters from the water edge. This study focused on the northernmost area of the lake, which has a public access. Because of the recent lo w water levels, this area of the lake has been disconnected from the main water body to the south. I collected surface sediments from the southwest side ( tra nsect 1 ) an d southeast side (transect 2) of this northern sub basin of the lake (Figure 4 1 ) Head Branch State Park, about five miles NE of Keystone Heights. It is a relatively sma ll shallow oligotrophic to mesotrophic lake. It has an average maximum depth of 4.5 m and an area of 0.11 km 2 Groundwater fed Gold Head Branch Stream flows into Little Lake Johnson and at high stage the lake connects with larger Lake Johnson. I collect ed surface sediments along one SW/NE tre nding transect, starting on the southwest side of the lake (Fig ure 4 1 ). 4.2.2 Sample C ollection I collected surface sediments at Lake Geneva on 3 March 2012 and at Little Lake Johnson on 29 March 2012. At that time, water elevations were extremely low (~81 ft NAVD 1988). Relative to the periods of record for Lake Geneva (55 years) and Little Lake Johnson (67 years), water levels were similar to the minimum historic lake

PAGE 119

119 elevation (Little Lake Johnson 80.6 ft; Lake Geneva 81.1 ft) (Figure 4 2). These low water levels occurred less than 2% of the time during the period of record. There was no standing water at Little Lake Johnson and only a few small, disconnected pools in the north basin of Lake Geneva. At each lak e the approximate location of a previous lake level highstand was identified by the presence of dried, dark organic matter on top of quartz sand. At Lake Geneva I collected samples every 10 m along two 50 m transects using acid cleaned, 4x water rinsed, 20 ml plastic vials. Along both transects, the 50 m sample was collected at the At each sampling location three samples were collected, one on the transect, and one each, a meter either side of the transect. At Little Lake Johnson I collected samples every 10 m over a 100 m transect. Along this transect, three samples were collected at each location as described above. At each sampl ing location at the two lakes, the 20 ml vial was pressed into the surface sediment and the top ~3 cm of sediment w ere collected. Samples were place d in a cooler and sediment pH was measured with in 48 hours of collection using a 1:1 mix of DI water and sediment. One sample from each sampling location along all three transects was wet sieved to determine the fractions of sediment > 63 m and < 63 m. I treated the < 63 m fraction with hydrogen peroxide to remove organic matter and determine perce nt clay/silt. To determine percent dry mass, I weighed wet samples and re weighed them af ter freeze drying. D ry s amples were ground to a fine powder.

PAGE 120

120 4.2.3 Total Pb Analysis of Digested S ediment I prepared samples for total digestion and geochemical analysis in a class 1000 clean lab, equipped with class 10 laminar flow hoods, at the Department of Geological Science s, University of Florida. All reagents used for sample preparation were Optima grade. About 1 g of sediment was weighed in an acid cleaned ceramic crucible and ashed at 550 C to remove organic matter and determine weight loss on ignition (LOI) (H kanson and Jansson 1983) All elements of interest have boiling points >550 C, so ashing caused no element loss. About 0.05 g of the mineral ash was weighed, transferred to acid cleaned Teflon vials, and digested with 1 ml concentrated HF and 2 ml concentrated HNO 3 The samples were place d in a 90 o C oven for 48 hours. V ials were opened and the solution was evaporated to dryness on a hot plate. After evaporation samples were treated wit h 2 ml of 6 N HCl and vials were capped and heated on a hot plate overnight to ensure full dissolution. The solution was evaporated again to dryness. Five ml of 0.8 N HNO 3 with 100 ppm HF spike with 8 ppb Rh and Re was added to the samples to re dissolve the residue. A small aliquot of the solution was removed and diluted with the same spiked 0.8 N HNO 3 for a final dilution of ~2,000x for trace metal analyses. Trace element analysis was performed on a Thermo Element2 HR ICP MS in medium resolution with Rh and Re used as internal standards. I calibrated results using USGS rock standards AGV 1, BIR, BCR 2 and a procedural blank following Kamenov et al. (2008) Duplicate samples were also analyzed and results showed <10% variability. Metal concentrations are reported as ppm per gram dry sediment. Replicated mean Pb measurements (n=11) of Canadian Certified Refe rence Materials Project lake sediment standard LKSD 4 were 103 g/g with a standard

PAGE 121

121 deviation of 4.36 and an average recovery of 113% based on the reported value of 91 g/g. 4.2.4 BCR Sequential Extraction P rocedure Table 4 1 summarizes the BCR extractio n procedure used in this study ( See Rauret et al. 1999 for complete procedure) I used a starting weight of 1 g dried sediment for analysis. After each step I centr ifu ged the sample and removed 5 ml of extractant for analysis. Between steps, I washed each sample with 4x dist illed water. Each extractant was evaporated to dryness and 4.5 ml of 0.8 N HNO 3 with 100 ppm HF spike and 8 ppb Rh and Re was added to the sample s to re dissolve the residue The final dilution for the first three extracted fractions was ~5.5x. The residue was treated as described above for total rock digestion. Instrumental analysis was performed as described above. D uplicate samples, reagent blan k, procedural blanks, and standard reference materials were analyzed for quality control S tandard reference material LKSD 4 (Canadian Certified Reference Materials Project Lake Sediment Standard ) was used to verify the accuracy of the sequential extracti on method. C oncentrations from each extraction step were summed and compared to total digestion concentrations. Percent recovery was 138% based on the certified value of Pb concentration (91 g/g) in LKSD 4, and 119% based on the long term average (103 g/g; n=11). R eliability of the sequential extraction was determined by reproducibility of multiple leaching s (Table 4 2). Additionally, for each of the sediment samples the measured Pb concentration from total digestion was compared to the sum of the leach ed fractions. The percent recovery of the summed leached fraction s compared to the total digestion averaged 109.4%.

PAGE 122

122 4.2.5 Pb Isotope A nalysis After measurement of Pb concentrations, remaining sample solution s, both total digested and four extractants, w e re evaporated to dryness and dissolved in 1 N HBr and loaded onto columns packed with Do wex 1X 8 resin to separate Pb. The sample was washed 3x with 1 ml of 1N HBr and the Pb fraction was collected in 1 ml of 3N HNO 3 (Kamenov et al. 2009). I determined Pb isotope composition on a Nu Plasma MC ICP MS with Tl normalization following the procedure described in Kamenov et al. (2004). The Pb isotope data are expressed relative to values of reference material from the National Bureau of Standards (now National I nstitute of Standards and Technology) : NBS 981: 206 Pb/ 204 207 Pb/ 204 208 Pb/ 204 4.3 Results and Discussion Pb concentrations in Little Lake Johnson sediments were lowest at the historic shoreline (0 m) and increase d with distance from shore (Figure 4 3 and Appendix B Table B 1) Concentrations reached greatest values, ~3x increase, at 60 m and remained relatively constant to 100 m from shore In Lake Geneva transect 1 sediment Pb concentrations were low est closest to the historic shoreline, increased with distance from shore, and peaked at the location farthest from the old shore, an increase of >5x along the transect. In transect 2 there was no systematic increase in Pb concentration with distance from the old shore, but metal concentrations were comparable to the high concentrations measured in transect 1. Blair et al. (in review) determined a baseline, pre anthropogenic Pb concentration of 15 g/g from a 210 Pb dated sediment core in Little Lake Johnson, whereas modern

PAGE 123

123 values reached a peak of 65 g/g in the 1980s. In this study, mean Pb concentrations in surface sediments from lakes Little Johnson and Geneva were 41.0 and 40.3 g/g, respectively, indicating anthropog enic enrichment above baseline concentrations. Anthropogenic metals introduced into the aquatic environment occur in particulate form or are rapidly sorbed to particles and are deposited or remain in suspension, depending on the particle size and composi tion (Regnier and Wollast 1993; Stecko and Bendell Young 2000). Metals have potential for sorption onto clay minerals, hydrous oxides, and organic matter surfaces, all of which tend to be found in the smaller grain sizes (Frstner and Patchineelam 1980; Calmano et al. 1993 ; Ikem et al. 2003) Lead is known to form stable compounds with organic matter in peat (Logan et al. 1997) and Pb 2+ has a relatively high binding constant of 0.81 with humic substances (Drever 1997). Therefore, sediment characteristics, namely organic matter and relative proportion of fine grain material are important factors that influenc e metal distribution. Organic matter content in Little Lake Johnson surface sediment increased gradually from 25% to a distance about 50 m from the old shoreline. Thereafter, it increased abruptly and remained high (45 50%). At Lake Geneva, the pattern was different. Organic ma tter content in transect 1 increased gradually with distance from the old shoreline, from 4 18%. In transect 2, organic matter content averaged 15%, with no trend (Figure 4 3, Appendix B Table B 1). Sediments from Little Lake Johnson and transect 1 from Lake Geneva show significant (p<0.05) positive relationships between percent organic matter and Pb total concentrations, r = 0.81 and r = 0.86, respectively (Figure 4 4). As organic matter content in sediment increased toward the center of the lake, Pb al so increased In transect 2 from Lake Geneva there was no

PAGE 124

124 significant correlation between percent organic matter and Pb concentration suggesting that other factors controlled Pb distribution. In Little Lake Johnson and Lake Geneva transect 1, sediments t he <63 m sediment fraction incr eased with distance from shore and the >63 m fraction percent decreased (Figure 4 3) In sediments from Lake Geneva transect 2, grain size distribution was relatively constant, with the <63 m constituting ~35% and the >63 m fraction ~50%. Because of the limited amount of grain size data, data from all three lakes were pooled. The combined data showed a significant correlation (r = 0.87) between the <63 m fraction and total Pb concentration (Figure 4 5). Trace metals bound to flocculent material organic matter or fine grained sediments are easily suspended in to the lake water column (Nguyen et al. 2005) and can become mobilized and concentrated in sediments elsewhere in the basin when water level declines (Gottgens and Crisman 1992) Given the relationships between total Pb concentration and both organic matter and grain size in my study lakes, I suggest this resuspension mechanism accounts for the greater Pb concentrations in surface sediments with increasing distance from shore. Le ad is relatively immobile once sorbed to particulate matter and deposited. Indeed, widespread successful application of 210 Pb dating of lake sediments provides evidence against post depositional mass transport of Pb in sediment porewaters (Appleby et al. 1986) Both pH and redox, the main variables that control Pb adsorption onto organic matter and hydrous metal oxides (Foster and Charlesworth 1996 ; Matagi et al. 1998; Schulz Zunkel and Krueger 2009) must be considered in these sediments

PAGE 125

125 aerobic waters are colloidal hydrous ferric and ma nganese (Fe Mn) oxides (Harrison and Laxen 1984). The scavenging process is relatively rapid and mostly irreversible, but strongly pH dependent (Tipping et al. 2003 ; F rstner et al. 1994). Lead, like most cation metals (Cd, Cu, Zn) becomes more mobile at lower pH (2 4.5) because dissolution of hydrous Fe Mn oxides and carbonates increases and adsorption at sediment cation exchange surfaces decreases (Calmano et al. 1993) In simple systems, most Pb 2+ cations are strongly adsorbed to metal oxides above pH 5 (Drever 1997). At high pH (9 11), Pb solubility also increa ses because metals can form stable and soluble complexes with hydroxyl ions and dissolved organic carbon, or is released by dissolution of sulfide minerals (Tack et al. 1996; Ho et al. 2012). In my study, pH values in surface sediments were between 5 and 6 (Figure 4 3), and therefore unlikely to promote Pb solubility. Lead mobility under changing redox conditions must also be considered (Gallon et al. 2004; Pearson et al. 2010) Under reducing, anoxic conditions, hydrous Fe Mn oxides are readily reduced and soluble, and release associated metals. Soluble Pb i s removed from porewaters by re adsorption onto solid sulfides. Under oxidizing conditions, sulfide minerals become oxidized and release adsorbed metals. It is probable that reducing conditions in uppermost sediments increased toward the lake center, as wa ter content increased (Figure 4 3). All surface sediments were exposed to atmosphere and sampling was limited to 3 cm depth. Although o xidizing conditions prevailed at the sediment surface, reducing conditions c an persist in wet sub surface deposits, even at shallow depths in the sediment (Reddy and DeLaune 2008). In Lake Geneva, pH values increased from the historic shoreline to the lake center, providing evidence of changing redox conditions. The higher pH at the lake center likely resulted

PAGE 126

126 from reductiv e dissolution of hydrous Fe Mn oxides, whereas the low pH at the historic lake shoreline resulted from oxidation of sulfide minerals (Calmano et al. 1993 ; et al. 2007) The question remains whether the redox gradient caused mobility of Pb along the sediment tr ansects. A study by Ho et al. (2012) investigated mobility of trace metals under oxic and suboxic conditions. The authors measured the amount of Pb that was easily leachable under these conditions and showed there is very little difference in percent mobil e Pb between the oxic and suboxic sediments at pH 5 6. This suggests that under slightly reducing conditions Pb is not very mobile. Sequential Pb extraction methods can be used to provide additional evidence of potential redox mobilization and evaluate pot ential metal toxicity. 4.3.1 Pb Concentrations from Sequential Extractions The toxicity and fate of trace elements is dependent on their chemical form T herefore quantification of the different forms is often more meaningful than estimat es of total conce ntration (Tessier et al. 1979) One approach that is widely applied is fractionation of elements using sequential extractions. Th is approach ha s been c riticized because it is operationally defined and extracted fractions do no t necessarily correspond to specific mineralogical or chemical phases in the sediment ( Sutherland 2010 ). Nevertheless if each extracted metal fraction is defined by the chemical process use d to isolate it (e.g. reduction or oxidation) each fraction can be related to chemical processes within the s ediment. T hese processes control the release and mobilization of the metals and can provide information about the geochemistry of the s ediment (F rstner et al. 1994).

PAGE 127

127 In the absence of anthropogenic influences, trace elements are primarily associated with silicates and primary miner als, and display limited mobility. E lements that are introduced in to the en vironment by human activity bind with other sediment phases, such as carbonates, oxides, hydroxides, sulfides and/or organic matter. These phases are more readily extractable (Tessier et al. 1979 Zimmerman and Weindorf 2010) The hypothesis underlying these sequential extraction techniques is that mobile metals are removed in the first easily extractable fraction and the metals obtained from subseque nt extractions are more securely bound and less mobil e The Pb concentrations in surface sediments of my study lakes are enriched relative to baseline, pre anthropogenic values. Sequential extraction procedures, combined with Pb isotopic analysis, yielde d additional information about Pb source, potential toxicity and redox conditions. Using the BCR sequential extraction procedure, Pb concentrations were measured in four sediment fractions: 1) water/acid soluble or exchangeable (carbonates and phosphates), 2) reducible hydrous metal oxides, 3) oxidizable organic matter or sulfides, and 4) the residual mineral fraction (Table 4 2). For all three transects, Pb concentrations were highest in the reducible hydrous metal oxide fraction, followed by the residual mineral fraction, the oxidizable fraction, and a minor contribution from the exchangeable fraction. This order of distribution is comparable to that measured in organic rich soils, river and lake sediments and is indicative of a large anthropogenic compone nt of Pb in the sediments (Monna et al. 2000; Bacon et al. 2006 ; Ho et al. 2012). Pb concentrations in all four fracti ons increased with greater distance from shore. Sediments closest to the historic shoreline were dominated by quartz sand, which

PAGE 128

128 contains low geogenic Pb concentrations. This accounts for the low concentrations in the residual fraction at these locations. Little Lake Johnson and Lake Geneva have very low water column alkalinity, 0.8 and 1.7 mg/L as CaCO 3 respectively (Florida Lakewatch 2005), which may accoun t for the low Pb concentrations in the exchangeable fraction, which includes metals bound to car bonates. Higher concentrations in both the reducible hydrous metal oxide fraction and oxidizable organic/sulfide fraction suggest that conditions along transects were neither extremely reducing or oxidizing. Lead concentrations in all fractions increased a long transects with greater distant from the old shoreline, suggesting that all Pb phases were concentrated toward the lake center. This is consistent with the hypothesis that an increase in fine grain and organic material toward the lake center is respons ible for the higher concentrations of adsorbed metals at those sites. The percent distribution of Pb fractions along transects gives an indication of Pb mobility (Figure 4 6). In all three transects, percent Pb in the residual fraction was relatively cons istent, indicating the proportion of minerals that contain Pb did not change. Along the length of the Little Lake Johnson and Lake Geneva 1 transects, the proportion of Pb in the more mobile fractions varied slightly. The change in proportion is close to t he percent variability measured in replicate samples, but there i s a trend along the transects. The proportion of Pb in the exchangeable fraction decreased from >2% to ~1% in both lakes, whereas the oxidizable fraction increased from 5 to 17% in the Little Lake Johnson transect and 7 to 15% in Lake Geneva transect 1. The reducible fraction decreased from 60 to 44% in Little Lake Johnson and 67 to 56% in Lake Geneva 1. The relative proportion of oxidizable organic matter or sulfide sediment increased with

PAGE 129

129 in creasing distance from shore, but the proportion of exchangeable and metal hydroxide bound Pb decreased. D ry conditions closest to the historic high lake level could have caused organic matter degradation decreasing the proportion of oxidizable organic matter and the exchangeable fraction (Prica et al. 2009) Exposure, o xidization and mineralization of the deposits may have resulted in Pb mobilizati on, even via aeolian processes (Calmano et al. 2005). The increase in the proportion of Pb in the reducible fraction towards the center of the lake may indicate the reduction of oxides under anoxic conditions ( Chesworth et al. 2006 ). Although variability between Pb concentrations in the four sediment fractions was slight along the transects, I conclude: 1) th e proportion of Pb in the most mobile fractions (exchangeable and reducible) was greatest closest to the historic shoreline, 2) there was a slight change in redox conditions along the transects indicated by the change in proportion of Pb in the oxidizable vs. the reducible fraction, however 3) Pb concentrations increased in both the reducible and oxidizable fractions towards the center of the lake, indicating that extreme redox conditions did not occur and probably did not influence Pb mobilization. Other s tudies found that Pb, compared to Cd, Cu, and Zn, is less affected by changing redox conditions (Saeki et al. 1992; Nemati et al. 2011). Thus, changing redox conditions along the transects probably did not influence the Pb distribution. Instead, concentrat ions of organic matter and fine particulate matter, which display greater concentrations in topographic lows nearer the lake center, had the most influence on Pb distribution.

PAGE 130

130 4.3.2 Pb Isotopes Radiogenic Pb isotopes were measured from the four sequential extracted sediment fractions (Appendix B Table B 2). I will focus on 208 Pb/ 204 Pb, but similar spatial trends are seen for each of the Pb isotope ratios. The residual fraction had values between 38.6 38.9, similar to the other three fractions, with value s in the range 38.3 38.5 (Figure 4 7). Pb isotope ratios can be used to fingerprint the source of Pb in sediments. Pb ore used as a gasoline additive, Pb released during coal combustion, and Pb from local geologic sources, have different isotopic ratios Studies of wetland and sediment cores from Florida show two major shifts in 208 Pb/ 204 Pb pre 1900 to present (Figure 4 8) (Kamenov et al. 2009; Escobar et al. 2013; Blair et al. in review). The first gradual shift was from values >38.7 pre 1900 to values of 38.3 38.45 by 196 0. After 1960, there was a shift to higher values of 38.4 38.6, which are found today. Pre 1900, the Pb signature was controlled primarily by geogenic Pb both from local sources and Saharan dust (Kamenov et al. 2009). Total Pb conce ntrations over the last century increased dramatically as a consequence of Pb inputs from leaded gasoline. The initial increase in concentration was accompanied by a decrease in 208 Pb/ 204 Pb, reflecting the composition of the Idaho Pb ore that was originall y used to make the gasoline additive. The second shift is at tributed to the transition from Idaho ore to Mississippi Valley Type (MVT) deposits used as the gas additive from the 1960s through the 1980s (Graney et al. 1995; Kamenov et al. 2009). One assum ption of sequential leaching procedures is that anthropogenic metals are sorbed to the more labile fractions. Pb isotope measurements support this

PAGE 131

131 assumption, as isotope ratio values in the leached sediment fractions (exchangeable, reducible, and oxidizabl e) show distinct Pb isotopic ratios compared to the residual fraction (Figure 4 9). Where values for three most labile fractions overlap with the residual fraction indicates a geogenic Pb component. Recent (1970s to 2010) anthropogenic Pb isotopic signatur e modeled from bulk measurements of Florida lake sediments (Escobar et al. 2013) are similar to the three most labile fractions, and provide further evidence of the anthropogenic source of the Pb found in these lake sediments. The Pb isotope measurements s how that anthropogenically sourced Pb dominates in exposed surface deposits, even in sediments with relatively low Pb concentrations (<20 g/g). In these lake sediments, Pb concentrations were low compared to the Florida Department of Environmental Protec tion for residential areas which is 400 g/g. There may, however, be human health implications associated with the exposure to Pb in these sediments. Tegen et al. (2002) documented high dust emissions from recently dried lake beds, a consequence of the high proportion of fine grain material and low surface roughness. Lakes in my study are surrounded by quartz sand and the major mechanisms for dust generation is saltation or sandblasting across the sediment surface (Grini and Z ender 2004). A number of studies concluded that modern h uman Pb exposure occurs primarily through inhalation and oral ingestion ( Mielke and Reagan 1998; Kamenov 2008; Hu et al. 2014). Zahran et al. (2013) found that atmospheric soil and atmospheric Pb conc entrations show near identical seasonal lead levels in Detroit, MI USA. Blood lead levels are higher in summer and autumn and lower during winter and spring. The

PAGE 132

132 authors concluded that resuspended soil is a signif icant source of atmospheric Pb and that a Pb concentration of only 0.0069 g/m 3 in the air can increase blood lead levels in a one year old child by 10%. Although the problem of atmospheric Pb exposure is more acute in urban environments, my study shows that dry lake beds are potential sources of legacy anthropogenic Pb. 4.4 Conclusi ons Lead concentrations in sediments of lakes and wetlands worldwide have increased over the last 150 years. This is mostly attributed to combustion of gasoline containing Pb additive. Throughout the US Pb concentrations in sediments peaked in the 1970s after which Pb additive was prohibited (Siver and Wozniak 2001 ; Escobar et al. 2013) Climate change scenarios for Florida predict periods of more intense rainy seasons followed by pe riods of more inte nse droughts. Protracted dry conditions in north Florida caused severe decreases in lake levels and winter 2012, some lakes desiccated completely. Exposed sediments were collected along the dry lake beds to determine the fate of anthropog enic Pb that had accumulated in the lake deposits. Lead concentrations increased with greater distance from the historic high water level and highest Pb concentrations were found in surface sediments closest to the topographic low (center) of the lake. Con centrations in these deposits were up to 3x higher than pre anthropogenic, background concentrations. Lead distribution was primarily controlled by the distribution of fine grain sediments and organic matter. This material was concentrated toward the lake center as water levels fell. Although redox conditions are known to change in drying sediments, sequential extractions on these sediments provided no evidence for a strong redox gradient along the transects.

PAGE 133

133 Therefore, redox potential is thought to have ha d little influence on Pb distribution. Sequential extraction showed that most Pb is associated with the relatively mobile hydrous Fe Mn oxide sediment fraction, followed by the residual mineral phase and the organic matter/sulfide phases. Very little Pb is present in the acid exchangeable carbonate fraction. Pb isotopes in the extracted sediment fractions confirmed that Pb in the exposed sediments was sourced from anthropogenic activity, namely gasoline additive. Although the Pb concentrations measured in t his study were relatively low, they show the potential of lake sediments as sources of legacy Pb pollution.

PAGE 134

134 Figure 4 1. Maps of study locations. a) Map of Florida; star indicates study location. b) Map view of Lake Geneva and Little Lake Johnson. c) Outline of average water levels of Little Lake Johnson D ashed link indicates location of sampling transect with the measured distance from shoreline (0 100 m) marked d) Outline of average water levels of Lake Geneva with the location of the enlarged northern end of the lake. Finely dashed line on enlarged map was lake level in March 2012. Numbered dashed lines indicate location of sampling transect s 1 and 2 with the measured distances from shoreline (0 50 m) marked.

PAGE 135

135 Figure 4 2 Lake stage (ft NAVD88) for Lake Geneva and Little Lake Johnson sinc e 1957 and 1945, respectively. Lake stage measurement from 2005 to 2013 are shown in the second panel and arrow indicates the date lake se diment samples were collected. Data from SJRWMD.

PAGE 136

136 Step Sediment phase Extractant Shaking time and temperature EXT Water and acid soluble, and exchangeable 40 mL of 0.11 M CH 3 COOH 16 hr at room temperature RED Reducible metal oxides 40 mL of 0.5 M HONH 2 HCl (pH 1.5) 16 hr at room temperature OXI Oxidizable organic matter/sulfides 10 mL of 8.8 M H 2 O 2 (pH 2) 10 mL of 8.8 M H 2 O 2 (pH 2) add 50 mL of 1 M NH 4 OAc (pH 2) 1 h at room temperature and 1 hr at 85 o C 1 hr at 85 o C (after second 10mL) 16 hr at room temperature RES Residual a 1 mL HF and 2 mL HNO 3 36 hr at 90 o C oven a Digestion of the residual is not a step of the BCR protocol. Table 4 1. Modified BCR sequential extraction scheme used for metal speciation

PAGE 137

137 Lake Distance from shore (m) Exchang e able Reducible Oxidizable Residua l Sum of fractions Total 1 Little Lake Johnson 0 0.142 4.00 0.373 2.08 6.60 5.52 10 0.724 20.7 4.12 6.40 32.0 22.6 20 0.374 12.9 2.78 9.55 25.6 24.4 30 0.436 17.5 3.13 9.69 30.7 32.6 40 0.422 18.1 3.79 10.9 33.2 35.6 50 0.385 15.4 3.34 10.0 29.2 36.9 60 0.535 29.6 11.6 26.7 68.4 53.7 70 0.565 29.5 13.1 26.2 69.3 57.8 80 0.447 27.6 10.9 26.0 65.0 59.9 90 0.424 27.9 12.0 25.5 65.8 54.9 100 0.443 28.6 11.6 25.0 65.7 56.9 Lake Geneva 1 0 0.280 5.66 0.742 3.52 10.2 9.10 10 0.674 17.7 2.76 5.17 26.3 28.1 20 0.718 19.8 3.13 7.36 31.0 30.2 30 0.778 29.4 6.78 11.5 48.4 30.3 40 0.860 27.8 5.97 10.9 45.6 44.0 50 0.937 36.2 9.79 17.7 64.7 70.4 Lake Geneva 2 0 0.647 16.9 3.55 7.06 28.2 30.5 10 0.799 27.2 6.22 10.2 44.4 56.0 20 0.827 29.8 7.39 12.5 50.5 41.8 30 0.788 30.3 7.11 11.1 49.3 50.2 40 0.806 29.9 6.80 15.1 52.6 40.9 50 0.866 30.9 8.52 12.2 52.5 48.2 LKSD 4 mean 3.54 90.5 19.6 9.23 123 103 stnd. dev. 0.405 7.02 1.43 0.54 7 6.33 -% variability 11.4 7.76 7.29 5.89 5.15 -1 Total bulk digestion Table 4 2. Lead concentration ( g/g) in sediment fractions from BCR sequential extraction and total digestion in sediment samples from Little Lake Johnson and Lake Geneva transects, and LKSD 4 lake sediment standard

PAGE 138

138 Figure 4 3. Average pH, percent water weight, percent organic matter, and total Pb concentrations (n=3), and measured percent distribution of grain size (>63 m<) measured at 10 m intervals along sediment transects from Little Lake Johnson and Lake Geneva. Distance fr om shore indicates distance from the historic lake level high (0 m) to the lake center. Error bars represent one standard deviation. Lake Geneva Transect 1 is represented by solid lines, whereas Transect 2 is represented by dashed lines.

PAGE 139

139 Figure 4 4. Percent organic matter vs. total Pb concentration ( g/g) in surface sediments from Little Lake Johnson and Lake Geneva. Correlation coefficients are shown for each transect and are significant at p<0.05 for Little Lake Johnson and Lake Geneva 1, but not significant for Lake Geneva 2 (p=0.069)

PAGE 140

140 Figure 4 5. Percent <63 m grain size fraction vs. total Pb concentration ( g/g) in sediments from Little Lake Johnson and Lake Geneva. Combined data from all three lakes (n=23) yield an r value of 0.87 (p<0.01).

PAGE 141

141 Figure 4 6. Relative abundance of Pb in surface sediment fractions, determined by the BCR sequential extraction procedure along transects in Little Lake Johnson and Lake Geneva, from the historic high shoreline s (0 m) towards the center of the lakes. Stacked bar graphs indicate the percent of Pb in the residual, oxidizable metal oxide, reducible organic matter/sulfides, and exchangeable fractions of sediment along the transects.

PAGE 142

142 Figure 4 7. 208 Pb/ 204 Pb measu red in sequentially extracted sediment fractions from Little Lake Johnson and Lake Geneva transects from the historic high lake level (0 m) towards the lake center. Error bars are shown, but in some cases are smaller than the data point symbol.

PAGE 143

143 Figure 4 8 208 Pb/ 204 Pb measured in total digested sediment from FL lake and peat cores over the last century. Lake Sheelar, FL black squares (Blair et al. accepted), Little Lake Jackson red circles (Escobar et al. 2012), and Blue Cypress Marsh, FL blue tria ngles (Kamenov et al. 2009).

PAGE 144

144 Figure 4 9 Comparison between 206 Pb/ 204 Pb, 207 Pb/ 204 Pb, and 208 Pb/ 204 Pb in leached sediment fractions (exchangeable, reducible, oxidizable), residual fraction from this study, and modeled recent (1970s to present) anthropogenic values from Florida lake sediments and peat from Escobar et al. (2013).

PAGE 145

145 CHAPTER 5 PRESERVATION OF TRACE METAL S TRATIGRAPHY IN FLORIDA LAKE SEDIMENTS AFTER LOW LAKE STAGE EVENTS 5.1 Introduction Paleolimnological studies are use d to quantify anthropogenic influence s on the environment. The approach has d ocumented human alteration of trace metal cycling (e.g. Pb), as a consequence of smelting practices that began thousands of years ago (Norton and Kahl 1987; Shotyk et al. 2001; Bi ndler et al. 2008; Thevenon et al. 2011 ; Cooke et al. 2007 ). Today, trace metals are found in lake sediments worldwide delivered via atmospheric deposition. Since the mid 1850s many studies have shown a significant increase in trace metal accumulation in lake sediments. This increase is attributed to the Industrial Revolution, increasing automobile usage, and for P b specifically, the widespread use of Pb as a gasoline additive. L ake sediments accurately preserve a record of trace metal deposition in wa ter bodies (Siver and Wozniak 2001; Bindler et al. 2008) Several studies, however have investigat ed the degree to which early diagen e tic alteration and sediment transport may alter the metal profiles (Winderlund et al. 2002; Schottler and Engstrom 2006; Rydberg et al. 2008). Sediment records have been used to study pre anthropogenic and anthropogenic metal loadings, deterioration of watersheds as a consequence of land use change, and the effectiveness of remediation practices ( Van Metre et al. 2006 ). I f paleolimnological data are to be reliably interpreted, then sediment metal concentrations must not change appreciably after burial (Boyle et al. 2001b ). This study focus ed on this requirement with special attention to post depositional metal migration. M echanisms responsible for metal mobility are both physical and chemical. Physical mechanisms

PAGE 146

146 include wave generated turbulence from storm events and sediment focusing, which translocate sediments and associated bound elements (Lehman 1975; Foulds et al. 2 014). Other factors that may promote post depositional metal mobility include chemical changes e.g. shifts associated with seasonal stratification and turnover ( Anshumali et al. 2009 ) or biological processes at the sediment water interface that caus e red ox and pH changes ( Santschi et al. 1990 ; ). Future c limate scenarios include predicted changes in seasonal hydrologic patterns, namely the duration and frequency of wet and dr y events. In light of the potential future severity of these e vents, I address their potential for altering trace metal accumulation records in the sediments of shallow lakes S pecifically, I determine if drying events that cause extreme lake level declines and exposure of large swaths of lake sediment, cause enough sediment mobility, either physically or chemically, to alter the recorded history of trace metal accumulation. ave maximum depths < 5 m. There has been some concern that sediment stratigraphy may not be well preserved in the se shallow systems. It has been suggested that in large, shallow, wind stressed water bodies such as Lake Okeechobee (area ~1,800 km 2 mean depth ~2.7 m), south Florida, hurricanes and tropical storms may exert strong effects on the sediments (Havens et al. 2001; Bachman et al. 2003). Schottler and Engstrom (2006), however, used multiple dated markers and determined that sediments in L ake Okeechobee have not been severely mixed for at least the last ~75 years. Similarly, Kenney et al. (2010) showed that sediments o f four other shallow Florida lakes (max depth <10 m) had accumulated in an

PAGE 147

147 orderly fashion, with little evidence for sedimen t mixing. What is less well understood is the effect of lake level variability on sediment mobility in these shallow lakes. L ake level fluctuation is natural and beneficial for some biological functions, but extreme dry conditions and precipitous declines in lake level can impair ecosystem function ( Zohary and Ostrovsky 2011) and affect water quality (White et al. 2008; Martinez et al. 2012) particularly in shallow lake systems. Extreme dry conditions may cause stage reductions that result in physical m obilization and focusing of sediments (Gottgens and Crisman 1992 Brezonik and Engstrom 1998 ) Other consequences of dry conditions such as sediment exposure and drying, organic matter decomposition, and changes in sediment chemistry, may lead to diagenet ic changes in sediments and their associated metals (Foster and Charlesworth 1996 ; Matagi et al. 1998; Schulz Zunkel and Krueger 2009) From a paleolimnological perspective it is important to determine if shallow lakes maintain accurate records of tra ce metal accumu lation or if sediment drying alters these long term records. W ith respect to metal contamination f uture management of lake sediment and water quality will require accurate records of previous metal loading and hence quantification of potential metal mobil ity. To address th e question of how lake level decreases may impact the integrity of trace metal stratigraphy in the sediments of shallow Florida lakes, I utilized repeat cores from Lake Harris cored in 1999 and 2013, and Newnans Lake cored in 1997 and 2005 (Figure 5 1). T he lakes experienced periods of lake stage decline between the years when the repeat cores were collected (Figures 5 2 and 5 3 ). Metals commonly attributed to anthropogenic activities, V, Ni, Cu, Zn, and Pb, and conse rvative elements

PAGE 148

148 Sc and Nd, were measured in the cores Metal concentration profiles and inventories were compared between the repeat cores in each lake to determine the degree of similarity. Additionally, radiogenic Pb isotopes were measured to assess Pb mobility. I used these measure ments to determine if protracted drying events altered the record of metal deposition in shallow Florida lakes. Re coring efforts have been used elsewhere to determine the effectiveness of management efforts in a watershed (Be nnion and Battarbee 2007), assess carbon and nitrogen loss over time (Glman et al. 2008), and evaluate mercury stability in sediments (Rydberg et al. 2008). In Florida, Engstrom et al. (2006) used repeat coring to quantify nutrient enrichment of Lake Okee chobee. Stratigraphic patterns obtained from the second coring effort confirmed results from the first study and showed that except for the addition of new material, underlying sediments remained virtually unchanged with respect to the values found 15 year s earlier. Effective re coring studies must have an accurate and precise way to match the two sediment cores, such as scans of magnetic susceptibility (Blumentritt et al. 2013), diatom biostratigraphy (Battarbee 1978), or pollutant markers (Rose and Yang 2 007). In this study I used 210 Pb dat es to match sedim ent ages between cores. 5.2 Materials and Methods 5.2.1 Study Sites Lake Harris (28 o o (Figure 5 1 Table 5 1 ). It is one of seven large lakes in the Harris Chain of Lakes, which discharge into the Ocklawaha River. The major inflow into the lake is Palatlakaha River. The Dead Head River is the major outflow, however depending on stream discharge and wind

PAGE 149

149 direction t his river can also function as a lake inflow. Th e Dead Head River connects Lake Harris with other water bodies in the Oklawaha chain of lakes (SJRWMD 2014). The lake basin in underlain primarily by the Pliocene Cypresshead formation, however Tertiary Quat ernary u ndifferentiated Reworked Cypress, and Holocene sediments are also found in the watershed. A mixture of commercial, industrial, and private uses dominates land use on the north e dge of the lake. The city of Leesburg is located on the northwest sho re, wh ereas the south and east shores are a mixture of private residences, citrus groves, and wetlands. Newnans Lake (29 o o east of th e city of Gainesville (Figure 5 1 Table 5 1 ). The northern exten t of the drainage basin is largely undeveloped and supplies surface water inflow via Hatchet Creek, Little Hatchet Creek, and several smaller input streams. Prairie Creek, at the southwest corner of the lake is the only surface water outlet. The lake lies in the Bone Valley Formation, made up of phosphatic sand, clayey fine sand and clay, and the Hawthorne Formation, composed of phosphatic sand, silty sand and clay (Brenner and Whitmore 1998). 5.2.2 Sample C ollection R epeat cores in each lake w ere taken at approximately the same mid lake location as the original, determined using a hand held GPS. Methods for c ollecti on and dating of the 1999 Lake Harris core are described in Schelske et al. (2001). Th e 1999 Harris core was extruded and sampled at 5 cm inter vals. M ethod s for collecting and dating the 1997 Newnans Lake are described in Brenner and Whitmore (1998). Th e 1997 Newnans core was extruded and sampled at 4 cm intervals. The 2013 Lake

PAGE 150

150 Harris and 2005 Newnans Lake cores were collected using a sediment water interface corer (Fisher et al. 1992) and both repeat cores were extruded and sectioned into 4 cm intervals. 5.2.3 Dry Mass, Organic Matter Content, and Bulk D ensity For the earlier e xtracted cores dry mass and bulk density were already determined. I used acid cleaned containers and Optima grade reagents for preparation and analysis of samples from the cores For the two more recent cores I followed the same protocol for measurement of dry mass, and bulk d ensity that had been used in analysis of the original cores. To determine percent dry mass, wet samples were weighted and then re weighed after freeze drying. I ground the dry samples to a fine powder. For samples from all four cores I estimated percent organic matter for dry samples by loss on ignition (LOI) at 550 C (Hkanson and Jansson 1983 ) All elements of interest have boiling points >550 C, so ashing caus ed no elemental loss. I calculated sediment bulk density (g dry mass cm 3 ) from percent dry mass in wet sediment and the organic/inorganic proportion of dry sediment (Binford 1990) 5.2.4 210 Pb C hronology All cores were dated using 210 Pb and measured radioisotope activities ( 210 Pb, 226 Ra [ 214 Bi], 137 Cs) by gamma counting (Appleby et al. 1986) using EG&G Ortec GWL High Purity Germanium coaxial well detectors attached to a 4096 channel pulse height analyzer (Sch elske et al. 1994) Unsupported 210 Pb activity was calculated as the difference between total 210 Pb activity and supported 210 Pb ( 226 Ra) activity at each depth. Dates were determined using the constant rate of supply (CRS) model (Appleby and Oldfield 1978 1983; Oldfield and Appleby 1984)

PAGE 151

151 Total 210 Pb represents the sum of supported and unsupported 210 Pb activity in the sediments. Supported 210 Pb is produced by the decay of 226 Ra from erosional inputs and is assumed to be in secular equilibrium with 226 Ra in the sediment. Total 210 Pb and 226 Ra activities become equal at depths where excess or unsupported 210 Pb is not measurable because of unsupported 210 Pb decay. Unsupported 210 Pb is deposited from the atmosphere and its decay provides the basis for estimating the sediment age. With a half life of 22.3 years, it is immeasurable beyond ~120 years ago. To c ompare unsupported 210 Pb activity profiles between cores within lakes, the original activities in the 1999 Lake Harris and 1997 Newnans Lake cores were adjusted for radiogenic decay between the core collection dates. I assume that unsupported 210 Pb profile s and inventories, after correction, should be similar between the two within lake cores if there was not significant sediment mobilization. 5.2.5 Trace Metal A nalysis I prepared and digested all samples from the four cores for geochemical analysis in a c lass 1000 clean lab, equipped with class 10 laminar flow hoods, at the Department of Geological Sciences, University of Florida. All reagents used for sample preparation were Optima grade. About 0.05 g of sediment was weighed, transferred to acid cleaned Teflon vials, and digested with 1 ml concentrated HF and 2 ml concentrated HNO 3 The samples were placed in an oven for 48 hours at 90 C. The vials were opened and the solution was evaporated to dryness on a hot plate. After evaporation, samples were treated with 2 ml of 6 N HCl and vials were capped and heated on a hot plate overnight to ensure full dissolution. The solution was evaporated again to

PAGE 152

152 dryness. Five ml of 0.8 N HNO 3 with 100 ppm HF spike with 8 ppb Rh and Re, was added to the samples t o re dissolve the residue. A small aliquot of the solution was removed and diluted with the same spiked 0.8 N HNO 3 for a final dilution of ~2,000x for trace metal analyses (Kamenov et al. 2009). A p rocedural blank was prepared with the samples. Trace elem ent analysis was performed on an Element2 HR ICP MS in medium resolution with Rh and Re used as internal standards. Quantification of the results was done with external calibration using USGS rock standards AGV 1, BIR 1, BCR 2. The following elements were analyzed for all sample s and standards: Sc, V, Ni, Cu, Zn, Nd, and Pb Analytic recovery was ensured by including subsamples of USGS certified reference standard AGV 1, with recovery between 95 and 103%. Precision was measured by replicate measurement of 1 0% of the samples, and wa s <85 % for all elements. Metal concentrations are reported as g/g dry sediment. 5.2.6 Stable Pb Isotope A nalysis The remaining sample solution was evaporated to dryness and dissolved in 1 N HBr and loaded onto columns packed with Dowex 1X 8 resin to separate Pb. The sample was washed 3x with 1 ml of 1N HBr and the Pb fraction was collected in 1 ml of 3N HNO 3 (Kamenov et al. 2009). I determined Pb isotope composition on a Nu Plasma MC ICP MS with Tl normalization following the pr ocedure described in Kamenov et al. (2004). The Pb isotope data are expressed relative to the following values of National Bureau of Standards (now National Institute of Standards and Technology) Standard Reference Material for native Pb NBS 981: 206 Pb/ 204 207 Pb/ 204 208 Pb/ 204

PAGE 153

153 5.2.7 Statistical Analysis and Inventory Calculations All statistical analysis was conducted with JMP Pro Version 11.0. Prior to statistical analysis, I tested data distributions of sediment geochemical and physico chemical characteristics for normality. I used the Shapiro Wilks test to test for normality and found that the majority of distributions were not normal. To test for significant differences between cores, I used the Wilcoxon text, which is similar to the Kruskal Wallis test, which is the non parametric equivalent of a two sample t test. Significance was determined at the p < 0.05 levels. Inventories or total accu mulation were calculated for unsupported 210 Pb, sediment bulk density, organic matter, and for the seven trace metals. The inventories reported are from 1950 to 1999 for Lake Harris cores and 1951 to 1997 for Newnans Lake cores. The 1950 cut off date was chosen as a conservative approach because of the high dating error ( 30 years) associated with older sediments. Inventory errors would be very large if these older sediments were included. Inventories were calculated by multiplying the concentration value 1 dry) by the bulk density (g dry cm 3 wet) and the length of each interval, i.e. 4 or 5 cm depending on the core. These values were then summed to yield the amount under a cm 2 of sediment surface. Errors were calculate d using the 2 error from 210 Pb ages and concentration measurements. If intervals were not dated to 1950 values were interpolated to desired date.

PAGE 154

154 5.3 Results 5.3.1 Water Levels Over the period of record the average lak e stage at Lake Harris was 61.5 ft asl corresponding to 3.7 m average water depth (Table 5 2). M inimum lake stage corresponds to a water level decrease of slightly more than 1 m. Between sampling events in 1999 and 2013, lake stage decreased to near record lows during the low rainfall yea rs of 2000 and 2006 (Figure 5 2) L ake stage was lower during this 14 year interval compared to the ent ire period of record (Figure 5 4 ). A verage depth at Newnans L ake was 1.3 m which corr esponds to a lake stage of 64 ft asl (Table 5 2) The minimum lake stage recorded in 2001 resulted in near drying of the whole lake. The cumulative distribution of lake stage was not very different between the entire period of record and the eight years between coring events, however the lowest lake level occurred between the 1997 and 2005 coring events, during t he 2000 2001 dry spell (Figure 5 4 ). 5.3.2 210 Pb Chronology and Inventory In the two cores from Lake Harris, there is a relatively consi stent decrease in unsupported 210 Pb activity with depth, suggesting that the CRS modeled dates are reliable (Figure 5 5 ). Additionally, the activities at comparable depths in the two cores are similar. A ctivity profiles in the Newnans Lake core s show vari able, but not necessarily decreasing activity at the top of the sections. According to the activity profiles and corresponding dates, ~40 cm of material accumulated from 1997 to 2005.

PAGE 155

155 S ediment in Newnans L ake is very floccul e nt with high water content w h ich may explain the apparently high linear accumulation Unsupported 210 Pb inventories were calculated for the period from 1950 to 1999 in each core. The Lake Harris 1999 core had an inventory of 13 dpm/cm 2 and the 2013 core had an inventory of 10.2 dpm/cm 2 Inventories for the Newnans Lake cores were 3.4 (1997) and 3.8 (2005) dpm/cm 2 5.3.3 Sediment Physical Properties M ean percent organic matter in Lake Harris was 62.4 % in the 1999 core and 58.5% in the 2013 core ( Table 5 3 ) The highest percent organic matter, >65%, was measured at the surface in both Lake Harris cores (Figure 5 6). The inventory of organic matter for the period 1950 1999 was ~50% greater in the 1999 compared to the 2013 core (Figure 5 8). M ean bulk sediment mass for the period 1950 1999 in the 1999 and 2013 cores from Lake Harris were 0.037 and 0.032 g/cm 3 respectively (Table 5 3). Cumulative mass inventories for the 50 year time period in the two cores were 0.72 (1999) and 0.93 (2013) g/cm 2 (Figure 5 8). M ean percent organic matter in the two Newnans Lake cores w as 59% (1997) and 56.9% ( 2005 ) P ercent organic matter profi les are presented in Figure 5 7 In both cores the organic matter was lowest in the oldest sediments and increased up core. The inventory of organic matter was higher in 2005 (Figure 5 9). In Newnans Lake the mean bulk sediment mass in the 1997 core was 0.047 g/cm 3 and 0.053 g/cm 3 in the 2005 core (Table 5 3) and the cumulative mass inventories were 0.22 (1999) and 0.14 (2005) g/cm 2 (Figure 5 9).

PAGE 156

156 5.3.4 Trace Metal Conce ntrations and Inventories The metal concentrati ons in contiguous core intervals and average metal concentrations in the four cores are shown in Table 5 3 Mean within lake concentrations were compared using the Wilcoxon rank sum test and the z statistic i s presented in Table 5 3. Trace metal profiles for the Lake Harris cores are shown in Figure 5 6 and for Newnans Lake in Figure 5 7. In the Lake Harris cores, all the metals had a sub surface maximum in the 1960s and 1970s, with the lowest concentrations i n the early 1900s and at the sediment surface. Sc and Nd exhibit this same pattern. All metal concentrations from the 2013 core are within two standard deviations of the concentrations in the 1999 core. The only discrepancies are the Pb and Cu measurements pre 1950, for which concentrations measured in the 1999 core are lower than those in the 2013 core. The trace metal means measured in the two Lake Harris cores were not different to one another at p<0.05. Various metal inventories for the 1999 cores are s ystematically higher than those for the 2013 core, however the values are within or very close to within error (Figure 5 8). In general, metal concentrations in the Newnans Lake cores had relatively constant concentrations and did not show systematic trend s up core. The Pb and V profiles exhibited slight variability. Pb concentrations increased from the 1950s to the 1970s and decreased in both cores at the sediment surface. The V profiles exhibited a slight increase in concentrations during the 1940s and c oncentrations decreased at the surface of the 2005 core. Zn, Cu and Ni concentrations in both cores were constant throughout. Conservative elements Sc and Nd did not show any up core trend in either

PAGE 157

157 core. In the two Newnans Lake cores, Sc, V, Zn, Nd, and P b had significantly different means, whereas means for Ni and Cu were not significant at p<0.05 Several metal inventories from the 2005 core are systematically higher than those of the 1997 core inventories. Only the inventories for Cu, Ni, and Zn were not different between the two cores (Figure 5 9). 5.3.5 Stable Pb Isotopes Stable Pb isotope data are presented in Figure 5 10 and Table 5 4. Pb isotope composition varies with depth in both sets of cores. In Lake Harris, the 208 Pb/ 204 Pb ratio was greatest in the deepest intervals and generally decreased upcore. Values in sediments from 1940 and younger varied little, between 38.5 and 38.6. In the Newnans Lake cores the highest 208 Pb/ 204 Pb ratio (38.73) was measured in the oldest interv al of the 2005 core Values in the younger sediments varied little, between 38.4 and 38.55. In both cores, there is a shift in values around 1940 to 1950, from about 38.4 to 38.5. 5.4 Discussion 5.4.1 Lake Harris Lake Harris cores were retrieved in 1999 and 2013, and between these times there were two large decreases in water level. These low levels represented less than 2% of the period of record. Th e decline corresponds to about a 33% decrease in average water d epth. D own core profiles of trace metals were comparable between the two Lake Harris cores (Figure 5 6). In both cores the concentrations of metals commonly attributed to anthropogenic activity, Pb, Cu, Ni, V, and Zn, exhibit ed slight increases in concentr ation from 1900 to the 1970s, and decrease s near the sediment surface. The increases in concentration correspond to the period of increased development

PAGE 158

158 throughout Florida from 1900 to the 1950s (Kamenov et al. 2009; Blair et al. in review). A decrea se in s urface metal concentrations was also measured in New England l akes (Siver and Wozniak 2001) and the Great Lakes (Graney et al. 1995). A survey of 35 lakes across the U S showed that more lakes exhibited decrease s in Cu, Ni, V, Pb, and Zn concentrations fr om the 1970s to the 1990s, than showed incr easing concentration trends. This is attributed to legislation, regulation, changing demographics and industrial practices limiting metal pollution since the 1970s (Mahler et al. 2006). Sc and Nd are conservative elements that are commonly sequestered in mineral grains and therefore represent mineral material originating from local rock or atmospheric soil/rock dust ( Weiss et al. 1999 ; Shotyk et al. 2002; Kamenov et al. 2009) These elements reflect geogenic sourc es, are not anthropogenically sourced, and are largely unaltered by post depositional biogeochemical processes. Sources of these elements include weathering of local lithology or long range transport of mineral soil dust. These metals serve as a proxy for geogenically sourced mineral material. Constant concentration throughout a core indicates little to no change in the source or rate of mineral material delivered to the lake basin, or consistent dilution by non mineral (organic) material. An increase in co ncentration indicates either an increase in accumulation rate of mineral material, a decrease in the accumulation of organic material, or addition of another source with higher elemental concentrations. Increasing concentrations could also be attributed to mineralization of organic matter that is exposed to atmosphere, with corresponding enrichment of mineral material. The concentration profiles of Sc and Nd showed slight variability throughout both Lake Harris cores, but all values were within error measu rements. The most noticeable

PAGE 159

159 trend seen in both cores was a decrease in concentrations from the 1970s to the top of the cores. This decrease is most likely a consequence of dilution of these elements by an increase in organic matter over the same time peri od. Sediment character and metal inventories were calculated from 1950 to 1999 for both Lake Harris cores (Figure 5 8). The inventories calculated for the two cores are within or close to within error. Inventory index ratios, i.e. the inventory of 1999 co mpared to that of 2013, were calculated for all variables (Table 5 5). All metals have ratios of 1.1 or 1.2, whereas as cumulative mass, unsupported 210 Pb and amount of organic matter have ratios of 1.3 or 1.5. These differences between the two cores sugge st sediments in the 2013 core were altered, resulting in smaller inventories. This alteration could be a product of either physical or chemical mobilization. It is assumed that conservative elements Sc and Nd are unaltered by chemical change, and because a ll the metal index ratios are nearly identical, chemical mobilization was unlikely. Physical mobilization could result in homogenization, an increase or decrease in sediment characteristics and metal measurements. The smaller inventories of the 2013 core c ompared to the 1999 core indicate lower concentrations of the measured metals in the more recent core. The decrease could have resulted from dilution by organic or mineral material, but this is not supported by an increase in organic matter or mass invento ries. The small differences in inventories likely reflect natural variability. Shallow lakes can display highly variable sediment distribution because of wind stress (Whitmore et al. 1996; Kenney et al. 2014). Even in annually varved sediments, spatial var iability as a consequence of diagenesis was found. Whereas down core trends were similar, absolute concentrations of elements differed among multiple lake cores (Ryberg and Martinez Cortizas 2014).

PAGE 160

160 Similarities in concentration profiles and inventory calcu lations, as well as the preservation of stratigraphy between the repeated cores, strongly suggest that metal deposition records were not altered below the horizon dated 1999 even after pronounced dry periods in 2000 and 2006. Radiogenic Pb isotope rati os were measured in both Lake Harris and Newnans Lake cores to evaluate mobility of Pb within the sediments. These ratios can be used to fingerprint the source of Pb. The most common sources of Pb in the environment, Pb ore used as a gasoline additive, Pb released during coal combustion, and Pb from local geologic sources, have different isotope ratios. Studies of wetland and lake sediment cores from Florida show two major shifts in 208 Pb/ 204 Pb, pre 1900 to present (Figure 5 10) (Kamenov et al. 2009; Escobar et al. 2013; Blair et al. in review). These shifts are attributed to changing sources of Pb in the environment. Pre 1900, the Pb signature was controlled primarily by geogenic Pb from local sources and Saharan dust (Kamenov et al. 2009). The ratio of this pre anthropogenic Pb is >38.7 (Figure 5 10). Over the last century, total Pb concentrations increased dramatically as a consequence of Pb inputs from leaded gasoline. This addition caused a gradual shift from values >38.7 pre 1900 to values of 38.3 38.45. This shift reflects the use of Idaho Pb in gasoline additive. After 1960, there was a shift back to higher values of 38.4 38.6, which are found today. This shift is at tributed to the transition from Idaho ore to Mississippi Valley Type (MVT) de posits which were used in the gas additive from the 1960s through the 1980s (Graney et al. 1995; Kamenov et al. 2009).

PAGE 161

161 In Lake Harris, there is evidence of the shift from the pre anthropogenic 208 Pb/ 204 Pb signature to lower values, after the input of an thropogenic Pb (Figure 5 10). The 208 Pb/ 204 Pb value of ~38.5, from the 1950s to the core surfaces, is consistent with a signature for anthropogenic Pb pollution (Escobar et al. 2013). What is not evident, however, is the 1960 shift that reflects the change in the source of Pb ore, which is commonly recorded in other lake sediment cores throughout Florida. Absence of this shift could be a consequence of sampling strategy or addition of another source Pb to this lake. It may also indicate physical sediment tr ansport, which homogenized the isotope values, however, the otherwise preserved stratigraphy and increase in Pb concentrations during the 1960s 1970s (Figure 5 6), argue against mobilization and homogenization. 5.4.2 Newnans Lake Newnans Lake cores were retrieved in 1997 and 2005. In the eight years between these events water levels decreased to the lowest recorded stage since 1936. The 2001 dry period caused a profound drop in lake stage and widespread exposure of lake bottom sediment. It is likely that this lake level decrease caused sediment mobility. Prior to collection of either core, i n the spring of 1989 the St. Johns River Water Management District conducted a partial drawdown of the lake to expose organic rich littoral sediment in an effort to i mprove water quality. It was hypothesized that oxidation and compaction of the exposed littoral sediment would improve establishment of littoral plant communities and promote downstream flushing of resuspended flocculent organic matter. L ake level was l owered by 30 62 cm for eight weeks. T his study showed there was sediment mobility within the lake, specifically that organic material was eroded from

PAGE 162

162 the littoral zone. This short term drawdown was shown to greatly enhance resuspension of fine littoral s ubstrate, however it was unclear whether this material was redistribut ed to the deeper portions of the lake (Gottgens and Crisman 1992). Unsupported 210 Pb profiles in the two Newnans cores (Figure 5 5) do not show consistent, down core decay curves. This could be the result of sediment mixing and redistribution or variable sediment accumulation rates over time. Kenney et al. (2010) showed that 210 Pb can be used effectively to date sediments in shallow Florida lakes. Preservation of 210 Pb stratigraphy suggests the radionuclide records changing accumulation rates and has not been homogenized as a consequence of mobility. Concentration profiles of Cu, Ni, and Zn, especially in the 2005 core, were fairly homogenous top to bottom, which might be interpreted to reflect sediment mixing. It is difficult to determine if these profiles preserved the increase in metal pollution after 1900 that is seen in many other lakes throughout Florida. In contrast to Cu, Ni, and Zn, both V and Pb concentrations from the 2005 core show an increase, with a peak during the 1950s and a decrease at the core surface, suggesting anthropogenic metal enrichment, as found in other studies (Kamenov et al. 2009; Escobar et al. 2013). Conservative elements Sc and Nd have unchanging concent rations throughout both cores. This could indicate little change in accumulation of mineral material, consistent dilution by organic material, or sediment homogenization by mixing. Unlike Lake Harris, calculated metal inventories for the 1997 and 2005 Ne wnans cores were not systematically different (Figure 5 9). The inventory for cumulative mass and Zn was higher in the 1997 core, whereas for all the other variables, the 2005 inventories were greater. Index ratios for the inventories were calculated to de termine

PAGE 163

163 the degree of difference between the two cores (Table 5 5). Ratios for unsupported 210 Pb, organic matter, Cu and Ni in the cores were very similar (0.8 0.9), whereas Sc, Nd and cumulative mass were nearly 50% different. The 2005 Newnans Lake cor e has evidence for the pre anthropogenic geogenic 208 Pb/ 204 Pb ratio in the bottommost interval. In post 1970s sediments, both cores display very similar values, ~38.5, consistent with modern 208 Pb/ 204 Pb values measured in other systems (Figure 5 10). Both cores record the shift from lower Idaho ore values to higher MVT deposits. The timing of the shift is somewhat offset between the two cores. It is dated to the late 1950s in the 1997 core and the early 1940s in the 2005 core. It is possible that 210 Pb da ting error in these intervals, on the order of 5 10 years, may account for the apparent discrepancy between the timing of the shift in the two cores. In any case, the similar Pb isotope plots from the two cores indicate there was no physical perturbation of sediments deposited before the 1960s. Another consideration was the potential for chemical mobilization of elements in the sediment column. Alfaro de la Torre and Tessier (2002) recognized three types of geochemical diagenesis of metals that could res ult in their redistribution in the sediment column. These include: 1) diffusion of dissolved metals from the water column into sediment pore water, later to be precipitated by sulfides under reducing conditions; 2) dissolution or desorption of particle ass ociated metals and their diffusion through pore waters and subsequent re precipitation with sulfides or oxyhydroxides in a new location in the sediment column; and 3) dissolution of particulate metals because of redox related decomposition of oxyhydroxides or organic matter and re precipitation or resorption elsewhere in the sediment column. Additionally, changes in pH can also be a

PAGE 164

164 factor in mobility of metals ( F rstner et al. 1994). Lowering of lake levels can expose sediments, allowing for changing redox and pH conditions (Tack et al. 1996), and refilling of the lake can also cause changes to pH and redox conditions (Calmano et al. 1993; A number of studies show metal mo bility at the sediment water interface. There is evid ence of surface enrichment of V Cu, Pb and Zn attributed to cycling processes and in the case of V, multiple oxidation states (Calmano et al. 1993 ; Boyle 2001 a ; Gallon et al. 2004 ). Those elements that return to the water column are re adsorbed onto falling particles. Although surface remobilization is acknowledged, many studies conclude remobilization is not a major factor in most situations (Taylor and Davison 1995; Boudreau 1999; Couillard et al. 200 8; Percival and Outridge 2013). Boyle (2001 b ) there are two scales of trace element cycling one fast and temporary and the other slow and long lived. The scale of the former is not recorded because of the over riding longevity of the later. From previous studies, it is unlikely that metal migration occurs at great depth below the sediment water interface, and thus it is unlikely that metal profiles would be disrupted. R esults from the Newnans Lake cores are not as stra ightforward as results f ro m the Lake Harris cores with respect to post depositional mobility of metals S ome conclusions however, can be made. Similar Pb isotopes shifts in the two Newnans cores indicate a lack of physical translocation of sediments pre 1 960. After 1960, however, many metal profiles in the 1997 and 2005 core s are fairly homogenous. The lack of stratigraphy after 1960 could be due to unchanging metal inputs to the lake, relative metal dilution due to increasing non metal material to the lak e, or physical

PAGE 165

165 homogenization of any stratigraphic variability prior to the coring events. Independent of these reasons, because there were no discernable stratigraphic trends in the 2005 core, I cannot determine if there were physical sediment mobility be tween 1997 and 200 5. Metal inventories for the periods 1951 1997 indicate differences between the 2005 and 1997 cores. This could result from natural variability or mobility of specific elements u nder changing chemical conditions. 5.5 Conclusions Climate change predictions for Florida suggest a change in hydrologic patterns, namely in the timing and amount of seasonal precipitation. This variability will likely cause extreme shifts in lake levels. The goal of this study was to determine if recent climatic ally and anthropogenically driven shifts in lake water levels altered records of trace metal deposition in two shallow Florida lakes as a consequence of physical and/or chemical mobility. Trace metal concentrations were measured in cores from shallow Lake Harris and Newnans Lake. Lake Harris, central Florida, was cored in 1999 and 2013. Newnans Lake was cored in 1997 and 2005. Both lakes experienced extreme low lake level conditions between the coring events. To compare metal profiles between the two cores taken in each lake, metal concentrations, inventories, and radiogenic Pb isotope ratios were examined. Additionally, metal distributions were compared to sediment bulk density, fraction organic matter, and unsupported 210 Pb activities in the cores. Lake level variability over the 14 years between coring events did not alter metal accumulation in the sediments of Lake Harris. Profiles of multiple physical and geochemical variables in the sediments are similar. Metal inventories for the period 1950 to1999 were different between the two cores. The offset between cores was

PAGE 166

166 similar to that for inventories of unsupported 210 Pb, cumulative mass, and organic matter This consistent discrepancy wa s likely a result of natural system variability rather than chemic al or physical sediment mobility. Stable Pb isotopes measured in the sediments did not record shifts commonly seen in other Florida lake and peat sediments. This may be attributed to insufficient sampling resolution, an alternative local Pb source, or post depositional sediment mobility. The latter, however, is unlikely given the evidence for maintenance of stratigraphic integrity from other variables. Results from the Newnans Lake cores are not as definitive as those from Lake Harris, with respect to prese rvation of metal deposition records. Newnans Lake experienced a protracted dry spell between the coring events and much of the lake bottom was exposed. Because of the flocculent nature of organic rich sediments at the sediment water interface, surface sedi ments might be expected to mobilize as the water level dropped. Metal concentration profiles did not vary much throughout each core, so it is difficult to determine if profiles were altered, post depositionally. Inventories of several variables accumulated between 1951 and 1997 display differences between the two cores. The magnitude of difference for metals, organic matter, and unsupported 210 Pb in the two cores is similar. The Pb isotope shift that resulted from the change in the source of Pb ore used as a gasoline additive in the 1960s was preserved in the sediments of both cores. Preservation of this shift in both cores provides evidence that there was not wholesale redistribution of sediments before 1960 in this shallow lake. For sediments younger than 1960, it is less certain whether metals in sediments deposited after 1960 were subjected to either chemical or physical remobilization.

PAGE 167

167 Extreme lake level decreases will likely cause mobility of recently deposited previously deposited metals. Future paleolimnological studies of long term metal deposition in shallow lake systems should measure metals from multiple cores to assess natural variability in metal accumulation between core sites and evaluate evidence for potential metal mobility. Although metal concentrations measured in sediments of Lake Harris and Newnans Lake are not high enough to trigger management or remediation intervention today, it is worth considering that future lake level declines may result in oxidation of surface sediments and concentration of metals in uppermost deposits, perhaps t o toxic levels.

PAGE 168

168 Figure 5 1. Map of Florida showing the locations of Lake Harris and Newnans Lake.

PAGE 169

169 Lake Lake Area Watershed Max Depth Mean Depth Trophic State (ha) Area (ha) (m) (m) Lake Harris 6 700 87 300 4.8 3.7 eutrophic Newnans Lake 2 670 30 800 3.6 1.3 hypereutrophic Table 5 1. Lake Harris and Newnans Lake characteristics

PAGE 170

170 Figure 5 2. Lake stage (ft NAVD88) for Lake Harris from 1956 to 2014. Stage measurements for 1998 to 2014 are shown in the right panel and arrows indicate the dates the sediment cores were collected. Data from SJRWMD.

PAGE 171

171 Figure 5 3. Lake stage (ft NAVD88) for Newnans Lake from 1936 to 2014. Stage measurements for 1996 to 2007 are shown in the right panel and arrows indicate the dates the sediment cores were collected. Data from SJRWMD.

PAGE 172

172 Lake Number of days in period of record Minimum Q1 Mean Median Q3 Maximum ----------------------------------------ft NAVD 1988 -----------------------------------------Lake Harris 18 896 57.85 61.14 61.49 61.66 61.99 63.79 Newnans Lake 8 361 59.72 64.02 64.94 64.96 65.93 70.07 Table 5 2. Summary statistics for water levels from Lake Harris and Newnans Lake

PAGE 173

173 Figure 5 4. Cumulative distribution functions of water levels from Lake Harris and Newnans Lake. The black line is for the full period of record and the red dashed line is for the period between the coring events

PAGE 174

174 Figure 5 5. 210 Pb and 226 Ra activities and date versus depth from repeat sediment cores collected at Lake Harris (1999 and 2013) and Newnans Lake (1997 and 2005).

PAGE 175

175 Depth Pb 210 Date Mass Sed Rate (mg/cm 2 /yr) Bulk Sed Organic Sc V Ni Cu Zn Nd Pb (cm) Density (g/cm 3 ) Matter (%) ( g/g) ( g/g) ( g/g) ( g/g) ( g/g) ( g/g) ( g/g) Lake Harris 1999 12 1994 27.7 0.018 68.4 2.14 24.04 8.59 21.32 37.05 7.37 44.82 16 1991 25.1 0.021 67.9 2.23 25.14 8.87 22.29 36.65 7.80 46.44 20 1986 20.9 0.027 66.6 2.30 25.00 9.09 23.30 42.77 7.82 49.79 24 1979.6 19.2 0.029 65.4 2.41 26.71 9.35 23.89 38.09 8.15 54.87 28 1972 17.5 0.035 62.8 2.53 28.38 9.29 23.34 37.16 8.67 57.05 32 1961 13.8 0.039 58.8 2.33 28.08 9.28 23.64 37.74 8.93 59.60 36 1950.3 16.9 0.043 57.9 2.65 26.66 8.86 15.17 29.90 9.15 41.81 40 1941 18.4 0.043 60.2 2.67 25.14 8.66 9.28 23.88 9.46 30.17 44 1932 20.3 0.047 58.8 2.61 24.27 8.41 7.37 20.69 9.93 26.36 48 1915 11.6 0.049 59.9 2.65 23.35 8.46 6.97 20.52 9.36 24.68 52 1898 11.5 0.050 59.7 2.52 20.72 8.05 5.61 16.84 8.63 18.46 Mean 0.037 62.4 2.46 25.23 8.81 16.56 31.03 8.66 41.28 Std. Dev. 0.011 4.0 0.19 2.20 0.42 7.76 9.00 0.80 14.22 Lake Harris 2013 4 2011.9 38.0 0.014 64.2 1.95 18.31 8.67 17.18 35.98 5.96 26.80 8 2008.9 29.1 0.022 64.8 1.96 18.69 7.76 16.89 29.36 6.37 27.87 12 2004.7 22.6 0.024 65.4 2.12 22.93 8.45 18.42 31.83 6.94 32.34 16 1998.7 16.4 0.025 63.1 2.29 27.80 9.10 20.25 34.06 7.73 38.58 20 1992.9 18.3 0.026 59.7 2.43 28.08 9.53 20.99 40.28 9.65 44.14 24 1984.7 15.5 0.032 52.9 2.58 28.00 10.13 23.79 40.60 9.29 52.83 28 1975.0 14.7 0.035 52.7 2.72 30.32 10.39 25.17 41.48 9.72 62.08 32 1964.1 14.2 0.039 52.6 2.86 32.63 10.65 26.54 42.37 10.16 71.33 36 1952.4 13.4 0.039 54.7 2.35 26.77 8.78 22.19 33.19 8.28 58.22 40 1933.8 8.7 0.040 56.4 2.54 28.63 9.42 21.94 35.28 9.05 61.76 44 1912.2 7.5 0.041 57.9 2.22 23.78 8.20 15.81 27.68 7.98 44.96 48 1888.4 7.2 0.043 58.1 1.90 18.93 6.98 9.69 20.08 6.91 28.15 Mean 0.032 58.5 2.33 25.40 9.01 19.90 34.35 8.17 45.76 Std. Dev. 0.009 4.9 0.31 4.81 1.09 4.64 6.56 1.42 15.39 Table 5 3. Pb 210 date, mass sedimentation ra te, percent organic matter, trace metal concentrations from sediment core intervals from Lake Harris (1999 and 2013) and Newnans Lake (1997 and 2005) Means, standard deviation and z statistic from Wilcoxon rank sum test of means are reported for each core

PAGE 176

176 Table 5 3 Continued Depth Pb 210 Date Mass Sed Rate (mg/cm 2 /yr) Bulk Sed Organic Sc V Ni Cu Zn Nd Pb (cm) Density (g/cm 3 ) Matter (%) ( g/g) ( g/g) ( g/g) ( g/g) ( g/g) ( g/g) ( g/g) Newnans Lake 1997 4 1996 94.4 0.023 62.1 1.77 22.05 34.41 10.58 50.47 8.64 24.08 8 1995 89.4 0.030 62.2 2.26 22.51 14.04 10.15 50.90 10.25 26.64 12 1994 89.2 0.035 61.2 2.13 21.86 13.52 9.86 49.21 9.01 26.99 16 1992 78.5 0.038 60.4 2.17 22.71 13.86 10.23 53.61 9.96 27.38 20 1990 86.7 0.039 60.8 2.21 23.56 14.19 10.60 58.01 9.83 27.78 24 1988 78.6 0.039 60.8 2.15 22.81 13.98 10.45 58.54 9.27 28.58 28 1985 62.0 0.039 60.9 2.22 23.17 14.18 10.70 58.22 9.45 28.02 32 1983 62.5 0.039 60.9 2.27 22.97 14.19 10.59 59.84 10.15 28.91 36 1980 67.4 0.048 60.6 1.94 22.56 14.05 10.46 51.30 8.46 29.29 40 1987 52.9 0.042 60.5 2.14 22.03 13.97 10.55 52.00 10.12 30.18 44 1974 43.6 0.041 60.5 2.01 22.33 14.21 10.60 57.70 9.20 29.35 48 1970 45.1 0.044 60.7 2.03 21.76 13.81 10.22 50.03 9.43 28.46 52 1966 35.4 0.045 60.7 2.17 22.67 14.10 10.81 56.28 9.66 28.14 56 1960 38.6 0.052 60.3 2.15 21.63 14.00 10.26 49.75 10.00 27.77 60 1956 49.3 0.053 55.8 2.28 21.02 14.15 11.06 61.43 11.10 29.16 64 1951 47.3 0.055 55.2 2.08 20.41 13.84 10.64 53.38 5.17 27.56 68 1946 51.4 0.057 55.2 1.88 19.79 13.53 10.22 45.33 8.28 25.97 72 1940 47.6 0.061 55.9 1.97 18.98 13.28 9.87 44.50 10.40 24.41 76 1932 33.0 0.057 57.1 2.11 19.36 13.40 10.09 53.31 10.51 24.98 80 1921 18.6 0.058 56.5 2.15 19.49 13.62 10.25 55.19 5.83 24.84 84 1906 17.9 0.061 55.2 2.18 19.62 13.83 10.42 57.07 10.67 24.71 88 1887 10.2 0.061 55.3 2.01 18.81 13.33 10.42 59.33 10.52 23.88 Mean 0.046 59.0 0.046 21.46 14.79 10.41 53.88 9.36 27.14 Std. Dev. 0.011 2.6 0.011 1.50 4.39 0.29 4.67 1.45 1.92 Newnans Lake 2005 5 2005 217.7 0.0 20 61.7 2.65 19.95 13.30 10.73 43.12 10.23 24.75 10 2004 206.7 0.022 61.0 2.70 20.71 13.63 10.02 37.88 10.62 25.30 15 2003 149.1 0.031 60.8 2.73 21.68 13.54 10.05 38.94 10.84 26.98

PAGE 177

177 Table 5 3 Continued Depth Pb 210 Date Mass Sed Rate (mg/cm 2 /yr) Bulk Sed Organic Sc V Ni Cu Zn Nd Pb (cm) Density (g/cm 3 ) Matter (%) ( g/g) ( g/g) ( g/g) ( g/g) ( g/g) ( g/g) ( g/g) 20 2002 206.0 0.034 60.3 2.68 20.33 13.46 10.37 40.50 10.42 25.03 25 2001 164.8 0.039 59.1 2.58 23.17 12.24 9.21 33.72 10.46 27.24 30 2000 148.1 0.042 59.1 2.87 25.66 13.79 10.50 38.41 11.61 31.22 35 1998 177.8 0.043 58.7 2.93 26.07 14.03 10.83 39.11 11.64 31.51 40 1997 170.7 0.043 59.1 2.90 25.75 13.90 10.55 38.68 11.62 31.34 45 1996 212.3 0.043 58.8 2.92 25.65 13.76 10.74 39.02 11.77 31.45 50 1995 163.2 0.044 58.9 2.84 25.18 13.85 10.42 39.29 11.13 30.71 55 1993 138.4 0.048 58.7 2.79 25.34 13.67 10.36 38.28 11.40 31.58 60 1992 185.1 0.051 58.0 2.87 25.15 13.88 10.27 38.41 11.89 31.56 65 1989 95.2 0.049 57.5 2.95 26.14 13.80 10.26 38.99 11.87 32.78 70 1986 85.5 0.048 57.7 3.00 26.25 14.09 10.53 39.11 11.84 33.73 75 1982 61.6 0.051 57.6 3.01 26.68 14.29 10.54 39.69 12.06 33.91 80 1979 76.8 0.053 58.1 2.98 26.19 13.94 10.40 39.05 11.85 33.25 85 1975 70.9 0.051 57.9 2.93 25.48 13.39 10.14 37.54 11.69 32.89 90 1973 128.5 0.053 57.0 2.80 24.92 13.33 11.51 35.40 11.25 32.30 95 1969 71.7 0.063 54.7 2.98 26.31 14.25 9.57 34.75 12.17 36.53 100 1963 53.5 0.065 55.3 3.04 26.05 13.91 9.78 37.27 12.38 39.30 105 1955 42.2 0.063 55.7 2.92 26.82 14.03 10.11 38.90 11.68 41.28 110 1947 36.5 0.062 54.9 2.84 26.08 13.64 10.13 38.45 10.89 41.38 115 1939 44.7 0.069 50.8 2.94 23.11 12.88 10.06 38.25 12.28 32.59 120 1932 51.7 0.069 46.4 2.82 22.43 12.40 9.52 34.65 11.22 29.05 125 1920 26.5 0.064 54.5 2.91 22.88 12.95 10.08 34.93 11.94 30.13 130 1902 22.7 0.083 45.7 3.29 26.28 15.88 10.15 34.40 12.86 27.22 135 1892 56.9 0.117 45.1 3.92 33.77 15.70 8.82 20.34 14.92 9.86 Mean 0.053 56.4 2.92 24.96 13.76 10.21 37.30 11.65 30.92 Std. Dev. 0.019 4.5 0.25 2.71 0.77 0.53 3.99 0.91 6.04 Wilcoxon rank sum means test 1 Lake Harris 0.05 0.28 0.72 0.56 0.48 0.64 0.48 0.44 Newnans Lake 0.02 0.0001 0.0001 0.14 0.08 0.0001 0.0001 0.0003 1 Values of <0.05 have means that are significantly different

PAGE 178

178 Figure 5 6. Percent organic matter and trace element concentration profiles measured from Pb 210 dated sediment cores taken in 1999 (black) and 2013 (red) from Lake Harris. The bolded lines with the data points are the measured concentrations and the fine and dotted lines are two standard deviation based on replicate measurements.

PAGE 179

179 Figure 5 7. Perce nt organic matter and trace element concentration profiles measured from 210 Pb dated sediment cores taken in 1997 (black) and 2005 (red) from Newnans Lake. The bolded lines with the data points are the measured concentrations and the fine and dotted lines are two standard deviation based on replicate measurements.

PAGE 180

180 Figure 5 8. Inventories for unsupported 210 Pb, sediment bulk density, organic matter, and trace metals calculated from 1950 to 1999 from sediment cores collected in 1999 and 2013 from Lake Harris. Error was calculated from the 2 210 Pb age and concentration errors.

PAGE 181

181 Figure 5 9. Inventories for unsupported 210 Pb, sediment bulk density, organic matter, and trace metals calculated from 1951 to 1997 from sediment cores collect ed in 1997 and 2005 from Newnans Lake. Error was calculated from the 2 Pb 210 age and concentration errors.

PAGE 182

182 Figure 5 10. Changes in 208 Pb/ 204 Pb ratio with age in sediment cores from Lake Harris (1999 and 2013), Newnans Lake (1997 and 2005), Lake Sheelar (Blair et al. in review), Little Lake Jackson (Escobar et al. 2013), and a peat core from Blue Cypress Marsh (Kamenov et al. 2009).

PAGE 183

183 Date Stable Pb Isotope Ratios 206/204 207/204 208/204 Lake Harris 1999 1994 18.865 15.657 38.518 1986 18.871 15.658 38.520 1972 18.833 15.654 38.491 1961 18.841 15.654 38.507 1950 18.753 15.645 38.505 1941 18.744 15.645 38.566 1898 18.807 15.654 38.676 Lake Harris 2013 1994 18.850 15.659 38.567 1987 18.863 15.658 38.524 1968 18.876 15.659 38.517 1956 18.893 15.660 38.528 1940 18.898 15.660 38.531 1890 18.861 15.657 38.573 Newnans Lake 1997 1995 18.828 15.650 38.497 1992 18.829 15.649 38.492 1983 18.843 15.651 38.495 1977 18.848 15.653 38.502 1974 18.853 15.652 38.501 1970 18.846 15.652 38.505 1966 18.846 15.652 38.499 1960 18.821 15.651 38.494 1956 18.645 15.632 38.394 1951 18.600 15.625 38.369 1946 18.598 15.624 38.371 1940 18.614 15.626 38.385 1921 18.612 15.627 38.387 1906 18.608 15.626 38.391 Newnans Lake 2005 2000 18.822 15.654 38.507 1992 18.832 15.654 38.511 1982 18.842 15.654 38.506 1975 18.854 15.658 38.519 1969 18.883 15.659 38.533 1963 18.891 15.659 38.533 1955 18.892 15.662 38.536 1947 18.860 15.656 38.510 1939 18.654 15.635 38.407 1920 18.608 15.631 38.387 1892 18.949 15.665 38.738 Table 5 4 Stable Pb isotope data from sediment cores collected from Lake Harris (1999 and 2013) and Newnans L ake (1997 and 2005)

PAGE 184

184 Lake Cumulative Mass Unsupported Pb 210 Organic Matter Pb Cu Ni V Zn Sc Nd Lake Harris 1 1.3 1.3 1.5 1.1 1.1 1.2 1.1 1.2 1.2 1.1 Newnans Lake 2 1.6 0.9 0.8 0.6 0.8 0.8 0.6 1.1 0.5 0.6 1 I ndex ratios are the inventory from 1950 1999 for the 1999 core vs. the 2013 core 2 I ndex ratios are the inventory from 1951 1997 for the 1997core vs. the 2005 core Table 5 5. Index ratios of sediment character and trace metal concentration inventories from repeat coring events from Lake Harris (1999 and 2013) and Newnans Lake (1997 and 2005)

PAGE 185

185 CHAPTER 6 SUMMARY It has been argued that metal pollution is one of the earliest typ es of anthropogenic pollution. Records of atmospheric metal pollution from mining and smelting are preserved in Greenland ice layers and date back to Roman times (500 B.C. to A.D. 300). However, since the Industrial Revolution ( ca. AD 1850), metal production and atmospheric emissions have increased substantially (Nriagu 1996). Metal pollution is now ubiquitous, most commonly via atmospheric circulation and deposition (Dillon and E vans 1982). Much of the necessity for studying metal fate and transport stems from known human and ecosystem health implications. Lake sediments are excellent archives for studying regional and temporal deposition of metals Paleolimnological studies allo w for the examination of recent human activities on trace metal cycling, and for the reconstruction of watershed histories that includ e a long term perspective on metal variability, necessary to gauge the magnitude of the human influence (Norton et al 199 2; Bindler et al 2011). This dissertation investigated trace metal s in Florida lake sediments by measuring background or pre anthropogenic metal concentrations, and quantifying anthropogenic impact s on metal inputs. It also investigated the possible futu re environmental impacts of human mediated metal enrichment especially in light of changing climate Using paleolimnological techniques, metal concentration measurements, and Pb isotope analysis, this research contributed new knowledge about concentration s and accumulation rates of metal s in north lacustrine systems The four studies presented investigate long term records and modern abundances, as well as the fate of trace

PAGE 186

186 metals under hydrologic conditions that are expected to be altered as a consequence of climate change in Florida. Trace metal profiles in sediment cores from three north Florida Lakes provided information on local erosion and deposition, regional atmospheric trace metal deposition, and anthropogenic influence on trace metal accumulation. At the base of the cores, metal concentrations were controlled by sediment source and lake morphology. Natural sources of metals to these lakes include the local geology and atmospherically transported dust. Metals including V, Cr, Ni, Cu, Zn, Sn, Sb, B i, Pb show increased up core enrichment beginning ~1900, and coincide with population growth and development in Florida. One known source of these metals is fossil fuel combustion, recorded by changes in Pb concentration and isotope r atios The timing of shifts in Pb accumulation and Pb concentrations was similar to that measured in lightly impacted lakes elsewhere in the United States. Trace metals measured in lake sediment cores from this relatively remote location of north Florida, USA, provide records of; 1) the supply of pre anthropogenic metals, a baseline reference for this location, and 2) a record of recent metal accumulation governed by anthropogenically influenced atmospheric dust deposition. Lake sediments can be sinks for trace metals, as the s e deposits typically remain undisturbed and permanently buried. Climate change scenarios for Florida predict more intense rainy periods, alternating with more severe droughts. A f alling groundwater table combined with increased water need, will likely lea d to lower water levels in shallow Florida lakes. Falling lake levels and even subaerial exposure of sediments may promote physical and/ or chemical redistribution of legacy metal pollution. To test

PAGE 187

187 this hypothesis, I measured metal distributions in transe cts of surficial lake sediments in two north Florida lakes, exposed during extreme dry conditions in 2012 In two of three transects, metals associated with anthropogenic activities V, Cr, Ni, Cu Zn, Sn, Pb, and conservative elements Sc and Nd, exhibited increasing concentrations from deposits near the historic high water level to sediments near the center of the lake. Some basin specific differences in the metal trends were demonstrated, however significant relationships were determined between metals a nd some sediment characteristics. Conservative metals are often bound in the crystalline structure of minerals, whereas metals sourced from anthropogenic activities are often adsorbed onto sediments in hydrous oxides, organic matter, or sulfates. T he main mechanism influencing metal concentrations along the sediment transects was the focusing of fine grain material and organic matter toward the lake center when water levels fell. It is also possible that metals were mobilized to some degree within the sedim ents because of changing pH, redox conditions, and degradation of organic matter associated with the decline in water level S ediment exposed during lake level low stands were also used to determine the source and distribution of Pb in the lake sediments Lead distribution was controlled primarily by the distribution of fine grain and organic matter as t his material was concentrated toward the lake center as water levels fell. Concen trations in deposits closer to the lake center were up to 3x greater than concentrations in sediments near the old shoreline Although redox conditions are known to change in drying sediments, sequential extractions on these sediments provided no evidence for a strong redox gradient along the transects. Therefore, redox pot ential is thought to have had little

PAGE 188

188 influence on Pb distribution. Sequential extraction showed that most Pb is associated with the relatively mobile hydrous Fe Mn oxide sediment fraction. Pb isotopes in the extracted sediment fractions confirmed that Pb i n the exposed sediments was sourced from anthropogenic activity, namely gasoline additive. From a paleolimnological perspective it wa s important to determine if shallow lakes maintain accurate records of tra ce metal accumulation or if lake leve l fluctuations and sediment exposure alter these long term records. W ith respect to metal contamination f uture management of lake sediment and water quality will require accurate records of previous metal loading and hence quantification of potential meta l mobility. To address th e question of how lake stage decrease resulting from low rainfall, may impact the integrity of trace metal stratigraphy in the sediments of shallow Florida lakes, I utilized repeat cores from Lake Harris cored in 1999 and 2013, a nd Newnans Lake cored in 1997 and 2005 T he lakes experienced periods of dramatic stage decline between the years when the repeat cores were collected. Metals commonly attributed to anthropogenic activities, V, Ni, Cu, Zn, and Pb, and conservative element s Sc and Nd, were measured in the cores Metal concentration profiles and inventories were compared between the repeat cores in each lake to determine the degree of similarity. Additionally, radiogenic Pb isotopes were measured to assess Pb mobility. I use d these measures to determine if protracted drying events altered the record of metal deposition in shallow Florida lakes. Lake level variability over the 14 years between coring events did not alter metal accumulation in the sediments of Lake Harris. The consistent discrepancy of calculated inventories in metals and sediment characteristics wa s likely a result of natural system

PAGE 189

189 variability rather than chemical or physical sediment mobility. Results from the Newnans Lake cores are not as definitive as those from Lake Harris, with respect to preservation of metal deposition records. Metal concentration profiles did not vary much throughout each core, so it was difficult to determine if the profiles were post depositionally altered A Pb isotope shift caused by the change in the source of Pb ore used as a gasoline additive in the 1960s was preserved in the sediments of both cores. Preservation of this shift in both cores provides evidence that there was not wholesale redistribution of sediments deposited befo re 1960 in this shallow lake. I t is less certain whether metals in sediments deposited after 1960 were subjected to either chemical or physical remobilization. Extreme lake level decreases will likely cause mobility of recently deposited sediments in F previously deposited metals. Future paleolimnological studies of long term metal deposition in shallow lake systems should measure metals from multiple cores to assess natural variabi lity in metal accumulation between core sites and evaluate evidence for potential metal mobility. F uture management of lake sediment and water quality with respect to metal contamination, will require accurate records of previous metal loading and qu antification of potential metal mobility. These studies serve as a baseline of metal deposition in northern Florida and show that trace metals are enriched by anthropogenic activities. Although the Pb and other anthropogenically sourced metal concentration s measured in these studies were relatively low, they show the potential of lake sediments as sources of legacy metal pollution. The combined effects of changing hydrology and relic metal

PAGE 190

190 pollution may create future management concerns, including legacy me tal pollution in sediments and water, as well as the potential for increased metal laden dust.

PAGE 191

191 APPENDIX A CHAPTER 2 SUPPLEMENTARY MATERIAL Table A 1 Depth (cm) Date Sc V Cr Co Ni Cu Zn Ga Ge Rb Sr Y Zr Lake Sheelar 0 2 2007 2.95 17.0 24.9 2.23 12.9 12.0 117.7 6.81 0.63 5.25 30.1 5.45 61.5 2 4 2003 2.98 15.9 23.5 2.13 11.8 10.8 25.4 6.52 0.71 6.23 29.1 5.92 56.2 4 6 2000 2.98 15.9 23.5 2.13 11.8 10.7 23.8 6.44 0.66 6.27 28.9 6.00 56.4 6 8 1994 2.81 15.8 23.7 2.10 11.7 10.6 21.4 6.58 0.72 4.59 27.1 5.21 57.0 8 10 1986 3.07 16.6 24.7 2.14 12.1 10.8 20.5 6.79 0.83 5.76 30.1 6.25 56.2 10 12 1977 3.08 16.9 24.0 2.07 12.0 10.9 20.3 6.90 0.80 5.57 30.2 6.42 60.1 12 14 1968 2.77 17.6 24.1 2.10 11.9 10.6 19.9 6.84 0.92 3.69 25.3 4.59 60.5 14 16 1961 2.98 16.7 25.0 2.20 11.3 9.57 19.7 7.84 0.78 3.78 30.3 5.52 61.6 16 18 1953 2.86 16.9 25.4 2.22 11.1 9.18 19.8 7.86 0.92 2.91 27.1 4.74 63.0 18 20 1938 3.12 16.2 24.6 2.19 10.7 9.06 18.7 7.69 0.89 3.93 31.9 6.19 61.9 20 22 1923 3.25 16.9 25.2 2.29 10.8 9.46 19.8 7.85 0.92 4.90 34.5 6.64 62.4 22 24 1886 2.50 16.5 25.0 2.18 10.7 8.46 17.2 7.64 0.91 2.00 21.3 2.30 63.6 24 26 3.46 17.1 26.9 2.37 11.4 8.86 16.4 8.94 0.99 4.66 35.7 6.33 65.0 26 28 3.25 16.5 26.0 2.34 10.7 8.43 15.4 8.33 1.09 4.22 33.5 5.63 60.5 28 30 2.82 16.7 25.0 2.32 10.5 8.49 15.0 7.85 0.97 2.64 27.9 4.18 62.2 34 36 2.31 16.7 23.7 2.33 10.0 8.54 15.3 7.01 1.09 1.66 22.3 2.23 58.7 40 42 2.80 15.1 21.8 2.26 9.31 8.38 14.5 6.48 0.98 3.41 30.0 4.60 54.1 46 48 2.53 14.7 19.9 2.22 8.90 8.11 14.3 5.79 0.90 2.86 29.0 3.70 49.8 52 54 2.21 16.1 22.0 2.25 9.47 8.45 14.2 6.24 1.20 1.58 24.7 2.70 55.8 58 60 2.65 15.6 20.7 2.23 8.92 8.64 14.3 6.08 0.92 3.18 30.3 4.48 51.6 64 66 2.91 15.7 21.8 2.23 9.24 8.48 14.2 6.57 1.00 3.55 32.7 6.36 51.3 70 72 2.21 14.9 19.3 2.07 8.62 8.41 14.5 5.47 1.00 2.44 23.8 2.85 46.2 74 76 2.32 14.8 17.9 2.01 7.89 8.11 14.7 4.74 1.05 2.50 24.6 3.44 41.5 78 80 2.21 14.5 19.6 1.96 8.90 8.80 14.2 5.58 0.90 1.72 23.0 2.49 50.4

PAGE 192

192 Table A 1 C ontinu ed Depth (cm) Cd Sn Sb Cs Ba Hf Ta Tl Pb Bi Th U Lake Sheelar 0 2 0.073 1.92 0.23 0.45 51.9 1.72 0.42 0.090 50.7 0.27 4.19 1.50 2 4 0.063 1.69 0.22 0.61 53.3 1.57 0.34 0.088 49.3 0.25 4.41 1.44 4 6 0.065 1.88 0.20 0.66 54.0 1.59 0.34 0.077 49.3 0.24 4.30 1.43 6 8 0.068 2.08 0.21 0.40 43.7 1.60 0.35 0.11 48.2 0.24 4.27 1.45 8 10 0.060 2.05 0.21 0.63 55.6 1.60 0.39 0.089 51.6 0.25 4.47 1.46 10 12 0.056 2.05 0.24 0.62 53.2 1.64 0.36 0.089 56.9 0.25 4.59 1.50 12 14 0.057 1.74 0.24 0.26 37.4 1.72 0.38 0.081 54.1 0.26 4.18 1.59 14 16 0.081 1.27 0.24 0.26 39.5 1.71 0.39 0.075 40.9 0.26 4.78 1.72 16 18 0.080 1.20 0.22 0.17 33.5 1.77 0.42 0.089 34.5 0.26 4.49 1.77 18 20 0.080 1.20 0.24 0.31 43.2 1.70 0.39 0.076 33.6 0.26 4.79 1.72 20 22 0.096 1.25 0.21 0.51 58.9 1.76 0.40 0.077 34.7 0.27 5.05 1.79 22 24 0.086 0.91 0.18 0.13 25.4 1.77 0.40 0.074 26.7 0.24 3.84 1.78 24 26 0.070 0.74 0.15 0.52 58.7 1.81 0.41 0.075 31.9 0.20 5.40 1.79 26 28 0.065 0.57 0.14 0.45 53.7 1.63 0.36 0.072 27.2 0.18 4.80 1.61 28 30 0.071 0.59 0.16 0.19 38.1 1.74 0.38 0.073 25.6 0.18 4.33 1.65 34 36 0.077 0.51 0.16 0.13 35.1 1.61 0.85 0.085 23.5 0.16 3.47 1.61 40 42 0.070 0.45 0.13 0.34 52.8 1.53 0.33 0.065 22.4 0.16 4.32 1.56 46 48 0.070 0.42 0.12 0.24 50.5 1.36 0.33 0.060 19.1 0.15 4.07 1.51 52 54 0.069 0.44 0.14 0.12 40.3 1.49 0.53 0.072 19.7 0.15 4.03 1.66 58 60 0.076 0.42 0.14 0.26 55.5 1.40 0.32 0.066 19.4 0.14 4.03 1.53 64 66 0.075 0.44 0.12 0.39 64.4 1.37 0.32 0.071 21.4 0.14 5.55 1.84 70 72 0.070 0.39 0.13 0.16 45.9 1.27 0.30 0.066 17.6 0.12 3.16 1.41 74 76 0.064 0.33 0.11 0.22 48.5 1.10 0.29 0.053 15.7 0.11 4.27 1.27 78 80 0.060 0.38 0.10 0.13 42.5 1.35 0.31 0.068 18.1 0.14 3.14 1.53

PAGE 193

193 Table A 1 Continued Depth (cm) Date Sc V Cr Co Ni Cu Zn Ga Ge Rb Sr Y Zr Pebble Lake 0 2 2005 2.74 20.2 31.6 2.19 16.2 20.1 77.7 10.6 1.03 10.6 40.2 7.39 66.5 2 4 1994 3.58 25.0 34.6 2.32 17.2 15.7 49.3 13.8 1.37 12.1 39.9 9.39 86.3 4 6 1975 3.72 26.4 35.1 2.31 17.4 14.5 42.7 14.6 1.39 12.0 43.5 9.63 88.3 6 8 1948 3.53 28.2 36.4 2.35 17.9 14.4 43.1 15.1 1.30 5.32 40.2 9.02 98.4 8 10 1899 3.61 28.0 34.5 2.24 17.3 13.4 40.8 14.8 1.38 12.2 41.7 9.37 85.3 10 12 2.63 19.0 24.4 1.77 13.0 11.1 29.7 10.3 1.12 8.21 33.5 7.71 78.6 12 14 2.56 17.1 22.7 1.71 12.1 10.6 22.6 9.31 1.20 9.20 33.4 7.85 75.5 14 16 1.50 16.0 20.9 1.70 10.9 9.54 18.2 7.98 1.12 2.24 17.9 3.73 72.4 16 18 2.26 16.0 22.1 1.77 11.1 10.1 19.8 8.62 1.07 3.46 28.5 6.76 88.0 18 20 2.06 15.5 21.2 1.75 10.7 9.59 18.4 8.19 1.08 2.99 25.8 5.86 80.2 20 22 2.39 15.7 21.3 1.77 10.7 9.99 18.4 8.42 1.05 7.31 31.1 8.13 87.0 22 24 1.21 16.4 21.4 1.64 10.6 10.1 17.9 7.78 1.08 1.70 15.3 2.92 82.2 24 26 2.54 16.7 22.2 1.64 10.4 10.4 19.0 8.48 1.05 8.36 33.7 8.21 85.6 26 28 2.46 16.2 22.3 1.66 10.1 10.4 19.8 7.75 1.05 7.44 32.1 8.83 102.4 28 30 2.11 14.0 19.3 1.41 8.35 9.35 19.8 6.33 0.98 6.91 28.1 8.17 102.3 30 34 2.30 15.8 20.9 1.66 10.1 11.1 20.3 7.65 1.01 6.16 31.9 8.89 94.8 34 38 2.69 17.8 23.1 2.02 11.5 12.0 20.5 8.69 1.01 9.59 34.9 9.20 88.4 38 42 2.69 17.4 23.8 2.00 11.2 11.9 21.3 8.42 1.00 6.50 32.4 8.16 81.6 42 46 2.77 18.8 26.1 2.14 11.9 12.1 22.6 8.68 0.97 5.27 31.8 7.80 83.5 46 50 2.67 18.0 22.3 2.08 10.7 11.5 21.6 8.11 1.06 8.74 30.1 8.60 84.2 50 54 2.92 18.8 24.9 1.90 11.0 11.2 22.5 8.77 1.06 9.15 34.9 9.83 110.4 54 58 2.78 18.9 23.7 1.98 10.9 11.6 23.0 8.34 1.09 8.92 32.6 8.71 89.0 58 62 3.27 20.1 26.5 1.28 11.4 13.1 26.8 9.30 0.87 4.71 33.0 8.06 100.2 62 66 4.05 24.3 31.6 1.72 14.0 14.0 26.3 13.0 1.05 11.4 42.1 10.57 95.2 66 70 3.66 22.6 28.0 1.31 12.5 13.2 24.9 11.1 0.94 10.7 38.7 9.92 86.3 70 74 3.18 21.8 22.6 1.02 10.5 14.7 22.1 8.09 0.85 9.23 37.8 8.58 72.1 74 78 2.67 20.4 20.8 1.02 10.4 14.1 21.1 7.33 0.75 2.46 29.0 5.13 61.6 78 82 2.78 19.7 19.4 1.23 9.99 13.8 24.9 6.79 0.70 7.07 30.4 6.83 59.0

PAGE 194

194 Table A 1 Continued Depth (cm) Cd Sn Sb Cs Ba Hf Ta Tl Pb Bi Th U Pebble Lake 0 2 0.13 1.71 0.28 1.65 109.9 1.87 0.51 0.18 50.6 0.22 5.71 1.42 2 4 0.13 2.26 0.28 1.97 93.8 2.42 0.64 0.25 67.7 0.29 7.67 1.89 4 6 0.11 2.26 0.26 1.90 96.4 2.41 0.68 0.24 71.1 0.29 8.12 1.98 6 8 0.13 2.21 0.30 0.85 65.0 2.72 0.72 0.25 67.6 0.30 6.76 1.98 8 10 0.11 1.92 0.26 1.85 99.0 2.48 0.70 0.31 65.3 0.31 8.40 2.01 10 12 0.078 1.37 0.22 1.47 79.0 2.26 0.57 0.17 39.3 0.25 6.06 1.44 12 14 0.083 1.03 0.18 1.49 83.5 2.10 0.58 0.21 30.7 0.21 6.06 1.36 14 16 0.053 0.77 0.15 0.40 39.0 2.12 0.61 0.12 20.4 0.14 3.75 1.30 16 18 0.053 0.78 0.15 0.88 68.3 2.52 0.62 0.16 21.2 0.14 5.02 1.31 18 20 0.052 0.69 0.14 0.52 55.2 2.28 0.60 0.15 18.5 0.14 4.63 1.25 20 22 0.066 0.81 0.14 1.26 88.1 2.50 0.62 0.16 20.0 0.14 5.87 1.43 22 24 0.061 0.77 0.15 0.36 39.5 2.43 0.66 0.11 17.5 0.14 3.44 1.44 24 26 0.052 0.85 0.15 1.35 101.6 2.57 0.67 0.18 19.9 0.14 6.41 1.52 26 28 0.051 0.84 0.14 1.25 97.6 2.86 0.73 0.12 19.0 0.13 7.13 1.73 28 30 0.056 0.73 0.14 1.02 85.5 2.92 0.73 0.11 15.6 0.12 6.21 1.58 30 34 0.059 0.76 0.13 0.94 85.2 2.68 0.64 0.37 18.1 0.13 6.26 1.53 34 38 0.057 0.78 0.16 1.49 105.6 2.46 0.66 0.15 20.5 0.14 6.50 1.50 38 42 0.067 0.82 0.16 1.29 96.9 2.49 0.68 0.19 20.9 0.14 5.96 1.44 42 46 0.067 0.83 0.16 1.12 93.0 2.31 0.82 0.15 19.9 0.15 5.83 1.46 46 50 0.070 0.78 0.16 1.33 99.1 2.52 0.77 0.16 19.6 0.14 6.68 1.48 50 54 0.067 0.84 0.17 1.45 114.7 3.22 0.89 0.15 20.8 0.22 7.12 1.70 54 58 0.065 0.86 0.18 1.44 104.3 2.63 0.75 0.14 21.1 0.14 6.57 1.54 58 62 0.074 0.88 0.21 1.16 100.7 2.83 0.78 0.18 20.4 0.15 5.24 1.53 62 66 0.091 0.92 0.21 1.87 136.8 2.68 0.70 0.20 26.3 0.17 7.02 1.73 66 70 0.088 0.84 0.21 1.80 136.0 2.49 0.64 0.20 22.7 0.15 6.69 1.57 70 74 0.074 0.76 0.19 1.71 137.9 2.01 0.56 0.20 17.9 0.14 5.17 1.25 74 78 0.079 0.66 0.19 0.79 96.3 1.68 0.50 0.16 16.6 0.13 3.44 0.97 78 82 0.10 0.63 0.16 1.38 109.0 1.66 0.45 0.16 15.4 0.12 4.28 1.02

PAGE 195

195 Table A 1 C ontinued Depth (cm) Date Sc V Cr Co Ni Cu Zn Ga Ge Rb Sr Y Zr Pebble Lake 82 86 2.62 19.3 18.6 1.24 9.70 13.8 26.7 6.47 0.70 7.16 29.9 6.92 57.4 86 90 2.63 18.8 19.5 1.39 9.92 14.0 25.5 6.65 0.70 6.63 29.9 6.96 62.3 90 94 2.77 18.9 20.4 1.43 10.6 14.3 28.0 7.29 0.71 8.00 32.3 7.55 63.5 94 95 3.11 20.8 23.5 1.36 11.8 14.8 27.8 8.76 0.70 9.01 37.6 8.41 68.8 95 102 3.09 21.1 22.8 1.31 11.7 14.6 26.5 8.92 0.72 8.27 38.2 8.06 66.1 102 106 2.75 20.3 20.7 1.62 11.2 14.7 29.6 7.84 0.69 7.99 35.1 7.47 61.9 106 110 2.39 19.4 18.5 1.77 10.4 13.9 32.7 6.77 0.66 7.67 30.8 6.39 52.9 110 114 2.29 19.3 17.9 1.75 10.0 13.0 29.2 6.62 0.56 7.43 30.1 6.68 50.1 114 118 2.36 20.3 18.2 1.70 10.4 13.0 30.4 7.06 0.61 7.47 32.7 6.54 48.9 118 122 1.58 17.3 12.0 1.15 7.89 12.1 29.1 4.65 0.63 4.91 23.8 4.24 27.1 122 126 1.30 19.5 9.55 0.91 7.26 14.8 43.6 3.77 0.59 2.39 20.0 3.31 21.7 126 130 1.30 18.2 9.75 0.85 7.18 13.4 45.0 3.88 0.39 3.33 20.0 3.59 22.6 130 134 1.29 15.6 13.1 0.68 8.82 13.2 34.9 4.26 0.36 4.55 22.2 3.36 23.7 Little Lake Johnson 0 2 2001 2.28 23.0 26.4 0.87 11.78 17.09 45.7 8.83 0.76 8.20 51.9 11.6 74.3 2 4 1980 2.28 23.2 26.6 0.75 11.59 17.16 41.6 8.93 0.80 7.92 50.8 12.2 77.6 4 6 1943 2.28 23.6 25.9 0.74 11.33 16.76 41.4 8.86 0.78 7.57 51.2 12.1 70.5 6 8 1926 1.43 21.9 17.9 0.48 7.94 11.75 26.0 6.41 0.67 5.18 43.0 9.86 53.1 8 10 1901 1.05 21.5 14.4 0.38 6.43 9.76 16.5 5.18 0.62 4.22 39.6 8.74 40.0 10 12 1.20 21.8 15.3 0.45 6.79 10.33 15.3 5.41 0.62 4.40 41.1 9.03 43.7 12 14 1.19 20.2 16.1 0.54 6.60 9.66 14.9 5.30 0.65 4.61 38.7 8.52 47.1 14 16 1.14 17.2 15.1 0.61 6.25 9.01 14.5 5.01 0.62 4.62 35.7 7.97 45.3 16 18 1.25 17.5 16.6 0.71 6.62 9.17 15.7 5.28 0.64 4.92 36.1 8.10 51.7 18 20 1.37 17.4 16.9 1.07 6.69 9.58 15.7 5.59 0.64 5.08 36.1 8.36 55.4 20 22 1.02 15.8 15.4 0.81 6.05 8.41 14.9 4.99 0.65 4.57 31.6 7.48 49.5

PAGE 196

196 Table A 1 Continued Depth (cm) Cd Sn Sb Cs Ba Hf Ta Tl Pb Bi Th U Pebble Lake 82 86 0.11 0.62 0.18 1.29 104.5 1.66 0.45 0.16 15.1 0.12 4.82 1.03 86 90 0.10 0.64 0.19 1.24 103.4 1.74 0.48 0.16 15.6 0.12 4.41 1.09 90 94 0.10 0.76 0.20 1.41 113.2 1.77 0.51 0.18 17.0 0.12 4.80 1.09 94 95 0.10 0.81 0.16 1.67 130.7 1.88 0.54 0.19 20.4 0.15 5.33 1.19 95 102 0.10 0.83 0.19 1.63 134.9 1.80 0.53 0.19 20.1 0.16 5.00 1.13 102 106 0.12 0.73 0.19 1.56 122.0 1.64 0.45 0.21 18.1 0.13 4.67 1.03 106 110 0.15 0.54 0.19 1.39 106.0 1.42 0.37 0.18 16.2 0.12 4.10 0.88 110 114 0.14 0.57 0.18 1.39 105.3 1.34 0.36 0.24 15.6 0.11 5.50 0.94 114 118 0.14 0.58 0.19 1.51 115.1 1.31 0.36 0.18 16.3 0.13 4.18 0.90 118 122 0.14 0.34 0.17 0.94 82.5 0.75 0.21 0.13 10.5 0.09 2.71 0.60 122 126 0.20 0.28 0.20 0.62 67.7 0.60 0.16 0.16 8.38 0.08 1.92 0.50 126 130 0.19 0.26 0.18 0.69 68.8 0.61 0.15 0.16 8.68 0.08 2.14 0.50 130 134 0.13 0.31 0.15 0.84 72.9 0.62 0.18 0.16 9.35 0.08 2.64 0.52 Little Lake Johnson 0 2 0.19 2.95 0.25 1.28 108.5 2.05 0.63 0.19 63.6 0.30 6.82 2.50 2 4 0.21 2.54 0.24 1.27 103.5 2.18 0.63 0.22 66.3 0.33 7.21 2.63 4 6 0.19 2.47 0.24 1.22 105.0 2.04 0.62 0.16 65.1 0.31 7.10 2.64 6 8 0.13 1.32 0.21 0.89 98.6 1.54 0.49 0.12 34.5 0.26 5.32 2.18 8 10 0.10 0.74 0.19 0.70 90.6 1.27 0.41 0.20 19.4 0.18 4.46 1.94 10 12 0.11 0.80 0.20 0.78 93.2 1.39 0.44 0.13 20.1 0.19 4.82 2.00 12 14 0.10 0.70 0.19 0.79 89.8 1.40 0.46 0.14 17.9 0.16 4.64 1.83 14 16 0.076 0.66 0.16 0.73 82.1 1.41 0.47 0.10 16.9 0.12 4.65 1.70 16 18 0.10 0.88 0.18 0.81 86.8 1.58 0.51 0.12 17.9 0.14 4.77 1.71 18 20 0.089 0.76 0.16 0.85 84.0 1.63 0.54 0.11 16.6 0.13 5.29 1.79 20 22 0.092 0.67 0.15 0.76 76.2 1.47 0.49 0.11 14.9 0.13 4.94 1.57

PAGE 197

197 Table A 1 C ontinued Depth (cm) Date Sc V Cr Co Ni Cu Zn Ga Ge Rb Sr Y Zr Little Lake Johnson 22 24 0.77 14.3 14.3 0.69 5.65 8.07 14.3 4.46 0.58 4.15 29.1 6.55 42.3 24 26 1.29 12.3 13.2 0.77 5.37 8.16 13.5 3.45 0.56 3.55 23.1 5.93 41.9 26 28 1.82 10.3 12.0 0.85 5.08 8.25 12.7 2.44 0.53 2.95 17.1 5.31 41.6 28 30 1.67 8.67 10.9 0.42 4.41 7.95 14.3 1.49 0.51 2.42 13.4 4.43 35.0 30 34 1.52 7.00 9.76 0.013 3.75 7.65 15.8 0.54 0.49 1.89 9.64 3.56 28.5 34 38 0.81 2.08 4.91 0.76 1.86 5.41 11.3 1.32 0.50 0.57 1.41 0.83 15.2 38 42 0.94 3.00 7.06 0.62 2.20 5.97 12.7 0.99 0.41 0.86 2.69 2.28 15.8 42 46 0.84 2.37 5.47 0.65 1.77 5.48 16.5 1.12 0.48 0.44 0.69 0.76 12.3 46 50 0.88 3.62 5.82 0.75 1.73 5.03 13.4 0.89 0.48 0.40 0.84 0.76 16.3 50 54 0.79 1.72 5.35 0.90 1.83 5.33 15.7 1.29 0.51 0.43 2.19 1.08 16.7 Table A 1 Continued Depth (cm) Cd Sn Sb Cs Ba Hf Ta Tl Pb Bi Th U Little Lake Johnson 22 24 0.073 0.62 0.12 0.68 65.5 1.31 0.44 0.091 12.9 0.10 4.51 1.40 24 26 0.069 0.51 0.11 0.56 52.3 1.22 0.40 0.074 9.99 0.091 3.94 1.24 26 28 0.065 0.41 0.10 0.43 39.1 1.12 0.36 0.057 7.07 0.076 3.37 1.08 28 30 0.055 0.32 0.087 0.36 30.8 0.98 0.33 0.052 6.09 0.062 3.34 0.92 30 34 0.044 0.23 0.074 0.28 22.5 0.85 0.29 0.047 5.12 0.048 3.31 0.75 34 38 0.017 0.01 0.032 0.077 2.36 0.52 0.13 0.031 2.34 0.005 0.68 0.25 38 42 0.019 0.18 0.040 0.12 5.67 0.52 0.14 0.032 2.86 0.020 2.96 0.83 42 46 0.015 0.16 0.030 0.057 3.64 0.46 0.24 0.028 2.50 0.025 0.62 0.24 46 50 0.011 0.22 0.038 0.049 1.23 0.60 0.26 0.022 2.39 0.006 1.19 0.25 50 54 0.003 0.04 0.030 0.059 2.39 0.60 0.12 0.024 3.15 0.009 2.17 0.64

PAGE 198

198 Table A 2 Scandium normalized (M/Sc) concentrations and calculated enrichment factors (EF) for selected trace metals Depth (cm) V Sc V EF Cr Sc Cr EF Co Sc Co EF Ni Sc Ni EF Cu Sc Cu EF Zn Sc Zn EF Sr Sc Sr EF Y Sc Y EF Zr Sc Lake Sheelar 0 2 5.75 0.92 8.44 1.02 0.76 0.86 4.36 1.20 4.06 1.17 39.9 6.76 10.2 0.92 1.85 1.21 20.9 20 5 .35 0.86 7.91 0.95 0.71 0.81 3.96 1.09 3.64 1.05 8.54 1.45 9.76 0.88 1.99 1.30 18.9 4 6 5 .35 0.86 7.89 0.95 0.72 0.81 3.95 1.09 3.58 1.03 8.00 1.36 9.71 0.88 2.02 1.32 18.9 6 8 5 .62 0.90 8.43 1.02 0.75 0.85 4.18 1.15 3.78 1.09 7.60 1.29 9.66 0.87 1.85 1.21 20.3 8 10 5.41 0.87 8.04 0.97 0.70 0.79 3.95 1.09 3.52 1.02 6.70 1.14 9.82 0.89 2.04 1.33 18.3 10 12 5.47 0.88 7.78 0.94 0.67 0.76 3.90 1.07 3.53 1.02 6.58 1.11 9.80 0.89 2.08 1.36 19.5 12 14 6.35 1.02 8.69 1.05 0.76 0.86 4.29 1.18 3.82 1.10 7.20 1.22 9.12 0.83 1.66 1.09 21.9 14 16 5.59 0.90 8.38 1.01 0.74 0.84 3.80 1.04 3.21 0.93 6.62 1.12 10.2 0.92 1.85 1.21 20.7 16 18 5.89 0.94 8.87 1.07 0.77 0.88 3.89 1.07 3.21 0.93 6.92 1.17 9.47 0.86 1.66 1.08 22.0 18 20 5.18 0.83 7.88 0.95 0.70 0.80 3.42 0.94 2.90 0.84 5.97 1.01 10.2 0.92 1.98 1.30 19.8 20 22 5.19 0.83 7.76 0.94 0.71 0.80 3.33 0.91 2.91 0.84 6.09 1.03 10.6 0.96 2.04 1.34 19.2 22 24 6.59 1.05 9.97 1.20 0.87 0.99 4.25 1.17 3.38 0.98 6.88 1.17 8.52 0.77 0.92 0.60 25.4 24 26 4.94 0.79 7.78 0.94 0.68 0.78 3.30 0.91 2.56 0.74 4.73 0.80 10.3 0.93 1.83 1.20 18.8 26 28 5.09 0.82 8.00 0.97 0.72 0.82 3.31 0.91 2.60 0.75 4.73 0.80 10.3 0.94 1.73 1.14 18.6 28 30 5.93 0.95 8.87 1.07 0.82 0.94 3.73 1.02 3.01 0.87 5.32 0.90 9.89 0.90 1.48 0.97 22.0 34 36 7.24 1.16 10.3 1.24 1.01 1.15 4.35 1.19 3.70 1.07 6.62 1.12 9.68 0.88 0.96 0.63 25.4 40 42 5.41 0.87 7.78 0.94 0.81 0.92 3.33 0.92 3.00 0.87 5.20 0.88 10.7 0.97 1.65 1.08 19.4 46 48 5.83 0.93 7.86 0.95 0.88 1.00 3.52 0.97 3.21 0.93 5.68 0.96 11.5 1.04 1.47 0.96 19.7 52 54 7.29 1.17 9.95 1.20 1.02 1.16 4.28 1.18 3.82 1.10 6.43 1.09 11.2 1.01 1.22 0.80 25.3 58 60 5.89 0.94 7.81 0.94 0.84 0.96 3.37 0.93 3.27 0.94 5.41 0.92 11.4 1.04 1.69 1.11 19.5 64 66 5.40 0.87 7.49 0.90 0.77 0.87 3.18 0.87 2.92 0.84 4.87 0.83 11.3 1.02 2.19 1.43 17.7 70 72 6.76 1.08 8.74 1.05 0.94 1.06 3.90 1.07 3.80 1.10 6.54 1.11 10.8 0.98 1.29 0.84 20.9 74 76 6.38 1.02 7.72 0.93 0.87 0.99 3.40 0.94 3.49 1.01 6.35 1.08 10.6 0.96 1.48 0.97 17.9 78 80 6.54 1.05 8.85 1.07 0.89 1.01 4.02 1.11 3.98 1.15 6.43 1.09 10.4 0.94 1.12 0.74 22.8

PAGE 199

1 99 Table A 2 Continued Depth (cm) Cd Sc Cd EF Sn Sc Sn EF Sb Sc Sb EF Ba Sc Ba EF Pb Sc Pb EF Bi Sc Bi EF Th Sc Th EF U Sc U EF Lake Sheelar 0 2 0.025 0.87 0.65 3.92 0.076 1.51 17.58 0.86 17.2 2.23 0.092 1.63 1.42 0.86 0.51 0.80 20 0.021 0.75 0.57 3.43 0.074 1.47 17.90 0.88 16.6 2.15 0.083 1.48 1.48 0.89 0.48 0.77 4 6 0.022 0.77 0.63 3.80 0.068 1.34 18.14 0.89 16.6 2.15 0.080 1.42 1.45 0.87 0.48 0.76 6 8 0.024 0.85 0.74 4.45 0.075 1.48 15.55 0.76 17.1 2.23 0.086 1.53 1.52 0.91 0.52 0.82 8 10 0.020 0.69 0.67 4.02 0.069 1.36 18.13 0.89 16.8 2.19 0.082 1.45 1.46 0.88 0.48 0.76 10 12 0.01 8 0.64 0.67 4.02 0.078 1.55 17.25 0.85 18.4 2.40 0.083 1.47 1.49 0.90 0.49 0.77 12 14 0.02 1 0.73 0.63 3.79 0.087 1.71 13.52 0.66 19.5 2.54 0.093 1.66 1.51 0.91 0.58 0.91 14 16 0.027 0.95 0.43 2.58 0.081 1.60 13.27 0.65 13.7 1.78 0.087 1.55 1.60 0.97 0.58 0.92 16 18 0.02 8 0.98 0.42 2.53 0.075 1.49 11.69 0.57 12.0 1.57 0.092 1.64 1.57 0.95 0.62 0.98 18 20 0.02 6 0.90 0.38 2.31 0.077 1.52 13.82 0.68 10.8 1.40 0.083 1.48 1.53 0.92 0.55 0.87 20 22 0.030 1.04 0.39 2.33 0.066 1.30 18.12 0.89 10.7 1.39 0.083 1.48 1.56 0.94 0.55 0.87 22 24 0.03 4 1.21 0.36 2.20 0.073 1.45 10.15 0.50 10.6 1.38 0.097 1.73 1.53 0.92 0.71 1.13 24 26 0.02 0 0.71 0.21 1.29 0.042 0.84 16.95 0.83 9.23 1.20 0.059 1.05 1.56 0.94 0.52 0.82 26 28 0.020 0.71 0.17 1.05 0.043 0.85 16.52 0.81 8.39 1.09 0.054 0.96 1.48 0.89 0.49 0.78 28 30 0.02 5 0.88 0.21 1.25 0.058 1.15 13.51 0.66 9.08 1.18 0.064 1.14 1.54 0.93 0.59 0.93 34 36 0.03 4 1.18 0.22 1.33 0.069 1.36 15.22 0.75 10.2 1.32 0.071 1.27 1.51 0.91 0.70 1.10 40 42 0.025 0.88 0.16 0.97 0.045 0.89 18.90 0.93 8.00 1.04 0.056 1.00 1.54 0.93 0.56 0.89 46 48 0.02 8 0.98 0.17 1.01 0.048 0.94 19.98 0.98 7.55 0.98 0.058 1.03 1.61 0.97 0.60 0.95 52 54 0.03 1 1.10 0.20 1.20 0.066 1.29 18.22 0.89 8.92 1.16 0.069 1.22 1.82 1.10 0.75 1.19 58 60 0.02 9 1.00 0.16 0.95 0.051 1.01 20.96 1.03 7.34 0.95 0.054 0.96 1.52 0.92 0.58 0.92 64 66 0.02 6 0.91 0.15 0.91 0.043 0.84 22.16 1.09 7.37 0.96 0.049 0.87 1.91 1.15 0.63 1.00 70 72 0.03 2 1.12 0.18 1.07 0.057 1.13 20.74 1.02 7.95 1.03 0.056 1.00 1.43 0.86 0.64 1.01 74 76 0.028 0.97 0.14 0.87 0.048 0.95 20.90 1.02 6.79 0.88 0.048 0.85 1.84 1.11 0.55 0.87 78 80 0.02 7 0.95 0.17 1.03 0.045 0.89 19.20 0.94 8.17 1.06 0.063 1.11 1.42 0.85 0.69 1.10

PAGE 200

200 Table A 2 Continued Depth (cm) V Sc V EF Cr Sc Cr EF Co Sc Co EF Ni Sc Ni EF Cu Sc Cu EF Zn Sc Zn EF Sr Sc Sr EF Y Sc Y EF Zr Sc Pebble Lake 0 2 7.36 1.11 11.5 1.31 0.80 1.12 5.90 1.44 7.34 1.71 28.3 3.49 14.7 1.18 2.69 0.81 24.2 2 4 6.99 1.05 9.68 1.10 0.65 0.91 4.81 1.17 4.40 1.02 13.8 1.70 11.2 0.90 2.63 0.79 24.1 4 6 7.09 1.07 9.44 1.07 0.62 0.87 4.69 1.15 3.91 0.91 11.5 1.42 11.7 0.94 2.59 0.78 23.7 6 8 7.98 1.20 10.3 1.17 0.66 0.93 5.06 1.23 4.07 0.95 12.2 1.50 11.4 0.92 2.55 0.77 27.8 8 10 7.76 1.17 9.57 1.09 0.62 0.87 4.79 1.17 3.70 0.86 11.3 1.40 11.6 0.93 2.60 0.78 23.6 10 12 7.25 1.09 9.27 1.05 0.68 0.95 4.97 1.21 4.25 0.99 11.3 1.40 12.8 1.03 2.93 0.88 29.9 12 14 6.68 1.01 8.87 1.01 0.67 0.94 4.72 1.15 4.15 0.96 8.85 1.09 13.1 1.05 3.07 0.92 29.5 14 16 6.64 1.60 8.69 1.58 0.71 1.59 4.53 1.77 3.96 1.48 7.55 1.49 7.5 0.96 1.55 0.74 30.1 16 18 7.08 1.06 9.77 1.11 0.78 1.10 4.92 1.20 4.49 1.04 8.75 1.08 12.6 1.01 2.99 0.90 38.9 18 20 7.54 1.13 10.3 1.17 0.85 1.19 5.18 1.27 4.66 1.08 8.96 1.10 12.5 1.01 2.85 0.85 38.9 20 22 6.56 0.99 8.89 1.01 0.74 1.04 4.49 1.10 4.17 0.97 7.68 0.95 13.0 1.05 3.40 1.02 36.3 22 24 6.66 2.05 8.68 2.01 0.67 1.91 4.29 2.14 4.11 1.95 7.24 1.82 6.2 1.02 1.18 0.73 33.3 24 26 6.56 0.99 8.73 0.99 0.64 0.90 4.11 1.00 4.11 0.95 7.46 0.92 13.2 1.07 3.23 0.97 33.7 26 28 6.59 0.99 9.06 1.03 0.68 0.95 4.09 1.00 4.25 0.99 8.08 1.00 13.1 1.05 3.59 1.08 41.6 28 30 6.63 1.00 9.15 1.04 0.67 0.94 3.95 0.96 4.42 1.03 9.35 1.15 13.3 1.07 3.87 1.16 48.4 30 34 6.87 1.03 9.10 1.03 0.72 1.01 4.39 1.07 4.83 1.12 8.83 1.09 13.9 1.11 3.86 1.16 41.2 34 38 6.60 0.99 8.58 0.97 0.75 1.05 4.28 1.05 4.45 1.03 7.62 0.94 13.0 1.04 3.42 1.03 32.8 38 42 6.48 0.97 8.87 1.01 0.75 1.05 4.17 1.02 4.43 1.03 7.91 0.97 12.0 0.97 3.03 0.91 30.4 42 46 6.79 1.02 9.41 1.07 0.77 1.08 4.30 1.05 4.37 1.02 8.17 1.01 11.5 0.92 2.82 0.85 30.1 46 50 6.74 1.01 8.35 0.95 0.78 1.09 4.00 0.98 4.29 1.00 8.11 1.00 11.3 0.91 3.22 0.97 31.6 50 54 6.45 0.97 8.52 0.97 0.65 0.91 3.76 0.92 3.85 0.90 7.70 0.95 12.0 0.96 3.37 1.01 37.8 54 58 6.80 1.02 8.52 0.97 0.71 1.00 3.94 0.96 4.16 0.97 8.28 1.02 11.7 0.94 3.13 0.94 32.0 58 62 6.13 -8.09 -0.39 -3.49 -4.01 -8.19 -10.1 -2.46 -30.6 62 66 6.01 -7.81 -0.42 -3.46 -3.45 -6.49 -10.4 -2.61 -23.5 66 70 6.16 -7.66 -0.36 -3.42 -3.60 -6.80 -10.6 -2.71 -23.6 70 74 6.85 -7.11 -0.32 -3.32 -4.62 -6.96 -11.9 -2.70 -22.7

PAGE 201

201 Table A 2 Continued Depth (cm) Cd Sc Cd EF Sn Sc Sn EF Sb Sc Sb EF Ba Sc Ba EF Pb Sc Pb EF Bi Sc Bi EF Th Sc Th EF U Sc U EF Pebble Lake 0 2 0.048 2.05 0.62 2.00 0.10 1.73 40.0 1.06 18.4 2.45 0.081 1.45 2.08 0.84 0.52 0.87 2 4 0.036 1.52 0.63 2.03 0.079 1.32 26.2 0.69 18.9 2.52 0.081 1.44 2.15 0.86 0.53 0.88 4 6 0.029 1.24 0.61 1.95 0.070 1.17 25.9 0.68 19.1 2.54 0.078 1.40 2.19 0.88 0.53 0.89 6 8 0.037 1.58 0.63 2.01 0.085 1.42 18.4 0.49 19.1 2.54 0.084 1.51 1.91 0.77 0.56 0.94 8 10 0.030 1.27 0.53 1.71 0.072 1.20 27.4 0.72 18.1 2.40 0.086 1.53 2.33 0.93 0.56 0.94 10 12 0.030 1.27 0.52 1.67 0.083 1.40 30.1 0.79 15.0 1.99 0.094 1.68 2.31 0.93 0.55 0.92 12 14 0.033 1.38 0.40 1.29 0.071 1.19 32.7 0.86 12.0 1.60 0.083 1.47 2.37 0.95 0.53 0.89 14 16 0.022 1.50 0.32 1.64 0.060 1.62 16.2 0.68 8.49 1.81 0.060 1.71 1.56 1.00 0.54 1.45 16 18 0.023 0.99 0.34 1.11 0.067 1.11 30.2 0.80 9.36 1.24 0.064 1.14 2.22 0.89 0.58 0.97 18 20 0.025 1.07 0.34 1.08 0.068 1.13 26.8 0.71 8.97 1.19 0.066 1.19 2.25 0.90 0.60 1.01 20 22 0.028 1.17 0.34 1.08 0.058 0.97 36.8 0.97 8.34 1.11 0.058 1.04 2.45 0.98 0.60 1.00 22 24 0.025 2.16 0.31 2.06 0.060 2.05 16.0 0.86 7.08 1.92 0.059 2.14 1.40 1.14 0.58 2.00 24 26 0.021 0.88 0.33 1.07 0.059 0.99 40.0 1.05 7.82 1.04 0.056 0.99 2.52 1.01 0.60 1.00 26 28 0.021 0.87 0.34 1.10 0.057 0.96 39.7 1.05 7.74 1.03 0.053 0.94 2.90 1.16 0.70 1.18 28 30 0.027 1.13 0.34 1.10 0.066 1.11 40.4 1.07 7.38 0.98 0.057 1.01 2.94 1.18 0.75 1.25 30 34 0.026 1.08 0.33 1.06 0.058 0.97 37.0 0.98 7.87 1.04 0.056 1.01 2.72 1.09 0.66 1.11 34 38 0.021 0.90 0.29 0.93 0.059 0.99 39.2 1.03 7.61 1.01 0.051 0.91 2.41 0.97 0.56 0.93 38 42 0.025 1.06 0.30 0.97 0.060 1.00 36.0 0.95 7.76 1.03 0.053 0.95 2.22 0.89 0.54 0.90 42 46 0.024 1.02 0.30 0.96 0.056 0.94 33.6 0.89 7.17 0.95 0.054 0.97 2.10 0.84 0.53 0.88 46 50 0.026 1.11 0.29 0.94 0.059 0.98 37.1 0.98 7.34 0.97 0.051 0.92 2.50 1.00 0.56 0.93 50 54 0.023 0.97 0.29 0.92 0.060 1.00 39.3 1.04 7.13 0.95 0.076 1.36 2.44 0.98 0.58 0.97 54 58 0.023 0.99 0.31 0.99 0.064 1.07 37.5 0.99 7.57 1.01 0.051 0.91 2.36 0.95 0.55 0.93 58 62 0.023 -0.27 -0.064 -30.8 -6.23 -0.045 -1.60 -0.47 -62 66 0.022 -0.23 -0.053 -33.8 -6.50 -0.041 -1.73 -0.43 -66 70 0.024 -0.23 -0.057 -37.1 -6.19 -0.041 -1.83 -0.43 -70 74 0.023 -0.24 -0.060 -43.4 -5.64 -0.045 -1.63 -0.39 -

PAGE 202

202 Table A 2 C ontinued Depth (cm) V Sc V EF Cr Sc Cr EF Co Sc Co EF Ni Sc Ni EF Cu Sc Cu EF Zn Sc Zn EF Sr Sc Sr EF Y Sc Y EF Zr Sc Pebble Lake 74 78 7.64 -7.78 -0.38 -3.91 -5.29 -7.89 -10.9 -1.92 -23.1 78 82 7.10 -7.01 -0.44 -3.60 -4.98 -8.97 -11.0 -2.46 -21.2 82 86 7.36 -7.09 -0.47 -3.70 -5.25 -10.2 -11.4 -2.64 -21.9 86 90 7.16 -7.43 -0.53 -3.77 -5.31 -9.69 -11.4 -2.65 -23.7 90 94 6.85 -7.35 -0.52 -3.82 -5.17 -10.1 -11.7 -2.73 -22.9 94 95 6.68 -7.57 -0.44 -3.78 -4.77 -8.94 -12.1 -2.71 -22.1 95 102 6.83 -7.39 -0.42 -3.78 -4.74 -8.58 -12.4 -2.61 -21.4 102 106 7.39 -7.52 -0.59 -4.09 -5.35 -10.8 -12.8 -2.72 -22.5 106 110 8.11 -7.74 -0.74 -4.36 -5.83 -13.7 -12.9 -2.68 -22.2 110 114 8.44 -7.80 -0.76 -4.38 -5.68 -12.7 -13.2 -2.92 -21.9 114 118 8.60 -7.73 -0.72 -4.43 -5.50 -12.9 -13.9 -2.77 -20.8 118 122 10.9 -7.60 -0.73 -4.99 -7.67 -18.4 -15.0 -2.68 -17.1 122 126 14.9 -7.32 -0.69 -5.56 -11.3 -33.4 -15.3 -2.54 -16.6 126 130 14.1 -7.52 -0.66 -5.54 -10.3 -34.7 -15.4 -2.77 -17.5 130 134 12.1 -10.1 -0.52 -6.82 -10.2 -27.0 -17.2 -2.60 -18.3 Little Lake Johnson 0 2 10.1 0.73 11.6 0.87 0.38 0.56 5.17 0.96 7.49 0.98 20.1 1.55 22.8 0.81 5.10 0.79 32.6 2 4 10.2 0.74 11.7 0.87 0.33 0.49 5.09 0.95 7.54 0.98 18.3 1.41 22.3 0.80 5.37 0.83 34.1 4 6 10.4 0.75 11.4 0.85 0.33 0.48 4.97 0.93 7.36 0.96 18.2 1.40 22.5 0.80 5.32 0.82 31.0 6 8 15.3 1.11 12.5 0.94 0.33 0.49 5.54 1.04 8.21 1.07 18.2 1.40 30.0 1.07 6.89 1.06 37.1 8 10 20.5 1.48 13.7 1.03 0.36 0.53 6.14 1.15 9.32 1.22 15.7 1.21 37.8 1.35 8.34 1.28 38.2 10 12 18.2 1.32 12.8 0.95 0.37 0.55 5.67 1.06 8.64 1.13 12.8 0.99 34.4 1.23 7.55 1.16 36.6 12 14 17.1 1.24 13.6 1.01 0.45 0.66 5.57 1.04 8.15 1.06 12.6 0.97 32.6 1.16 7.19 1.11 39.7 14 16 15.0 1.09 13.2 0.99 0.54 0.79 5.47 1.02 7.89 1.03 12.7 0.98 31.3 1.12 6.98 1.08 39.7 16 18 14.1 1.02 13.3 0.99 0.57 0.84 5.31 0.99 7.36 0.96 12.6 0.97 28.9 1.03 6.50 1.00 41.5

PAGE 203

203 Table A 2 Continued Depth (cm) V Sc V EF Cr Sc Cr EF Co Sc Co EF Ni Sc Ni EF Cu Sc Cu EF Zn Sc Zn EF Sr Sc Sr EF Y Sc Y EF Zr Sc Little Lake Johnson 18 20 12.8 0.92 12.4 0.93 0.78 1.15 4.90 0.91 7.01 0.92 11.5 0.89 26.4 0.94 6.12 0.94 40.6 20 22 15.4 1.11 15.1 1.13 0.79 1.16 5.91 1.10 8.21 1.07 14.6 1.12 30.9 1.10 7.31 1.13 48.4 22 24 18.6 1.35 18.6 1.39 0.90 1.31 7.36 1.37 10.5 1.37 18.6 1.43 37.9 1.35 8.54 1.31 55.1 24 26 9.54 0.69 10.2 0.76 0.60 0.87 4.15 0.78 6.31 0.82 10.4 0.81 17.9 0.64 4.59 0.71 32.5 26 28 5.69 -6.62 -0.47 -2.80 -4.54 -7.00 -9.40 -2.92 -22.9 28 30 5.20 -6.53 -0.25 -2.65 -4.76 -8.56 -8.01 -2.66 -21.0 30 34 4.61 -6.42 -0.009 -2.47 -5.03 -10.4 -6.35 -2.34 -18.8 34 38 2.57 -6.05 -0.94 -2.29 -6.67 -14.0 -1.73 -1.03 -18.8 38 42 3.20 -7.52 -0.66 -2.34 -6.37 -13.6 -2.87 -2.43 -16.9 42 46 2.81 -6.49 -0.77 -2.10 -6.50 -19.6 -0.82 -0.91 -14.6 46 50 4.11 -6.60 -0.85 -1.96 -5.71 -15.2 -0.96 -0.86 -18.5 50 54 2.19 -6.80 -1.14 -2.33 -6.77 -19.9 -2.79 -1.38 -21.2

PAGE 204

204 Table A 2 C ontinued Depth (cm) Cd Sc Cd EF Sn Sc Sn EF Sb Sc Sb EF Ba Sc Ba EF Pb Sc Pb EF Bi Sc Bi EF Th Sc Th EF U Sc U EF Litt le Lake Johnson 18 20 0.065 0.89 0.55 0.92 0.12 0.92 61.5 0.94 12.2 0.93 0.10 0.92 3.87 0.94 1.31 0.95 20 22 0.090 1.22 0.65 1.09 0.14 1.12 74.4 1.14 14.5 1.11 0.13 1.21 4.82 1.17 1.53 1.11 22 24 0.095 1.29 0.80 1.34 0.16 1.23 85.3 1.30 16.8 1.29 0.14 1.30 5.87 1.43 1.82 1.33 24 26 0.053 0.73 0.40 0.66 0.09 0.67 40.5 0.62 7.73 0.59 0.070 0.67 3.05 0.74 0.96 0.70 26 28 0.036 -0.22 -0.05 -21.5 -3.89 -0.042 -1.85 -0.59 -28 30 0.033 -0.19 -0.05 -18.5 -3.65 -0.037 -2.00 -0.55 -30 34 0.029 -0.15 -0.05 -14.8 -3.37 -0.032 -2.18 -0.50 -34 38 0.022 -0.01 -0.04 -2.91 -2.89 -0.006 -0.84 -0.30 -38 42 0.020 -0.20 -0.04 -6.05 -3.05 -0.021 -3.16 -0.89 -42 46 0.017 -0.19 -0.04 -4.32 -2.97 -0.029 -0.73 -0.28 -46 50 0.013 -0.25 -0.04 -1.39 -2.71 -0.007 -1.35 -0.28 -50 54 0.004 -0.05 -0.04 -3.03 -4.01 -0.011 -2.76 -0.82 -

PAGE 205

205 Table A 3 208 Pb/ 204 Pb, 206 Pb/ 204 Pb, and 207 Pb/ 204 Pb values at selected depths and corresponding dates from the Lake Sheelar, Pebble Lake and Little Lake Johnson sediment cores. Dates are for bottom of depth interval Depth Date 206 Pb/ 204 Pb 207 Pb/ 204 Pb 208 Pb/ 204 Pb Lake Sheelar 0 2 2007 18.737 15.642 38.433 2 4 2003 18.736 15.641 38.436 6 8 1994 18.741 15.643 38.443 10 12 1977 18.772 15.649 38.458 12 14 1968 18.737 15.641 38.431 14 16 1961 18.669 15.636 38.447 16 18 1953 18.685 15.637 38.481 18 20 1938 18.701 15.636 38.496 24 26 18.722 15.640 38.523 34 36 18.716 15.638 38.520 52 54 18.745 15.642 38.588 64 66 18.758 15.645 38.608 74 76 18.749 15.647 38.654 78 80 18.721 15.640 38.518 Pebble Lake 0 2 2005 18.699 15.639 38.442 2 4 1994 18.702 15.640 38.444 4 6 1975 18.703 15.639 38.446 6 8 1948 18.695 15.639 38.442 8 10 1899 18.670 15.636 38.444 10 12 18.724 15.640 38.534 12 14 18.755 15.641 38.589 14 16 18.815 15.650 38.679 16 18 18.828 15.651 38.684 18 20 18.853 15.654 38.715 Little Lake Johnson 0 2 2001 18.751 15.643 38.476 2 4 1980 18.753 15.644 38.477 4 6 1943 18.754 15.643 38.485 6 8 1926 18.797 15.647 38.599 8 10 1902 18.876 15.653 38.766 10 12 18.880 15.653 38.772 12 14 18.905 15.656 38.795 22 24 18.915 15.659 38.830

PAGE 206

206 APPENDIX B CHAPTER 4 SUPPLEMENTARY MATERIAL Table B 1 Descriptive statistics for pH, percent water and organic matter, and total Pb concentrations ( g/g) from all samples from Little Lake Johnson (n=33) and Lake Geneva (n=36). Descriptive statistics for pH, percent water and organic matter, and Pb concentrations ( g/g) from sampling locations at 10 m intervals along transects from the historic high water shoreline (0 m) at Littl e Lake Johnson and Lake Geneva (n=3). Values are means one standard deviation. Grain size distribution, > and < 63 m, was measured from one sample from each sampling location. Distance from shore (m) pH Water weight (%) LOI (%) > 63 m < 63 m Pb SD SD SD (%) (%) SD Little Lake Johnson all 5.12 0.178 6.35 7.45 33.3 15.6 --41.0 15.9 Lake Geneva all 5.53 0.442 29.5 16.7 13.1 4.69 --40.3 15.5 Little Lake Johnson 0 5.29 0.150 1.22 0.678 10.1 14.0 84.8 2.35 17.6 13.0 10 5.30 0.110 0.930 0.386 10.1 5.16 79.3 10.6 17.6 8.84 20 5.20 0.080 1.64 0.196 20.3 3.89 55.8 23.9 24.4 4.85 30 4.73 0.160 2.8 1.27 20.0 6.03 67.4 12.6 33.8 6.54 40 5.05 0.200 2.44 1.03 25.9 3.92 49.9 24.2 39.4 19.3 50 5.24 0.160 4.75 4.01 25.2 3.17 32.8 42.0 40.3 11.4 60 5.11 0.140 4.52 1.70 47.7 1.39 3.74 48.6 53.7 5.64 70 4.97 0.110 23.3 5.52 48.8 0.24 0.97 50.2 57.8 7.93 80 4.95 0.090 4.29 1.10 49.5 0.57 1.34 49.2 59.9 2.73 90 5.23 0.170 5.33 3.15 50.0 0.16 0.60 49.4 54.9 6.99 100 5.24 0.150 18.6 9.66 49.4 1.36 1.02 49.6 51.1 2.02 Lake Geneva 1 0 4.92 0.139 7.66 0.78 3.6 0.27 3.60 8.52 9.11 1.12 10 4.93 0.112 8.52 0.21 8.3 1.72 8.30 19.2 33.6 10.0 20 5.09 0.158 20.3 4.28 10.6 1.68 10.6 20.5 30.2 8.27 30 5.80 0.257 39.4 1.82 11.3 0.42 11.3 25.6 30.3 1.78 40 5.96 0.140 45.9 1.55 14.8 2.04 14.8 29.0 44.0 6.91 50 6.09 0.126 57.0 3.69 19.9 2.42 19.9 38.3 70.4 4.32

PAGE 207

207 Table B 1. Continued Distance from shore (m) pH Water weight (%) Organic matter (%) > 63 m < 63 m Pb SD SD SD SD (%) (%) Lake Geneva 2 0 5.25 0.055 16.0 4.89 10.0 2.62 10.0 11.3 30.5 13.6 10 5.25 0.274 16.5 2.55 19.2 5.61 19.2 22.4 56.1 1.15 20 5.44 0.061 22.9 2.01 14.4 0.60 14.4 29.4 41.8 17.5 30 5.55 0.090 28.3 4.57 12.7 1.55 12.7 26.2 50.2 30.7 40 6.01 0.080 42.2 1.56 15.9 1.27 15.9 31.6 40.9 6.85 50 6.04 0.116 49.4 0.63 16.8 1.27 16.8 30.1 46.9 2.38

PAGE 208

208 Table B 2. Pb isotope data from exposed sediment collected along transects from Little Lake Johnson and Lake Geneva Distance from shore (m) Sediment Fraction* Pb Isotope Ratios 208/204 207/204 206/204 208/206 207/206 Little Lake Johnson 0 EXT 38.287 15.602 18.672 2.051 0.836 0 RED 38.386 15.640 18.705 2.052 0.836 0 OXI 38.419 15.638 18.718 2.053 0.835 0 RES 38.688 15.659 18.967 2.040 0.826 20 EXT 38.322 15.618 18.683 2.051 0.836 20 RED 38.397 15.644 18.714 2.052 0.836 20 OXI 38.397 15.646 18.696 2.054 0.837 20 RES 38.631 15.653 18.843 2.050 0.831 40 EXT 38.354 15.629 18.737 2.051 0.836 40 RED 38.403 15.647 18.718 2.052 0.836 40 OXI 38.406 15.646 18.712 2.052 0.836 40 RES 38.706 15.660 18.858 2.053 0.830 60 EXT 38.385 15.639 18.714 2.051 0.836 60 RED 38.395 15.644 18.717 2.051 0.836 60 OXI 38.404 15.645 18.714 2.052 0.836 60 RES 38.604 15.655 18.815 2.052 0.832 80 EXT 38.325 15.619 18.695 2.050 0.835 80 RED 38.391 15.641 18.713 2.052 0.836 80 OXI 38.404 15.645 18.710 2.052 0.836 80 RES 38.593 15.651 18.809 2.052 0.832 100 EXT 38.396 15.642 18.712 2.052 0.836 100 RED 38.390 15.641 18.712 2.052 0.836 100 OXI 38.393 15.640 18.707 2.052 0.836 100 RES 38.583 15.649 18.803 2.052 0.832 Lake Geneva 1 0 EXT 38.447 15.640 18.761 2.049 0.834 0 RED 38.453 15.644 18.742 2.052 0.835 0 OXI 38.468 15.641 18.746 2.052 0.834 0 RES 38.883 15.671 19.162 2.029 0.818 30 EXT 38.472 15.652 18.826 2.044 0.831 30 RED 38.469 15.651 18.825 2.044 0.831 30 OXI 38.500 15.660 18.840 2.043 0.831 30 RES 38.720 15.660 18.961 2.042 0.826 50 EXT 38.482 15.658 18.828 2.044 0.832 50 RED 38.465 15.651 18.821 2.044 0.832 50 OXI 38.479 15.655 18.826 2.044 0.832 50 RES 38.703 15.659 18.944 2.043 0.827

PAGE 209

209 Table B 2. Continued Distance from shore (m) Sediment Fraction* Pb Isotope Ratios 208/204 207/204 206/204 208/206 207/206 Lake Geneva 2 0 EXT 38.449 15.644 18.805 2.045 0.832 0 RED 38.468 15.652 18.814 2.045 0.832 0 OXI 38.475 15.650 18.814 2.045 0.832 0 RES 38.897 15.681 19.360 2.009 0.810 30 EXT 38.470 15.654 18.813 2.045 0.832 30 RED 38.464 15.651 18.815 2.044 0.832 30 OXI 38.484 15.656 18.821 2.045 0.832 30 RES 38.700 15.657 18.927 2.045 0.827 50 EXT 38.424 15.653 18.746 2.050 0.835 50 RED 38.463 15.651 18.810 2.045 0.832 50 OXI 38.475 15.654 18.812 2.045 0.832 50 RES 38.836 15.661 19.013 2.043 0.824 *Sediment Fraction Explanation EXT Exchangeable, RED Reducilbe Fe Mn oxides, Oxidizable organic matter/sulfides, RES Residual mineral

PAGE 210

210 LIST OF REFERENCES Abi Ghanem C, Chiffoleau JF, Bermond A, Nakhl K, Khalaf G, Borschneck D, Cossa D (2009) Lead and its isotopes in the sediment of three sites on the Lebanese coast: Identification of contamination sources and mobility. Applied Geochem 24:1990 1999 Adkinson A, Watmough SA Dillon PJ (2008) Drought induced metal release from a we tland at Plastic Lake, central Ontario. Can J Fish Aquat Sci 65:834 845 Alfaro de la Torre MC and Tessier A (2002) Cadmium deposition and mobility in the sediments of acidic oligotrophic lake. Geochm Cosmochim Acta 66:3549 3562 Alloway BJ (1997) The mobi lisation of trace elements in soils. p. 133 145. In R. Prost (ed.) Contaminated Soils, Proc. of the 3rd Int. Conf. on the Biogeochemistry of Trace Elements, Paris. 15 19 May 1995. INRA, Paris accumulation and temporal trends in leaves of urban deciduous trees ( Aesculus hippocastanum and Tilia spp.). Ecol Indic 11: 824 830 Annable MD, Motz LH, Knapp DS, Sousa GD Beddow WD (1996) Investigation of Lake and Surficial Aquifer Interaction in Theupper Etonia Creek Basin.1 240 Anshumali, Ramanathan AL, Singh G, Singh G, Ranjan R Tripathi P (2009) Chemodynamics of trace metal fractions in surface sediments of the Pandoh Lake, Lesser Him alaya, India. Environ mental Geology 57:1865 1879 Appleby PG, Oldfield F (1978) The calculation of Lead 210 dates assuming a constant rate of supply of unsupported 210 Pb to the sediment. Catena 5: 1 8 Appleby PG, Oldfield F (1983) The assessment of 210 Pb data from sites with varying sediment accumulation rates. Hydrobiologia 103: 29 35 Appleby PG, Nolan PJ, Gifford DW, Godfrey MJ, Oldfield F, Anderson NJ, Battarbee RW (1986) 210 Pb dating by low background gamma counting. Hydrobiologia 143: 21 27 Augus tsson A, Peltola P, Bergbck B, Saarinen T, Haltia Hovi E (2010) Trace metal and geochemical variability during 5,500 years in the sediment of Lake Lehmilampi, Finland. J Paleolimnol 44: 1025 1038 Ault WA, Senechal RG, Erlebach WE (1970) Isotopic composit ion as a natural tracer of lead in the environment. Environ Sci Technol 4: 305 313

PAGE 211

211 Bachman RW, Hoyer MV, Fernandez C, Canfield Jr DE (2003) An alternative to proposed phosphorus TMDLs of the management of Lake Okeechobee. Lake Reservoir Manag 19:251 Bac on JR, Farmer JG, Dunn SM and Graham MC (2006) Sequential extraction combined with isotope analysis as a tool for the investigation of lead mobilisation in soils: application to organic rich soils in an upland catchment in Scotland. Environ Poll 141:469 48 1 Battarbee RW (1978) Biostratigraphical evidence for variations in the recent pattern of sediment accumulation in Lough Neagh, N Ireland. Verh Internat Verein Limnol 20:624 629 Batterman S, Su F C, Jia C, Naidoo RN, Robins T Naik I (2011) Manganese and lead in children's blood and airborne particulate matter in Durban, South Africa. Sc i Total Environ 409:1058 1068 Bennion H, Battarbee R (2007) The European union water framework directive: op portunities for palaeolimnology. J Paleolimnol 38:285 295 Bilali El L, Rasmussen PE, Hall GEM, Fortin D (2002) Role of sediment composition in trace metal distribution in lake sediments. Appl Geol 17: 1171 1181 Bindler R, Renberg I, Klaminder J (2008) Bridging the gap between ancient met al pollution and contemporary biogeochemistry. J Paleolimnol 40: 755 770 Bindler R, Rydberg J, Renberg I (2011) Establishing natural sediment reference conditions for metals and the legacy of long range and local pollution on lakes in Europe. J Paleolimno l 45: 519 531 Binford M (1990) Calculation and uncertainty analysis of 210 Pb dates for PIRLA project lake sediment cores. J Paleolimnol 3: 253 267 Blake WH, Walsh RPD, Barnsley MJ, Palmer G, Dyrynda P, James JG (2003) Heavy metal concentrations during st orm events in a rehabilitated industrialized catchment. Hydrol Process 17: 1923 1939 Blumentritt DJ, Engstrom DR, Balogh SJ (2013) A novel repeat coring approach to reconstruct recent sediment, phosphorus, and mercury loading from the upper Mississippi Ri ver to Lake Pepin, USA. J Paleolimnol 50:293 304 Bos X, Rydberg J, Martinez Cortizas A, Bindler R, Renberg I (2011) Evaluation of conservative lithogenic elements (Ti, Zr, Al, and Rb) to study anthropogenic element enrichments in lake sediments. J Paleolimnol 46: 75 87

PAGE 212

212 Boudreau BP (1999) Metals and models: diagenetic modeling in freshwater lacustrine sediments. J Paleolimnol 22:227 251 Boyle JF and Birks HJB (1999) Predicting Heavy Metal Concentrations in the Surface Sediments of Norwegian Headwa ter Lakes From Atmospheric Deposition: an Application of a Simple Sediment Water Partitioning Model. Water Air Soil Pollut 114: 27 51 Boyle JF (2001a) Redox remobilization and the heavy metal record in lake sediments: a modelling approach. J Paleolimnol 2 6: 423 431 Boyle JF (2001b) Tracking Environmental Change Using Lake Sediments. Volume 2: Physical and Geochemical Methods Last WM and Smol JP (eds) Dordrecht, The Netherlands: Kluwer Academic Publishers Brallier S, Harrison RB, Henry CL, Dongsen X (199 6) Liming effects on availability of Cd, Cu, Ni and Zn in a soil amended with sewage sludge 16 years previously. Water Air Soil Pollut. 86:195 206 Brenner M, Binford MW, Deevey ES (1990) Lakes. Pp. 364 391, In R.L. Myers and J.J. Ewel (eds.), Ecosystems of Florida. University of Central Florida Press, Orlando. Brenner M and Whitmore TJ (1998) Historical sediment and nutrient accumulation rates and past water quality in Newnans Lake. Fina l Report for St. Johns River Water Management District Brezonik PL, Engstrom DR (1998) Modern and historic accumulation rtes of phosphorus in Lake Okeechobee, Florida. J Paleolimnol 20:31 46 Burger J, Gochfeld M, Rooney AA, Orlando EF, Woodward AR, Guillette LJ Jr (2000) Metals and Metalloids in Tissues of American Alligators in Three Florida Lakes. Arch Environ Con Tox 38: 501 508 Caccia VG, Millero FJ, Palanques A (2003) The distribution of trace metals in Florida Bay sediments. Mar Pollut Bull 46: 1420 1433 Callender E. and P.C. Van Metre. 1997. Reservoir sediment cores show U.S. lead declines. Envi ron. Sci. Technol. 31:424A 428A Calmano W, Hong J, Forstner U (1993) Binding and mobilizati on of heavy metals in contaminated sediments affected by pH and redox potential. Water Sci Technol 28: 223 235 Calmano W, von der Kammer F, Schwartz R (2005) Characterization of redox conditions in soils and sediments: Heavy metals. IWA Publ., London, UK

PAGE 213

213 Camarero L, Botev I, Muri G, Psenner R, Rose N, Stuchlik E (2009) Trace elements in alpine and arctic lake sediments as a record of diffuse atmospheric contamination across Europe. Freshwater Biol 54: 2518 2532 Carignan R, Tessier A (1985) Zinc depositio n in Acid Lakes: The role of diffusion. Science 228: 1524 1526 Chadwick OA, Derry LA, Vitousek PM, Huebert BJ, Hedin LO (1996) Changing sources of nutrients during four million years of ecosystem development. Nature 397 : 491 497 Cheng S (2003) Heavy metal pollution in China: Origin, pattern and control. Environ Sci Poll Res 10:192 198 Chesworth W, Martnez Cortizas A, Garca Rodeja E (2006) The redox pH approach to Develo pments in Earth Surface Processes 9: Peatlands Evolution and Records of Evironmental and Climate Changes (eds. P. Martini, A.Martnez Cortizas and W. Chesworth). Elsevier, 175 195 Cooke CA, Abbott MB, Wolfe AP, Kittleson LJ (2007) A millennium of metallur gy recorded by lake sediments in the Peruvian Andes. Envrion Sci Tech 41:3469 3474 Couillard Y, Cattaneo A, Gallon C, Courcelles M (2008) Sources and chronology of fifteen elements in the sediments of lakes affected by metal deposition in a mining area. J Paleolim 40:97 114 Davis RB, Anderson DS, Dixit SS, Appleby PG, Schauffler M (2006) Responses of Two New Hampshire (USA) Lakes to Human Impacts in Recent Centuries. J Paleolimnol 35: 669 697 Deevey EJ (1988) Estimation of downward leakage from Florida l akes. Limnol Oceanogr 33: 1308 1320 Demographia (2013) U S Population by State. (Consulted October 2013 http://www.demographia.com/db state1900.htm) Desenfant F, Veron AJ, Camoin GF, Nyberg J (2006) Reconstruction of pollutant lead invasion into the tropical North Atlantic during the twentieth century. Coral Reefs 25: 473 484 Dexter DT, Carayon A, Javoy Agid F, Agid Y, Wells FR, Daniel SE, Lees AJ, Jenner P, Marsden CD (1991) Alteration in the levels of iron, ferritin and other trace metals in ganglia. Brain 114:1953 1975

PAGE 214

214 Dillon PJ and Evans RD (1982) Whole lake lead burdens in sediments of l akes in southern Ontario, Canada. Hydrobiologia 91:121 130 Drever JI (1997) The Geochemistry of Natural Waters: Surface and Groundwater Environments Prentice Hall Duffus JH (2002) "Heavy metals" A meaningless term (IUPAC Technical Report) Pure Appl Ch em 74:793 807 Engstrom DR and Wright HE (1984) Chemical stratigraphy of lake sediments as a record of environmental change. In Lake Sediments and Environmental History: Studies in Palaelimnology and Palaeoecology in Honour of Winifred Tutin. Haworth EY an d Lund JW (Eds) Leicester University Press Engstrom DR, Schottler SP, Leavitt PR, Havens KE (2006) A reevaluation of the cultural eutrophication of Lake Okeechobee using multiproxy sediment records. Ecological Applica 16:1194 1206 Escobar J, Whitmore TJ Kamenov GD, Riedinger Whitmore MA (2013) Isotope record of anthropogenic lead pollution in lake sediments of Florida, USA. J Paleolimnol 49: 237 252 FLDEP (2010) Mike Roess Gold Head Branch State Park Unit Management Plan Division of Recreation and Par ks. Ferguson G and Gleeson T (2012) Vulnerability of coastal aquifers to groundwater use and climate change. Nature Climate C hange 2 : 342 345 Fisher MM, Brenner M, Reddy RK (1992) A simple, inexpensive piston corer for collection undisturbed sediment/ wat er interface profiles. J Paleolimnol 7: 157 161 Fleishman E, Blockstein DE, Hall JA, Mascia MB, Rudd MA, Scott M, Sutherland WF, Bartusk AM, Brown AG, Christen CA, Clement JP, Dellasala D, Duke CS, Eaton M, Fiske SJ, Gosnell H, Haney JC, Hutchins M, Klein ML, Marqusee, Noon BR, Nordgren JR, Orbuch PM, Powell J, Quarles SP, Saterson KA, Savitt CC, Stein BA, Webster MS, Vebber A (2011) Top 40 Priorities for science to inform U S conservation and management policy. BioScience 61:290 300 Florida Industrial and Phosphate Research Institute (2013) Phosphate Primer. (Consulted October 2013 http://www1.fipr.state.fl.us/PhosphatePrimer/ Florida Lakewatch (2005) Annual Data Summaries 2005. UF/IFAS School of Forest Resources and Conservation

PAGE 215

215 Florida Memory (201 3) Tracks Toward the Future, Scenes from Florida's Railroad History. Library Services and Technology Act, Institute of Museum and Library Services, Florida Department of State. (Consulted October 2013 http://www.floridamemory.com/photographiccollection/ph oto_exhibits/trains/trains4. php) Florida Oceans and Coastal Council (2010) Climate change and sea level rise in Florida; An update of the effects of climate change on Florida's Ocean and Coastal Resources. Tallahassee, Florida Frstner U and Patchineelam SR (1980) Chemical associations of heavy metals in polluted sediments from the Lower Rhine River in Kavanaugh, MC and Lecke JO (Eds) Particulates in Water; Characterization, Fate, Effects, and Removal, Avd. Chem Ser., 189, American Chemical Society, Washi ngton. 177 193 F rstner U., and G.T.W. Wittmann. 1981. Metal pollution in the aquatic environment. Springer Verlag, New York. F rstner U Calmano W, Hong J, Kersten M (1994) Effects of redox variations on metal speciation: Implications on sediment qualit y criteria assessment. In: Littlejohn D, Burns DT, editors. Reviews on analytical chemistry. Proceedings of Euroanalysis VIII; 1 993 September Edi nburgh, UK: Cambridge 83 102. Foster IDL and Charlesworth SM (1996) Heavy metals in the Hydrological cycle: Tr ends and explainations. Hydrol Process 10:227 261 Foulds SA, Brewer PA, Macklin MG, Haresign W, Betson RE, Rassner SME (2014) flood related contamination in catchmetns affected by historical metal mining: An unexpected and emerging hazard of climate change. Science of the Total Environment 476 477:165 180 Fujiwara FNG, Gmez DR, Dawidowski L, Perelman P, Faggi A (2011) Metals associated with airborne particulate matter in road dust and tree bark collected in a megacity (Buenos Aires, Argentina). Ecol Indic 11: 240 247 Furey PC, Nordin RN Mazumder A (2004) Water Level Drawdown Affects Physical and Biogeochemical Properties of Littoral Sediments of a Reservoir and a Natural Lake. Lake and Reservoi r Management 20 : 280 295 Gallon C, Tessier A, Gobeil C, Beaudin L (2005) Sources and chronology of atmospheric lead deposition to a Canadian Shield lake: Inferences from Pb isotopes and PAH profiles. Geochim Cosmochim Ac 69: 3199 3210 Gallon C, Tessier A, Gobeil C, Alfaro de la Torre M (2004) Modeling diagen esis of lead in sediments of a Canadian Shield lake. Geochim Cosmochim Acta 68:3531 3545

PAGE 216

216 Galloway JN, Thornton JD, Norton SA, Volchok H L, McLean RAN (1982) Trace metals in atmospheric deposition: A review and assessment. Atmospheric Environment (1967) 16 : 1677 1700 G lman V, Rydberg J, de Luna SS, Bindler R, Renberg I (2008) Carbon and nitrogen loss rates during aging of Lake Sediment: changes over 27 years studied in varved Lake Sediment. Limnol Oceanogr 53:1076 1082 Gelinas Y, Schmit J P (1998) Estimati on of the bulk atmospheric deposition of major and trace elements to a rural watershed. Atmos Environ 32: 1473 1483 Gelinas T, Lucotte M, Schmit JP (2000) History of the atmospheric deposition of major and trace elements in the industrialized St. Lawrence Valley, Quebec, Canada. Atmosph Environ 34:1797 1810 Gott gens JF and Crisman TL (1993) Quantitative impacts of lake level stabilization on material transfer between water and sediment in Newnans Lake, Florida. Can J Fish Aquatic Sc 50:1610 1616 Gottgens JF, Rood BE, Delfino JJ, Simmers BS (1999) Uncertainty in paleoecological studies of mercury in sediment cores. Water Air Soil Pollut 110:313 333 Graney JR, Halliday AN, Keeler GJ, Nriagu JO, Robbins JA, Norton SA (1995) Isotopic record of lead pollution in the lake sediments from the northeastern United States. Geochem Cosmochem Acta 59: 1715 1728 Grini A and Zender CS (2004) Roles of saltation, sandblasting, and wind speed variability on mineral dust aerosol size distribution during the Puerto Rican dust experiment (PRIDE). Journal of Geophysica l Research Atmospheres 109 (D7) Grybos M, Davranche M, Gruau G Petitjean P (2007) Is trace metal release in wetland soils controlled by organic matter mobility or Fe oxyhydroxides reduction? Journal of Colloi d and Interface Science 314 : 490 501 Gulson BL (2008) Stable lead isotopes in environmental health with emphasis on human investigations. Sci. Total Env 400:75 92 Gunawardana C, Egodawatta P, Goonetilleke A (2014) Role of particle size and composition in metal adsorption by solids deposited on urban road surfaces. Environ Poll 184 : 44 53 Havens KE, Jin KR, Rodusky AJ, Sharfstein B, Brady MA, East TL, Iricanin N, James RT, Harwell MC, Steinman AD (2001) Hurricane effects on a shallow lake ecosystem and its response to a controlled manipulation of water level. The Scientific World J 1:44 70

PAGE 217

217 H kanson L and Jansson M (1983 ) Principles of Lake Sedimentology Springer Hamilton Taylor J, Davison W (1995) Redox driven cycling of trace elements in lakes Lerman A Imboden D and Gat J (eds) Springer Verlag Hamilton Taylor J, Davison W, Morfett K (1996) A laboratory study of the biogeochemical cycling of Fe, Mn, Zn and Cu across the sediment water interface of a productive lake. Aquat Sci 58: 191 209 Hamilton Taylo r J, Giusti L, Tych DW, Hewitt CN (1997) Sorption of trace metals (Cu, Pb, Zn) by suspended lake particles in artifical (0.005 M NaNO 3 ) and natural (Esthwaite Water) freshwaters. Colloids and Surfaces A: Physicochem Engineer Aspects 120: 205 219 Hamilton Taylor J, Smith E, Davison W Sugiyama M (2005) Resolving and modeling the effects of Fe and Mn redox cycling on trace metal behavior in a seasonally anoxic lake. Geochim Cosmochim Acta 69 : 1947 1960 Hargreaves AL, Whiteside DP, Gilchrist G (2011) Concentr ations of 17 elements, including mercury, in the tissues, food and abiotic environment of Arctic shorebirds. Sci Total Environ 409: 3757 3770 Hart BT (1982) Uptake of trace metals by sediments and suspended particulates: a review. Hydrobiologia 91:1 15 H einz GH, Percival HF, Jennings ML (2012) Contaminants in American alligator eggs from Lake Apopka, Lake Griffin, and Lake Okeechobee, Florida. Environ Monit Assess 16: 277 285 Hilscherova K, Dusek L, Kubik V, Cupr P, Hofman J, Klanova J, Holoubek I (2007) Redistribution of organic pollutants in river sediments and alluvial soils related to major floods. J Soil Sed 7 : 167 177 Historic Florida Mainlines (2013) The Interstate System: 1959 to the Present. (Consulted October 2013 http://www.us highways.com/fli .htm) Ho HH, Swennen R, Cappuyns V, Vassilieva E, Van Gerven T, Van Tran T (2012) Potential release of selected trace elements (As, Cd, Cu, Mn, Pb and Zn) from sediments in Cam River mouth (Vietnam) under influence of pH and oxidation. Sci Total Environ 435 436:487 498 Hofmann P, Duckensell M, Chpitsglous A, Schwark L (2005) Geochemical and organic petrological characterization of the organic matter of lacustrine Eocene oil shales (Prinz von Hessen, Germany): reconstruction of the depositional environmen t. J Paleolimnol 33:155 168

PAGE 218

218 Horowitz AJ (1985) A primer on trace metal sediment chemistry Alexandria, VA: United States Government Printing Office Hu X, Sun Y, Ding Z, Zhang Y, Wu J, Lian H, Wang T (2014) Lead contamination and transfer in urban enviro nmental compartments analyzed by lead levels and isotopic composition. Enviro Poll 187 : 42 48 Huang X, Moir RD, Tanzi RE, Bush AI, Rogers JT (2006) Redox active metals, Academy of Sciences 1012:53 163 Hunter KA (1980) Microelectric properites of natural surface active organic matter in coastal seawater. Limnol Oceanogr 25:807 822 IPCC (2007) Climate Change 2007: The Physical Basis. In: Solomon S, Qin D, Manning M, Chen Z, Marquis M, Averyt KB, Tignor M, Miller HL (Eds.), Contribution of the Working Group I to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change. Cambridge University Press, Cambridge, United Kingdom Ikem A, Egiebor NO, Nyavor K (2003) Trace Elements In Water, Fish and Sediment from Tuskegee Lake, Southeastern U SA Juracek KE and Ziegler AC (2006) The Legacy of Leaded Gasoline in Bottom Sediment of Small Rural Reservoirs. J Environ Quality 35:2092 Kabata Pendias, A (2001) Trace elements in soils and plants. CRC Press, Boca Raton, FL Kamenov GD, Perfit MR, Mueller PA, Jonasson IR (2008) Controls on magmatism in an island arc environment: study of lavas and sub arc xenoliths from the Tabar Lihir Tanga Feni island chain, Papua New Guinea. Contr Mineral Petrol 155: 635 656 Kamenov GD (2008) High precision Pb isotopic measurements of teeth and environmental samples from Sofia (Bulgaria): insights for regional lead sources and possible pathways to the human body. Environmental Geology 55:669 680 Kamenov GD, Brenner M, Tucker JL (2009) Anthropogenic versus natural control on trace element and Sr Nd Pb isotope stratigraphy in peat sediments of southeast Florida (USA), 1500 AD to present. Geochim Cosmochim Ac 73: 3549 3567 Kamenov GD, M ueller PA, Perfit MR (2004) Optimization of mixed Pb Tl solutions for high precision isotopic analysis by MC ICP MS. J Anal Atom Spectrom 19: 1262 1267

PAGE 219

219 Karlen DJ, Ash T, Markham S, Goetting B, Barron J, Perez S, Thorne L (2009) Sediment metals status in Lake Thonotosassa Hillsborough County, Florida. Technical Report; Environmental Protection Commision of Hillsborough County : 1 61 Kawahata H, Okamoto T, Matsumoto E, Ujiie H (2000) Fluctuations of eolian flux and ocean productivity in the mid latitude North Pacific during the last 200 kyr. Quat Sci Rev 19 : 1279 1291 Kibar M (200 9) Wind control on the accumulation of heavy metals in sediment of Lake Ulubat, Anatolia, Turkey. J of Paleolimn 43 : 89 110 Kenney WF, Brenner M, Curtis JH, Schelske CL (2010) Identifying sources of organic matter in sediments of shallow lakes using multi ple geochemical varaibles. J Paleolimnol 44:1039 1052 Kenney WF, Whimore TJ, Buck DG, Brenner M, Curtis JH, Di JJ, Kenney PL, Schelske CL (2014) Whole basin, mass balance approach for identifying critical phosphorus loading thresholds in shallow lakes. J Paleolimnol. Published online DOI 10.1007/s10933 014 9771 9 Kistemann T, Classen T, Koch C, Dagendorf F, Fischeder R, Gebel J, Vacata V, Exner M (2002) Microbial load of drinking water res ervoir tributaries during extreme rainfall and runoff. Appl Envir on Microbiol 68:2188 2197 Kidd PM (2000) Attention Deficit/Hyperactivity Disorder (ADHD) in children: rationale for its integrative management. Alternative Medicine Review 5:402 425 Koinig KA, Shotyk W, Lotter AF, Ohlendorf C, Sturm M (2003) 9000 years of geochemical evolution of lithogenic major and trace elements in the sediment of an alpine lake the role of climate, vegetation, and land use history. J Paleolimnol 30: 307 320 Kol ak JJ, Long DT, Beals TM, Eisenreich SJ, Swackhamer DL (1998) Anthropogenic inventories and historical and present accumulation rates of copper in Great Lakes sediments. App Geochem 13:59 75 Kratzer CR, Brezonik PL (1981) A Carlson type trophic state index for Nitrogen in Florida lakes. J Am Water Resour Assoc 17: 713 715 Kreider ML, Panko JM, McAtee BL, Sweet LI, Finley BL (2010) Physical and chemical characterization of tire relatated parti cles generated using different methodologies. Sci Total Environ 408: 652 659

PAGE 220

220 Krueger F, Meissner R, Groengroeft A, Grunewald K (2005) Flood induced heavy metal and arsenic contamination of Elbe river floodplain soils. Acta Hydrochim. Hydrobiol. 33:455 46 5 Landing WM, Caffrey JM, Nolek SD, Gosnell KJ, Parker WC (2010) Atmospheric wet deposition of mercury and other trace elements in Pensacola, Florida. Atmos Chem Phys 10: 4867 4877 Landre AL, Watmough SA, Dillon PJ (2010) Metal Pools, Fluxes, and Budge ts in an Acidified Forested Catchment on the Precambrian Shield, Central Ontario, Canada. Water Air Soil Poll 209: 209 228 Lehman JT (1975) Reconstructing the rate of accumulation of lake sediments: the effect of sediment focusing. Quat Res 5:541 550 Li kens GE and Davis MB (1975) Post glacial history of Mirror Lake and its watershed in New Hampshire, USA: an initial report. Int Ver Theor Angew Limnil Verh 19: 982 993 Liu E, Shen J, Birch GF, Yang X, Wu Y, Xue B (2012) Human induced change in sedimentary trace metals and phosphorus in Chaohu Lake, China, over the past half millennium. J Paleolimnol 47: 677 691 Livingston I and Warren A (1996) Aeolian Geomorphology, Addison Wesley Longman, Reading, MA Logan EM, Pulford ID, Cook GT, MacKanzie AB (1997) Complexation of Cu 2+ and Pb 2+ by peat and humic acid. Eur. J. Soil Sci. 48:685 696 Luettich RA Jr, Harleman DRF, Somyody L (1990) Dynamic behavior of suspended sediment concentrations in a shallow lake perturbed by episodic wind events. Limnnol Oceanogr 3 5:1050 1067 Luo X S, Yu S and Li X D (2011) Distribution, availability, and sources of trace metals in different particle size fractions of urban soils in Hong Kong: Implications for assessing the risk to human health. Environmental Pollution 159 : 1317 1326 MacDonald DD, Scott CR, Calder FD, Long ER (1996) Development and evaluation of sediment quality guidelines for Florida coastal waters. Ecotoxicol 5: 253 278 Mahler BJ, VanMetre PC, Callendar E (2006) Trends in metals in urban and reference lak e sediments across the United States, 1970 to 2001. Environ Toxicol Chem 25:1698 1709

PAGE 221

221 Maher BA, Prospero JM, Mackie D, Gaiero D, Hesse P, Balkanski Y (2010) Global connections between aeolian dust, climate and ocean biogeochemistry at the present day an d at the last glacial maximum. Earth Science Rev 99 : 61 97 Martinez CJ, Maleski JJ Miller MF (2012) Trends in precipitation and temperature in Florida, USA. Journal of Hydrology Massey DD, Kulshrestha A, Taneja A (2013) Particulate matter concentrations and their related metal toxicity in rural residential environment of semi arid region of India. Atmosph Environ 67:278 286 Matagi S, Swai D, Mugabe R (1998) A review of heavy metal removal mechanisms in wetlands. African J Trop Hydrobiol Fish 8:23 35 McK enzie ER, Money JE, Green PG, Young TM (2009) Metals associated with stormwater relevant brake and tire samples. Sci Total Environ 407: 5955 5860 Middleton NJ (2984) Dust storms in Australia: Frequency, distribution and seasonality. Search 15:46 47 Mielke HW and Reagan PL (1998) Soil is an important pathway of human lead exposure. Envir Health Perspectives 106: 217 229 Mimikou MA, Baltas E, Varanou E Pantazis K (2000) Regional impacts of climate change on water resources quantity and quality indica tors Monna F, Clauer N, Toulkeridis T, Lancelot JR (2000) Influence of anthropogenic activity on the lead isotope signature of Thau Lake sediments (southern France): origin and temporal evolution. Appl Geochem 15: 1291 1305 Moses CS, Anderson WT, Saunders C Sklar F (2012) Regional climate gradients in precipitation and temperature in response to climate teleconnections in the Greater Everglades ecosystem of South Florida. J of Paleol im 49 : 5 14 Motz LH and Heaney JP (1991) Upper Etonia Creek Hyrdo logic Study Palatka: St. Johns River Water Management District Mulholland PJ, Best GR, Coutant CC, Hornberger GM, Meyer JL, Robinson PJ, Stenberg JR, Turner RE, Vera Herrera F Wetzel RG (1997) Effects of climate change on freshwater ecosystems of the so uth eastern United States and the gulf coast of Mexico. Hydrol. Process. 1 1: 949 970 N'guessan YM, Probst JL, Bur T, Probst A (2009) Trace elements in stream bed sediments from agricultural catchments (Gascogne region, S W France): Where do they come from? Sci Total Environ 407: 2939 2952

PAGE 222

222 NOAA (2014) National Weather Service National Hurricane Center 2004 Atlantic Hurricane Season. Accessed 2/2014 http://www.nhc.noaa.gov/2004atlan.shtml Nagel R and Loskill R (Eds) (1991) Bioaccumulation in Aquatic Systems VCH Verlagsgesellschaft, Weinheim Needleman HL, Bellinger D (1991) The health effects of low level exposure to lead. Annu Rev Public Health 12: 111 140 Nemati K, Bakar NKA, Abas MR, Sobhanzadeh E (2011) Speciation of heavy metals by modified BCR sequential extraction procedure in different depths of sediment from Sungai Bulah, Selangor, Malaysia. J Haz Materials 192:402 410 Nguyen HL, Leermakers M, Osn J and Trk S (2005) Heavy metals in Lake Balaton: water column, suspended matter, sediment an d biota. Sci Total Environ 340:213:230 Norton SA, Kahl JS (1987) New Approaches to Monitoring Aquatic Ecosystems Boyle TP (ed.) Philadelphia: American Society for Testing and Materials Norton SA, Bienert RW Jr, Binford MW, Kahl JS (1992) Stratigraphy of total metals in PIRLA sediment cores. J Paleolimnol 7: 191 214 Nriagu JO (1996) A history of global metal pollution. Science 272: 223 224. Nriagu JO, Pacyna JM (1988) Quantitative asses sment of worldwide contamination of air, water and soils by trace elements. Nature 333: 134 139 Obeysekera J, Irizarry M, Park J, Barnes J Dessalegne T (2010) Climate change and its implications for water resources management in south Florida. Stoch Envi ron Res Risk Assess 25 : 495 516 Oldfield F, Appleby PG (1984) Empirical testing of 210 Pb dating models for lake sediments Haworth ET and Lund J (eds) University of Minnesota Press Olivie Lauquet G (2001) Release of Trace Elements in Wetlands: Role of Se asonal Variability. Water Res 35 : 943 952 Outridge PM, Stern GA, Hamilton PB, Percival JB, McNeely R, Lockhart WL (2005) Trace metal profiles in the varved sediment of an Arctic lake. Geochim Cosmochim Ac 69: 4881 4894 Patterson CC and Settle DM (1987) Review of data on eolian fluxes of industrial and natural lead to the lands and seas in remote regions on a global scale. Marine Chem 22:137 162

PAGE 223

223 Pearson LK, Hendy CH, Hamilton DP,Pickett RC (2010) Natural and anthropogenic lead in sediments of the Rotorua lakes, New Zealand. Earth Planet Sc Lett 297: 536 544 Pelletier JD (2006) Sensitivity of playa windblown dust emissions to climatic and anthropogenic change. J Arid Environ 66 : 62 75 Percival JB and Outridge PM (2013) A test of the stability of Cd, Cu, Hg, Pb and Zn profiles over two decades in lake sediments near the Flin Flon Smelter, Manitoba, Canada. Sci Total Environ 454 455:307 318 Phips DA (1981) Chemistry and biochemistry of trace metals in biolgoical systems in Effect of heavy metal pollution o n plants. NW Lepp (Ed) Applied Science Publishers, Barking Pirkle EC, Pirkle FL, Pirkle WA, Stayert PR (1984) The Yulle heavy mineral sand deposits of northeaster n Florida. Eco Geol 79: 725 737 Pope III CA and Dockery DW (2006) Critical Review: Health effects of fine particulate air pollution: Lines that connect. J Air Waste Manag Assoc 56 : 709 742 Prica M, Dalmacija B, Dalmacija M, Agbaba J, Krcmar D, Trickovic J, Karlovic E (2010) Changes in me tal availability during sediment oxidation and the correlation with the immobilization potential 73:1370 1377 Prospero JM, Ginoux P, Torres O, Nicholson SE (2002) Environmental characteristics of global sources of atmospheric soil dust derived from the NI MBUS 7 TOMS absorbing aerosol product. Rev Geophys 40 : 2 1 2 31 Provencher JF, Braune BM, Gilchrist HG, Forbes MR, Mallory ML (2014) Trace element concentrations and gastrointestinal parasites of Arctic terns breeding in the Canadian High Arctic. Sci Total Environ 476 477:308 316 Pye K (1987) Aeolian Dust and Dust Deposits, Academic, San Diego, C A Rauret G, Lpez Snchez JF, Sahuquillo A, Rubio R, Davidson C, Ure A, Quevauviller P (1999) Improvement of the BCR three step sequential extraction procedu re prior to the certification of new sediment and soil reference materials. J. Environ. Monitor 1: 57 61 Reddy KR and DeLaune RD (2008) Biogeochemistry of Wetlands, Science and Applications. CRC Press, Boca Raton, FL Regnier P and Wollast R (1993) Distri bution of trace metals in suspended matter of the Scheldt estuary. Mar Chem 43 : 3 19

PAGE 224

224 Reheis MC, Budahn RJ, Lamothe P (2002) Geochemical evidence for diversity of dust sources in the southwestern United States. Geochim Cosmochim Acta 66 : 1569 1587 Reimann C, Caritat P de (2000) Intrinsic Flaws of Element Enrichment Factors (EFs) in Environmental Geochemistry. Environ Sci Technol 34: 5084 5091 Rhodes SL and Wiley KB (1993) Great Lakes toxic sediments and climate change. Global Environmental Change 292 305. Riedinger Whitmore MA, Whitmore TJ, Smoak JM, Brenner M, Moore A, Curtis JH, Schelske CL (2005) Cyanobacterial Proliferation is a Recent Response to Eutrophication in Many Florida Lakes: A Paleolimnological Assessment. Lake Reserv Manage 21: 423 435 Rockstrom J, Steffen W, Noone K, Persoon A, Chapin III FS, Lambin EF, Lenton TM, Scheffer M, Folke C, Schellnhuber HJ, Nykvist B, de Wit CA, Hughes T, van der Leeuw S, Rodhe H, Sorlin S, Snyder PK, Costanza R, Svedin U, Falkenmark M, Karlberg L, Corell RW, Fabry VJ, Hansen J, Walker B, Liverman D, Richardson K, Crutzen P, Foley JA (2009) A safe operating space for humanity. Nature 461: 472 475 Rose NL, Yang H (2007) Temporal and spatial patterns of spheroidal carbonaceous particles (SCPs) in sediments, soils and deposition at Lochnagar. In: Rose NL (ed) Lochnagar: the natural history of a Mountain Lake.Springer, Dordrecht, pp 403 423 Rydberg J, Galman V, Rengerg I, Bindler R, Lambertsson L, Martinez Cortizas A (2008) Assessing the stability of Mercury a nd Methylmercury in a varved lake sediment deposit. Environ Sci Technol 42:4391 4396 Ryberg J and Martinez Cortizas A (2014) Geochemical assessment of an annually laminated lake sediment record from northern Sweden: a multi core, multi element approach. J Paleolimnol Published online DOI 10.1007/s10933 014 9770 x SFDEP (State of Florida Department of Environmental Protection) (2010) Mike Roess Gold Head Branch State Park Unit Management Plan Division of Recreation and Parks. SJRWMD (St. Johns River Wate r Management District) (2013 a ) St. Johns River Water Management District Online Tools, GIS, Data for SHEEL (Lake Sheelar) Accessed February 2013: http://www.sjrwmd.com/wate rshedfacts/factPages/SHEEL.html SJRWMD (St. Johns River Water Management District) (2013 b ) Status of Keystone Heights lakes and aquifer protection projects. Accessed November 2013. http://floridaswater.com/keystoneheights/projectsstatus.html

PAGE 225

225 SJRWMD (2014) Lake Harris; St. Johns River Water Management District. Accessed Jan 2014 http://www.sjrwmd.com/watershedfacts/factPages/HAR.html Saeki K, Okazaki M, Matsumoto S (1992) The chemical phase changes in heavy metal with drying and oxidation of t he lake sediments. Water Research 27:1243 1251 Sakata M (1985) Diagenetic remobilization of Manganese, Iron, Copper, and Lead in anoxic sediment of a freshwater pond. Water Res 19: 1033 1038 Salomans W and Frstner U (1984) Metals in the Hdrocycle. Springer Verlag, Berlin Santos IR, Silva Filho EV, Schaefer CEGR, Albuquerque Filho MR, Campos LS (2005) Heavy metal contamination in coastal sediments and soils near the Brazilian Antarctic Station, King George Island. Mar Pollut Bull 50: 185 194 Santsc hi P, Hohener P, Benoit G, Brink MB (1990) Chemical processes at the sediment water interface. Marine Chem 30:26 315 Schelske CL, Peplow A, Brenner M, Spencer CN (1994) Low background gamma counting: applications for 210 Pb dating of sediments. J Paleolimn ol 10:115 128 Schelske CL, Kenney WF, Whitmore TJ (2001) Final Report, Sediment and Nutrient Deposition in Harris Chain of Lakes. St. Johns River Water Management District Contract No 99W208. Special Publication SJ2001 SP7 Schiffer DM (1999) Hydrology o f Central Florida Lakes: A Primer U.S. Geological Survey Circular 1137 Schulz Zunkel C and Krueger F (2009) Trace Metal Dynamics in Floodplain Soils of the River Elbe: A Review. Journal of Environment Quality 38 : 1349 Applied Geochem 22:2569 2594 Shaw TJ, Gieskes JM, Jahnke RA (1990) Early diagenesis in differing depositional en vironments The response of transition metals in pore water. Geochim Cosmochim Acta 54:1233 1246 Shaw DJ (1992) Colloid and Surface Chemistry, fourth ed. Butterworth Heinemann, Boston, MA Shen GT, Boyle EA (1987) Lead in corals: reconstruction of histo rical industrial fluxes to the surface ocean. Earth Planet Sc Lett 82: 289 304

PAGE 226

226 Shirahata H, Elias RW, Patterson CC (1980) Chronological variations in concentrations and isotopic compositions of anthropogenic atmospheric lead in sediments of a remote subalpine pond. Geochim Cosmochim Ac 44: 149 162 Shottler and Engstrom (2006) A chro nological assessment of Lake Okeechobee (Florida) sediments using multiple dating markers. J Paleolimnol 36:19 36 Shotyk W, WEISS D, Appleby PG, Cheburkin AK, Frei R, Gloor M, Kramers JD, Reese S Van Der Knaap WO (1998) History of Atmospheric Lead Deposition Since 12,370 14 C yr BP from a Peat Bog, Jura Mountains, Switzerland. Science 281 : 1635 1640 Shotyk W, Weiss D, Kramers JD, Frei R, Cheburkin AK, Gloor M, Reese S (2001) Geochemistry of the peat bog at Etang de la Grure, Jura Mountains, Switzerland, and its record of atmospheric Pb and lithogenic trace metals (Sc, Ti, Y, Zr, and REE) since 12,370 14C yr BP. Geochim Cosmochim Ac 65: 2337 2360 Shotyk W, Krachler M, Martnez Cortizas A, Cheburkin AK, Emons H (2002) A peat bog record of natural, pre anthropogenic enrichments of trace elements in 14 C yr BP, and their variation with Holocene climate change. Earth Planet Sc Lett 199: 21 37 Sigg L, Sturm M, Kistler D (1987) Vertical transport of heavy metals by settling particles in Lake Zurich. Limnol Oceanogr 32: 112 130 Siver PA,Wozniak JA (2001) Lead analysis of sediment cores from seven Connecticut lakes. J Paleolimnol 26: 1 10 Smith JL and Lee K (2003) Soil as a Source of Dust and Implications for Human Health. In Advances in Agronomy Advances in Agronomy. Elsevier, 1 32 Smith J O Smith P, Wattenbach M, Zaehle S, Hiederer R, Jones R JA Montanarella L, Rounsevell M DA Reginster I Ewert F (2005) Projected changes in mineral soil carbon of European croplands and grasslands, 1990 2080. Global Change Biol 11 : 2141 2152 Sparks DL (2003) Environmental Soil Chemistry, second ed. Academic Press, San Diego, CA Stecko JRP and Bendell Young LI (2000) Contrasting geochemistry of suspended particulate matter Sutherland RA (2010) BCR 701: A review of 10 years of sequential extraction analysis. Analytica Chimica Acta 680:10 20 Sweets PR, Bienert RW, Crisman TL, Binford MW (1990) Paleoecological investigations

PAGE 227

227 of recent lake acidification in Northern Florida. J Paleolimnol 4: 103 137 Syers JK and Goldfelds M (Eds) (2000) Environmental cadmium in the food chain: source, pathways and risks, Proceedings of the SCPOPE workshop. Scient ific committee on the problems of the environment/international council of scientific unions Tack TM, Callewaert OWJJ, Verloo MG (1996) metal solubility as a function of pH in contaminated, dredged sediment affected by oxidation. Evniron Pollut 91:199 208 Tanaka TY and Chiba MA (2006) A numerical study of the contributions of dust source regions to the global dust budget. Glob Planet Change 5 2: 88 104 Tao Y, Yuan Z, Wei M Xiaona H (2011) Characterization of heavy metals in water and sediments in Taihu L ake, China. Environmental Monitoring and Assessment 184 : 4367 4382 Tegen I, Lacis AA, Fung I (1996) The influence on climate forcing of mineral aerosols from disturbed soils. Nature 380 : 419 422 Tegen I, Harrison SP, Kohfeld K, Prentic IC, Coe M, Heimann M (2002) Impact of vegetation and preferential source areas on global dust aerosol: Results from a model study. J Geophysical Res107 : 14 27 Tessier A, Fortin D, Belzile N, DeVitre RR, Leppard GG (1996) Metal sorption to diagenetic iron and manganese oxyhydroxides and associated organic matter: Narrowing the gap between field and laboratory measurements. Geochim Cosmochim Ac 60: 387 404 Tessie r A, Campbell PGC, Bisson M (1979) Sequential extraction procedure for the speciation of particulate trace metals. Anal Chem 51:844 851 Thevenon F, Gudron S, Chiaradia M (2010) (Pre ) historic changes in natural and anthropogenic heavy metals deposition inferred from two contrasting Swiss Alpine lakes. Quaternary Sci Rev 30: 224 233 Tipping E (1981) The adsorption of aquatic humic substances by iron oxides. Geochim Cosmochim Acta 45:191 199 Tipping E, Sm ith EJ, Lawlor AJ, Hughes S, Stevens PA (2003) Pre dicting the release of metals from ombrotrophic peat due to drought induced acidification. Environ Poll 123:239 253 Toutain JP, Sortino F, Reynier B, Dupre B, Munoz M, Nonell A, Polve M, Chancha Do Vale S (2003) A new collector for sampling volcanic aeros ols. J Volcanol Geoth Res 123: 95 103

PAGE 228

228 Trivedi P, Axe L (2000) Modeling Cd and Zn sorption to hydrous metal oxides. Environ Sci Technol 34:2215 2223 Tylmann W (2005) Lithological and geochemical record of anthropogenic changes in recent sediments of a small and shallow lake (Lake Pusty Staw, northern Poland). J Paleolimnol 33: 313 325 USEPA. 2000. National air pollutant emission trends, 1900 1998. EPA /454/R 00/002. USEPA, Washington, DC VanMetre PC, Mahler BJ, Wilson JT, Callender E (2006) Major and Trace Elements in 35 Lake and Reservoir Sediment Cores From Across the United States, 1994 2001. United States Geolo gical Society. Data Series 166 Vorosmarty CJ (2000) Global Water Resources: Vulnerability from Climate Change and Population Growth. Science 289 : 284 288 Wang X, Yang H, Gong P, Zhao X, Wu G, Turner S, Yao T (2010) One century sedimentary records of polycyclic aromatic hydrocarbons, mer cury and trace elements in the Qinghai Lake, Tibetan Plateau. Environ Pollut 158: 3065 3070 Watts WA, Stuiver M (1980) Late Wisconsin climate of Northern Florida and the origin of species rich deciduous forest. Science 210: 325 327 Weatherspoon RL, Cumm ings E, Wittstruck WH (1989) Soil Survey of Clay County, Florida. United States Department of Agriculture Soil Conservation Service Weiss D, Shotyk W, Appleby PG, Kramers JD, Cheburkin AK (1999) Atmospheric Pb Deposition since the Industrial Revolution Recorded by Five Swiss Peat Profiles: Enrichment Factors, Fluxes, Isotopic Composition, and Sources. Environ Sci Technol 33: 1340 1352 White MS, Xenopoulos MA, Hogsden K, Metcalfe RA Dillon PJ (2008) Natural lake level fluctuation and associated concordance with water quality and aquatic communities within small lakes of the Laurentian Great Lakes region. Hydrobiologia 613 : 21 31 Whit more TJ, Brenner M, Schelske CL (1996) Highly variable sediment distribution in shallow, wind stressed lakes: a case for sediment mapping surveys in paleolimnological studies. J Paleolimnol 15:207 221 Winderlund A, Roos P, Gunneriusson L, Ingri J, Holmstr om H (2002) Early diagenesis and isotopic composition of lead in Lake Laisan, northern Sweden. Chem Geol 189:183 197

PAGE 229

229 Yakes D, Towler E (2011) Global climate change screening analysis for the St. Jhns River Water Supply Impact Study. Report delivered to t he St. Johns River Water Management District. National Center for Atmospheric Reserach Wu Y, Liu E, Yao S, Zhu Y, Xia W (2010) Recent heavy metal accumulation in Dongjiu and Xijiu lakes, East China. J Paleolimnol 43: 385 392 Zahran S, Laidlaw MA, McElmur ry SP, Filippelli GM, Taylor M (2013) Linking source Mic higan. Env. Sci. Technology 47: 2839 2845 Zhang H, Davison W, Miller S, Tych W (1995) In situ high resolution meas urements in fluxes of Ni, Cu, Fe, and Mn and concentrations of Zn and Cd in porewaters by DGT. Geochim Cosmochim Ac 59: 4181 4192 Zimmerman AJ and Weindorf DC (2010) Heavy Metal and Trace Metal Analysis in Soil by Sequential Extraction: A Review of Procedures. International J Analytical Chem 3 4:1 7 Zobeck TM and Van Pelt RS (2006) Wind induced dust generation and transport mechanics on a bare agricultural field. Journal of Hazardous Materials 132:26 38 Zohary T and Ostrovsky I (2011) Ecological impacts of excessive water level fluctuations in strati fied freshwater lakes. Plenary Lecture in Inland Waters 1 : 47 59

PAGE 230

230 BIOGRAPHICAL SKETCH Susanna Whitman Blair was born and raised in Baltimore, Maryland. She graduated from Baltimore City College High school i n 1999 and that fall began her bachelor degree at Colgate University in Hamilton, New York. In 2003 she graduated with a Bachelor of Arts in geology. Her love of science took her to the University of Florida w degree in the Department of Geological Sciences. Her thesis project focused on ancient ocean circulation as a proxy for climate change. In 2006 she graduated and began working for Golder Associates Inc, an environmental consulting firm in Jacksonville, Florida. This experience allowed Susanna to see geolo resource management implications of the physical sciences. Although, rewarding Susanna was anxious to return to school to pursue a PhD. She was awarded a National Science Foundation Integra tive Graduate Education and Research Traineeship (IGERT) Fellowship focused on the Adaptive Management of Water, Wetlands, and Watersheds and began her PhD in 2008 in the Department of Geological Sciences under the advisement of Dr. Mark Brenner. Her rese arch focus has been on the cycling of trace metals in Florida lake sediments and the compounding effects of climate change on metal chemistry. During her time as a PhD student she also worked extensively with Tom Ankersen and the Conservation Clinic at th e Levin College of Law and she was also awarded a National Science Foundation Science Partnerships in Collaborative Education (SPICE) Fellowship where for two years she taught in a local public middle school. She received her Doctor of Philosophy from the University of Florida in the spring of 2014. She is very interested and hopes to pursue a career at the intersection between science and policy and society.