|UFDC Home||myUFDC Home | Help|
This item has the following downloads:
1 ORIGIN AND STRATIGRAPHIC SIGNIFICANC E OF KAOLINITIC SEDIMENTS FROM THE CYPRESSHEAD FORMATION: A SEDI MENTOLOGICAL, MINERALOGICAL AND GEOCHEMICAL INVESTIGATION By KENDALL BRIAN FOUNTAIN A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLOR IDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2009
2 2009 Kendall Brian Fountain
3 "The Ridge is the Florida Divide, the peni nsular watershed, and, to hear Floridians describe it, the world's most stupendous mountai n range after the Himalayas and the Andes. Soaring two hundred and forty feet into the sub-tr opical sky, the Ridge is difficult to distinguish from the surrounding lowlands, but it differs more in soil conditions than in altitude, and citrus trees cover it like a long streamer, sometimes as litt le as a mile and never more than twenty-five miles wide, running south, from Leesburg to Sebring, for roughly a hundred miles." John McPhee, 1967
4 ACKNOWLEDGMENTS I would like to acknowledge the financial support and in-kind service of the Clay Minerals Society, the Graduate School of th e University of Florida, the Fl orida Geological Survey, E.I. du Pont de Nemours & Company, and Edgar Minerals (formerly Feldspar Corporation). Without their assistance, this research would have neve r been possible. I would also like to thank my committee, but in particular, Dr. Guerry H. McClellan, my committee chairman, for constantly challenging me during my time at the University of Florida and serving as my mentor both academically and professionally. Additionally, I woul d like to thank some of the other sources of academic, and sometimes personal, advice that impact ed this research and my life as a whole. They include, but are not limited to, Dr. J.L. Eades, Dr. T.M. Scott, Dr. V.J. Hurst, Dr. F.N. Blanchard and Dr. W.G. Harris. There were, of course, many peopl e that assisted with the acqu isition of data necessary to complete this research, or simply were a great source of conversation over a beer (or two). They include Dr. Michael (Mike) Rosenmeier, Dr. Phillip (Phil) Neuhoff, Dr. George Kamenov, William Kenney, Dr. Jehangir (Jango) Bhadha, Ren Bohren, Dr. Richard (Rich) Hisert, Dr. Todd Kincaid, Brickman (Bricky) Way, Christ ian George and Dr. Craig Oyen. Special recognition is afforded to George Kamenov for hi s assistance with the collection of neodymium isotopic data. Without his help, this portion of the study would have never been completed. Additionally, a debt of gratitude is also owed to the staff of the Major Analytical Instrumentation Center at the University of Florida (Wayne Acree in particular) for their assistance with collecting SEM data for this study, often at redu ced rates and sometimes as bartered services. Lastly, I would like to thank my family, for they are the ones that have always been there for me through this extended process, even when it seemed doubtful that I would ever finish. My geologist father has been a huge influence in my career, having introduced me to the wonder that
5 is this science at an early age, and helping to raise me with a strong work ethic and a sense of professionalism. On the other hand, it is my moth er and brother who have been my rocks for many years, and have kept faith when I would fall off the path to finishing. They have also been the ones that have offered moral and emotional support when it has been needed the most during trying times. Without them, I would not be the man I am today.
6 TABLE OF CONTENTS page ACKNOWLEDGMENTS...............................................................................................................4 LIST OF TABLES................................................................................................................. ........10 LIST OF FIGURES.......................................................................................................................12 ABSTRACT...................................................................................................................................17 CHAP TER 1 INTRODUCTION..................................................................................................................20 Hypothesis..............................................................................................................................22 Purpose and Scope..................................................................................................................22 Geologic Setting............................................................................................................... ......24 Neogene Stratigraphic and Structural Fram ework.......................................................... 24 Regional Physiography.................................................................................................... 25 Regional Stratigraphy......................................................................................................30 Miocene series.......................................................................................................... 30 Pliocene series.......................................................................................................... 33 Pleistocene series...................................................................................................... 38 2 REVIEW OF LITERATURE.................................................................................................39 Cypresshead Origin and Age.................................................................................................. 39 Facies Associations.........................................................................................................39 Paleoenvironment............................................................................................................ 41 Age Estimates.................................................................................................................. 42 Correlative Siliciclastics..................................................................................................... ....43 Citronelle Formation....................................................................................................... 43 Miccosukee......................................................................................................................45 Kaolin Origin and Provenance................................................................................................46 Georgia-South Carolina Kaolin District .......................................................................... 46 Cypresshead Formation................................................................................................... 50 3 METHODS.............................................................................................................................53 Sample Localities.............................................................................................................. ......53 Cypresshead Formation (Florida).................................................................................... 53 Cypresshead Formation (Georgia)..................................................................................58 Middle Georgia Kaolin District....................................................................................... 58 Sample Preparation.................................................................................................................59 Analytical Procedures.......................................................................................................... ...60 Grain-Size Analysis......................................................................................................... 60
7 Hydrometer and sieve analysis................................................................................. 61 SediGraph analysis................................................................................................... 62 X-ray Diffraction Analysis..............................................................................................64 Oriented samples......................................................................................................65 Random samples...................................................................................................... 66 Petrographic/Scanning Electron Microscope Analysis................................................... 67 Geochemical Analysis..................................................................................................... 68 Major and trace element analysis............................................................................. 68 Nd isotopic analysis................................................................................................. 68 4 SEDIMENTOLOGICAL AND MINERALOGICAL EVIDENCE FOR THE ORIGIN AND STRATIGRAPHIC SIGNI FICANCE OF THE CYPRESSHE AD FORMATION..... 73 Introduction................................................................................................................... ..........73 Related Deposits and the Siliciclastic Conveyor............................................................. 74 Age Constraints on Cypresshead Deposition.................................................................. 77 Pliocene Paleoclimate...................................................................................................... 78 Results.....................................................................................................................................80 Outcrop and Core Descriptions.......................................................................................80 Sedimentary Framework................................................................................................. 80 Grain-size distributions............................................................................................81 Sedimentary structures............................................................................................. 84 Mineralogy and Petrography........................................................................................... 91 Sand size-fraction..................................................................................................... 92 Clay size-fraction.....................................................................................................94 Facies Architecture..........................................................................................................97 North-central Florida facies................................................................................... 101 Southeastern Georgia facies...................................................................................104 Discussion.............................................................................................................................105 Depositional Environment.............................................................................................105 Cypresshead Formation model............................................................................... 106 Model consistency and the siliciclastic conveyor ..................................................111 Timing and Regional Stratigraphic Correlation............................................................ 115 Pre-Cypresshead siliciclastic flux..........................................................................115 Cypresshead deposition and reworking..................................................................119 Paleoclimate Forcing of Cypresshead Deposition......................................................... 124 Middle to late Miocene clim ate and sedim ent supply............................................ 125 Pliocene climate and the transition toward Northe rn Hemisphere Glaciation (NHG)................................................................................................................. 126 Evidence for increased current and storm activity................................................. 127 Conclusions...........................................................................................................................129 5 EVIDENCE FOR NEOFORMATION AND RE CRYSTALLIZATION OF KAOLINITE IN THE CYPRESSHEAD FORMATION..................................................... 131 Introduction................................................................................................................... ........131 Results...................................................................................................................................132
8 Mineralogy....................................................................................................................132 Kaolinite Disorder and Crystallite Size.........................................................................137 Kaolinite Microtexture.................................................................................................. 149 Particle-Size Analysis....................................................................................................153 Geochemistry................................................................................................................. 154 Major element data................................................................................................. 155 Rare earth element (REE) data............................................................................... 157 Discussion.............................................................................................................................164 Kaolinite Origin............................................................................................................. 164 Kaolinite neoformation..........................................................................................165 Kaolinite recrystallization and disorder................................................................. 170 Accessory Phase Paragenesis........................................................................................ 172 Conclusions...........................................................................................................................175 6 TRACE ELEMENT AND Nd ISOTOPIC EV I DENCE FOR THE PROVENANCE OF CYPRESSHEAD FORMATION KAOLINITIC SANDS................................................... 177 Introduction................................................................................................................... ........177 Southern Piedmont........................................................................................................179 Georgia-South Carolina Kaolin District ........................................................................ 180 Results...................................................................................................................................182 Trace Elements.............................................................................................................. 182 Neodymium (Nd) Isotopes............................................................................................ 185 Discussion.............................................................................................................................190 Trace Element Mobility and Enrich ment...................................................................... 191 Comparison to Georgia-South Ca rolina Kaolin Provenance ......................................... 193 Cypresshead Provenance...............................................................................................194 Conclusions...........................................................................................................................199 7 SUMMARY AND CONCLUSIONS...................................................................................201 Cypresshead Formation Stratigraphy...................................................................................201 Cypresshead Formation Mineralogy..................................................................................... 202 Cypresshead Formation Provenance..................................................................................... 203 APPENDIX A MINE SITE MAPS...............................................................................................................204 B ANALYTICAL PROCEDURES FOR ICP-AES, ICP-MS, AND MC-ICPMS.................. 208 C STRATIGRAPHIC SECTIONS........................................................................................... 215 D GRAIN-SIZE DATA (HYDR OMETER AND SI EVE)...................................................... 223 E X-RAY DIFFRACTION DATA (ORIENTED).................................................................. 233 F X-RAY DIFFRACTION DATA (RANDOM)..................................................................... 249
9 G MINUS-200 MESH PARTICLE-SIZE DATA (SEDIGRAPH).......................................... 266 H MAJOR AND TRACE ELEMENT DATA......................................................................... 278 LIST OF REFERENCES.............................................................................................................283 BIOGRAPHICAL SKETCH.......................................................................................................306
10 LIST OF TABLES Table page 1-1 Southern Atlantic Coastal Plain terraces (Florida, Georgia, and South Carolina).............28 3-1 Sample list and corresponding an alyses perform ed for this study.....................................55 3-2 MC-ICP-MS analyses of the Ames Nd in-house standard................................................ 70 3-3 MC-ICP-MS analyses of the JNdi -1, LaJolla Nd, and BCR-1 standards. ......................... 71 3-4 TIMS analyses of the Ames Nd in-house standard............................................................ 71 4-1 Summary of the lithostratigraphic and sequence stratigraphic nom enclature applied to South Florida siliciclastics............................................................................................. 75 4-2 Clay size-fraction mineralogy of Cypre sshead F ormation and reworked Cypresshead Formation sediments..........................................................................................................95 4-3 Facies summary of the Cypresshead Form ation in north-central peninsular Florida and southeastern Georgia. ..................................................................................................98 4-4 Correlation of siliciclastic depositional events on the Florida Platform with Haq et al. (1988) sequence boundaries, sequence boundari es of Eberli (2000), and sea-level falls identified by Miller et al. (2005). ............................................................................. 120 5-1 Results of disorder calculations for no rth-central Florida Cyp resshead and reworked Cypresshead kaolinite...................................................................................................... 138 5-2 Statistical summary of kaolinite or der and crystallite size calculations. ......................... 141 5-3 Results of crystallite size calculati ons for north-central Florida and Georgia Cypresshead and reworked Cypresshead kaolinite. .........................................................145 5-4 Major element concentrati ons as oxides for Cypresshead Formation clay (< 2 m ) samples........................................................................................................................ .....156 5-5 Correlation matrix of major elements and REE for both Florida and Georgia Cypresshead sam ples....................................................................................................... 157 5-6 REE concentrations of Cypresshead Form ation and related samples.............................. 159 6-1 Trace element concentrations (ppm) and elem ental ratios for Cypresshead Formation and comparison samples.................................................................................................. 183 6-2 Correlation matrix of select trace elements and elem ental ratios for all Cypresshead Formation samples........................................................................................................... 186
11 6-3 Nd isotope data for Cypresshead Formation and comparison samples........................... 187 D-1 Grain-size distributions a nd m oment statistics for the Grandin and Goldhead sand mines in north-central Florida.......................................................................................... 224 D-2 Grain-size distributions a nd m oment statistics for the EPK kaolin mine in northcentral Florida..................................................................................................................225 D-3 Grain-size distributions a nd m oment statistics for the Davenport and Joshua sand mines in central Florida................................................................................................... 225 D-4 Grain-size distributions a nd m oment statistics for Cypresshead Formation sampling locations in southeastern Georgia.................................................................................... 226 H-1 Raw major element concentrations for sam ples used in this study................................. 279 H-2 Minor element concentrations for samples used in this study......................................... 281
12 LIST OF FIGURES Figure page 1-1 Major structural features of Florida a nd southern Georgia influencing Neogene to Holocene sedim entation..................................................................................................... 26 1-2 Pleistocene marine terraces and shorelines of Florida and Georgia. ................................. 27 1-3 Regional stratigraphic correlation char t for Florida and southeast Georgia. .....................31 3-2 Sample processing and analysis flow chart. ...................................................................... 60 3-3 SediGraph concentration te st results of sa mple FRL-19 at concentrations of 1.8 g, 2.0 g, 2.2 g and 2.4 g mixed with 70 ml of dispersant solution......................................... 63 3-4 SediGraph precision test results of sample EPK36-J-12 (56-59) com paring two replicate samples................................................................................................................64 4-1 Histogram illustrating the distribution of m ean grain-si ze values for samples of the Cypresshead Formation and reworked Cypresshead sediments........................................ 82 4-2 Representative grain-size distribution cu rves for Cypresshead Form ation sediments...... 83 4-3 Scatter plot illustrating the relationshi p between mean grain-size and skew ness for samples of Cypresshead Formation a nd reworked Cypresshead sediments......................84 4-4 Example sedimentary structures from the Cypresshead Formation.................................. 85 4-5 Paleocurent rose diagrams for repres entative Cypresshead F ormation exposures............ 88 4-6 Examples of trace fossil a nd bivalve mold occurrences. ................................................... 90 4-7 Photomicrographs of accessory sand-sized phases from the Cypresshead Formation...... 93 4-8 Correlation of Cypresshead Formati on facies in north-central Florida. ............................ 99 4-9 Correlation of Cypresshead Formati on facies in southeastern Georgia........................... 100 4-10 Examples of Cypresshead Formation facies.................................................................... 103 4-11 Correlation chart for Late Miocene through Pliocene siliciclastic units evaluated in this study. .........................................................................................................................116 4-12 Maps illustrating the aerial extent of success ive siliciclastic deposition events impacting the Florida Platform from 8.6 Ma to 1.8 Ma................................................... 117
13 4-13 Relationship between the Cypresshead Form ation, related siliciclastics in southern Florida, seism ic sequence boundaries of Eberli (2000), sequen ce chronostratigraphy, and the sea-level curves of Haq et al. (1988) and Miller et al. (2005)............................. 118 5-1 XRD patterns of example Cypresshead Formation clays................................................ 134 5-2 SEM photomicrographs illustrating charac teristic secondary weathering phases and textures. ............................................................................................................................135 5-3 Histograms illustrating the distribution of results for both kaolinite order (HI and R2) and crystallite size calculations........................................................................................142 5-4 Box-and-whisker diagrams for kaolinite disorder (HI and R2) and crystallite size calculations. .....................................................................................................................143 5-5 Scatterplots illustrating the positive co rrelation be tween the Hinkley Index (HI) and measured CSD values for north-central Florida Cypresshead (and reworked Cypresshead) samples...................................................................................................... 149 5-6 Measured CSD distribution curves and f itted th eoretical lognormal curves (red) for select EPK36-J-12 samples.............................................................................................. 150 5-7 Example CSD distributions illustrating changing c urve shape with increased depth from the surface............................................................................................................... 151 5-8 SEM photomicrographs illustrating the micr otextural variatio n noted in Cypresshead and reworked Cypresshead Formation sediments............................................................ 152 5-9 Example SediGraph particle-size distri butions for select Cypresshead Form ation samples........................................................................................................................ .....154 5-10 Scatterplots illustrating mixing trends relate d to the presence of crandallite-florencite series minerals in Cypresshead Formation clays............................................................. 158 5-11 Chondrite-normalized REE distribution pa tterns for select Cypresshead F ormation clay (< 2 m) fraction and related samples...................................................................... 161 5-12 Histogram illustrating the distribution of Eu/Eu* values for both Georgia and Florida Cypresshead Form ation clays.......................................................................................... 164 5-13 SEM photomicrographs of Cypresshead and reworked Cypresshead Formation kaolin ite textures associated with the in situ weathering of muscovite...........................167 5-14 Correlation of Eu/Eu* values to CS D (volum e-weighted mean thickness) calculations......................................................................................................................173 6-1 Tectonostratigraphic terran es and granites proposed as potential source m aterials for Cypresshead Formation sediments.................................................................................. 181
14 6-2 Multi-element normalized diagrams for Cypresshead and comparison samples, norm alized against average continental crust.................................................................. 184 6-3 143Nd/144Nd versus 147Sm/144Nd isochron diagram for Cypresshead Formation samples........................................................................................................................ .....189 6-4 Nd(t) versus Nd concentration scatterplo t for Cypresshead Formation samples............. 189 6-5 Histogram illustrating the distribution of Nd model age (TDM) results for both Cypresshead Formation clay samples and comparison formations................................. 190 6-6 Trace element plots of Cypresshead Fo rm ation clays illustrating evidence of weathering, provenance, and sedi ment recycling processes............................................ 192 6-7 Comparison of Cypresshead Formation trace ele ment concentrations to the results of Dombrowski (1992; 1993)...............................................................................................195 6-8 Plot of Nd(t) versus Th/Sc for Cypresshead Formation samples based on the model of McLennan et al. (1990; 1993)..................................................................................... 197 6-9 Plot of Nd(t) versus stratigraphic age for the C ypresshead Formation compared to the Nd isotopic evolution of potential sources.......................................................................197 A-1 Site map for the Edgar Minerals EPK Mine....................................................................205 A-2 Site map for the VMC Goldhead Sand Mine...................................................................205 A-3 Site map for the VMC Grandin Sand Mine..................................................................... 206 A-4 Site map for the CEMEX Davenport Sand Mine.............................................................206 A-5 Site map for the CEMEX Joshua Sand Mine...................................................................207 C-1 Stratigraphic section for Gr andin Sand Mine section FRG-1. .........................................216 C-2 Stratigraphic section for Gr andin Sand Mine section FRG-2. .........................................217 C-3 Stratigraphic section for Gol dhead Sand Mine section FRL-1. .......................................218 C-4 Stratigraphic section for Jo shua Sand Mine core SSJ-1. ................................................. 219 C-5 Stratigraphic section for Da venport Sand Mine core SSD-1. .......................................... 220 C-6 Stratigraphic section for Jesup section J-1.......................................................................221 C-7 Stratigraphic section for Birds section B-1......................................................................221 C-8 Stratigraphic section fo r Linden Bluff section L-1. ......................................................... 222
15 D-1 Grain-size distribution curves for EPK Mine core EPK36-J-12. .....................................227 D-2 Grain-size distribution curves for EPK Mine core EPK31-P-40. ....................................227 D-3 Grain-size distribution curves for EPK Mine core EPK30-V-6. ..................................... 228 D-4 Grain-size distribution curves for Grandin Sand Mine section FGR-1. .......................... 228 D-5 Grain-size distribution curves for Grandin Sand Mine section FRG-2. .......................... 229 D-6 Grain-size distribution curves for Goldhead Sand Mine section FRL-1. ........................229 D-7 Grain-size distribution curves for Joshua Sand Mine core SSJ-1. ...................................230 D-8 Grain-size distribution curves for Davenport Sand Mine core SSD-1. ........................... 230 D-9 Grain-size distribution curv es for Jesup section J-1. .......................................................231 D-10 Grain-size distribution curves for Linden Bluff s ection L-1...........................................231 D-11 Grain-size distribution curv es for Birds section B-1. ...................................................... 232 G-1 SediGraph particle-size plots for the m inus-200 mesh (< 75 m) fraction of select EPK Mine samples (EPK36-J-12)................................................................................... 267 G-2 SediGraph particle-size plots for the m inus-200 mesh (< 75 m) fraction of select EPK Mine samples (EPK31-P-40)................................................................................... 268 G-3 SediGraph particle-size plots for the m inus-200 mesh (< 75 m) fraction of select EPK Mine samples (EPK30-V-6).................................................................................... 269 G-4 SediGraph particle-size plots for the m inus-200 mesh (< 75 m) fraction of select Grandin Sand Mine samples (FRG-1)............................................................................. 270 G-5 SediGraph particle-size plots for the m inus-200 mesh (< 75 m) fraction of select Grandin Sand Mine samples (FRG-2)............................................................................. 271 G-6 SediGraph particle-size plots for the m inus-200 mesh (< 75 m) fraction of select Goldhead Sand Mine samples (FRL-1)........................................................................... 272 G-7 SediGraph particle-size plots for the m inus-200 mesh (< 75 m) fraction of select Joshua Sand Mine samples (SSJ-1)................................................................................. 273 G-8 SediGraph particle-size plots for the m inus-200 mesh (< 75 m) fraction of select Davenport Sand Mine samples (SSD-1).......................................................................... 274 G-9 SediGraph particle-size plots for the m inus-200 mesh (< 75 m) fraction of select Jesup type locality samples (J-1)..................................................................................... 275
16 G-10 SediGraph particle-size plots for the m inus-200 mesh (< 75 m) fraction of select Linden Bluff reference locality samples (L-1)................................................................. 276 G-11 SediGraph particle-size plots for the m inus-200 mesh (< 75 m) fraction of select Birds reference locality samples (B-1)............................................................................ 277
17 Abstract of Dissertation Pres ented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy ORIGIN AND STRATIGRAPHIC SIGNIFICANC E OF KAOLINITIC SEDIMENTS FROM THE CYPRESSHEAD FORMATION: A SEDI MENTALOGICAL, MINERALOGICAL AND GEOCHEMICAL INVESTIGATION By Kendall Brian Fountain December 2009 Chair: Guerry H. McClellan Major: Geology Kaolinitic sediments of the Cypresshead Form ation (3.4.3 Ma), deposited as two distinct shoreface-shelf parasequences in response to se a level falls at 3.3 Ma and 2.5 Ma, define the final stages of a period of siliciclastic deposit ion which dominated the Florida Platform between 8.6 Ma and 1.8 Ma. Beginning with deposition of the Late Mio cene SS2 siliciclastics of Cunningham et al. (2003) and ending with the Late Pliocene cessation of Cypresshead deposition and reworking at approximately 1.8 Ma, episodes of sediment accumulati on correlate with two paleoclimatic transitions from; (1) arid conditions during the Late Miocen e to continual El Nio conditions during the early Pliocene warm period (~4.5.0 Ma), and (2) continual El Nio conditions to global cooling and the onset of si gnificant Northern Hemisphere Glaciation (NHG) (~3.0.5 Ma). Responding to the interplay of se a-level, sediment supply and accommodation associated with climate instability, the silicicl astics deposited during th is Late Miocene through Pliocene interval define a retrogradational parase quence set episodically deposited on the Florida platform over a 6.8 Ma period. Facies identified within the Cypresshead Form ation define the depositional environment as nearshore marine, consistent with a wave-domin ated coastline in north-central Florida and a
18 mixed energy coastline in southeastern Georgia. Composed of coarsening-upward sequences, both facies sets define a progradational response to the coastal delivery and subsequent coastparallel transportation of a subs tantial siliciclastic flux, which, in turn, correlates with the nonequilibrium landscape conditions predicted by a transitional mid-Pliocene paleoclimate. A potential fluvial-deltaic compone nt of the Cypresshead Formati on in southeastern Georgia is proposed based on paleocurrent indicators and the similarities noted for the offshore transition (OST) facies in Georgia and the prodeltaic Miccosukee Formation. Mineralogically, Cypresshead Formation sediments are highly weathered, but particularly in north-central Florida where local hydrologic c onditions coupled with the depositional fabric and microtexture of the sediments have resulted in the formation of at least three kaolinite fractions; (1) in situ kaolinite formed at the expense of fe ldspars and mica, (2) detrital kaolinite deposited as part of the original clay mineral suite, and (3) near surface recrystallized kaolinite. In situ kaolinite crystallizes via the combined topotactic (transformation) and epitactic (neoformation) weathering of muscovite mica under saturate d conditions to produce an enrichment of vermicular kaolinite in Cypressh ead sediments. Kaolinite formed in this way exhibits a high degree of order in response to the epitactic nucleation of crystallites on the structurally similar muscovite su rface, explaining the correlati on between low disorder and small particle-size that has confounded researchers fo r decades. Additionally, the feldspar dissolution sourcing of the Al and Si necessary for in situ kaolinite crystallization is confirmed by inherited positive Eu anomalies reported for these kaolinites and the presence (and geochemistry) of residual feldspars in basal Cypresshead sediments. Weathering occurring under oxic, vadose or mixed vadose/satu rated conditions results in the degradation of neoformed kaolinite as ev idenced by a decrease in kaolinite order and
19 coherent scattering domain (CSD) values consis tent with the formation of the near surface recrystallized kaolinite fraction. Forming under conditions of extreme leaching and recystallization, this kaolinite fraction possesses microtextural characteristics suggestive of a pedogenic origin. The trace phases gibbsite, halloys ite and crandallite-flore ncite also form under similar near surface, oxic weatheri ng conditions. The origin of the cr andallite-florencite phase is uncertain, but most likely originated from th e decomposition of post-depositional pore water organics coupled with detr ital mineral dissolution. Both trace elements and Nd isotopes were an alyzed in order to constrain the provenance and broader geochemical characteristics of th e Cypresshead Formation and to compare these results with the studies by Dombrowski ( 1992; 1993) on the provenance of Cretaceous and Tertiary kaolins from the Georgia-South Carolina kaolin district. Of the trace elements used by Dombrowski, only Th and Sc appear to be resistant to significant mobilization and depletion in Cypresshead sediments, with the Th/Sc rati o a robust indicator of provenance composition. Additionally, a significant propor tion of Florida Cypresshead clays possess Th/U ratios well below that of average upper crust, and appear to have undergone redox-driven U enrichment dictated by pore water organics. As for Nd resu lts, Cypresshead Formation samples appear to originate from sources intermediate between those associated with Cretaceous soft kaolins and Tertiary hard kaolins of Georgia and South Caro lina. This is consistent with these sediments being a mixture of materials originating from Ca rolina terrane and Alleghanian granite sources. Additionally, Nd model ages (TDM) for Florida and Georgia Cypr esshead samples range between 1.4-0.9 Ga and 1.5-0.8 Ga respectively, with both possessing an average TDM value of 1.1 Ga consistent with the age of Grenville crust.
20 CHAPTER 1 INTRODUCTION Kaolin deposits in the sout heastern United States are important economically as an industrial mineral and geol ogically as an indicator of past climatic and depositional conditions. As a result, the Cretaceous and Tertiary deposit s of the Georgia-South Ca rolina kaolin district, one of the largest and most valu able coastal plain kaolin deposits in the world (Patterson and Murray, 1984), have been the focus of extensive re search. In comparison, the kaolinitic sands of the Cypresshead Formation, actively exploited for kaolin at the Edgar Minerals EPK Mine in north-central Florida, have rece ived only limited study due in part to their minor economic value and poorly known geologic history. From an industrial perspectiv e, kaolinite, the essential clay mineral component in kaolin (or china clay), ball clay, fire clay (or refractory clay), and flint clay, possesses an assortment of unique properties which make it suitable for a vari ety of applications. In fact, the kaolin-bearing sediments of the Georgia-South Carolina kaolin district constitute one of the worlds leading sources of high quality kaolin, representing a ne arly $1 billion/year i ndustry. These high quality Coastal Plain kaolins are exploited due to s uperior properties suitable to film formation necessary in paper coating, by far the most economically important application of kaolin. This use commands the most stringent specifications defined by high brightness, suitable particle size, and low abrasion. Other kaol in applications include paint, ink, fiber extension, polymer extension and reinforcement, catalysts, fiber gla ss, and use as a carrier, adsorbent, or diluent (Bundy, 1993). Kaolin associated with the Edgar Minerals EPK deposit lacks the properties suitable for use in the paper i ndustry, but possesses unusually good forming characteristics and high green strength which make it a superior product for use in the ceramics industry.
21 Of particular significance to this study is the un certainty that exists as to the origin of the kaolin within the Cypresshead Fo rmation, the origin of the unit it self, and the importance of the unit within the context of Florida Platform and broader Coastal Plain deposition. In particular, confusion persists as to whether th e kaolin is primary, having formed in situ through postdepositional alteration of detrital feldspars, micas, and/or primary clays, or secondary, having formed elsewhere only to be eroded, transported, and deposited as the kaolinite-rich siliciclastics now observed (Sellards, 1912; Bell, 1924; Pirkle, 1960). Of further inte rest is the significance of the general low disorder/fine part icle size properties of kaolin from this unit. First noted by Fountain and McClellan (1993), this characteristic is inconsistent with the existing models of Georgia-South Carolina kaolin formation, for which low defect/coarse particle size and high defect/fine particle size are the rule (Pickering and Hurst, 1989; Pickering et al., 1997). In order to answer these questions, this study seeks to define the depositional framework, provenance, mineralogy, and weathering history of the kaolinit ic sediments associated with the Cypresshead Formation, and to constrain this deposit within a regional cont ext of siliciclastic deposition. This dissertation has been orga nized into three separate pape rs for future publication. The first of these papers, entitled Sedimentological and mineralogical evidence for the origin and stratigraphic significance of the Cypresshead Formation defi nes the regional stratigraphy and sedimentology of the Cypresshead Formation, and correlates the unit with genetically related siliciclastics in order to define a sequence st ratigraphic model for Late Miocene to Pliocene deposition on the Florida Platform. The second paper, entitled Evidence for neoformation and recrystallization of kaol inite in the Cypresshead Formati on defines the recrystallization processes impacting the mineralogy and industr ial properties of Cypresshead Formation kaolinite, and answers the primary verses secondary origin is sue associated with these clays.
22 Lastly, the third paper, entitled Trace element and Nd isotopic evidence for the provenance of Cypresshead Formation kaolinitic sands identifies the ultimate provenance of Cypresshead Formation kaolinite, addressing the role of sedi ment sources on kaolin physical properties. Hypothesis The hypothesis to be tested in th is study is that the Cypresshead Formation in north-central Florida and eastern Georgia fits into a regional model of ep isodic kaolin deposition on the southern Atlantic coastal plain, which began in the Cretaceous and continued through the Pliocene under specific envir onmental conditions favorable for formation and subsequent preservation of kaolinitic sediments. As part of this model, it is proposed that the kaolinite contained within the Cypresshead Forma tion possesses both primary and secondary characteristics consistent with deposition as pa rt of an episodic siliciclastic flux, followed by post-depositional weathering and recr ystallization. It is hypothesized that: (1) precursor materials eroded from source areas located within the Pi edmont-Blue Ridge region, the Georgia-South Carolina kaolin district, and/or the southeastern Georgia Coastal Plain are responsible for the bulk of kaolinite accumulation in Cypresshead Fo rmation sediments, (2) subsequent weathering is responsible for increased kaolin ite content and a substantial modi fication of the original fabric and mineralogy of the kaolinitic sediments, and (3) the Cypresshead Formation is part of a genetically related sequence of siliciclastics de posited episodica lly during the Late Miocene to Pliocene, which formed in response to a favorable interplay of paleoclimate, weathering and sealevel fluctuations. Purpose and Scope The ultimate industrial potential of a kaolin deposit is governed, in part, by the geological conditions operating both before and after depos ition. As a consequence, this study seeks to define the geological factors influencing Pliocene siliciclas tic deposition and subsequent
23 weathering on the Florida Platform and southeaste rn Georgia Coastal Plain using a combination of stratigraphic, mineralogical and geochemical techniques. As part of this effort, the stratigraphic framework for the Cypresshead Formation will be evaluated in light of recent studies which have added signifi cant insight into the deposition of related Late Miocene through Pliocene siliciclastics in southern Florida. To accomplish this, both sedimentological and mineralogical indicators of Cypresshead Form ation timing and deposition were reassessed through field-based observation and measurement, and through a review of previous studies on the unit and its correlatives. B ecause siliciclastic sedimentation has dominated much of the southeastern United States episodically since the Cretaceous, it is vital to understand sedimentation patterns and the changing nature of clastic sediment sources. By defining the genetic characteristics of these de posits, the processes influencing both the origin and subsequent modification of kaolin-beari ng units are better understood. Additionally, the integration of geochemical techniques and de tailed clay mineralogy of kaolin deposits as an effective method for provenance determination was evaluated. This approach has not been previously employe d in the study of Neogene Coastal Plain sedimentation, and was undertaken in response to previous stud ies of Cypresshead Formation sediments using traditional methods of provena nce analysis (paleocurrent indicators, heavy mineral analysis, and standard petrography), wh ich have resulted in ambiguous results and conflicting interpretations (Sellards, 1912; Bell, 1924; Pirkle, 1960; Pirkle et al., 1964; Kane, 1984). To this end, trace elements, particularly ra re earth elements (REEs), and neodymium (Nd) isotopic systematics were employed to constrain kaolinite provenance. In general, geochemical approaches to sedimentary provenance of fine-grained sediments are more useful than traditional petrographic techniques due to li mitations associated with part icle size (McLennan et al., 1993).
24 Geologic Setting Neogene Stratigraphic and Structural Framework The Florida Peninsula and portions of southeastern Georgia are underlain by as much as 1220 m of sedimentary rocks. This sedimentar y sequence of predominately carbonate units comprises the broad, relatively flat Florida Platfo rm, of which, the Florida Peninsula, veneered by a relatively thin sequence of predominately si liciclastic sediments, is the emergent portion (Scott et al., 1980). Beginning in the Oligocene, siliciclastics, incl uding quartz sands, silts, and clays, began to supplant carbonate deposition on the Florida Plat form in response to changes in both the structural controls on depositi on and sediment sources (Sco tt, 1992a; Cunningham et al., 2003). Prior to that time, from the Middle Cretaceous to the Late Paleogene, siliciclastic sediment supply from Appalachian sources to the north had been restricted by the Georgia Channel System (Huddlestun, 1993; includes the Gulf Tr ough and Suwannee Straits), effectively isolating the carbonate platform from a siliciclastic influx (Scott, 1992a; Cunningham et al., 2003). However, during the Late Eocene through Oligocene, uplift and/or crustal arching associated with rejuvenation of the Appalachians (Stuck ey, 1965; Dennison and Stewart, 2001; Stewart and Dennison, 2006), along with weakening of the cu rrent through the Georgia Channel System (Huddlestun, 1993), led to a renewed supply of siliciclastics flooding th e southeastern North American coastline. As a result, a major infl ux of sediments eventua lly filled the channel, blocking the current from reoccupying th e Georgia Channel System, and permitted encroachment of the siliciclastic wedge onto th e Florida Platform, and by the Late Miocene, siliciclastic deposition dominated the Florida Pl atform, particularly in the north and central peninsular areas (Scott, 1992a; 1997; Cunningham et al., 2003). Th e present highlands of the northern peninsula and panhandle of Florida are the dissected rema ins of the fluvial, deltaic, and
25 shallow-water marine deposits associated with this Neogene siliciclastic influx (Schmidt, 1997). Ultimately, these sediments were reworked and/or reshaped by subsequent sea-level fluctuations and associated nearshore, coast-parallel currents into the elongate system of upland ridges seen today (Schmidt, 1997). The Neogene structural setting of the Flor ida Peninsula and the southeastern Georgia Coastal Plain is shown in Figure 1-1. In the study area, Neogene to Holocene sediments unconformably overlie an eroded and karstified Paleogene surface characterized by a multitude of structural highs and basi ns (Scott, 1997). Among the highs are the Ocala Platform (or uplift), the Chattahoochee Antic line, the Sanford High, the St. J ohns Platform, and the Brevard Platform. Among the basins are the Georgia Ch annel System (comprising the Apalachicola Embayment, Gulf Trough, and Southeast Georgia Embayment), the Osceola Low, and the Okeechobee Basin. These features variably affect ed Neogene sedimentation, in part, directing siliciclastic fluxes on the Florida Platform. The reader is referre d to Huddlestun (1988) and Scott (1988b) for a detailed review of the structural framework of Florida and southeastern Georgia. Regional Physiography Some of the most distinguishable geomorphi c features of peninsular Florida and the coastal plain of Georgia are the marine terrace s and ancient shorelines generated by sea-level fluctuations that drastically a ltered the physiography of the region. Six to eight coastal terraces of Pleistocene age approximately parallel the Atlan tic coast of Georgia and Florida (Fig. 1-2, Table 1-1). Historically, various aut hors have assigned diffe rent elevations and terminology to these terraces, based mainly on physiographic evidence. These terraces were fi rst identified by Cooke (1930; 1945),who along with others (Hoyt and Hails 1974), held that each terrace was, in fact, a formation consisting of a barrie r island facies on the seaward si de and a marsh-lagoon facies on the landward side. This was later deemed to be an incorrect observation by Huddlestun (1988),
26 Figure 1-1. Major structural feat ures of Florida and southern Georgia influencing Neogene to Holocene sedimentation (modi fied after Scott, 1997). who noted that instead of being stratigraphic un its, the marine terraces are simply geomorphic features. As a result, the use of terrace terminology in a lithostr atigraphic context has ceased by most authors because there is no genetic relati onship between any singl e terrace surface and any single lithostratigraphic unit along the Florida and Georgia coastal plains.
27 Figure 1-2. Pleistocene marine terraces and shorelines of Florida and Georgia (modified after Healy, 1975; Huddlestun, 1988).
28 Table 1-1. Southern Atlantic Coastal Plain te rraces (Florida, Georgia, and South Carolina) (modified after Pirkle et al., 1970; Healy, 1975). (ft)(m) After Cooke (1939; 1945) Hazlehurst/Brandywine27082.3 Aftonian Coharie 21565.5 Yarmouth Sunderland 17051.8 Yarmouth Wicomico 10030.5 Sangamon Penholoway 7021.3 Sangamon Talbot 4212.8 Sangamon Pamlico 257.6 Wisconsin After MacNeil (1950) High Pliocene terrace28085.4 Aftonian Okefenokee 15045.7 Yarmouth Wicomico 10030.5 Sangamon Pamlico 25-357.6-10.7 Wisconsin Silver Bluff 8-102.4-3 Recent Hoyt and Hails (1974) Wicomico 95-10029-30.5 Pleistocene Penholoway 70-7521.3-22.9Pleistocene Talbot 40-4512.2-13.7Pleistocene Pamlico 247.3 Pleistocene Princess Anne 13 4 Pleistocene Silver Bluff 4.51.4 Pleistocene After Healy (1975) Hazlehurst 215-32065.5-97.6Miocene or Pliocene Coharie 170-21551.8-65.5Pleistocene Sunderland/Okefenokee100-17030.5-51.8Pleistocene Wicomico 70-10021.3-30.5Pleistocene Penholoway 42-7012.8-21.3Pleistocene Talbot 25-427.6-12.8 Pleistocene Pamlico 8-252.4-7.6 Pleistocene Silver Bluff 1-100.3-3 Pleistocene Elevation above msl Age Terrace Regional correlation of the marine terraces has been the goal of many authors. However, as noted by Winker and Howard (1977a; b), the terraces in the Carolinas and northern Florida have been uplifted, confusing correlati ons made on the basis of eleva tion alone. Opdyke et al. (1984) have explained that the uplift in Florida is due to isostatic rebound accompanying solution of limestone in the karst regions of central Florida. Based on the m easurement of dissolved solids in Floridas springs, Opdyke et al. (1984) estimated that the uplift in the north-central Florida
29 peninsula has been at least 36 m during the Pl eistocene and Holocene, in agreement with observations of marine fossil occurr ences associated with the terraces. The distribution of marine terrace sediments in peninsular Florida and southeastern Georgia is dominated by a ridge morphology. The hi ghest of these ridges is Trail Ridge (Fig. 12), located on the western flank of the D uval Upland, an area underlain by Cypresshead Formation sediments. It is believed to be a relict barrier or spit corre sponding to a Wicomico shoreline (29 m) that may have formed at the crest of a marine transgression from erosion of the Northern Highlands located further to the we st (Pirkle et al., 1974). The age of Trail Ridge has long been debated, with the discovery of shal low marine fossils in Tr ail Ridge indicating an age no older than late Pliocene or Pleistocene (Pirkle and Czel, 1983). The Penholoway shoreline marks the eastern limit of the Cypresshead Formation exposed at the surface in eastern Ge orgia and northern Florida, and is bounded by a 9 m scarp along its eastern edge. The scarp likely indicates a peri od of sea-level transgression responsible for the present distribution and ridge mo rphology of Cypresshead Formation sediments in north-central Florida and eastern Georgia. A lthough eroded to form a younger shoreline than the Wicomico, Cypresshead Formation sediments are st ratigraphically older than Trail Ridge. The Lake Wales Ridge (Fig. 12) is a distinct topographi c feature running roughly northsouth for more than 100 mi (161 km) along the center of north-central peninsular Florida (White, 1958; White, 1970; Scott, 1980). This feature contains some of the highest elevations in Florida, rising over 320 ft (97.6 m) above sea-level at Iron Mountain in Polk County, and contains the thickest sections of Cypresshead sediments in th e state. It is a well defined topographic high from the central part of Lake County southward to its terminus near Lake Placid, in Highlands County, but is not well defined in its northern pa rt, most likely as a result of
30 Pleistocene erosion and karsti fication (Pirkle, 1960; White 1970). The Winter Haven and Lakeland ridges in Polk County, Florida, ar e, like the Lake Wales Ridge, underlain by Cypresshead Formation sediments, and interpre ted to be erosional fe atures produced during Pleistocene sea-level highstands. The northern re mnant of the Lake Wales Ridge, located in the area of the Interlachen Karstic Highlands of A rrington (1985), is flanked on the west and north by Trail Ridge, and on the east by the Baywood Prom ontory (Pirkle et al., 1970) (Fig. 1-2). As the base of Trail Ridge sediments are located at 150 ft (45.7 m) ms l, it is likely that the seas which deposited the sediments of this ridge also inundated the northern extension of the Lake Wales Ridge, thereby destroying, via wave planation, the ridge morphology observed in central peninsular Florida. Regional Stratigraphy The Neogene stratigraphy of southeastern Georgia and peninsular Florida records a major transition toward episodic siliciclastic deposi tion in response to fluctuating sea-level and sediment sources. The timing and nomenclature a ssigned to the various units that record this history have long been a source of controversy among earlier wo rkers (Scott, 1992b). Within the region included with this study, the presently ac cepted stratigraphic nomenclature is outlined in Figure 1-3. Miocene series Hawthorn Group: northern peninsular Florida and southeastern Georgia. From northern peninsular Florida to southeastern Ge orgia, Hawthorn Group sediments, in general, disconformably underlie the Cypresshead Formation. The Hawthorn Group is easily distinguished from the overlying Cypresshead Fo rmation in being typically thick-bedded and massive, commonly phosphatic (except where it grades into the Altamaha Formation), argillaceous, and locally dolomitic, calcareous and siliceous (Huddlestun, 1988). Three
31 Figure 1-3. Regional stratigraphic correlation chart for Florida a nd southeast Georgia (modified after Huddlestun, 1988; Scott, 1992a; 2001; Scott et al., 2001). Hawthorn Group formations share disconformable contacts with the overlying Cypresshead Formation in northern peninsular Florida and south eastern Georgia. These units are, from oldest to youngest, the Parachucla Formation, the Marks Head Formation, a nd the Coosawhatchie Formation. The reader is referred to Huddlest un (1988) and Scott (1988b) for a more detailed review of these units, including type localities, lithologies, and stratigraphic relationships. Hawthorn Group: central and southern peninsular Florida. In central and southern peninsular Florida, the Cypresshead Formati on disconformably overlies either undifferentiated Hawthorn Group sediments or those of the Peace River Formation of the Hawthorn Group. Undifferentiated Hawthorn Group se diments are limited to the vicin ity of Lake County in central Florida where the transition between northern Florida Coosawhatchie Formation and central Florida Peace River Formation lithologies occurs (Scott, 1988b). As for the middle Miocene to
32 early Pliocene Peace River Formation of Scott ( 1988b), lithologies consist of interbedded quartz sands, clays, and carbonates, with the dominant (two-thirds or more) siliciclastic component distinguishing this unit. The very fineto medium-grained quartz sands are characteristically clayey, calcareous to dolomitic, and variably pho sphatic, with phosphate concentrations greatest in the Bone Valley Member of the unit. Clay beds are common in the Peace River Formation, and tend to be dominated by smectite and pal ygorskite (Reynolds, 1962; McClellan and Van Kauwenbergh, 1990; Scott, 1988b). Scott (1988b) corre lates the lower part of the Peace River Formation with the Coosawhatchie and Statenv ille formations in northern Florida, based on stratigraphic position, diatom s of middle Miocene age, and the occurrence of an early to middle Barstovian age vertebrate fauna (Webb and Crissi nger, 1983) at the base of the unit. The reader is referred to Scott (1988b) for a more deta iled review of the Peace River Formation, including type locality, lithology, and stratigraphic relationships. Altamaha Formation. The Altamaha Formation of Huddlestun (1988), the principle Miocene fluvial lithofacies found in Georgia, is the most widely occurring outcropping formation in the Georgia Coastal Plain (Huddl estun, 1993). The blanket deposit of kaolinitic sands and sandy kaolins of this formation are consistent with deposit ion by braided streams disgorging their sediment loads from the Piedmont (Huddlestun, 1988; 1993). Lithofacies include both overbank and channel-fill deposits c onsisting of variably indurated to nonindurated, pebbly, feldspathic, kaolinitic sands and sandy kaolins that ar e interpreted to have been generated by braided streams de positing their sediment loads from the Piedmont onto the coastal plain (Huddlestun, 1988). As noted, the clay mineral suite of the Altamaha Formation is dominated by kaolinite with illite and smectite as minor constituents.
33 In some regions of Georgia, the Altamaha Fo rmation is divisible in to an upper and lower part (Huddlestun, 1988). The lower part is typically thick bedded, massive, sandy clays and argillaceous sands to claystones a nd sandstones, while the upper pa rt is typically a prominently cross-bedded, pebbly to gravelly sand with clay lenses. The latter of these units, assigned by Huddlestun (1988) to the Screven Member, is particularly well developed in the Altamaha and Satilla River area, and consists of a maze of fl uvial channels and cut-and-fill structures with corresponding channel-fill lithologies. Given an average thickness of between 100 and 200 feet (30-60 m) (Huddlestun, 1988), and its widespread occurrence in southeastern Georgia, the Altamaha Formation represents a large reservoir of siliciclastics similar in character to the Pliocene siliciclastics of the Cypresshead Formati on. The reader is referred to Huddlestun (1988) for a more detailed review of the Altamaha Fo rmation, including type locality, lithology, and stratigraphic relationships. Pliocene series Cypresshead Formation. The name Cypresshead Formation was first used by Huddlestun (1988) to describe a prominently thinto th ick-bedded and massive, planarto cross-bedded, variably burrowed and bioturbated, fine-grained to pebbly, coarse-grained sand formation in the terrace region of eastern Georgia. Subsequently the name was extended into Florida by Scott (1988a) to encompass sediments in peninsular Florida previously assi gned to the Citronelle Formation of Matson (1916). As defined by Huddlestun (1988) and Scott (1988 a), the Late Pliocene (early Piacenzian) to early Pleistocene (Calabrian) Cypresshead Formation is composed entirely of siliciclastics; predominately quartz and clay minerals, with qu artz and/or quartzite pebbles locally abundant (Pirkle et al., 1970). The unit is characteristically a mottled, fineto very coarse-grained, often gravelly, variably clayey qua rtz sand, containing minor amount s of feldspar, mica and heavy
34 minerals (Scott, 1988a). Sediments vary from po orlyto well-sorted a nd angular to subrounded, with induration generally poor to nonindurate d. The binding matrix or cementing agent is normally clay, although iron oxide cement is known to occur. In areas where the Cypresshead outcrops, the sediments are charac teristically oxidized and mottl ed, exhibiting shades of red, orange, and white (Scott, 1988a). As noted, clays are present throughout the C ypresshead Formation as a binding agent and occasionally as a primary lithology. Clay content of the sediments appears to decrease in a general north to south trend with higher averag e clay contents in southern Georgia than in northern Florida. The clay mineral present in th e oxidized, mottled portion is characteristically kaolinite while in the downdip unoxidized portion illite and smectite are reported to dominate (Scott, 1988a; 1992a). Although not yet recognized in Florida, shells are reported to occur very sporadically near the base of the Cypresshead Formation in southeastern Georgia (Huddlestun, 1988). The Cypresshead Formation is known to extend as far north as Dorchester County, South Carolina, and to be widespread in southeastern Georgia and in the Central Highlands of the Florida peninsula, south to Highlands County, although the extent of the Cypresshead Formation has not been accurately mapped in this area (Sco tt, 1992a). In Florida, the unit thins toward the west onto the flanks of the Ocala Platform, and appears to extend into the subsurface south of Highlands County. In Georgia, north of the Altamaha River, the western limit of the Cypresshead Formation occurs at or a few kilome ters west of the Orangeburg Escarpment, while south of the river, the formation occurs west of the escarpment in nor thern Wayne County, and immediately west of Trail Ridge (Huddlestun, 19 88). In both Florida and Georgia, the eastern edge of the Cypresshead appears to be truncated or grad es laterally into age equivalent sands and
35 marls (e.g. Nashua Formation and/or Raysor Form ation). The Cypresshead is thickest in the Central Highlands of Florida, where it crops out and may attain thicknesses in excess of 200 ft (~60 m) in Lake County. The reader is referre d to Huddlestun (1988) and Scott (1988a; 1992a; 1997) for a more detailed review of the Cypr esshead Formation, incl uding type locality, lithology, and stratigraphic relationships. Tamiami Formation. The late Early Pliocene to Late Pliocene Tamiami Formation of Mansfield (1939) and Missimer (1992; 1993) is a poor ly defined unit contai ning a wide range of mixed carbonate/siliciclastic lithologies, including limestone, sandstone, quartz sand, marl, shell, and clay (Missimer, 1992). This unit disconf ormably overlies Hawthorn Group sediments along an erosional contact marking a major Early Pl iocene (Zanclean) regression. Illustrating its lithological diversity, the Tamiami Formati on contains a number of named and unnamed members, including, but not limited to, the Bu ckingham Limestone, the Ochopee Limestone, the Bonita Springs Marl, the Golden Gate Reef, and the Pinecrest beds (Sand) of Olsson and Petit (1964) which are, in fact, a series of biofacies and lithologies that are mappable only over limited areas (Missimer, 1992; Scott, 1992a). The complex relationships exhibited by these facies are due, in part, to the diverse depositional environm ents involved in formation of the unit (Scott, 1992a). The reader is referred to Missimer (1992; 1993; 1997; 2001a; b) for a more detailed review of the Tamiami Formation, includi ng type locality, lith ology, and stratigraphic relationships. Raysor Formation and the unnamed Raysor-equivalent shelly sand. The early Late Pliocene (early Piacenzian) Raysor Formation (Raysor Marl) of C ooke (1936) is a soft, variably shelly, slightly argillaceous, calcareous quartz sa nd in southeastern Georgia. The quartz sand is typically fine-grained and wellsorted, although coarse sands and pebbles (quartz and feldspar)
36 have been reported to occur in basal sediment s of the unit (Huddlest un, 1988). The early Late Pliocene (Piacenzian) unnamed Raysor-equivalent shelly sand of Huddlestun (1988) is similar lithologically to the Raysor Formation, but is re stricted to the coasta l area of Georgia. Huddlestun (1988) has described the Cypre sshead Formation as disconformably or paraconformably overlying the Raysor Formati on in both Effingham and Wayne Counties, and the unnamed Raysor-equivalent shelly sand along the Georgia coast. Planktonic foraminifera recovered by Huddlestun (1988) from the Raysor Formation and the unnamed Raysor-equivalent shelly sand in Georgia are consistent with Zone PL3 of Berggren (1973), or in the case of the unnamed Raysor-equivalent shelly sand, PL3 or PL 4, with PL3 roughly equivalent to Zone N20 of Blow (1969). The reader is referred to Huddl estun (1988) for a more detailed review of the Raysor Formation and the unnamed Raysor-equiva lent shelly sand, incl uding type locality, lithology, and stratigraphic relationships. Nashua Formation. Formerly the Nashua Marl of Matson and Clapp (1909), the late Pliocene (Piacenzian) to early Pl eistocene (Calabrian) Nashua Formation of Huddlestun (1988) is a fossiliferous, variably calcareous, someti mes clayey, quartz sand, with mollusks as the dominant fossil type (Scott, 1992a). Quartz sa nd is the dominant lith ic component of the formation, ranging in grain-size from medium to fine, and constitutes the bulk of the unit to the west where it seems to grade laterally into the Cypresshead Formation (Huddlestun, 1988). The Nashua Formation appears to underlie the St. Johns River area at least as far south as Deland in Volusia County, Florida (Huddlestun, 1988), and ex tends north into Georgia in the subsurface near Jacksonville (Scott, 1992a). Bedding within the unit is massive and appears to be devoid of primary sedimentary or biogenic structures, possessing a maximum estimated thickness of between 40-60 feet (12-18 m), a lthough the type locality is on ly 6-8 feet (1.8-2.4 m) thick
37 (Huddlestun, 1988). Based on stratigraphic position and elevation, the Nashua Formation appears to represent deposition in an ope n-marine, shallow-water, inner neritic continental shelf setting consistent with interpretation as an offshore faci es of the coastal marine Cypresshead Formation (Huddlestun, 1988). Planktonic foraminifera rec overed by Huddlestun (1988) from two Florida Geological Survey core sites (W-8400 and W-13815) are consistent with Z one PL5 of Berggren (1973) or the middle of N21 of Blow (1969), while a suite recovered from near the type locality are consistent with an early Plei stocene (Calabrian) age, equivale nt to Zone N22 of Blow (1969). The reader is referred to Huddles tun (1988) and Scott (1992a) for a more detailed review of the Nashua Formation, including type locality, lithology, and stratigraphic relationships. Caloosahatchee Formation. The Plio-Pleistocene Caloosahatchee Formation (Missimer, 1993; Scott and Wingard, 1995), informally referre d to as the Caloosahatchee Marl (Cooke and Mossom, 1929) or the Bermont Formation (DuBar 1974), consists of fo ssiliferous quartz sand with variable amounts of carbonate matrix interb edded with variably sandy, shelly limestone, some of which has a fresh water origin (Scott, 19 92a). The unit is reported to extend on the west coast of Florida from north of Tampa south to Lee County, and then to extend eastward to the east coast then northward into northern Flor ida (DuBar, 1974; Scott, 1992a). However, Huddlestun (1988) suggests that the Caloos ahatchee Formation is neither a mappable lithostratigraphic unit of formati on rank nor continuous in the s ubsurface. Because the unit has historically been recognized on the basis of fossil content ra ther than lithology, Scott (1992b) now includes it in the informal Okeechobee formation, along with the overlying, faunallyderived Bermont and Ft. Thompson formations. The reader is referred to Scott (1992b) and Missimer (1993) for a more detailed review of the Caloosahatchee Formation, including type locality, lithology, and stra tigraphic relationships.
38 Pleistocene series The Cypresshead Formation is overlain in much of the study area by Pleistocene to Holocene undifferentiated surficial sand of variable origin. Much of this sand is loose, generally structureless and massive, and ranges in color from pale gray to buff to white (Huddlestun, 1988). These sands include those of marine, pedogenic, and windblow n origin that occur at the top of local geologic sections, a nd underlie, or are part of, the local soil profile (White, 1958; Huddlestun, 1988; Scott, 1988a). Trail Ridge sa nds are included with these sediments.
39 CHAPTER 2 REVIEW OF LITERATURE Cypresshead Origin and Age A review of the history of stratigraphic nom enclature applied to Cypresshead Formation sediments in north-central Florida is given by Kane (1984) and for southeastern Georgia by Huddlestun (1988). Some of the many stratigraphic designation s applied to the Cypresshead Formation in Florida include the terms Citronelle Formation first used by Doering (1960), Fort Preston Formation used by Puri and Vernon (1964), and the Grandin Sands, a designation used by Kane (1984). The first of these, Citronelle Formation, evolved from the belief of Cooke (1945), Doering (1960), and Pirkle (1960) that the sediments of peni nsular Florida correlated to the Citronelle Formation described by Matson (1916) as reddish-orange quartz clastics located in Mobile County, Alabama. Otvos (1998b) has suggested including both the Cypresshead and Miccosukee formations in the Citronelle Formati on. However, use of that term in northern and central peninsular Florida implies a direct correlation to southern Alabama and western Florida that has not been clearly dem onstrated. In Georgia, Cypresshead Formation sediments have previously been included with the Okefenokee and Altamaha Formations by Veatch and Stephenson (1911) and in various shoreline complexes, among others (Huddlestun, 1988). Facies Associations Pirkle (1960) was the first to describe C ypresshead Formation lithofacies, basing his conclusions on the physical appe arance of these sediments in Florida. This stratigraphic technique divided the unit into three zones from the surface downward; (1) loose surface sands (now considered to be, in part, Pleistocene cove r of marine, pedogenic, or windblown origin), (2) red and yellow clayey sands, and (3) white clayey sands, also referred to as the kaolin zone. Subsequently, Kane (1984) defined four facies, distinct from those of Pirkle (1960), based on
40 lithologic and biogenic associations observed in nor th-central Florida. In ascending order, these facies are: bivalve and burrowed, burrowed and trough cross-bedded, burrowed and planar crossbedded, and unstructured. This vertical facies progression was interpreted by Kane (1984) to indicate a prograding shoreline, consistent w ith deposition within a coastal or nearshore environment. In southeastern Georgia, Huddlestun (1988) described the Cypresshead Formation as consisting of two gross lithofacies independent of those defined by Kane (1984). The first of these, the updip lithofacies, is described as coarse-grained and pebbly, with the sand-size fraction ranging from fine to coarse with scattered gravel stringe rs. Sorting in this facies ranges from well-sorted to poorly sorted with typically pr ominent bedding ranging from thick to thin in thickness. Crossbedding is also conspicuous in this lithofacies, with the largest scale crossbedding associated with the coarsest and most poorly sorted sands. Ophiomorpha nodosa, a trace fossil, is locally common, and is especially co mmon in the massive, structureless, medium to coarse sands. This lithofacies is interpreted by Huddlestun (1988) to be similar in appearance to the Citronelle Formation in the panhandle of wester n Florida, and is described as being typically developed in updip areas and near large rivers. The second lithofacies of Huddlestun (1988), the downdip lithofacies, consists of finegrained sand and clay. It is ch aracterized by thinly-bedded, fine -grained, well-sorted sand with thin layers, laminae, or partings of clay disper sed through the sand. In some areas, the bulk of the formation consists of massive, argillaceous, fine-grained sand that is devoid of any primary sedimentary or biogenic structur es. In such cases, the sediment is interpreted by Huddlestun (1988) as being completely mixed and homogenized by burrowing organisms. Intermediate lithologies consist of bioturbated, poorly mixed sediments commonly associated with a
41 discontinuous, gray, thinly laye red, silty, diatomaceous clay. This lithofacies is interpreted by Huddlestun (1988) to resemble the Miccosukee Fo rmation of southwestern Georgia and western Florida, and is described as be ing typically developed in downdip areas and between large rivers. Paleoenvironment The environment of deposition for the Cypre sshead Formation in Georgia and Florida has been interpreted as flood-plain (Davis, 1916), coas tal or nearshore marine (Bell, 1924; Martens, 1928; Kane, 1984; Huddlestun, 1988), and alluvial or fluvial-deltaic (B ishop, 1956; Pirkle, 1960; Pirkle et al., 1964). The theory of an alluvial or fluvial-delta ic origin suggests deposition associated with a large delta, with terrestrial se diments in some areas laterally continuous with marine fossiliferous strata (Bishop, 1956). Pirkle (1960) supported the alluvial model based on the intimate mixing of sediments ranging in size from coarse quartzite pebbles to very fine clay, as well as the irregular stratificat ion characterizing these sediments. Observations of vertical and horizontal irregularities in sediment mixtures, ex tensive cross-bedding and cut-and-fill structures were all considered compatible with an a lluvial origin (Pirkle, 1960). However, this interpretation was based, in part, on a lack of marine fossils; an observation proven since to be incorrect (Kane, 1984; Huddlest un, 1988). Also, as noted by Alt ( 1974), the alluvial or fluvialdeltaic model is inconsistent wi th the long, relatively straight geometry of the deposit and its orientation along the crest of the Florida peninsula. Although a coastal or nearshore marine model no w appears most appropriate, it is unclear whether the Cypresshead Formati on was deposited in a large sound or lagoon, partially isolated from the open ocean as suggested by Huddlest un (1988), or whether it was deposited in a nearshore marine setting as part of a barrier island complex as suggested by others (Kane, 1984). Evidence is mixed, as the presence of abundant Ammonia beccarii and Elphidium spp. at the base of the Cypresshead Formation in Effingha m County, Georgia, indi cate brackish water
42 conditions, while the presence of sparse plankton ic foraminiferal assemblages in Georgia and kaolinite molds of pelecypod shell morphologies resembling Mercenaria spp. and Ensis spp. in Florida suggest that near normal salinities must have prevailed in some areas (Kane, 1984; Huddlestun, 1988). Additionally, the pr esence of locally abundant Ophiomorpha spp. trace fossils in both Georgia and Florida (Kane, 1984; Huddlestun, 1988) sugges t that the associated sediments were deposited in shallow water, near to sea-level. Thus, it seems most likely that Cypresshead Formation sediments were deposited in a mix of shallow water marine and coastal environments which appear to have varied in their characteristics in a north-south trend, a condition similar to what is seen along the modern coasts of Georgia and Florida. Age Estimates Age estimates for sediments composing the Cypresshead Formation have varied from Miocene to Pleistocene, with Cooke (1945) the first to assi gn a Pliocene age to the unit. Additionally, Cooke (1945) made the observation at that time that the unit was essentially contemporaneous with other Pliocene deposits in Florida, including the Caloosahatchee Formation, Nashua Formation and Tamiami Form ation, among others, and merely represented a littoral lithofacies of the other units (Matson and Clapp, 1909; Sellards, 1914; Cooke and Mossom, 1929). The best estimate of the maximum age range for the Cypresshead Formation prior to this study, based on stratigraphic positio n, limited internal paleontology, and physical correlation, has been interpreted as late Pliocene (early Piacenzi an) to early Pleistocene (Calabrian) (Huddlestun, 1988), although deposition in Georgia is most lik ely restricted to the late Pliocene (late Piacenzian to late Gelasian). This age range is based, in part, on two microfossil (planktonic and benthic foraminifera) assemblages recovered fr om the Cypresshead Formation in Wayne and Chatham Counties, Georgia, and evidence from a third assemblage recovered from the Nashua
43 Formation in northern Florida. Each of these asse mblages is consistent with a late Pliocene age, but based on the contention of Huddlestun (1988) that the Cypre sshead correlates laterally with the Nashua Formation, timing for potential Cypr esshead Formation deposition was extended into the early Pleistocene (Calabrian). Correlative Siliciclastics Although the Cypresshead Formati on is thought to be the same age as the Citronelle and Miccosukee formations (Scott, 1988a; Otvos, 1998b), and is similar to these units lithologically (Scott, 1988a), the application of the name Cypresshead to the sili ciclastics which are the focus of this study is desirable since the units are not traceable into e ach other (Scott, 1988a). In fact, the Cypresshead in southeastern Georgia and peni nsular Florida is separated from the Citronelle Formation by both the Miccosukee Formation and by a large area where these sediments have either been removed by erosion or where nondeposition occurred. Citronelle Formation Named by Matson (1916) for a town in Mobile County, Alabama, the Citronelle Formation, the most extensive northeastern Gulf of Mexico coastal plain unit, can be traced westward from Alabama across Mississippi and L ouisiana into Texas. Ea stward, the Citronelle can be traced through Florida to the Little River in central Gadsden County, east of the Apalachicola River. Including sediments previously considered part of the Lafayette (Matson, 1916; Cooke, 1945) and Bristol (Sellards, 1918) fo rmations by earlier workers, the Citronelle was extended by Cooke and Mossom (1929) to incl ude the red sand of the lake region of peninsular Florida, from Clay County southw ard to Highlands County. In 1988, these sediments were redefined as Cypresshead Formation (Scott, 1988a). Cooke (1945) initially interpreted the Citronelle Formation as a littoral or near-sho re accumulation of sand and clay brought down by rivers and distributed by waves along the shore of the Gulf. He referred to that portion of the
44 Citronelle (Cypresshead Formation) in the peninsula as likewise a littoral deposit but one formed far from any large river. He further sugg ested that most of the sediment comprising the Cypresshead Formation likely drifted southwar d from Georgia along the Atlantic coast, envisioning the Cypresshead as nearshore and beach deposits of the same sea in which the shell marl of the Caloosahatchee Formation accumulated father out. As with the Cypresshead Formation, the relative paucity of age-diagnostic fo ssils has hampered efforts to constrain the age of the Citronelle Formation. However, recent da ting of the underlying and/or laterally correlative Jackson Bluff Formation and a few enclosed Japane se umbrella pine pollen specimens, as well as constraints of global climate, sea-level hist ory, and surface elevations now provide a Late Pliocene (3.4-2.7 Ma) age for the Citronelle (Otvos, 1998b). Considered to be time equivalent to the C ypresshead Formation (Scott, 1988a; Huddlestun, 1988; Otvos, 1998b), the Citronelle Formation consis ts mostly of angular to subangular, very poorly sorted, fineto very coarse-grained quartz sand intercalated with lenses of gravel and clay (Scott et al., 1980; Scott, 1992a). Co arseto fine-grained, gravelly alluvial facies sands, rarely more than 20-30 m thick, were deposited in majo r and minor stream channels and floodplains, with cyclic interlayering of fineand coarse -grained beds, and indications of channel bank erosion and reworking common (Otvos, 1998a). Pebbles and intraformational mudclasts are common, with pebbles consisting mainly of chert, quartz, flint, and jasp er (Otvos, 1998b). Chert pebbles are known to reach >10 cm in length in inland exposures. Burrowed paralic-nearshore facies sediments are recognized in a semi-contin uous band from Mobile Bay, Alabama, into the western panhandle of Florida (Otvos, 1998a; b) Burrow traces in these sediments include Ophiomorpha sp. ( Ophiomorpha nodosa) and Skolithos sp., believed to be associated with callianassid (ghost shrimp) and pol ychaete worm activity, respectivel y. Internal molds of coastal-
45 nearshore veneride and other shallow water biva lves also have been identified at several locations (Otvos, 1998b). Intense post-depositiona l alteration of the Citr onelle Formation has resulted in iron and silica mob ilization, producing the typical bright orange-red, reddish, and pale orangeto yellowish-brown coloration associated with the unit, as well as the tripolitization of chert pebbles (Otvos, 1998b). The re ader is referred to Otvos (1998b) for a more detailed review of the Citronelle Formation, in cluding type locality, lithology, and stratigraphic relationships. Miccosukee The Citronelle Formation grades to the east, through a broad facies transition, into the time equivalent Miccosukee Formation of Hendry and Yon (1967). This fo rmation includes all clastic sediments of the Tallahassee Hills that occu r above the Miocene Hawthorn Group but below the Pleistocene sands in the Northern Highlands region of the central panhandle of Florida (Scott et al., 1980). Considered a time equivalent of the Cypresshead Formation (Huddlestun, 1988; Otvos, 1998b), sediments of the Miccosukee Form ation occur in the eastern panhandle of Florida, extending east from the Little River in central Gadsden Count y to eastern Madison County (Rupert, 1990; Scott, 1992a). Various au thors (Otvos, 1998b; and others) have argued that the Miccosukee and Cypresshead Formati ons were once continuous, only to have the connection eroded during terrace construc tion west of the Okefenokee Swamp. Lithologically, the Miccosukee is dominated by well-sorted, fineto medium-grained sand with scattered layers or lamin ae of white to gray clay (Huddl estun, 1988; Scott, 1992a). More rarely, very fineand coarse-g rained sand beds occur in the unit, as do poorly-sorted, pebbly, cross-bedded coarse-grained sands and clays (H uddlestun, 1988). Limon ite pebbles also are common in the unit (Scott, 1992a). Where mode rately to deeply weathered, Miccosukee Formation sands are typically mottled reddish-orange to reddish-brown in color, with associated
46 clay layers white (Huddlestun, 1988; Otvos, 1998b). This produces a characteristic appearance analogous to the downdip lithofac ies of Huddlestun (1988) for the Cypresshead Formation. Huddlestun (1988) has suggested a coasta l marine, possibly bay-sound, depositional environment for the Miccosukee Formation based on the scattered occurrence of burrows ( Ophiomorpha nodosa), bioturbated sediments, and tidal channel scour-and-fill structures. However, others have suggested a prodeltaic origin for the Miccosukee based on cross-bedding (Yon, 1966; Scott, 2001). In summary, the Miccosukee appears to represent a down-dip facies of the Citronelle Formation as suggested by Otvos (1 988b), and shares depositional affinity with the down-dip facies of the Cypresshead Formation in Georgia as defined by Huddlestun (1988). The reader is referred to Huddlestun (1988) for a more detailed revi ew of the Miccosukee Formation, including type localit y, lithology, and stratigraphic relationships. Kaolin Origin and Provenance Georgia-South Carolina Kaolin District As a proxy for potential models of kaolin or igin in the Cypresshead Formation, GeorgiaSouth Carolina kaolin district deposits are th e most likely correlative. Disagreement over the origin of the Cretaceous "soft" kaolins and the Tertiary "hard" kaolins has centered on two major problems; (1) the locatio n of kaolinization, and (2) the factors responsib le for the differences between the two kaolin types (D ombrowski, 1992). Reviews of th e detailed differences between the two types of kaolin are given by Patters on and Murray (1984) and Pickering and Hurst (1989). Disagreement as to the origin of the co arse-grained (up to 65% coarser than 2 m) "soft" kaolins has centered on whether the kaolinite formed in situ through the alteration of crystalline rocks in the southern Piedmont-Blue Ridge regi on followed by subsequent transportation to the Cretaceous shoreline, or whether the kaolinite fo rmed by the alteration of arkosic sands at the
47 Cretaceous shoreline. Keller (1977) was one of the first to conclude that both mechanisms played a role in the origin of the "soft" kaolins, comb ining data on the geologic history of the area, the morphology of the kaolinite developed from arkosic sediments, and the stability of kaolinite crystals undergoing transportation. Recent research concerning sealevel fluctuations during the Cretaceous to Early Eocene have further substantiated the idea that multiple mechanisms were responsible for the formation of th ese deposits (Dombrowski, 1992; 1993). Pickering and Hurst (1989) summarized a theory for the origin of the "soft" kaolins, noting as follows: "The initial source material was aluminous detr itus from weathered feldspathic rocks in the adjacent Piedmont Upland. The detritus cons isted mainly of kaolinite, quartz, and metahalloysite. Lesser constituents were mica, feldspar, smectite, ferric pigments, and anatase." They further indicate that these sediments, exposed during regressi ons, would have been subjected to intense weathering and some erosio n and redeposition. They conclude their theory, stating: "The Cretaceous kaolins consist partly of remn ants of kaolinite-rich zones of weathering profiles overprinting kaolini tic sediments, but they consist mainly of eroded and redeposited materials derived from such profiles." Such a model is consistent with kaolin microt extures assigned by Picker ing and Hurst (1989) and Pickering et al. (1997) to post-de positional leaching, oxidation, diag enesis, and authigenesis that have subsequently impacted these deposits. The origin of fine-grained (> 80% finer than 2 m) "hard" kaolins became a focus of study after that of the Cretaceous deposits. Stull and Boles (1926) and Smith (1929) were some of the first researchers to investigate the differences between the two types of kaolin. Their theories suggested that variations in de position and diagenesis accounted for the observed differences in the deposits. Later work by Kesler (1956) suggested that the water salinit y in which the kaolins
48 were deposited controlled many of the differences between the "hard" and "soft" kaolins. This observation was supported by Hinckley (1961), who observed face-toface grain orientation in "hard" kaolins and face-to-edge grain orientation in "soft" kaolins. He believed this was evidence to support the idea that "hard" ka olins were deposited in saline water, while "soft" kaolins were deposited in fresh water. However, as noted by Pi ckering and Hurst (1989), the latter observation relating to soft kao lins is a function of in situ recrystallization rather than original sedimentation. Subsequent microtextural observations of Huber Formation hard kaolins support the observation of Pickerin g and Hurst (1989), which states: "In these kaolins, the fine particle size, good sorting and face-to-f ace association of the kaolinite platelets, as well as the presence of hystrichospherids and di noflagellates indicate that the hard kaolins are a marine sediment." Furthermore, the preservation of such microtex tures indicates that the younger hard kaolins have undergone far less post-depositional a lteration than older soft kaolins. Although Georgia-South Carolina district kaol ins have commonly been referred to as sedimentary in origin (Ladd, 1898; Veatch, 1909; Kesler, 1956; Murray and Keller, 1993), Hurst and Pickering (1997) and Pickering et al. (1997) argue against describing these deposits as such. Rather, they suggest the term Coastal Plai n, to account for the hydrogeologic controls and many post-depositional processes ultimately responsible for kao lin formation. As for regional hydrogeology, they note that GeorgiaSouth Carolina kaolins are restricted to within the recharge area of the regional groundwater system, and ar e variably impacted by oxic and/or dysoxic weathering reactions as a function of saturated ve rses unsaturated conditions. As many as twelve distinct post-depositional proces ses have been identified by Hurst and Pickering (1997) as impacting the formation of Georgia kaolins under these variable groundwater conditions, including the following:
49 Bacterially mediated stripping of Fe from or ganic matter, kaolinite-metahalloysite, illite, and other minerals by HS-, with sequestration of Fe as sulfide (pyrite) under dysoxic conditions Destruction or organic matter via bacterially mediated pyrite formation and oxic weathering Recrystallization of kaolin ite-metahalloysite and coarse ning of kaolinite under oxic weathering conditions (Ostwa ld ripening and/or open-sy stem recrystallization) In situ weathering/kaolinization of feldspar, mica illite, and smectite (montmorillonite) to kaolinite and/or halloysite, particularly under oxic leaching conditions Removal through leaching of alkalis, al kaline earths, silica, Fe, and Mn Diagenetic changes under long-maintained dysoxic conditions, in cluding pyrite and kaolinite coarsening Fe oxidation under oxic weathe ring conditions, including bacterially mediated breakdown of pyrite under oxic vadose conditions and oxidation of octahedral Fe in kaolinitemetahalloysite These mechanisms, along with other weathering a nd diagenetic processes outlined by Hurst and Pickering (1997) and Pickering et al. (1997), are believed to be responsible for transforming high-alumina, kaolinite(and metaha lloysite) bearing sedimentary de tritus into kaolin deposits of commercial quality. Early research on the provenance of the "sof t" and "hard" kaolin s suggested that the Tertiary kaolins were derived from the fine fraction of the Cretaceous kaolins by reworking, as evidenced by rounded kaolin balls in Tertia ry sediments (Smith, 1929; Murray, 1976). This concept was first challenged by Hassanipak and Eslinger (1985), who stud ied the crystallinity and oxygen isotopes in different size -fractions of the two kaolin type s. They concluded from this data that the Tertiary kaolins were not the same as the fine fraction of the Cretaceous kaolins, but failed to elaborate on the causes of the differenc es. It was not until the work of Dombrowski (1982; 1992; 1993) and Dombrowski and Murray (1984) that the differences in the two kaolins was attributed, in part, to differences in source provenance. These studies used trace element
50 geochemistry (La, Th, Co, Sc) to show that the signature of the Cretaceo us "soft" kaolins is identical to kaolinite derived from local granite and gneiss, while the signature of the Tertiary "hard" kaolins from eastern Georgia and South Carolina are consistent with a mixture of approximately 70% metavolcanic rocks (Little River Group) and 30% granite/gneiss source rocks. Trace element analyses of Tertiary "hard" kaolins from central Geor gia indicate that these deposits are derived from pre dominately metavolcanic source rocks with distinct zones dominated by detritus from granite/gneiss sources. The differences in sources appear to be contro lled by the availability of material exposed to alteration and erosion during different time pe riods (Dombrowski, 1992). During the Cretaceous, high sea-level stands would have submerged th e metavolcanic rocks of the Little River Group, leaving exposed granite and gnei ss as sources for kaolinitic sedi ment. During the Early Tertiary, the metavolcanic rocks were exposed to kaoliniza tion processes, and contributed substantially to the accumulation of Tertiary "hard" kaolins. Cypresshead Formation Kaolinitic clays present in the Cypresshead Formation occur as irregular thin beds or lenses and stringers of clay, or as a binding ma trix for sands and gravels, some of which are commonly cross-bedded. Clay content may vary from absent to >50 percent in sandy clay lithologies, although the average content of clay-rich lithologies is 10-20 percent (Pirkle, 1960; Kane, 1984; Huddlestun, 1988; Scott, 1988a). Armored clay balls and clay rip-up clasts roughly 1-5 cm in diameter are also common (Pirkle, 1960; Kane, 1984), with Pirkle (1960) assigning their origin to the current fragmentation of desiccated kaolinite stringers and lenses following intervals of subaerial exposure. Observations by Pirkle (1960) of the flocculation and settling capacity of these clays during industrial processing suggest that bot h clay balls and rip-up clasts could have formed under subaqueous conditions as well, without the need for subaerial exposure.
51 Concentration of these intraforma tional clasts in the upper porti on of the Cypresshead has been suggested by Kane (1984). The lone commercially exploited Cypresshead Formation kaolin deposit, located at the Edgar Minerals EPK facility in north-central Florida, averages approximately 15-18% 2 m (pers. com., Edgar Minerals), and commonly ha s been referred to as a kaolinitic sand. Mineralogy of the clay fraction has been assumed to be kaol inite, although recent evidence suggests the likelihood of a minor halloysite an d/or metahalloysite component (Fountain and McClellan, 1993). According to Pirk le (1960), the highest percentage s of kaolinite appear to be associated with the lower part of the Cypr esshead Formation in north-central Florida. Prior to this study, the exis ting understanding as to the origin and provenance of Cypresshead Formation kaolinitic sa nds was similar to that of Ge orgia-South Carolina district kaolins during the 1970s and 1980s, with ongoing questions focused on; (1) the location of kaolinization, and (2) the source and sedimentological significance of associated sediments. As is the case with the kaolinitic sedimentary deposits in Georgia, several theo ries on the origin of Cypresshead Formation kaolinite have been proposed, including in situ formation from arkosic sands, weathering of smectitic precursor clays, or direct deposition as a sedimentary kaolin. Arguments supporting the in situ formation of the kaolinitic sediments were first introduced by Sellards (1912), who considered the kaolinite to ha ve formed by the weathering of arkosic sands transported from Piedmont and Bl ue Ridge sources. Bell (1924) was the first to contradict this view, supporting the alternative model of a sedi mentary origin for Cypresshead kaolinite based on th e observation that: "Feldspar would have become completely d ecomposed long before it could have been transported from the crystalline area on the nor th to the sedimentary kaolin region, several hundred miles southward."
52 Building on the observations of Bell (1924), Pirkle (1960) outlined various lines of evidence supporting a sedimentary origi n, including field relationships and sedimentary features incompatible with in situ models. Among these arguments were the lack of undecomposed or partially altered feldspar in the un it, the lack of secondary silica phases, the presence of clay balls and rip-up clasts, and the occurren ce of stringers and lenses of n early pure kaolinite in the unit. Although a sedimentary, or rather detrital, or igin for much of the clay content in Cypresshead Formation sediments is likely, Austin (1998) suggests an important role for groundwater leaching of aluminous components (m ica, feldspar) in the formation of an in situ kaolinite fraction, with the associated removal or preservation of Fe and organic compounds dependent on groundwater acidity, redox, and biological content. Such an observation correlates with those made by Hurst and Pickering (1997 ) for the Georgia-South Carolina kaolins. A comprehensive review of arguments against the in situ theory of Sellards (1912), as well as the possibility for the in situ weathering of smectitic clays as a po tential kaolinite source is given by Pirkle (1960). Provenance has only been addressed in the most general of terms re lative to Cypresshead Formation sediments. Observations have been lim ited to those of Pirkle (1960); Pirkle et al. (1964), and Kane (1984), with both focusing on the be lief that crystalline rocks of the Piedmont and Blue Ridge are the most likel y source of kaolinitic sediments. Heavy minerals suites (Pirkle et al., 1964; Kane, 1984) and quartz grain textures (Kane, 1984) evaluated by these researchers support this contention, suggesti ng a mixed igneous/metamorphic sediment source. To date, no detailed provenance evaluation specific to the ka olinitic clay component of the unit has been attempted.
53 CHAPTER 3 METHODS Sample Localities Samples collected for this study include mine exposure, road/railroad outcrop, and drilling samples from Peninsular Florida and southeaste rn Georgia which were collected in cooperation with the Florida Geological Survey (FGS) and the following mining companies: Vulcan Materials Company (formerly Florida Rock Indus tries), Edgar Minerals (formerly Feldspar Corporation), CEMEX (formerly Standard Sand & Silica/Rinker Materials) and E.I. du Pont de Nemours & Company (DuPont). Sampling locations are illustrated in Figure 3-1, with a complete sample list and corresponding analys es performed for this study in Table 3-1. Cypresshead Formation (Florida) Sampling sites were selected along the length of the Lake Wales Ridge and its northern extension, the Interlachen Ka rstic Highlands, to permit th e thorough description of the Cypresshead Formation in Florida. Both outcr op and drill core/auger samples acquired from mine sites in Florida were used. Sampling sites in northern Florida includ ed the Edgar Minerals EPK Mine located off CR 20A near Edgar, Fl orida, the Vulcan Materials Company (VMC, formerly Florida Rock Industries) Grandin Sand Mine located off SR 100 between Grandin and Putnam Hall, Florida, the Vulcan Materials Co mpany (VMC, formerly Florida Rock Industries) Goldhead Mine located off SR 21 north of Keys tone Heights, Florida, and the DuPont Trail Ridge, Highland and Maxville heavy mineral mining areas located northeast of Starke, Florida (TRF2214, WEX164, WEX366 and MCB109). Sampling si tes along the southern portion of the ridge include the Joshua and Davenport mines operated by CEMEX. These mines are located near Haines City and Davenport, Florida, resp ectively. Additionally, th e Paran Church site located across SR 100 from the Grandin Sand Mine and described by Pirk le (1960) and Kane
54 Figure 3-1. Sample locations ev aluated in this study (modified after Huddlestun, 1988; Scott et al., 2001).
55Table 3-1. Sample list and correspondi ng analyses performed for this study. Hydrometer/ Sieve SediGraphSEMPLM/ RLMOrientedRandomOrientedRandom Major/Trace/ REEs Nd Isotopes Cypresshead Formation FL EPK Mine EPK36-J-1225 27 35 40 44 46 48 50 53 56 59 EPK31-P-4027 35 45 50 62 EPK30-V-616 22 24 30 35 39 43 48 53 58 63 68 73 EPK Vermiforms EPK Mica EPK Feldspar Grandin MineFRG-1 1 2 3 4 5 6 7 8 9 Microscopy XRD (< 2 m) XRD (2-30 m/Other) Geochemistry Particle-Size Analysis Sample Location/ Source Sample ID Interval (ID/ft)
56Table 3-1. (continued). Hydrometer/ Sieve SediGraphSEMPLM/ RLMOrientedRandomOrientedRandom Major/Trace/ REEs Nd Isotopes Grandin MineFRG-1 10 11 12 13 14 15 FRG-2 1 2 3 4 5 6 7 8 9 10 11 12 13 Goldhead MineFRL-1 1 2 3 4 5 6 7 8 9 Joshua MineSSJ-1 1 2 3 4 5 6 7 8 9 10 11 DuPont TRF2214 60.0.5 WEX164 18.0.0 WEX366 9.0.0 Microscopy XRD (< 2 m) XRD (2-30 m/Other) Geochemistry Sample Location/ Source Sample ID Interval (ID/ft) Particle-Size Analysis
57Table 3-1. (continued). Hydrometer/ Sieve SediGraphSEMPLM/ RLMOrientedRandomOrientedRandom Major/Trace/ REEs Nd Isotopes Reworked Cypresshead Formation FL Davenport MineSSD-1 1 2 3 4 5 6 7 8 9 10 Cypresshead Formation GA Jesup J-1 1 2 3 4 5 6 Linden Bluf f L-1 1 2 3 4 5 6 7 Birds B-1 1 2 3 4 5 Hawthorn Group, Coosawhatchie Formation FL/GA DuPont MCB109 15.0.0 Jesup J-1 BC Huber Formation GA CMS StandardKGa-2 Congo Boone MineECCI-CB Buffalo Creek Formation GA CMS StandardKGa-1 Buffalo-China MineECCI-BC Ennis Avant MineTKC-E A Brooks 93 MineDBK-B93 Microscopy XRD (< 2 m) XRD (2-30 m/Other) Geochemistry Sample Location/ Source Sample ID Interval (ID/ft) Particle-Size Analysis
58 (1984) was investigated to asse ss sedimentary features exposed in the pit for comparison to sampling sites investigated in this study. For more detailed reference, site maps for the mines are included in Appendix A. Cypresshead Formation (Georgia) In order to compare data from Florida sample s with that of the Cypresshead Formation in Georgia, samples were collected from the type section at the type locality designation of Huddlestun (1988) located near the town of Je sup, in Wayne County, Georgia (Fig. 3-1). Two reference localities designated by Huddlestun (1988) also were sample d as a part of this effort. The first, Linden Bluff, is located on the south ba nk of the Altamaha River, east-southeast of the type locality. The second reference locality is a CSX railroad cut along Ebenezer Creek, located at Birds in Effingham County, Georgia. The reader is referred to Huddlestun (1988) for a detailed description of the Cypresshead Formati on type section/locality and the two reference localities sampled for this study. Middle Georgia Kaolin District Kaolin samples from the Middle Georgia Kaolin District were collected from several mine sites in order to compare both mineralogy and ge ochemistry to Cypresshead Formation samples. Selected Cretaceous kaolin samples include th ose from the IMERYS (formerly ECCI) BuffaloChina Mine (ECCI-BC), the IMERYS (formerly Dry Branch Kaolin) Brooks 93 Mine (DBKB93) and the Thiele Kaolin Ennis Avant Mine (TKC-EA). One Tertiary kaolin sample was collected from the Jeffersonville Member of the Huber Formation e xposed in the IMERYS (formerly ECCI) Congo Boone Mine (ECCI-CB) in Hancock County. Additionally, two Clay Mineral Society (CMS) Source Clay standards representative of low defect, Cretaceous (KGa-1) and high defect, Tertiary (KGa-2) Georgia ka olins were evaluated (Note: KGa-1b has now replaced KGa-1 due to exhaustion of the origin al standard). These Source Clay standards are
59 carefully selected from large, commercial, r easonably homogeneous deposits, and have been characterized for common physical and chemical properties to allow for their use as wellcharacterized standards in clay research. The r eader is referred to Volume 49, Number 5 (2001) of Clays and Clay Minerals for a collection of baseline studies addressing the physical and chemical properties of the Clay Mineral Society Source Clays. Sample Preparation Following detailed field desc ription and collection, bulk sa mples from the Cypresshead Formation and the Middle Georgia Kaolin District were split into subsamples for processing on return to the Industrial Mine rals Laboratory at the Depart ment of Geological Sciences, University of Florida. A flow chart outlining sample processing for the various analytical techniques used in this study is shown in Figure 3-2. Samples collected for analysis were initially crushed, as needed, prior to additional pretreatment to facilitate dispersion and avoid flocculation. Samples for hydrometerand sieve-based par ticle-size analysis were processed using standard procedures outlined in ASTM D42263, which use a 4% Na-metaphosphate solution for sample dispersal. For SediGraph, mineralogical and geochemical analyses, sample splits were processed using a modification of the ASTM D4 22-63 procedure, where concentrated, Optimagrade ammonium hydroxide (NH4OH) was added after introduction of the sample to the settling column to adjust the pH of the clay-water solution, thereby facil itating dispersion. This modification of the ASTM D422-63 procedure prev ents contamination of the clay samples by Na-metaphosphate, which is difficult or impossi ble to completely remove from clays with washing, and would have interfered with isotopic and trace element analyses. The clay (< 2 m) and fine silt (2-30 m) size-frac tions were then separated using standard gravity settling and centrifuge techniques (Hathaway, 1956), and washed repeatedly to remove excess NH4OH. Once
60 Figure 3-2. Sample processing and analysis flow chart. washing was complete, samples were freeze-dried, as needed, for mineralogical and geochemical analysis. Analytical Procedures Grain-Size Analysis Grain-size characteristics were evaluated usi ng hydrometer-, sieveand SediGraph-based techniques in order to evaluate general sedimentological parame ters and to evaluate specific weathering-induced characteristics associated with the fine silt/clay distribution within select samples. As both weathering and illuviation were considered to have likely had a significant effect on the details of grainsize distributions, the degree of accuracy afforded by the pipette method of analysis was deemed unnecessary.
61 Hydrometer and sieve analysis Grain-size analyses were performed in acco rdance with ASTM D422-63 Standard Method for Particle-size Analysis using the hydrometer method in accordance with Stokes Law. For this study, a representative sample of approximately 30 g was used for each analysis along with a sample split to determine moisture content. The sample for hydrometer analysis was first soaked in 125 ml of a 4% Na-metaphosphate solution fo r no less than 16 hours. Following that step, the sample was transferred to an ASTM specified high-speed mixer in a stainless steel cup with stationary baffles. Following agitation for one minut e, the sample slurry was transferred to a 1000 ml glass cylinder and brought to volume with deionized (DI) water. Each sample was allowed to stand to check fo r flocculation or thixotropy. Next, the sample was stirred for one minute us ing a brass stirring rod with a perforated disk at the base. The hydrometer was then placed in the cylinder prior to each reading, and measurements were recorded at 1, 2, 4, 8, 15, 30, 60, 120, 300, and 1440 minutes. A reference hydrometer consisting of a hydrometer placed in a 1000 ml cylinder filled with a blank solution of 125 ml Na-metaphosphate and DI water was used as a modification of the ASTM D422-63 procedure. This allows for direct determination of the effects of temperature and atmospheric pressure on the density of the solution. Temperature also was recorded during each hydrometer reading in order to monitor any changes. After the 24 hour (1440 min.) reading, samples were washed over a U.S. Standard 200 mesh sieve to remove clayand silt-size material s. The plus-200 mesh fraction was then dried at 105 C. The dried sand fraction was then sieved over U.S. Standard 20, 40, 60, 80, 100, and 200 mesh sieves using a Ro-Tap shaker. Weights fo r each size fraction were then recorded and the total weight of the plus-200 fraction determined. Subsequently, the relati ve percentages of sand, silt and clay were calculated using standard graphical techniques in association with a plot of the
62 grain-size distribution of each sample. Additionall y, moment statistics were calculated for each set of results in accordance with the met hod outlined by Balsillie (1995). In combining hydrometerand sieve-based grainsize data, datasets were normalized as needed. As noted by Coakley and Syvitski (1991), there is a noted breakdown in Stokes Law near the sand-silt size boundary at 63 m, which often corresponds to a dip in the size freque ncy distribution of a merged sample. This observation is commonly cau sed by the distributions of the two techniques not being overlapped and renorma lized, but rather being abutted. SediGraph analysis Grain-size distribution test procedures based on gravitational sedimentation are relatively inaccurate for clay particles below 1 m in e.s.d., particularly if the sedimenting slurry is disturbed by insertion and removal of a hydrometer. As a result, supplemental grain-size analyses were performed using a Micromeritics Se diGraph 5100 at the University of Florida in order to more thoroughly characteri ze the fine silt and clay size fraction of select samples. The aim of these analyses was to determine the presence or absence of fine siltand clay-size populations correlated to weatheri ng processes. For this purpose, the SediGraph 5100 uses an Xray/sedimentation method of grain-size analysis th at is accurate and reproducible for grain-sizes ranging from 1-70 m e.s.d., with the lower limit for acceptably accurate results being 1 m as noted by Hendrix and Orr (1972) (the manufacturer reports accuracy to 0.1 m). Below 1-0.5 m, particle settling behavior becomes governed more by Brownian motion rather than Stokes Law. Stein (1985) has noted that montmorilloni te-rich samples are difficult to analyze on the SediGraph due the ability of the thixotropic prop erties of montmorillonite to change viscosity and hinder grain settling. Following standard SediGraph procedures, samples were prepared as sediment suspensions on the order of 0.02-0.1 g/ml in concentration to avoid hindere d settling effects (Stein, 1985;
63 Coakley and Syvitski, 1991). This was accomplis hed by combining 2.2 g of air-dried minus-200 mesh (< 75 m) sample with 70 ml of a 0.05% Na-metaphosphate dispersant solution. This sample concentration was found to be the most su itable for SediGraph analysis following testing of a range of concentrations using sample FRL-1-9 (Fig. 3-3). Followi ng overnight dispersion, samples were agitated ultras onically prior to introduction to the SediGraph 5100 sample reservoir. Using standard test parameters for silicate materials, triplicate analyses were performed for the size range 0.5-64 m, with periodic collection of baseline data on the 0.05% Na-metaphosphate dispersant/rise solution. Subs equently, the resulting mean data were graphically and statistically eval uated, with test precision assessed via analysis of sample EPK36-J-12 (56-59) dup licates (Fig. 3-4). 0.0 0.5 1.0 1.5 2.0 2.5 0.1 1 10 100Particle Diameter (um)Mass Frequency (%) 1.8 g 2.0 g 2.2 g 2.4 g Figure 3-3. SediGraph concentratio n test results of sample FRL1-9 at concentrations of 1.8 g, 2.0 g, 2.2 g and 2.4 g mixed with 70 ml of dispersant solution (curve for each concentration is the aver age of two analyses).
64 0.0 2.0 4.0 6.0 8.0 0.1 1 10 100Particle Diameter (um)Mass Frequency (%) 36-J-12 (56-59) Replicate 1 36-J-12 (56-59) Replicate 2 Figure 3-4. SediGraph precision test results of sample EPK36-J-12 (56-59) comparing two replicate samples (curve for each replicate is the average of three analyses). X-ray Diffraction Analysis X-ray diffraction (XRD) analysis of clay (< 2 m) and fine silt (2-30 m) size-fractions separated by standard gravity settling and centrifugation tech niques was accomplished using a Philips APD 3600/XRG3100 diffractometer located at the Department of Geological Sciences, University of Florida. The diffractometer was equipped with a monochrometer and scintillation detector, and employing CuK radiation at 45 KV and 30 mA for all analyses. Software used to process the resulting data was Jade 3.1+ published by Materials Data, Inc. As accurate peak shape definition was essential for the extraction of information about crystallite size distributions and structural order from the XRD data, a fixed-slit goniometer conf iguration employing a 2 divergence slit and a 0.5 receiving slit was used. The use of a wide receiving slit aperture (0.5 ) has been shown by Madsen and Hill (1988) to provide increased peak intensities, slightly more
65 Gaussian peak shapes, and slightly wider peaks. Moreover, divergence slits 2 have been deemed satisfactory for pattern ev aluation purposes, while slits > 2 have been shown to generate serious deficiencies in Voigt and pseudo-Voigt peak shape models for low angle peaks (Madsen and Hill, 1988). Step size and scanning speed also have been shown to have an influence on peak shape (Wang, 1994), with the parameters of 0.02 2 and 1 sec respectively, chosen for this study in order to minimize these effect s and still perm it collection of XRD data in a timely fashion. Data was collected over a range of 4-65 2 and 4-30 2 for random and oriented samples, respectively, with external XR D reference standards (Arkansas novaculite quartz plate and SRM 675) periodically analyzed to correct for instrument drift. Oriented samples Oriented samples of the < 2 m size fraction were prepar ed and analyzed using the Millipore filter transfer met hod (Drever, 1973; Moore and Reynolds 1997). Wet sample splits of the < 2 m mineralogy/geochemistry concentrate outline d in Figure 3-2 were separated prior to freeze-drying the remainder, and resuspended/dispersed using approximately 15 ml of 4% Nametaphosphate solution to 5 ml of the ~70% soli ds concentrate in a 50 ml centrifuge tube. Samples were allowed to soak overnight prior to further disaggregati on/dispersion by a Model 450 Branson Sonifier, and were then agitated and an aliquot of the disp ersed suspension, ranging from 10 to 20 ml was transferred to a 0.45 m Millipore filter under vacuum. The sample was then filtered rapidly, agitating as necessary, in order to prevent size segregation and keep the sample homogenous as the flow rate slows during filtration. After collection as a filter cak e on the Millipore filter, each sample was then Mg saturated using a 20 ml aliquot of 0.1 M MgCl2 solution as outlined by Moore and Reynolds (1997) and rinsed with DI water to remove any excess solubl e salts, and the filter and clay removed from the apparatus. The filter then was inverted onto a clean glass slide, briefly dried, and the filter peeled
66 off following a modification of the Drever (1973) method similar to that outlined by Moore and Reynolds (1997). Drever (1973) rolle d the clay cake onto the glass slide using a glass cylinder. However, Moore and Reynolds (1997) note that this can cause a non-Gaussian particle orientation, which, in turn, produ ces an unknowable Lorentz factor. Samples were analyzed air-dried, solvated with ethylene glycol, and heat-treated (as necessary). Pretreatment of orie nted clay samples by ethylene gl ycol is useful in identifying expansive phases such as smectites. After XRD an alysis in the air-dried state, samples were placed on a platform in a Pyrex dessicator cont aining 2 cm of ethylene glycol in the bottom, and exposed to the reagent at 60C for a mini mum of 12 hours per the method outlined by Moore and Reynolds (1997). On removal of the sample s from the dessicator, XRD analysis was performed within 1 hour to prevent evaporation of the ethylene glycol. Heat treatments to 105C for a minimum of four hours were employed as needed to define the presence of halloysite in some of the samples examined in this study. On completion of data collection, XRD data were evaluated using the MudMaster computer program of Eberl et al. (1996) to calculate crystallite size distributions. The reader is referred to Eberl et al. (1996) for a detailed review of the MudMaster program and data requirements. Random samples Samples of freeze-dried clay (< 2 m) and fi ne silt (2-30 m) were powdered to minus-200 mesh using an agate mortar and pestle prior to analysis as random clay mineral aggregates. Care was taken to minimize grinding in order to prev ent the creation of grinding induced crystalline defects as illustrated by Kris tf et al. (1993). Random powde r mounts were prepared using aluminum holders and a side-drift ed technique as a means of produc ing results consistent with a randomly oriented sample. On completion of data collection, the qualitative mineralogy of the clay and fine silt size-fractions was determin ed, with kaolinite reflections evaluated for
67 crystalline disorder characteristics using the me thods of Hinkley (1963), Litard (1977) and the expert system of Planon and Zacharie (1990) Quantitative XRD analyses were not performed due to the mineral suite consistency no ted for most of the samples studied. Petrographic/Scanning Electron Microscope Analysis Standard petrographic and scanning electron mi croscope (SEM) analysis of kaolinitic sediments were performed at the University of Florida using equipment made available at the Department of Geological Scien ces and the Major Analytical In strumentation Center (MAIC). Additionally, sample preparation facilities at th e Department of Soil and Water Science were also employed. Reflected light microscopy (RLM) photomicrographs were taken with an Olympus binocular microscope equippe d with a digital camera attachment. Thin sections for petrographic analysis were prepared using standard epoxy impregnation techniques (Araldite epoxy resi n), employing a 2% Automate Blue 8HF dye to indicate sample porosity. Once impregnated chips were prepared, they were sent to Spectrum Petrographics, Inc., in Vancouver, Washington for final preparation. However, rather than finish ing the thin sections to a 30 m thickness, sections used in this study we re finished to 50 m in order to more readily delineate kaolinite. The thin sections were then analyzed via polarized light microscopy (PLM) using a Nikon Eclipse E600W PO L microscope equipped with a Nikon Digital Still Camera DXM1200. Samples taken for SEM analysis were embedded in a colloidal silver paste on the surface of an aluminum SEM stub, then prepared using st andard coating techniques (C and/or Au-Pd), with analysis performed at the Major Analytical Instrumentation Center (MAIC) using a JEOL JSM-6400 scanning electron microsc ope equipped with an EDS system with a thin window solid state detector.
68 Geochemical Analysis Freeze-dried clay fraction (<2 m) samples for geochemical analysis were powdered to minus-200 mesh using an agate mortar and pestle. In between each sample, the equipment was thoroughly cleaned. Agate was selected as the grinding medium in order to minimize the potential for trace element contamination of the samples. Major and trace element analysis Major and trace element analyses were perf ormed by XRAL Laboratories, in Toronto, Ontario, Canada. Powdered clay-size fraction (< 2 m) separates were analyzed using an ICPAES method (ICP40) developed for the USGS along with a complimentary ICP-MS (MS95) analyses as outlined by Methods A and B, Appe ndix B. Samples for ICP-AES were decomposed using a multi-acid total digestion procedure at low temperature as outlined by Crock et al. (1983). Digested samples were then aspirated in to the ICP-AES discharge of a Perkin Elmer Optima 3000 ICP-AES where the elemental emissi on signal was measured simultaneously for 40 elements. Calibration was performed by standardizi ng with digested rock reference materials and a series of multi-element soluti on standards (Lichte et al., 1987). Data was deemed acceptable if recovery for all 40 elements was 15 % at five times the Lower Limit of Determination (LOD) and the calculated Relative Standard Deviation (RSD) of duplicate sample s is no greater than 15 %. CRM SO-3 was analyzed by ICP-AES and IC P-MS as a means of evaluating laboratory accuracy. Nd isotopic analysis Clay fraction (<2 m) separates were analyzed for neodymium (Nd) isotopic values using a Nu-Plasma multiple-collector magnetic-secto r inductively coupled mass spectrometer (MCICP-MS) located at the Department of Geological Sciences, University of Florida. All samples were prepared in a class 1000 clean lab, equi pped with class 10 laminar flow hoods, using
69 standard cation exchange column chemistry as outlined by Method C, Appendix B, then diluted to approximately 250 ppb in Optima-grade 2% HNO3. Both samples and standard solutions were then aspirated into the plasma source via a Micromist nebulizer with GE cinnabar spray chamber. The instrument settings were carefully tuned to maximize the signal intensities on a daily basis. Preamplifier gain calibration was performed before each analytical session. Nd isotope measurements were conducted for 60 ratios in static mode acquiring simultaneously for 142Nd on low-2, 143Nd on low-1, 143Nd on Axial, 145Nd on high-1, 146Nd on high-2, 147Sm on high-3, 148Nd on high-4, and 150Nd on high-5 Faraday detectors. The measured 144Nd, 148Nd, and 150Nd beams were corrected for isoba ric interference from Sm using 147Sm/144Sm = 4.88, 147Sm/148Sm = 1.33, and 147Sm/150Sm = 2.03. All measured ratios were normalized to 146Nd/144Nd = 0.7219 using an exponential law for mass-bias corre ction (ONions et al ., 1977; Belshaw et al., 1998). 143Nd/144Nd values are cited in Nd notation as parts in 104 deviation from CHUR (chondritic uniform reservoi r) defined by present-day 143Nd/144Nd = 0.512638 and 147Sm/144Nd = 0.1967 (Jacobsen and Wasserberg, 1980). The mean value of 143Nd/144Nd for the Ames Nd in-house standard based on 23 repeat analyses during the time that study samples were measured is 0.512140 (2 = 0.000011), corresponding to reproducib ility of better than 0.3units at 2 levels (95% confidence) (Table 3-2). Three repeat analyses each of the JNdi-1 and LaJolla Nd standards during the same time interval produced mean values of 0.512106 (2 = 0.000013) and 0.511856 (2 = 0.000013), respectively (Table 3-3). Three samples of US GS SRM BCR-1 (Columbia River basalt) were prepared via Method C, Appendix B, and analyzed for Nd isotopes together with the samples in order to further evaluate the an alytical protocol. As shown in Table 3-3, the mean value of 143Nd/144Nd for the analyses was 0.512645 (2 = 0.000011), which is indistinguishable from the
70 Table 3-2. MC-ICP-MS analyses of the Ames Nd in-house standard. Date Nd (ppb)143Nd/144Nd Error x 10-6 142Nd/144Nd145Nd/144Nd148Nd/144Nd150Nd/144Nd146Nd/144Ndraw Nd MC-ICPMS 06/02/03~2500.51214551.1417460.3484240.2415610.2363900.738502-9.6 06/02/03~2500.51215051.1417710.3484110.2415480.2363730.738353-9.5 06/30/03~2500.51213261.1417610.3484100.2415570.2364110.740981-9.9 06/30/03~2500.51214251.1417750.3484090.2415500.2363970.740617-9.7 06/30/03~2500.51214861.1417630.3484030.2415440.2363870.740618-9.6 07/01/03~2500.51214541.1417850.3484130.2415430.2363800.739758-9.6 07/01/03~2500.51213951.1417470.3484120.2415480.2363890.739771-9.7 07/01/03~2500.51214461.1417810.3484120.2415390.2363710.739859-9.6 07/01/03~2500.51214541.1417820.3484180.2415540.2363750.739979-9.6 07/01/03~2500.51214641.1417690.3484160.2415470.2363740.739947-9.6 07/02/03~2500.51214561.1417580.3486530.2415510.2363860.739881-9.6 07/02/03~2500.51214061.1417790.3484090.2415580.2363860.740251-9.7 07/16/03~2500.51213431.1417270.3484070.2415390.2363760.739788-9.8 07/16/03~2500.51213741.1417150.3484080.2415510.2363820.739614-9.8 07/16/03~2500.51214061.1417660.3484170.2415460.2363600.739190-9.7 07/18/03~2500.51213551.1417510.3484040.2415520.2363920.740067-9.8 07/18/03~2500.51214041.1417640.3484190.2415510.2363700.739495-9.7 08/08/03~2500.51213441.1417600.3484060.2415390.2363940.740858-9.8 08/08/03~2500.51213161.1417430.3484110.2415510.2363970.740721-9.9 08/08/03~2500.51214351.1417700.3484220.2415430.2363830.740813-9.7 09/03/03~2500.51214041.1417460.3484050.2415510.2363880.739989-9.7 09/03/03~2500.51213441.1417470.3484030.2415500.2363850.739873-9.8 09/03/03~2500.51213151.1417510.3484100.2415480.2363790.739799-9.9 Mean 0.512140 1.1417590.3484220.2415490.2363840.739944-9.7 Std. Dev. 5.7E-06 1.8E-055.1E-055.8E-061.1E-056.7E-041.1E-01 Note: Errors on 143Nd/144Nd measurements are 2 Measured Nd values calculated assuming a present-day 143Nd/144Nd(CHUR) = 0.512638 (Jacobsen and Wasserberg, 1980). published thermal ionization mass spectrometer (T IMS) measurement of 0.51264 (Gladney et al., 1990). Given the agreement between measured a nd published values for these standards, no further corrections were performed. Internal m easurement errors of standards are reported in Table 3-2 and Table 3-3. During the time interval of MC-ICP-MS analyses, a Microma ss Sector 54 TIMS located at the University of Florida was used to collect Nd isotopic values for the Ames Nd in-house standard, following methods described in Scher and Martin (2004). Operated in dynamic mode and equipped with seven Faraday collectors a nd one Daly collector, the mean value of 143Nd/144Nd for 18 repeat analyses run as NdO was 0.512140 (2 = 0.000012), in agreement with MC-ICP-MS results (Table 3-4). Internal measur ement errors for the Ames Nd standard are included in Table 3-4. Advantages of MC-ICPMS over the standard TIMS method include
71 Table 3-3. MC-ICP-MS analyses of the JNdi-1, LaJolla Nd, and BCR-1 standards. Date Nd (ppb)143Nd/144Nd Error x 10-6 142Nd/144Nd145Nd/144Nd148Nd/144Nd150Nd/144Nd146Nd/144Ndraw Nd JNdi-1 0.5121157 -10.2 ~2500.51210261.1417440.3484080.2415540.2363860.739760-10.5 ~2500.51210351.1417440.3484030.2415460.2363880.739633-10.4 ~2500.51211471.1477700.3484060.2415470.2363690.739793-10.2 Mean 0.512106 1.1437530.3484050.2415490.2363810.739729-10.4 Std. Dev. 6.5E-06 3.5E-032.4E-064.3E-061.1E-058.5E-051.3E-01 LaJolla Nd 0.5118587 -15.2 ~2500.51184961.1417780.3484090.2415520.2364130.739819-15.4 ~2500.51185641.1417720.3484170.2415460.2364350.739635-15.3 ~2500.51186251.1418070.3484120.2415460.2364330.739839-15.1 Mean 0.511856 1.1417860.3484130.2415480.2364270.739764-15.3 Std. Dev. 6.6E-06 1.9E-053.8E-063.7E-061.2E-051.1E-041.3E-01 Date Nd (ppm)143Nd/144Nd0Error x 10-6 142Nd/144Nd145Nd/144Nd148Nd/144Nd150Nd/144Nd146Nd/144Ndraw Nd BCR-1 28.800.5126430 0.0 0.51264851.2446300.3484290.2416490.3105360.7403110.2 *0.512649 (2.4)61.3676370.3484380.2416180.2933220.7400960.2 *0.512639 (3.2)61.1867700.3484290.2416660.3152060.7400520.0 Mean 0.512645 1.2663460.3484320.2416440.3063550.7401530.1 Std. Dev. 5.4E-06 9.2E-025.3E-062.4E-051.2E-021.4E-041.1E-01 Note: Errors on 143Nd/144Nd measurements are 2 Measured Nd values calculated assuming a present-day 143Nd/144Nd(CHUR) = 0.512638 (Jacobsen and Wasserberg, 1980). JNdi-1: Nd isotopic reference standard of Tanaka et al. (2000). 143Nd/144NdJNdi-1 = 1.000503 143Nd/144NdLaJolla Nd. LaJolla Nd: Nd isotopic reference standard of Lugmair and Carlson (1978). BCR-1: Columbia River Basalt USGS SRM data source reference of Gladney et al. (1990). Average of duplicate analyses (n=2). Standard deviations for 143Nd/144Nd shown in parenthesis (10-6). Table 3-4. TIMS analyses of the Ames Nd in-house standard. Date NdO (ng)143Nd/144Nd Error x 10-6 142Nd/144Nd145Nd/144Nd148Nd/144Nd150Nd/144Nd162NdO/160NdO Nd 06/04/03750.51212781.1418400.3483820.2415740.2365230.720799-10.0 06/04/03750.51213981.1418690.3483770.2415860.2365090.723820-9.7 06/06/031500.51214481.1418950.3483790.2415690.2365160.720284-9.6 06/09/03750.51214491.1418890.3483890.2415740.2365000.719983-9.6 06/12/03750.51213881.1418660.3483950.2415790.2364840.720113-9.8 06/12/03750.51214981.1418800.3483810.2415580.2364840.720506-9.5 06/13/03750.51214681.1418720.3483970.2415720.2365050.720598-9.6 06/20/03750.51213481.1418690.3483800.2415630.2365100.719821-9.8 06/21/03750.51213691.1418380.3483870.2415660.2365020.719876-9.8 06/21/03750.51213771.1418550.3483760.2415620.2365010.720107-9.8 06/28/031500.51213481.1419140.3483840.2415650.2364880.720333-9.8 06/29/031500.51214681.1418950.3483880.2415760.2364930.720795-9.6 06/30/031500.51213381.1418610.3483810.2415750.2364950.720897-9.8 07/09/03750.51213981.1418810.3483840.2415690.2364920.720537-9.7 07/08/03750.51213991.1418740.3483870.2415860.2365000.723172-9.7 07/07/03750.51213891.1418730.3483840.2415800.2364940.720884-9.8 07/13/03750.51214481.1418970.3483880.2415710.2365010.720763-9.6 07/15/03750.51214971.1419010.3483820.2415590.2364970.720503-9.5 Mean 0.512140 1.1418760.3483840.2415710.2365000.720766-9.7 Std. Dev. 5.9E-06 2.0E-055.7E-068.2E-061.0E-051.1E-031.2E-01 Note: Errors on 143Nd/1 44Nd measurements are 2 Measured Nd values calculated assuming a present-day 143Nd/144Nd(CHUR) = 0.512638 (Jacobsen and Wasserberg, 1980).
72 increased sample throughput without noticea ble loss of precision, and simplification in separation procedures due to the chemical isol ation of Nd from Sm no longer being necessary (Luais et al., 1997).
73 CHAPTER 4 SEDIMENTOLOGICAL AND MINERALOGICAL EVIDENCE FOR THE ORIGIN AND STRATIGRAPHIC SIGNIFICANCE OF THE CYPRESSHEAD FORMATION Introduction The timing and mechanism of Late Miocene th rough Pliocene siliciclastic transport and deposition on the Florida Platform has been a focus of many studies (Cunningham et al., 1998; Guertin et al., 1999; Cunningham et al., 2003), yet remain in dispute, particularly as relates to depositional model inconsistencie s and the significance of this sediment flux in relation to changes in global climate and sea-level. Illustra ting this dispute are th e questions surrounding the Cypresshead Formation of Huddlestun (1988), a s iliciclastic unit in pe ninsular Florida and southeastern Georgia which is characteristically a mottled, fineto coarse-grained, often gravelly, variably clayey quartz sand. The formati on has been variously described as Miocene to Pleistocene in age (Puri and Vernon, 1964; Scott, 1988a), and originating e ither through fluvialdeltaic processes (Bishop, 1956; Pirkle, 1960; Pirk le et al., 1964; Klein et al., 1964; Peacock, 1983; Cunningham et al., 2003) or via longshore current transport and deposition in a nearshore, marine to brackish environment (Bell, 1924; Martens, 1928; Alt, 1974; Winker and Howard, 1977b; Peck et al., 1979; Kane, 1984; Huddlestun, 1988; Scott, 1988a). Even the stratigraphic nomenclature applied to the uni t has varied, with the term coarse clastics first applied by Cooke (1945) shifting through severa l stratigraphic designations pr ior to Huddlestuns work in southeastern Georgia and extens ion of the Cypresshead Formati on name into Florida by Scott (1988a). The Cypresshead Formation and correlative siliciclastics in the Florida panhandle, including the Citronelle Formation of Matson (1916), were once viewed as representing a unique, isolated episode of siliciclastic deposition during the Pliocene (Pirkle; 1960; Pirkle et al., 1964). That view, however, has been erased by the work of Cunningha m et al. (1998; 2001;
74 2003) and others over the past de cade, which indicates that silic iclastic flooding of the Florida peninsula, involving sediments which share an affinity with those of th e Cypresshead Formation, extends from the Late Miocene through at least the end of the Pliocene. Thus, the Cypresshead Formation represents just a compone nt of an apparent 6 to 7 Ma interval of episodic siliciclastic deposition (and reworking), and as a consequence, may be viewed as an indicator of the sedimentary dynamics which drove depositi on throughout this period. Through a renewed evaluation of sedimentological a nd mineralogical indicators, th is study defines both the timing and mechanism of Cypresshead Formation depositi on, and incorporates that understanding into a synthesis of recent studies on re lated siliciclastics of the Flor ida Platform. This approach provides a basis for constraining the regional co rrelation and stratigraphi c significance of the episodic siliciclastic deposition that dominated the Florida peni nsula during this time interval. Related Deposits and the Siliciclastic Conveyor The Cypresshead Formation has long been corr elated with the Pliocene age (3.4 2.7 Ma) sediments of the Citronelle Formation (Otvos, 19 98b). Composed of unconsolidated siliciclastics consistent with an alluvial to nearshore (or shor eface) marine origin, Citronelle sediments can be traced from the central Florida panhandle to as far west as eastern Texas (Otvos, 1998a; b), and exhibit many similarities (e.g., trace fossils, cr oss bedding, etc.) to those of the Cypresshead. However, these two deposits also exhibit dist inct differences in both sediment sourcing and sedimentary architecture which confirm th e stratigraphic delineation accepted by most researchers (Kane, 1984; Scott, 1988a), although deposition of the units appears to have been concurrent. It is the siliciclastics in southern Florida (Table 4-1), and particularly the Late Miocene through Pliocene sequence studied by Cunningha m (1998; 2001; 2003) and others, which offer the greatest insight into the origin, timing and significance of Cypresshead deposition. The oldest
75 Table 4-1. Summary of the lit hostratigraphic and sequence stratig raphic nomenclature applied to South Florida siliciclastics. Study Formation Member/Interval/ Sequence Age/Time (Ma) Environment Scott (1992b) Okeechobee Formationa--Pleistocene --Cunningham et al. (1998) Okeechobee Formationa--Late Pliocene (Gelasian) to Pleistocene --Long Key Formationb--Late Miocene (Messinian) to Pliocene --Guertin et al. (1999) Long Key FormationbIII ~2.0 inner shelf II 4.5-3.5 outer shelf to inner shelf transition I 6.2-5.5 outer shelf to inner shelf transition Missimer (2001a;b) Caloosahatchee Formation --2.14-0.6 shallow shelf/ramp, open bay, lagoonal/embayment Tamiami Formation Pinecrest 3.22-2.15shallow coastal marine Sand Facies 4.29-3.22shallow marine Cunningham et al. (2001)Tamiami FormationUnnamed SandLate Plioceneinner shelf Ochopee Limestone (DS4) late Early Pliocenecarbonate ramp Peace River Formation DS3 5.23-3.83 --DS2 Late Miocene (latest Tortonian and Messinian) --DS1Late Miocene (Tortonian)outer shelf Cunningham et al. (2003)Tamiami Formation SS3 late Early to Late Pliocene --Upper Peace River Formation SS2* 8.6-3.75 outer shelf to inner shelf transition/lagoonal Arcadia/Lower Peace River Formations SS1 24.31-8.6 aggradational, mixed carbonate-s iliciclastic ramp a The Okeechobee Formation is an informal stratigraphic unit introduced by Scott (1992b), and includes the faunally derived Caloosahatchee, Bermont and Ft. Thompson Formations. b The Long Key Formation is a stratigraphic unit proposed by Cunningham et al. (1998). SS2 siliciclastics can be subdivided into eastern (S S2-E: 8.6-5.04 Ma) and western (SS2-W: 5.04-3.75 Ma) components. of these sediments are assigned to the upper Peace River Formation of the Hawthorn Group by Cunningham et al. (2003) and the proposed Long Key Formation of Cunningham et al. (1998), and extend as far south as the Florida Keys, ach ieving a thickness of greater than 145 m. The latter of these units, the Long Key Formation, has been proposed as a potential correlative of the Cypresshead Formation, and perhap s a partial correlat ive of the Tamiami, Caloosahatchee and upper Peace River Formations (Scott and Winga rd, 1995; Cunningham et al., 1998) based, in
76 part, on a late Late Miocene (Messinian) to Late Pliocene (Gelasian) age determination (Cunningham et al., 1998; Guertin et al., 1999). Fu rthermore, the unit appears to represent a shallowing-upward succession corresponding to progradation in response to a substantial siliciclastic flux (Guertin et al., 1999). Lithologically, the L ong Key Formation consists of a single quartz sand facies consisting primarily of moderately to we ll sorted, angular to subrounded, very fine to fine qua rtz sand, with carbonate grains ( 50%), minor feldspar ( 5%), mica, and phosphate. Quartzite pebbles, of ten discoid in shape, are also common. Of the Late Miocene siliciclastics associ ated with the upper Peace River and Long Key Formations, the SS2 sequence (8.6 3.75 Ma) of Cunningham et al. (2003), along with the Interval I siliciclastics (6.2 5.5 Ma) of Guertin et al. (1999), mark the first evidence for the onset of siliciclastic deposition which shares compositional affinity with the Cypresshead Formation. Consisting primarily of quartz sand with varying admixtures of terrigenous mudstone and diatomaceous mudstone, the initial pulse of SS2 deposition is widespread over a broad portion of southern Florida, and correlates with the DS2 siliciclastics from an earlier study by Cunningham et al. (2001). Silicicl astic supply appears to have been restricted during the subsequent early Pliocene (Cunningham et al., 2001), and as a consequence, deposition during this interval is more spatially limited (SS2-W and DS3) or exhibits si gnificant evidence of reworking from earlier deposited siliciclastics (I nterval II) (Guertin et al., 1999; Cunningham et al. 2003). During the remainder of the Pliocene, siliciclastic accumulation continued episodically in southern Florida (SS3/Interval III), but with intervening peri ods of carbonate deposition as characterized by the Ochopee Limestone Member of the Tamiami Formation (DS4) and the lower Caloosahatchee Formation (Guertin et al., 1999; Cunningham et al., 2001; 2003).
77 The relative volume and architectu re of south Florida siliciclastics have been used as the basis for a siliciclastic conveyor model first proposed by Cunningham et al. (2003). This model, based primarily on the occurrence of 50-100 m th ick siliciclastic cli noforms from the SS2 sequence, employs a fluvial-deltaic system as the principle means by which the bulk of siliciclastics were transported and deposited on the southeastern Florida Platform during this interval. Such an interpretation is in agreemen t with earlier held views related to the deposition of Cypresshead Formation sediments (Bishop, 1956; Pirkle, 1960; Pirkle et al., 1964), but these views have since been shown to be inconsis tent with field observations (Kane, 1984; Huddlestun, 1988). Proposed altern atives to the fluvial-deltaic transport model proposed by Cunningham et al. (2003) include channeled depositi on by strong paleocurrents associated with a southward prograding shorelin e (Warzeski et al., 1996) or via longshore transport, potentially as a prograding spit (Winker and Howard, 1977b; War zeski et al., 1996). Identification of the transport mechanism aside, the siliciclastic conveyor model constrains a significant and potentially unique period of deposition on the Florida peninsula, of which the Cypresshead Formation is the most readily studied component. Age Constraints on Cypresshead Deposition Temporal control for the Cypresshead Formati on is poor, resulting from a paucity of fossil material of known age, a lack of radiometrically dateable volcanic rock s, and the lack of a sufficiently long, continuous stratigraphic section suitable to magnetostratigraphic study. However, lithologic correlation can be used to constrain the age of th e unit in both central peninsular Florida and southeaste rn Georgia. Researchers have for some time correlated the unit with the Citronelle and Miccos ukee Formations of the Florid a panhandle (Scott, 1988a; Otvos, 1998b), the Nashua Formation in northern Florid a (Scott, 1988a) and southeastern Georgia (Huddlestun, 1988), and at least in part, with the sediments of the Tamiami Formation (Bishop,
78 1956; Cunningham et al., 1998) and possibly the Caloosahatchee Formation (Scott and Wingard, 1995; Cunningham, 1998) in southern Florida. Recently, Scott (2001) also noted that the Cypresshead Formation appears to occur in the subsurface south of Highlands County, Florida, and suggests correlation, at leas t in part, with the Long Key Formation of Cunningham et al. (1998). Based on these relationships alone, th e timing of Cypresshead deposition can be constrained to the Late Pliocen e, although the possibility for an early Pleistocene component must be considered if the unit is determined to fully correlate with the Nashua Formation, the type section of which is early Pleist ocene (Calabrian) in age (Huddlestun, 1988). Pliocene Paleoclimate The Pliocene represents a transitional period of global cooling following the warmer, low amplitude, low frequency climate fluctuations of the Miocene, and preceding the high amplitude, high frequency climate fluctuations of the more recent Pleistocene. Although representative of a general cooling trend, warm episodes did occur, particularly during the early to mid-Pliocene (4.8-3.2 Ma), when climate is believed to have been warmer than at any other time since the Miocene (Dowsett and Loubere, 19 92; Kennett and Hodell, 1995; Cr owley, 1996; Dowsett et al., 1996; Haywood et al., 2000). During this interv al, enhanced thermohaline circulation (heat transport) has been proposed as the primary cause for warming, rather than increased CO2 concentrations (Crowley, 1996; Haywood et al., 2000), resulting in a general reduction in the latitudinal gradient of both atmospheric and sea-surface temperature (Willard, 1994; Dowsett et al., 1996). Pollen data collected by Willard (1994) show that temp eratures were much warmer than present at high latitudes, yet remained si milar to conditions today at lower latitudes. Following this warm period, steady temperature d ecline is believed to ha ve occurred between 2.8-2.4 Ma (McDougall, 1994), with a significant drop in temper atures at 2.6 Ma corresponding to the onset of major Northern Hemisphere ice sheet formation.
79 As for sea-level during episodes of mid-Plio cene warmth, some climate models predict increases of more than 25 m above present levels (Dowsett and Croni n, 1990; Dowsett et al., 1994). However, Kennett and Hodell (1993; 1995) ha ve suggested that rela tive climate stability in the early to mid-Pliocene would have limited the fluctuation in Antarctic ice sheet volume during this time such that maximum sea-level would have been <25 m above present levels. Alternatively, Naish and Wilson (2009) suggest th at as much as a 30 m rise in sea-level may have been expected during the mid-Pliocene in response to a proposed 30 percent loss of the present-day mass of the East Antarctic Ice Sheet (EAIS) and complete deglaciation of the West Antarctic Ice Sheet (WAIS) and Greenland. Lastly, increased aridity has long been sugge sted for portions of the Late Miocene and Pliocene (Axelrod and Raven, 1985; Jiang et al ., 2005). Evidence for such late Pliocene increased aridity, associated w ith the onset of regression afte r 3.3 Ma, has been recorded for tropical regions of the Southern Hemisphere (Ste in and Sarnthein, 1984). Alt (1974) was the first to support the proposal for increas ed aridity impacting the south eastern United States during the Miocene and subsequent Pliocene, suggesting that arid (or semi-arid) climatic conditions may have influenced deposition of the coarse upland siliciclastics associated with the Citronelle and Cypresshead Formations. As part of this model, Alt (1974) suggested that blanketing siliciclastics were deposited in response to peri odic flooding under arid cond itions similar to that associated with modern central and northern Au stralia. Alt (1974) furt her speculated that the driving mechanism for these conditions was a southward displacement of prevailing wind patterns, bringing the so utheastern United States into the path of the prevailing westerlies, thereby isolating the region from sources of mois ture. However, more recent studies support the concept that continual El Nio conditions prevailed at least during the early Pliocene warm
80 period (~4.5.0 Ma) (Ravelo and Wara, 2004; Fe dorov et al., 2006). Under such conditions, regional climate in the southeastern United States would have been much wetter than previously thought. Results Outcrop and Core Descriptions A total of six sections and five cores totaling approximately 139 m of Cypresshead Formation and reworked Cypresshead Formation sediments were evaluated for this study from the sites shown on Figure 3-1. Detailed cross-secti ons and descriptions of the six exposures from Florida and Georgia, along with two cores collected from central Florida locations are included in Appendix C. Core samples collected from the Edgar Minerals EPK Mine at surface elevations of 27 28 m were not suitabl e for stratigraphic analysis. Of the three sections studied in north-central Fl orida, two are located at the same mine site (Grandin), with the third locat ed approximately 12 km northnorthwest near the town of Keystone Heights. Other than the Paran Church site, located across the road from the Grandin pit, these sites represent the most complete e xposures of the Cypressh ead Formation in northcentral Florida. The three sites evaluated in Geor gia include the type locality at Jesup (J-1) and two reference localities as defined by Huddl estun (1988). Additionally, three Cypresshead Formation samples supplied by E.I. du Pont de Nemours & Company (DuPont) from the Trail Ridge (1) and Highland (2) heavy mineral mining ar eas located northeast of Starke, Florida, were also evaluated for clay mineralogy. Sedimentary Framework Standard sedimentological parameters for th e Cypresshead Formation including grain-size characteristics, the distribution and types of sedimentary struct ures, and information regarding sediment transport mechanisms were determined fo r this study in order to assist in defining a
81 comprehensive depositional model for the unit. Th e data outlined here were supplemented by the previous observations of Kane (1984) and others. Grain-size distributions Cypresshead Formation sediments consist of poor ly to well sorted, very fine to coarse, clayey sands, sandy clays, gravels and clays. Disc oid quartzite pebbles up to 4 cm in diameter and clay rip-up clasts 1 cm to 5 cm in diam eter are common as well. Mean grain-size for Cypresshead sediments ranges from 0.8 to 4.6 (coarse sand to silt), with percent of sand, silt, and clay ranging from 15 to 98, 0 to 41, and 1 to 63, respectively (Fig. 41). Mean grain-size for reworked Cypresshead sediments (Davenpor t Sand Mine: SSD-1) ranges from 1.3 to 3.3 (medium sand to very fine sand), with percent of sand, silt, and clay ranging from 88 to 93, 2 to 7, and 3 to 8, respectively. Summary tables and figures defining percent sand, silt and clay, along with moment statistics and grain-size distribution curves fo r each sample included in this study are included in Appendix D. Additionally, select grain-size distribution curves are shown in Figure 4-2 for discussion purposes. Pervasive weathering has modified the original grain-size distri bution of Cypresshead sediments, preventing detailed environmental interpretation. In particular, processes which include illuviation and the in-situ formation of clays at the ex pense of the coarser mineral fraction (Fountain and McClel lan, 1993; Heuberger, 1995) have modified the depositional signature of the deposit, imparti ng a fines tail to most grain-size distribution curves (Fig. 4-2) which is inconsistent with hydrodynamic sorti ng. Further evidence for a post-depositional increase in the fines content of most samples is illustrated by the high positive skewness values seen with Cypresshead sediments (Fig. 4-3). Not only do coarser sediments (low value) exhibit the highest values, but differences ar e noted between reworked and non-reworked
82 0 2 4 6 8 10 12 14 160.00 0.25 0.50 0.75 1.00 1.25 1.50 1.75 2.00 2.25 2.50 2.75 3.00 3.25 3.50 3.75 4.00 4.25 4.50 4.75 5.00Mean Grain-Size (phi)Frequency Reworked Cypresshead Cypresshead Figure 4-1. Histogram illustrating the distribution of mean grain-size valu es for samples of the Cypresshead Formation and reworked Cypresshead sediments. sediments, with reworked sediments exhibiting an overall reduction in fine s likely in response to winnowing activity during the reworking process. Another feature noted for the gr ain-size distributions are the coarse tails associated with individual sand-size modes within total sample curv es (Fig. 4-2). This feature is consistent with shoreline hydrodynamic sorting generated by swash and/or wave action, and appears to be best developed higher up in the sections evaluated for this study, except at the Birds reference locality (Fig. 4-2F). The coarse tails are also greatest in coarse-grained samples and diminished in finergrained materials, the latter of which are asso ciated with the lower portion of each section. Lastly, both unimodal and bimodal distributions are seen with the grain-size curves. In the case of the unimodal curves, they tend to either correspond to coarse grained sands near the top of the section or fine to very fine sands near the base. Samples with intermediate grain-size characteristics exhibit the greatest degree of bi modality, usually with a dominant coarse or
83 Figure 4-2. Representative grain-size distribution curves for Cypresshead Formation sediments. A) Grandin Sand Mine (FRG-1), B) Goldhead Sand Mine (FRL-1), C) EPK Mine (EPK30-V-6), D) Joshua Sand Mine (SSJ-1), E) Jesup type locality (J-1) and F) Birds reference locality (B-1). medium sand mode paired with a secondary fine sand component. Bimodal distributions such as these are indicative of mixed energy environments during deposition.
84 0.0 2.0 4.0 6.0 8.0 10.0 12.0 14.0 0.00.51.01.52.02.53.03.54.04.55.0Mean Grain-Size (phi)Skewness Cypresshead Reworked Cypresshead Figure 4-3. Scatter plot illust rating the relationship between m ean grain-size and skewness for samples of Cypresshead Formation a nd reworked Cypresshead sediments. Sedimentary structures A variety of environment and process sensitive sedimentary structures are associated with Cypresshead Formation sediments. Included among these are various types of stratification and bedforms, discontinuity surfaces and trace fossils. Stratification and bedforms Several forms of stratificat ion indicative of depositional environment are well preserved in Cypresshead sediments. The firs t of these, cross-stratification, has been described in detail by Kane (1984), and consists of three principle types; tabular, trough and hummocky. Tabular cross-stratifi cation is small to large in scal e, ranging from 5 cm to 3 m in bed thickness, with foreset dips generally in the 20-25 range (Fig. 4-4). Cross-strata are generally symmetrical, with angular to tangen tial lower bounding surface contacts consistent with formation by the migration of straight-c rested sand waves and dunes (megaripples). The smaller of these features (5-60 cm ) are concentrated near the top of the sections where they are
85 Figure 4-4. Example sedimentary structures fr om the Cypresshead Formation. A) Tabular and trough cross-bedding associated with sample interval FRG-1-5, illustrating scour and fill and horizontal bedding surfaces (arrows) alon g with escape structures (circles), B) Tabular cross-bedding associated with sa mple intervals FRG-1-9 through FRG-1-11 containing 0.5-1 m long Ophiomorpha spp. burrows (arrows) inclined in response to current activity, C) Hummocky cross-strati fication from sample interval FRG-1-13, D) Large-scale infilling of a nearshore ch annel at the Grandin Sand Mine, E) Clay lense nondepositional discontinuity (arrow) from sample interval L-1-6 (offset due to slumping along exposure face), F) .Clay bed nondepositional discontinuity (bracket) from sample interval J-1-4.
86 present, but may occur sporadically near the base as well. Larger features (0.5-3 m) were noted only in north-central Florida at the Grandin and Goldhead Sand Mi nes, and are consistent with the migration of dune or megaripple bedforms a ssociated with nearshore bars (Fig. 4-4B). Trough cross-stratification is small to medium in scale, ranging from 5 cm to 60 cm in bed thickness, with foreset dips in the same range as seen with tabular crossstrata (Fig. 4-4). Crossstrata are lenticular and asy mmetrical, and exhibit tangential contacts with er osional lower bounding surfaces consistent with formation by the migration of undulatory and lunate sand waves. Graded bedding within in dividual bed sets is common. As was noted for tabular crossstrata, these features are concentrated near th e top of most sections, with a less common and more sporadic distribution with increased de pth. Scour and fill structures are commonly associated with this type of cross-stratification (Fig. 4-4), suggesting significant flow velocities. Large-scale scoured surfaces with up to 15 m in ve rtical relief, likely representative of nearshore current or tidal channels, are also evident in limited Cypresshead expos ures in north-central Florida (Fig. 4-4D). The last type of cross-stra tification noted here, hummocky, was not previously noted by Kane (1984), but was observed in the basal fine sands of both e xposures studied at the Grandin Sand Mine locality (Fig. 4-4). Thes e cross-strata are medium in sc ale, ranging from 0.3 m to 1 m in bed thickness, with foreset dips and trunca tion angles < 15. As was noted for trough crossstratification, cross-strata fo rm tangential contacts along er osional lower bo unding surfaces. Diagnostic traits which differentiate these beds from other forms of cross-stratification are the antiformal hummocks and synformal swales wh ich are defined by randomly oriented, even lamination (Dott and Bourgeois, 1982). Humm ocky cross-stratification is most commonly associated with redeposition of fine sand below normal fair-weather wave base by large waves.
87 Paleocurrent information derived from the orie ntation of cross-strati fication offers insight into the transportation directi on of Cypresshead siliciclastics Representative measurements collected for this study at the Grandin Sand Mi ne to compliment data accumulated by Kane (1984) for other sites in north-central Florida s upport earlier observations indicating a dominant southerly flow orientation with a secondary bim odal, bipolar flow component generally oriented east-west (Fig 4-5A). Composite measurements made of cross-stratific ation in southeastern Georgia, however, exhibit substantially differe nt results, with the dominant flow orientation generally to the east, and a sec ondary component suggesting a subordi nate flow orientation to the south. Paleocurrent measurements at Linden Bl uff (72, n = 15) exhibit the most pronounced eastern flow characteristics, with measurements made at the Jesup type locality (121, n = 8) more consistent with the secondary flow orientation to the south (Fig. 4-5B). The second form of stratification noted in Cypresshead sediments is horizontal, and sometimes massive, bedding. Where noted, horizontal bedding is more concentrated at the base of exposures, with the Birds reference loca lity exhibiting horizontal bedding throughout the section. This appears to be c onsistent with deposition under lo w flow velocity conditions insufficient to develop ripple or larger bedforms. Graded bedd ing within horizontally bedded strata is also common. Massively bedded Cypresshead strata appear to result from the destruction of original sedimentary fabric through bioturbation, or in so me cases, weathering. These beds tend to be concentrated near the top of exposures where weathering is strongest, or in association with other bioturbated beds which have retained evidence of original stratification. Discontinuity surfaces Two types of stratigraphic di scontinuities are recognized in Cypresshead siliciclastics; nonde positional and erosional. The first of these, nondepositional discontinuities, mark abrupt decreases in se diment accumulation rates and are commonly
88 Figure 4-5. Paleocurent rose di agrams for representa tive Cypresshead Formation exposures. A) FRG-1, B) Georgia composite based on Linde n Bluff (L-1: n = 15) and Jesup type locality (J-1: n = 8) measurements, C) FRG-1 USF facies, D) FRG-1 LSF (pLSF + dLSF) facies. associated with increased concentrations of burrowing activity (e.g. FRG-1-6) or the deposition of discrete clay lenses or beds (Fig. 4-4E a nd F). Erosional discontinuities are more common in Cypresshead sediments, and mark an abrupt in crease in sediment accumulation, grain-size (e.g. graded bedding) and corresponding erosive scouring, either by current s or waves. These features are particularly pronounced at th e base of medium to large-s cale planar cross-stratification associated with dune (or megari pple) bedforms (Fig. 4-4B).
89 Trace fossils Kane (1984) summarized the occurrence of Ophiomorpha spp. trace fossils, bivalve molds and fecal pellets as the sum foss il assemblage associated with the Cypresshead Formation. However, a more careful evaluation of exposures in Florida and southeastern Georgia indicate a slightly more diverse a ssemblage than previously described. Ophiomorpha spp. burrows, normally consistent with Ophiomorpha nodosa, continue to be the most common trace fossil in both Florida and Georgia sections, and are typically 3-4 cm in diameter and can exceed 1 m in length (Figs. 4-4B and 4-6). Specimens retain a characteristic knobby exterior and may exhibit minimal branching oriented at an obl ique angle to bedding. In most instances, Ophiomorpha spp. burrows are vertically oriented a nd non-branching although occasionally burrows may be inclined in response to strong current conditions during coeval deposition and burrowing activity (Fig. 4-6B). Burrows stand out in relief, consisting of clay cemented sands more resistant than the surrounding sediments. Wh ere sediment clay contents are low, escape structures are most common (Fig. 4-6A). Burrows are normally sand filled with rare instances of kaolinite replaced fecal remains preserved in the base. In highly bioturbated units, bedding structures are lacking due to high concentrations of burrows (Fig. 4-6A and B). Along the modern Florida and Georgia coastlines, Callianassa major burrows are considered to be a modern analog, and are accepted as shoreline indicators. Although they primarily occur in the sandy, open marine littoral to shallow neritic enviro nment, they have been reported on protected beaches, sandy tidal flats, and shoals (Frey, 1970) tidal deltas (Warme, 1971), and offshore bars (Weimer and Hoyt, 1964). Thallasinoides spp. burrows observed in north-cen tral Florida basal Cypresshead sediments (FRG-1, FRG-2 and FRL-1) are of ten similar in size and appearance to Ophiomorpha spp. burrows, but occur in units c onsisting of fine sand with greater clay contents, have greater
90 Figure 4-6. Examples of trace fossil and biva lve mold occurrences. A) Densely bioturbated sample interval FRG-1-6, B) Close-up of interval FRG-1-6 from (A) showing the morphology of Ophiomorpha spp. (left arrow) and Skolithos spp. (right arrow) traces, C) Thallasinoides spp. burrows from the Paran Church site exhibiting secondary burrowing of shaft walls and burrow interiors, D) Fossil bivalve molds from the Paran Church site. developed horizontal branching, a nd lack a knobby exterior (Fig. 46C). Differences in structure likely reflect differences in the substr ate in which the burrow was constructed. Thallasinoides spp. burrows examined at the Paran Church si te commonly exhibit secondary boring of shaft walls and backfilled burrow interiors (Fig. 46C), producing a trace which is similar in appearance to the Skolithos spp. traces observed in the same exposure interval from the nearby Grandin Sand Mine. Skolithos spp. traces are commonl y associated with both Ophiomorpha spp. and Thallasinoides spp. burrows. These traces have been described by Chamberlain (1978) as any
91 simple, even width vertical tube varying in di ameter from 2mm to 10 mm, with walls which are usually smooth, but may be segmented or striat ed. In the study area, these tubes range in diameter from 2mm to 5 mm and possess burrow wa lls formed from agglutinated sand grains. Polychaete (annelid) worms are likely responsible for these structures, and are indicative of a marginal marine facies (Seilacher, 1967). Along the Georgia Sea Isles co ast, analogous tubes associated with the polychaete species Onuphis microcephala often occur in association with Callianassa major burrows (Curran, 1985). Environmentally, Skolithos spp. traces appear consistent with burrowing activity during periods of quiescent to highly reduced sedimentation. Additionally, the low biodiversity exhibited by this trace fossil assemblage is consistent with a high stress environment associated with ep isodic high sedimentati on rates dictated by fluctuating, but often high, current flow velocities. The last fossils of note which have been described in detail by Kane (1984) are clay (kaolin) molds of marine bivalves similar in morphology to the modern surf clam Mercenaria spp. and the razor clam Ensis spp. Not seen in sections eval uated for this study, but readily visible in horizontal exposures at the north-centr al Florida Paran Church site (Fig. 4-6D), molds represent disarticulated valves which appear to have been transported by current and/or tidal activity from a proximal estuarine source. Preservation is poor, with stratigraphic positioning of the fossils in basal Cypresshead sediments potentially favored by the relatively high clay content of the fine sands. Mineralogy and Petrography Details of the mineralogy and petrography of both the sand and clay size-fractions were evaluated for this study for corr elation to previous studies on the Cypresshead Formation and comparison to complimentary data on south Florida siliciclastics. X-ray diffraction (XRD) data for the samples (oriented and random) are includ ed with Appendices E and F. A comprehensive
92 evaluation of the clay mineralogy of Cypresshead Formation sediments and the implications for weathering and recrystalli zation processes is addressed in the Chapter 5. Sand size-fraction Evaluation of the sand size-fraction confirms pr evious observations related to the mineral suite, indicating quartz as the dominant lithic component. Coarse to fine quartz grains are angular to subrounded in shape, consistent with pa rticle morphologies noted for Long Key and SS2 siliciclastics (Cunningham et al., 1998; 2003), with angularity decreasing with increased grainsize as a consequence of transp ortation mechanism (i.e. suspensi on vs. saltation or traction). Accessory components of the sand size-fraction incl ude feldspar, mica, kaolinite vermiforms and heavy minerals (Fig. 4-7), the latter of which has been discussed in deta il by Pirkle et al. (1964). Feldspars, confirmed as K-feldspar (microcline) via energy dispersive X-ray spectroscopy (EDS), petrographic examination, and XRD, are f ound at the base of all southeastern Georgia sampling locations (J-1-6, L-1-5, L-1-7 and B-1-5) and only in lower sections of the EPK Mine and Joshua Sand Mine cores collected in Flor ida. Since feldspars grains would correlate depositionally with quartz grains of similar size, feldspar occurrences are related to preservation as a function of weathering rather than hydrodyna mic sorting. At the north-central Florida EPK Mine site, feldspars are preserved in all cores approximately 16 m below the surface elevation of 27-28 m (11-12 m msl), while at the central Flor ida Joshua Sand Mine preservation is seen approximately 11.5 m below the surface elevatio n of 40 m (28.5 m msl). For the Jesup, Linden Bluff, and Birds locations, feldspars are pres erved at 23.7 m msl, 21.7 m msl and 15.2 m msl, respectively. Feldspar concentrations, where s een, range from 1% to 5% of the sand sizefraction. The distribution of mica, confirmed as muscovite via EDS, petrographic examination, and XRD, appears related to hydrodynamic sorting, as it is notable as a trace or minor component in
93 Figure 4-7. Photomicrographs of accessory sand-sized phases from the Cypresshead Formation. A) PLM image of K-feldspar (microcline) grain with twinning from SSJ-1-8, B) RLM image of K-feldspar concentrate from EPK Mine cores, C) PLM image of mica (muscovite) grains (arrows) from FRG-1-8, with the lower grain exhibiting expansion (exfoliation) in response to kaolinization, D) RLM image of mica concentrate from EPK Mine cores, E) SEM image of kao linite vermiforms from FRL-1-4, F) RLM image of large vermiform (book) con centrate of from EPK Mine cores.
94 most Cypresshead Formation sediments. The excep tions to this observati on are highly scoured and/or cross-bedded units within the Cypresshead which appear to have been deposited under relatively high flow velocities. Such depositi onal conditions appear to have preferentially winnowed mica grains, preventing their incorporation into the sediments, with the highest concentrations of mica seen to be associated with more quiescent depos ition, particularly with basal medium to fine clayey sands and the entirety of the EPK Mine cores and Birds section. Kaolinite vermiforms (or books) occurring in both the sand and silt size-fractions (Fig. 47E and F), are observed throughout Cypresshead sediments in Florid a, but are relatively rare in southeastern Georgia Cypresshead se diments. The distribution of verm iforms is similar to that of the feldspar grains in Florida, with the largest an d highest concentrations seen toward the base of exposures, although the distributi on appears, in part, correlate d with intervals of mica concentration, a preferred mineralogical templa te on which vermiforms commonly form during weathering (Jeong, 1998b). Clay size-fraction Kaolinite is the dominant clay mineral found in both Cypresshead Formation and reworked Cypresshead Formation sediments (Table 4-2). In Florida, the clay (< 2 m) size-fraction also includes lesser amounts of quartz, gibbsite, halloys ite, hydroxyl-interlayere d vermiculite (HIV), and crandallite-florencite, with occurrences of rutile, anatase, boehmite, and diaspore in discrete samples. Furthermore, metahalloysite is inferred to be a part of the suite based on the occurrence of halloysite, and the likelihood for halloysite dehydration in res ponse to fluctuating water table conditions. Both smectite and illite are indicated to occur in north-central Florida Cypresshead sediments in association with the Trail Ridge and Highland mining area samples collected updip from the principle sampling sites employed in this region. Overlain by Pleistocene sediments associated with Trail Ridge, these clays may be indicative of a less weathered Cypresshead suite
95 Table 4-2. Clay size-fraction mineralogy of Cypresshead Formation and reworked Cypresshead Formation sediments. Sample ID Interval% ClayKaoliniteGibbsiteHalloysiteIlliteHIVQuartz CrandalliteFlorencite Other Cypresshead Formation FL EPK36-J-1225-2713.1****** **EPK36-J-1227-307.5***** **EPK36-J-1235-408.2***** *EPK36-J-1240-4414.7***EPK36-J-1244-4612.6***EPK36-J-1246-4812.9***EPK36-J-1248-5013.7***EPK36-J-1250-534.7***EPK36-J-1253-569.7***EPK36-J-1256-5915.6***EPK36-J-1259-6220.3***EPK31-P-4027-3513.8***EPK31-P-4035-4514.7***EPK31-P-4045-5016.0***EPK31-P-4050-6218.2***EPK31-P-4062-6516.8***EPK30-V-616-225.0*** *EPK30-V-622-243.8*** *EPK30-V-624-273.0** *EPK30-V-630-355.1**** ****EPK30-V-635-3910.1**** ***EPK30-V-639-436.9*** ***EPK30-V-643-484.8*** **EPK30-V-648-539.0*** **EPK30-V-653-588.9***EPK30-V-658-6310.5***EPK30-V-663-6813.8***EPK30-V-668-7313.0***EPK30-V-673-7811.9***FRG-1 114.0**** *FRG-1 210.4***** *FRG-1 39.3**** *FRG-1 41.0***** *FRG-1 51.7***** **FRG-1 61.8***FRG-1 73.7***FRG-1 811.5***FRG-1 92.7*** *FRG-1 1034.2*** *FRG-1 119.3***FRG-1 128.3***FRG-1 137.5***FRG-1 149.4***FRG-1 152.5***FRG-2 18.7**** *FRG-2 24.7**** *FRG-2 36.6***FRG-2 42.0***FRG-2 510.4***FRG-2 68.3***FRG-2 76.6***FRG-2 85.6***FRG-2 94.4***FRG-2 108.6*** *FRG-2 115.4*** *FRG-2 1210.3*** *FRG-2 1310.2*** *R (*) Note: Mineral component designations correspond to the following semi-quantitative sc heme: *** = major (> 50%), ** = minor (10-50%), and = trace (< 10%). R = rutile, S = smectite, A = anatase, B = boehmite, D = diaspore, Gr = goethite
96 Table 4-2. (continued). Sample ID Interval% ClayKaoliniteGibbsiteHalloysiteIlliteHIVQuartz CrandalliteFlorencite Other Cypresshead Formation FL (cont.) FRL-1 119.5***FRL-1 210.6***FRL-1 39.7***FRL-1 410.6***FRL-1 51.7***FRL-1 69.2***FRL-1 713.9***FRL-1 89.5***FRL-1 910.5***SSJ-1 116.9***R (*) SSJ-1 217.5*** *R (*) SSJ-1 310.6*** *SSJ-1 411.0*** *SSJ-1 52.0***SSJ-1 68.8***SSJ-1 710.0***SSJ-1 89.9***SSJ-1 98.5***SSJ-1 1010.6***SSJ-1 1112.8***TRF2214 60.0-62.5---*** **S (**) WEX164 18.0-26.0---*** *S (**) WEX366 9.0-10.0---*** **Reworked Cypresshead Formation FL SSD-118.3*** *****A (**)/R (*) SSD-123.3***** **B (*)/D (*) SSD-1 33.0**** *SSD-1 43.0***** *SSD-1 54.7**** *SSD-1 66.5***SSD-1 73.6***SSD-1 86.7***SSD-1 95.2***SSD-1 104.9***Cypresshead Formation GA J-1119.6***** **Gr (*)/R (*) J-1227.8**** *Gr (*) J-1 319.1**** *Gr (*) J-1 463.0*** *Gr (*) J-1 512.3*** *Gr (*) J-1 66.3*** *LB-1 120.4**** *LB-1 224.6**** *LB-1 315.6****LB-1 412.4***LB-1 52.4***LB-1 663.0***LB-1 72.2***B-1 126.9*** **Gr (*) B-1 233.3*** ***Gr (*) B-1 327.6*** ***Gr (*) B-1 433.4*** **Gr (*) B-1 524.0*** ** Note: Mineral component designations correspond to the following semi-quantitative sc heme: *** = major (> 50%), ** = minor (10-50%), and = trace (< 10%). R = rutile, S = smectite, A = anatase, B = boehmite, D = diaspore, Gr = goethite consistent with what would have been originally deposited with the unit prior to weathering, or the clays may have been derived locally through reworking of proximal Hawthorn Group
97 sediments which would have been exposed adjacent to these locations. The trace illite noted for FRG-2-13 is most likely reworked from Hawt horn Group sediments immediately below the sampling interval. The mineralogy of Georgia clay size-fractions di ffer slightly from those in Florida, with the absence of halloysite and crandallite, and the addition of significant illite concentrations at both the Jesup type locality and the Birds site Based on evaluation of the 060 reflection, this illite is of the 2M1 polytype and therefore of detrital orig in. Goethite is also a common trace component in southeastern Georgia Cypresshead sediments (Jesup and Birds), and reflects higher iron (Fe) concentrations associated with the depo sition of these sediments. Additionally, vertical trends in the clay mineral suite suggest an in situ weathering origin for se veral clay size-fraction phases, favoring the formation of HIV and gibbsite in the unsaturated z one and halloysite under saturated conditions. Facies Architecture Pirkle (1960), Kane (1984) a nd Huddlestun (1988) have made past attempts at describing the lithofacies which compose th e Cypresshead Formation in north -central Florida and Georgia based primarily on field observati ons. In this study, information from field observations were combined with a detailed evaluation of the sedimentary framework of the unit to arrive at the facies outlined in Table 4-3. The distribution of stratigraphic disc ontinuities was then used to define facies boundaries and corr esponding facies architecture as shown in Figures 4-8 and 4-9. A detailed evaluation of the facies associated with reworked Cypresshead sediments from the Davenport Sand Mine core (SSD-1) was beyond the focus of this study, but appears, based on limited data, to mimic the characteri stics of the Cypresshead Formation.
98Table 4-3. Facies summary of the Cypresshead Formation in north-central peninsular Florid a and southeastern Georgia. Facies Lithology Sedimentary structures Bioturbation North-central Florida Upper shoreface (USF) Quartz sand, medium to coarse, with minor gravel throughout. Sand is slightly clayey to clayey near the top due to illuviation Tabular and trough cross-bedding (10-60 cm) dominant except where bioturbation or weathering has destroyed primary sedimentary structure (massive bedding) Dense to sparse, consisting of Ophiomorpha spp., Skolithos spp. and undiff. traces with escape structures Proximal lower shoreface (pLSF)Quartz sand, medium to coarse, with gravel throughout. Sand is slightly clayey, with clay more common as stringers and lenses. Trough cross-bedding (10-40 cm) transitions into 0.5-3 m tabular cross-bedding associated with the migration of dune (megaripple) bedforms Dense near top, consisting of Ophiomorpha spp. and Skolithos spp. traces Distal lower shoreface (dLSF)Quartz sand, medium to fine, with interbedded coarse sand and gravel. Sand is slightly to moderately clayey. Minor feldspar (<5%) and mica Horizontal bedding with 10-40 cm trough cross-bedding interbedded with 0.3-1 m hummocky stratification and storm derived coarse sand and gravel graded beds Minor bioturbation consisting of Thalassinoides spp. and Skolithos spp. traces Offshore transition (OST)Quartz sand, medium to fine, is the dominant lithic component. Sand is clayey, with up to 20% or more clay. Minor feldspar (< 10%) and mica Horizontal to massive bedding is dominant unknown Offshore inner shelf (OSI)*Quartz sand, medium to fine, is the dominant lithic component. Sand is fossiliferous, variably calcareous, and sometimes clayey. Mollusks are the dominant fossil type Massive bedding which appears to be devoid of primary sedimentary or biogenic structures unknown Southeastern Georgia Proximal lower shoreface (pLSF)Quartz sand, medium to coarse, with minor gravel throughout. Sand is slightly clayey to clayey near the top due to illuviation. Clay common as stringers and lenses. Tabular and trough cross-bedding (10-60 cm) dominant except where weathering has destroyed primary sedimentary structure (massive bedding) Moderate to sparse near top, consisting of undiff. traces Distal lower shoreface (dLSF)Quartz sand, medium to fine, with interbedded coarse sand and gravel. Sand is moderately clayey. Minor mica Horizontal to massive bedding is dominant Minor bioturbation consisting of Ophiomorpha spp. traces Offshore transition (OST)Quartz sand, fine, is the dominant lithic component. Sand is clayey, with up to 30% or more clay. Clay common as stringers and partings. Minor mica Horizontal to massive bedding is dominant Moderate to sparse bioturbation, consisting of undiff. traces Offshore inner shelf (OSI)*Quartz sand, medium to fine, is the dominant lithic component. Sand is fossiliferous, variably calcareous, and sometimes clayey. Mollusks are the dominant fossil type Massive bedding which appears to be devoid of primary sedimentary or biogenic structures unknown Corresponds to the Nashua Formation of Huddlestun (1988) and Sc ott (1992a), which is considered a facies of the Cypresshead F ormation for this study.
99 Figure 4-8. Correlation of Cypresshead Form ation facies in north-central Florida.
100 Figure 4-9. Correlation of Cypresshead Form ation facies in southeastern Georgia.
101 North-central Florida facies In north-central Florida, a to tal of five facies were iden tified based on sedimentological factors which define a single coarsening-upwar d cycle consistent wi th the progradational siliciclastic deposition of a s horeface-shelf parasequence. Incl uded with these facies is the Nashua Formation of Huddlest un (1988), which has been previ ously defined by both Huddlestun (1988) and Scott (1992a) as an offshore facies of the Cypresshead Formation in both northcentral Florida and southeastern Georgia. For th e purpose of this study, the Nashua is assigned to the offshore inner shelf (OSI) facies and defi nes the basal facies unit of the Cypresshead Formation for both north-central Florida and southeastern Georgia. Continuing upsequence, the second facies identifi ed in north-central Florida is an offshore transition (OST) facies for which only a small section may be exposed at the base of the Grandin Sand Mine (FRG-2-12?). However, it is this faci es, which appears to correlate to the kaolinitic sands associated with the EPK Mine orebody. Ba sed on sedimentological factors such as total clay and mica content, apparent horizontal to massive bedding characteristics, and a downdip position of the kaolinitic sand orebody relative to the Grandin Sa nd Mine exposure, a transitional offshore environment is likely. Cores sampled from this location were site d at surface elevations of 27 m to 28 m above msl, with the kaolinitic orebody encounter ed an additional 7.5 m to 9 m below the surface. This places the top of the or ebody at no greater than 20.5 m above msl, an elevation which would be consistent with the positioning of an OST facies relative to the elevation-facies depth characteristics noted for Grandin Sand Mine exposures (Fig. 4-8). General characteristics of the OST facies include quart z sand in the medium to fine mean grain-size range, with a minor coarse sand component and up to 20% or more clay (< 2 m) content. In north-central Florida, the clay content of this facies occurs primarily as a binding matrix.
102 Additionally, as much as 10% of the sand fraction may consist of feldspar and/or mica, with both phases concentrated in the medium and fine sand fraction. The third facies, identified as a distal lowe r shoreface (dLSF) facies, is characterized by mediumto fine-grained quart z sands which commonly exhibit hummocky cross-stratification, and are interbedded with graded and trough crossbedded storm-derived coarse sands with minor gravel (Figs. 4-8 and 4-10A). Biot urbation associated with this facies is sparse, and is dominated by Thalassinoides spp. and Skolithos spp. traces. Clay content of the dLSF facies can be as much as 10% or more in the medium to fine sand component where it occurs as a binding matrix. Additionally, minor mica and/or feldspar are common, and may repr esent up to 5% or more of the sand fraction, the mica content being primar ily a function of hydrodynamic sorting while the feldspar content is most likely related to pres ervation relative to weathe ring. Lastly, grain-size distributions for this facies are mainly unimoda l, consistent with deposition below normal wave base (Balsillie, 1995). The proximal lower shoreface (pLSF) facies is well developed in north-central Florida mine exposures due in no small part to the relativ e concentration of coarse sands associated with the largeto medium-scale tabular cross-stratification generate d by the migration of dune (or megaripple) bedforms in this environment (Fig. 4-8, Fig. 4-10B). This facies is characterized by an increase in quartz sand grain-size (medium to coarse) with commonly associated gravel, a reduction in the overall clay content except for the occurrence of discrete clay lenses and stringers, and the extensive development of crossstratification. The base of the facies is defined by an erosional disconformity which indicates a significant increase in scouring and sediment transport velocity, which most likely correspond ed to a strong southerly oriented longshore current (Fig. 4-5D). During periods of reduced curre nt flow, clay stringers and lenses could have
103 Figure 4-10. Examples of Cypresshead Formati on facies. A) Exposure (~ 2 m) in the Grandin Sand Mine illustrating the ve rtical juxtaposition of dLSF, pLSF and USF facies (large-scale tabular cross-stratification is not developed in this section), B) FRL-1 exposure (~ 13 m) exhibiting well developed tabular cro ss-stratification associated with the pLSF facies, C) Upper portion of the B-1 exposure (~ 2.5 m) illustrating the OST facies, D) J-1 exposure (~ 5.5 m) illu strating an example of the dLSF and pLSF facies, with the J-1-4 clay bed (arrow) indicated for reference. developed in response to quiescent conditions, with reactivation of current activity corresponding to the incorporation of clay ri p-up clasts in overlying sediments. Based on the bimodal character of grain-size distributions for the pLSF facies, energies associated with current sorting and transport were mixed with wave, and possibly tid al, influences. Additiona lly, the top of this facies commonly exhibits dense bioturbation associated with Ophiomorpha spp. and Skolithos spp. traces (Fig. 4-10A), marki ng a non-depositional discontinuity between the pLSF facies and the overlying facies.
104 The last facies, the upper shoreface (USF) faci es, is characterized by medium to coarse quartz sands with minor gravel, which exhibit well developed smallto medium-scale tabular and trough cross-stratification consistent with a southerly directed wa ve and current flow orientation (Figs. 4-5C and 4-10A). Many coarse beds exhibit graded bedding, and bioturbation varies from dense to sparse, dominated by Ophiomorpha spp., Skolithos spp. and undifferentiated traces. Additionally, sands are generally slightly clayey, except near the top of exposures where illuviation has resulted in the post-de positional concentration of clays. Southeastern Georgia facies Cypresshead Formation exposures in southeaste rn Georgia are not as well developed as what is available in north-central Florida (i.e., sand mines). As a result, only four facies have been identified for the Cypresshead in this region (Table 4-3), with two of the exposures used in this study exhibiting only a single facies (Fig. 4-9). However, as is seen in north-central Florida, these facies define a single coarsening-upward cycle consistent with a progradational shorefaceshelf parasequence. As with Florida, the basal facies of the C ypresshead Formation in southeastern Georgia, the OSI facies, corresponds to the Nashua Form ation per the previous discussion. Upsequence of the OSI facies, the OST facies shares many simila rities with the description of the facies in Florida, including a high clay content, sometime s exceeding 30%. However, as exposed at Birds, the facies in Georgia appears to be generally fi ner grained, consisting of often thinly-bedded fine quartz sand with discrete layers or partings of clay, and a signifi cant clay matrix component (Fig. 4-10C). Dominated by horizontal bedding, the OST facies exhibits evidence of sparse to moderate bioturbation, and a notab le lack of current or wave generated scouring or bedforms. In southeastern Georgia, approximately 1.5 m of the dLSF facies is exposed at the Jesup type locality (Figs. 4-9 and 4-10D). Sedimentological characteristics in Georgia are similar to
105 that seen in Florida, with the facies characterized by mediumto fine-grained quartz sands with interbedded coarse sand and gravel, although the mean grain-size of the facies is coarser in Georgia. This coarser substrate is reflected in the presence of Ophiomorpha spp. traces rather than Thalassinoides spp. as is observed for the dLSF facies in Florida. Lastly, similar to what is seen in Florida, the pLSF facies in southeastern Georgia is characterized by medium to coarse quartz sands with minor grav el throughout. Clay content in the facies is concentrated at th e top by illuviation, but exhibits a significant matrix component (520%) and concentrated deposition of clays as stringers, lenses and beds (Fig. 4-10D). Additionally, the facies possesses well deve loped tabular and trough cross-stratification indicating significant current and/ or wave action consistent w ith a dominant southerly flow direction at the type locality a nd an easterly direction at Linde n Bluff. Clay rip-up clasts are common in cross-bedded sands overlying clay lenses or beds (Fig. 4-9). Discussion Depositional Environment It is probable that nearshore conditions not unlike those se en along the modern Georgia and Florida east coasts are res ponsible for deposition of the Cypresshead Formation, except for the fact that climate, and consequently its im pact on sea-level and sedi mentation, differed from what is experienced today. As a result, a review of modern coastal conditions offers the most suitable basis for evaluating Cypresshead depositional environments. The present coastline of Georgia and extrem e northern Florida in the vicinity of the Georgia Bight is characterized by barrier islands broken by tidal inlets (Kussel and Jones, 1986). Impacted by semidiurnal tides with an average ra nge of 2 m, spring tidal ranges up to 3 m, and a southerly longshore current (Davis et al., 1992), this coastline is consistent with a mesotidal mixed energy system. Further south, along the re mainder of eastern Florida, are marine
106 depositional coasts dominated by barrier beaches barrier islands, splits, and overwash fans (Schmidt, 1997). This modern wave-dominated to mixed energy east coast barrier system is bounded by a relatively narrow and st eep sloping shelf, and is impacted by wave energies associated with a mean annual wave height of up to 0.7 m (Nummedal et al., 1977). Longshore transport is dominantly to the south as a result of strong northerly winds associated with winter frontal systems that domi nate longshore currents, even thoug h prevailing winds have a distinct southerly component (Davis et al., 1992). Cypresshead Formation model Based on the collective sedimentological and mine ralogical characteristics outlined for this study of the Cypresshead Formati on in both north-central Florida and southeastern Georgia, the unit was deposited in a nearshore marine setting as tw o distinct shoreface-shelf parasequences, one of which was in a wave-dominated environmen t (north-central Florid a) and the other in a mixed energy environment (southeastern Georgi a). In both instances, deposition took place in response to relative sea-level fall, resulting in a single coarsening-upward cy cle of siliciclastics. Internally within the unit, this model for depos ition is expressed by the presence of clinoform surfaces within the unit that dip gently at 2-3 to the east in northern-central Florida and southeastern Georgia. The most likely nearshor e marine environment for Cypresshead deposition appears consistent with a strand plain setting, lacking well-developed lagoons or marshes. In north-central Florida, the Cypresshead Form ation thins toward the west onto the flanks of the Ocala Platform where it is absent (Sco tt, 1988a). Thus, the Ocala Platform acts as a depositional basin divide between the Cypresshead Formation to the east and the time equivalent Miccosuckee and Citronelle Form ations to the west. Deposition of the Cypresshead would have commenced during a sea-level highstand, with the we sternmost, updip extent of the unit in close proximity to the Trail Ridge and Highland sample locations discussed in this study. This
107 location, proximal to exposed uplands consis ting of Hawthorn Group (i.e. smectiteand illitebearing) sediments, along with the subsequent burial of these sediments by early Pleistocene Trail Ridge sediments helped preserve the clay mineral suite noted herein. Whether a similar clay mineral suite was initially deposited elsewhere with Cypresshead sediments in north-central Florida is uncertain, but it is pr obable that the original suite va ried considerably from the highly weathered, kaolinite-domin ated suite seen today. Subsequent deposition in north-central Florida would have been dictated by relative regressing sea-level, sediment supply, current positioning, and available accommodation. In north-central Florida, the latt er of these would have been related to antecedent topography developed on top of Hawthorn Group (Coosawhatchee Formation) sediments. Dictating available accommodation for Cypresshead siliciclastics, th is surface is known to have been scoured by pre-Cypresshead erosion and submarine curren t activity, creating irre gularities which would have acted as preferred nearshor e conduits for siliciclastic tr ansportation by longshore current and storm activity, or topogra phic lows for the preferential deposition of hydrodynamically sorted fines (e.g. clays and mica). As indicated by the well deve loped cross-stratification a nd overall grain-size of Cypresshead siliciclastics, str ong longshore current activity, signi ficantly greater than seen along the modern Florida and southeastern Georgia coas ts, is proposed as a major factor in the coast parallel transportation of thes e sediments. Evidence for similar processes extending down the Florida peninsula are noted from an outcrop desc ribed by Johnson (1989) from Lady Lake pit in Lake County, Florida, which contains two cros s-bedded, coarse-grained sand beds (Beds 3 and 4), four and three feet thick, which appear to be similar to the dune (o r megaripple) bedforms (Figs. 4-4B and 4-10B) described at in the Grandin and Goldhead exposures. In fact, estimates
108 by Kane (1984) for current velocities based on the maximum grain-sizes and sedimentary structures noted in Cypresshead sediments indicate values of 40+ cm/sec, which are consistent with strong longshore current activity cap able of transporting even the largest fraction (i.e., discoid quartzite pebbles) noted for the C ypresshead. As indicated by Dobkins and Folk (1970), the development of a discoid pebble shap e, as seen in Cypresshead sediments, is consistent with a high wave-energy shoreface en vironment. Under such conditions, the discoid (oblate) shape can be generated through abrasion as pebbles slide back and forth over sand or smaller pebbles in the surf zone. Although Pirkle et al. (1964) and others have argued against using the shape of quartzite pebbles solely as an indicator of depositi onal environment, noting that some component of shape is likely influe nced by the inherited textural and structural characteristics, the proposed current velocities of Kane (1984) help to ex plain their o ccurrence in Cypresshead sediments. Based on facies characteris tics discussed in this study (e.g., dLSF gravels and hummocky cross-stratification), storm-induced sedimentation appears to have played a significant role in deposition of Cypresshead sediments in north-centr al Florida. In fact, support for storm-induced depositional process in Florid a playing a significant role dur ing the mid-Pliocene is found elsewhere in the observations of Allmon (1992) and Missimer (2001a;b), as they correlate the deposition of massive shell beds from the Tami ami Formation to storm processes occurring in shallow coastal waters at that time. Hummocky cross-stratification in the lower portion of the Grandin Sand Mine exposures is further support for significant levels of storm activity, as it occurs in repetitive successions separated by horizontally bedded clayey fine sands. The hummocky beds themselves are relatively de void of burrows, with evidence of sparse Thalassinoides spp. burrows limited to the interbedded hor izontal clayey fine sands. This appears
109 consistent with multiple storm-generated depos itional events being separated by periods of quiescent deposition during which burrowing would have been favored. Further supporting the view of a high-stress depositional environment for the Cypresshead Formation is the low diversity macrobenthic infaunal trace fossil assemblage outlined in this study. Characteristic of shallow-water facies with relatively coarse substrates, this assemblage possesses relatively low diversity in comparison to the modern shelf off south-central Texas (Hill, 1985), but is similar to that described by Kussel and Jones (1986) for the late Pleistocene to Holocene Satilla Formation deposited seaward of the Cypresshead in southeastern Georgia and northernmost Florida. Although Ophiomorpha spp. and Skolithos spp. traces are dense within select beds correlated to nondepositiona l discontinuities, the overall density of traces appears to decrease significantly with increased water depth. This is inconsistent with the opinion of Howard and Reineck ( 1972) that under ideal conditions bioturbation should increase with increasing water depth, due to fewer bottom disturbances such as storms. Possible causes for this trend might include deeper water wave base impacts of storms or nutrient conditions unsuitable for supporting a more de nse and/or diverse assemblage. For southeastern Georgia, the nearshore marine depositional model proposed by this study differs from that applied in Florida only with respect to differences in coastal environment (wave-dominated vs. mixed energy) and timing. For Cypresshead sediments deposited proximal to the riverine discharge of s iliciclastics along the southeastern Georgia coast, detrital clay content is expected to be gr eater than the current and wave winnowed sediments transported down the Florida coastline. Additionally, the poten tial for a fluvial-deltaic component associated with deposition is more likely as reflected in the flow orientation of crossstratification in Linden Bluff sediments (Fig. 4-5B). However, once siliciclastics reached the coastline, southerly
110 directed longshore currents would have driven co ast parallel migration of sediments. Huddlestun (1988) used a depositional model with a very br oad sound, at least 80 km wide, to explain many of the sedimentological characteri stics described in this study, but a mesotidal coast (2-4 m tidal range) characterized by the development of short (or stunted) barriers cut by common tidal inlets describes the spatial distribution of Cypr esshead sediments in Georgia as well. If significant sediment fluxes are necessary to explain the overall volume of Cypresshead and related siliciclastics deposit ed on the Florida Platform, then total sediment loads carried by rivers to the Georgia coastline would have been large during periodic storm and flooding related erosional events. As such, homopycnal or even hyperpycnal river plumes would have been possible (Addington et al., 2007; Summerfield and Wheatcroft, 2007), favoring nearshore deposition of bed and suspended loads, includ ing muds, which storm, wave, and longshore currents would then have moved southward along the coastline. Under such conditions favorable to nearshore mixing, salinities proximal to rive r outflows were likely brackish, favoring the occurrence of Ammonia beccarii and Elphidium spp. seen in Cypresshead sediments from Effingham County, Georgia (Huddlestun, 1988). Additi onally, high physico-chemical stresses in such a setting would have resu lted in a paucity of infaunal burrowing as is noted for the Cypresshead Formation in southeastern Georgia.Fu rthermore. However, rather than indicating an enclosed sound environment as suggested by Huddlestun (1988), such conditions would have favored river outflow along a strand plain or mixed barrier-estuary coastline significantly different than the modern mesotidal Georgia coast of today. Corresponding to the downdip faci es of Huddlestun (1988), and exhibiting similarities to the Miccosukee Formation in southwestern Geor gia and the eastern panh andle of Florida as noted by Huddlestun (1988), the OST facies in Geor gia may possess, at least in part, a prodeltaic
111 origin. Argued by Scott (2001) as the deposit ional origin of the Miccosukee Formation, a prodeltaic origin for the OST facies would be c onsistent with the well-sorted sands, discrete occurrence of layers or laminae of white to gray clay, overall high clay content, lack of scoured contacts, and sparse to modera te bioturbation observed with the facies. As such, it appears reasonable to propose a prodeltaic or igin for the OST facies seen at the Birds site. This would be consistent with the general spatial distribution of this facies as descri bed by Huddlestun (1988) as being distal from major rivers presently ex iting along the Georgia co astline. If the coarser sand lithologies observed at the type locality (J-1) and the Linde n Bluff site (L-1) are to be considered as proximal to points of coastal sedime nt delivery, then it is consistent for the OST facies at Birds to have a prodeltaic origin, with deposition occurring distally from river outflows. Model consistency and the siliciclastic conveyor Other than exhibiting evidence for far more post-depositional weathering, Cypresshead sediments share significant mine ralogical similarities with Late Miocene through Pliocene south Florida siliciclastics, particularly with respect to the occurrence of feldspar and mica noted in both. Cunningham et al. (1998) fi rst noted minor feldspar ( 5%) and mica, with the particle shape of the dominant quartz fraction angular to subrounde d, which is also a shared characteristic of Cypresshead sediments. Additionally, it is rec ognized as a consequence of this study that the Cypresshead Formation contains two shoreface-shelf parasequences marking major episodes of shoreline regression during the Pliocene. Other than for timing, this is similar to the models that have been proposed for La te Miocene through Plio cene siliciclastics in southern Florida. Based on such a correlation, th e lack of Late Miocene through early Pliocene preservation of siliciclastics in southeastern Georgia and north-central Florida point to the preferred mobilization and coast-parallel tran sport of these sediments southward as the siliciclastic conveyor of Cunni ngham et al. (2003). Throughout the Late Miocene to latest
112 Early Pliocene, siliciclastic depo sition dominated the eastern Flor ida Platform at the expense of carbonate deposition (Scott, 1988b; Cunningham et al., 1998; 2001; 2003). In response to a rise in Pliocene eustatic sea-level and/or loadingbased subsidence, mixed carbonate and siliciclastic deposition returned to the southern penins ula throughout the remai nder of the Pliocene (Missimer, 1992; Cunningham et al., 2001; 2003), wh ile siliciclastic depos ition continued to dominate to the north in response to retrograd ation. This resulted in the deposition of the Cypresshead Formation in north-central Florida and southeastern Georgia. Similar to what has been noted for the Cypresshead Formation, the spatial record of Late Miocene through Pliocene siliciclastic deposition in southern Florida was also controlled by the interplay of sediment supply, relative sea-leve l, current activity, and antecedent accommodation (Guertin et al., 1999). As illustrated by Cunningham et al. (1998), a shallow marine basin or low at the top of the Arcadia Formation existed from west of Lake Okeechobee to the Florida Keys, which acted as a preferred c onduit of accommodation for silicicl astics transported southward by longshore currents and storms. During periods of reduced siliciclasti c flux, sediments would have been subjected to current reworking a nd winnowing, imparting char acteristic sorting and gradation to these sediments, c onsistent with the observations of both Guertin et al. (1999) and Cunningham et al. (1998; 2003). Overall, the Late Miocene to Pliocene sili ciclastic deposition from south Florida to southeastern Georgia represents a retrogr adational parasequence set with individual parasequences progradational in character as a result of regression-based deposition. Retrogradational stacking results when the l ong-term rate of accommodation exceeds the longterm rate of sedimentation. In this way, accommodation space is created more rapidly than it is filled, water depth becomes deeper, and facies increasingly move farther landward. Although
113 each parasequence is shallowing-upward, the amount of deepening at the flooding surface exceeds the amount of shallowing in the follo wing parasequence, producing a net overall deepening within the parasequence set. A consequence of this observation is that it points to a period of subsidence as a conse quence of Paleogene through Early Neogene sediment loading of the Florida peninsula, prior to the Plei stocene uplift noted by Opdyke et al. (1984). In considering consistency for the Late Miocene through Pliocene siliciclastics that compose the retrogradational se quence tract of which the Cypresshead Formation is a part, depositional processes should be considered. Of pa rticular interest is th e assigning of a fluvialdeltaic model to the deposition of the SS2 sili ciclastics of Cunningham et al. (2003). Whereas interest in assigning an alluvial or fluvial-deltaic origin to C ypresshead sediments extends back to the work of Bishop (1956), Pi rkle (1960), and Pirkle et al. (1964), Kane (1984) successfully challenged this perception on the basis of paleocu rrent and facies analys is, deposit geometry, and the distribution of marine pa leontological evidence. However, similarities between the Cypresshead and the predominately alluvial Citr onelle Formation to the west have kept the alluvial model for the Cypresshead and simila r sediments alive as evidenced by the fluvialdeltaic interpretation for the SS2 si liciclastics (Cunningham et al., 2003). The interpretation of the SS2 sequence as representing a fluvial-deltaic depositional system was based on seismic reflection geometries, t hus suggesting the presen ce of a N-S oriented fluvial system along the Florida Platform respons ible for deposition in a nearshore, marine-tobrackish environment. As noted, such a fluvial-d eltaic model has been suggested by previous authors for other Florida siliciclastics (Bishop, 19 56; Pirkle, 1960; Pirkle et al., 1964; Klein et al., 1964; Peacock, 1983; Missimer and Maliva, 2 006), but was first challenged by Winker and Howard (1977b), who noted the unlikelihood of a major river flowing the length of the Florida
114 Platform rather than seeking a more direct rout e to the sea. Furthermore, evidence reported by Cunningham et al. (2003) does not seem to suppor t fresh to brackish water conditions one would associate with a deltaic complex. Rather, foraminiferal data for SS2 sediments correspond to a shallowing upward sequence from outer-shelf to inner-shelf and possibly lagoon environments, which can be interpreted as favoring the prograding spit model proposed by Ginsburg et al. (1989) and Warzeski et al. (1996). In fact, Ginsburg et al. (1989) sought to combine both fluvialdeltaic and spit models, suggesting that silic iclastics were transported southward by a combination of longshore and riverine processes, eventually to be redistributed by wave and current activity to form a giant progradational sp it with sediments extending as far south as the Florida Keys. Warzeski et al. (1996) proposed that a pathway of maximum paleocurrents and a coincident trend of coarse-grained siliciclas tics trend down the peninsula. This trend, later refined by Cunningham et al. (1998) offers a means by which sea-level fluctuations, strong currents, and storms could have redistributed sili ciclastics to the south and east (Guertin et al., 1999). Further difficulty with employing the fluvial-del taic model to SS2 sediments is the failure of Cunningham et al. (2003) to provide evidence for a riverine source of sediments landward (to the north) of the SS2 sediments. In fact, the observation by Bishop (1956) of deltaic beds, which is used to support the hypothesis of Cunni ngham et al. (2003), actu ally corresponds to sediments of the Cypresshead Formation near the SSD-1 and SSJ-1 drill hole sites used in this study, and are known today to be cross-bedded sands similar to those of marine origin noted in this study at several exposures in north-central Florida (FRG and FRL). Also, as noted by Cunningham et al. (2003), the di scoid pebbles within SS2 sediments are consistent with deposition in a beach environment (Dobkins and Folk, 1970). These observations fit the
115 spit/nearshore marine depositional model more suitably than the fluvial-deltaic model of Cunningham et al. (2003). Timing and Regional Stratigraphic Correlation Based on field observations, sequence stratigraphic an alysis, and a review of the works of Huddlestun (1988), Cunningham et al. (1998; 2001; 2003), Guertin et al. (1999), and others on the late Neogene siliciclastics of Florida and Ge orgia, the timing for de position and reworking of the Cypresshead Formation was determined for this study. The results are summarized in Figure 4-11 relative to the Berggren et al. (1995a; b) time scale, and cons train the age of the Cypresshead Formation deposition to about 3.4 to 2.3 Ma and significant reworking of the unit to about 2.8 to 1.8 Ma. Pre-Cypresshead siliciclastic flux As noted previously, the initial onset of the siliciclastic flux ultimately associated with the Cypresshead Formation is first seen with deposition of the Late Miocene SS2 siliciclastics of Cunningham et al. (2003). Initially deposited over a broad swath of southern Florida beginning at 8.6 Ma (Fig. 4-12), this sequence is equivalent, in part, to the Interval I siliciclastics (6.2 5.5 Ma) of Guertin et al. (1999). This latter un it has been included with the upper Peace River Formation for this study (Fig. 4-11), and is te ntatively correlated to bracket the TB3.3-TB3.4 boundary of supercycle TB3 of Ha q et al. (1988) corresponding to the 5.7 Ma sea-level fall of Miller et al. (2005) (Fig. 4-13). Increased siliciclastic sedimentat ion during this interval appears to be coincident with the late Miocene intensif ication of the Florida Current-Gulf Stream system (Mullins et al., 1980; Eberli et al., 1997), which, in turn, is linked to a eustatic lowstand correlated to late Miocene glac iation on Antarctica. With loweri ng of sea-level under conditions of increased aridity in the southeastern Unite d States (Alt, 1974), a significant volume of siliciclastics would have been gene rated in response to renewed incision and erosion of early to
116 Figure 4-11. Correlation chart for Late Miocene through Pliocene siliciclastic units evaluated in this study. Correlation is to the geomagneti c polarity time scale of Berggren et al. (1995a; b), with calcareous nannoplankton zones according to Okada and Bukry (1980) and Bukry (1991). middle Miocene Coastal Plain sediments and Ap palachian sources. Subsequent mobilization, transport and deposition is reflected in the accu mulation of the more than 100 m of siliciclastics in southernmost Florida associated with the lower SS2 (SS2-E) and Interval I sequences. The succeeding TB3.4 early Pliocene transgress ion, believed to have occurred between 5.3 to 4.9 Ma (Willard et al., 1993; McNeill et al., 2001; Miller et al., 2005), marks a transition toward a more restricted depositional pattern be ginning at ~5 Ma. As illustrated in Figure 4-12,
117 Figure 4-12. Maps illustrating the aerial extent of successive siliciclastic deposition events impacting the Florida Platform from 8.6 Ma to 1.8 Ma. A) SS2-E and Interval I, B) SS2-W and Interval II, C) Cypresshead Formation, D) Cypresshead Formation (Georgia only), reworked Cypresshead sediments, and Interval III.
118 Figure 4-13. Relationship between the Cypresshea d Formation, related siliciclastics in southern Florida, seismic sequence boundaries of Eberli (2000), sequen ce chronostratigraphy, and the sea-level curves of Haq et al. (1988) and Miller et al. (2005). The Haq et al (1988) eustatic curve is adjusted to the Berggren et al. (1995b) time scale (modified after Cunningham et al., 2003). the overall reduction in sediment supply is expressed by the restriction of upper SS2 (SS2-W) siliciclastics to southwestern Florida, with a corresponding absence in the southernmost peninsula, and the sourcing of reworked Miocene sediments as a dominant component of Interval II siliciclastics (Guertin et al ., 1999; Cunningham et al., 2003). This observation appears not to be a response to a reduction in sediment flux being delivered to the Florida peninsula, but rather represents retrogradation to the north as a function of rising Plio cene sea-level or subsidence in response to sediment loading in southern Florida. Further supporting this vi ew of a restriction in sediment supply is the aggradational accumula tion of carbonate-rich sediments during this period, including the Ochopee Limestone Member of the Tamiami Formation in southwest Florida (Cunningham et al., 2001).
119 During the TB3.4 transgression, which appears to have flooded the Florida Platform, siliciclastics to the north would have continued to be mobilized as was the case during the Late Miocene (Guertin et al., 1999). Subs equent regressions associated with sea-level falls at 4.9 Ma and 4.0 Ma (Fig. 4-13; Table 4-4) would have then transported the sediments southward, resulting in the initia l deposition of SS2-W and Interval II siliciclastics bracketing the TB3.4 TB3.5 transition, with cessation of deposition occurring during the TB3.6 cycle at about 3.6 Ma (Guertin et al., 1999). Confirmation for the tim ing of SS2-W deposition during this Early Pliocene (Zanclean) interval is based on diatoms that indicate an age of less than 5.5 Ma, and coccoliths characteristic of Zones CN10c-11 (Fig. 4-12) (Cunningham et al., 2003). Cypresshead deposition and reworking Further retrogradation in association with a continued rise in sealevel or subsidence following deposition of SS2 and Interval II silic iclastics is required to deposit the updip sediments which correspond to the Cypresshead Formation. Therefore, given the regressionbased depositional characteristics of the unit, it is likely that Cypresshead siliciclastics in central and north-central Florida correlate to the TB3.6 TB3.7 boundary (Fig. 4-13), suggesting an early Late Pliocene (Piacenzian) age (3.4.8 Ma) consistent with the 3.3 Ma sea-level fall of Miller et al. (2005) (Figs. 4-11 a nd 4-13; Table 4-4). This correlates closely with the most recent early Late Pliocene (3.4-2.7 Ma) age estimate for the Citronelle Formation (Otvos, 1998b), and also supports correlation of the unit, in part, with the timing of s iliciclastic deposition associated with the Tamiami Formation (SS3/Pinecrest Sand s) in southwest Florida (Missimer, 2001a; b; Cunningham et al., 2001; 2003) (Fig. 4-11). However, it should be noted that correlation of the Cypresshead to the Haq et al. (1988) sea-level cu rve should be considered tentative due to poor age resolution of the siliciclastics and uncertain ty in the age resoluti on and amplitude of the curve.
120 Table 4-4. Correlation of siliciclastic depositional ev ents on the Florida Platform with Haq et al. (1988) sequence boundaries, sequence boundaries of Eberli (2000), and sea-level falls identified by Miller et al. (2005). Sen et al. (1999)aWornardt et al. (2001)aEberli (2000)bN/A ------2.2/2.1/1.9CH reworking/III deposition TB 3.82.4 2.50 --2.5CH deposition/reworking TB 3.73.1 3.21 3.1 3.3 CH/SS3 deposition TB 3.63.9 3.95 3.6 4.0 TB 3.54.1 4.37 ----SS2-W/II deposition N/A ------4.9 TB 3.45.6 5.73 5.4 5.7 TB 3.37.1 6.98 ----SS2-E/I deposition Sequence Boundary Age (Ma) Third-order cycles Miller et al. (2005) Sea-level Falls (Ma) Florida Platform Stratigraphic Event a Based on revised Haq et al. (1988) seismic sequence boundaries corrected to the Berggren et al. (1995b) time scale. b ODP Leg 166 seismic sequence boundaries identified from Great Bahama Bank drilling results. Supporting evidence for the Cypresshead Forma tion being related to a continuation in retrogradation similar to that recorded in SS2 siliciclastics (SS2-E to SS2-W) of Cunningham et al. (2003) was first noted by Klei n et al. (1964), who observed th at sediments now associated with the Cypresshead in Highl ands County, Florida, appear to extend southward into the subsurface of Glades County. This observation was later reinforced by the belief of Scott (2001) that the Cypresshead appears to grade downdip in to siliciclastics of th e Long Key Formation of Guertin et al. (1999) in southern Florida. Howe ver, based on timing and the relative elevation of the base of the Cypresshead, it seems more likely that any cont act with the Long Key Formation (Interval II) or upper Peace Rive r SS2 siliciclastics is disconf ormable or paraconformable in nature (Fig. 4-12), favoring a re trogradational hypothesis. This position is corroborated by the age results from Wingard et al. (1994) a nd Weedman et al. (1995), which support partial correlation of Long Key Formation sediments w ith the upper Peace River Formation in central Florida. As the top of SS2 siliciclastics are located near present day sea-level, and little epeirogenic uplift is expected during the Quaternary near 27 N latitude in the vicinity of the Highlands-Glades County border (W inker and Howard, 1977; Opdyke et al., 1984), sea-level at the time of SS2 deposition was most likely slight ly to moderately above present day levels
121 relative to the present vertical positioning of the Fl orida Platform, with Interval II siliciclastics of the Long Key Formation deposited under deeper marine conditions. This would require ongoing subsidence of central and sout hern Florida to permit retrogradation and deposition of Cypresshead sediments in a nearshore mari ne setting through the development of updip accommodation, as the middle Pliocene sea-level rise (25 m) which culminated at 3.3 Ma was likely of a lower magnitude than that asso ciated with deposition of SS2 and Interval II sediments based on the Haq et al. (1988) curve. Even with the accepted uncertainties in the absolute magnitudes of these sea-level highstands as seen in comparisons to the Miller et al. (2005) curve (Fig. 4-13), retrog radation would remain a prerequi site for Cypresshead deposition. The TB3.6 sea-level highstand associated with deposition of the Cypresshead Formation is well expressed along the southeastern Atlantic co astal plain. In particul ar, this highstand is responsible, in part, for eroding the Orangeburg scarp in Georgia (Dowsett and Cronin, 1990) concurrent with deposition of the Raysor Formation and unnamed Raysor-e quivalent shelly sand of Huddlestun (1988). Given the oc currence of planktonic foraminiferal assemblages in each of these units which correlate to zone PL3 of Berggren (1973), or PL4 in the case of the unnamed Raysor-equivalent shelly sand, these units are early Late Pliocene (early Piacenzian to Piacenzian) in age, and limit the initial onset of Cypresshead deposition in southeastern Georgia (Fig. 4-11). In the case of the Raysor Forma tion, which disconformably or paraconformably underlies the Cypresshead Formation in Georgia, the unit most likely corresponds to a condensed section deposited during the TB 3.6 (and potentially TB3.7) transg ression episode preceding the regression-based deposition of the Cypresshead. The same is likely for the unnamed Raysorequivalent shelly sand, although the potential foraminiferal co rrelation to PL4 of Berggren (1973) suggests the possibility that the unit is instead representative, in part, of an offshore (inner
122 to middle shelf) facies of the Cypresshead Formation, similar to the relationship proposed by Huddlestun (1988) and Scott (1988a) for the Na shua Formation in northern Florida. This contention that the Nashua correlates, in part with the Cypresshead Formation in northern Florida and may represent an offshore facies of the unit appears consistent with field observations and depositional timing for both units. This idea was built upon the earlier observation of Pirkle (1960) that Cypresshead sediments overlay pos t-Hawthorn, sandy shellbearing marls (Nashua Formation) along the easte rn flank of the Lake Wales Ridge topographic trend in northern Florida. However, timing of deposition for the actual type section of the Nashua as early Pleistocene (Calabrian) is incons istent with foraminiferal assemblages collected from two Florida Geological Survey core sites (Cassidy 1: W-13815 and Baywood 1: W-8400) which penetrate the Nashua in Nassau and Putn am Counties proximal to north-central Florida exposures of the Cypresshead. These assemblages, collected from core depths consistent with the position of a potential downdip Cypresshead equi valent facies, correlate with Zone PL5 of Berggren (1973) (Huddlestun, 1988). Although H uddlestun (1988) proposed a multideposit origin of the Nashua to explain this contradi ction, it appears more likely that the Nashua sediments sampled from the two cores are simply representative of an offshore (inner shelf) facies of the Cypresshead Formation, and shoul d be included with th e unit as discussed previously, while the type s ection is a discrete deposit unrelated to the Cypresshead. Palynological data reported by Hansen et al. (2001) from the Peace Creek site sinkhole in Polk County, Florida, constrains the minimum age of the Cypressh ead in central Florida. These sediments correlate with reworked Cypresshead sediments that would have accumulated during the late Pliocene TB3.7 TB3.8 transition at 2.8.8 Ma (Haq et al., 1988) which correlates to the 2.5 Ma sea-level fall of Miller et al. (2005) (Fig. 4-13; Table 4-4) Marking the onset of
123 significant Eurasian and North Am erican glaciation (Hansen et al., 2001), deposit ion of these sediments would have likely taken place near or even below modern sea-level given the present Peace Creek elevation of 34 m and the minimum es timated Quaternary uplift of 36 m calculated by Opdyke et al. (1984) for this region of Florida. Given that marine Cypresshead Formation sediments occur at elevations a bove approximately 52 m in central peninsular Florida, a sea-level highstand of at least 16 m above present msl wo uld be required during deposition, a figure well within the magnitude predicted by both the Haq et al. (1988) and Miller et al. (2005) curves. Thus, based on the constraints placed on depos itional timing by the Raysor Formation, the unnamed Raysor-equivalent shelly sand, reworked Cypresshead sediments, and south Florida siliciclastics, along with strong evidence for correlation of the Cypresshead Formation with the Citronelle Formation in the Florida panhandle, deposition of the Cypresshead in central and north-central Florida appears to have occu rred between 3.4.8 Ma (Figs. 4-11 and 4-12). In southeastern Georgia, deposition of the C ypresshead Formation appears to be restricted to the late Pliocene (late Piacenzian to early Gelasian) from about 3.15.3 Ma. This conclusion is based on a Raysor-equivalent benthic fora miniferal assemblage recovered by Huddlestun (1988) from basal sediments of the unit in so utheastern Georgia (Chatham County), and the noted correlation of north Florid a and southeastern Ge orgia Cypresshead sediments to those of the Nashua sediments discussed previously. Such timing correlates with the TB3.7 TB3.8 boundary (Haq et al., 1988) or the 2.5 Ma sealevel fall of Miller et al. (2005), which is associated with a period of Cypresshead rework ing in central Florida. Although some deposition likely occurred in response to the regression following the sea-level fall at 3.3 Ma, Cypresshead siliciclastics in southeastern Geor gia (and potentially por tions of north-central Florida) appear to be primarily deposited following further retrograd ation north in response to a continuation of
124 subsidence in central Florida and the infilling of available accommodation to the south. This observation supports the contenti on that the Cypresshead Formation, as had been proposed by Huddlestun (1988) for his Nashua Formation, is a multideposit unit (i.e., it was deposited during more than one episode of deposition), a theme reflected in many of the Neogene deposits of Florida (Tamiami, Caloosahatchee, etc.). As for subsequent re working of the Cypresshead in southeastern Georgia, that like ly continued throughout the remainder of th e Pliocene in concert with reworking in Florida (Figs. 4-11 and 4-12) Evidence for this is seen in the timing of deposition for Caloosahatchee Formation siliciclastics and Interval III of Guertin et al. (1999), the latter of which began to accumulate during TB3.8 near the end of th e Pliocene (~2 Ma) and ceased deposition in the area of the Florida Keys near the Pliocene-Pleistocene transition. As Interval III appears to be sourced from Cypressh ead reworking, the timing for the cessation of deposition at ~1.8 Ma likely correlates with th e end of major Cypressh ead reworking in both Florida and Georgia (Fig. 4-11) Continued accumulation of siliciclastics seen in the Caloosahatchee Formation are likely more restri cted to local reworking and appear to cease during the early Pleistocene (Calabrian) (Fig. 4-11). Paleoclimate Forcing of Cypresshead Deposition Significant accumulations of siliciclastics into a given deposystem can be attributed to both regional tectonic factors and paleoclimate (Molnar and Engl and, 1990; Zhang et al., 2001; Harris and Mix, 2002). In the case of the Florida Platfo rm, paleoclimatic transitions toward periods of instability and the corresponding ch anges in temperature, precipit ation and vegetation appear to be the dominant driving mechanism behind the anomalous accumulations of siliciclastics associated with the conveyor model of C unningham et al. (2003). Two such episodes are believed to have impacted the depositional timing of Cypresshead and related Late Miocene siliciclastics; (1) the transition from arid conditions during the Late Miocene to continual El Nio
125 conditions during the early Pliocene warm period (~4.5.0 Ma), and (2) the transition from continual El Nio conditions to a period of global cooling and the onset of significant Northern Hemisphere Glaciation (NHG) (~3.0.5 Ma). Combined with the Late Eocene through Oligocene rejuvenation of the Appalachians (Stuckey, 1965; Dennison and Stewart, 2001; Stewart and Dennison, 2006) which ultimately pr ovided the sediment source and potential energy for erosion, such periods of transitional paleoclimate would have resulted in landscape disequilibrium, thereby enhancing er osion and resultant sedimentation rates as has been noted to occur globally during the middle to late Pliocene (~4 Ma) (Zhang et al., 2001). Middle to late Miocene cl imate and sediment supply Two factors dictated by mid-Miocene paleoc limate were likely prerequisites for the conveyor model of Cunningham et al. (2003) to function; intense weat hering and increased aridity. The first of these, inte nse, even lateritic, weathering, has long been proposed to have occurred in the southeastern United States (Cooke and MacNeil, 1952; Isphording, 1970; 1971), and appears to correlate worldw ide with the development of d eep weathering sequences during the middle to late Miocene (10-13 Ma) (Ashley and Silberman, 1976; Alpers and Brimhall, 1988; Vasconcelos et al., 1994a; b). The second, a worldw ide transition toward ar id conditions during the late Miocene (Yemane et al., 1985; van Zi nderen Bakker and Mercer 1986; Vasconcelos et al., 1994b) is supported by evidence for the increa sed supply of terrigenous dust in oceanic sediments (Ruddiman et al., 1989; Rea et al., 1990) and by the expansion of arid climate flora and fauna (Tedford, 1985; Wolfe, 1985; Yomane et al., 1985; Axelrod and Raven, 1986). In fact, the Miocene ended with the driest climate of the Tertiary (both regional and global), and was accompanied by conversion of savanna to steppe or scrub desert, spread of C4 grasses, and the greatest mammal extinction of the Neogene (Cha pman, 2008). Together with Appalachian uplift and/or crustal arching, Miocene weathering an d aridity are responsible for producing the
126 sediment reservoir required to supply the silic iclastic conveyor model during the Late Miocene and Pliocene. Increased aridity likely facilitated erosion and subsequent deposition of late Miocene through Pliocene siliciclastics in two ways; (1) through a reducti on in overall vegetative cover, and (2) via an increase in the seasonality of prec ipitation. Fluvial transp ort is known to be minimal in both extremely arid and peri-humid (e verwet) climates under e quilibrium, with recent studies indicating that climates characterized by strong seasonal ity of annual precipitation are capable of generating a significant sediment flux (Edgar and Cecil, 2003). Under such conditions, the effect of vegetati on on erosion is minimized, leavi ng soil vulnerable to rainsplash and sheetwash erosion, which favors highly competent runoff and high rates of denudation (Alt, 1974). Additionally, arid c onditions may have increased the rela tive magnitude of rare floods, or conversely, increased the frequency of large flood s (Molnar, 2001). As a consequence, incision rates would have been significant, despite a decrea se in total precipitation and discharge. If the climate instability associated w ith the transition toward early Pliocene warm period conditions is then superimposed on an arid Late Miocene lands cape, the siliciclastic flux directed toward the Florida platform would have been further enhanced given that relatively m odest shifts in climate can have large impacts on the sediment yield of rivers (Syvitski, 2004). As noted by Goodbred and Kuehl (2000) for the Ganges-Brahmaputra, a period of several thousand years of intensified discharge can lead to signif icant continental margin depos ition (e.g. 50 m along the Bengal margin). Pliocene climate and the transition towa rd Northern Hemisphere Glaciation (NHG) Several authors have proposed that paleoc limate during the early Pliocene warm period (~4.5.0 Ma) was consistent with continual El Nio conditions in the tropics rather than intermittent as is seen today (Ravelo and Wara, 2004; Fedorov et al., 2006). Under such
127 conditions, regional climate in the southeastern U.S., including Florida an d Georgia, would have been characterized by cooler temperatures, abov e average rainfall, a reduction in hurricane activity, and an increase in winter storm activity (cyclogenesis). The latter of these factors would have contributed to an overall increase in nearsh ore erosion and transport during this interval, but the relative climate stability during this time woul d have reduced the delivery of siliciclastics to the coast. With onset of the climatic instability associated with the transition toward significant Northern Hemisphere Glaciation th at occurred after ~3.0 Ma, the siliciclastic fl ux to the coast would have increased as part of the global increase in sediment ation rates and grain sizes noted by Zhang et al. (2001) for the middle to late Pliocene (~2 million years ago). Zhang et al. (2001) and Molnar (2004) postulate that an over all increase in the amplitude and frequency of climate change is the root caus e for the increase in both sedimentation rates and sediment coarsening observed during the Pliocen e as a result of a persistent state of disequilibrium in sediment source regions. Sinc e the response time of landscapes dominated by fluvial incision is longer than th e periodicity of major climate fl uctuations during this interval (Hancock and Kirwan, 2007), upland sediment sour ces remain in disequilibrium, driving the increased siliciclastic flux toward the coast. In creased fluvial incision ra tes are likely to be magnified during regressive peri ods of base-level lowering (Mills, 2000), which coincide with overall late Miocene through Pliocene cooling. Along the ea stern North America margin, sedimentation rates are known to have doubled be tween the late Miocene and the Quaternary in offshore basins (Poag and Sevon, 1989), with Ma tmon et al. (2003) indicating that basinaveraged erosion rates of ~30 m/m.y. are common in the southern Appalachians during this time. Evidence for increased current and storm activity Coincident with deposition of the Cypresshead Formation, multiple climate models suggest that higher annual sea-surface temperatures and re duced ice cover in the Northern Hemisphere
128 likely led to intensification of the Icelandic lo w-pressure system and the Azores high-pressure system (Dowsett et al., 1994; Haywood et al., 2 000; Jiang et al., 2005). As a consequence, corresponding strengthening of we sterly wind velocities and wind st ress over the North Atlantic Ocean would have enhanced the flow of the Gu lf Stream, with evidence of enhanced oceanic upwelling along the east coast of North America and southwestern Florida during this time (Cronin and Dowsett, 1990; Cr onin, 1991) consistent with this result. Additionally, intensification of the Azores high likely resulted in a strengthening of easterly Trade Winds and increased tropical storm activity, with the later resulting in the Florida coast experiencing more hurricanes than present (Hobgood and Cerveny, 1988; Barron, 1989). Storm activity associated with either hurricanes or wint er storms, perhaps at increased levels of intensity and/or occurrence, has been used to explain the concentra tion of Pliocene shell be ds associated with the Pinecrest Sand (Allmon, 1992), a time equivalent of the Cypresshead Formation in Florida. With a proposed increase in th e strength of prevailing winds impacting the east coast of Florida, conditions favorable to a substantial in crease in longshore tran sport likely persisted during the Pliocene. This contrast s with modern conditions along the east coast of Florida which experience a mean significant wave height during winter months of only 1.2 m and a relatively low frequency of high-intensity storms consiste nt with a microtidal, wave-dominated coast (Davis et al., 1992). Although these conditions ar e capable of a relatively high gross longshore transport (~600,000 m3/yr), there is relatively lo w net transport (50,000-150,000 m3/yr) along the east coast of Florida due to the seasonally bimodal wave climate that impacts the coastline (Walton, 1976; Davis et al., 1992). During late Miocene through Pliocene warming events, intensified wave, storm, and fr ontal activity may have acted as major forcing mechanisms for increased coast parallel siliciclastic fluxes, with increased mean wave height and stronger
129 longshore transport acting to mobilized and tr ansport significant volumes of siliciclastics concentrated by fluvial processes along the sout heastern Georgia coast. Such conditions would not only explain the quantity of siliciclastics deposited on the Florida Platform, but also the relative coarseness of these sedi ments, including discoid pebbles and the common occurrence of storm generated sedimentary features. Conclusions Focusing on the stratigraphy and sedimentology of the Cypresshead Formation, the results outlined in this chapter highlight significant observations which clarify the nature, timing and significance of siliciclastic deposition impacting the Florida Platform during the Late Miocene through Pliocene. Among the results are th e following: Cypresshead sediments were deposited in a near shore marine environment, most likely in a strand plain setting, as two distinct progradational shor eface-shelf parasequences. Cypresshead facies define coarsening-upward sequences consis tent with a wave-dominated environment in north-central Florida and a mixed energy environment in southeastern Georgia. Deposition of the Cypresshead took place in re sponse to sea-level falls at 3.3 Ma and 2.5 Ma as a consequence of the interplay of sea-level, sediment supply and accommodation. Deposition in Florida at 3.4.8 Ma with reworking at 2.8.8 Ma. Deposition in Georgia at 3.15.3 Ma with reworking at 2.3.8 Ma. Timing of Cypresshead deposition at 3.4.3 Ma during the Late Plio cene (Piacenzian to early Gelsian) correlates w ith age estimates of the Citronelle Formation (3.4.7 Ma) as defined by Otvos (1988b) and timing of siliciclastic deposition associated with the Tamiami Formation (SS3/Pinecrest Sands). Viewed collectively with the Late Miocene SS 2 siliciclastics of Cunningham et al. (2003), Cypresshead and associated siliciclastics de fine a retrogradational parasequence which was deposited on the Florida Pl atform over a 6.8 Ma period. Cypresshead deposition correlates with a pale oclimate transition from continual El Nio conditions associated with the Pliocene warm period (~4.5.0 Ma) to conditions associated with the onset of significant Northern Hemisphere Glaciation (NHG) (~3.0.5 Ma).
130 The driving mechanism behind the anomalous accumulation of siliciclastics associated with the Cypresshead and related Late Miocene siliciclastics (SS2) is the shift from periods of climate stability to periods of climate transition (instability) characterized by changes in temperature, precip itation and vegetation.
131 CHAPTER 5 EVIDENCE FOR NEOFORMATION AND RE CRYSTALLIZATION OF KAOLINITE IN THE CYPRESSHEAD FORMATION Introduction As previously noted in this study, the ka olinitic clays present in the Cypresshead Formation occur as irregular thin beds, lenses, an d stringers of clay, or as a binding matrix for sands and gravels, with clay contents that may vary from absent to > 50% (Pirkle, 1960; Kane, 1984; Huddlestun, 1988; Scott, 1988 a). Although much of the ka olinitic clay possesses obvious sedimentological characteristics indicative of a detrital or sedime ntary origin, discrepancies in grain-size relationships (i.e. clays with coarse sands and grav els) and overall distribution of clay in the formation continue to raise questions regarding the potential in situ or weathering origin for some portion of the clay mineral assemblage. While a detrital origin for much of the clay content in Cypresshead Formation sediments is li kely, Austin (1998) has suggested an important role for groundwater leaching of aluminous components (mica, feldspar) in the formation of an in situ kaolinite fraction, with the associated removal or preservation of Fe and organic compounds dependent on groundwater acidity, re dox, and biological content. This view is similar to that proposed for Georgia-South Caro lina kaolins by Hurst and Pickering (1997), and reflects the significant role that post-depositional alteration plays in maturing kaolin assemblages through a complex series of early diagenetic a nd weathering reactions that increase overall kaolinite content of sediments over time. Arguments supporting the in situ formation of the kaolinitic sediments of the Cypresshead Formation were first introduced by Sellards (1912), who considered the kaolinite to have formed by the weathering of arkosic sands. The main ai m of this proposed mechanism for kaolinite origin was to explain the presence of fine clay mixed with coarse sand and gravel, evidence of which is noted by the high positive skewness values reported for Cypresshead sediments in
132 Chapter 4 of this study. However, Sellards ( 1912) model still does not fully explain the distribution of all fines mixed with coarse sand and gravel, particularly as relates to the occurrence of muscovite throughout these sediments, a considerable quantity of which has been identified in the 44 to 10 m range (Pirkle, 1960) As a result, this study seeks to define the detailed role of weathering reac tions in the overall modification of Cypresshead clays, and the likelihood for an in situ kaolinite fraction in respons e to both transformational and neoformational weathering processe s in order to constrain observe d grain-size anomalies. The reader is referred to Chapter 2 of this study for a more detailed review of the previous arguments associated with the detrital versus in situ origin of Cypresshead Formation clays. Results X-ray diffraction (XRD) analysis of clay (< 2 m) separates obtained from the Cypresshead Formation in both Florida and Georgia (Fig. 3-1) was performed in order to fully characterize the detailed clay mineralogy of the formation, something which has already been done for Georgia-South Carolina ka olin district kaolinites (Kelle r, 1977; 1978; Hurst et al., 1979; Hassanipak and Eslinger, 1985; Pi ckering and Hurst, 1989; and ot hers). Both oriented and randomly mounted clay mineral samples were pr epared and analyzed in order to define qualitative sample mineralogy and to characteri ze kaolinite disorder and crystallite size characteristics. Scanning electron microscope (SEM) analysis was aimed at discerning microtextural characteristics of kaolinites and as sociated sediments in order to facilitate recognition of kaolinite microm orphologies of detrital and in situ origin. XRD data for the samples used in this study (oriented and random) are included with Appendices E and F. Mineralogy A table outlining the clay mineralogy of C ypresshead Formation samples is included in Chapter 4 for reference (Table 4-2). Based on this data, kaolinite is the do minant clay mineral in
133 both Cypresshead Formation and re worked Cypresshead Formation sediments. In Florida, the clay (< 2 m) size-fraction also includes le ssor amounts of quartz, gibbsite, halloysite, metahalloysite (inferred), hydroxyl-interlayered ve rmiculite (HIV), and cr andallite-florencite, with additional occu rrences of rutile, anatase, boehmite, and diaspore. Georgia clay mineralogy differs slightly, with the absence of halloysite and crandallite, and the addition of significant illite at both the Jesup type locality and the Birds site. Based on eval uation of the 060 reflection, this illite is of the 2M1 polytype and therefore of detrital orig in. Goethite is also a common trace component in both Jesup and Birds samples, a nd reflects higher iron (Fe) concentrations associated with the deposition of these sediments. Clay mineralogical trends of note in north -central Florida include the overlapping occurrence of gibbsite and halloys ite at the near surface. Forming in proximity to fluctuating water table conditions, gibbsite (above) and halloysite (below) likely crystallize at the expense of allophane and amorphous Al hydroxide gels. Gibb site occurs as either a trace or minor constituent in near surface C ypresshead sediments, with the formation of gibbsite favored by well drained soil conditions consiste nt with what is observed with the Entisols and Ultisols of the Cypresshead (Heuberger, 1995). Where present, the phase is commonly concen trated in the near surface zone of pedogenic clay accumulation via illuviation. For north-central Florida sampling locations, gibbsite is notably absent in th e Goldhead Sand Mine and Joshua Sand Mine exposures. In Georgia, gibbsite is absent only from the Birds locality. The occurrence of halloysite in Cypresshead sedi ments, noted as a trace clay mineral phase in north-central Florida samples, was confirme d by XRD and SEM (Figs. 5-1 and 5-2). Based on heat treatment testing for halloysite loosely bound interlayer water is lost, resulting in structural collapse of the basal ( d001) spacing from 10 to 7 consis tent with the formation of
134 Figure 5-1. XRD patterns of example Cypressh ead Formation clays. A) Sample FRG-1-5 containing halloysite which is partially co llapsed when air dried, B) Sample EPK36J-12 (25-27) containing hall oysite which requires completed heat treatment to collapse structure to 7 basal spacing, C) Comparison of north-central Florida (EPJ36-J-12 (48-50)) and southeastern Georgia (L-1-6) kaolinite, illustrating significant differences in disorder, D) Vertic al variation in kao linite disorder (G = gibbsite; Q = quartz; C = crandallite-florencite). metahalloysite (halloysite-(7 )) as defined by Hu rst and Pickering (1997). Behavior related to dehydration varied, with some samples exhibiting water loss with air drying (Fig 5-1A), while others required full-cycle heating to fully collapse the structure (F ig. 5-1B). The distribution of halloysite was restricted to near surface samples in close proximity to the modern water table, with water table levels closer to surface elevation at the EPK Mine site than what is observed at the Grandin Sand Mine. Halloysite morphology in Cypresshead sediments is similar with that observed in other southeastern U.S. kaolin deposits, possessing a tubular morphology consistent with formation
135 Figure 5-2. SEM photomicrographs illustrating ch aracteristic secondary weathering phases and textures. A) halloysite (EPK30V-6 (22-24)), B) crandallite-florencite (FRG-1-5), C) etched quartz (EPK30-V-6 (48-53)), D) sk eletal K-feldspar grain (SSJ-1-11). under low Fe conditions (Joussein et al., 2005). Tubes are generally 1 m in length and 0.2 m in diameter, and appear to be crys tallized on the basal surface of weathered muscovite mica (Fig. 52A), a common association for halloysite neof ormation (Robertson and Eggleston, 1991; Singh and Gilkes, 1992). Consistent with the ubiquit ous character of HIV in sandy soils along the southern U.S. coastal plain (Harris et al., 1992), HIV occurrences in the Cypresshead Formation are restricted to the near surface, and apparently above recent water table levels. Additi onally where noted as a minor rather than trace phase (J-1), HIV concentr ation is at the top of the stratigraphic section,
136 suggesting that the phase is prefer entially concentrated in upper sec tions of the unit most effected by pedogenic processes. As noted by Harris et al. (1992) and Heuberg er (1995), HIV forms preferentially in the soil profile from sandand si lt-sized mica precursor grains which retain their position in the profile while colloidal component s such as kaolinite illuviate to deeper weathering zones. Thus, HIV appears to be the preferred vadose weathering product for the near surface detrital mica component of the Cypresshead Formation. The occurrence of a crandallite -florencite phase ((Ca,REE)Al3(PO4)2(OH)5H2O) was limited to Cypresshead and reworked Cypresshead sediments in north-central Florida. Known to occur in near surface sediments of central Florida as a secondary weathering product of phosphatic (francolite-bearing) clayey sediment s (Blanchard, 1972), cran dallite-florencite is primarily associated with near surface Cypre sshead samples, except for the Grandin Sand Mine exposures (FRG-1 and FRG-2), wh ere crandallite is also noted at depth. Of note, crandalliteflorencite was identified near both the top and bottom of FRG-2. Crandalliteflorencite observed in Cypresshead samples possesses a lath or ac icular morphology (Fig. 5-2B) commonly noted in weathered phosphorites (F licoteaux and Lucas, 1984). Other phases of interest, which are not neces sarily included with the clay fraction, are quartz, mica, and K-feldspar. Sand-size quartz, the dominant lithic component associated with Cypresshead sediments varies in grain shap e from angular to subrounded, and exhibits significant evidence of etching when viewed via SEM (Fig. 5-2C). This indicates extreme leaching conditions in Cypresshead sediments in response to weathering, and corroborates the evidence for silica mobilization, a factor previous ly indicated by the occu rrence of gibbsite in near surface samples. Clay-size quartz also occurs in Cypresshead sediments, but was limited to north-central Florida sites, with EPK30-V-6 exhibiting the most si gnificant concentrations to a
137 depth of ~ 16 m (53 ft). In other instances, cl ay-size quartz was isolat ed to near surface occurrences as has been noted previously fo r the Cypresshead in Florida (Heuberger, 1995). Feldspars, confirmed as K-feldspar (microcline) via energy dispersive X-ray spectroscopy (EDS), petrographic examination, and XRD, were found to occur at both Florida and Georgia sampling locations. The distribution of K-feldspar is addressed in Chapter 4, and appears to be most significantly impacted by preservation poten tial in relation to weathering. When evaluated by SEM, K-feldspars were found to possess skelet al microtextures consistent with significant dissolution activity (Fig. 5-2D), with no evidence for pseudomor phic replacement of feldspar grains by kaolinite. Kaolinite Disorder and Crystallite Size Random clay mineral aggregates were prepared and analyzed by XRD in order to define kaolinite disorder using computer-based methods to describe the diso rder occurring in the kaolinite structure (i.e. enantiomo rphic distortions and displacement of octahedral Al vacancies), and to interpret both qualitatively and quantitativ ely the total density of stacking faults in the crystallite ensemble (Hinckley, 1963; Bish a nd Von Dreele, 1989; Plan on and Zacharie, 1990; Artioli et al., 1995). To accomplish this, the Hinkley Index (HI), the Lit ard Index (R2), and the expert system of Planon and Zacharie (1990) we re calculated (Table 5-1). The HI is the most widely used of these measurements, and evaluates changes in the 02 l and 11 l peaks of kaolinite (20-30 using Cu K ), which are sensitive to random and specific interlayer displacements of type b/3 (Aparicio and Galn, 1999). Normal HI va lues range from <0.5 (disordered) to 1.5 (ordered). The R2, on the other ha nd, measures changes in the 13 l peak sequence of kaolinite (37-40 using Cu K ), which are affected by random di splacements (Cases et al., 1982; Aparicio and Galn, 1999). Reported R2 values nor mally range from <0.7 (disordered) to 1.2
138 Table 5-1. Results of disorder calculations for north-central Fl orida Cypresshead and reworked Cypresshead kaolinite. # PhasesMWcdp%ldp Cypresshead Formation FL EPK36-J-1225-27H/Q/G0.310.781280.000.040.35 EPK36-J-1227-30H/Q/G0.860.902 30.66 EPK36-J-1235-40H/Q/G0.690.852 21.94 EPK36-J-1240-44 1.160.942 46.21 EPK36-J-1244-46 1.170.912 47.22 EPK36-J-1246-48 1.100.902 45.08 EPK36-J-1248-50 1.170.902 42.78 EPK36-J-1250-53 1.160.892 43.79 EPK36-J-1253-56 1.130.852 42.97 EPK36-J-1256-59 1.050.822 38.68 EPK36-J-1259-62 1.070.812 38.54 EPK31-P-4027-35 1.020.882 38.61 EPK31-P-4035-45 1.200.932 47.20 EPK31-P-4045-50 1.230.892 49.77 EPK31-P-4050-62 0.970.782 38.30 EPK31-P-4062-65 0.640.652 21.73 EPK30-V-616-22H/Q ----------------------EPK30-V-622-24H/Q ----------------------EPK30-V-624-27Q ----------------------EPK30-V-630-35H/Q/G0.55--------------------EPK30-V-635-39Q/G0.340.681300.000.040.35 EPK30-V-639-43Q 0.510.732 10.79 EPK30-V-643-48Q 0.560.781380.000.030.24 EPK30-V-648-53 0.540.752 14.49 EPK30-V-653-58 1.240.782 47.19 EPK30-V-658-63 1.300.932 51.43 EPK30-V-663-68 1.360.942 55.16 EPK30-V-668-73 1.230.702 49.93 EPK30-V-673-78 0.910.712 42.42 FRG-1 1 G 0.540.892 14.67 FRG-1 2HIV/G0.56---2 14.04 FRG-1 3HIV/G0.560.851340.000.030.29 FRG-1 4HIV/G0.64---1270.000.040.27 FRG-1 5 H/G0.230.74-----------------FRG-1 6 1.190.991450.000.010.08 FRG-1 7 1.130.892 43.06 FRG-1 8 0.620.772 19.79 FRG-1 9 1.271.022 54.48 FRG-1 10 1.190.872 47.21 FRG-1 11 1.451.042 65.46 FRG-1 12 1.511.092 69.11 FRG-1 13 1.561.072 67.25 FRG-1 14 1.561.062 69.71 FRG-1 15 1.491.052 62.04 FRG-2 1HIV/G0.520.732 10.57 FRG-2 2HIV/G1.031.011410.000.020.12 FRG-2 3 1.390.972 56.33 FRG-2 4 1.340.952 56.60 FRG-2 5 1.491.032 62.96 FRG-2 6 1.491.022 65.25 FRG-2 7 1.491.042 65.46 FRG-2 8 1.511.102 67.11 FRG-2 9 1.431.002 62.68 FRG-2 10 1.320.972 54.86 FRG-2 11 1.200.882 48.23 FRG-2 12 0.990.842 34.97 FRG-2 13 0.920.792 10.96 Litard Index (R2) Expert System Parameters Sample ID Interval Mineral Interferences Hinkley Index (HI) Note: For expert system calculations, %ldp = percentage of low-defect kaolinite, M = average layer number in the coherent domains along the c -axis, Wc = percentage of layers with vacant octahedral position, = small random translations between adjacent layers, an d p = great translation defects between adjacent layers. Mineral interferences include: quartz (Q), halloysite (H), hydroxyl-interlayered vermiculite (HIV), gibbsite (G), and anatase (A).
139 Table 5-1. (continued). # PhasesMWcdp%ldp Cypresshead Formation FL (cont.) FRL-111.010.82236.55 FRL-1 2 1.100.852 41.56 FRL-1 3 1.110.862 38.37 FRL-1 4 1.080.862 40.07 FRL-1 5 1.230.962 44.46 FRL-1 6 1.140.882 46.60 FRL-1 7 1.150.852 44.93 FRL-1 8 1.110.822 44.57 FRL-1 9 1.000.822 36.52 SSJ-1 1 0.820.672 13.50 SSJ-1 2 0.860.752 26.25 SSJ-1 3 0.940.832 31.90 SSJ-1 4 1.130.932 43.27 SSJ-1 5 1.280.992 49.34 SSJ-1 6 1.200.842 49.22 SSJ-1 7 1.260.892 51.45 SSJ-1 8 1.240.892 50.74 SSJ-1 9 1.240.912 50.12 SSJ-1 10 1.130.892 45.58 SSJ-1 11 1.060.822 41.45 Reworked Cypresshead Formation FL SSD-1 1Q/HIV/G/A----------------------SSD-1 2 Q/G0.490.88-----------------SSD-1 3 G 0.680.882 33.24 SSD-1 4 G 0.690.912 34.32 SSD-1 5 G 1.130.882 46.42 SSD-1 6 1.080.742 41.18 SSD-1 7 1.060.672 40.87 SSD-1 8 0.860.632 30.00 SSD-1 9 0.610.552 14.01 SSD-1 10 0.460.501210.000.050.31 Cypresshead Formation GA J-1 1HIV/G0.69--------------------J-1 2HIV/G0.42--------------------J-1 3HIV/G0.50--------------------J-1 4 I 0.290.51-----------------J-1 5 I 0.430.53-----------------J-1 6 I 0.240.46-----------------LB-1 1HIV/G0.180.45-----------------LB-1 2HIV/G0.150.45-----------------LB-1 3 G 0.290.46-----------------LB-1 4 0.350.461210.000.050.40 LB-1 5 0.430.551240.000.050.32 LB-1 6 0.280.46-----------------LB-1 7 0.380.581250.000.050.35 B-1 1 HIV/I----------------------B-1 2 HIV/I----------------------B-1 3 HIV/I----------------------B-1 4 I ----------------------B-1 5 I ----------------------Litard Index (R2) Expert System Parameters Sample ID Interval Mineral Interferences Hinkley Index (HI) Note: For expert system calculations, %ldp = percentage of low-defect kaolinite, M = average layer number in the coherent domains along the c -axis, Wc = percentage of layers with vacant octahedral position, = small random translations between adjacent layers, an d p = great translation defects between adjacent layers. Mineral interferences include: quartz (Q), halloysite (H), hydroxyl-interlayered vermiculite (HIV), gibbsite (G), and anatase (A). (ordered). An alternative to these methods, the expert system of Plan on and Zacharie (1990), uses multiple measurements from the XRD pattern to describe kaolinite defects, and provides an abundance estimate of translation defects and the potential for a two-phase kaolinite mixture.
140 Accessory phases encountered with Cypresshead samples which are capable of interfering with disorder calculations include halloysite, quart z, gibbsite, HIV, and anatase (Table 5-1). Based on XRD results, significant differences exis t between the disorder characteristics of north-central Florida and Georgi a Cypresshead kaolinites (Fig. 51C; Table 5-2), with Florida Cypresshead samples consistent w ith ordered (low-defect) kaolinite similar to what is seen with the Cretaceous soft kaolins form Georgia, and Georgia Cypresshead samples exhibiting disordered (high-defect) characteristics. Fo r Florida samples, including SSD-1 reworked Cypresshead sediments, HI and R2 values range from 0.23 to 1.56 and 0.50 to 1.10, respectively, with corresponding means of 1.03 and 0.86. For Georgi a samples, HI and R2 values range from 0.15 to 0.69 and 0.45 to 0.58, respectively, with m eans of 0.36 and 0.49. In the case of the distribution of HI values illustrated in Figure 53A, Florida samples exhibit a bimodal character, with one group of HI values corresponding to di sordered kaolinite values of <0.7, and a second group possessing moderately to well ordered HI values >0.8. Geor gia samples do not exhibit this characteristic, but rather are rest ricted to disordered HI values <0.7. R2 results, believed to be sensitive to random defects only (Cases et al ., 1982), do not exhibit a bimodal character for either the Florida or Georgia samples (Fig. 5-3B). Box-and-whisker plots of both the HI (Fig. 54A) and R2 (Fig. 5-4B) results for Florida and Georgia Cypresshead samples illustrate several notable trends. First, the Florida sample sites exhibit consistency in the median values of both HI and R2, except for SSD-1 which possesses a lower median value than the othe rs. Additionally, the HI and R2 results for the EPK, FRG, and SSD-1 localities exhibit a greate r spread in data than what is observed for FRL-1 and SSJ-1 samples. This greater variation in both HI and R2 correlates to a consistent vertical trend noted for these sites, where kaolinite order increases with depth from the surface (Fig. 5-1D). In most
141 Table 5-2. Statistical summary of kaolinit e order and crystallite size calculations. Locality ID/ Statistic Hinkley Index (HI) Litard Index (R2) Area-wt Mean Thickness (nm) Volume-wt Mean Thickness (nm) EPK Max1.360.947.211.0 Min0.310.652.94.7 Mean0.940.835.58.9 FRG Max1.561.108.2 11.4 Min0.230.732.7 4.2 Mean1.130.955.7 8.8 FRL-1 Max1.230.9610.0 13.6 Min1.000.826.6 10.1 Mean1.100.867.8 11.1 SSJ-1 Max1.280.998.1 11.0 Min0.820.676.3 9.4 Mean1.100.867.4 10.5 SSD-1 Max1.130.917.2 9.6 Min0.460.502.8 4.7 Mean0.780.745.0 7.5 J-1 Max0.690.536.6 8.9 Min0.240.465.4 7.5 Mean0.430.506.2 8.4 L-1 Max0.430.586.9 9.3 Min0.150.454.5 7.1 Mean0.300.496.2 8.5 B-1 Max-----4.4 6.4 Min-----4.0 5.8 Mean-----4.2 6.0 instances, however, kaolinites appear to reverse this trend near the base of each section studied, becoming more disordered prior to reaching the basal su rface of the formation (Table 5-1). Even for the Florida sample sites which do not exhibit the near-surface occurrence of disordered kaolinite (FRL-1 and SSJ-1), the la tter observation holds true, as in both cases HI and R2 values exhibit a notable decrease in value near the ba se of the section. Excepting the EPK data, the Florida sample sites with the greatest variability in HI and R2 results correlate to a lower average
142 Figure 5-3. Histograms illustrating the distributi on of results for both kaolinite order (HI and R2) and crystallite size calculations. A) Hinkley Index (HI), B) Litard Index (R2), C) area-weighted thickness, D) volume-weighted thickness. clay content (FRG = 7.7% and SSD-1 = 4.9%) than t hose with little variabili ty in results (FRL-1 = 10.4% and SSJ-1 = 10.8%) (Appendix D). Whereas clay content at the EPK site is more concentrated toward the base of the formation, clay content noted for both FRL-1 and SSJ-1 is highest near the top of the formation, likely impacting vertical in filtration rates and consequently leaching/weathering. Box-and-whisker plots for the Georgia sample sites, for which HI and R2 results were measurable, exhibit consistently low HI (Fig. 5-4A) and R2 (Fig. 5-4B) median values, with little variation in results. A review of the results obtained via the expert system of Pl anon and Zacharie (1990) indicate that Florida Cypresshead kaolinite, incl uding SSD-1, is primarily a two-phase (at least) mixture, achieving a high relative percentage (~38-69%) of well-ordered (l ow defect) kaolinite
143 Figure 5-4. Box-and-whisker diag rams for kaolinite disorder (H I and R2) and crystallite size calculations. (%ldp) in the middle of the sect ions evaluated by this study. Howe ver, as with the HI and R2 results, the percentage of well-ordered kaolinite for two-phase mixt ures decreases at both the top (11-32%) and bottom (~11-42%) of most sections For the Florida sect ions exhibiting the greatest variability in HI and R2 values (EPK and FRG), single-phase kaolinite is common at the top of the sections consistent with a disordered phase exhibiti ng a moderate concentration of translation defects of probability p (Table 5-1). Octahedral C -site vacancies with an abundance of Wc are absent. Additionally, (in fraction of unit-cell), wh ich corresponds to stacking disorders, also appears to be a minimal component of the disorder in th ese kaolinites. One basal
144 sample (SSD-1-10) was also determined to be a single-phase kaolinite si milar in character to those described from the near surface of EPK an d FRG sections. In general, results from the expert system correlated well with HI results, in dicating a general vertical trend of increased kaolinite order with depth from the surface follo wed by degradation in order near the base of most sections (Table 5-1). As for Cypresshead samples from southeastern Georgia, only three collected from the Birds reference locality could be evaluated by the expert system due to the highly disordered (high defect) character of th e kaolinite. In each case, the results indicate a single-phase kaolinite similar in defect characteristics to those not ed from Florida sections where translation defects (p = 0.32-0.40) appear to be responsible for the high degree of disorder noted for these samples (Table 5-1). Crystallite size, or rather the coherent scattering domain (CSD) size, determined for Cypresshead Formation kaolinite yields geologic al information related to weathering processes which have impacted the unit (Table 5-3). For ka olinite, CSD values measured for the 001 peak identifies the crystallite dimens ion perpendicular to the ab plane of the unit cell such that crystallite size is equal to (N-1) d(hkl), where N is the number of hkl planes responsible for the reflection. CSD values were collected for all Ge orgia samples, as analysis are based on the diffraction characteristics of the basal d(001) reflection and does not suffer from the same mineral phase interferences encountered with disorder calculations. Best mean thickness values determined graphically v ia the Warren-Averbach method (1953) are included for reference, while the area-weighted mean thickness and vol ume-weighted mean thickness values calculated using the modified Bertaut-Warren-Averbach method employed by the MudMaster computer program of Eberl et al. (1996) were the focu s of evaluation (Table 5-3). The values of and 2 define the theoretical lognormal distribution for crystallite size fitted to the measured data
145 Table 5-3. Results of crystallite size calcu lations for north-central Florida and Georgia Cypresshead and reworked Cypresshead kaolinite. Sample ID Interval Mineral Interferences Best Mean Thickness (nm)aArea-wt Mean Thickness (nm)b 2Volume-wt Mean Thickness (nm)cDistribution Limit Cypresshead Formation FL EPK36-J-1225-27H 2.9 2.90.720.606.5 20 EPK36-J-1227-30H 4.5 4.51.080.829.2 23 EPK36-J-1235-40H 4.3 4.31.070.768.7 22 EPK36-J-1240-44 5.6 5.71.340.8210.3 23 EPK36-J-1244-46 6.1 6.11.490.689.9 21 EPK36-J-1246-48 5.6 5.81.410.749.9 22 EPK36-J-1248-50 6.3 6.31.510.7310.2 22 EPK36-J-1250-53 6.5 6.41.550.6910.0 20 EPK36-J-1253-56 6.6 6.61.590.6610.2 21 EPK36-J-1256-59 6.8 6.91.660.6110.1 20 EPK36-J-1259-62 6.7 6.91.690.5510.0 20 EPK31-P-4027-35 6.0 5.91.490.629.3 20 EPK31-P-4035-45 7.2 7.21.660.7111.0 22 EPK31-P-4045-50 6.4 6.61.610.629.8 20 EPK31-P-4050-62 6.8 6.91.730.439.4 20 EPK31-P-4062-65 5.6 6.31.640.439.0 22 EPK30-V-616-22H 2.8 3.00.860.454.7 15 EPK30-V-622-24H 2.9 2.90.800.524.9 15 EPK30-V-624-27 -------------EPK30-V-630-35H 4.3 220.127.116.117.2 17 EPK30-V-635-39 3.1 3.30.890.546.0 20 EPK30-V-639-43 3.6 3.81.030.597.1 20 EPK30-V-643-48 4.2 18.104.22.1687.7 20 EPK30-V-648-53 4.2 22.214.171.1247.3 19 EPK30-V-653-58 6.0 6.21.450.7910.7 25 EPK30-V-658-63 6.6 6.71.590.7210.5 22 EPK30-V-663-68 6.3 6.31.530.7110.0 21 EPK30-V-668-73 6.9 6.91.670.599.9 20 EPK30-V-673-78 5.8 6.21.580.509.2 21 FRG-1 1 3.5 3.41.020.435.1 12 FRG-1 2HIV 3.6 126.96.36.1995.5 13 FRG-1 3HIV 4.3 4.51.260.496.7 15 FRG-1 4HIV 3.8 188.8.131.526.4 16 FRG-1 5 H 2.5 2.70.780.404.2 14 FRG-1 6 4.5 184.108.40.2067.5 17 FRG-1 7 4.8 220.127.116.117.3 16 FRG-1 8 3.7 3.91.090.516.9 20 FRG-1 9 5.2 5.11.320.628.4 20 FRG-1 10 6.3 6.31.600.549.1 18 FRG-1 11 7.1 6.71.560.7810.7 21 FRG-1 12 7.6 6.91.590.8010.7 20 FRG-1 13 6.2 6.01.390.8310.6 23 FRG-1 14 6.5 5.81.380.809.9 20 FRG-1 15 6.3 5.81.390.779.7 20 FRG-2 1HIV 3.8 18.104.22.1685.8 13 FRG-2 2HIV 4.2 22.214.171.1247.0 17 FRG-2 3 6.8 6.61.550.7610.4 20 FRG-2 4 5.8 5.61.370.749.5 20 FRG-2 5 7.7 7.01.620.7510.5 19 FRG-2 6 8.7 8.21.820.6811.4 20 FRG-2 7 7.9 7.21.630.7910.9 20 FRG-2 8 7.5 6.91.610.7510.6 20 FRG-2 9 7.6 7.51.710.7211.2 22 FRG-2 10 8.4 8.11.870.5010.8 20 FRG-2 11 7.8 7.81.810.5711.1 22 FRG-2 12 7.5 7.51.810.4510.1 20 FRG-2 13 5.1 5.61.480.518.5 20 Note: Crystallite size analyses based on the Be rtaut-Warren-Averbach method. The parameters and 2 define the theoretical lognormal distribution for crystallite size fitte d to the measured data. Mineral interferences include: halloysite (H), hydroxyinterlayed vermiculite (HIV), and illite (I). a Determined by extrapolation of a line along the steepest slope of a plot of the Fourier coefficient of the K interference function H(n) verses n, where n = the number of data points divided by 2. b Calculated as the area-weighted mean of a plot of crystallite size normalized to the frequency of occurrence. c Calculated as the volume-weighted mean of a plot of crystallite size normalized to the frequency of occurrence.
146 Table 5-3. (continued). Sample ID Interval Mineral Interference Best Mean Thickness (nm)aArea-wt Mean Thickness (nm)b 2Volume-wt Mean Thickness (nm)cDistribution Limit Cypresshead Formation FL (cont.) FRL-1 1 10.1 10.02.090.4813.6 30 FRL-1 2 7.1 6.81.630.6410.1 20 FRL-1 3 7.5 7.31.730.6110.6 20 FRL-1 4 9.0 9.01.970.5312.4 24 FRL-1 5 6.7 6.61.540.7610.7 22 FRL-1 6 7.2 7.31.750.5310.3 20 FRL-1 7 7.4 7.51.790.5110.3 20 FRL-1 8 6.8 7.41.730.6111.0 23 FRL-1 9 7.8 8.01.870.5010.9 21 SSJ-1 1 8.0 8.01.880.4410.7 22 SSJ-1 2 7.5 7.61.820.4410.3 21 SSJ-1 3 6.5 6.61.660.499.4 20 SSJ-1 4 7.0 7.01.680.6110.4 22 SSJ-1 5 6.4 6.31.520.699.8 20 SSJ-1 6 7.1 7.31.710.6410.9 22 SSJ-1 7 7.4 7.51.750.6310.9 21 SSJ-1 8 7.2 7.41.730.6110.8 22 SSJ-1 9 7.2 7.31.730.5810.6 21 SSJ-1 10 7.6 7.91.840.5011.0 22 SSJ-1 11 8.0 8.11.890.4510.9 22 Reworked Cypresshead Formation FL SSD-1 1 2.9 2.80.800.474.7 15 SSD-1 2 3.1 3.10.850.535.6 17 SSD-1 3 3.4 3.40.930.586.2 16 SSD-1 4 4.0 3.91.050.647.0 17 SSD-1 5 4.3 126.96.36.1997.4 17 SSD-1 6 6.4 6.91.750.429.2 18 SSD-1 7 7.0 7.21.780.439.6 19 SSD-1 8 6.6 7.01.780.409.6 23 SSD-1 9 5.5 6.01.610.378.1 19 SSD-1 10 5.3 5.81.590.337.6 18 Cypresshead Formation GA J-1 1 4.9 5.41.470.447.5 16 J-1 2 6.1 6.31.680.378.3 17 J-1 3 5.9 6.41.670.398.5 18 J-1 4 I 5.7 6.31.630.458.9 20 J-1 5 I 5.6 6.11.630.398.4 19 J-1 6 I 6.4 6.61.710.388.9 20 LB-1 1 6.1 6.41.690.368.5 18 LB-1 2 6.6 6.81.760.348.9 20 LB-1 3 6.5 6.71.730.378.9 20 LB-1 4 6.7 6.91.770.369.3 22 LB-1 5 4.8 5.31.420.518.0 18 LB-1 6 6.3 6.51.690.388.8 20 LB-1 7 4.1 4.51.260.507.1 18 B-1 1 I 3.5 188.8.131.525.8 17 B-1 2 I 3.4 4.01.200.405.8 16 B-1 3 I 3.4 4.01.150.465.9 15 B-1 4 I 3.9 4.41.280.436.4 16 B-1 5 I 3.7 4.31.260.416.3 17 Note: Crystallite size analyses based on the Be rtaut-Warren-Averbach method. The parameters and 2 define the theoretical lognormal distribution for crystallite size fitte d to the measured data. Mineral interferences include: halloysite (H), hydroxyinterlayed vermiculite (HIV), and illite (I). a Determined by extrapolation of a line along the steepest slope of a plot of the Fourier coefficient of the K interference function H(n) verses n, where n = the number of data points divided by 2. b Calculated as the area-weighted mean of a plot of crystallite size normalized to the frequency of occurrence. c Calculated as the volume-weighted mean of a plot of crystallite size normalized to the frequency of occurrence.
147 corresponding to the area-weighted mean thickne ss curve. Values related to instrumental broadening and the LpG2 factor used in these calculations were derived from standard kaolinite values included with the MudMaster program. For CSD calculations of Florida samples, including SSD-1 reworked Cypresshead sediments, the area-weighted mean and volume-weighted mean values range from 2.7 nm to 10.0 nm and 4.2 nm to 13.6 nm, respectively, with corresponding means of 6.0 nm and 9.1 nm. For Georgia samples, the area-weighted mean and volume-weighted mean values range from 4.0 nm to 6.9 nm and 5.8 nm to 9.3 nm, respectively, wi th means of 5.6 nm and 7.8 nm. Unlike what was noted with disorder indices (HI and R2), CSD dist ributions for Florida and Georgia samples exhibit significant overlap for both area-weighted (Fig. 5-3C) and volume-weighted (Fig. 5-3D) thickness calculations, resulting in no differen tiation between the two sa mple populations based on this parameter. This is further reflected in the lack of different iation between individual sample localities, although the spr ead within the data for individual sites is similar that observed for the disorder indices (Table 5-2). This la tter observation is further supported by box-andwhisker plots of both the area-weighted (Fig. 5-4C) and volume-weighted (Fig. 5-4D) thickness calculations for Florida and Georgia Cypresshead samples. Additionally, several other trends are notable, including the relative coar seness of the FRL-1 and SSJ-1 samples and the fine crystallite thicknesses measured for the Birds locality (B-1). In general, site specific trends related to the CSD results are consistent with those noted for disorder indices (HI and R2) and the expert system of Planon and Zacharie (1990). For localities where significant trends in disorder were noted (EPK, FRG, and SSD-1), CSD values increase with depth from the surface. Additiona lly, where disorder indi ces indicate a reversal near the base of select localities, CSD results exhibit a simila r trend, recording a decrease in
148 value prior to reaching the basal contact of the Cypresshead Formation. Cont rasting with what is observed for the Florida localities, none of the Georgia localitie s exhibit significant vertical trends in CSD values. Correlation of the areaweighted mean thickness and volume-weighted mean thickness calculations to HI values for Flor ida Cypresshead samples is illustrated in Figure 5-5A and B. With an increase in depth from the surface in north-central Florida, CSD distributions diverge significantly from the theoretical lognormal distribution curves defined by and 2 (Fig. 5-6). In near surface samples, curves are most similar to a lognormal dist ribution, but tend to be narrower than predicted, often with a circumflex (or hat) that extends above the mode of the theoretical lognormal curve. This feature indicates that the curves may have started out with an asymptotic shape, a feature which is commonly attributed to the early stag es of crystallization (Eberl et al., 1998; Simi and Uhlk, 2006), with loss of the low end of the asymptotic curve potentially related to dissolution of a fine crystallite fraction. With dept h, the curves take on a broader shape toward greater crystallite thic knesses, sometimes toward distribution curves possessing a bimodal or polymodal shape (Figs. 5-6 and 5-7). For north-central Florida Cypresshead (and reworked Cypresshead) sediments, this shift in CSD thickness distributions is associated with 2 values remaining constant with increasing which correlates to an open system supply-controlled growth method according to the Law of Proportionate Effect (Eberl et al., 1998). Georgia samples exhibit no significant change in values and more closely correlate to a lognormal distribution shape than north-central Florida kaolinites, suggesting minimal evidence for recrystallization (or ne oformation). Examples of verti cal trends in CSD distribution curves for select north-central Florida localities characterized by significant spreads in CSD and disorder values (EPK, FRG, and SSD) are included in Figure 5-7 for reference.
149 Figure 5-5. Scatterplots illustrating the positiv e correlation between the Hinkley Index (HI) and measured CSD values for north-central Florida Cypressh ead (and reworked Cypresshead) samples. A) HI plotted agains t the area-weighted mean thickness, B) HI plotted against the volume -weighted mean thickness. Kaolinite Microtexture Microtextures within a kaolin record eviden ce of both deposit origin and post-depositional alteration processes. Furthermore, microtextures allow in situ crystallization to be differentiated from sedimentary fabrics (Hurst and Rigsby, 1984). Kaolins with a high concentration of parallel intergrowths, random intergrowths, coarse booklets, stacks, and vermiforms are indicative of in situ crystallization (Keller, 1977; Hurst a nd Pickering, 1989), while those possessing a
150 Figure 5-6. Measured CSD distribution curves a nd fitted theoretical lognormal curves (red) for select EPK36-J-12 samples. A) narrow cu rve possessing a circumflex characteristic of an initial asymptotic shape, B) curve exhibits initial broadening and development of a secondary mode at ~8 nm, C) continued development of a polymodal shape, D) same as (C). sedimentary fabric of crystallites in face-t o-face association are in dicative of little postdepositional recrystallization (H urst and Pickering, 1989). In Ge orgia, Cretaceous kaolins possess the textures consistent with in situ recrystallizati on, with Tertiary kaolins possessing a notably fine-grained and face-to-face oriented fabric (Keller, 1977; Hu rst and Pickering, 1989), thus characteristic of l ittle recrystallization. SEM data from Cypresshead Formation samp les indicate varia tions in kaolinite morphology for sediments both vertically and be tween sampling areas, particularly for northcentral Florida sites (Fig. 5-8). Near surface north -central Florida kaolinite consists of small ( 1
151 Figure 5-7. Example CSD distributions illustrating changing curve shape with increased depth from the surface. A) EPK36-J-12, B) EPK30-V-6, C) FRG-1, D) SSD-1. m) rounded aggregates and anhedral to subhe dral crystallites in an open microtexture associated with combined face-to-face and face-t o-edge floccules (Fig. 5-8A). Where present, vermicular (or vermiform) kaolinite is highly degraded, or potentiall y recrystallized, and commonly appears coated by either secondary overgrowths of fine (<0.5 m) anhedral to subhedral kaolinite or transloc ated (illuviated) kaolinite. Fo r both FRL-1 and SSJ-1 sample localities, this microtexture is less developed, and correlates with a redu ction in the apparent porosity of near surface samples. Elsewhere, this kaolinite morphology tends to transition downsection into significant accumulati ons of vermicular kaolinite com posed of slightly coarser (1-2 m) subhedral crystallites reta ining a similar open microtexture (Fig. 5-8B) and samples indicating the development of ove rgrowth crystallization textures consisting of subhedral to
152 Figure 5-8. SEM photomicrographs illustrating the microtextural variation noted in Cypresshead and reworked Cypresshead Formation sediments. A) Near surface rounded aggregates of kaolinite (FRG-1-6), B) Degraded ve rmicular kaolinite (FRG-1-13), C) Fine secondary overgrowths of kaolinite (SSD-1 -10), D) Well formed kaolinite vermiform (FRG-1-11), E) Matrix image of samp le EPK36-J-12 (59-62), illustrating the dominance of vermicular kaolinite, F) Dense microtextue inconsistent with recrystallization fabric (L-1-6).
153 euhedral crystallites (Fig. 5-8C). Where overgrowths are developed on preexisting vermicular or discrete kaolinite, crys tallization appears to be favored along grain edges, orient ed face-to-edge or edge-to-edge at an oblique an gle. At depth, north-central Flor ida kaolinite is dominated by the accumulation of vermicular kaolinite (Fig. 5-8D ), which can compose the bulk of the clay matrix, particularly at the EPK Mi ne site (Fig. 5-8E). Evidence of this microtexture trend is well documented in the EPK36-J-12 core, where vermicular kaolinite appears degraded (recrystallized) or coated with overgrowths or tr anslocated clay down to 48 ft, while matrix clay retains an aggregate morphology observed at the top of the core. Below 50 ft, a more characteristic open microtexture with large ve rmiforms is dominant, but with individual crystallites still fine in particle-size (<2 m). Cypresshead samples from southeastern Georgia differ significantly, consisting of small ( 1 m) subhedral kaolinite crystallites in a closed, faceto-face flocculated microtexture inconsistent with significant recrystallizaton (Fig. 5-8F). Where a more open face-to-edge microtexture is developed in Georgia sediments (L-1), it tends to be restricted to sandy lithologies possessing mini mal clay content. The general difference in sediment microtexture (microfabric) between north-central Florida and southern Georgia Cypresshead Formation sediments indicates signif icant differences in weathering susceptibility as a function of porosity (and likely permeability), which in turn, appears to be related to spatial variations in the initial clay content of the unit at the time of deposition. Particle-Size Analysis SediGraph analyses were performed in order to characterize silt and clay particle-size distributions as a means of confirming XRD a nd SEM based observations related to general trends of increased structural order, crystallite (CSD) size, and clay part icle-size with increased depth of sampling. Figure 5-9 is included for re ference, and indicates a general trend of
154 Figure 5-9. Example SediGraph particle-size di stributions for select Cypresshead Formation samples. A) EPK36-J-12, B) FRG-1, C) SSJ-1, D) J-1. increasing coarsening of particle-size with depth. This is agrees with the previously noted observation related to both disord er and crystallite size. Additiona l data is included in Appendix G for reference. Geochemistry Major element and rare earth element (REE) da ta were collected using a combination of ICP-AES and ICP-MS methodologies to assist in defining weathering r eactions impacting the Cypresshead Formation. Of particular interest to this study is the source of Al and Si necessary for the neoformation of kaolinite and coeval weathering reactions involving secondary accessory phases (e.g., crandallite-florencite gibbsite, halloysite). The latter of these has the potential of
155 impacting the dissolution and/or crystallization ki netics of kaolinite and the chemical signature (i.e., REE distributions and Nd isotopes) of the clay fraction. Major element data Major element data for the clay (< 2 m) si ze-fraction of both north-central Florida and southeastern Georgia Cypresshead Formation and reworked Cypresshead sediments converted to oxides (%) is reported in Table 5-4. Original ma jor element data is included in Appendix H. Additionally, a correlation matrix indicating th e correlation coefficients (r) between major elements and REE is shown in Table 5-5. Due to the complexity of weathering reactions occurring within the unit, correla tion coefficient values (r > 0.6) were considered significant. However, due to the low concen trations encountered with Na2O for the total data set and CaO for the Georgia samples, no correlations for thes e elements were considered significant. Correlations among major elements for Cypressh ead samples are related to variations in sample mineralogy and weathering reactions im pacting Cypresshead sediments. The most significant of these as it relates to the overall geochemical signature of Cypresshead clays is the occurrence of a cranda llite-florencite phase in some samples. P2O5 concentrations approach almost 2% in several samples (SSD-1), with sign ificant enrichment evident in others (EPK and FRG). P2O5 exhibits a range of 0.046% to 1.978% and 0.034% to 0.213% in north-central Florida and southeastern Georgia samples, respectively, with m eans of 0.228% and 0.092%. It is significant to note that P2O5 and CaO show no correlation for the Florida samples (Table 5-5), as both crandallite and florencite appear to be present in these sediments (Fig. 5-10A). The only major element correlations of signi ficance are noted for the Georgia samples, with MgO, Fe2O3, and K2O exhibiting positive correlations in response to the concentration of both illite and hydroxyl-interlayered vermiculite (HIV) in these sediments. Negative correlations
156 Table 5-4. Major element concentrations as ox ides for Cypresshead Formation clay (< 2 m) samples (concentrations in %). Sample ID Interval Na2O MgO Al2O3P2O5K2O CaO TiO2Fe2O3Cypresshead Formation FL* EPK36-J-1225-270.0110.07636.6970.1370.040.0480.2500.27 EPK36-J-1235-400.0130.05033.8060.1490.050.0490.2340.36 EPK36-J-1246-480.0150.07337.6040.0710.050.0350.2820.49 EPK36-J-1250-530.0200.05032.2000.0460.080.0280.3170.51 EPK36-J-1259-620.0200.09934.1840.0690.100.0770.5500.77 EPK31-P-4035-450.0180.08337.6230.0620.060.0450.3470.66 EPK31-P-4050-620.0200.15832.4450.0690.080.1470.3090.84 EPK31-P-4062-650.0180.39334.5430.1720.230.2880.2851.24 EPK30-V-622-240.0202.4299.5110.2220.280.6410.5922.55 EPK30-V-630-350.0341.27725.1131.1800.290.8400.3001.53 EPK30-V-648-530.0340.22432.6720.6300.160.4970.4170.89 EPK30-V-658-630.0200.06037.2450.0500.080.0220.2520.41 EPK30-V-668-730.0200.11632.4640.0570.140.0700.4340.67 FRG-130.0120.08335.6390.2800.050.0420.3122.99 FRG-160.0070.02533.7300.1600.020.0210.0670.99 FRG-17n.a.0.02535.0720.0920.02n.a.0.0750.64 FRG-1100.0130.09134.4860.2410.100.0420.2170.61 FRG-1110.0070.03335.1660.0690.040.0140.1080.23 FRG-1130.0070.02535.7710.0460.040.0070.1170.21 FRG-1150.0070.02533.3520.0800.050.0070.2250.23 FRG-230.0670.04134.8640.0570.040.0140.1080.27 FRG-25n.a.0.02534.1080.0550.04n.a.0.1070.36 FRG-270.0610.03335.4880.0460.050.0140.1500.27 FRG-2100.0130.05834.8260.1260.080.0210.2670.63 FRG-2120.0110.13833.5410.1440.110.0140.3601.40 FRL-120.0070.05533.2010.0570.06n.a.0.3090.83 FRL-140.0070.05633.2960.0640.07n.a.0.3150.94 FRL-160.0070.07533.8810.0690.050.0140.2500.61 FRL-170.0070.08332.2370.0460.050.0140.2170.59 SSJ-120.0190.15134.2780.1240.100.0381.1543.22 SSJ-140.0160.16233.5600.1400.080.0240.5822.14 SSJ-160.0200.09135.0530.0570.080.0280.1580.47 SSJ-180.0220.10934.7880.0530.200.0410.2540.40 SSJ-1100.0270.15834.6560.0570.160.0700.4090.51 SSD-110.0810.24028.4581.9780.410.1017.9702.12 SSD-130.0130.07037.8311.6680.080.0700.7240.56 SSD-160.0110.10133.0500.1080.050.0080.2290.66 SSD-170.0200.10833.7680.0800.070.0140.2590.73 SSD-1100.0200.29033.6360.0800.110.0980.2251.04 Max 0.0812.42937.8311.9780.4100.8407.9703.217 Min 0.0070.0259.5110.0460.0240.0070.0670.214 Mean 0.0200.19133.4320.2280.0990.1000.5060.894 Cypresshead Formation GA J-120.0230.24715.2460.1050.190.0131.4017.65 J-140.0340.20734.2400.0800.520.0141.0264.48 J-160.0270.18233.6170.0340.370.0070.7671.90 L-130.0190.13136.3570.1120.18n.a.1.0281.72 L-15n.a.0.05035.3550.0440.05n.a.0.3740.84 L-160.0200.12433.8060.0570.270.0070.8511.62 B-120.0510.63028.2690.0871.200.0140.6665.62 B-130.0630.70329.3270.2131.540.0130.71710.91 B-150.0540.70533.2960.0921.070.0910.7423.63 Max 0.0630.70536.3570.2131.5420.0911.40110.909 Min 0.0190.05015.2460.0340.0480.0070.3740.844 Mean 0.0360.33131.0570.0920.6000.0230.8414.262 Note: n.a., element concentration was below the detection limit. Florida Cypresshead Formation samples include reworked Cypresshead sediments collected at the SSD-1 locality.
157 Table 5-5. Correlation matrix of major elements and REE for both Florida and Georgia Cypresshead samples. Na2O MgO Al2O3P2O5K2O CaO TiO2Fe2O3 REE Cypresshead FL* N a2 O 1.00 MgO 0.101.00 Al 2 O 3-0.16-0.921.00 P 2 O 50.460.21-0.201.00 K 2 O 0.510.59-0.600.621.00 CaO 0.180.82-0.710.370.631.00 TiO 20.600.06-0.210.700.650.031.00 F e2 O 30.110.46-0.460.280.440.340.361.00 REE0.500.29-0.350.860.740.460.760.341.00 Cypresshead GA N a2 O 1.00 MgO 0.961.00 Al 2 O 3-0.01-0.211.00 P 2 O 50.590.61-0.291.00 K 2 O 0.990.95-0.080.661.00 CaO 0.440.550.200.020.321.00 TiO 2-0.62-0.17-0.600.18-0.26-0.221.00 F e2 O 30.630.66-0.640.840.68-0.130.291.00 REE-0.200.020.460.17-0.030.760.15-0.301.00 Florida Cypresshead Formation samples include reworked Cypresshead sediments collected at the SSD-1 locality. noted between Al2O3 and MgO, K2O and CaO for Florida samples are insignificant because they are related to the high clay fraction quartz c ontent at the top of EPK30-V-6 diluting the Al2O3 content. Consequently, the correlations among CaO, MgO and K2O are similarly related. Rare earth element (REE) data Although rare earth elements (REEs) can be re liable geochemical tracers of sedimentary provenance because of their charac teristic immobility and resistan ce to elemental fractionation in the supracrustal environment (Wildeman a nd Condie, 1973; Piper, 1974; Nesbitt, 1979; Chaudhrui and Cullers, 1979), studies suggest that mobilization and local fractionation of the REEs can occur under extreme weathering and di agenesis (McLennan, 1989). In fact, relative concentrations of REEs in neoformed and recrystallized phases will reflect, to varying degrees, the geochemical conditions and redox potential within local pore wa ters at the time of formation (Henderson et al., 1983; McLennan, 1989).
158 Figure 5-10. Scatterplots illustrating mixing trends related to the presence of crandalliteflorencite series minerals in Cypresshead Formation clays. A) relationship between P2O5 and CaO for crandallite and florencite crystallization tren ds, B) correlation between P2O5 and REE (r = 0.86). REE concentrations for the clay (< 2 m) si ze-fraction of both north-central Florida and southeastern Georgia Cypresshead Formation a nd reworked Cypresshead sediments along with relevant REE ratios are reported in Table 56. Additionally, REE data for three samples representative of updip, less weathered C ypresshead clays (TRF 2214, WEX164, and WEX366), two samples of underlying Hawthorn Group clays (MCB109 and J-1-BC), and three EPK
159Table 5-6. REE concentrations of Cypresshead Formation and related samples (concentra tions in ppm). Sample IDIntervalLaCePrNdSmEuGdTbDyHoErTmYbLu REEEu/Eu* (La/Yb)n(La/Sm)n(Gd/Yb)n LREE/ HREE Cypresshead Formation FL* EPK36-J-1225-2718.431.93.9013.72.20.471.880.2184.108.40.2060.070.40.0575.240.7130.765.163.7515.01 EPK36-J-1235-4029.653.46.66220.127.116.114.340.592.900.501.260.150.80.09129.760.6924.744.254.3211.12 EPK36-J-1246-4822.639.24.8018.32.70.772.360.291.360.270.640.090.50.0793.950.9418.104.22.16815.70 EPK36-J-1250-5320.032.53.8222.214.171.1241.630.190.940.160.420.060.40.0676.260.9433.435.613.2518.61 EPK36-J-1259-6234.476.69.3236.07.32.955.500.652.760.380.9126.96.36.19977.871.4328.752.915.4814.48 EPK31-P-4035-4520.238.04.1115.32.51.042.030.251.120.190.470.070.50.0885.861.4227.024.993.2417.01 EPK31-P-4050-6228.367.47.6188.8.131.525.050.713.300.601.4184.108.40.20653.411.7114.563.123.1010.72 EPK31-P-4062-6545.9115.014.5064.113.56.2313.501.999.911.693.910.472.90.42294.021.4210.582.103.717.27 EPK30-V-622-2454.4112.014.0055.510.32.0010.101.437.311.353.350.442.60.37275.150.6013.993.263.109.14 EPK30-V-630-35243.0516.072.20306.060.212.7060.708.3843.308.0120.402.3512.71.601367.540.6512.792.493.817.61 EPK30-V-648-5389.9169.022.5093.419.54.6918.902.5013.002.456.040.734.10.48447.190.7514.662.843.678.18 EPK30-V-658-6330.451.45.3619.03.41.412.980.391.740.290.700.090.50.09117.751.3640.665.524.7516.16 EPK30-V-668-7324.847.06.1021.84.31.753.710.532.730.481.310.191.30.17116.171.3512.763.562.279.98 FRG-1344.386.710.9043.98.31.667.851.115.420.992.480.291.70.26215.860.6317.433.293.689.66 FRG-1636.076.39.86220.127.116.115.330.693.010.481.070.140.60.08177.950.8940.123.477.0814.46 FRG-1731.260.97.2318.104.22.168.680.481.940.280.550.070.40.06138.990.9252.164.387.3317.47 FRG-110139.0281.035.60139.023.16.8516.802.027.220.971.840.191.00.12654.711.0792.953.7113.3920.48 FRG-11140.671.48.022.214.171.1243.510.451.690.230.45n.a.0.3n.a.162.921.1490.505.579.3323.35 FRG-11320.333.84.0014.21.90.781.490.200.880.120.24n.a.0.2n.a.78.111.4367.876.595.9423.71 FRG-11555.1122.014.3047.37.72.765.240.712.580.330.640.070.4n.a.259.131.3492.114.4210.4424.71 FRG-2314.325.12.94126.96.36.1992.010.250.970.150.34n.a.0.2n.a.61.721.0947.813.048.0114.53 FRG-2521.643.34.7217.22.71.042.020.260.930.150.31n.a.0.2n.a.94.431.3772.224.948.0523.13 FRG-2717.135.24.0114.92.61.001.870.240.940.130.26n.a.0.2n.a.78.451.3957.174.067.4520.28 FRG-21052.4112.013.2052.610.64.3188.8.131.520.821.890.201.00.12265.031.3735.043.057.2812.13 FRG-21253.3120.015.0063.713.54.5112.801.808.401.362.850.281.60.19299.291.0522.282.446.389.07 FRL-1184.108.40.206220.127.116.112.080.311.260.200.400.060.40.07114.001.1550.326.404.1422.66 FRL-1445.588.79.3418.104.22.1683.300.461.880.260.590.070.50.07187.781.2760.856.105.2625.11 FRL-1645.0108.013.3050.18.72.725.850.742.710.370.840.110.60.08239.121.1722.214.171.12419.92 FRL-1733.578.29.8937.46.01.864.550.602.480.360.870.090.60.07176.471.0937.343.456.0417.15 SSD-11493.0850.085.00286.044.97.2536.905.5428.905.9617.802.7419.63.211886.800.5516.826.771.5014.58 SSD-13123.0229.023.5080.513.31.989.611.304.960.761.7126.96.36.19991.740.5451.415.714.7922.94 SSD-1643.093.511.4048.69.22.707.060.883.290.450.880.100.60.09221.751.0347.922.889.3815.41 SSD-1750.1123.014.5058.511.13.388.401.044.240.531.090.110.60.07276.661.0855.842.7811.1616.00 SSD-11022.857.75.79188.8.131.523.700.492.090.340.760.090.50.05123.001.2630.493.065.9014.13 SSJ-1253.786.87.84184.108.40.2062.970.472.800.621.840.302.10.33186.900.6217.1010.041.1315.30 SSJ-1432.456.15.4518.02.90.542.460.361.970.3220.127.116.11.19123.320.6218.056.891.6314.48 SSJ-1610.718.104.22.168.20.490.890.150.620.110.22n.a.0.2n.a.41.951.4635.785.503.5517.93 SSJ-189.616.31.722.214.171.1240.780.130.550.110.26n.a.0.2n.a.36.911.5832.106.583.1116.97 SSJ-11016.429.83.45126.96.36.1991.510.231.050.170.490.060.40.0667.291.4427.425.623.0115.76 TRF2214 60.0-62.529.543.34.17188.8.131.522.540.422.620.541.5184.108.40.20603.820.6512.337.581.279.62 WEX164 18.0-26.076.1131.015.0058.910.92.5511.801.8010.102.206.050.855.10.72333.070.699.984.311.847.56 WEX366 9.0-10.046.168.06.9323.94.00.854.420.683.860.902.710.412.80.41165.970.6211.017.111.269.20 EPK Vermiforms ---220.127.116.1118.104.22.1681.040.191.330.411.400.241.70.3322.214.171.1244.140.492.61 EPK Mica ---126.96.36.199188.8.131.521.380.211.380.290.920.141.00.1737.511.154.813.701.105.74 EPK Feldspar ---3.65.00.59184.108.40.2060.500.070.380.070.21n.a.0.2n.a.14.255.0212.045.551.998.45 Cypresshead Formation GA J-1249.283.37.28220.127.116.112.710.513.190.772.520.392.60.43177.860.6112.6510.470.8312.51 J-1462.2121.011.2032.34.50.843.250.441.800.310.910.130.90.12239.900.6846.218.532.8829.41 J-1651.5134.012.5041.55.91.174.230.531.980.300.718.104.22.16855.250.7249.205.394.8228.27 L-13158.0367.028.8073.78.51.575.250.722.740.471.222.214.171.12449.570.7288.0411.473.4953.00 L-1547.288.46.4317.02.10.391.490.210.920.170.410.070.40.07165.260.6878.9113.872.9743.08 L-16112.0189.015.3036.14.40.833.370.461.670.280.7126.96.36.19965.180.6693.6215.713.3647.26 B-1253.998.29.6188.8.131.523.840.542.610.511.4184.108.40.206210.410.6721.206.931.8017.77 B-1363.6131.012.9046.97.61.596.340.894.580.872.540.392.60.40282.200.7016.365.161.9414.08 B-1589.3198.021.2088.219.34.5516.802.2811.302.115.960.865.50.72466.080.7810.862.852.439.14 Hawthorn Group, Coosawhatchie Formation FL/GA MCB109 15.0-20.056.394.59.8536.06.31.416.510.955.471.233.420.523.60.55226.610.6810.465.511.449.12 J-1BC93.8187.023.1095.619.25.0521.303.3619.004.1011.801.6210.41.52496.850.776.033.011.635.73 Florida Cypresshead Formation samples include reworked Cypresshead sediments collected at the SSD-1 locality.
160 composite samples consisting of vermicular (vermiform) kaolinite, muscovite mica, and Kfeldspar concentrates are also included in Table 5-6. All sample localities are shown in Figure 31. REE data for Cypresshead clays and accessory samples have been normalized to the chondrite values of Nakamura (1974) for plotting purposes. Data for REE concentrations of Florida Cypr esshead sediments (including SSD-1) correlate strongly with P2O5 (r = 0.86) due to the occurrence of crandallite-florencite series minerals ((Ca,REE)Al3(PO4)2(OH)5H2O) in the clay fraction (Fig. 5-10B). REE values for Florida samples vary significantly, ra nging between 36.91 ppm and 1886.80 ppm, whereas Georgia samples are more constrained, ra nging between 165.26 ppm and 649.57 ppm, with means of 207.96 ppm and 312.41 ppm, respectively. The distribution in REE values correlated to secondary crandallite-florencite supports mob ilization and reconcentration of REEs in northcentral Florida sediments as a consequence of weathering. Furthermore, given the high concentration of REEs in the cr andallite-florencite phase, it has the potential to impart a change in shape to chondrite-normalized REE distribution patterns should th e phase preferen tially enrich light (L) REEs as has been indi cated in other studies (Dill et al., 1995; Rasmussen et al., 1998) or retain the REE signature of a precursor marine phosphate phase. As such, Florida clay fraction samples exhibiting evidence of crandallite-florencite via XRD or anomalous P2O5 and/or anomalous REE concentrations may be considered to possess a mixed REE signature if either of these conditions are observed. Chondrite-normalized REE distri bution patterns for Cypresshead Formation and associated samples are shown in Figure 5-11. Both Florid a and Georgia Cypresshead samples (including SSD-1) exhibit a wide range in the slope of REE di stributions, with (La/Yb)n values varying from 10.58 to 92.95 and 10.86 to 93.62, respectively. For Florida samples, LREE enrichment
161 Figure 5-11. Chondrite-normalized REE distributi on patterns for select Cypresshead Formation clay (< 2 m) fraction and related sample s. A) EPK36-J-12, B) EPK30-V-6, C) FRG1, D) SSJ-1, E) SSD-1, F) J-1 and B-1, G) EPK vermiform, mica, and feldspar composite concentrates, H) updip C ypresshead and Hawthorn Group clays.
162 relative to chondrite (La/Sm)n is moderate (2.10-10.04) as is heavy (H) REE depletion ((Gd/Yb)n = 1.13 to 13.39). No significant difference in LR EE enrichment is apparent between phosphate and non-phosphate bearing samples, thereby ex cluding any evidence of a fractionation or REE mixing effect associated with th e crandallite-florencite phase. Fo r Florida samples, one of the most notable REE trends corresponds to the vari able HREE depletion values seen between some near surface samples and samples collected at depth. For shallow samples associated with FRG1, SSJ-1 and SSD-1 sampling locations, HREE trends are only slightly depleted to flat, whereas all other Florida Cypresshead samples exhibit moderate HREE depletion. The reason for this observation is most likely related to the prefer ential adsorption of HR EEs under reducing (anoxic to dysoxic) conditions by pore water organics (humic acid) (Sonke and Salters, 2006; Wan and Liu, 2006). Below the water table, in situ derived clays would exhibit HREE depletion in response to the preferential uptake of the HREEs by the organics. With exposure to oxic weathering conditions above the water table fa voring decomposition of organics, clays would uptake liberated HREEs, resulting in the flat HREE trends noted for some near surface samples. For sampling locations where the wa ter table is at or near the su rface (EPK), this trend is not observed. For Georgia samples, LREE enrichment is slig htly greater than that seen in Florida ((La/Sm)n = 2.85 15.71), likely as a consequence of LREE enrichment by goethite, while HREEs are flat to only sl ightly depleted ((Gd/Yb)n = 0.83 to 4.82), reflecting the overall reduced severity of leaching (and potential for in situ clay formation) impacting these sediments. As for the less weathered updip Cypresshead Formation and Hawthorn Group clays (Fig. 5-11H), their REE patterns are consistent, possessing moderate LREE enrichment, flat HREE, and moderate negative Eu anomalies similar to what is noted for Georgia Cypresshead clays. Furthermore, the
163 similarity in the REE distribut ion of these clays suggests a re worked origin for the updip Cypresshead clays in agreement with observations outlined in Chapter 4. Whereas Eu anomalies (Eu/Eu*) are most commonly used to discern intracrustal igneous differentiation processes (McLennan et al., 1993), the vertical distributi on of Eu anomalies in Florida Cypresshead sediments are more significan tly an indicator of w eathering processes and kaolinite origin. For Florida samples, Eu/E u* values range between 0.54 and 1.71, reflecting both positive and negative Eu anom alies of moderate magnitude, w ith the frequency distribution of values suggesting the occurr ence of three modes, a positive Eu/Eu* mode at ~1.5, a flat to slightly positive Eu/Eu* mode at ~1.1, and a negative Eu/Eu* mode at ~0.7 (Figure. 5-12). The most positive Eu anomalies occu r at of near the base of th e Cypresshead Formation, where residual feldspar has been noted (Chapter 4) and where microtextural evaluation of these sediments suggests the alteration of mica into verm icular kaolinite. Given these associations, and the positive Eu anomaly determined for the EPK feldspar concentrate (Fig. 5-11G), it appears that the dissolution of feldspars possessing an original plagioclase component (EPK Feldspar CaO = 0.432 %) imparted the observed Eu anomaly to the clay fraction. Occurring below the water table under dysoxic to anox ic conditions due to pore water organics, positive Eu anomalies would be preserved by clays formed in situ from the dissolution and/or weathering of feldspar and micas. Further evidence for this process is seen in the REE distributions of both the mica and vermicular kaolin (vermiform) concentrates (F ig. 5-11G), which show evidence of slight Eu enrichment, supporting their weathering (mica) or formation (vermiform) in proximity to feldspar dissolution. With decreasing depth fr om the surface, Eu anomalies decrease in magnitude, and ultimately transiti on into negative values in near surface clays (Fig. 5-11). This transition is likely in response to two near surface factors impacting th e clay fraction and pore
164 0 2 4 6 8 10 120.5 0.6 0.7 0.8 0.9 1.0 1.1 1.2 1.3 1.4 1.5 1.6 1.7 1.8 1.9 2.0Eu/Eu*Frequency Cypresshead GA Cypresshead FL* Figure 5-12. Histogram illustrating the distri bution of Eu/Eu* values for both Georgia and Florida Cypresshead Formation clays. water organics; exposure to oxic weathering c onditions and kaolinite recrystallization under vadose or mixed vadose/saturated conditions near the water table. Contrasting with Florida observations, Eu/Eu* values for Georgia Cypressh ead samples are fairly constant (Eu/Eu* = 0.61-0.78), consistent with a moderate negative Eu anomaly. Discussion Hurst and Pickering (1997) have noted that sedimentary kaolins do not originate exclusively through sedimentation processes, but rather through a complicated assortment of post-depositional altera tion processes acting on sediments. This is true, although original sedimentation characteristics can have a si gnificant impact on pos t-depositional reaction pathways and the relative susceptibility of sediments to alteration processes. Kaolinite Origin Kaolinite in the Cypresshead Formation consis ts of at least three separate components; detrital kaolinite associated with discrete clay beds, stringers, and lenses that were deposited
165 during the initial sedimentation of the unit, neoformed kaolinite associated with the topotactic replacement and/or epitactic ov ergrowth of muscovite mica (and kaolinite) at the expense of dissolved K-feldspar, mica, and clays, and near surface recrystallized kaolinite which formed at the expense of precursor clays, including detrital and neofor med kaolinite, and other labile components under pedogenic (vadose) and fluctua ting water table (vadose/ saturated) conditions. Although a significant portion of the kaolinite fraction found in Cypresshead Formation sediments is most certainly detrital in origi n, the original mineralogical maturity of those precursor clays remains a question, as even the kaolins of the Georgia-South Carolina kaolin district were modified mineral ogically following deposition as a less mature clay mineral suite. Based on REE data evaluated for less weathe red updip Cypresshead clay samples in northcentral Florida and Georgia Cypresshead clays, it appears likely that the detrital clay component of the formation was derived from local sources, with little coast parallel reworking of the clay fraction from Georgia southward. However, limiti ng the origin of Cypresshead Formation clays to a detrital source is not corroborated by the mineralogical and geochemical observations compiled in this study. Among the possible pathways for the formation of an in situ kaolinite fraction are transformation (topotaxy), neoforma tion (epitaxy), and recrystallization weathering processes. Kaolinite neoformation Although mica is rather resistant to weatheri ng under normal conditions, the kaolinization of micas, including muscovite, by topotactic replacement and epitactic overgrowths has been well documented in the literature as outlined by Jeong (1998b) and Arostegui et al. (2001). Mica, and particularly a muscovite mica substrate, fits the requirements for t opotactic replacement by kaolinite in possessing a high de gree of three-dimensional structural accord between both reactant and product. Additionally, mica grain surfaces are well suited to the epitactic
166 nonrandom overgrowth of kaolinite given that the mica cell dimensions in the ab plane match those of kaolinite to within 4% (Bailey, 1980). As such, the basal su rface is an excellent template for epitactic neoformation. Based on SEM analysis, kaolinization of muscovite in Cypresshead sediments, and the corresponding formation of vermic ular kaolinite, appears to take place under reducing (dysoxic to anoxic), saturated conditions according to the model proposed by Jeong (1998b). For Cypresshead sediments, initial weathering of th e muscovite takes place along grain edges leading to the development of the characteristic spla ying noted by both SEM and standard petrographic microscopy (Figs. 5-13A and B). This first phase of the weathering pro cess appears to involve the topotactic replacement of muscovite grain edge s, which generates tensional stresses within the grain interior in response edge splaying. These stresse s, in turn, result in the formation of lenticular voids along interior basal cleavage surfaces and smaller voids along grain edges in response parting, accessing basal grain surface area suitable for the crystallization of kaolinite via an epitactic overgrowth model. Dissolution stud ies suggest that during this initial phase of weathering the muscovite surface is modified by the precipitation on an Al-hydroxide or kaolinite monolayer when surface K+ is exchanging with H+ in solution, an early stage alteration process consistent with what is observed in so il when mica transforms to HIV (Harris et al., 1992; Nagy and Pevear, 1993). With continued repl acement of grain edges inward, splaying and void development continues, resulting in the significant expansion of the muscovite flake along the c-axis (Fig. 5-13C). Eventually exfoliation of discrete vermiforms takes place as topotactic replacement continues to migrat e toward the core of the grain, exposing additional grain edges and basal surfaces to continued kaolinization (Fig. 5-13D).
167 Figure 5-13. SEM photomicrographs of Cypresshead and reworked Cypresshead Formation kaolinite textures associated with the in situ weathering of muscovite. A) Neoformed kaolinite developing along the edges (arrow) and on the basal surface (arrow) of muscovite (FRG-1-10), B) Broader view of same sample described in (A), C) Characteristic splayed edge of muscovite flake with re sulting lenticular void development (arrow) and formation of pr e-exfoliated vermiform, D) Detached vermicular kaolinite with residual mu scovite core as confirmed by EDS. This weathering process also provides an answer to the question surrounding the distribution of fine mica in co arseto medium-sands and grav el. Not adequately explained by hydrodynamic sorting during sediment ation, the actual cause of this association is due to the previously described weathering process. As mica flakes are constantly exfoliated during kaolinization and vermicular kaolinite formati on, the grains will constantly decrease in size. Thus, mica flakes which may have been of a gr ain-size suitable to co-deposition with coarse
168 sediments will degrade through weathering, eventual ly decomposing into fine sand and silt sizes consistent with earlier noted observations (Pirkl e, 1960). Observations related to a decrease in kaolinite order and CSD values in basal north-cen tral Florida clays are also explained by this mica weathering model. If topotac tic replacement is favored during the early stages of mica weathering, with epitactic overgrowth becoming the dominant neoformation processes over time, then early stage kaolinites associated with a hi gh percentage of topotac tic replacement are likely to possess greater disorder and smaller CSD values consistent with observations. With continued weathering, and an increased role for epitactic ne oformation of kaolinite, kaolinite populations should exhibit increased structural order and CSD values under saturated groundwater conditions. Once exposed to fluctuating water ta ble and/or vadose conditions, this weathering model is likely to be replaced by dissolution and/or recrystallization reactions. Kaolinizing muscovite grains altered according to this model exhibit a substantial increase in grain volume. Since conserva tion of Al during the weathering process would dictate a volume reduction following a topotactic onl y replacement model for muscov ite, a substantial import of dissolved Al (and Si) from an external weathe ring source is required. Several studies have described the weathering of muscovite to kaolinite via dissolution-recrystallization (Banfield and Eggleton, 1990; Jiang and Peacor, 1991; Singh an d Gilkes, 1991), thus, participation of an imported Al and Si component during epitact ic neoformation is expected. Based on the likelihood that Cypresshead Formations sands co ntained up to 5% or more K-feldspar during initial deposition, the presence of an inherited Eu anomaly in basa l Cypresshead clays, and given the highly weathered character of skeletal feldsp ar grains noted by SEM, there should have been sufficient Al and Si delivered into the aqueous environment to drive ne oformation of kaolinite.
169 The dissolution rate constant of K-feldspar is one order of magnitude greater than that for muscovite (Arostegui et al., 2001), supporting the view that K-feldspars were the most likely source of Al and Si necessary to crystallize neoformed kaolinite via epitaxy onto muscovite surfaces. Al and Si which would have been m obilized during muscovite weathering would have been consumed by the topotactic replacement process. In order for this crys tallization reaction to work, H+ would need to be supplied, and pore water circulation would need to be continuous to facilitate the removal of K+ and support a low aK+/ aH+ ratio promoting kaolinite crystallization (Arostegui et al., 2001). As a major aquifer rech arge area, hydrologic conditions in north-central Florida would be well suited to satisfy these requirements as evidenced by the open system supply-controlled growth characte ristics of Cypresshead clays. Jeong (1998b) suggests up to a 9-fold increase in volume from primary micas to resultant kaolinite through the vermiform crystallization pr ocess. Not only does this indicate that the epitactic method of kaolinite neoformation is dom inant with respect to topotactic replacement, but this also represents a signi ficant method by which to tal kaolinite (clay) content can increase within a given sedimentary sequence. In order to achieve the 20% clay c ontent as seen in the EPK deposit, this would require the complete weathering of only a 2.2% mica fraction. The only limiting factor for this being of significant importa nce in the formation of Cypresshead kaolinite is the availability of dissolved Al and Si. Although dissolved Al is likely in Cypresshead Formation pore waters, freshwater concentration in equilibrium with kaolinite is typically on the order of 0.1 g/l to 1 g/l (Drever, 1982), with the dissolved concentration of silica commonly up to five orders of magnitude higher (1-100 mg /l; Hem, 1985). However, the concentration of dissolved aluminum in pore waters may be greatly increased by complexing with organic compounds, particularly monofunctional and difu nctional carboxylic acid s (acetic and oxalic
170 acids). Studies by Fein (1991) and Fein and Hest rin (1994) have demonstrated that aluminumoxalate complexation can dramatically increase aqueous aluminum concentrations. Subsequent destabilization of these aluminum complexes ma y result in the formation of kaolinite in siliciclastic units (Surdam et al., 1984; Small and Manning, 1994). With the Cypresshead Formation known to concentrate secondary organic co mpounds in basal pore waters as is seen at the EPK Mine site (Pirkle and Yoho, 1961), bacter ial degradation of organic compounds may be a likely mechanism by which aluminum is released (Maliva et al., 1999). Kaolinite recrystallization and disorder Other than the origin of kaolinitic clays associated with the Cypresshead Formation, the second most common question regarding clays in this unit is the unusually high degree of structural order exhibited by the kaolinite fraction relative to the small particle-size of these clays. As first noted by Pirkle (1960), it is th e small particle size of Cypresshead clays which contributes to the overall greater strength, plasticity, su rface area and base-exc hange capacity of these clays as an industrial mineral product. In fact variations in the struct ural order of kaolinite have been known for some time to correlate to industrial properties including plasticity, brightness and viscosity (Murra y and Lyons, 1956; Velho and Go mes, 1991; Chvez and Johns, 1995; Galn et al., 1998; Aparicio and Galn, 1999)). Based on th e results of this study, it appears that the in situ neoformed origin of kaolinite in th e Cypresshead is the answer to this question. If a major component of the kaolinite fraction in north-central Florid a sediments is derived via the combined topotactic and epitactic repl acement of mica at the expense of dissolved feldspar, then a high degree of crystalline order is to be expected, particularly if the greatest volume of kaolinite originates through epitaxi al overgrowth on mica template surface. Such a reaction pathway would favor the formation of a well ordered nucleated pha se, particularly when
171 the highly leached nature of Cypresshead sediments below th e water table is considered. With few competing cations (e.g., Fe2+) capable of incorporation in th e kaolinite structure at depth within the unit, high purity and resultant, well-ordered kaolinite is predicted. As evidenced by trends in CSD values, diso rder, and microtextura l characteristics of Cypresshead kaolinite, recrystallization processes app ear superimposed on kaolinite neoformation in near surface sediments. Under vadose or mixed vadose/saturated conditions, oxic weathering would prevail, favoring leaching of labile components, including clays, and the recrystallization of precu rsor phases (Hurst and Pickering, 1997). Through dissolution of Siand Al-bearing phases, mobilized Si and Al would be allowed to participate in vadose crystallization or could migrate vertically to below the ex isting water table. Evid ence for strong leaching conditions persisting in near surf ace sediments is confirmed by th e presence of gibbsite, which is evidence for laterization/feralization processes impacting Cypresshead sedi ments. Halloysite at the near surface, along with disordered kaolinite consistent with pedogenic formation, is further evidence for the extreme leaching, particularly at the north-central Florida localities which exhibited the greatest spread in disorder and CSD values (EPK, FRG, and SSD-1). The concentration of fine aggregates (or mi croaggregates) of a nhedral to subhedral kaolinite noted for near surface sediments in north -central Florida is further evidence for oxic weathering, and likely a pedogenic-re lated origin. Consisting primar ily of disordered and often single-phase kaolinite, th ese clays possess negative Eu anomalies consistent with formation and/or recrystallization from precursor phases, including kaolinite, unde r oxic conditions. This shift toward negative Eu anomalies in near surfa ce sediments correlates well with trends in CSD values and disorder (Fig. 5-14), with the near surface decrease in Eu/Eu* values corresponding to the oxidation of Eu2+ to Eu3+ during dissolution, inhibiting th e incorporation of Eu into
172 recrystallized phases. For near surface sediments associated with the Rio Jari kaolin deposit in Brazil, disordered kaolinite sh aring the same rounded aggregat e morphology corresponds with an increase in the amount of structural Fe substitu ted for Al in the kaolin ite (Montes et al., 2002). Although no measurements of Fe in Cypresshead kaolinites have been performed, incorporation of trace Fe could be possible due to the weathe ring of accessory heavy minerals. However, as noted previously, the halloysite morphology observe d in near surface Cypresshead sediments is inconsistent with elevated Fe concentrations. Of additional significance to observed trends in disorder and CSD values are the local hydrologic conditions existing at the various Cypr esshead localities evaluated for this study. As noted by Hurst and Pickering (1997), location of kaolins within the groundwater system can have a significant impact on the compositional variations of the deposit. This s eems particularly true for north-central Florida Cypresshead localities, where increased concentrations of detrital or residual clays in near surface sediments (FRL-1 and SSJ-1) appear to impede vadose and mixed vadose/saturated leaching and recrys tallization, resulting in a lack of significant spread in near surface disorder and CSD, and the absence of gibbsite, halloysite, and crandallite-florencite phases which appear to indicate strong, even lateritic/feralitic, leaching conditions. Accessory Phase Paragenesis Significant weathering reactions driving the post-depositional alteration of Cypresshead sediments occur under both saturated and unsatur ated groundwater conditions. Below or at the water table, the formation of halloysite as a ki netically favored metastab le precursor phase to kaolinite with subsequent recrystallization to disordered kaolinite as the thermodynamically stable phase is likely favored. The main factors favoring kinetic control of halloysite formation are intense, but short wet periods follow ed by prolonged extremely dry seasons, and microenvironmental conditions lead ing to immediate uptake of re leased Al by the halloysite-
173 R2 = 0.47 2.0 4.0 6.0 8.0 10.0 12.0 14.0 0 40. 60. 81 01. 21. 41 61 8Eu/Eu*Volume-Weighted Mean Thickness (nm) Cypresshead FL* Cypresshead GA Cypresshead FL* Figure 5-14. Correlation of Eu/Eu* values to CSD (volume-weighted mean thickness) calculations (* Florida Cypresshead Formation samples include SSD-1). precursor mineral allophane. Halloysite is known to be a common metastable precursor to kaolinite (Jeong, 1988a; Joussein et al., 2005), with the crystallization of metastable phases often favored in low-temperature geological envir onments. Although, nucleation of more soluble halloysite is kinetically easier due to a high surface free ener gy (Morse and Casey, 1988; Stumm, 1992), metastable phases such as halloysite will ultimately transform to stable phase such as kaolinite with time. In response to eventual dehydration and recrystallization, the halloysite crystallized in Cypresshead sediments with ultimately alte r into kaolinite via the halloysite metahalloysite kaolinite reaction pathway described by Hurst and Pickering (1997). This is consistent with thermodynamic considerations as the Gibbs free energy of formation of halloysite is approximately 4 k cal/mol higher than that of ka olinite (Anovitz et al., 1991). The crystallization of halloysite under near su rface water table condi tions is most likely related to the interplay of pH (H+ activ ity), and the activities of Al species, K+, and H4SiO4 under conditions favorable to repeated wetting an d drying (Ziegler et al., 2003). As indicated by
174 the curved or tubular morphology of Cypresshead halloysite, the relative uptake of tetrahedrally coordinated Al as a function of pH may also be of importance to the preferred cr ystallization of this phase, with the change in coordination stat e from an octahedral st ate taking place between pH 5.5 and 6.5 at 25C (Merino et al., 1989). With a sh ifting of this pH range to lower values at higher temperature, one would predict tetrahedra lly coordinated Al under near surface conditions in north-central Florida based on the pH range of 4.5 to 5.5 noted by Heuberger (1995) for soils developed on the Cypresshead Forma tion in the Ocala National Forest. Two likely mechanisms exist that could poten tially explain the origin of crandalliteflorencite phases ((Ca,REE)Al3(PO4)2(OH)6) in Cypresshead Form ation sediments; (1) diagenetic precipitation in response to organic decomposition, or (2) crystallization in response to the weathering of a precursor phosphate pha se (francolite). As originally proposed by Rasmussen (1996) for sandstones, the first mode l suggests that authigen ic precipitation could occur in response to organic matter decompositi on coupled with detrital mineral dissolution shortly after sediment burial. This seems unlikely as the sedimentary carbonate-fluorapatite, francolite, is kinetically favored for early diagen etic crystallization duri ng marine sedimentation (Flicoteaux and Lucas, 1984). Additionally, tra ce fossil assemblages found in Cypresshead sediments (Chapter 4) do not suggest a high organic matter accumu lation rate in these sediments during their initial depositi on. However, an alternative source of organics is the post-depositional concentration of pore water organics noted at the EPK Mine and othe r sites. Continuously delivered to the groundwater system via the surface decomposition and infiltration of organic detritus, pore water organics would periodi cally decomposed in response to water table fluctuations and corresponding re dox conditions. As suggested by the vertical distribution of P2O5 concentrations in most sampled sections (Tab le 5-4), this would lib erate phosphorous (P) to
175 surrounding pore waters near the water table interf ace where crandallite-florencite appears to be concentrated. Additionally, the disso lution of plagioclase feldspars, postulated as a source of Al, Si and the inherited positive Eu anoma lies associated with the formation of in situ kaolinite could potentially be the source of calcium (Ca) requi red for crandallite-florencite crystallization. As for the weathering of trace reworked phosphate (francolite), dissolu tion of francolite would supply both Ca and P to surrounding pore wate rs, with the addition of Al necessary for crandallite-florencite crystallization supplied by coeval clay, mica, and/or feldspar weathering (Flicoteaux and Lucas, 1984). Proposed original de position of minor francolite with Cypresshead sediments would be consistent with the minor accumulations of phosphate grains noted for southern Florida Late Miocene through Pliocene siliciclastics, where reworking of Hawthorn Group phosphate is the most likely source (Cunn ingham et al., 1998). However, geochemical evidence is lacking to support th is model given no negative Ce anomalies associated with the chondrite-normalized REE patterns of Florida Cypresshead clays. Conclusions Weathering processes are known have a profound impact on original sediments which host commercial kaolin deposits. Among the effects at tributed to the inte nse post-depositional alteration of such deposits are; (1) the strong compositional and textural modification of the original sediments, (2) whitening of the sediment s by partial removal of organic matter, Fe and Mn, and (3) recrystallizati on of kaolinite (Hurst and Pickering, 1997). In the case of Cypresshead Formation, complex hydrogeologic controls and post-depositional processes are ultimately responsible for the formation and recrystallization of the ka olinites. Focusing on the clay mineralogy and microtexture of Cypresshead sedi ments, the results outl ined in this chapter highlight evidence for the origin and weathering history of Cypresshead kaolinites in order to
176 define processes impacting their mineralogy and industrial properties, and answers the primary verses secondary origin issue associated with these clays. Among the results are the following: Cypresshead kaolinite consists of at least three fractions; in situ kaolinite formed at the expense of feldspars and mica, detrital kaolin ite deposited as part of the original clay mineral suite, and near surf ace recrystallized kaolinite. In situ kaolinite crystallizes via the combined topotactic (transformation) and epitactic (neoformation) weathering of muscovite mica consistent with the model of Jeong (1988b) and confirmed by vertical trends in kaolin ite disorder and cohe rent scattering domain (CSD) values. The muscovite weathering model of Jeong (1988 b) explains the unusual low disorder/small particle-size characteristics of Florida Cypre sshead kaolinite, the distribution of fine mica in coarseto medium-sands and gravels, and the relatively hi gh (> 20%) total clay content observed in some basal Cypresshead sediments. Feldspar dissolution as a source of Al a nd Si necessary for the crystallization of in situ kaolinite and post-depositional increase in Cypresshead clay content is confirmed by the presence of residual feldspars in basal Cypre sshead sediments and the inherited positive Eu anomalies likely resulting from an original plagioclase component to the feldspar suite. Pore water organics are like ly important in concentrati ng dissolved Al in order to crystallize in situ kaolinite and in co ntrolling the pore wate r redox environment of Cypresshead sediments as indicated by the vertical trends in Eu anomalies and HREE depletion. Near surface recrystallized ka olinite formed under oxic, vado se or mixed vadose/saturated conditions are disordered and possess microtextural characteristics consistent with a pedogenic origin. Occurrences of near surface formed halloysite and gibbs ite are consistent with Cypresshead sampling locations characterized by extensive evidence of vadose and mixed vadose/saturated leaching and recrystallization under oxic conditions. Trace crandallite-florencite minerals likely originated from the decomposition of postdepositional pore water organics couple d with detrital mi neral dissolution.
177 CHAPTER 6 TRACE ELEMENT AND ND ISOTOPIC EV IDENCE FOR THE PROVENANCE OF CYPRESSHEAD FORMATION KAOLINITIC SANDS Introduction The geological significance of kaolinitic sa nds occurring in the Pliocene Cypresshead Formation (3.4.3 Ma) of north-central Florida and southeastern Georgia has long been a question confounding Coastal Plain geologists. Although the formati on contains economic kaolin concentrations in north-central Florida, little is known concerning these deposits, particularly the provenance of the clay fraction. Overshadowed by the large economic kaolin deposits of the Georgia-South Carolina kaolin district (Fig. 3-1), the Cypres shead has received limited study, and remains poorly understood within the overall c ontext of Coastal Plain deposition. As is the case with the kaolin deposits in Georgia and South Carolina (Keller, 1978; Patterson and Murray, 1984; Dombrowski, 1992; Hurst and Pickering, 1997), several theories on the origin of these kaolinitic sands have been proposed, but have fail ed to fully define provenance. Existing data on the provenance of Cypresshead sediments prior to this study has been limited to information on monocrystalline and polycrystall ine quartz grain textures id entified by Kane (1984), which suggest sources consistent with granitoids, quartzo-feldspathic gneisses and /or schists, and heavy mineral suites collected from the Cypr esshead (Pirkle et al., 1964; Kane, 1984) which have supported the previous view of a mixed i gneous and metamorphic source terrane for these sediments. In this study, both trace element and neodymium (Nd) isotopic data were used to answer the question of Cypresshead provena nce with respect to possible crystalline (metamorphic and/or plutonic) source rocks in the southern Piedmont and reworked kaolinite from the Georgia-South Ca rolina kaolin district. In the past two decades, attention has focused on the use of trace elements, particularly the high-field-strength elements (HFSE) such as the ra re earth elements (REEs), Zr, Th, and U, and
178 Nd isotopic systematics as constraints on provenance (Tabbutt, 1990; Linn et al., 1991; Dombrowski, 1992; Toulkeridis et al., 1994; Condie et al., 19 95; Gleason et al., 1995). The REEs are often reliable geochemical tracers of sedimentary provenance because of their characteristic immobility and resistance to elemental fractionation in the supracrustal environment (Wildeman and Condie, 1973; Piper, 1974; Nesbitt, 1979; Chaudhrui and Cullers, 1979). However, evidence to the contrary of i mmobility has been given by Nesbitt (1979), Nesbitt and Markovics (1997), a nd Wade (2002). As shown by Nesbitt and Markovics (1997), careful monitoring of REE distributions in a w eathering profile reveals a tendency toward light (L) REE enrichment (relative to heavy (H) REEs) in the most extremely weathered materials of the profile. Therefore, profiles showing slight enrichment in LREEs can be explained by this mechanism. Additionally, as outlined in Chapter 5 of this study, variations in the redox potential of pore waters can have a significant impact on the REE distribution of fine-grained sediments. In general though, certain trace elements, such as the REEs with a valence of +3, Th, and Sc, and to a lesser extent Co and Cr, are believed to be transported quantitatively from source rocks into sediments (McLennan, 1989). These elements also are the least prone to diagenetic redistribution, thus providing useful information on the composition of sediment sources (Taylor and McLennan, 1985; McLennan, 1989). REE distribu tions appear to be sensitive to tectonic setting (Cullers et al., 1987; Cullers, 1988; McLennan, 1989; McLe nnan et al., 1993; and others), and are particularly useful when evaluating fine-grained sediments. Although many studies have used the distribution of REEs and other trace elements to constrain provenance (Dypvik and Brunfelt, 1979; Bhatia, 1985; Tabbutt, 1990), it is the Nd isotopic character of sediments which has receiv ed the most attention recently in identifying potential source terrains for clastic sediment s (Frost and O'Nions, 1984; Miller and O'Nions,
179 1984; 1985; Frost and Winston, 1987; Nelson an d DePaolo, 1988; Ghosh and Lambert, 1989; Bouquillon et al., 1990; Tabbutt, 1990; Linn et al ., 1991; Gleason et al., 1994; 1995). Because weathering and diagenesis can affect Sm-Nd is otopic ratios in sediments (Ohr et al., 1991; McDaniel et al., 1994; Bock et al., 1994), spurious Nd model ages (TDM) may result. Therefore, initial Nd values ( Nd(t)) calculated for the stratigraphic age of a sediment are often emphasized. This approach assumes that any diagenetic effect s resulting in the disturbance of Sm-Nd isotopic ratios would have occurred near the time of deposition; thus, calculated Nd(t) values should reflect true Nd of the material at that time (Gl eason et al., 1995). Furthermore, using Nd notation permits the comparison of data sets that are nor malized to different Nd isotopic ratios without accounting for normalization differences. Southern Piedmont The deeply weathered rocks of the Piedmont, stretching from Alabama in the southwest to southernmost New York in the northeast, have long been considered a major source for sediments deposited along the Atlantic Coastal Plain, and as the ultimate source of sediments which are now the Cretaceous soft and Tertiary hard kaolins in central and eastern Georgia (Murray, 1976; Keller, 1977; Austin, 1978; Hu rst and Pickering, 1989). Weathering of these crystalline rocks produced kaolin ite-metahalloysite-rich saprolites, which on erosion produced detritus delivered to the ancient Georgia coastline via fluvial transport. Subs equent laterization of these sediments played an important role in kao lin formation (Austin, 1978), with periods of sea level regression favoring intens e weathering, particularly in updip sediments (Lowe, 1991). The southern Piedmont can be divided into tectonostratigraphic terranes on the basis of age, origin, and shared affin ity, and further divided into metamorphic belts based mainly on lithological characteristics, metamorphic grade, a nd in some areas, structural boundaries. For this study, the terranes of interest incl ude the Carolina terrane, the I nner Piedmont, and Eastern slate
180 belt (Fig. 6-1). Possible source ro cks for kaolinitic sediments include both metamorphic and igneous rocks within these terra nes. Of particular interest are the Carboniferous-Permian Alleghanian (285-340 Ma) granites found throughout the southern Piedmont (Fig. 6-1), which are but one of three groups of plutons emplaced in the area. The other groups include a Late Proterozoic-Cambrian (495-734 Ma) mafic to felsic calc-alkaline series and a Silurian-Devonian (436-375 Ma) mafic-felsic bimodal series with al kalic tendencies, both occurring mainly in the Charlotte belt portion of the Carolina terrane (McSween et al., 1991). Georgia-South Carolina Kaolin District The Cretaceous and Tertiary kaolin-bearing sedimentary rocks of the Georgia-South Carolina kaolin district lie to the southeast of the Fall Line, the contact between the crystalline rocks of the southern Piedmont and the sediment ary assemblage of the Atlantic Coastal Plain. A review of the stratigraph y associated with the Cretaceous a nd Tertiary kaolins of the GeorgiaSouth Carolina kaolin district is given by Pi ckering and Hurst (1989) and Dombrowski (1992). Dombrowski (1992; 1993) identified La, Th, Co, a nd Sc as effective provenance indicators for Cretaceous and Tertiary kaolins based on the concentr ation of Th and La in felsic (acidic) source rocks, and Sc and Co in mafic (basic) source ro cks, as well as the relative immobility of these elements during weathering processes. As such, results from the work of Dombrowski (1992; 1993) are addressed in the disc ussion section of this study. The Cretaceous-Tertiary boundary is a major unconformity in the Georgia-South Carolina kaolin district (Buie and Fountain, 1967; Murr ay, 1976; Patterson and Murray, 1984; Nystrom et al., 1986; Pickering and Hurst, 1989), marking th e boundary between Cretaceous "soft" kaolins and Tertiary "hard" kaolins. Above the unconfor mity in middle and eastern Georgia and South Carolina are Paleocene through Middle Eocene sediments assigned to the Huber Formation (Buie, 1978). Nystrom et al. (1986) restricts this term to Early to Middle Eocene sediments.
181 Figure 6-1. Tectonostratigraphic te rranes and granites proposed as potential source materials for Cypresshead Formation sediments (modified after Horton et al., 1989; Samson et al., 1995). Commercial kaolin deposits of Tertiary age are concentrated near the t op of the Huber Formation and are elongated from northeast to southwest, parallel to the or ientation of the paleoshoreline (Dombrowski, 1992). Below the unconformity in Georgia and South Carolina lie Late Cretaceous sediments composed of continental to near-shore marginal marine deposits adjacent to the Fall Line, with marine deposits in the remainder of the region (Kesler, 1956; Herrick, 1961; Herrick and Vorhis, 1963; Applin and Ap plin, 1964; Gohn et al., 1979; Nystrom et al., 1986; and others). Those sediments located adjace nt to the Fall Line ar e referred to as the Buffalo Creek Formation west of the Ocmulgee Ri ver in western Georgia (Pickering and Hurst, 1989) and as "Cretaceous Undifferentiated" or "Unnamed Cretaceous Sediments" east of the river in South Carolina (Ear gle, 1955; Nystrom et al., 1986; Pickering and Hurst, 1989).
182 Results Provenance investigations of the kaolin deposits of the Georgia-South Carolina kaolin district have focused on both plutonic and metamor phic source rocks in the southern Piedmont as primary sources for kaolinitic sediments deliver ed to these locations (Dombrowski, 1982; 1992; Dombrowski and Murray, 1984). A similar provenance is possible for the kao linitic sands of the Cypresshead Formation in southeastern Georgia and peninsular Florida, or that kaolinite accumulated in response to reworking of kaolin di strict sediments. In order to answer that question, the trace elemental and Nd isotopic char acteristics of possible source rocks in the southern Piedmont and Georgia-So uth Carolina kaolin district are used in conjunction with similar data from Cypresshead clays to deci pher characteristics of sediment provenance. Trace Elements Trace element concentrations of Cypresshead Formation clay (< 2 m) separates and comparison samples are reported in Table 6-1 along with select major elements, REE, and applicable elemental ratios. A complete summary of all trace elements anal yzed for this study is included in Appendix H. In comparison with aver age continental crust (ACC), the concentrations of most of the trace elements for Florida Cypresshead samples are relatively low, except for Pb, Th, and U (Fig. 6-2A). Rb and Co are particular ly depleted, averaging a relative concentration ratio of ~ 0.1 or greater. Zr is also relatively depleted in most samples other than the Davenport Mine composite. This sample possesses an anom alously high Zr concentration in response to zircon enrichment in several of the near surface samples used to calculate the average used in Figure 6-2A. Trace element concentrations for Georgia Cypresshead samples are slightly more enriched than Florida samples (Fig. 6-2B). As with the Florida samples, Pb, Th, and U remain enriched relative to ACC while Cr, V, and Rb, although depleted to equi valent with ACC, are more enriched relative to Florida samples. Sr, however, is more depleted in Georgia samples
183Table 6-1. Trace element concentrations (ppm) and elementa l ratios for Cypresshead Formation and comparison samples. Sample ID IntervalRbSrThUZrScVCrCoTh/ScTh/UZr/ScLa/ScLa/ThP2O5 (%)TiO2 (%)Fe2O3 (%) REEEu/Eu*(La/Yb)n(La/Sm)n(Gd/Yb)nCypresshead Formation FL EPK36-J-1225-272.7220.127.116.1145.2324433.11.72.015.18.104.22.168.250.2775.240.7130.765.163.75 EPK36-J-1235-40312174.7973.751822.214.171.124126.96.36.199.150.230.36129.760.6924.744.254.32 EPK36-J-1246-4188.8.131.52.063210364184.108.40.206.22.34.00.070.280.4993.950.94220.127.116.11 EPK36-J-1250-518.104.22.168.9324.9103222.214.171.124.52.05.70.050.320.5176.260.9433.435.613.25 EPK36-J-1259-6126.96.36.199.863816604188.8.131.52.42.25.00.070.550.77177.871.4328.752.915.48 EPK31-P-4035-454.521.34.717.722.212524184.108.40.206.91.74.30.060.350.6685.861.4227.024.993.24 EPK31-P-4050-6220.127.116.110318.52147212.30.20.00.91.36.40.070.310.84153.411.7114.563.123.10 EPK31-P-4062-6517.61237.892.930.52077618.104.22.168.22.214.171.124.291.24294.021.4210.582.103.71 EPK30-V-622-2425.734815.15.581776636126.96.36.199188.8.131.52.220.592.55275.150.6013.993.263.10 EPK30-V-630-3523.1204056.946.325513794184.108.40.206.6220.127.116.11.301.531367.540.6512.792.493.81 EPK30-V-648-5312.75492329.31511279318.104.22.16822.214.171.124.630.420.89447.190.7514.662.843.67 EPK30-V-658-63619.83.818.923.114844126.96.36.199.72.28.00.050.250.41117.751.3640.665.524.75 EPK30-V-668-738.925.94.316.82322113188.8.131.52.01.15.80.060.430.67116.171.3512.763.562.27 FRG-134.622917.75.8184.9761594.62.53.012.184.108.40.2060.312.99215.860.6317.433.293.68 FRG-160.812561.6219.1516220.127.116.11.87.26.00.160.070.99177.950.8940.123.477.08 FRG-171.8767.41.1616.4418418.104.22.168.17.84.20.090.080.64138.990.9252.164.387.33 FRG-1106.745411.71.5724.9621103.52.07.54.22.214.171.124.220.61654.711.0792.953.7113.39 FRG-1111.7126.96.36.19917.7516188.8.131.52.58.15.70.070.110.23162.921.1490.505.579.33 FRG-1131.733.15.61.0421.5518184.108.40.206.34.13.60.050.120.2178.111.4367.876.595.94 FRG-1152.3220.127.116.1118.2112718.104.22.168.75.08.50.080.230.23259.131.3492.114.4210.44 FRG-231.722.214.171.12422.7511126.96.36.199.52.92.70.060.110.2761.721.0947.813.048.01 FRG-252.428.15.91.0116.35203188.8.131.52.34.33.70.050.110.3694.431.3772.224.948.05 FRG-272.3184.108.40.2065.6716220.127.116.11.22.44.40.050.150.2778.451.3957.174.067.45 FRG-2105.866.9102.2723.3113918.104.22.168.22.214.171.124.270.63265.031.3735.043.057.28 FRG-2121099.294.6229.915574126.96.36.199.03.65.90.140.361.40299.291.0522.282.446.38 FRL-188.8.131.52.28461450503.70.66.23.32.23.80.060.310.83114.001.1550.326.404.14 FRL-144.726.810.61.1649.614615184.108.40.206.53.34.30.060.320.94187.781.2760.856.105.26 FRL-163.4220.127.116.113.3153518.104.22.168.93.05.40.070.250.61239.121.1722.214.171.124 FRL-173.6126.96.36.19932.11330188.8.131.52.52.65.50.050.220.59176.471.0937.343.456.04 SSJ-125.88729.43.112165789184.108.40.206220.127.116.11.121.153.22186.900.6217.1010.041.13 SSJ-148.318.104.22.168117661522.214.171.124126.96.36.199.140.582.14123.320.6218.056.891.63 SSJ-165.949.22.23.0318.8627188.8.131.52.11.84.90.060.160.4741.951.4635.785.503.55 SSJ-1810.8184.108.40.20623.912564220.127.116.11.00.84.00.050.250.4036.911.5832.106.583.11 SSJ-1101143.24.614.133.71885318.104.22.168.90.93.60.060.410.5167.291.4427.425.623.01 SSD-1121.377714531.31630222451605.56.64.6722.214.171.124.987.972.121886.800.5516.826.771.50 SSD-135.214456.87.541807417126.96.36.1995.7188.8.131.520.720.56491.740.5451.415.714.79 SSD-164.141.2101.6434.8844503.41.36.14.184.108.40.206.230.66221.751.0347.922.889.38 SSD-174.5416.82.7128.8104516220.127.116.11.07.40.080.260.73276.661.0855.842.7811.16 SSD-1107.8384.42.042393318.104.22.168.62.55.20.080.231.04123.001.2630.493.065.90 TRF2214 60.0-62.529.26610.62.4312715119603.60.74.48.52.02.80.070.841.79103.820.6512.337.581.27 WEX164 18.0-26.087.732022.313.363.92618916910.20.91.72.22.214.171.1240.844.35333.070.699.984.311.84 WEX366 9.0-10.012411826.69.9238231831126.96.36.1990.32.01.70.131.219.42165.970.6211.017.111.26 Cypresshead Formation GA J-1223.335.936.65.21231211951106.31.77.011.02.31.30.111.407.65177.860.6112.6510.470.83 J-1439.434.821.23.0474.724190775.40.97.03.12.62.90.081.034.48239.900.6846.218.532.88 J-1628.223.914.42.0357.5176745188.8.131.52.03.60.030.771.90255.250.7249.205.394.82 L-1317.440.534.32.8710419100805.81.812.05.58.34.60.111.031.72649.570.7288.0411.473.49 L-156.410.710.81.1638.710685184.108.40.206.94.74.40.040.370.84165.260.6878.9113.872.97 L-1621.528.816.61.8560.31680246.11.09.03.87.06.70.060.851.62365.180.6693.6215.713.36 B-1297.452.615.94.0182.325149847.30.64.03.32.23.40.090.675.62210.410.6721.206.931.80 B-1311677.420.45.5583.6311989220.127.116.11.18.104.22.168.7210.91282.200.7016.365.161.94 B-1588.511218.54.0374.4361683080.54.62.12.54.80.090.743.63466.080.7810.862.852.43 Hawthorn Group, Coosawhatchie Formation FL/GA MCB109 15.0-20.013911212.74.3884.92117712722.214.171.124.02.74.40.080.695.98226.610.6810.465.511.44 J-1BC42.960.811.31.7237.62213965126.96.36.199.74.38.30.030.626.03496.850.776.033.011.63 Huber Formation GA KGa-21.251.2133.5278.6151052188.8.131.52.23.13.60.051.430.94173.530.7717.356.701.86 ECCI-CB2.6184.108.40.20651.92076220.127.116.11.62.22.90.061.110.90198.540.7232.022.835.81 Buffalo Creek Formation GA KGa-18.104.22.16898.119223383.31.916.05.21.80.90.061.570.19194.950.7437.561.7814.75 ECCI-BC50.531.41.531151412722.214.171.124.24.21.90.081.250.16434.500.5665.871.5218.60 TKC-EA0.636.731.18.72111211022126.96.36.199.30.90.60.061.450.1787.860.5925.682.7610.04 DBK-B930.3188.8.131.52149222499011.54.96.81.71.10.072.210.24262.400.6324.471.3610.20 Note: Detailed major element and REE data ( REE in ppm) are included in Tables 5-4 and 5-6 of Chapter 5.
184 Figure 6-2. Multi-element normalized diagrams for Cypresshead and comparison samples, normalized against average continenta l crust (Wedepohl, 1995). A) Florida Cypresshead composite samples, B) Geor gia Cypresshead composite samples, C) Cypresshead Formation (FL and GA) composite and comparison units.
185 relative to Florida. In comparing a composite of all Cypresshead Formation samples to Hawthorn Group, Huber Formation, and Buffalo Creek Formation composites (Fig. 6-2C), the most notable difference is related to the si gnificant depletion of Rb for both the Huber and Buffalo Creek composites. For comparison to the work of Dombrowski (1992; 1993), the concentration ranges of Sc, Co, and Th for Florida Cypressh ead samples are 3 ppm, 1.3.2 ppm, and 2.2.0 ppm respectively. Georgia Cypresshead samples possess concentrati on ranges of 10 ppm, 5.4.4 ppm, and 10.8.6 ppm. A detailed summary of REE distributions in Cypresshead clays is discussed in Chapter 5. A correlation matrix based on the trace elements, major elements, and elemental ratios of all Cypresshead Formation samples reported in Ta ble 6-1 is included in Table 6-2. Significant correlations (r > 0.6) highlight the effect of accessory mineral phases on the trace element concentrations of Cypresshead clays, and include correlations between P2O5 and Th (r = 0.92), Zr (r = 0.76), and Sr (r = 0.63). TiO2 also correlates with both Th (r = 0.90) and Zr (r = 0.98), but can be differentiated from P2O5 by additional correlations with V (r = 0.72) and Cr (r = 0.63). Fe2O3 exhibits significant correla tions with Rb (r = 0.90), V (r = 0.66), Cr (r = 0.69), and Co (r = 0.64). The correlation between P2O5 and REE was previously addressed in Chapter 5. Additionally, the geochemical results for CMS So urce Clays (KGa-1 and KGa-2) are considered to be of questionable quality due to uncertain ty surrounding the preparation of these standards which were originally generated for co mparative X-ray diffraction (XRD) studies. Neodymium (Nd) Isotopes Research has shown that the Nd isotopic composition of sedimentary rocks reflects the integrated composition of cont ributing source terranes (Nelson and DePaolo, 1993; McLennan et al., 1990; Linn et al., 1991). For this study, Nd isot ope data from the Cypresshead Formation and comparison clay-bearing strata (Hawthorn Gr oup, Huber Formation, and Buffalo Creek
186 Table 6-2. Correlation matrix of select trace elements and elemental ratios for all Cypresshead Formation samples. RbSrThUZrScVCrCoTh/ScTh/U Rb1.00 Sr0.181.00 Th 0.220.581.00 U 0.100.300.191.00 Zr 0.190.420.940.161.00 Sc 0.570.130.260.500.291.00 V 0.710.310.650.260.680.761.00 Cr 0.680.240.640.160.600.510.861.00 Co 0.640.420.460.090.350.290.600.771.00 Th/Sc0.020.460.780.020.61-0.180.280.470.441.00 Th/U-0.18-0.120.15-0.450.08-0.41-0.200.010.080.411.00 Zr/Sc0.100.420.870.060.870.000.520.590.430.830.24 La/Sc-0.040.650.710.030.57-0.200.170.250.370.800.37 La/Th-0.29-0.02-0.270.12-0.230.03-0.29-0.48-0.34-0.38-0.13 P2O5 (%) 0.130.630.920.220.760.170.480.520.440.820.08 TiO2 (%) 0.210.290.900.150.980.360.720.630.320.530.07 Fe2O3 (%) 0.900.150.270.050.250.410.660.690.640.19-0.02 REE 0.130.790.900.290.820.270.530.440.400.630.05 Eu/Eu*-0.36-0.40-0.520.21-0.410.13-0.37-0.54-0.72-0.66-0.26 (La/Yb)n-0.40-0.24-0.23-0.41-0.25-0.47-0.56-0.50-0.42-0.110.55 (La/Sm)n0.18-0.190.22-0.270.26-0.060.280.360.100.320.35 (Gd/Yb)n-0.40-0.12-0.29-0.29-0.31-0.42-0.59-0.57-0.39-0.200.37 Cypresshead Formation samples include reworked Cy presshead sediments collect ed at the SSD-1 locality. Table 6-2. (continued). Zr/ScLa/ScLa/Th P2O5 (%)TiO2 (%)Fe2O3 (%) REEEu/Eu* (La/Yb)n(La/Sm)n(Gd/Yb)nRb Sr Th U Zr Sc V Cr Co Th/Sc Th/U Zr/Sc1.00 La/Sc0.671.00 La/Th-0.390.121.00 P2O5 (%) 0.730.76-0.221.00 TiO2 (%) 0.820.49-0.200.691.00 Fe2O3 (%) 0.280.03-0.400.110.261.00 REE 0.700.800.050.860.750.131.00 Eu/Eu*-0.63-0.530.45-0.51-0.34-0.49-0.401.00 (La/Yb)n-0.300.160.48-0.18-0.25-0.46-0.140.331.00 (La/Sm)n0.390.04-0.440.050.300.28-0.07-0.220.011.00 (Gd/Yb)n-0.390.180.65-0.21-0.32-0.45-0.060.300.83-0.451.00 Cypresshead Formation samples include reworked Cy presshead sediments collect ed at the SSD-1 locality. Formation) were corrected to their stratigraphi c age as shown in Table 6-3 to calculate for Nd(t). For the Cypresshead Formation, stratigraphic ages of 3.3 Ma and 2.5 Ma were used for Florida and Georgia samples, respectively.
187 Table 6-3. Nd isotope data for Cypresshead Formation and comparison samples. Sample IDIntervalSm* (ppm)Nd* (ppm)Sm/Nd147Sm/144Nd143Nd/144Nd0Error x 10-6Nd(0)143Nd/144NdtNd(t)TDM (Ga) Cypresshead Formation FL (3.4.8 Ma: Fountain, this study) EPK36-J-1225-2184.108.40.206.09670.51229112-6.8 0.512288 -6.71.0 EPK36-J-1235-404.3220.127.116.1100.5122496-7.6 0.512247 -7.61.1 EPK36-J-1246-482.718.30.150.08880.5122666-7.3 0.512264 -7.21.0 EPK36-J-1250-518.104.22.168.09960.51229310-6.7 0.512290 -6.71.0 EPK36-J-1259-627.336.00.200.12210.5123564-5.5 0.512353 -5.51.2 EPK30-V-622-2410.322.214.171.12470.5122154-8.2 0.512213 -8.21.2 EPK30-V-630-3560.2306.00.200.11850.5122465-7.6 0.512243 -7.61.3 EPK30-V-648-5319.5126.96.36.19970.5122595-7.4 0.512256 -7.41.4 EPK30-V-658-633.419.00.180.10770.5123725-5.2 0.512369 -5.21.0 EPK30-V-668-734.3188.8.131.5280.5123794-5.1 0.512376 -5.01.1 FRG-1 38.3184.108.40.20680.5121835-8.9 0.512180 -8.81.3 FRG-1 66.4220.127.116.1120.5122526-7.5 0.512250 -7.51.1 FRG-1 74.426.60.170.09960.5122465-7.7 0.512243 -7.61.1 FRG-1 1023.1139.00.170.10010.5122605-7.4 0.512258 -7.31.1 FRG-1 114.530.30.150.08940.5122744-7.1 0.512272 -7.01.0 FRG-1 131.914.20.130.08060.5122885-6.8 0.512286 -6.80.9 FRG-1 157.747.30.160.09800.5123456-5.7 0.512343 -5.70.9 FRL-1 44.630.90.150.08960.5123096-6.4 0.512307 -6.40.9 FRL-1 68.718.104.22.16860.5123566-5.5 0.512353 -5.51.0 FRL-1 76.037.40.160.09660.5123325-6.0 0.512330 -5.91.0 SSD-1 313.380.50.170.09950.5121915-8.7 0.512189 -8.71.1 SSD-1 722.214.171.124.11420.5123383-5.8 0.512336 -5.81.1 SSD-1 104.6126.96.36.19960.5123436-5.8 0.512340 -5.71.2 SSJ-1 42.918.00.160.09700.5121717-9.1 0.512169 -9.11.1 SSJ-1 188.8.131.52.10470.5122257-8.1 0.512222 -8.01.2 SSJ-1 101.811.10.160.09760.5123156-6.3 0.512313 -6.31.0 TRF221460.0-62.52.4184.108.40.20680.51214021-9.7 0.512138 -9.71.2 WEX16418.0-26.010.9220.127.116.1140.51223814-7.8 0.512235 -7.81.2 WEX3669.0-10.04.023.90.170.10080.51216011-9.3 0.512158 -9.31.2 Cypresshead Formation GA (3.15-2.3 Ma: Fountain, this study) J-1 44.532.30.140.08390.5122535-7.5 0.512252 -7.51.0 J-1 65.941.50.140.08560.5122805-7.0 0.512279 -6.90.9 L-1 52.117.00.120.07440.5122995-6.6 0.512298 -6.60.8 L-1 64.436.10.120.07340.5122754-7.1 0.512273 -7.00.9 B -1 37.646.90.160.09760.51215719-9.4 0.512155 -9.41.2 B-1 519.318.104.22.16880.5122074-8.4 0.512205 -8.41.5 Hawthorn Group, Coosawhatchie Formation FL/GA (13.5-15 Ma: Huddlestun, 1988) MCB10915.0-20.06.336.00.180.10540.51217716-9.0 0.512167 -8.81.2 J-1 BC19.222.214.171.12490.5123474-5.7 0.512336 -5.51.2 Huber Formation GA (40-50 Ma: Al-Sanabani, 1991; Dombrowski, 1992; Pickering et al., 1997) KGa-2a---4.3126.96.36.19970.5123115-6.40.512281-5.81.0 ECCI-CBa---9.441.00.230.13800.5124935-2.8 0.512453 -2.51.1 Buffalo Creek Formation GA (70-80 Ma: Tschudy and Patterson, 1975; Nystrom et al., 1986) KGa-1a---11.741.70.280.16890.5122824-6.9 0.512199 -6.72.4 ECCI-BC---24.0131.00.180.11030.5122163-8.2 0.512162 -7.41.2 TKC-EA ---4.312.60.340.20550.5122525-7.5 0.512151 -7.67.8 DBK-B93---16.6188.8.131.5260.5122533-7.5 0.512185 -6.91.6 Note: Initial Nd isotopic values have been normalized to 146Nd/144Nd = 0.7219. Errors on 143Nd/144Nd measurements are 2 Measured Nd values calculated assuming a present-day 143Nd/144Nd(CHUR) = 0.512638 (Jacobsen and Wasserberg, 1980). Analysis by LiBO2 fusion ICP-MS. a Average of duplicate analyses (KGa -1 and KGa-2, n=2; ECCI-CB, n=3). Calculated Nd(t) values for both Florida and Geor gia Cypresshead Formation samples exhibit moderate variation, ranging between Nd(3.3 Ma) = -5.0 and -9.7 and Nd(2.5 Ma) = -6.6 and -9.4 respectively. Corresponding Nd(t) average values are -7.1 and -7.6, and show consistency between the sample groups. The variation noted for Nd(t) values in Florida samples
188 appears to exhibit a vertical trend, with near surface Nd(t) more negative than samples collected from increasing depth. Sm/Nd ratios range fr om 0.13 to 0.21 and 0.12 to 0.22 for Florida and Georgia samples respectively. Cypres shead Formation samples plotted on a 143Nd/144Nd versus 147Sm/144Nd isochron diagram (Fig. 6-3) exhibit no linear trends suggest ive of diagenetic resetting. Additionally, the lack of variation noted for Nd(t) when plotted against Nd concentration (Fig. 6-4) indicates there is no evidence for a multiple component Nd isotopic signature in Cypresshead samples. Rather, Cypr esshead samples appear to originate from a single, well mixed source. Nd model ages based on Nd(t) results exhibit a high degree of consistency (Fig. 6-5) for Florid a and Georgia samples as well. TDM values for Florida and Georgia Cypresshead samples range between 1.4 -0.9 Ga and 1.5-0.8 Ga respectively, with both possessing an average TDM value of 1.1 Ga. Comparison clay samples from the Hawthorn Gr oup in Florida and Georgia, and samples of kaolin from the Huber and Buffalo Creek form ations in Georgia possess similar Nd isotopic characteristics to Cypresshead samples. Hawthorn and Huber Nd(t) data corrected for stratigraphic age exhibit m oderate variation as noted with Cy presshead samples (Table 6-3). For the two Hawthorn samples evaluated, Nd(14.25 Ma) values area -5.5 and -8.8, and for the two Huber samples, Nd(45 Ma) values are -2.5 and -5.8. As for the four Buffalo Creek samples evaluated, Nd(75 Ma) values are less varied, ranging from -6.7 to -7.4. However, two of the Buffalo Creek samples exhibit anomalous Sm/N d ratios of 0.28 and 0.34. Correspondingly, these two samples also exhibit spurious TDM ages of 2.4 Ga and 7.8 Ga. The remainder of the comparison samples exhibit TDM ages consistent with results from the Cypresshead, with the remaining TDM ages ranging between 1.0 Ga and 1.6 Ga.
189 Figure 6-3. 143Nd/144Nd versus 147Sm/144Nd isochron diagram for Cypresshead Formation samples. Figure 6-4. Nd(t) versus Nd concentration scatterplo t for Cypresshead Formation samples.
190 0 2 4 6 8 10 120.7 0.8 0.9 1.0 1.1 1.2 1.3 1.4 1.5 1.6TDM(Ga)Frequency Buffalo Creek Fm.* Huber Fm. Hawthorn Gr. Cypresshead GA Cypresshead FL Figure 6-5. Histogram illustrating th e distribution of Nd model age (TDM) results for both Cypresshead Formation clay samples and comparison formations (* spurious TDM values associated with samples KGa-1 and TKC-EA not included with Buffalo Creek Formation results). Discussion Trace elements are the primary geochemi cal tools used for paleoenvironmental interpretation, yet many are mobile under vari able redox conditions accompanying diagenesis and weathering. Thus, trace element and Nd isotopic results for Cypresshead Formation clays are interpreted in light of the apparent mobility of many trace elements (including REEs) in weathering profiles developed on Cypresshead Formation sediments and materials similar to proposed sources (Cullers, 1988; Condie et al., 1995; Lev et al., 1998). Although many trace element systems previously assumed to be sens itive to source composition and immobile during weathering have proven to be unr eliable for provenance studies (Condie et al., 1995), several do offer a means of interpreting kaolinite provenan ce or weathering and/or diagenetic pathways.
191 Trace Element Mobility and Enrichment HFSEs such as Zr, Th, and U are preferentially partitioned into melts during crystallization (Feng and Kerrich, 1990). As such, these elements are enriched in felsic rather than mafic sources, and are thought to reflect provenance com positions as a consequence of their relatively immobile character (T aylor and McLennan, 1985). For most upper crustal rocks, Th/U is typically 3.5-4.0 (McLennan et al., 1993), with weathering an d sedimentary recycling under oxidizing conditions responsible fo r mobilization and loss of U (U6+). As such, weathering elevates Th/U ratios (Fig. 6-6A), although the add ition of heavy minerals (zircon) or secondary phosphates can complicate this trend. In the case of Florida Cypresshead clays, a significant proportion of the samples analyzed possess Th/U ratios well below that of average upper crust (Fig. 6-6A). This appears most likely a consequence of redox-dr iven U enrichment dictated by pore water organics associated with basal Cypresshead sediment s. As seen in Table 6-1, low Th/U ratios are concentrated at th e base of most sample locations where pore water organics and corresponding reducing conditions would be most favored. Additi onally, there appears to be no correlation between Th/U values and either Zr or P2O5 concentrations, ruling out a role for U enrichment associated with either detrital zi rcons or secondary cranda llite-florencite. Georgia Cypresshead samples correspond to the weathering trend in Figure 6-6A as would be predicted given the highly oxidized appearance of these sample localities. Since Zr is preferentially concentrated in the detrital heavy mineral zircon, which is known to occur in Cypresshead sediments (Chapter 4), it is possible to evaluate the role of heavy mineral concentration during sedimentary sorting of the unit (McLennan et al., 1993). For this purpose, the Zr/Sc ratio is plotted against Th/S c (Fig. 6-6B). Zr/Sc is an index for zircon enrichment given the concentrati on of Zr in zircon and the corres ponding behavior of Sc, which
192 Figure 6-6. Trace element plots of Cypresshead Formation clays illu strating evidence of weathering, provenance, and sedi ment recycling processes. A) Th/U versus Th (ppm), and B) Th/Sc versus Zr/Sc. is not enriched but tends to preserve the provenance signature of its source. In contrast, the Th/Sc ratio is thought to be particularly sensitive to average provenance due to the incompatibility of Th and the relative compatibility of Sc, and the fact that both elements tend to be transferred
193 quantitatively into terrigenous sediments duri ng sedimentation processes (Taylor and McLennan, 1985; McLennan et al., 1990), although enrichments can also be indicative of a relatively severe weathering regime (McLennan et al., 1993). The trend illustrated in Figure 6-6B supports a mix in provenance compositions supplying Cypre sshead sediments based on the sympathetic relationship between Zr/Sc and Th/Sc at low Zr /Sc values. For higher Zr/Sc values above 10, Th/Sc ratios increase at a much lower rate cons istent with zircon enrichment. Such a conclusion is in agreement with X-ray diffraction and REE da ta presented in Chapter 4 and Chapter 5 which indicate near surface enrichment of many C ypresshead sample locations by detrital heavy minerals. Comparison to Georgia-South Carolina Kaolin Provenance Based on the assumption of relative immobility of Co, La, Sc, and Th, Dombrowski (1992; 1993) focused on the use of these elements as provenance indicators for the commercial kaolin deposits located in the Georgia-So uth Carolina kaolin district. REE (La) and Th abundances are believed to be higher in felsic source rocks a nd their weathering products, whereas Sc and Co are more concentrated in mafic sources. As a result Dombrowski was able to discriminate between felsic and mafic sources based on preferential en richment. For kaolins derived from granite and gneiss source rocks adjacent to the kaolin district, Th (25.5.6 ppm) and La (46.0 ppm) concentrations we found to be high relati ve to Co (7.5.2 ppm) and Sc (13.8.3 ppm). In contrast, kaolins derived from metavolcanic s ources (e.g., Little River Group) exhibited the opposite characteristics, with Th (5.6.7 ppm) and La (17.5 ppm) possessing relatively low concentrations as compared to Co ( 18.6.5 ppm) and Sc (13.8.9 ppm). This allowed Dombrowski to propose that felsic lithologies (granites/gneiss) ar e the principle source rocks for Cretaceous soft kaolins in both central Georgi a and South Carolina, with eastern Georgia and South Carolina Tertiary hard kaolins origin ating from a mixed meta volcanic/felsic (70/30)
194 source, and middle Georgia Tertiary kaolins derived almost exclusively from metavolcanic sources. Comparing the results of Dombro wski (1992; 1993) to those in this study, it is apparent that this approach is not suitable as a basis for provenance determination father south on the Florida peninsula. Th, Sc, and Co results for Cypr esshead samples illustrate this problem (Fig. 67). For Cypresshead samples, the fields correspon ding to Sc and Th values for felsic (soft kaolin) and metavolcanic (hard ka olin) sources overlap convincingl y with the results from this study (Fig. 6-7A). However, the range in values for Sc noted with the Dombrowski (1992; 1993) study far exceeds values noted for either Florid a (max = 26 ppm) or Georgia (max = 36 ppm), even if the bulk of Cypresshead clays were derived from metavolcanic sources. Complicating this interpretation is the occurr ence of both heavy minerals (zir con and rutile) and crandalliteflorencite in Cypresshead sediments. Both pha ses correlate with Th, potentially impacting relative concentration of the element beyond what might be indicative of provenance (e.g., SSD1-1). The plot of Th versus Co shows a far wors e fit to Cypresshead data (Fig. 6-7B). Whereas Th concentrations do fall within the range of values reported by Domb rowski (1992; 1993) for both felsic and metavolcanic sources, Co values for Cypresshead samples are extremely depleted. In fact, if a component of Cypresshead clays or iginated from a metavolcanic or reworked hard kaolin source, then Co has been significantly mobilized and removed from the resultant deposit. As such, Co appears to be an unsatisfactory indi cator of Cypresshead provenance. La exhibits the same characteristics as Co, and is highly de pleted in Cypresshead samples relative to concentrations reported by Dombrowski (1992; 1993). Cypresshead Provenance Of the trace elements reported in the Dombrowski studies, Th and Sc appear to be the most resistant to depletion relative to concentrations consistent with potenti al provenance sources.
195 Figure 6-7. Comparison of Cypresshead Formation trace element concentrations to the results of Dombrowski (1992; 1993). A) Th versus Sc plot illustrating the fields for metavolcanic and felsic sources relative to Cypresshead results (arrow indicates the field for metavolcanics extends beyond the concentration range shown on the plot), B) Th versus Co plot illu strating the fields for meta volcanic and felsic sources relative to Cypresshead results (arrows indi cates the fields for both metavolcanics and felsic sources extend beyond the conc entration range shown on the plot).
196 Based on this observation, and the coherent mixing trend illustrated in Figure 6-6B for Th/Sc, this ratio appears to be a robust indicator of provenance composition. As such, Th/Sc was plotted against Nd(t) using the model of McLennan et al. (1990) in order to clarify the bulk chemical composition of Cypresshead clays at the time of deposition (Figure 6-8). Based on this plot, Cypresshead provenance appears consistent with a dominantly felsic composition originating from an upper crust source, al though some of the high Th/Sc va lues are likely due to heavy mineral concentration rather than provenance bulk chemistry (Fig. 6-6B). This conclusion differs somewhat from the model suggested by Figure 6-7A and may be due to a slight depletion of both Th and Sc dictating the high degree of correlation to metavolcanics in Figure 6-7A. In comparing granites and metavolcanics associated with the Carolina terrane to Cypresshead Nd isotope results, initial 143Nd/144Nd ratios of both granites and Carolina terrane metavolcanics reported in the lite rature are considered to be robust (Kozuch, 1994; Samson et al., 1995a; 1995b; Fullagar et al., 1997). Additionall y, even large uncertainties in granite ages have little effect on the calculated Nd(t) because of the low Sm/Nd ratio of the granites and the 106 Ga half-life of 147Sm. The Nd(t) values of Alleghanian granite s range from -8.2 to +3.0, with more restricted ranges for individual terranes (Samson et al., 1995a; Fu llagar et al., 1997). Carolina terrane granites range from -6.7 to +1.9, Eastern slate belt granites from -2.7 to +2.4, Kiokee belt granites from 2.3 to +2.0, Raleigh belt granites from -4.3 to +3.0, and Inner Piedmont granites from -8.2 to -3.4. The bulk of the Nd(t) values for the A lleghanian granites range between -3 and +3, indicati ng that a significant proportion of their source material is not old, evolved crust, but rather likely from partial melting of depleted mantle or wholesale melting of terrane rocks (Samson et al., 1995a). The Nd isotopic evolution trend for Alleghanian granites is plotted in Figure 6-9 for comparison to both Cypresshead and comparison samples.
197 Figure 6-8. Plot of Nd(t) versus Th/Sc for Cypresshead Formation samples based on the model of McLennan et al. (1990; 1993). Figure 6-9. Plot of Nd(t) versus stratigraphic age for the Cypresshead Formation compared to the Nd isotopic evolution of potential sour ces (data for Grenville crust, Alleghanian granites, and the Carolina terrane ar e from Kozuch, 1994; Samson et al., 1995a; 1995b; Fullagar et al., 1997).
198 Carolina terrane rocks have Nd(t) values (most from -2 to +3) indistinguishable from those of most Alleghanian granites which intrude the Caro lina terrane (most from -2 to +2), suggesting them as a likely source for the granites (Koz uch, 1994; Samson et al., 1995b; Fullagar et al., 1997). Supporting this idea are the Nd isotopic com positions of Eastern slate belt volcanic rocks which, although lower than Carolina terrane values, are similar to granites intruding the terrane (Samson et al., 1995a). Some granites possess Nd(t) values suggesting a more evolved source, possibly Grenville basement. The chemical and isotopic data for granit es intruding the Kiokee belt, Eastern slate belt, and Carolina terrane are sim ilar, suggesting that the terranes are similar as well (Samson et al., 1995a). As with the availabl e Nd isotopic data for Alleghanian granites, the Nd isotopic evolution trend for Carolina terrane rocks is also plotted in Figure 6-9. Based on a comparison of Cypresshead, Hawthorn, Huber, and Buffalo Creek Nd(t) values to the Nd isotopic evolution trends of Alleghanian granites, Carolina terrane rocks, and Grenville crust, Cypresshead samples appear to origin ate from sources intermediate between those associated with Cretaceous soft kaolins and Te rtiary hard kaolins (Fig. 6-9). Samples from the Cretaceous Buffalo Creek Formation appear c onsistent with the conclusions of Dombrowski (1992; 1993), likely origin ating from Alleghanian granite sour ces, but apparently from granites having a significant portion of their source material originatin g from old, evolved Grenville crust. Tertiary hard kaolin samples from the Huber Formation are more consistent with a mixed sourcing from both Alleghanian and Carolina terrane lithologies (Fig. 6-9). Again, this conclusion is consistent with th e work of Dombrowski (1992; 1993). Cypresshead and Hawthorn clays exhibit sim ilar provenance characteristics, and are apparently a mixture of Carolina terrane a nd Alleghanian granite sources (Fig. 6-9). Additionally, both units appear to possess a significant affinity to an evolved crustal source, an
199 observation consistent with the distribution of TDM values similar in age to Grenville crust (Fig. 6-5). This conclusion is fu rther supported by the low Nd(t) values associated with Carolina terrane (min = -6.7) and Inner Piedmont (min = -8.2) granites, the most likely source for both Buffalo Creek clays and a felsic component of C ypresshead clays. The similarity in provenance between Cypresshead Formation and Hawthorn Gr oup clays is further evidence supporting the detrital sourcing of some Cypresshead clays via reworking of Hawthorn sediments as discussed in Chapter 5. Furthermore, the similarity in provenance is also a predictable outcome of the significant sediment mixing expe cted for deposits located distally from their source. Conclusions The results outlined in this chapter combined trace element and Nd isotopic analyses to constrain the provenance and broader geochemical characteristics of the Cypresshead Formation and associated units. Among th e results are the following: Trace element concentrations of Florida Cypres shead clays relative to average continental crust are relatively low, except for P, Th, and U, with Georgia Cypresshead clays slightly more enriched than Florida samples. Significant geochemical correlations exist for P2O5 (Th, Zr, and Sr), TiO2 (Th, Zr, V, and Cr), and Fe2O3 (Rb, V, Cr, and Co), the most significant of which are correlated to accessory phase (zircon and cranda llite-florencite) enrichment. Cypresshead samples exhibit moderate variation in Nd(t) values, ranging between Nd(3.3 Ma) = -5.0 and -9.7 and Nd(2.5 Ma) = -6.6 and -9.4 for Florida and Georgia samples respectively Nd model ages (TDM) for Florida and Georgia Cypressh ead samples range between 1.4-0.9 Ga and 1.5-0.8 Ga respectively, with both possessing an average TDM value of 1.1 Ga consistent with the age of Grenville crust. A significant proportion of Florid a Cypresshead clays possess Th/U ratios well below that of average upper crust, and a ppear to have undergone redox-dr iven U enrichment dictated by pore water organics in ba sal Cypresshead sediments. Of the trace elements reported in the Dombro wski (1992; 1993) studies, Th and Sc appear to be the most resistant to depletion, with the Th/Sc ratio a robust indicator of provenance composition.
200 Based on Nd isotopic results, the Buffalo Creek Formation appears to have originated from Alleghanian granite sources having a signi ficant portion of their source material originating from old, evolved Grenville crust, while the Huber Formation is consistent with a mixed sourcing from both Alleghanian and Carolina terrane lithologies. Cypresshead Formation (and Hawthorn Group) sa mples appear to originate from sources intermediate between those associated with Cr etaceous soft kaolins and Tertiary hard kaolins, possessing provenance characteristics consistent with a mixture of Carolina terrane and Alleghani an granite sources.
201 CHAPTER 7 SUMMARY AND CONCLUSIONS Cypresshead Formation Stratigraphy Focusing on the stratigraphy and sedimentology of the Cypresshead Formation, the results outlined in Chapter 4 highlight significant obser vations which clarify the nature, timing and significance of siliciclastic deposition impacting the Florida Platform during the Late Miocene through Pliocene. Among the results are th e following: Cypresshead sediments were deposited in a near shore marine environment, most likely in a strand plain setting, as two distinct progradational shor eface-shelf parasequences. Cypresshead facies define coarsening-upward sequences consis tent with a wave-dominated environment in north-central Florida and a mixed energy environment in southeastern Georgia. Deposition of the Cypresshead took place in re sponse to sea-level falls at 3.3 Ma and 2.5 Ma as a consequence of the interplay of sea-level, sediment supply and accommodation. Deposition in Florida at 3.4.8 Ma with reworking at 2.8.8 Ma. Deposition in Georgia at 3.15.3 Ma with reworking at 2.3.8 Ma. Timing of Cypresshead deposition at 3.4.3 Ma during the Late Plio cene (Piacenzian to early Gelsian) correlates w ith age estimates of the Citronelle Formation (3.4.7 Ma) as defined by Otvos (1988b) and timing of siliciclastic deposition associated with the Tamiami Formation (SS3/Pinecrest Sands). Viewed collectively with the Late Miocene SS 2 siliciclastics of Cunningham et al. (2003), Cypresshead and associated siliciclastics de fine a retrogradational parasequence which was deposited on the Florida Pl atform over a 6.8 Ma period. Cypresshead deposition correlates with a pale oclimate transition from continual El Nio conditions associated with the Pliocene warm period (~4.5.0 Ma) to conditions associated with the onset of significant Northern Hemisphere Glaciation (NHG) (~3.0.5 Ma). The driving mechanism behind the anomalous accumulation of siliciclastics associated with the Cypresshead and related Late Miocene siliciclastics (SS2) is the shift from periods of climate stability to periods of climate transition (instability) characterized by changes in temperature, precip itation and vegetation.
202 Cypresshead Formation Mineralogy Weathering processes are known have a profound impact on original sediments which host commercial kaolin deposits. Focu sing on the clay mineralogy and microtexture of Cypresshead sediments, the results outlined in Chapter 5 highlight evidence for the origin and weathering history of Cypresshead kaolinite s in order to define processe s impacting their mineralogy and industrial properties, and answers the primary verses secondary origin issue associated with these clays. Among the results are the following: Cypresshead kaolinite consists of at least three fractions; in situ kaolinite formed at the expense of feldspars and mica, detrital kaolin ite deposited as part of the original clay mineral suite, and near surf ace recrystallized kaolinite. In situ kaolinite crystallizes via the combined topotactic (transformation) and epitactic (neoformation) weathering of muscovite mica consistent with the model of Jeong (1988b) and confirmed by vertical trends in kaolin ite disorder and cohe rent scattering domain (CSD) values. The muscovite weathering model of Jeong (1988 b) explains the unusual low disorder/small particle-size characteristics of Florida Cypre sshead kaolinite, the distribution of fine mica in coarseto medium-sands and gravels, and the relatively hi gh (> 20%) total clay content observed in some basal Cypresshead sediments. Feldspar dissolution as a source of Al a nd Si necessary for the crystallization of in situ kaolinite and post-depositional increase in Cypresshead clay content is confirmed by the presence of residual feldspars in basal Cypre sshead sediments and the inherited positive Eu anomalies likely resulting from an original plagioclase component to the feldspar suite. Pore water organics are like ly important in concentrati ng dissolved Al in order to crystallize in situ kaolinite and in co ntrolling the pore wate r redox environment of Cypresshead sediments as indicated by the vertical trends in Eu anomalies and HREE depletion. Near surface recrystallized ka olinite formed under oxic, vado se or mixed vadose/saturated conditions are disordered and possess microtextural characteristics consistent with a pedogenic origin. Occurrences of near surface formed halloysite and gibbs ite are consistent with Cypresshead sampling locations characterized by extensive evidence of vadose and mixed vadose/saturated leaching and recrystallization under oxic conditions.
203 Trace crandallite-florencite minerals likely originated from the decomposition of postdepositional pore water organics couple d with detrital mi neral dissolution. Cypresshead Formation Provenance The results outlined in Chapter 6 combined trace element and Nd isotopic analyses to constrain the provenance and broader geochemical characteristics of the Cypresshead Formation and associated units. Among th e results are the following: Trace element concentrations of Florida Cypres shead clays relative to average continental crust are relatively low, except for P, Th, and U, with Georgia Cypresshead clays slightly more enriched than Florida samples. Significant geochemical correlations exist for P2O5 (Th, Zr, and Sr), TiO2 (Th, Zr, V, and Cr), and Fe2O3 (Rb, V, Cr, and Co), the most significant of which are correlated to accessory phase (zircon and cranda llite-florencite) enrichment. Cypresshead samples exhibit moderate variation in Nd(t) values, ranging between Nd(3.3 Ma) = -5.0 and -9.7 and Nd(2.5 Ma) = -6.6 and -9.4 for Florida and Georgia samples respectively Nd model ages (TDM) for Florida and Georgia Cypressh ead samples range between 1.4-0.9 Ga and 1.5-0.8 Ga respectively, with both possessing an average TDM value of 1.1 Ga consistent with the age of Grenville crust. A significant proportion of Florid a Cypresshead clays possess Th/U ratios well below that of average upper crust, and a ppear to have undergone redox-dr iven U enrichment dictated by pore water organics in ba sal Cypresshead sediments. Of the trace elements reported in the Dombro wski (1992; 1993) studies, Th and Sc appear to be the most resistant to depletion, with the Th/Sc ratio a robust indicator of provenance composition. Based on Nd isotopic results, the Buffalo Creek Formation appears to have originated from Alleghanian granite sources having a signi ficant portion of their source material originating from old, evolved Grenville crust, while the Huber Formation is consistent with a mixed sourcing from both Alleghanian and Carolina terrane lithologies. Cypresshead Formation (and Hawthorn Group) sa mples appear to originate from sources intermediate between those associated with Cr etaceous soft kaolins and Tertiary hard kaolins, possessing provenance characteristics consistent with a mixture of Carolina terrane and Alleghani an granite sources.
204 APPENDIX A MINE SITE MAPS
205 Figure A-1. Site map for the Edga r Minerals EPK Mine (DOQQ Source: http://data.labins.org/200 3/MappingD ata/DOQQ/doqq_04_utm.cfm Last accessed October, 2007). Figure A-2. Site map for the VMC Goldhead Sand Mine (DOQQ Source: http://data.labins.org/200 3/MappingD ata/DOQQ/doqq_04_utm.cfm Last accessed October, 2007).
206 Figure A-3. Site map for the VMC Grandin Sand Mine (DOQQ Source: http://data.labins.org/200 3/MappingD ata/DOQQ/doqq_04_utm.cfm Last accessed October, 2007). Figure A-4. Site map for the CEME X Davenport Sand Mine (DOQQ Source: http://data.labins.org/200 3/MappingD ata/DOQQ/doqq_04_utm.cfm Last accessed October, 2007).
207 Figure A-5. Site map for the CEME X Joshua Sand Mine (DOQQ Source: http://data.labins.org/200 3/MappingD ata/DOQQ/doqq_04_utm.cfm Last accessed October, 2007).
208 APPENDIX B ANALYTICAL PROCEDURES FOR ICP-AES, ICP-MS, AND MC-ICPMS
209 METHOD A: ICP-AES MULTI-ACID AN ALYSIS (SGS/XRAL: ICP40) SUMMARY: This method involves the determination of 40 major, mi nor, and trace elements in geologic materials by inductively coupled plasma-atomic emissi on spectrometry (ICP-AES). The sample is decomposed using a mixture of hydrochloric, nitric, perchloric and hydrofluoric acids at low temperature (Crock et al., 1983). The digested sample is aspirated into the ICP-AES discharge where the elemental emission signal is measured simultaneously for the 40 elements. Calibration is performed by standa rdizing with digested rock re ference materials (BCR-2) and a series of multi-element solution standards (Lichte et al., 1987). APPARATUS: 1) Sample preparation 35 ml Teflon tube with screw cap. Temperature controlled aluminum block machined to accept 1 inch diameter Teflon vessels with screw caps. 2) Sample analysis Perkin-Elmer Optima 300 ICP-AES. Perkin-Elmer Cross-flow nebulizer. Perkin-Elmer peristaltic pump. CETAC autosampler. REAGEANTS: Hydrochloric acid (HCl) 36-38%. Nitric acid (HNO3) 69-71%. Perchloric acid (HClO4) 62-70%. Hydrofluoric acid (HF) 48-51%. Lu internal standard 500 ppm. PROCEDURE: 1) Weigh 0.200 g ( 2 mg) sample into Teflon dish. 2) Add 100 ml of 500 g/ml Lu internal standard to each dish with repeating pipet. 3) Rinse side walls of Teflon dish with a minimum amount of deionized (DI) water (~ 2 ml). 4) In the fume hood, slowly add 3 ml HCl and allow any reaction to subside. 5) Add 2 ml HNO3, 1 ml HClO4, and 2 ml HF. 6) Place sample solution vessel on hot plate wi th aluminum heat block at a controlled temperature of 110 C in a perchloric acid fume hood. 7) Evaporate sample solution to hard dryness on hot plate (usually overnight). 8) Remove from hot plate, c ool to touch and add 1 ml HClO4 and 2-3 ml DI water. 9) Return to hot plate and evaporate to hard dryness. The temperature of the hot plate is increased to 160 C. This step usually takes a few hours. 10) Remove dried sample from hot plate and cool. 11) Add 1 ml aqua regia with repea ting pipet and let react for 15 minutes.
210 12) Add 9 ml 1% HNO3 and heat for a few minutes. 13) Cool, transfer solution into labeled dispos able polypropylene test tube and dilute to 10 ml. 14) Analyze sample solution by ICP-AES. ANALYTICAL PERFORMANCE: Data is deemed acceptable if recovery is 15% at five times the Lower Limit of Determination (LOD), and the calculated Relative Standard Deviation (RSD) of duplicate samples is 15%. CRM SO-3 was analyzed as a standard to evaluate accuracy. REPORTING LIMITS: Element Concentration Range Element Concentration Range Aluminum, Al 0.005 50% Gallium, Ga 4 50,000 ppm Calcium, Ca 0.005 50% Holmium, Ho 4 5,000 ppm Iron, Fe 0.02 25% Lanthanum, La 2 50,000 ppm Potassium, K 0.01 50% Lithium, Li 2 50,000 ppm Magnesium, Mg 0.005 5% Manganese, Mn 4 50,000 ppm Sodium, Na 0.005 50% Molybdenum, Mo 2 50,000 ppm Phosphorous, P 0.005 50% Niobium, Nb 4 50,000 ppm Titanium, Ti 0.005 25% Neodymium, Nd 9 50,000 ppm Silver, Ag 2 10,000 ppm Nickel, Ni 3 50,000 ppm Arsenic, As 10 50,000 ppm Lead, Pb 4 50,000 ppm Gold, Au 8 50,000 ppm Scandium, Sc 2 50,000 ppm Barium, Ba 1 35,000 ppm Tin, Sn 5 50,000 ppm Beryllium, Be 1 5,000 ppm Strontium, Sr 2 15,000 ppm Bismuth, Bi 10 50,000 ppm Tantalum, Ta 40 50,000 ppm Cadmium, Cd 2 25,000 ppm Thorium, Th 6 50,000 ppm Cerium, Ce 5 50,000 ppm Uranium, U 100 100,000 ppm Cobalt, Co 2 25,000 ppm Vanadium, V 2 30,000 ppm Chromium, Cr 2 50,000 ppm Yttrium, Y 2 25,000 ppm Copper, Cu 2 15,000 ppm Ytterbium, Yb 1 5,000 ppm Europium, Eu 2 5,000 ppm Zinc, Zn 2 15,000 ppm REFERENCES: Crock, J.G., Lichte, F.E., and Br iggs, P.H., 1983, Determination of elements in National Bureau of Standards geological reference materi als SRM278 obsidian and SRM688 basalt by inductively coupled argon plasma-atomi c emission spectrometry: Geostandards Newsletter, v. 7, p. 335. Lichte, F.E., Golightly, D.W., and Lamothe, P.J., 1987, Inductively coupled plasma-atomic emission spectrometry, in Baedecker, P.A., ed., Methods for Geochemical Analysis, USGS Bulletin, Report: B1770, USGS, Reston, VA, p. B1B10.
211 METHOD B: ICP-MS LiBO2 FUSION (SGS/XRAL: MS95) SUMMARY: This method involves the determination of 37 trace a nd rare earth elements in geologic materials by inductively coupled plasma-mass spectrometry (ICP-MS). The sample is fused with LiBO2 at 950C for twenty minutes, and then the cake is dissolved in dilute nitric acid prior to aspiration into the ICP-MS. APPARATUS: 1) Sample preparation Graphite crucible. Muffle furnace. Teflon vials. 2) Sample analysis Perkin-Elmer Elan 6100 ICP-MS, VG ThermoElemental PlasmaQuad 2 ICP-MS, or VG ThermoElemental PlasmaQuad 3 ICP-MS. REAGEANTS: Lithium borate (LiBO2) Nitric acid (HNO3) 10%. Tartaric acid 2%. Lu internal standard 1000 ppm. Re and Rh internal standards 50 ppb and 10 ppb, respectively. Calibration standards diluted with 2% HNO3 and 4 ml of LiBO2 solution (2.8 g flux in 10% HNO3) into a 200 ml volumetric flask o #1 blank (all elements at 0 ppb) o #2 10.0 ppb all elements except Ag (1 ppb) o #3 25.0 ppb all elements except Ag (2.5 ppb) o #4 50.0 ppb all elements except Ag (5 ppb) PROCEDURE: 1) Weigh 0.1 g sample into a graphite crucible. 2) Add 0.70 g of LiBO2 and mix. 3) Put crucibles in muffle furnace (preheat ed at 950C) for twenty (20) minutes. 4) Take out the crucibles and pour in to vial containi ng 50 ml of 10% HNO3, 2% tartaric acid. 5) Shake for ten (10) minutes. 6) Add 0.25 ml of 1000 ppm Lu internal standard and mix. 7) Transfer 15 ml into a plastic test tube. 8) Transfer the rack to the ICP room. 9) Pipette 0.2 ml sample digest + 0.8 ml DDW into centrifuge tube. 10) Add 9 ml internal standard solution (R e + Rh) to correct for matrix effects. 11) Cover sample with parafilm and shake. 12) Load the sample(s) into a rack and set into the autosampler station.
212 13) Analyze sample solution by ICP-MS. ANALYTICAL PERFORMANCE: CRM SO-3 was wei ghed, fused, and analyzed as a standard to evaluate accuracy. Calibration standard #2 is run after every 16 samples as a QC check. REPORTING LIMITS: Element Detection Limits Element Detection Limits Barium, Ba 0.5 ppm 1% Pras eodymium, Pr 0.05 ppm 0.1% Cerium, Ce 0.1 ppm 1% R ubidium, Rb 0.2 ppm 1% Cesium, Cs 0.1 ppm 1% Sa marium, Sm 0.1 ppm 0.1% Cobalt, Co 0.5 ppm 1% Silver, Ag 1 ppm 0.1% Copper, Cu 5 ppm 1% Strontium, Sr 0.1 ppm 1% Dysprosium, Dy 0.05 ppm 0.1% Tantalum, Ta 0.5 ppm 1% Erbium, Er 0.05 ppm 0.1% Te rbium, Tb 0.05 ppm 0.1% Europium, Eu 0.05 ppm 0.1% Thallium, Tl 0.5 ppm 0.1% Gadolinium, Gd 0.05 ppm 0.1% Thorium, Th 0.1 ppm 0.1% Gallium, Ga 1 ppm 0.1% Thulium, Tm 0.05 ppm 0.1% Hafnium, Hf 1 ppm 1% Tin, Sn 1 ppm 1% Holmium, Ho 0.05 ppm 0.1% Tungsten, W 1 ppm 1% Lanthanum, La 0.1 ppm 1% Uranium, U 0.05 ppm 0.1% Lead, Pb 5 ppm 1% Va nadium, V 5 ppm 1% Lutetium, Lu 0.05 ppm 0.1% Y tterbium, Yb 0.1 ppm 0.1% Molybdenum, Mo 2 ppm 1% Yttrium, Y 0.5 ppm 0.1% Neodymium, Nd 0.1 ppm 1% Zinc, Zn 5 ppm 1% Nickel, Ni 5 ppm 1% Zirconium, Zr 0.5 ppm 1% Niobium, Nb 1 ppm 1%
213 METHOD C: Nd ISOTOPIC ANALYSIS: SAMPLE PREPARATION SUMMARY: This method involves procedures followe d in the dissolution of silicate whole-rock powders and Nd collection via column chemistry. PROCEDURE: 1) Clean tall Savillex Teflon hex-cap beakers: Put a little 6N trace metal (TM) grade HCl in the bottom of the beakers, cap them, and leave them on a warm hot plate for at least one hour. Rinse 3 times with 4 H2O. Dry by opening them up and leaving them on the hot plate for up to an hour. The caps must be left concave side up in this situation. 2) Add sample: Put the beaker on the balance in the outer clean lab room. Allow to settle, then push the tare button. Remove the beaker from the balance and put it on a clean, lint free wiper on the balance table. (No rock powders are allowed in the laminar flow hoods.) Open it up, putting the cap in a safe clean place. Using a clean spatula, add a small amount of rock powder to the beaker. Cap the beaker, and put it in the balance to check the weight. Keep adding rock pow der and checking the weight until you get approximately 50 mg (0.050 g). Record the weight on the weigh sheet. 3) Begin dissolving the sample: Take the capped beakers over to the exhaust hood. Open them up and add 1-2 drops of Optima HNO3 from the dropper bottle. Then add 3 ml of 1x TM HF (once-distilled trace metal grade HF). 4) Heat the sample: Put the cap on the b eaker and tighten it with the green plastic wrenches. Do not overtighten or you will stri p the threads. Put th e beaker in the small oven in the anteroom of the clean lab. Turn the temperature to 100 C and leave it there for 2-3 days. 5) Convert to chloride salts: Remove the samp le from the oven and let it cool. Put it in the exhaust hood. Taking all safety precautions, remove the cap with the green wrenches. Dry the sample by putting the beaker on the hotplate at 250 F (setting 3). The cap should be put convex side down on a piece of parafilm. It will take a day or so to dry. Then, add 2 ml of 6N TM grade HCl. Replace th e cap and tighten it down. Put the sample back into the oven overnight. Then remove it, let it cool, and dry it on the hotplate again. 6) Cation columns separation of bulk rare earth elements (REEs): This step separates the REEs from the rest of the rock using chromatographic methods on Dowex 50 X12 cation exchange resin (see below). Before the sample is loaded on the columns, it must be centrifuged to remove any solid material. After dissolving the sample in 500 l of 3.5N HCl, pour it into a clean centrif uge tube. Centrifuge the samp le for 2 minutes, then pipet out just the clean solution off the top and load it. Once the REE are separated, they should be dried down on the hotplate at 200-250 F. 1 Equilibrate 5 ml 3.5N HCl 2 Load sample dissolved in 250 l 3.5N HCl
214 3 Wash 500 l 3.5N HCl 4 Wash 500 l 3.5N HCl 5 Wash 500 l 3.5N HCl 6 Wash 23 ml 3.5N HCl 7 Wash 4 ml 6N HCl 8 Collect 10 ml 6N HCl (REE) 9 Clean by washing with approximately 50 ml 6N HCl 10 Re-equilibrate with 10 ml 4 H2O 11 Store columns in 4 H2O 7) REE columns separation of Nd: This step separates Nd from the rest of the REE using HCl elution on quartz columns packed with Teflon beads coated with bisethylhexyl phosphoric acid (after Richard et al., 1976) (see below). Normally, the sample will not have to be centrifuged before loading. Once th e Nd is separated, it should be dried on the hotplate. 1 Equilibrate 5 ml 0.18N HCl 2 Equilibrate 5 ml 0.18N HCl 3 Load sample dissolved in 200 l 0.18N HCl 4 Wash 200 l 0.18N HCl 5 Wash 200 l 0.18N HCl 6 Wash 200 l 0.18N HCl 7 Wash 8 ml 0.18N HCl 8 Collect 5 ml 0.18N HCl (Nd) 9 Clean by washing with approximately 50 ml 6N HCl 10 Re-equilibrate with 10 ml ~0.2N HCl 11 Store columns in ~0.2N HCl REFERENCES: Richard, P., Shimazu, N., and Allegre, C.J., 1976, 143Nd/144Nd, a natural tracer: An application to oceanic basalt: Earth and Planet ary Science Letters, v. 31, p. 269.
215 APPENDIX C STRATIGRAPHIC SECTIONS
216 Figure C-1. Stratigraphi c section for Grandin Sand Mine section FRG-1.
217 Figure C-2. Stratigraphi c section for Grandin Sand Mine section FRG-2.
218 Figure C-3. Stratigraphic section for Goldhead Sand Mine section FRL-1.
219 Figure C-4. Stratigraphi c section for Joshua Sa nd Mine core SSJ-1.
220 Figure C-5. Stratigraphic section for Davenport Sand Mine core SSD-1.
221 Figure C-6. Stratigraphic s ection for Jesup section J-1. Figure C-7. Stratigraphic s ection for Birds section B-1.
222 Figure C-8. Stratigraphic secti on for Linden Bluff section L-1.
223 APPENDIX D GRAIN-SIZE DATA (HYDROMETER AND SIEVE)
224 Table D-1. Grain-size di stributions and moment statistics for the Grandin and Goldhead sand mines in north-central Florida. Core/SectionInterval/Sample Mean ( )Std. Dev.SkewnessKurtosis FRG-1 185.80.214.01.01.912.298.3 287.72.010.41.184.108.40.206 3220.127.116.11.01.24.431.2 498.00.91.01.51.24.432.0 596.02.31.71.18.104.22.168 622.214.171.124.126.96.36.199 7188.8.131.52.184.108.40.206 8220.127.116.11.92.210.573.2 918.104.22.168.22.214.171.124 1047.6126.96.36.199.95.246.4 1149.741.09.33.31.83.030.2 12188.8.131.52.184.108.40.206 13220.127.116.11.02.16.045.8 1474.718.104.22.168.53.117.9 1596.01.62.51.22.214.171.124 FRG-2 126.96.36.199.188.8.131.52 2184.108.40.206.11.26.450.8 3220.127.116.11.91.86.347.1 495.32.72.01.21.33.317.0 582.17.510.42.62.15.644.8 687.74.08.32.22.15.948.1 718.104.22.168.81.72.318.7 889.45.05.62.22.214.171.124 9126.96.36.199.71.43.017.5 1087.73.88.62.81.43.326.1 1188.8.131.52.61.85.944.6 1286.13.610.32.91.63.528.9 1369.420.510.22.61.63.527.9 FGL-1 178.81.6184.108.40.206.168.7 284.74.710.62.31.97.249.1 3220.127.116.11.21.85.841.7 4A82.94.718.104.22.168.845.6 4B22.214.171.124.51.79.065.9 597.01.31.71.21.37.349.6 6A126.96.36.199.41.58.764.4 6B90.40.09.60.81.610.482.7 786.7-0.6188.8.131.52.848.7 8184.108.40.206.01.912.093.4 973.715.810.53.71.64.522.5 Sample ID % Sand% Silt% Clay Moment Statistics
225 Table D-2. Grain-size di stributions and moment statistics for the EPK kaolin mine in northcentral Florida. Core/SectionInterval/Sample Mean ( )Std. Dev.SkewnessKurtosis EPK36-J-1225-2781.35.6220.127.116.11.645.1 27-3086.85.77.51.61.55.740.0 35-4081.810.08.21.71.96.845.6 40-4484.01.318.104.22.168.437.4 44-4683.93.512.62.32.08.457.8 46-4822.214.171.124.91.99.769.8 48-5080.85.4126.96.36.199.745.2 50-53188.8.131.52.184.108.40.206 53-56220.127.116.11.72.18.256.5 56-5976.67.818.104.22.168.341.0 59-6269.010.722.214.171.124.924.7 EPK31-P-4027-3582.73.5126.96.36.199.567.4 35-4575.010.3188.8.131.52.340.8 45-5077.86.216.03.11.86.340.7 50-6280.31.518.23.01.76.744.3 62-6580.03.216.82.51.22.425.1 EPK30-V-616-2286.78.35.02.51.44.627.1 22-24184.108.40.206.31.02.515.3 24-2785.611.43.02.61.33.318.0 30-35220.127.116.11.18.104.22.168 35-3987.32.710.12.01.46.843.4 39-4322.214.171.124.01.44.126.1 43-48126.96.36.199.61.02.010.5 48-5387.63.49.01.61.98.960.6 53-58188.8.131.52.31.84.939.1 58-6377.012.510.53.31.94.231.6 63-6878.47.913.83.12.04.939.6 68-7379.08.013.03.11.84.434.3 73-7883.05.211.92.41.93.836.2 Sample ID % Sand% Silt% Clay Moment Statistics Table D-3. Grain-size di stributions and moment statistics for the Davenport and Joshua sand mines in central Florida. Core/SectionInterval/Sample Mean ( )Std. Dev.SkewnessKurtosis SSD-1 1 184.108.40.206.220.127.116.11 2 18.104.22.168.31.45.032.6 3 92.74.43.01.31.53.720.1 4 89.77.33.02.61.53.421.1 5 93.32.04.71.22.214.171.124 6 126.96.36.199.71.73.931.7 7 188.8.131.52.31.44.024.9 8 184.108.40.206.11.01.510.3 9 220.127.116.11.11.12.815.2 10 18.104.22.168.11.13.014.9 SSJ-1 1 80.62.516.91.01.13.418.3 2 81.31.322.214.171.124.825.1 3 86.72.610.61.31.67.850.8 4 85.43.611.01.91.55.735.8 5 97.10.82.01.126.96.36.199 6 188.8.131.52.184.108.40.206 7 86.53.510.02.11.53.828.1 8 220.127.116.11.31.53.528.6 9 87.54.08.52.31.52.923.1 10 87.71.710.62.41.44.229.9 11 85.21.918.104.22.168.027.6 Moment Statistics Sample ID % Sand% Silt% Clay
226 Table D-4. Grain-size di stributions and moment statistics for Cypresshead Formation sampling locations in southeastern Georgia. Core/SectionInterval/Sample Mean ( )Std. Dev.SkewnessKurtosis J-1 1 69.322.214.171.124.84.932.8 2 68.93.3126.96.36.199.033.1 3 73.47.619.11.72.07.247.5 4 14.922.163.04.61.94.223.9 5 83.34.4188.8.131.52.511.8 6 184.108.40.206.71.34.429.2 L-1 1 220.127.116.11.01.33.020.0 2 72.62.918.104.22.168.832.3 3 80.53.822.214.171.124.730.9 4 126.96.36.199.81.54.835.5 5 188.8.131.52.91.33.924.9 6 18.618.463.04.62.90.639.5 7 184.108.40.206.220.127.116.11 B-1 1 18.104.22.168.41.13.918.8 2 56.110.622.214.171.124.427.7 3 55.516.9126.96.36.199.023.6 4 56.010.6188.8.131.52.548.9 5 67.98.124.03.72.06.238.7 % Silt% Clay Sample ID % Sand Moment Statistics
227 0 10 20 30 40 50 60 0.001 0.01 0.1 1Particle Size (mm)Mass Frequency (%) (25-27) (27-30) (35-40) (40-44) (44-46) (46-48) (48-50) (50-53) (53-56) (56-59) (59-62) Figure D-1. Grain-size distribution curves for EPK Mine core EPK36-J-12. 0 10 20 30 40 50 0.001 0.01 0.1 1 Particle Size (mm)Mass Frequency (%) (27-35) (35-45) (45-50) (50-62) (62-65) Figure D-2. Grain-size distribution curves for EPK Mine core EPK31-P-40.
228 0 5 10 15 20 25 30 35 40 45 50 0.001 0.01 0.1 1Particle Size (mm)Mass Frequency (%) (16-22) (22-24) (24-27) (30-35) (35-39) (39-43) (43-48) (48-53) (53-58) (58-63) (63-68) (68-73) (73-78) Figure D-3. Grain-size distribution curves for EPK Mine core EPK30-V-6. 0 10 20 30 40 50 0.001 0.01 0.1 1 Particle Size (mm)Mass Frequency (%) FRG-1-1 FRG-1-2 FRG-1-3 FRG-1-4 FRG-1-5 FRG-1-6 FRG-1-7 FRG-1-8 FRG-1-9 FRG-1-10 FRG-1-11 FRG-1-12 FRG-1-13 FRG-1-14 FRG-1-15 Figure D-4. Grain-size distribution curves for Grandin Sand Mine section FGR-1.
229 0 10 20 30 40 50 60 0.001 0.01 0.1 1 Particle Size (mm)Mass Frequency (%) FRG-2-1 FRG-2-2 FRG-2-3 FRG-2-4 FRG-2-5 FRG-2-6 FRG-2-7 FRG-2-8 FRG-2-9 FRG-2-10 FRG-2-11 FRG-2-12 FRG-2-13 Figure D-5. Grain-size distribution curves for Grandin Sand Mine section FRG-2. 0 10 20 30 40 50 60 70 80 0.001 0.01 0.1 1 Particle Size (mm)Mass Frequency (%) FRL-1-1 FRL-1-2 FRL-1-3 FRL-1-4 FRL-1-5 FRL-1-6 FRL-1-7 FRL-1-8 FRL-1-9 Figure D-6. Grain-size distribution curves for Goldh ead Sand Mine section FRL-1.
230 0 10 20 30 40 50 0.001 0.01 0.1 1 Particle Size (mm)Mass Frequency (%) SSJ-1-1 SSJ-1-2 SSJ-1-3 SSJ-1-4 SSJ-1-5 SSJ-1-6 SSJ-1-7 SSJ-1-8 SSJ-1-9 SSJ-1-10 SSJ-1-11 Figure D-7. Grain-size distribution curves for Joshua Sand Mine core SSJ-1. 0 10 20 30 40 50 60 70 0.001 0.01 0.1 1 Particle Size (mm)Mass Frequency (%) SSD-1-1 SSD-1-2 SSD-1-3 SSD-1-4 SSD-1-5 SSD-1-6 SSD-1-7 SSD-1-8 SSD-1-9 SSD-1-10 Figure D-8. Grain-size distribution curves for Davenport Sand Mine core SSD-1.
231 0 10 20 30 40 0.001 0.01 0.1 1 Particle Size (mm)Mass Frequency (%) J-1-1 J-1-2 J-1-3 J-1-4 J-1-5 J-1-6 Figure D-9. Grain-size distribution curves for Jesup section J-1. 0 10 20 30 40 50 0.001 0.01 0.1 1 Particle Size (mm)Mass Frequency (%) L-1-1 L-1-2 L-1-3 L-1-4 L-1-5 L-1-6 L-1-7 Figure D-10. Grain-size distribution curves for Linden Bluff section L-1.
232 0 10 20 30 40 50 60 0.001 0.01 0.1 1 Particle Size (mm)Mass Frequency (%) B-1-1 B-1-2 B-1-3 B-1-4 B-1-5 Figure D-11. Grain-size distribut ion curves for Birds section B-1.
233 APPENDIX E X-RAY DIFFRACTION DATA (ORIENTED) Note: If not marked, diffract ion peaks are assigned to kaol inite. The legend for other minerals is as follows: Q quartz G gibbsite C crandallite-florencite H halloysite HIV hydroxyl-interlayered vermiculite I illite S smectite B boehmite D diaspore A anatase
249 APPENDIX F X-RAY DIFFRACTION DATA (RANDOM) Note: If not marked, diffract ion peaks are assigned to kaol inite. The legend for other minerals is as follows: Q quartz G gibbsite C crandallite-florencite H halloysite HIV hydroxyl-interlayered vermiculite I illite S smectite R rutile B boehmite D diaspore A anatase Gr goethite
266 APPENDIX G MINUS-200 MESH PARTICLE-SIZE DATA (SEDIGRAPH)
267 0 20 40 60 80 100 0.1 1 10 100Particle Diameter (um)Cumulative Mass Finer (%) 36-J-12 (25-27) 36-J-12 (35-40) 36-J-12 (46-48) 36-J-12 (50-53) 36-J-12 (56-59) 0.0 2.0 4.0 6.0 8.0 0.1 1 10 100Particle Diameter (um)Mass Frequency (%) 36-J-12 (25-27) 36-J-12 (35-40) 36-J-12 (46-48) 36-J-12 (50-53) 36-J-12 (56-59) Figure G-1. SediGraph particle-s ize plots for the minus-200 mesh (< 75 m) fraction of select EPK Mine samples (EPK36-J-12).
268 0 20 40 60 80 100 0.1 1 10 100Particle Diameter (um)Cumulative Mass Finer (%) 31-P-40 (35-45) 31-P-40 (50-62) 31-P-40 (62-65) 0.0 0.5 1.0 1.5 2.0 2.5 0.1 1 10 100Particle Diameter (um)Mass Frequency (%) 31-P-40 (35-45) 31-P-40 (50-62) 31-P-40 (62-65) Figure G-2. SediGraph particle-s ize plots for the minus-200 mesh (< 75 m) fraction of select EPK Mine samples (EPK31-P-40).
269 0 20 40 60 80 100 0.1 1 10 100Particle Diameter (um)Cumulative Mass Finer (%) 30-V-6 (22-24) 30-V-6 (30-35) 30-V-6 (48-53) 30-V-6 (58-63) 30-V-6 (68-73) 0.0 0.5 1.0 1.5 2.0 2.5 0.1 1 10 100Particle Diameter (um)Mass Frequency (%) 30-V-6 (22-24) 30-V-6 (30-35) 30-V-6 (48-53) 30-V-6 (58-63) 30-V-6 (68-73) Figure G-3. SediGraph particle-s ize plots for the minus-200 mesh (< 75 m) fraction of select EPK Mine samples (EPK30-V-6).
270 0 20 40 60 80 100 0.1 1 10 100Particle Diameter (um)Cumulative Mass Finer (%) FRG-1-3 FRG-1-6 FRG-1-7 FRG-1-10 FRG-1-11 FRG-1-13 FRG-1-16 0.0 0.5 1.0 1.5 2.0 2.5 3.0 0.1 1 10 100Particle Diameter (um)Mass Frequency (%) FRG-1-3 FRG-1-6 FRG-1-7 FRG-1-10 FRG-1-11 FRG-1-13 FRG-1-16 Figure G-4. SediGraph particle-s ize plots for the minus-200 mesh (< 75 m) fraction of select Grandin Sand Mine samples (FRG-1).
271 0 20 40 60 80 100 0.1 1 10 100Particle Diameter (um)Cumulative Mass Finer (%) FRG-2-3 FRG-2-5 FRG-2-7 FRG-2-10 FRG-2-12 0.0 0.5 1.0 1.5 2.0 2.5 0.1 1 10 100Particle Diameter (um)Mass Frequency (%) FRG-2-3 FRG-2-5 FRG-2-7 FRG-2-10 FRG-2-12 Figure G-5. SediGraph particle-s ize plots for the minus-200 mesh (< 75 m) fraction of select Grandin Sand Mine samples (FRG-2).
272 0 20 40 60 80 100 0.1 1 10 100Particle Diameter (um)Cumulative Mass Finer (%) FRL-1-2 FRL-1-4 FRL-1-6 FRL-1-7 FRL-1-9 0.0 0.5 1.0 1.5 2.0 2.5 0.1 1 10 100Particle Diameter (um)Mass Frequency (%) FRL-1-2 FRL-1-4 FRL-1-6 FRL-1-7 FRL-1-9 Figure G-6. SediGraph particle-s ize plots for the minus-200 mesh (< 75 m) fraction of select Goldhead Sand Mine samples (FRL-1).
273 0 20 40 60 80 100 0.1 1 10 100Particle Diameter (um)Cumulative Mass Finer (%) SSJ-1-2 SSJ-1-4 SSJ-1-6 SSJ-1-8 SSJ-1-10 0.0 2.0 4.0 6.0 8.0 10.0 0.1 1 10 100Particle Diameter (um)Mass Frequency (%) SSJ-1-2 SSJ-1-4 SSJ-1-6 SSJ-1-8 SSJ-1-10 Figure G-7. SediGraph particle-s ize plots for the minus-200 mesh (< 75 m) fraction of select Joshua Sand Mine samples (SSJ-1).
274 0 20 40 60 80 100 0.1 1 10 100Particle Diameter (um)Cumulative Mass Finer (%) SSD-1-1 SSD-1-3 SSD-1-6 SSD-1-7 SSD-1-10 0.0 2.0 4.0 6.0 8.0 10.0 0.1 1 10 100Particle Diameter (um)Mass Frequency (%) SSD-1-1 SSD-1-3 SSD-1-6 SSD-1-7 SSD-1-10 Figure G-8. SediGraph particle-s ize plots for the minus-200 mesh (< 75 m) fraction of select Davenport Sand Mine samples (SSD-1).
275 0 20 40 60 80 100 0.1 1 10 100Particle Diameter (um)Cumulative Mass Finer (%) J-1-2 J-1-4 J-1-6 0.0 2.0 4.0 6.0 8.0 10.0 0.1 1 10 100Particle Diameter (um)Mass Frequency (%) J-1-2 J-1-4 J-1-6 Figure G-9. SediGraph particle-s ize plots for the minus-200 mesh (< 75 m) fraction of select Jesup type locality samples (J-1).
276 0 20 40 60 80 100 0.1 1 10 100Particle Diameter (um)Cumulative Mass Finer (%) LB-1-3 LB-1-5 LB-1-6 0.0 2.0 4.0 6.0 8.0 10.0 0.1 1 10 100Particle Diameter (um)Mass Frequency (%) LB-1-3 LB-1-5 LB-1-6 Figure G-10. SediGraph particle-size plots for the minus-200 mesh (< 75 m) fraction of select Linden Bluff reference locality samples (L-1).
277 0 20 40 60 80 100 0.1 1 10 100Particle Diameter (um)Cumulative Mass Finer (%) B-1-2 B-1-3 B-1-5 0.0 0.5 1.0 1.5 2.0 2.5 0.1 1 10 100Particle Diameter (um)Mass Frequency (%) B-1-2 B-1-3 B-1-5 Figure G-11. SediGraph particle-size plots for the minus-200 mesh (< 75 m) fraction of select Birds reference locality samples (B-1).
278 APPENDIX H MAJOR AND TRACE ELEMENT DATA
279 Table H-1. Raw major element concentrations for samples used in this study. Sample ID IntervalNa (%)Mg (%)Al (%)P (%)K (%)Ca (%)Ti (%)Fe (%) Cypresshead Formation FL EPK36-J-1225-270.0080.04619.420.060.030.0340.150.19 EPK36-J-1235-400.010.0317.890.0650.040.0350.140.25 EPK36-J-1246-480.0110.04419.90.0310.040.0250.1690.34 EPK36-J-1250-530.0150.0317.040.020.070.020.190.36 EPK36-J-1259-620.0150.0618.090.030.080.0550.330.54 EPK31-P-4035-450.0130.0519.910.0270.050.0320.2080.46 EPK31-P-4050-620.0150.09517.170.030.070.1050.1850.59 EPK31-P-4062-650.0130.23718.280.0750.190.2060.1710.87 EPK30-V-622-240.0151.4655.0330.0970.230.4580.3551.78 EPK30-V-630-350.0250.7713.290.5184.108.40.206.07 EPK30-V-648-530.0250.13517.290.2750.130.3550.250.62 EPK30-V-658-630.0150.03619.710.0220.070.0160.1510.29 EPK30-V-668-730.0150.0717.180.0250.120.050.260.47 FRG-130.0090.0518.860.1220.040.030.1872.09 FRG-160.0050.01517.850.070.020.0150.040.69 FRG-17n.a.0.01518.560.040.02n.a.0.0450.45 FRG-1100.010.05518.250.1050.080.030.130.43 FRG-1110.0050.0218.610.030.030.010.0650.16 FRG-1130.0050.01518.930.020.030.0050.070.15 FRG-1150.0050.01517.650.0350.040.0050.1350.16 FRG-230.050.02518.450.0250.030.010.0650.19 FRG-25n.a.0.01518.050.0240.03n.a.0.0640.25 FRG-270.0450.0218.780.020.040.010.090.19 FRG-2100.010.03518.430.0550.070.0150.160.44 FRG-2120.0080.08317.750.0630.090.010.2160.98 FRL-120.0050.03317.570.0250.05n.a.0.1850.58 FRL-140.0050.03417.620.0280.06n.a.0.1890.66 FRL-160.0050.04517.930.030.040.010.150.43 FRL-170.0050.0517.060.020.040.010.130.41 SSJ-120.0140.09118.140.0540.080.0270.6922.25 SSJ-140.0120.09817.760.0610.070.0170.3491.5 SSJ-160.0150.05518.550.0250.070.020.0950.33 SSJ-180.0160.06618.410.0230.170.0290.1520.28 SSJ-1100.020.09518.340.0250.130.050.2450.36 TRF2214 60.0-62.50.0750.44614.890.030.190.3540.5011.25 WEX164 18.0-26.00.1310.70414.050.1780.680.2170.5063.04 WEX366 9.0-10.00.1670.37814.380.0561.110.0470.7246.59 EPK Vermiforms 0.010.01118.190.0160.010.0260.0190.3 EPK Mica 0.3490.4417.190.0085.750.0130.3931.62 EPK Feldspar 0.3050.0175.7030.1186.160.3090.010.07 Reworked Cypresshead Formation FL SSD-110.060.14515.060.8630.340.0724.7781.48 SSD-130.010.04220.020.7280.070.050.4340.39 SSD-160.0080.06117.490.0470.040.0060.1370.46 SSD-170.0150.06517.870.0350.060.010.1550.51 SSD-1100.0150.17517.80.0350.090.070.1350.73 Note: n.a., element concentration was below the detection limit.
280 Table H-1. (continued). Sample ID IntervalNa (%)Mg (%)Al (%)P (%)K (%)Ca (%)Ti (%)Fe (%) Cypresshead Formation GA J-1 20.0170.1498.0680.0460.160.0090.845.35 J-1 40.0250.12518.120.0350.430.010.6153.13 J-1 60.020.1117.790.0150.310.0050.461.33 L-1 30.0140.07919.240.0490.15n.a.0.6161.2 L-1 5n.a.0.0318.710.0190.04n.a.0.2240.59 L-1 60.0150.07517.890.0250.220.0050.511.13 B-1 20.0380.3814.960.03810.010.3993.93 B-1 30.0470.42415.520.0931.280.0090.437.63 B-1 50.040.42517.620.040.890.0650.4452.54 Hawthorn Group, Coosawhatchie Formation FL/GA MCB109 15.0-20.00.0661.01511.690.0371.250.7430.4164.18 J-1 BC0.0571.07111.770.0130.30.3120.3714.22 Huber Formation GA KGa-2 0.010.0217.960.020.030.010.8550.66 ECCI-CB 0.020.02517.560.0250.070.0050.6650.63 Buffalo Creek Formation GA KGa-1 0.020.01518.950.025n.a.0.020.940.13 ECCI-BC 0.0150.0118.650.035n.a.0.0150.750.11 TKC-EA 0.010.0118.970.0250.020.0050.870.12 DBK-B93 0.0150.01517.20.03n.a.0.0151.3270.17 Note: n.a., element concentration was below the detection limit.
281Table H-2. Minor element concentr ations for samples used in this study (concentration are in ppm). Sample ID IntervalRbSrCsBaScVCrMnCoNiCuZnGaYZrNbPbThU Cypresshead Formation FL EPK36-J-1225-272.748.80.312932443 3.145814228645.26625.12.51 EPK36-J-1235-4031210.372.851816103.73993472814.273.756374.79 EPK36-J-1246-4220.127.116.113.110364363.12276463673261175.72.06 EPK36-J-1250-518.104.22.1681.110321672.4182121354.224.96693.51.93 EPK36-J-1259-622.214.171.1248.316604592.6204933398.738111166.94.86 EPK31-P-4035-454.521.30.337.212524561.8142320364.622.261194.717.7 EPK31-P-4050-6126.96.36.1997.7214721102.31836243416.118.561064.4103 EPK31-P-4062-6517.61231.4277207767123.42837453341.230.561157.892.9 EPK30-V-622-2425.73482.3127663651785.232471661541.61771611315.15.58 EPK30-V-630-3523.120402.5419137942915.2515316833257255229856.946.3 EPK30-V-648-5312.75491.8189127938264.65043463273.2151121032329.3 EPK30-V-658-63619.80.344.4148446 1.7137132358.223.151153.818.9 EPK30-V-668-738.925.90.571.32211320112.51715153712.6238764.316.8 FRG-1 34.62290.811476159114.63067452326.984.995917.75.81 FRG-1 60.81250.180.851617115.53915376198.519.1210861.62 FRG-1 71.8760.244.841841 3.5195837224.816.421307.41.16 FRG-1 106.74540.61516211073.51744293013.824.9536711.71.57 FRG-1 111.7188.8.131.52166 2.1185530263.617.721257.10.91 FRG-1 131.733.1 20.7518841.91612255271.721.52635.61.04 FRG-1 152.3184.108.40.2061271051.820291130345.518.252556.51.38 FRG-2 31.7220.127.116.11111352.1181411292.622.72405.30.83 FRG-2 52.418.104.22.1682033 1.5144324312.416.32845.91.01 FRG-2 72.322.214.171.12416741.3145125312.115.63753.91.19 FRG-2 105.866.90.517811391381.9185027361623.36151102.27 FRG-2 121099.20.9228155747102.62040303425.929.9710894.62 FRL-1 126.96.36.1997.614505053.7227442354466837.91.28 FRL-1 44.726.80.432.914615353.8255741355.349.6615210.61.16 FRL-1 63.432.10.435.815351382.4227639347.843.351278.31.13 FRL-1 73.622.80.42613301462.5165137338.432.15656.11.27 SSJ-1 25.887160.657892154.64247284817.6216307529.43.11 SSJ-1 48.370.11.548.366158144.33456364210.4117166218.32.67 SSJ-1 65.949.20.629.56271192.2414620332.618.84442.23.03 SSJ-1 810.847.30.792.1125642 1.4104422362.423.95502.47.93 SSJ-1 101143.20.870.2188537101.4115320374.133.79484.614.1 SSD-1 121.37772.4505222451602895.5338454183167163025351414531.3 SSD-1 35.21440.610374177214.547103603716.41801813656.87.54 SSD-1 188.8.131.521.784450 3.418613729834.8599101.64 SSD-1 74.5410.466.910451692225431338.928.851176.82.71 SSD-1 107.8380.782.293318221.7202720327.3235874.42.04 TRF2214 60.0-62.529.2663.71841511960323.6823283716.8127264410.62.43 WEX164 18.0-26.087.73209.6305261891695110.22362873510563.9194222.313.3 WEX366 9.0-10.012411810.932223183122505.9181071004232.6238294326.69.9 EPK Vermiforms 184.108.40.206.5113917 1.8840313920.77.9 2511.21 EPK Mica 23138.52.616103815365872.478453529.313031312.91.46 EPK Feldspar 1912651.31870 61280.5121717102.921.5 580.51.63
282Table H-2. (continued). Sample ID IntervalRbSrCsBaScVCrMnCoNiCuZnGaYZrNbPbThU Cypresshead Formation GA J-1223.335.94.790.221195110356.34759514922.3231375136.65.21 J-1439.434.83.51432419077355.422322164417.974.7223621.23.04 J-16220.127.116.110217674533652879397376.657.5164314.42.03 L-1317.440.52.41161910080205.83865474010.71042211334.32.87 L-156.410.70.734.710685987.15617374293.938.776510.81.16 L-1621.518.104.22.16868024216.1397547316.860.3198716.61.85 B-1297.452.69.22272514984547.32966773214.482.3155715.94.01 B-1311677.410.8270311989110810.43368963424.383.6156420.45.55 B-1588.51128.120136168305183480753850.274.4156018.54.03 Hawthorn Group, Coosawhatchie Formation FL/GA MCB109 15.0-20.013911210.9304211771275611.5243186294184.9152512.74.38 J-1BC42.960.82.6180221396513644.44642832614737.6122911.31.72 Huber Formation GA KGa-22.214.171.1240.3151052959.43972707020.178.63743133.52 ECCI-CB2.645.20.379.420761669.93569594810.151.9242914.74.39 Buffalo Creek Formation GA KGa-1n.a.43.2n.a.41.81922338n.a.3.3176324538.498.1391836.32.27 ECCI-BCn.a.50.5n.a.64.61412716n.a.3.11811655535.811530n.a.31.41.53 TKC-EA0.636.7n.a.64.62110226n.a.4.4208134468.1111371331.18.72 DBK-B930.341.2n.a.1552224990n.a.1537104910.314951n.a.31.96.52
283 LIST OF REFERENCES Addington, L.D., Kuehl, S.A., and McNinch, J.E ., 2007, Contrasting modes of shelf sediment dispersal off a high-yield ri ver: Waiapu River, New Zealand: Marine Geology, v. 243, p. 18. Allmon, W.D., 1992, Whence southern Flor idas Plio-Pleisto cene shell beds? in Scott, T.M., and Allmon, W.D., eds., Plio-Pleisto cene stratigraphy and paleontology of southern Florida: Florida Geological Survey Sp ecial Publication 36, p. 1. Alpers, C.N., and Brimhall, G.H., 1988, Middle Mio cene climatic change in the Atacama Desert, northern Chile; Evidence from supergene mineralization at La Escondida. Geological Society of America, Bulletin, v. 100, p. 1640. Alt, D., 1974, Arid climate control of Miocene sedimentation and origin of modern drainage, southeastern United States, in Oaks, R.Q., Jr., and Dunbar, J.R., eds., Post-Miocene Stratigraphy, Central and Sout hern Atlantic Coastal Plai n: Logan, Utah, Utah State University Press, p. 21. Anovitz, L.M., Perkins, D., and Essene, E.J ., 1991, Metastability in near-surface rocks of minerals in the system Al2O3-SiO2-H2O: Clays and Clay Minerals, v. 39, p. 225. Aparicio, P., and Galn, E., 1999, Mineralogical interference on kaolinite crystallinity index measurements: Clays and Clay Minerals, v. 47, p. 12. Applin, E.R., and Applin, P.L., 1964, Logs of selected wells in the coastal plain of Georgia: Georgia Geological Survey, Bulletin 74, 229 p. Arostegui, J., Irabien, M.I., Ni eto, F., Sangesa, J., and Zuluaga, M.C., 2001, Microtextures and the origin of muscovite-kaolinite intergrowt hs in sandstones of the Utrillas Formation, Basque Cantabrian Basin, Spain: Clays and Clay Minerals, v. 49, p. 529. Arrington, D.V., 1985, Geology of the Interlachen Karstic Highlands, Florida [unpublished M.S. thesis]: University of Florid a, Gainesville, Florida, 130 p. Artioli, G., Bellotto, M., Gualtieri, A., and Pavese, A., 1955, Nature of structural disorder in natural kaolinites: a new model based on com puter simulation of powder diffraction data and electrostatic energy calculation: Clays and Clay Minerals, v. 43, p. 438. Ashley, R.P., and Silberman, M.L., 1976, Direct da ting of mineralization at Goldfield, Nevada, by potassium-argon and fission-track methods: Economic Geology, v. 71, p. 904. Austin, R.S., 1978, The origin of Georgias kaolin deposits: Georgia Geological Survey, Information Circular 49, p. 10. Austin, R.S., 1998, Origin of kaolin of the s outheastern U.S.: Mining Engineering, v. 50, p. 52 57.
284 Axelrod, D.I., and Raven, P.H., 1985, Origins of the Cordilleran flora: Journal of Biogeography, v. 12, p. 21. Axelrod, D.I., and Raven, P.H., 1986, Late Cretaceous and Tertiary vegeta tion history of Africa, in Werger, M.J., and Bruggen, A.C., eds., Biogeography and Ecology of Southern Africa: The Hague, Junk, p. 77. Bailey, S.W., 1980, Structure of layer silicates, in Brindley, G.W., and Goodman, G., eds., Crystal structures of clay mine rals and their X-ray identifica tion: Mineralogical Society of London, p. 1. Balsillie, J.H., 1995, Will iam F. Tanner on environmental clastic granulometry: Florida Geological Survey, Special Publication 40, 144 p. Banfield, J.F., and Eggleton, R.A., 1990, Analytical transmission electron mi croscope studies of plagioclase, muscovite, and K-feldspar weat hering: Clays and Clay Minerals, v. 38, p. 77 89. Barron, E.J., 1989, Severe storms during earth histor y: Geological Society of America, Bulletin, v. 101, p. 601. Bell O.G., 1924, A preliminary report on the clays of Florida (exclusive of Fuller's earth): Florida Geological Surve y, 15th Annual Report, p. 53. Belshaw, N.S., Freedman, P.A., ONions, R.K., Frank, M., and Guo, Y., 1998, A new variable dispersion double-focusing plasma mass spectrom eter with performance illustrated for Pb isotopes: International Journa l of Mass Spectrometry, v. 181, p. 51. Berggren, W.A., 1973, The Pliocene time-scale: calibration of planktonic foraminiferal and calcareous nannoplankton z ones: Nature, v. 243, p. 391. Berggren, W.A., Hilgen, F.J., Langereis, C.G., Kent, D.V., Obradovich, J.D., Raffi, I., Raymo, M.E., and Shackleton, N.J., 1995a, Late Neogene chronology: new perspectives in highresolution stratigraphy: Geological Society of Am erica, Bulletin, v. 107, p. 1272. Berggren, W.A., Kent, D.V., Swisher, C.C ., III, and Aubry, M.-P., 1995b, A revised Cenozoic geochronology and ch ronostratigraphy, in Berggren, W.A., Kent, D.V., Aubry, M.-P., and Hardenbol, J., eds., Geochronology, Time Scal es and Global Stratigraphic Correlation: SEPM, Special Publication 54, p. 129. Bertaut, F., 1950, Rais de Debye-Scherrer et re partition des dimensions des domains de Bragg dans les poudres polycristallines: Acta Crystallographica, v. 3, p. 14. Bhatia M.R., 1985, Rare earth element geochemist ry of Australian Paleozoic graywackes and mudrocks: provenance and tectonic c ontrol: Sedimentary Geology, v. 45, p. 97. Bish, D.L., and Von Dreele, R.B., 1989, Rietveld refinement of non-hydrogen atomic positions in kaolinite: Clays and Clay Minerals, v. 37, p. 289.
285 Bishop E.W., 1956, Geology and ground water resources of Highlands County, Florida: Florida Geological Survey, Report of Investigation 15. Blanchard, F.N., 1972, Physical and chemical data for crandallite from Alachua County, Florida: American Mineralogist, v. 57, p. 473. Blow, W., 1969, Late middle Eocene to Recent planktonic foraminiferal biostratigraphy, in Brnnimann, P., and Renz, H.H., eds., Proceedings of the First International Conference on Planktonic Microfossils, v. 1, p. 199. Bock, B., McLennan, S.M., and Hanson, G.N., 1994, Rare earth element redistribution and its effects on the neodymium isotope system in the Austin Glen Member of the Normanskill Formation, New York, USA: Geochimi ca et Cosmochimica Acta, v. 58, p. 5245. Bouquillon, A., France-Lanord, C., Michard, A., and Tiercelin, J-.J., 1990, Sedimentology and isotopic chemistry of the Bengal fan sediments: the denudation of the Himalaya, in Cochran, J.R., et al., eds., Proceedings of th e Ocean Drilling Program, Scientific Results 116, p. 43. Buie B.F., 1978, The Huber Formation of eastern Georgia, in Short Contributions to the Geology of Georgia: Georgia Geologi cal Survey, Bulletin 93, p. 1. Buie B.F., and Fountain R.C., 1967, Tertiary and Cretaceous age of kaolin deposits in Georgia and South Carolina (abstract): Geological Society of America, Program with Abstracts, v. 19, p. 465. Bukry, D., 1991, Paleoecological transect of western Pacific Ocean late Pliocene coccolith flora: Part I Tropical Ontong-Java Plateau at ODP 806B: U.S. Geological Survey, Open File Report 91-552, 35 p. Bundy, W.M., 1993, The diverse industr ial applications of kaolin, in Murray, H.H., et al., eds., Kaolin Genesis and Utilization: The Clay Mi nerals Society, Special Publication No. 1, p. 43. Cases, J.M. Litard, O., Yvon, J., and Delon, J. F., 1982, tude des propirts cristallochimiques, morphologiques, superficielles de kaolinites dsordonnes: Bulletin of Minralogico, v. 105, p. 439. Chamberlain, C.K., 1978, Recogniti on of trace fossils in cores, in Basan, P.B., ed., Trace Fossil Concepts (SEPM Short Course no. 5), SEPM, Tulsa, Oklahoma, p. 119. Chapman, C.E., 2008, Interplay of oceanographic and paleoclimate events with tectonism during middle to late Miocene sedimentation acro ss the southwestern USA: Geosphere, v. 4, p. 976. Chaudhrui, S., and Cullers, R.L., 1979, The distribu tion of rare-earth elements in deeply buried Gulf coast sediments: Chemical Geology, v. 24, p. 327.
286 Chvez, C.L., and Johns, W., 1995, Mineralogical and ceramic properties of refractory clays from central Missouri (USA): A pplied Clay Science, v. 9, p. 407. Coakley, J.P., and Syvitski, J.P.M., 1991, SediGraph technique, in Syvitski, J.P.M., ed., Principles, Methods, and Application of Particle Size Analysis: Cambridge University Press, New York, p. 129. Condie, K.C., Dengate, J., and Cullers, R.L., 1995, Behavior of rare earth elements in a paleoweathering profile on granodiorite in the Front Range, Colorado, USA: Geochimica et Cosmochimica Acta, v. 59, p. 279. Cooke, C.W., 1930, Correlation of coastal terraces: Journal of Geology, v. 38, p. 577. Cooke, C.W., 1936, Geology of the Coastal Plain of South Carolina: Geological Survey, Bulletin 867, p. 132. Cooke, C.W., 1945, Geology of Florida: Flor ida Geological Survey, Bulletin 29, 339 p. Cooke, C.W., and MacNeil, F.S., 1952, Tertiary stratigraphy of South Carolina: U.S. Geological Survey, Professional Paper 243-B, 29 p. Cooke, C.W., and Mossom, S., 1929, Geology of Florida: Florida Geological Survey, Annual Report 20, p. 29. Cronin, T.M., 1991, Pliocene shallow water paleo ceanography of the North Atlantic Ocean based on marine ostracods: Quater nary Science Reviews, v. 10, p. 75. Cronin, T.M., and Dowsett, H.J., 1990, A quantit ative micropaleontological method for shallow marine paleoclimatology: appli cation to Pliocene deposits of the western North Atlantic ocean: Marine Micropaleontology, v. 16, p. 117. Crowley, T.J., 1996, Pliocene climates: the nature of the problem: Marine Micropaleontology: v. 27, p. 3. Cullers, R.L., 1988, Mineralogical and chemical ch anges of soil and stream sediment formed by intense weathering of the Danburg granite Georgia, U.S.A.: Lithos, v. 21, p. 301. Cullers, R.L., Barrett, T., Carlson, R., and Robi nson, R., 1987, Rare-earth di stributions in size fractions of Holocene soil and stream sediment, Wet Mountains region, Colorado, U.S.A.: Chemical Geology, v. 63, p. 275. Cunningham, K.J., Bukry, D., Sato, T., Barron, J.A., Guertin, L.A., and Reese, R.S., 2001, Lithostratigraphy, sequence stratigraphy and biostratigraphy of a carbonate ramp and bounding siliciclastics (Late Mio cene-Pliocene), southern Fl orida: Florida Geological Survey, Special Publication 49, p. 35.
287 Cunningham, K.J., Locker, S.D., Hine, A.C., Bukry, D., Barron, J.A., and Guertin, L.A., 2003, Interplay of Late Cenozoic siliciclastic supply and carbonate response on the southeast Florida Platform: Journal of Sedimentary Research, v. 73, p. 31. Cunningham, K.J., McNeill, D.F., Guertin, L.A., Ciesielski, P.F., Scott, T.M., and de Verteuil, L., 1998, New Tertiary stratigraphy for the Florid a Keys and southern peninsula of Florida: Geological Society of Am erica, Bulletin, v. 110, p. 231. Curran, H.A., 1985, The trace fossil assemblage of a Cretaceous nearshore environment: Englishtown Formation of Delaware, U.S.A., in Curran, H.A., ed., Biogenic Structures: Their use in Interpreting Depositional Environm ents: SEPM, Special Publication 35, Tulsa, Oklahoma, p. 261. Davis N.B., 1916, The plasticity of clay and its relation to mode of origin. American Institute of Mining, Metallurgical, and Petroleum Engineers, Transactions, v. 51, p. 451. Davis, R.A., Hine, A.C., and Shinn, E.A., 1992, Holocene coastal develo pment on the Florida peninsula, in Fletcher, C.H., III, and Wehmiller, J.F., eds., Quaternary Coasts of the United States: Marine and Lacustrine System s, SEPM, Special Publication 48, p. 193. Dennison, J.M., and Stewart, K.G., 2001, Regional structural and stratigraphic evidence for dating Cenozoic uplift of Southern Appalach ian highlands (abstrac t): Abstracts with Programs, 50th Annual Meeting, Southeastern Sec tion of the Geological Society of America, v. 33, p. 6. Dill, H.G., Fricke, A., and Henning, K.-H., 1995, Th e origin of Baand REE-bearing aluminumphosphate-sulphate minerals from the Lohrhe im kaolinitic clay deposit (Rheinisches Schiefergebirge, Germany): A pplied Clay Science, v. 10, p. 231. Dobkins, J.E., Jr., and Folk, R.L., 1970, Shape development on Tahiti-Nui: Journal of Sedimentary Petrology, v. 40, p. 1167. Doering, J.A., 1960, Quaternary surface formations of southern part of Atlantic Coastal Plain: Journal of Geology, v. 68, p. 182. Dombrowski T., 1982, Abundance, di stribution and origin of thorium in the Georgia kaolins [unpublished M.S. thesis]: Indiana Un iversity, South Bend, Indiana, 85 p. Dombrowski T., 1992, The use of trace elemen ts to determine provenance relations among different types of Georgia kaolins [unpublishe d Ph.D. dissertation]: Indiana University, South Bend, Indiana, 212 p. Dombrowski T., 1993, Theories of origin for the Georgia kaolins: a review, in Murray, H.H., et al., eds., Kaolin Genesis and Utilization: The Clay Minerals Society, Special Publication 1, p. 75.
288 Dombrowski T., and Murray H.H., 1984, Thorium a key element in differentiating Cretaceous and Tertiary kaolins in Georgia and South Caro lina: Proceedings of the 27th International Geological Congress, v. 15, p. 305. Dott, R.H., Jr., and Bourgeois, J., 1982, Hummocky stratification: signifi cance of its variable bedding sequences: Geological Societ y of America, Bulletin, v. 93, p. 663. Dowsett, H.J., Barron, J., and Poore, R., 1996, Middle Pliocene sea surface temperatures: a global reconstruction: Marine Micropaleontology, v. 27, p. 13. Dowsett, H.J., and Cronin, T., 1990, High eustatic sea level during the middl e Pliocene: evidence from the Southeastern U.S. Atlant ic Coastal Plain: Geology, v. 18, p. 435. Dowsett, H.J., and Laubere, P., 1992, High resolution Late Pliocene sea-surface temperature record from the Northwest Atlantic Ocean: Marine Micropaleontology, v. 20, p. 91. Dowsett, H., Thompson, R., Barron, J., Cronin, T., Fleming, F., Ishman, S, Poore, R., Willard, D., and Holtz, T., Jr., 1994, Joint investigatio ns of the Middle Pliocene climate I: PRISM paleoenvironmental reconstructions: Global and Planetary Change, v. 9, p. 169. Drever, J.I., 1973, The preparation of oriented clay mineral specimens for X-ray diffraction analysis by a filter-membrane peel techni que: American Mineralogist, v. 58, p. 553. Drever, J.I., 1982, The Geochemistry of Natura l Waters: Prentice-Hall, Englewood Cliffs, New Jersey, 388 p. DuBar, J.R., 1974, Summary of the Neogene stratigraphy of southern Florida, in Oaks, R.Q., Jr., and DuBar, J.R., eds., Post-Miocene Stratigrap hy of the Central and Southern Atlantic Coastal Plain: Utah State Univ ersity Press, Logan, Utah, 206 p. Dypvik, H., and Brunfelt, A.O., 1979, Distribution of rare earth elem ents in some North Atlantic Kimmeridgian black shales: Nature, v. 278, p. 339. Eargle P.H., 1955, Stratigraphy of the outcropping Cr etaceous rocks of Georgia: U.S. Geological Survey, Bulletin 1014, 101 p. Eberl, D.D., Drits, V.A., and rodo, J., 1998, Deducing growth mechanisms for minerals from the shapes of crystal size distributions: American Journal of Science, v. 298, p. 499. Eberl, D.D., Drits, V.A., rodo, J., and Nesch, R., 1996, MudMaster: a program for calculating crystallite size dist ributions and strain from th e shapes of x-ray diffraction peaks: U.S. Geological Surve y, Open-File Report 96-171, 46 p. Eberli, G.P., 2000, The record of Neogene sealevel in the prograding carbonates along the Bahamas transect Leg 166 synthesis, in Swart, P.K., Eberli, G.P., Malone, M.J., and Sarg, J.F., eds., Proceedings of the Ocean drilling Program, Scientific Results, v. 166, p. 167.
289 Eberli, G.P., Swart, P.K., Malone, M., and Leg 166 Shipboard Party, 1997, Proceedings of the Ocean Drilling Program, Initial Reports 166: College Station, Texas, Ocean Drilling Program. Edgar, N.T., and Cecil, C.B., 2003, Influence of climate on deep-water clastic sedimentation: application of a modern mode l, Peru-Chile Trough, to an ancient system, Ouachita Trough, in Cecil, C.B., and Edgar, N.T., eds., Clim ate Controls on Stratigraphy: SEPM, Special Publication 77, p. 185. Fedorov, A.V., Dekens, P.S., McCarthy, M., Ravelo, A.C., deMenocal, P.B., Barreiro, M., Pacanowski, R.C., and Phila nder, S.G., 2006, The Pliocene paradox (mechanisms for a permanent El Nio: Science, v. 312, p. 1485. Fein, J.B., 1991, Experimental study of aluminum-oxalate complexing at 80 C: implications for the formation of secondary porosity with in sedimentary reservoirs: Geology, v. 19, p. 1037. Fein, J.B., and Hestrin, J.E., 1994, Experiment al studies of oxalate complexation at 80 C: gibbsite, amorphous silica, and quartz solubiliti es in oxalate-bearing fluids: Geochimica et Cosmochimica Acta, v. 58, p. 4817. Feng, R, and Kerrich, R., 1990, Geochemistry of fi ne-grained clastic sediments in the Archean Abitibi greenstone belt, Canada: Implications for provenance and tectonic setting: Geochimica et Cosmochimica Acta, v. 54, p. 1061. Flicoteaux, R., and Lucas, J., 1984, W eathering of phosphate minerals, in Nriagu, J.O., and Moore, P.B., eds., Phosphate Minerals, Springer-Verlag, New York, p. 292. Frey, R.W., 1970, Environmental significance of recen t marine lebenspuren near Beaufort, North Carolina: Journal of Paleontology, v. 44, p. 507. Frost, C.D., and O'Nions, R.K., 1984, Nd evidence for Proterozoic crustal development in the Belt-Purcell Supergroup: Nature, v. 312, p. 53. Frost, C.D., and Winston, D., 1987, Nd isotope systematics of coarseand fine-grained sediments: examples from the middle Proterozoic Belt-Purcell Supe rgroup: Journal of Geology, v. 95, p. 309. Fountain, K.B., and McClellan, G.H., 1993, Origin of northcentral Florida kaolinite deposits in light of mineralogical and geochemical evid ence (abstract): 1993 A nnual Meeting of the Clay Minerals Society, San Diego, California, p. 132. Fullagar, P.D., Goldberg, S.A., and Butler, J.R., 1997, Nd and Sr isotopic characterization of crystalline rocks from the southern Appal achian Piedmont and Blue Ridge, North and South Carolina, in Sinha, A.K., Whalen, J.B., and Hogan, J.P., eds., The Nature of Magmatism in the Appalachian Orogen: Ge ological Society of America Memoir 191, p. 165.
290 Galn E., Aparicio, P., Gonzlez, I., and Miras, A., 1998, Contribution of mu ltivariate analysis to the correlation of some properties of kao lin with its mineralogical and chemical composition: Clay Minerals, v. 33, p. 65. Ghosh, D.K., and Lambert, R.StJ., 1989, Nd-Sr isotopic study of Proterozoic to Triassic sediments from southeastern British Columbia : Earth and Planetary Science Letters, v. 94, p. 29. Ginsburg, R.N., Brown, K.M., and Chung, G.S ., 1989, Siliciclastic f oundations of South Floridas Quaternary carbonates (abstract): Ge ological Society of America, Programs with Abstracts, v. 21, p. A-290. Gladney, E.S., Jones, E.A., Nickell, E.J., and Roelandts, I., 1990, 1988 Compilation of elemental concentration data for USGS Basalt BCR1: Geostandards Newsletter, v. 14, p. 209. Gleason, J.D., Patchett, P.J., Dickinson, W.R., and Ruiz, J., 1994, Nd isotopes link Ouachita turbidites to Appalachia n sources: Geology, v. 22, p. 347. Gleason, J.D., Patchett, P.J., Di ckinson, W.R., and Ruiz J., 1995, Nd isotopic constraints on sediment sources of the Ouachita-Marathon fold belt: Geological Society of America, Bulletin, v. 107, p. 1192. Gohn G.S., Bybell L.M., Christopher R.A., Owens J.P., and Smith C.C., 1979, Cretaceous and Paleogene stratigraphic framework of the South Carolina and Georgia coastal margins (abstract): Programs and Abstracts for the 2nd Symposium on the Geology of the Southeastern Coastal Plain, Americus, Georgia, p. 11. Goodbred, S.L., Jr., and Kuehl, S.A., 2000, Enorm ous Ganges-Brahmaputra sediment discharge during strengthened early Holo cene monsoon: Geology, v. 28, p. 1083. Guertin, L.A., McNeill, D.F., Lidz, B.H., a nd Cunningham, K.J., 1999, Chronologic model and transgressive-regressive signa tures in the late Neogene si liciclastic foundation (Long Key Formation) of the Florida Keys: Jour nal of Sedimentary Research, v. 69, p. 653. Hancock, G., and Kirwan, M., 2007, Summit erosion rates deduced from 10Be: implications for relief production in the central Appalachians: Geology, v. 35, p. 89. Hansen, B.C.S., Grimm, E.C., and Watts, W.A ., 2001, Palynology of the Peace Creek site, Polk County, Florida: Geological Society of America, Bulletin, v. 113, p. 682. Haq, B.U., Hardenbol, J., and Vail, P.R., 1988, Mesozoic and Cenozoic chronostratigraphy and cycles of sea-level change, in Wilgus, C.K., Hastings, B.S., Kendall, C.G.St.C., Posamentier, H.W., Ross, C.A., and Van Wa goner, J.C., eds., Sea Level ChangesAn Integrated Approach: SEPM, Special Publication 42, p. 72. Harris, S.E., and Mix, A.C., 2002, Climate and te ctonic influences on c ontinental erosion of tropical South America, 0-13 Ma: Geology, v. 30, p. 447.
291 Harris, W.G., Morrone, A.A., and Coleman, S. E., 1992, Occluded mica in hydroxyl-interlayered vermiculite grains from a highly-weathere d soil: Clays and Clay Minerals, v. 40, p. 32. Hassanipak, A.A., and Eslinger, E. V., 1985, Mineralogy, crystallinity, 18O/16O, and D/H of Georgia kaolins: Clays and Clay Minerals, v. 33, p. 99 Hathaway J.C., 1956, Procedure for clay mineral analysis used in the sedimentary petrology laboratory of the USGS: Clay Minerals Bulletin, v. 3, p. 8. Haywood, A.M., Sellwood, B.W., and Valdes, P. J., 2000, Regional warming: Pliocene (3 Ma) paleoclimate of Europe and the Mediterranean: Geology, v. 28, p. 1063. Healy, H.G., 1975, Terraces and shorelines of Flor ida: Florida Geological Survey, Map Series 71. Hem, J.D., 1985, Study and interpretation of the ch emical characteristics of natural water, 3rd Edition: U.S. Geological Survey, Water-Supply Paper 2254, 263 p. Henderson, P., Marlow, C.A., Molleson, T.I., an d Williams, C.T., 1983, Patterns of chemical change during bone fossilization: Nature, v. 306, p. 358. Hendrix, W.P., and Orr, C., 1972, Automatic sedimentation size analysis instrument, in Groves, M.J., and Wyatt-Sargent, J.L., eds., Particle Size Analysis 1970: Society of Analytical Chemistry, London, p. 133. Hendry, C.W., Jr., and Yon, J.W., Jr., 1967, St ratigraphy of Upper Miocene Miccosukee Formation, Jefferson and Leon counties, Florid a: American Association of Petroleum Geologists, Bulletin, v. 51, p. 250. Herrick S.M., 1961, Well logs of the coastal pl ain of Georgia. Geor gia Geological Survey, Bulletin 70, 461 p. Herrick, S.M., and Vorhis, R.C., 1963, Subsurface geology of the Georgia Coastal Plain: Georgia Geological Survey, Information Circular 25, 78 p. Heuberger, D., 1995, The influence of post-Miocene sediments on soil formation in Big Scrub, Ocala National Forest [unpublished M.S. thesis ]: University of Florida, Gainesville, Florida, 195 p. Hill, G.W., 1985, Ichnofacies of a modern size-g raded shelf, northwestern Gulf of Mexico, in Curran, H.A., ed., Biogenic Structures: Their use in Interpreting Depositional Environments: SEPM, Special Publication 35, Tulsa, Oklahoma, p. 195. Hinkley, D.N., 1961, Mineralogical a nd chemical variations in the kaolin deposits of the coastal plain of Georgia and South Carolina: N.S. F. Technical Report, Pennsylvania State University, University Park, Pennsylvania, 180 p.
292 Hinkley, D.N., 1963, Variability in crystallinity values among the kaolin deposits of the coastal plain of Georgia and South Caro lina: Clays and Clay Minerals, v. 13, p. 229. Hobgood, J.S., and Cerveny, R.S., 1988, Ice-age hurricanes and tropical storms: Nature, v. 333, p. 243. Horton, J.W., Jr., Drake, A.A., Jr., and Rankin, D.W., 1989, Techtonostratig raphic terranes and their Paleozoic boundaries in the central and southern Appala chians: Geological Society of America Special Paper 230, p. 213. Howard, J.D., and Reineck, H.E., 1972, Georgi a coastal region, Sapelo Island, U.S.A.: sedimentology and biology. IV. Physical and biogenic sedi mentary structures of the nearshore shelf: Senckenb ergiana Maritima, v. 4, p. 81. Hoyt, J.H., and Hails, J.R., 1974, Pleistocene stratigraphy of southeastern Georgia, in Oaks, R.Q., Jr., and DuBar, J.R., eds., Post-Miocene Stratigraphy Central and Southern Atlantic Coastal Plain: Utah State Univ ersity Press, Logan, Utah, p. 191. Huddlestun, P.F., 1988, A revision of the lithostratigra phic units of the coastal plain of Georgia: The Miocene through Holocene: Georgia Geological Survey, Bulletin 104, 162 p. Huddlestun, P.F., 1993, An overview of the Neogene of Georgia, in Zullo, V.A., Harris, W.B., Scott, T.M., and Portell, R.W., eds., Th e Neogene of Florida and Adjacent Regions, Proceedings of the Third Bald Head Island Conference on Coastal Plain Geology: Florida Geological Survey, Special Publication 37, p. 91. Hurst, V.J., and Pickering, S.M., Jr., 1989, Signi ficance of crystalline size and habit in determining origin and industrial applications of Georgia kaolins: Geological Society Southeast Section, Abstracts with Programs, p. 22. Hurst, V.J., and Pickering, S.M., Jr., 1997, Origin and classification of coastal plain kaolins, southeastern USA, and the role of groundwater and microbial action: Clays and Clay Minerals, v. 45, p. 274. Hurst, V.J., and Rigsby, W.E., 1984, Intergrowths and aggregates of natural kaolinite, in Bailey, G.W., ed., Proceedings of the 42nd Annual meeting of the Electron Microscopy Society of America, p. 22. Isphording, W.C., 1970, Late Tertiary paleoclimate of eastern United States: American Association of Petroleum Ge ologists, Bulletin, v. 54, p. 334. Isphording, W.C., 1971, Provenance and petrogra phy of Gulf Coast Miocene sediments, in Geological Review of Some No rth Florida Mineral Resources: Southeastern Geological Society, Fifteenth Field Conference Guidebook, p. 43. Jacobsen, S.B., and Wasserberg, G.J., 1980, Sm Nd isotopic evolution of chondrites: Earth and Planetary Science Letters, v. 41, p. 139.
293 Jeong, G.Y., 1998a, Formation of vermicular kao linite from halloysite aggregates in the weathering of plagioclase: Clay s and Clay Minerals, v. 46, p 270. Jeong, G.Y., 1998b, Vermicular kaolinite epitactic on primary phyllosilicates in the weathering profiles of anorthosite: Clay s and Clay Minerals, v. 46, p. 509. Jiang, W-.T., and Peacor, D.R., 1991, Transmission electron microscopic study of the kaolinization of musc ovite: Clays and Clay Minerals, v. 39, 1. Jiang, D., Wang, H., Ding, Z., Lang, X., and Drange, H., 2005, Modeling the middle Pliocene climate with a global atmospheric general circulation model: Journal of Geophysical Research, v. 110, p 14. Johnson, R.A., 1989, Geologic descriptions of select ed exposures in Florida: Florida Geological Survey, Special Publication 30, 175 p. Joussein, E., Petit, S., Churchman, J., Theng, B., Righi, D., and Delvaux, B., 2005, Halloysite clay minerals a review: Clay Minerals, v. 40, p. 383. Kane B.C., 1984, Origin of the Grandin (Plio -Pleistocene) sands, western Putnam County, Florida [unpublished M.S. thesis]: University of Florida, Gainesville, Florida, 85 p. Keller W.D., 1977, Scan electron micrographs of kao lins collected from dive rse environments of origin-IV Georgia kaolins and kaolinizing sour ce rocks: Clays and Clay Minerals, v. 25, p. 311. Keller W.D., 1978, Classification of kaolins exemp lified by their textures in scanning electron micrographs: Clays and Clay Minerals, v. 26, p. 1. Kennett, J.P., and Hodell, D.A., 1993, Evidence for relative climatic stability of Antarctica during the early Pliocene: a marine perspe ctive: Geografiska Annaler., v.75A, p. 205. Kennett, J.P., and Hodell, D.A., 1995, Stability or instability of Antarctic ice sheets during warm climates of the Pliocene?: GSA Today, v. 5, p. 10. Kesler T.L., 1956, Environment and origin of th e Cretaceous kaolin deposits of Georgia and South Carolina: Economic Geology, v. 51, p. 541. Ketner, K.B., and McGreevy, L.J., 1959, Stratigraphy of the area between Hernando and Hardee counties, Florida: U.S. Geological Survey, Bulletin 1074-C, p. 49. Klein, H.B., Schroeder, M.C., and Lichtler, W.F., 1964, Geology a nd ground water resources of Glades and Hendry Counties: Fl orida Geological Survey, Report of Investigations 37, 132 p. Kozuch, M., 1994, Age, isotopic, and geochemical characterization of the Carolina slate and Charlotte belts: Implications for stratigraphy and petrogenesi s [unpublished M.S. thesis]: University of Florida, Gainesville, Florida, 115 p.
294 Kristf, E., Juhsz, A.Z., and Vassanyi, I., 1993, The effect of mechanical treatment on the crystal structure and thermal be havior of kaolinite, v. 41, p. 608. Kussel, C.M., and Jones, D.S., 1986, Depositional history of three Pleistocene bluffs in northeastern Florida: Flor ida Scientist, v. 49, p. 242. Ladd G.E., 1898, A preliminary report on a part of the clays of Georgi a: Georgia Geological Survey, Bulletin 6A, 204 p. Land Boundary Information System [doqq_04_utm.cf m], 2004, Tallahassee, Florida: Florida Department of Environmental Protection. Available FTP: http://data.labins.org/200 3/MappingD ata/DOQQ/doqq_04_utm.cfm [October 1, 2007]. Lev, S.M., McLennan, S.M., Meyers, W.J., a nd Hanson, G.N., 1998, A petrographic approach for evaluating trace-element mobility in black shale: Journal of Sedimentary Research, v. 68, p. 970. Lichte, F.E., Golightly, D.W., and Lamothe, P.J., 1987, Inductively coupled plasma-atomic emission spectrometry, in Baedecker, P.A., ed., Methods for Geochemical Analysis, U.S. Geological Survey, Report B1770, Reston, Virginia, p. B1B10. Litard, O., 1977, Contribution l tude des propits phisicochim iques, cristallographiques et morphologiques des kaolins [unpublished P h.D. thesis]: Nancy, France, 345 p. Linn, A.M., DePaolo, D.J., and Ingersoll, R.V., 1991, Nd-Sr isotopic provenance analysis of Upper Cretaceous Great Valley fore -arc sandstones: Geology, v. 19, p. 803. Lowe, R.A., 1991, Microtextural an d mineralogical differences of Georgias kaolins and the search for Ostwald ripening [unpublished M.S. thesis]: University of Georgia, Athens, Georgia, 81 p. Luais, B., Telouk, P., and Albarede, F., 1997, Precise and accurate neodymium isotopic measurements by plasma-source mass spectrometry: Geochimica et Cosmochimica Acta, v. 61, p. 4847. Lugmair, G.W., and Carlson, R.W., 1978, Th e Sm-Nd history of KREEP: Proceedings 9th Lunar and Planetary Science Conference, p. 689. Madsen, I.C., and Hill, R.J., 1988, Effect of dive rgence and receiving slit dimensions on peak profile parameters in Rietveld analysis of X-ray diffractometer data : Journal of Applied Crystallography, v. 21, p. 398. Maliva, R.G., Dickson, J.A.D., and Fallick, A.E ., 1999, Kaolin cements in limestones: potential indicators or organic-rich por e waters during diagenesis: Journal of Sedimentary Research, v. 69, p. 158. Mansfield, W.C., 1939, Notes on the upper Tertiary and Pleistocene mollusks of peninsular Florida: Florida Geological Survey, Bulletin 18, 75 p.
295 Martins, J.H.C., 1928, Sand and gravel deposits of Florida. Florida Geological Survey, 19th Annual Report, p. 33. Matmon, A., Bierman, P.R., Larsen, J., Southw orth, S., Pavich, M.J ., and Caffe, M.W., 2003, Temporally and spatially uniform rates of erosion in the southern Appalachian Great Smoky Mountains: Geology, v. 31, p. 155. Matson, G.C., and Clapp, F.G., 1909, A preliminary re port on the geology of Florida with special reference to the stratigraphy: Florida Geological Survey, Annual Report 2, p. 25. Matson, G.C., 1916, The Pliocene C itronelle formation of the Gulf Coastal Plain: U.S. Geological Survey, Prof essional Paper 98, p. 167. McDaniel, D.K., Hemming, S. R., McLennan, S.M., and Hanson, G.N., 1994, Resetting of neodymium isotopes and redistri bution of REEs during sedime ntary processes: the Early Proterozoic Chelmsford Formation, Sudbury Basin, Ontario, Canada: Geochimica et Cosmochimica Acta, v. 58, p. 931. McDougall, K., 1994, Late Cenozoic benthic foramini fers of HLA Borehole Series, Beaufort Sea Shelf, Alaska: U.S. Geological Survey, Bulletin 2055, 100 p. McClellan, G.H., and Van Kauwenbergh, S.J., 1990, Clay mineralogy of the phosphorites of the southeastern United States, in Burnett, W.C., and Riggs, S.R., eds., Phosphate Deposits of the World, v. 3: Neogene to Modern phosphorites, p. 337. McLennan S.M., 1989, Rare earth elements in sedi mentary rocks: influence of provenance and sedimentary processes, in Lipin, B.R., and McKay, G.A., eds., Geochemistry and Mineralogy of Rare Earth Elements: Minera logical Society of America Reviews in Mineralogy, v.21, p. 169. McLennan, S.M., Taylor, S.R., McCulloch, M. T., and Maynard, J.B., 1990, Geochemical and Nd-Sr isotopic composition of deep-sea turbid ites: crustal evoluti on and plate tectonic associations: Geochimica et Cosmochimica Acta, v. 54, p. 2015. McLennan, S.M., Hemming, S., McDaniel, D.K., and Hanson, G.N., 1993, Geochemical approaches to sedimentation, provenance, and tectonics, in Johnsson, M.J., and Basu, A., eds., Processes Controlling the Composition of Clastic Sediments: Geological Society of America, Special Paper 284, p. 21. McNeill, D.F., Eberli, G.P., Lidz, B.H., Swart, P.K., and Kenter, J.A.M., 2001, Chronostratigraphy of prograding carbonate platform margins: a record of dynamic slope sedimentation, western Great Bahama Bank, in Ginsburg, R.N., ed., Subsurface Geology of a Prograding Carbonate Platform Margin, Great Bahama Bank: Results of the Bahamas Drilling Project: SEPM, Speci al Publication 70, p. 101. McSween, H.Y., Jr., Speer, A., and Fu llagar, P.D., 1991, Plutonic rocks, in Horton, J.W., Jr., and Zullo, V.A., eds., Geology of the Caro linas: Carolina Geolog ical Society, p. 109.
296 Merino, E., Harvey, C., and Murray, H.H., 1989, Aqueous-chemical control of the tetrahedralaluminum content of quartz, halloysite, and other low-temperature silicates: Clays and Clay Minerals, v. 37, p. 135. Miller, K.G., Kominz, M.A., Browning, J.V., Wright, J.D., Mountain, G.S., Katz, M.E., Sugarman, P.J., Cramer, B.S., Christie-Blic k, N., and Pekar, S.F., 2005, The Phanerozoic record of global sea-level change: Science, v. 310, p. 1293. Miller, R.G., and O'Nions, R.K., 1984, The provena nce and crustal residence ages of British sediments in relation to pal aeogeographic reconstructions: Earth and Planetary Science Letters, v. 68, p. 459. Miller, R.G., and O'Nions, R.K., 1985, Source of Precambrian chemical and clastic sediments: Nature, v. 314, p. 325. Mills, H.H., 2000, Apparent increasing rates of stream incision in the eastern United States during the late Cenozoic: Geology, v. 28, p. 955. Missimer, T.M., 1992, Stratigraphic relationships of sediment facies within the Tamiami Formation of southwest Florida: pr oposed intraformational correlations, in Scott, T.M., and Allmon, W.D., eds., Plio-Pleis tocene Stratigraphy and Paleontology of Southern Florida: Florida Geological Survey, Sp ecial Publication 36, p. 63. Missimer, T.M., 1993, Pliocene stratigraphy of southern Florida: unresolved issues of facies correlation in time, in Zullo, V.A., Harris, W.B., Scott, T.M., and Portell, R.W., eds., The Neogene of Florida and Adjacent Regions, Proc eedings of the Third Bald Head Island Conference on Coastal Plains Geology: Florid a Geological Survey, Special Publication 37, p. 33. Missimer, T.M., 1997, Late Oligocene to Pliocene evolut ion of the central portion of the South Florida Platform: Mixing of siliciclastic and carbonate sediments [unpublished Ph.D. dissertation]: University of Miami, Co ral Gables, Florida, 2 volumes, 1001 p. Missimer, T.M., 2001a, Late Neogene geology of northwestern Lee County, Florida, in Missimer, T.M., and Scott, T.M., eds., Ge ology and hydrology of Lee County, Florida: Florida Geological Survey, Sp ecial Publication 49, p. 21. Missimer, T.M., 2001b, Late Paleogene and Neogene Chronostratigraphy of Lee County, Florida, in Missimer, T.M., and Scott, T.M., ed s., Geology and hydrology of Lee County, Florida: Florida Geological Surv ey, Special Publication 49, p. 67. Missimer, T.M., and Maliva, R.G., 2006, Late Mio cene to early Pliocene fluvial transport of siliciclastic sediment onto th e southern Florida Platform, in Lock, B.E., Willis, J.J., and Hammes, U., eds., Transactions Gulf Coast Association of Geologi cal Societies, v. 56, p. 605. Molnar, P., 2001, Climate change, flooding in arid environments, and erosion rates: Geology, v. 29, p. 1071.
297 Molnar, P., 2004, Late Cenozoic increase in accu mulation rates of terrestrial sediment: How might climate change have affected erosion rates?: Annual Review of Earth and Planetary Sciences, v. 32, p. 67. Molnar, P., and England, P., 1990, Late Cenozoic uplift of mountain ranges and global climatic change: chicken or egg?: Nature, v. 346, p. 29. Montes, C.R., Melfi, A.J., Carvalho, A., Vi eira-Coelho, A.C., and Formoso, M.L.L., 2002, Genesis, mineralogy and geochemistry of kaolin deposits of the Jari River, Amap State, Brazil: Clays and Clay Minerals, v. 50, p. 494. Moore, D.M., and Reynolds, R.C., Jr., 1997, X-ray diffraction and the identification and analysis of clay minerals, 2nd Ed., Oxford University Press, New York, 378 p. Morse, J.W., and Casey, W.H., 1988, Ostwald pro cesses and mineral paragenesis in sediments: American Journal of Science, v. 288, p. 537. Mullins, H.T., Neumann, A.C., Wilber, R.J., Hine, A.C., and Chinburg, S.J., 1980, Carbonate sediment drifts in Northern Straits of Fl orida: American Association of Petroleum Geologists, Bulletin, v. 64, p. 1701. Murray H.H., 1976, The Georgia sedimentary kaolins, in Shimoda, S., ed., The 7th Symposium on Kaolin: Tokyo, p. 114. Murray, H.H., and Keller, W.D., 1993, Kaolins, kaolins, and kaolins, in Murray, H.H., et al., eds., Kaolin Genesis and Utilization, Special Publication No. 1: The Clay Minerals Society, Boulder, Colorado, p. 1. Murray, H.H., and Lyons, S.C., 1956, Correlation of paper-coating qual ity with degree of crystal perfection of kaolinite, in Swineford, A., ed., Clays and Clay Minerals, Proceedings of the 4th National Conference, p. 31. Nagy, K.L., and Pevear, D.R., 1993, Kinetics of mu scovite dissolution and kaolinite precipitation onto muscovite at 80C and pH 3 (abstract): Annual Meeting of the Geological Society of America, Programs with Abstract, v. 25, p. A-437. Naish, T.R., and Wilson, G.S., 2009, Constraints on the amplitude of Mid-Pliocene (3.6-2.4 Ma) eustatic sea-level fluctuations from the Ne w Zealand shallow-marine sediment record: Philosophical Transaction of the Royal Society A: Mathematical, Physical and Engineering Sciences, v. 367, p. 169. Nakamura, N., 1974, Determination of REE, Ba, Fe, Mg, Na, and K in carbonaceous and ordinary chondrites: Geochimica et Cosmochimica Acta, v. 38, p. 757. Nelson, B.K., and DePaolo, D.J., 1988, Comparison of isotopic and petrographic provenance indicators in sediments from Tertiary con tinental basins of New Mexico: Journal of Sedimentary Petrology, v. 58, p. 348.
298 Nesbitt, H.W., 1979, Mobility and fractionation of rare earth elements during weathering of a granodiorite: Nature, v. 279, p. 206. Nesbitt, H.W., and Markovics, G., 1997, Weathering of granodioritic crust, long-term storage of elements in weathering profiles, and petrogenesi s of siliciclastic sediments: Geochimica et Cosmochimica Acta, v. 61, p. 1653. Nummedal, D., Oertel, G.F., Hubba rd, D.K., and Hine, A.C., 1977, Tidal inlet variability Cape Hatteras to Cape Canaveral: Proceedings, Coastal Sediments ASCE, Charleston, South Carolina, p. 543. Nystrom P.G.Jr., Willoughby R.H., and Kite L.E ., 1986, Cretaceous-Tertiary stratigraphy of the upper edge of the coastal plain between No rth Augusta and Lexington, South Carolina: South Carolina Geological Survey, Field Trip Guidebook, October 11-12, Columbia, South Carolina, 82 p. Ohr, M., Halliday, A.N., and Peacor D.R., 1991, Sr and Nd isotopic evidence for punctuated clay diagenesis, Texas Gulf Coast: Earth a nd Planetary Science Letters, v. 105, p. 110. Okada, H., and Buhry, D., 1980, Supplemental modifi cation and introduction of code numbers to the low-latitude coccolith biostratigraphi c zonation: Marine Mi cropaleontology, v. 5, p. 321. Olssson, A.A., and Petit, R.E., 1964, Some Neogene Mollusca from Florida and the Carolinas: Bulletin of American Paleontology, v. 47, p. 509. ONions, R.K., Hamilton, P.J., and Ev enson, N.M., 1977, Variations in the 143Nd/144Nd and 87Sr/86Sr ratios in oceanic basalts: Earth and Planetary Science Letters, v. 34, p. 13. Opdyke, N.D., Spangler, D.A., Smith, D.L., Jones, D.S., and Lindquist, R.C., 1984, Origin of the epeirogenic uplift of Plio-Pleistocene beach ri dges in Florida and development of Florida karst: Geology, v. 12, p. 226. Otvos, E.G., 1998a, Citronelle Form ation, northeast Gulf Coastal Plain; stratigraphic and age issues (abstract): American Association of Petroleum Geologists, Bulletin, v. 82, p. 1789. Otvos, E.G., 1998b, Citronelle Formati on, northeastern Gulf Coastal Plain: Pliocene stratigraphic framework and age issues: Gulf Coast Associa tion of Geological Societies Transactions, v. 48, p. 321. Patterson S.H., and Murray H.H., 1984, Kaolin, refract ory clay, ball clay a nd halloysite in North America, Hawaii, and the Caribbean region. U.S. Geological Survey, Professional Paper 1306, 56 p. Peacock, R., 1983, The post-Eocene stratigraphy of southern Collier County, Florida: South Florida Water Management District Technical Publication 83-5, 42 p.
299 Peck, D.M., Missimer, T.M., Slater, D.H., Wise S.W., and ODonnell, T.H., 1979, Late Miocene glacio-eustatic lowering of sea level: evid ence from the Tamiami Formation of South Florida: Geology, v. 7, p. 285. Pickering S.M.Jr., and Hurst V.J., 1989, Comm ercial kaolins in Georgia occurrence, mineralogy, origin, use, in Fritz, W.J., ed., Excursions in Georgia Geology: Georgia Geological Society, Guidebook 9-1, p. 29. Pickering, S.M., Jr., Hurst, V.J., and Elzea, J.M., 1997, Mineralogy stratigraphy, and origin of Georgia kaolins: Field Excursion Guidebook, 11th International Clay Conference, Ottawa, Canada, 69 p. Piper, D.Z., 1974, Rare earth elements in the se dimentary cycle: a summary: Chemical Geology, v. 14, p. 285. Pirkle E.C., 1960, Kaolinitic sediments in peninsul ar Florida and origin of the kaolin. Economic Geology, v. 55, p. 1382-1405. Pirkle, E.C., Pirkle, W.A., and Yoho, W.H., 1974, The Green Cove Springs and Boulougne heavy-mineral sand deposits of Florida: Economic Geology, v. 69, p. 1129. Pirkle E.C., and Yoho W.H., 1961, Folding or warp ing resulting from solution with associated joints and organic zones in cl ayey sands at Edgar, Florida: Florida Academy of Science, Quarterly Journal, v. 24, p. 247. Pirkle E.C., Yoho W.H., and Allen A.T., 1964, Orig in of the silica sand deposits of the Lake Wales Ridge area of Florida: Economic Geology, v. 59, p. 1107. Pirkle, E.C., Yoho, W.H., and Hendr y, C.W., 1970, Ancient sea level stands in Florida: Florida Geological Survey, Bulletin 52, 61 p. Pirkle, F.L., and Czel, L.J., 1983, Marine fossils from region of Trail Ridge, a Georgia-Florida landform: Southeastern Geology, v. 24, p. 31. Planon A., and Zacharie, C., 1990, An expert sy stem for the structural characterization of kaolinites: Clays and Clay Minerals, v. 25, p. 249. Poag, C.W., and Sevon, W.D., 1989, A record of Appalachian denudation in postrift Mesozoic and Cenozoic sedimentary deposits of the U.S. middle Atlantic continental margin: Geomorphology, v. 2, p. 119. Puri H.S., and Vernon R.O., 1964, Summary of the ge ology of Florida and a guide to the classic exposures: Florida Geological Su rvey, Special Publication 5, 312 p. Rasmussen, B., 1996, Early-diagenetic REE-phosphate minerals (florencite, gorceixite, crandallite, and xenotime) in marine sandstones: a major sink for oceanic phosphorous: American Journal of Science, v. 296, p. 601.
300 Rasmussen, B., Buick, R., and Taylor, W.R., 1998, Removal of oceanic REE by authigenic precipitation of phosphatic minerals: Earth and Planetary Science Letters, v. 164, p. 135 149. Ravelo, A.C., and Wara, M.W., 2004, The role of the tropical oceans on global climate during a warm period and a major climate transition: Oceanography v. 17, p. 32 Rea, D.K., Zachos, J.C., Owen, R.M., and Gingerich, P.D., 1990, Global change at the Paleocene-Eocene boundary: Climatic and evolut ionary consequences of tectonic events. Palaeogeography Palaeoclimat ology Palaeoecology, v. 79, p. 117. Reynolds, W.R., 1962, The lithostratigraphy and clay mineralogy of the Tampa-Hawthorn sequence of peninsular Florida [unpublished M.S. thesis]: Florida State University, Tallahassee, Florida, 126 p. Robertson, I.D.M., and Eggleston, R.A., 1991, Weathe ring of granites muscov ite to kaolinite and halloysite and of plagioclasederived kaolinite to halloysite: Clays and Clay Minerals, v. 39, p. 113. Ruddiman W., Sarnthein, M., et al., 1989, Late Mio cene to Pleistocene evol ution of climate in Africa and the low-latitude Atlantic: overview of Leg 108 results: Proceedings ODP Scientific Results, v. 108, p. 463. Rupert, F.R., 1990, Geology of Gadsden County, Flor ida: Florida Geological Survey, Bulletin 62, 61 p. Samson, S.D., Coler, D.G., and Speer, J.A., 1995a, Geochemical and Nd-Sr-Pb isotopic composition of Alleghanian granite s of the southern Appalachia ns: Origin, tectonic setting, and source characterization: Earth and Planetary Sc ience Letters, v. 134, p. 359. Samson, S.D., Hibbard, J.P., and Wortman, G.L., 1995b, Nd isotopic evidence for juvenile crust in the Carolina terrane, southern Appa lachians: Contributions to Mineralogy and Petrology, v. 121, p. 171. Scher, H.D., and Martin, E.E., 2004, Circulation in the Southern Ocean during the Paleogene inferred from Nd isotopes: Earth and Planetary Science Letters, v. 228, p. 391. Schmidt, W., 1997, Geomorphology and physiography of Florida, in Randazzo, A.F., and Jones, D.S., eds., The Geology of Florida: Universi ty Press of Florida, Gainesville, p. 1. Scott, T.M., 1980, The sand and gravel resources of Florida: Florida Geological Survey, Report of Investigations 90, 41 p. Scott, T.M., 1988a, The Cypresshead Forma tion in northern peninsular Florida, in Pirkle, F.L., and Reynolds, J.G., eds., Southeastern Geol ogical Society, Annual Field Trip Guidebook, Feb. 19-29, p. 70.
301 Scott, T.M., 1988b, The lithostratigraphy of the Ha wthorn Group (Miocene) of Florida: Florida Geological Survey, Bulletin 59, 148 p. Scott, T.M., 1992a, A geological overview of Fl orida. Florida Geological Survey, Open-File Report 50, 78 p. Scott, T.M., 1992b, Coastal Plains stratigra phy: The dichotomy of biostratigraphy and lithostratigraphy A ph ilosophical approach to an old problem, in Scott, T.M., and Allmon, W.D., eds., Plio-Pleisto cene stratigraphy and paleontology of southern Florida: Florida Geological Survey, Sp ecial Publication 36, p. 21. Scott, T.M., 1997, Miocene to Holocene history of Florida, in Randazzo, A.F., and Jones, D.S., eds., The Geology of Florida: University Press of Florida, Gainesville, p. 57. Scott, T.M., 2001, Text to accompany the geologic map of Florida: Florida Geological Survey, Open-File Report 80, 29 p. Scott, T.M., Campbell, K.M., Rupert, F.R., Arthur, J.D., Missimer, T.M., Lloyd, J.M., Yon, J.W., and Duncan, J.G., 2001, Geologic map of the state of Florida: Florida Geological Survey, Map Series 146. Scott, T.M., Hoestine, R.W., Knapp, M.S., Lane E., Ogden, G.M., Jr., Deuerling, R., and Neel, H.E., 1980, The sand and gravel resources of Fl orida. Florida Geological Survey, Report of Investigation 90, 41 p. Scott, T.M., and Wingard, G.L., 1995, Facies, fossils, and time A discussion of the lithoand biostratigraphic problems in the Plio-Pleistocene sediments in southern Florida, in Scott, T.M., ed., Stratigraphy and paleontology of th e Plio-Pleistocene shell beds, southwest Florida: Southeastern Geological Society, Guidebook 35. Seilacher, A., 1967, Bathymetry of trace fossils. Marine Geology, v. 5, p. 413. Sellards E.H., 1912, The soils and ot her surface residual materials of Florida. Florida Geological Survey, 4th Annual Report, p. 1. Sellards, E.H., 1914, The relation between the D unnellon formation and the Alachua clays of Florida: Florida Geological Survey, 6th Annual Report, p. 161. Sellards, E.H., 1918, Geology between the Apalach icola and Ochlocknee Rivers in Florida: Florida Geological Survey, 10th Annual Report, p. 9. Sen, A., Kendall, C.G.St.C., and Levine, P., 1999, Co mbining a computer simulation and eustatic events to date seismic sequence boundaries: a case study of the Neogene of the Bahamas: Sedimentary Geology, v. 125, p. 47. Simi V., and Uhlk, P., 2006, Crystallite size distri bution of clay minerals from selected Serbian clay deposits: Annales Gologiques de la Pninsule Balkanique, v. 67, p. 109.
302 Singh, B., and Gilkes, R.J., 1992, an electron optical in vestigation of the alte ration of kaolinite to halloysite: Clays and Clay Minerals, v. 40, p. 212. Small, J.S., and Manning, D.A.C., 1994, On-line monitoring of clay mineral precipitation in sandstone pore space under flow conditi ons: Mineralogical Magazine, v. 58A, p. 852. Smith R.W., 1929, Sedimentary kaolins of the co astal plain of Georgi a. Georgia Geological Survey, Bulletin 44, 48 p. Sommerfield, C.K., and Wheatcr oft, R.A., 2007, Late Holocene sediment accumulation on the northern California shelf: Oceanic, fluvial and anthropogenic influences: Geological Society of America, Bulletin, v. 119, p. 1120. Sonke, J.E., and Salters, V.J.M., 2006, Lantha nide-humic substances complexation. I. Experimental evidence for a lanthanide contraction effect: Geochimica et Cosmochimica Acta, v. 70, p. 1495. Stein, R., 1985, Rapid grain-size analyses of clay and silt fraction by SediGraph 5000D: comparison with Coulter Counter and Atte rberg methods: Journal of Sedimentary Petrology, v. 55, p. 590. Stein, R., and Sarnthein, M., 1984, Late Neogene ev ents of atmospheric and oceanic circulation offshore northwest Africa: high resoluti on record from deep-sea sediments, in Coetzee, J.A. and van Zinderen Bakker, E.M., eds., Palaeoecology of Africa, v. 16: Balkema, Rotterdam, p. 9. Stewart, K.G., and Dennison, J.M., 2006, Tertiary-t o-Recent arching and the age and origin of fracture-controlled lineaments in the Southe rn Appalachians (abstract): Abstracts with Programs, 55th Annual Meeting, Southeastern Sec tion of the Geological Society of America, v. 38, p. 27. Stuckey, J.L., 1965, North Carolina: Its geology and mineral resour ces: North Carolina Department of Conservation and Development, 550 p. Stull, R.T., and Boles, G.A., 1926, Beneficiation and utilization of Georgia clays. U.S. Bureau of Mines, Bulletin 252, 72 p. Stumm, W., 1992, Chemistry of the Solid-Water Interface: Wiley & Sons, New York, 346 p. Surdam, R.C., Crossey, L.J., and Boese, S.W ., 1984, The chemistry of secondary porosity, in McDonald, D.A., and Surdam, R.C., eds., Clas tic Diagenesis: American Association of Petroleum Geologists, Memoir 37, p. 127. Syvitski, J.P.M., 2004, The influence of climate on the sediment load of rivers (abstract): Annual Meeting Expanded Abstracts, American Association of Petroleum Geologists Annual Meeting, v. 13, p. 135.
303 Tabbutt, K.D., 1990, Provenance of some Mesozoic terrigenous clastic st rata in the western interior of North America: a mineralogical and geochemical approach [unpublished Ph.D. dissertation], Dartmouth College, 162 p. Tanaka, T., Togashi, S., Kamioka, H., Amakawa, H., Kagami, H., Hamamoto, T., Yuhara, M., Orihashi, Y., Yoneda, S., Shimizu, H., Kunimaru, T., Takahashi, K., Yanagi, T., Nakano, T., Fujimaki, H., Shinjo, R., Asahara, Y., Tanimizu, M., and Dragusanu, C., 2000, JNdi-1: a neodymium isotopic reference in consiste ncy with LaJolla neodymium: Chemical Geology, v. 168, p. 279. Taylor S.R., and McLennan S.M., 1985, The Con tinental Crust: Its Composition and Evolution: Blackwell Scientific Pub lications, Oxford, 312 p. Tedford, T.H., 1985, Late Miocene turnover of the Australian mammal fauna: South Africa Journal of Science, v. 81, p. 262. Toulkeridis, T., Goldstein, S.L., Clauer, N., Kr ner, A., and Lowe D.R., 1994, Sm-Nd dating of Fig Tree clay minerals of the Barberton greenstone belt, South Africa: Geology, v. 22, p. 199. van Zinderen Bakker, E.M., and Mercer, J.H., 19 86, Major late Cenozoic climatic events and palaeoenvironmental changes in Africa viewed in a world wide context: Palaeogeography, Palaeoclimatology, Palaeoecology, v. 56, p. 217. Vasconcelos, P.M., Becker, T.A., Renne, P.R., and Brimhall, G.H., 1994a, Direct dating of weathering phenomena by 40Ar/39Ar and K-Ar analysis of supergene K-Mn oxides. Geochimica et Cosmochimica Acta, v. 58, p. 1635. Vasconcelos, P.M., Brimhall, G.H., Becker, T.A., and Renne, P.R., 1994b, 40Ar/39Ar analysis of supergene jarosite and alunite: implications to the paleoweathering hi story of the western USA and West Africa: Geochimica et Cosmochimica Acta, v. 58, p. 401. Veatch O., 1909, Second report on th e clay deposits of Georgia. Georgia Geological Survey, Bulletin 18, p. 92. Veatch, O., and Stephenson, L.W., 1911, Preliminar y report on the geology of the Coastal Plain of Georgia: Georgia Geologi cal Survey, Bulletin 26, 466 p. Velho, J.A., and Gomes, C., 1991, Characterization of Portuguese kaolins for the paper industry: beneficiation through new delamination tech niques: Applied Clay Science, v. 6, p. 155 170. Wade, J.A., 2002, Element mobility and secondary mi neral formation during the early stages of alteration in rocks from the Tecuamburro volcanic complex, southeast Guatemala: Ph.D. Dissertation, Michigan State University. Walton, T.L., 1976, Littoral drift es timates along the coastline of Florida: Florida Sea Grant Program, Report 13, 41 p.
304 Wan, Y., and Liu, C., 2006, The effect of humic acid on the adsorption of REEs on kaolin: Colloids and Surfaces A: Physicochemical and Engineering Aspects, v. 290, p. 112. Wang, H., 1994, Step size, scanning speed and shape of X-ray diffraction peak: Journal of Applied Crystallography, v. 27, p. 716. Warme, J.E., 1971, Paleoecological aspects of a m odern coastal lagoon: Univ ersity of California Publication in Geological Sciences, v. 87, p. 1. Warren, B.E., and Averbach, B.L., 1953, The effect of cold-work distortion on X-ray patterns: Journal of Applied Physics, v. 21, p. 595. Warzeski, E.R., Cunningham, K.J., Ginsburg, R.N., Anderson, J.B., and Ding, Z.-D., 1996, A Neogene mixed siliciclastic a nd carbonate foundation for the Qu aternary carbonate shelf, Florida Keys: Journal of Se dimentary Research, v. 66, p. 788. Webb, S.D., and Crissinger, D.B., 1983, Stratigraphy and vertebrate paleontology of the central and southern Phosphate District of Florida, in Central Florida P hosphate District: Geological Society of America, Southeast Section Field Trip Guidebook, p. 28. Wedepohl, K.H., 1995, The composition of the con tinental crust: Geochimica et Cosmochimica Acta, v. 59, p. 1217. Weedman, S.D., Scott, T.M., Edwards, L.E., Br ewster-Wingard, G.L., and Libarkin, J.C., 1995, Preliminary analysis of integrated stratigra phic data from the Phred #1 corehole, Indian River County, Florida: U.S. Geological Survey, Open-File Report 95-824, 63 p. Weimer, R.J., and Hoyt, J.H., 1964, Burrows of Callianassa major Say, geological indicators of littoral and shallow neritic environm ents. Journal of Paleontology, v. 38, p. 761. White, W.A., 1958, Some geomorphic features of cen tral peninsular Florid a: Florida Geological Survey, Bulletin 41, 92 p. White, W.A., 1970, The geomorphology of the Florid a peninsula: Florid a Geological Survey, Bulletin 51, 164 p. Wildeman, T.R., and Condie, K.C., 1973, Rare ea rths in Archean graywackes from Wyoming and from the Fig Tree Group, South Africa: Ge ochimica Cosmochimica et Acta, v. 37, p. 439. Willard, D.A., 1994, Palynological record from the North Atlantic region at 3 Ma: vegetational distribution during a period of global warmth: Review of Palaeobotany and Palynology, v. 83, p. 275. Willard, D.A., Cronin, T.M., Ishman, S.E., and Litwin, R.J., 1993, Terrestrial and marine records of climatic and environmental changes dur ing the Pliocene in s ubtropical Florida: Geology, v. 21, p. 679.
305 Wingard, G.L., Weedman, S.D., Scott, T.M., Edwards, L.E., and Green, R.C., 1994, Preliminary analysis of integrated stratigraphic data from the South Venice corehole, Sarasota County, Florida: U.S. Geological Survey, Open-File Report 95-3, 129 p. Winker, C.D., and Howard, J.D., 1977a, Correlation of tectonically deformed shorelines on the southern Atlantic coasta l plain: Geology, v. 5, p. 123. Winker, C.D., and Howard, J.D., 1977b, Plio-Pleistocene paleogeography of the Florida Gulf Coast interpreted from relict shorelines: Gulf Coast Association of Geological Societies, Transactions, v. 27, p. 409. Wolfe, J.A., 1985, Distribution of major vegetation types during the Tertiary, in The Carbon Cycle and Atmospheric CO2: Natural Variations Archaen to Present, American Geophysical Union, p. 357. Wornardt, W.W., Jr., Shaffer, B ., and Vail, P.R., 2001, Revision of the Late Mio cene, Pliocene and Pleistocene sequence cycles (abstract): Am erican Association of Petroleum Geologists, Bulletin, v. 85, p. 1710. Yemane, K., Bonnefille, R., and Faure, H., 1985, Palaeoclimatic and tectonic implications of Neogene microflora from the northwester n Ethiopian highlands: Nature, v. 318, p. 653 656. Yomane, K., Bonnefille, R., and Faure, H., 1985, Paleoclimatic and tectonic implications of Neogene microflora from the northwester n Ethiopian Highlands: Nature, v. 318, p. 653 656. Yon, J.W., Jr., 1966, Geology of Jefferson County, Fl orida: Florida Geological Survey, Bulletin 48, 115 p. Zhang, P., Molnar, P., and Downs, W.R., 2001, Incr eased sedimentation ra tes and grain sizes 2 4 Myr ago due to the influence of climate change on erosion rates: Nature, v. 410, p. 891 897. Ziegler, Z., Hsieh, J.C.C., Chadwick, O.A., Kelly E.F., Hendricks, D.M., and Savin, S.M., 2003, Halloysite as a kinetically cont rolled end product of arid-zone basalt weathering: Chemical Geology, v. 202, p. 461.
306 BIOGRAPHICAL SKETCH Kendall Fountain was born in Winter Haven, Florida, where he was exposed to the geology of the Central Florida Phosphate District at an early age via a geologist father. He earned a Bachelor of Science degree with honors at the University of Florida in May 1989, majoring in geology. He then began Graduate School at the University of Florida and received a Master of Science degree in April 1994, w ith a major in geology. Kendall subsequently continued his education at the University of Flor ida in order to receive a Ph.D. in geology, with a minor in environmental engineering, focusing on l ong held questions related to the origin and significance of Cypresshead Formation sediments in Florida and Georgia. Presently employed by Plum Creek Timber Company as their Seni or Manager Mineral Resources, Kendalls responsibilities include identification and development of mi neral resources associated with Plum Creeks land base across the United States, and coordination of group activities with Plum Creeks timber resource, real estate, and business development divisions. Prior to joining Plum Creek, and while attending the Un iversity of Florida, Kendall started a sole proprietorship consulting firm, Fountain Geological, which provid ed consulting expertise in the identification and development of industrial mineral prospect s and addressed mineral resource recovery and quality issues through the application of advanced analytical techniques.