Evolution of karst in the Lower Suwannee River Basin, Florida

MISSING IMAGE

Material Information

Title:
Evolution of karst in the Lower Suwannee River Basin, Florida
Physical Description:
xiii, 214 leaves : ill. ; 29 cm.
Language:
English
Creator:
Denizman, Can
Publication Date:

Subjects

Subjects / Keywords:
Karst -- Suwannee River Valley (Ga. and Fla.)   ( lcsh )
Hydrogeology -- Suwannee River Valley (Ga. and Fla.)   ( lcsh )
Dissertations, Academic -- Geology -- UF   ( lcsh )
Geology thesis, Ph. D   ( lcsh )
Genre:
bibliography   ( marcgt )
non-fiction   ( marcgt )

Notes

Thesis:
Thesis (Ph. D.)--University of Florida, 1998.
Bibliography:
Includes bibliographical references (leaves 205-213).
General Note:
Typescript.
General Note:
Vita.
Statement of Responsibility:
by Can Denizman.

Record Information

Source Institution:
University of Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
aleph - 029225151
oclc - 41186956
System ID:
AA00022869:00001

Table of Contents
    Title Page
        Page i
    Dedication
        Page ii
    Acknowledgement
        Page iii
    Table of Contents
        Page iv
        Page v
        Page vi
    List of Tables
        Page vii
        Page viii
    List of Figures
        Page ix
        Page x
        Page xi
    Abstract
        Page xii
        Page xiii
    Chapter 1. Introduction
        Page 1
        Page 2
        Page 3
        Page 4
        Page 5
        Page 6
        Page 7
        Page 8
        Page 9
        Page 10
        Page 11
        Page 12
        Page 13
        Page 14
        Page 15
        Page 16
        Page 17
        Page 18
        Page 19
        Page 20
    Chapter 2. Morphometric and spatial distribution parameters of karstic depressions
        Page 21
        Page 22
        Page 23
        Page 24
        Page 25
        Page 26
        Page 27
        Page 28
        Page 29
        Page 30
        Page 31
        Page 32
        Page 33
        Page 34
        Page 35
        Page 36
        Page 37
        Page 38
        Page 39
        Page 40
        Page 41
        Page 42
        Page 43
        Page 44
        Page 45
    Chapter 3. Factors affecting depression morphometry and distribution
        Page 46
        Page 47
        Page 48
        Page 49
        Page 50
        Page 51
        Page 52
        Page 53
        Page 54
        Page 55
        Page 56
        Page 57
        Page 58
        Page 59
        Page 60
        Page 61
        Page 62
        Page 63
        Page 64
        Page 65
        Page 66
        Page 67
        Page 68
        Page 69
        Page 70
        Page 71
        Page 72
        Page 73
        Page 74
        Page 75
        Page 76
        Page 77
        Page 78
        Page 79
        Page 80
        Page 81
        Page 82
        Page 83
        Page 84
        Page 85
        Page 86
        Page 87
        Page 88
        Page 89
        Page 90
        Page 91
        Page 92
        Page 93
        Page 94
        Page 95
        Page 96
    Chapter 4. Karstic erosion rate calculations
        Page 97
        Page 98
        Page 99
        Page 100
        Page 101
        Page 102
        Page 103
        Page 104
        Page 105
        Page 106
        Page 107
        Page 108
        Page 109
        Page 110
        Page 111
        Page 112
        Page 113
        Page 114
    Chapter 5. Evolution of karst in the Suwannee River area
        Page 115
        Page 116
        Page 117
        Page 118
        Page 119
        Page 120
        Page 121
        Page 122
        Page 123
        Page 124
        Page 125
        Page 126
        Page 127
        Page 128
        Page 129
        Page 130
        Page 131
        Page 132
        Page 133
        Page 134
        Page 135
        Page 136
        Page 137
        Page 138
        Page 139
        Page 140
        Page 141
        Page 142
        Page 143
        Page 144
        Page 145
        Page 146
        Page 147
        Page 148
        Page 149
        Page 150
        Page 151
        Page 152
        Page 153
        Page 154
        Page 155
        Page 156
        Page 157
    Chapter 6. Summary and conclusions
        Page 158
        Page 159
        Page 160
        Page 161
        Page 162
    Appendix A. Depression area and length-width relationship for topographic quadrangles
        Page 163
        Page 164
        Page 165
        Page 166
        Page 167
        Page 168
        Page 169
        Page 170
        Page 171
        Page 172
        Page 173
        Page 174
        Page 175
    Appendix B. Depression length and width relationship for topographic quadrangles
        Page 176
        Page 177
        Page 178
        Page 179
        Page 180
        Page 181
        Page 182
        Page 183
        Page 184
        Page 185
        Page 186
        Page 187
        Page 188
    Appendix C. Flow duration curves
        Page 189
        Page 190
        Page 191
    Appendix D. Allometric relationships for elevation ranges and marine terraces
        Page 192
        Page 193
        Page 194
        Page 195
        Page 196
        Page 197
        Page 198
        Page 199
    Appendix E. Length-width ratio and area relationship within elevation ranges
        Page 200
        Page 201
        Page 202
        Page 203
        Page 204
    List of references
        Page 205
        Page 206
        Page 207
        Page 208
        Page 209
        Page 210
        Page 211
        Page 212
        Page 213
    Biographical sketch
        Page 214
        Page 215
        Page 216
Full Text










EVOLUTION OF KARST IN THE LOWER SUWANNEE RIVER BASIN, FLORIDA


By

CAN DENIZMAN













A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
DOCTOR OF PHILOSOPHY

UNIVERSITY OF FLORIDA


1998































To the memory of my brother, Levent Denizman.
Nous sommes du soleil.













ACKNOWLEDGEMENT S


I would like to express my sincere thanks to my supervisor, Dr. Anthony

Randazzo, for his guidance and encouragement during various stages of this study. I am

also grateful to my other committee members for their assistance and input, especially to

Dr. Dan Spangler from whom I learned much of my knowledge on the hydrogeology of

Florida.

Acknowledgements for assistance in data collection are extended to Ron Ceryak

from the Suwannee River Water Management District and Jonathan Arthur from the

Florida Geological Survey.

I am grateful to David Reed from the St. Johns River Water Management District

(SJRWMD) for his generous assistance and creative ideas regarding many GIS operations

performed in this study. Thanks are also due to Palmer Kinser and Marc Minno from the

SJRWMD for their understanding and encouragement.

Additional thanks go to Yasar Yesilcay and Fuat Sener for their assistance in the

application of statistical techniques. I am also grateful to Ron Ozbun from the Geology

Department office of the University of Florida for his effort and time spent in resolving

many complicated issues regarding my appointment as a research assistant.

Very special thanks to my wife, Isik, for her love, support and patience.

Finally, to John McLaughlin and many others for the music.













TABLE OF CONTENTS
page


ACKNOWLEDGEMENTS ........................................... iii

L IS T O F T A B L E S ...................................................... ....... ....................................... vii

L IS T O F F IG U R E S ........................................................................................................ ix

A B S T R A C T ......... ................................................................................................... . x ii

CHAPTERS

1 IN T R O D U C T IO N ........................................................................................................ 1

C lim ate ................................................. ................................................................. . 4
P hysiographic Setting ............................................................................................. . 4
G eologic Structure ................................................................................................. . 7
G eological Setting .................................................... ............................................. . 10
Hydrogeologic Setting ................................................. 12
K arst in F lo rid a ............... .................................................................................... . 15

2 MORPHOMETRIC AND SPATIAL DISTRIBUTION PARAMETERS OF
KARSTIC DEPRESSIONS ................................................................................ 21

P rospectu s ............................................................................................ ...... . . 2 1
Background ........................................................ 22
M eth o d o lo gy .................................................................. ................. ............... . .2 6
R esu lts .. .......... ................................................................................... ......... 29
D epression D ensity ......................................................................................... . 29
P lanim etric Shap e ........................................................................................... ...37
M ean D iam eter ................................... ............................................................ . 40
Depth of Karstic Depressions .......................................... 41
Discussion ..... ................................................ 43

3 FACTORS AFFECTING DEPRESSION MORPHOMETRY AND
D ISTR IB U T IO N ............................................................................. 46

P ro sp e ctu s ................................................................................................................... 4 6
M etho d o lo gy ......... ............................................................................................... . 50








R esults ........................................................... ... .................... . . 52
Lithology of the Vadose Zone ............................................................................ 52
Thickness of the Overburden M aterial ............................................................... 54
Floridan Aquifer Potentiometric Level Fluctuation ............................................. 60
S o il T y p e ............................................................................................................... 6 4
Depth to W ater Table ........................................................................................ 68
Photolineaments and Orientation of Depressions ............................. 73
T opographic Slope ..................................... ................................................... .. .76
R ep o rted S ink h o les ............................................................................................... 7 8
D iscu ssio n ............................................. ............................................................ . 8 8

4 KARSTIC EROSION RATE CALCULATIONS .................................................... 97

P ro sp e c tu s 9 7............................................................................................................... 9 7
B a c k g ro u n d ................................................................................................................. 9 8
M eth o d o lo g y ................................................................................................... ........ 10 0
R e su lts ....................................................................................................................... 1 0 6
D iscu ssio n ............................................................................................................. .. 1 1 2

5 EVOLUTION OF KARST IN THE SUWANNEE RIVER AREA...................... 115

P ro sp e c tu s ................................................................................................................. 1 1 5
B a c k g ro u n d .............................................................................................................. 1 1 6
M ethodology ... ............................................... 118
R e su lts ....................................................................................................................... 1 2 0
M orphometric Evolution of Karstic Depressions ................................................. 120
D epression area ....................................................................................... . 123
M ean depression diameter ............................................................................... 126
Planimetric shapes and allometry of depressions .............................................. 128
Evolution of the Spatial Distribution of Depressions .................. ... 136
D iscu ssio n ......................................................................................................... . 14 4
Conceptual M odel of Karst Evolution ....................................................................... 148

6 SUMMARY AND CONCLUSIONS ........................................ 158

APPENDICES

A DEPRESSION AREA AND LENGTH-WIDTH RELATIONSHIP FOR
TOPOGRAPHIC QUADRANGLES ........ ....................... ................................ 164

B DEPRESSION LENGTH AND WIDTH RELATIONSHIPS FOR
TOPOGRAPHI C QUADRANGLES ................. ....................... 177

C FLOW DURATION CURVES ............................................................................... 190








D ALLOMETRIC RELATIONSHIPS FOR ELEVATION RANGES AND
M A R IN E T E R R A C E S ....................................................... ................................ 193

E LENGTH-WIDTH RATIO AND AREA RELATIONSHIP WITHIN
ELE V A TIO N R A N G E S ................ ................... ..................... ................... 201

L IST O F R E FE R E N C E S ............................................... ........................................... 205

B IO G R A PH IC A L SK E TCH ...................................................................................... 214













LIST OF TABLES


Table

2.1. Morphometric and spatial distribution parameters for topographic quadrangles ....... 30
2.2. Statistical summary of morphometric parameters ................................................ 44
2.3. Depression density and nearest neighbor statistics for various karst areas ........... 45
3.1. Percentages of areas corresponding to depression density ranges for the vadose zone
lith o lo g ie s ......................................................................................................... 5 2
3.2. Percentages of areas corresponding to depression density ranges for the zones of the
overburden thickness ...................................................................................... 57
3.3. Percentages of areas corresponding to depression density ranges for the zones of
Floridan aquifer potentiometric level fluctuations ............................. 60
3.5. Percentages of areas corresponding to depression density ranges for the zones of the
depth to w ater table ........................................................................................ . 7 1
3.6. Summary of the mean morphometric parameters and the clay content within the zones
of the overburden thickness ................................................................................ 93
3.7. Summary of the mean morphometric parameters and the clay contents within the
zones of the depth to water table ................................... 95
4.1. Statistical summary of hydrologic and chemical parameters for the Santa Fe River
basin ........................................................... ............................ . . . 109
4.2. Statistical summary of hydrologic and chemical parameters for the Upper Suwannee
R iv e r b a sin ........................................................................................................... 1 1 0
4.3. Statistical summary of hydrologic and chemical parameters for the Lower Suwannee
R iv er b asin ...................................................................................................... . 111
4.4. Published erosion rates of different geographic locations ....................................... 113
5.1. Statistical summary of depression area (M2) within the ranges of elevation ............ 124
5.2. Statistical summary of the depression area (M) within the marine terraces .......... 125
5.3. Statistical summary of mean depression diameter (in) within the ranges of
e le v a tio n .............................................................................................................. 12 6
5.4. Statistical summary of the mean diameter (in) within the marine terraces .............. 127
5.5. Statistical summary of the length/width ratio within the ranges of elevation .......... 128
5.6. Statistical summary of the circularity index within the ranges of elevation ............. 129
5.6. Statistical summary of the length/width ratio within the marine terraces ............... 130
5.7. Statistical summary of the circularity index within the marine terraces .................. 131
5.8. Allom etric relationships in elevation ranges.......................................................... 132
5.9. Allometric relationships within marine terraces ..................................................... 133
5.10. Statistical summary of depression density within elevation ranges ....................... 137
5.11. Statistical summary of depression density within marine terraces ........................ 138








5.12. Statistical summary of mean nearest neighbor distance (in) within the elevation
ra n g e s .................................................................................................................. 1 3 9
5.13. Statistical summary of nearest neighbor distance (m) within marine terraces ........ 140
5.14. Summary of mean spatial parameters ................................... 141
5.15. Summary of mean morphometric parameters ............................. 148













LIST OF FIGURES


igurepg

1.1. T he stu dy area ............................................................................................. . . 3
1.2 P hysiographic regions ....... ................................................................................. . 5
1.3. Structural features of Florida .............................................................................. 8
1.4. Hydrostratigraphic column for the Suwannee River Basin .............................. 13
1.5. Drawdown doline initiation in the subcutaneous (epikarstic) zone 19
2.1. Names and location of topographic quadrangles within the study area ................. 27
2.2. Histogram of depression density distribution ..................................................... 31
2.3. D istribution of depression density.. ... ........................................................... 32
2.4. Histogram of nearest neighbor index ................................................................... 34
2.5. Distribution of nearest neighbor index within the study area ...................... 35
2.6. Percent frequency of nearest neighbor distance .................................................... 36
2.7. Cumulative percentage of nearest neighbor distance ........................... 36
2.8. Percent frequency of L/W ratio ........................................................................... 38
2.9. Cumulative percentage of L/W ratio ................................. 38
2.10. Percent frequency of circularity index ............................................................... 39
2.11. Cumulative percentage of circularity index ........................................................ 39
2.12. Percent frequency of m ean diam eter ................................................................. 40
2.13. Cumulative percentage of mean diameter ................................. 40
2.14. Cumulative percentage of depression area ................................. 41
2.15. Depth-frequency distribution for the depressions ............................................... 42
3.1. Distribution of the vadose zone lithology within the study area ............................ 53
3.2. Distribution of depression density within the vadose zone lithologies ................... 54
3.3. Distribution of overburden thickness within the study area .................................. 55
3.4. Changes in depression density within the zones of the overburden thickness 56
3.5. Changes in vadose zone lithology with the thickness of the overburden .............. 58
3.6. Changes in mean depression area with the overburden thickness ..................... 58
3.7. Changes in mean depression diameter with the overburden thickness ...................... 59
3.8 Changes in mean length/width ratio with the overburden thickness......................... 59
3.9. Fluctuation of the Floridan aquifer potentiometric level ...................................... 61
3.10. Changes in mean depression density with the potentiometric fluctuation ............. 62
3.11. Changes in mean depression area with potentiometric fluctation .........................63
3.12. Changes in mean depression diameter with potentiometric fluctation ................ 63
3.13. Changes in mean length/width ratio with potentiometric fluctation ................... 64
3.14. Distribution of soil types within the study area ................................................. 65
3.15. Changes in mean depression density with soil types ........................................... 66
3.16. Changes in mean depression area with soil types ............................................... 67








3.17. Changes in mean depression diameter with soil types ......................................... 67
3.18. Changes in mean length/width ratio with soil types ........................................... 68
3.19. Distribution of depth to water table within the study area ...................................... 69
3.20. Changes in mean depression density with depth to water table ............................... 70
3.21. Changes in vadose zone lithology with depth to water table .............................. 70
3.22. Changes in mean depression area with depth to water table ............................... 72
3.23. Changes in mean depression diameter with depth to water table .................. 72
3.24. Changes in mean depression length/width ratio with depth to water table ....... 73
3.25. Map of photolinears and karstic depression density within the study area ....... 74
3.26. Histogram of depression major axis orientations ............................................... 75
3.27. Histogram of nearest neighbor directions .................................. 76
3.28. Map of topographic slope within the study area .............................. 77
3.29. Histogram of topographic slope distribution within the study area ................. 78
3.30. Distribution of reported sinkholes within the study area .................................. 79
3.31. Changes in reported sinkhole frequency with vadose zone lithology ................... 81
3.32. Changes in mean diameter with vadose zone lithology for reported sinkholes ........ 81
3.33. Changes in mean depth with vadose zone lithology for reported sinkholes. 83
3.34. Changes in mean length/width ratio with vadose zone lithology for reported
sin k h o le s ...................................................................................................... .... 8 3
3.35. Changes in reported sinkhole frequency with potentiomeric level fluctuation of the
F loridan aqu ifer ................................................................................................ . 84
3.36. Changes in mean diameter with potentiomeric level fluctuation of the Floridan
aquifer for reported sinkholes ..................................................................................... 84
3.37. Changes in mean length/width ratio with potentiometric level fluctuation of the
Floridan aquifer for reported sinkholes .............................................................. 85
3.38. Changes in mean depth with potentiometric level fluctuation of the Floridan
aquifer for reported sinkholes ..................................................................................... 85
3.39. Changes in reported sinkhole frequency with depth to water table ........................ 86
3.40. Changes in mean diameter with depth to water table for reported sinkholes. 86
3.41. Changes in mean length/width ratio with depth to water table for reported
sin k h o le s ..................................................................... .......... ........ .............................. 8 7
3.42. Changes in mean depth with depth to water table for reported sinkholes................ 87
3.43. Changes in reported sinkhole frequency with the thickness of the overburden
m a te ria l .................................................................................................................. 8 9
3.44. Changes in mean depth with the thickness of the overburden material for reported
sin k h o le s ....................................................................................... ........................ 8 9
3.45. Changes in mean diameter with the thickness of the overburden material for
reported sinkholes .................................................................................................... .90
3.46. Changes in mean length/width ratio with the thickness of the overburden
m aterial for reported sinkholes ......................................................................... 90
3.47. Changes in reported sinkhole frequency with the soil type ................... 91
3.48. Changes in mean length/width ratio with the soil type ....................................... 91
3.49. Changes in mean diameter with the soil type for reported sinkholes ................... 92
3.50. Changes in mean area with the soil type for reported sinkholes ......................... 92









4.1. Location map of flow and water quality stations .............................. 101
4.2. Discharge hardness relationship for the Upper Suwannee River basin ................ 103
4.3. Discharge hardness relationship for the Santa Fe River basin ............................. 103
4.4. Discharge hardness relationship for the Lower Suwannee River basin ................ 104
4.5. Discharge mass flux relationship for the Upper Suwannee River basin............... 104
4.6. Discharge mass flux relationship for the Santa Fe River basin ............................ 105
4.7. Discharge mass flux relationship for the Lower Suwannee River basin ............... 106
4.8. Annual dissolved load carried by different flows in the Upper Suwannee River
b a sin ................ ................ ............................... ............................................... 1 0 8
4.9. Annual dissolved load carried by different flows in the Santa Fe River basin. 108
4.10. Annual dissolved load carried by different flows in the Lower Suwannee River
b a sin ................................. ........................ ...... .............................................. 1 0 9
4.11. Seasonal changes of the flow and mass flux for the Upper Suwannee River basin. 110
4.12. Seasonal changes of the flow and mass flux for the Santa Fe river basin ............... 111
4.13. Seasonal changes of the flow and mass flux for the Lower Suwannee River basin. 111
5.1. Map of elevation .............................................. 121
5.2. M ap of the marine terraces within the study area ................................................. 122
5.3. Changes in mean depression area with elevation ........................ ........................ 124
5.4. Changes in depression area with marine terraces ................................................ 125
5.5. Changes in mean depression diameter with topography ......................................... 127
5.6. Changes in mean depression diameter with marine terraces .................................... 127
5.7. Changes in mean length/width ratio with topography ............................................. 129
5.8. Changes in mean circularity index with topography ............................................... 129
5.9. Changes in mean length/width ratio with marine terraces ....................... 131
5.10. Changes in mean circularity index with elevation ................................................ 131
5.11. Changes in allometric growth rate (a) with elevation ....................................... 134
5.12. Changes in allometric growth rate (a) with marine terraces .................................. 136
5.13. Changes in depression density w ith elevation ....................................................... 138
5.14. Changes in depression density with marine terraces ............................................. 139
5.15. Changes in mean nearest neighbor distance with elevation ................................... 140
5.16, Changes in mean nearest neighbor distance with marine terraces ........................ 141
5.17. Changes in pitting index with distance from the Cody Escarpment ............... 142
5.18. Zones of equal distance from the Cody Escarpment ......................... 143
5.19 (a-f). Evolution of karst in the Suwannee river area, Late Oligocene to recent. 150













Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy

EVOLUTION OF KARST IN THE LOWER SUWANNEE RIVER BASIN, FLORIDA

By

Can Denizman

May 1998


Chairman: Anthony F. Randazzo
Major Department: Geology

This study represents an application of Geographic Information Systems to

examine the morphometric and spatial distribution of karstic depressions and the factors

controlling the karst development in the Lower Suwannee River basin. Because of the

generally unconfined or semiconfined hydrogeologic conditions of the Floridan aquifer, the

Lower Suwannee River basin presents an extraordinary set of karst features illustrating a

complex evolutionary history. Analysis of the morphometric and spatial distribution

parameters of karstic depressions reveals that the Florida karst is represented by broad,

shallow depressions with an average density of 6.07 depressions/km2 and an average

pitting index of 14.5. A simple morphoclimatic classification of karst landforms is

precluded by the great variation of morphometric and spatial ditribution parameters of

karstic depressions within the Lower Suwannee River basin.








Comparison of morphometric and spatial distribution parameters of depressions

with geologic and hydrogeologic factors such as potentiometric level fluctuation of the

Floridan aquifer, thickness of overburden material above the Floridan aquifer, depth to

water table, soil type, and the vadose zone lithology suggests that all of the factors

examined in this study collectively affect the morphometric characteristics and spatial

distribution of karstic depressions. Analysis of depression major axis orientations does not

show a structural control on the karst development

Application of the mass flux technique to available data indicates that the best

estimate of the net karstic dissolution rate in the Lower Suwannee River basin is 40

m3/km2/year. This is equal to a denudation rate of 40 mm/1000 years.

Morphometric and spatial distribution characteristics of karstic depressions suggest

that the last phase of the post-Miocene karstic evolution within the Lower Suwannee

River basin has been controlled by the retreat of the Pleistocene sea level stands. During

the Pleistocene, as interglacial seas retreated, marine terraces were formed by sequential

sea level lowstands and the total period of subaerial exposure diminished towards the sea.

Consequently, geomorphologically younger karst landforms formed as the elevation of

marine terraces decreased. This evolutionary pattern of karst landforms has resulted in the

development of more frequent and/or larger and more complex depressions at higher

elevations.













CHAPTER 1
INTRODUCTION

Karst can be defined as a terrain with distinctive hydrology and landforms formed

by a combination of high rock solubility and well developed secondary porosity (Ford and

Williams, 1989). Approximately 20 percent of the earth's land surface consists of karstic

terrains and about 25 percent of the global population supplies its water from wells and

springs in karst aquifers (Ford and Williams, 1989; White et al., 1995). The highly

complicated surficial and underground karst network of the subtropical Florida karst not

only hosts one of the most productive aquifers of the world, but also reveals a variety of

karst features such as caves, dolines, uvalas, poljes, sinking streams and dry valleys which

are world-famous. The Suwannee River basin presents an extraordinary set of karst

features of Florida. Here, the generally unconfined or semiconfined hyrogeologic

conditions of the Floridan aquifer have resulted in a fascinating collection of karst

depressions illustrating a complex evolutionary history.

The Florida Platform, on which extensive carbonate deposition took place for

much of the Mesozoic and Cenozoic, has been variously subject to periods of extensive

karstification resulting from the glacio-eustatic fluctuations. Numerous studies on the

hydrogeochemical, structural, diagenetic and hydrogeologic characteristics of the Florida

karst have provided a great deal of information which may serve as a basis for

understanding the karstic evolution of the Suwannee River basin (Vernon, 1951; White,

1958; Ceryak et al., 1983; Randazzo and Bloom, 1985; Scott, 1997). Yet, a detailed








geomorphologic approach with particular emphasis on the analysis of the spatial and

temporal characteristics of karstic depressions and studies on karst has rarely been

attempted. Despite considerable study on the classification and formation mechanisms of

sinkholes in the Florida Platform, much confusion still exists as to the evolution of karst in

the Suwannee River basin. Better understanding of the karstic evolution of the Suwannee

River basin requires recognition of paleoenvironments and associated karstic dissolution

rates, identification of glacio-eustatic controls on the karst base level changes, evaluation

of the influence of paleokarst phases, and particularly assessment of spatial and temporal

distribution of sinkholes.

This study integrates morphometric techniques and karstic erosion rate studies in

an attempt to describe the subtropical Suwannee River karst area and to address the post-

Miocene evolution of karst. The study area is located in the Suwannee River basin,

northwest peninsular Florida, where sinkhole (doline) karst dominates (Figure 1.1). The

hypothesis to be tested is that the post-Miocene evolution of karst in the study area has

been predominantly controlled by the sea level changes represented by marine terraces at

various elevations and by changes from confined to water table aquifer conditions of the

Floridan aquifer. It is further hypothesized that the fluvial erosion of the overlying

impermeable siliciclastic deposits is responsible for the post-Miocene karstic evolution.

The morphometric and spatial distribution parameters of karstic depressions were

determined based on a Geographic Information System database created by delineating

depressions on twenty-four 7.5 minute topographic maps of 1:24,000 scale along the

Suwannee River (Chapter 2). The effect of geologic, structural, and hydrogeologic

variables on the morphometry and spatial distribution of depressions are discussed in



























































Figure 1.1. The study area.








Chapter 3. Factors addressed within this context include; fluctuation of the Floridan

aquifer potentiometric level, thickness of the overburden material, depth to water table,

type of soil media, lithological conditions of the vadose zone, and regional and local

fracture traces. Chapter 4 includes a thorough discussion of karstic dissolution rate

evaluation and its seasonal changes within the lower Suwannee River drainage basin.

Chapter 5 describes the substitution of space for time to address the problem of post-

Miocene karstic evolution in the study area by addressing how depression morphology and

distribution pattern have changed through time, that is, on surfaces of successively greater

age.



Climate


Humid subtropical climate prevails in the study area. Average annual temperature

varies between 680 720 F. Mean annual precipitation is 1422 mm. Most rainfall occurs as

localized thunderstorm activities in the summer months from June through August. In

winter, convectional systems provide rainfall of longer duration and uniform spatial

distribution (Fernald and Patton, 1985). Average evapotranspiration rate estimated for the

area is 1092 mm per year (Fisk, 1977; Ceryak et al., 1983).



Physiographic Setting


The Northern Highlands and the Gulf Coastal Lowlands constitute the major

physiographic provinces in the area (Figure 1.2) (Schmidt, 1997). As one of the most

distinct physiographic features in Florida, the Northern Highlands occupies most of the





























8 0 8 16 Kilometers

VCody Escarpme
Major rivers


Figure 1.2. Physiographic regions.








north and east of the study area. Elevations for the Northern Highlands range between 100

- 240 feet (30.5 73.2 m) asl within the study area. The plateau, which originally covered

the entire study area has been eroded by headward erosion through surface drainage as

well as by karstic dissolution within the underlying carbonate units (Scott, 1997). The

Cody Escarpment, which is described as ".. the most persistent topographic break in

Florida" (Puri and Vernon, 1964, p. 11), separates the Northern Highlands from the Gulf

Coastal Lowlands. The scarp generally coincides with the 100-foot topographic contour

except when it is interrupted by rivers such as the Suwannee, Withlacooche, and Santa Fe.

Hydrologically, the Northern Highlands contains a thick confining unit of

Hawthorn Group sediments and represents confined conditions of the Floridan aquifer.

Occasionally, cover collapse doline development takes place near the margin of the

plateau (e.g., Devil's Millhopper). In general, karstic depressions in this zone are marked

by low relief slumpage and erosional infilling due to thick cover sediments.

The Gulf Coastal Lowlands, covering large areas within the study area, consists of

both erosional and depositional features. Broad plains of a series of Pleistocene surfaces

and shorelines are pitted with karstic depressions within the limestones at or near land

surface. It represents a typical mature karst terrane with a thin mantle of permeable marine

terrace deposits. Because of the low topographic relief and rapid infiltration of rainfall by

diffuse recharge to the karst aquifer, surficial runoff is limited to major rivers. Remnants of

the Northern Highlands occur as outlier hills and sinkhole fill. Important remnant features

in the Gulf Coastal Lowlands are the Bell and Brooksville ridges (Figure 1.2). The coastal

swamps consist of muddy and silty deposits which support a growth of both fresh water

and saltwater marshes.








The transition zone or "Marginal Zone" between the Gulf Coastal Lowlands and

the Northern Highlands is an active zone of extensive karstification caused by allogenic

recharge from the covered karst terrains of the Northern Highlands. Because of their

highly acidic chemical properties, the waters of these allogenic streams rapidly dissolve the

carbonates and disappear underground along the Cody Escarpment. The area is

represented by numerous sinking streams, sinkholes of various sizes, closed karst praires

(poIjes), karst resurgences, and extremely complicated surface water-ground water

interactions. Point recharge to the Floridan aquifer occurs via discrete conduits connected

to well-developed phreatic cave systems of anastomotic pattern. The hydrogeologic and

geomorphologic structure of the karst system is further complicated by paleokarst

horizons developed during low sea-level stands (Hunn and Slack, 1983).

Geologic Structure


Despite its relatively stable geologic history on the passive margin of the North

American Plate, the Florida Platform is considered to have been subject to tectonic

activities occurring in the Caribbean and Gulf of Mexico (Randazzo, 1997; Scott, 1997;

Safko and Hickey, 1992). Structural features associated with these tectonic forces are

explained by possible fault and fracture systems, folding, uplift and subsidence.

Major structural features of the Suwannee River basin are the Ocala "Uplift",

Peninsular Arch, and the Suwannee Straits (Figure 1.3). The term "Ocala Uplift" was first

used by OB. Hopkins in the USGS press release in 1920 to describe the Eocene limestone

outcrops in the western peninsular Florida. Vernon (1951, p.55), generally agreeing with

Hopkins, described this feature as having ...... developed in Tertiary sediments as a gentle






8



















SOUTHEAST
-'"GEORGIA
CHATTAHOOCHEE I Q EMBAYMENT
ANTICLINE "
%- ] NASSAU NOSE

JACKSONVILLE
BASIN
ST. JOHNS
PLATFORM
GULF 0SANFORD
BASIN APALACHICOHIG
EMBAYMEN B REVARD
LEGEND PAFR


AXIS OF POSITIVE
FEATURE

AXIS OF NEGATIVE LOW
FEATURE
APPROXIMATE UPDIP IMIT
AND AREA UNDERLAIN BY THE
FLORIDAN AQUIFER SYSTEM

0 50 100 150 200 MILES


0 100 200 300 KILOMETERS

SCALE


Figure 1.3. Structural features of Florida (from Scott, 1997).








flexure, 230 miles long, and 70 miles wide where exposed in central peninsular Florida."

He suggested that the "uplift" consists of two well-defined shallow folds trending

northwest-southeast formed by tectonic activities occurring from Late Eocene through

Early Miocene. Noting that the rocks older than Middle Eocene were not affected by the

proposed Ocala "uplift", Winston (1976) suggested that this feature was not an uplift but

was formed by a thickening of the Lake City Formation and the eastward tilting of the

Florida Peninsula (Ceryak et al, 1983). In this study, the term "Ocala Platform" will be

used in order to avoid any generic connotation of this feature whose origin is yet to be

clearly defined. Regardless of its origin, the Ocala Platform has significantly affected the

Neogene deposition and the karstic evolution of the area. Miocene sediments once

covering the whole area are much thinner on the Ocala Platform. Fluvial and karstic

erosion of the Hawthorn Group sediments exposed the Eocene carbonates along the crest

of the Ocala Platform (Scott, 1997).

The Peninsular Arch is approximately 275 mile long subsurface feature trending

northwest-southeast and forming the axis of the Florida peninsula (Applin, 1951). Its crest

is formed by the pre-Mesozoic rocks which were topographically high during the Early

Cretaceous (Puri and Vernon, 1964) and intermittently positive during the Cenozoic

(Miller, 1986). Based on the maps by Miller (1986), Scott (1997 p. 58) states that the

Peninsular Arch has affected sedimentation as late as Oligocene, but "... did not affect

deposition of Neogene to Holocene sediments."

The Suwannee Channel located to the north of and adjacent to the Upper

Suwannee River basin was a topographically low feature separating the Florida Platform

from the continental landmass. It is described as an approximately 200 miles long








and 20 to 30 miles wide channel developed on top of the Cretaceous rocks (Hull, 1962).

The Suwannee current flowing within the channel has isolated the Florida carbonate

Platform from the influence of siliciclastic sedimentation until the Late Oligocene (Scott,

1997).

Geologic Setting


After its separation from the African plate during the Jurassic, the Florida Platform

has been a depocenter for thick carbonate sediments and minor amounts of evaporite

throughout the Paleogene (Randazzo, 1997). The carbonate factory of the warm, shallow

seas was effectively isolated from the northerly siliciclastic sediment transport by the Gulf

Trough and the Suwannee Channel in which the Suwannee current was active until the

Late Oligocene. Encroachment of siliciclastic transport onto the Florida Platform began to

suppress carbonate deposition when the Suwannee current ceased to occur following a

major sea level drop in the Oligocene. Siliciclastic sediments covered the entire Platform

by mid-Pliocene and continued to be predominant except for the southernmost Florida,

where carbonate sedimentation took over during the Late Pliocene (Randazzo, 1987;

Scott, 1997).

Surrounded by submarine escarpments on both east and west, the Florida Platform

consists of a thick sequence of limestone and dolomite deposited during the Tertiary

period. Carbonate deposition in the warm, shallow seas of the Paleocene formed the Cedar

Keys Formation. It was overlain by the Oldsmar, Avon Park, and Ocala carbonate

sequences (Eocene) and Suwannee Limestone (Oligocene). Repeated sea level

fluctuations during the Eocene have brought about short episodes of nondeposition and












subaerial erosion within the carbonate rock sequence as evidenced by numerous

unconformities (Randazzo, 1997). The oldest unit exposed in the study area is the Ocala

Limestone.

In general, Miocene Hawthorn Group deposits are comprised of siliciclastics

interbedded with carbonates. Phosphate is abundant in the Hawthorn Group sediments of

the study area. The siliciclastic content increases in the younger sediments (Scott, 1997).

On the lithology of the Hawthorn Group sediments Scott and MacGill (1981 p. 23)

reported that the "...most predominant lithology in the study area is a silty, sandy,

phosphatic dolomite, ...and comprises approximately 90% of the volume of sediments."

Thickness of the Hawthorn Group increases to the north and east within the study area. It

is virtually absent throughout the Gulf Coastal Lowlands physiographic province whereas

its thickness may exceed 200 feet (61 m) in the northeasternmost part of the study area.

Ceryak et al. (1983) used the term "Undifferentiated Marine Terrace Deposits" for

the Plio-Pleistocene clastic materials which overlie the Hawthorn Group sediments, where

present, and the carbonate rocks of the Gulf Coastal Lowlands. These terrace deposits are

formed by the Plio-Pleistocene eustatic sea level fluctuations and are composed of fine-to

medium-grained quartz sand with minor amounts of organic material, clay, and heavy

minerals.








Hydrogeologic Setting


Three aquifer systems are located in the study area. A surficial (water table)

aquifer and an artesian "intermediate" aquifer exist in the Northern Highlands (Figure 1.4).

The Floridan aquifer underlies the whole peninsular Florida and the panhandle, displaying

both artesian and water table conditions.

The surficial aquifer is an unconfined aquifer consisting of Miocene and younger

sediments. Its thickness varies between 20 to 150 feet (6 to 45.7 in). Ground water is

recharged directly from precipitation. In the study area, due to the greater hydraulic head

of the surficial aquifer, part of the percolating water leaks downward through confining

beds to recharge the intermediate and the Floridan aquifer system. Because of the

complex interbedding of aquifer material, the permeability and transmissivity of the

surficial aquifer are extremely variable (Miller, 1997).

The intermediate aquifer system occurs between the surficial aquifer and the

Floridan Aquifer. Lithologically, the aquifer material contains phosphatic sands, clays, and

carbonates. Its thickness and depth varies with the lateral and vertical continuity of

hydraulic conductivity. Thickness increases from the west to the northeast ranging from

80 feet to 234 feet (24.3 to 71.3 in). Confined hydraulic conditions occur except locally

where overlying impermeable units are thin through erosion. Most recharge to the

intermeadiate aquifer occurs as downward leakage through the confining unit (Ceryak et

al., 1983)










Era Period Epoch Group/Formation Hydrogeologic Unit


Holocene


________________________________________________________ I


I Pleistocene


Pliocene


1.-I.


Miocene


6.1


6.1
0
-3


undifferentiated


undifferentiated


Hawthorn
Group


___ _ ___ I ~ 3


Oligocene


Eocene


-4


Suwannee
Limestone


Ocala Limestone


Avon Park
Formation


____ 1 4


Oldsmar
Formation


I_ J i


Paleocene


______ _______ 1 1


Cretaceous


Cedar Keys
Formation


undifferentiated


Surficial Aquifer


Intermediate Aquifer/
Confining Layer


Upper
Floridan
Aquifer


Lower Floridan
Aquifer


Lower Confining
Unit


Figure 1.4. Hydrostratigraphic column for the Suwannee River Basin (after Hirten, 1996).








As one of the most productive karst aquifers of the world, the Floridan aquifer is

the principal aquifer in the study area. It consists of a thick sequence of Tertiary carbonate

rocks. Because of the highly variable lateral and vertical hydraulic conductivity of the

aquifer material, there are zones of high productivity and low transmissivity within the

Floridan aquifer system. Its vertical extension is determined by the relative permeability

difference between the aquifer and the upper and lower boundaries of the system. The

permeability of the aquifer material is at least one order of magnitude greater than that of

the upper and lower confining units. Naturally, this permeability difference does not

correspond everywhere to a particular lithologic unit. In general, massive anhydrite beds

of the Lower Cedar Key Formation constitute the base of the aquifer. The top of the

Ocala Limestone, or Suwannee Limestone, where present, marks the upper limit of the

aquifer. In the study area, the most productive zones of the Floridan aquifer occur within

the Suwannee Limestone and Ocala Limestone.

The Floridan aquifer system exhibits both confined and unconfined conditions.

Overlain by the thick impermeable zones of the Hawthorn Group, the aquifer acts as an

artesian system within the Northern Highlands physiographic region. Recharge to aquifer

is maintained through slow leakage through the confining unit or as point recharge

through collapse depressions breaching the impermeable layers of the overlying Hawthorn

Group. In areas where the Hawthorn Group has been eroded, the Floridan aquifer is

unconfined and represents water table conditions. This situation occurs over a great

portion of the study area within the Gulf Coastal Lowlands physiographic province

(Western Karst Plain). Here, diffuse recharge takes place through a thin blanket of terrace

deposits covering the karst aquifer.








Because of the relatively high primary porosity coupled with extensive karstic

dissolution conduits within the carbonate rocks of the Floridan aquifer, ground water

storage and flow take place through a complex system of intergranular openings and

cavities. Both diffuse and conduit ground water flows occur. Preferential ground water

flow along solutionally enlarged joints, fractures, and bedding planes have been

documented by numerous studies (e.g., Rosenau et. al, 1977; Hunn and Slack, 1983; Beck

and Arden, 1983). Unlike deep karst ground water flow in south Florida, most of the

ground water circulation in the study area occurs within the 200 to 300 feet (60 to 90 m)

of saturated limestone. Most of the cave systems are reported to have formed at the

contacts between the Hawthorn Group-Suwannee Limestone or Ocala Limestone-

Suwannee Limestone (Beck, 1986). Rapid ground water flow takes place within the

complex cave systems from recharge areas to major karst springs along the Suwannee

River. Because of the complicated interaction between the surface water and ground

water, recharge and discharge areas show strong temporal variations. Numerous ground

water tracing studies were carried out to better understand the ground water-surface

water interactions and to delineate ground water catchment areas for karst springs in the

area (Hisert, 1994; Kincaid, 1994; Hirten, 1996).



Karst in Florida


Sea level is the absolute base level of karstification and its changing positions have

played an important role in the vertical extension of karst processes, giving rise to

repeated shifts of vadose and phreatic zones. Numerous subaerial exposures and








accompanying dissolution processes caused by sea level fluctuations have extensively

karstified the thick Tertiary carbonate rocks of the Florida Platform, resulting in a multi-

cyclic, complex network of karst development (White, 1958; Hanshaw and Back, 1979;

Randazzo and Bloom, 1985; Upchurch, 1989). Dissolution of carbonate rocks has been

prevalent especially along the unconformities, marked by paleokarst horizons or extensive

phreatic cave systems. In accordance with the sea level fluctuations, shifts in the sea

water-fresh water mixing zone caused significant diagenetic changes in the carbonate

rocks (Randazzo and Bloom, 1985; Back et. al, 1986). Various hydrochemical processes

including dolomitization, dissolution and precipitation of carbonate minerals have taken

place in the Floridan aquifer (Hanshaw and Back, 1979).

Sea level fluctuations have also brought about changes in recharge and discharge

zones of the Floridan Aquifer. Most of the springs discharging from the Floridan Aquifer

have sinkhole morphologies with shallow conduit networks. They represent previous point

recharge locations to the Floridan Aquifer and currently function as discharge points due

to rising sea levels (Ceryak et al, 1983).

Karst in Florida has developed in a relatively stable tectonic setting and is covered

by either thick, impermeable siliciclastics or thin layers of Pleistocene terrace deposits.

Therefore, structural control on karst development can not be readily observed.

Nevertheless, preferential karst development along joints, fractures, and photolineaments

were reported by White (1958), Pirkle and Brooks, (1959), and Littlefield et al (1984).

Regardless of the existence of structural control on the karst development, major

dissolution of carbonate rocks occurs at shallow depths of recharge areas with soil cover

where water is undersaturated with respect to calcite, or within the sea water-fresh water








mixing zone where undersaturation results from mixing of two different water bodies

(Back et al, 1986; Randazzo, 1997). Major conduit development has also been observed

within the zones of potentiometric level fluctuation (Ceryak et al, 1983).

Unlike tectonically deformed, uplifted, bare karst of the temperate Mediterranean

region where almost all the drainage takes place underground, or tropical karst with

positive features surrounded by polygonal doline fields, the Florida karst displays a gently

rolling topography with shallow depressions. However, this gentle topographic relief and

relatively low depression density in Florida represent a muted surficial expression of much

denser doline network covered by at least several meters of soil material or as much as

hundreds of meters of impermeable confining layer. In other words, karst in Florida is

much more extensive and complicated than it is observed on the ground surface.

Karstification has occurred as a multisequential process through which vertical and lateral

dissolution zones associated with different sea level stands have been superimposed to

form a complex three dimensional network of conduits throughout the thick carbonate

rock sequence of the Florida Platform (Davies and Legrand, 1972; Randazzo, 1997).

Sinkholes (dolines) are the ubiquitous landform in the temperate karst of Florida.

Beck (1986) classifies them into three types: ponors (swallets, stream-sinks), subsidence

sinkholes, and cenotes. Most of the sinking streams represent allogenic recharge to the

Floridan aquifer in an area confined to the transition zone between the Northern Highlands

and the Coastal Lowlands physiographic provinces. He further relates these ponors to the

paleokarst dissolution pipes reactivated by the erosion of the cover material.

Ponors, as described by Beck (1986), corresponds to the solution sinkholes of

Ford and Williams (1989). They subclassify solution sinkholes into point recharge and








drawdown sinkholes. Point recharge sinkholes form as the overlying material on an

undeformed carbonate rock sequence is eroded by fluvial erosion. Points of high fissure

frequency are attacked by chemically aggressive allogenic runoff. Once the connection

between the input and output is established by a proto-cave system, then the removal of

carbonate material is maintained by the focused recharge. Removal of rock mass is greater

at the center of depressions than their sides. More water is focused as a result of

increasing drainage areas and a positive feedback mechanism is established. Density of

point recharge sites is increased by the continued erosion of the overlying material and by

more sinkhole initiation at the expense of reduced drainage areas of individual depressions.

Drawdown sinkholes can be observed in uplifted and previously karstified areas

where the overburden material has long been eroded. Focused corrosion is established not

by point recharge, but by drawdown within the epikarstic (subcutaneous) aquifer (Figure

1.5). Williams (1983) has shown that diffuse recharge through the soil cover forms a

perched water table in the extremely karstified epikarstic zone by virtue of the relatively

low permeability of the underlying bedrock. However, preferred vertical paths of

dissolution develop within this zone because of the initial variabilities in permeability.

Further development of these paths brings about a cone of depression in the overlying

epikarstic water table, resulting in focused corrosion and initiation of drawdown sinkhole

development in the epikarstic zone with no requirement for a caprock point recharge.

Once the positive feedback is established, zones of focused corrosion are represented by

topographic depressions whose diameters are controlled by epikarstic water table

drawdown cone (Williams, 1983).









In discussing the origin of solution sinkholes Ford and Williams (1989, p.402)

point out that there is a ".. necessity to distinguish (a) between doline initiation where

there has been no proto-cave development and that where a ready made, permeable

vadose zone is inherited from an earlier phase of karstification, and (b) between doline

corrosion focused by point recharge as opposed to that focused by epikarstic drawdown."

They further suggest that both successive caprock stripping and formation of secondary

drawdown within a major epikarstic drawdown can account for the formation of different

generations of sinkholes (parent and daughter, primary and secondary).

Subcut neous infiltration rate
p i z metr controlled by soil
surface Subsoil korren





l ow percol ati on
in tight fissures t


19Eff)cient drainage v
connected pipe


Enlarged fissures
permit rapid
infiltration
Do line Doline




C a Pill r ..-- I diminishes.- ,.. :
barrier 'I ;,
preventsI.f with / :.-
rapid i I
percolation d- e, tI



Figure 1. 5. Drawdown doline initiation in the subcutaneous (epikarstic) zone (Williams,
1983).








Solution sinkholes constitute the predominant sinkholes type in the study area

where a thin blanket of soil material overlies the carbonate rocks of the unconfined or

semiconfined Floridan aquifer. Within the transition zone between the Northern highlands

and the Gulf Coastal Lowlands where overlying siliciclastic material has been constantly

eroded along the Cody Escarpment, focused corrosion by point recharge of allogenic

runoff forms solution sinkholes of different generations. Epikarstic drawdown sinkholes

are also expected to occur within the Gulf Coastal Lowlands, where the caprock has long

been eroded and a paleokarst template is available to be reactivated by focused corrosion.

After the initiation of sinkholes, their development is further encouraged by the increased

aggresiveness of water as it percolates through thick soils accumulated in the bottoms of

depressions. In some cases, accumulation of low permeability material within depressions

may retard the sinkhole development, forming ponds and lakes. A majority of lakes in

Florida has formed within karstic depressions lined by impermeable material.












CHAPTER 2
MORPHOMETRIC AND SPATIAL DISTRIBUTION PARAMETERS OF KARSTIC
DEPRESSIONS


Prospectus


In light of the morphoclimatic approach to karst geomorphology, much research

has been directed towards a better understanding of karst landform evolution in different

climatic regions. Numerous attempts have been made to relate different karst landform

patterns and corresponding dissolution rates to climate (Corbel, 1957; Lehman, 1964;

Jakucs, 1973). It has been argued that variations in karst landforms were caused by

different karst processes related to climatic variations in soil and vegetation types, and

amounts and patterns of runoff. These early applications of a morphoclimatic approach to

karst were based upon visual comparisons of karst landforms formed in contrasting

climates, and on assertions, rather than measurements, of differences in erosion processes

(Smith and Atkinson, 1976). Nevertheless, the morphoclimatic approach to karst

processes and landforms has given rise to new perspectives in understanding the operation

of karst systems. Its application to various karst regions has also brought the necessity for

an objective system of karst landform description and analysis.

An objective and quantitative description of karst landforms is achieved by karst

morphometry. It has been widely applied to a variety of karst regions and proved to be a

very effective technique to describe and characterize karst landforms, especially

depressions (e.g. Williams, 1972; White and White, 1979; Day, 1983; Troester et al.,








1984). Evolution of karst landforms was also addressed by using morphometric and spatial

distribution parameters of karstic depressions (e.g. Drake and Ford, 1972; Kemmerly and

Towe, 1978; Kemmerly, 1982).

Smith and Atkinson (1976, p. 403) explain the importance of morphometric

techniques and erosion rate calculations as ".... the tools of process studies and

morphometric analysis, when used together, may prove powerful enough to unravel the

interaction of lithology (or Structure), erosion rate and distribution (Process), and time

(Stage) in sculpting the complex variety of landscapes found in limestone regions."

Despite its variety of karst landforms, the subtropical karst of Florida has rarely

enjoyed a morphoclimatic approach with special attention to its morphometric parameters

and erosion rates. The few studies on depression morphometry and spatial distribution

were of local scale with small depression populations. In this study, morphometric analysis

of a relatively large portion of the Suwannee River karst area was made possible by using

the Geographic Information Systems.

Background


Morphometry can be defined as the measurement and mathematical analysis of the

configuration of the Earth's surface and of the shape and dimensions of its landforms

(Bates and Jackson, 1987).

Based on hundreds of field measurements, Cvijic (1893) classified dolines

according to their side slope angles and depth-diameter ratios. He described them as bowl,

funnel-, and well-shaped depressions. His quantitative approach was continued by Cramer








(1941), who summarized statistical results of previous workers and presented

morphometric information for diverse karst regions of the world.

Other important contributions to karst morphometry were brought by Williams's

(1966, 197 7) works on the temperate and tropical karst of Yorkshire and New Guinea,

respectively. Considering dolines as the smallest drainage basins, he mapped topographic

divides, summits, channels and stream sinks on the polygonal karst of New Guinea. He

applied Strahler's (1957) channel ordering system to depressions and classified them based

on their drainage characteristics. Williams (1971) not only introduced a hydrological

perspective to karst morphometry, but also drew attention to the spatial distribution

patterns of karstic depressions. Following methods developed by plant ecologists, he

applied Clark Evans' (1954) index on the nearest neighbor distances as a measure of

randomness in depression distribution (Williams, 1971, 1972).

The nearest neighbor analysis is used to describe a pattern with a nearest neighbor

index that compares the actual nearest neighbor distance (La) with the average expected

distance (L.). The expected nearest neighbor distance in a randomly distributed depression

population is given by,

Le = 1/(24D) where D is the depression density.

The nearest neighbor index R, described as the ratio of La / L, ranges from 0 for

maximum clustering to 2.149 for a regular pattern in which depressions are as evenly and

widely spaced as possible. A nearest neighbor index value of 1 indicates random

distribution (Williams, 1972).

Drake and Ford (1972) applied morphometry to karst landform evolution. Using

quadrat analysis, they verified the existence of two generations of dolines in the Mendip








Hills, UK. It was concluded that an average of four daughter depressions cluster around

randomly distributed mothers.

Statistically comparing two random samples of karstic depressions based on the

absence or presence of natural ponds, Kemmerly (1976) used morphometric measures

such as long-axis orientation, length of long axis versus mean depression width. He

concluded that the two random depression samples in Tennessee are not from the same

population. He further analyzed a contagious karst mechanism and constructed a

conceptual model for depression initiation and development by defining the parent and

daughter depressions with different spatial characteristics (Kemmerly, 1982, 1986).

White and White (1979) performed a comprehensive statistical analysis on

parameters of karst development such as relief factor, drainage factors, size and shape

factors, carbonate rock fractions, measures of doline development, and measures of

internal drainage. They described relationships between sinking stream catchment area and

stream length, area of internal drainage into depressions and number of depressions, and

also compared morphometric measures of karst with different rock type.

Among the more recent studies, Magdalena and Alexander (1995) analyzed

sinkhole distribution in Minnesota. They applied nearest neighbor analysis to a sinkhole

data set consisting of field-located karstic depressions. The randomness of sinkhole

distribution was also tested by comparing it with artificially created random data sets. The

observed sinkhole locations were found to be significantly different from a random spatial

distribution. Rather they tend to occur in clusters, in which new sinkhole development is

observed.








Despite an apparent absence of systematic application of morphometric techniques

to karstic depressions of Florida, there are several important studies characterizing the

subtropical Florida karst. Most research of karstic depressions in Florida is based on the

sinkhole database compiled by the Florida Sinkhole Research Institute. The database

consists of information on reported sinkholes. Therefore, distribution of sinkholes is

strongly biased towards highly populated areas where sinkhole occurrence was detected

and reported to the Institute. Nevertheless, studies based on this database may provide a

better understanding of sinkhole-prone areas. One exception to these studies is the work

of Troester et al. (1984) in which doline depth distribution was examined for two tropical

and four temperate karst areas including Florida, based on data obtained from topographic

maps of various scales. Considering the depth information as the most useful parameter in

distinguishing temperate and tropical karst, they produced depth frequency plots for

karstic depressions. For all areas considered in the study, the depth distribution was

approximately exponential with different values of the exponential coefficient. Unlike

tropical karst regions of Puerto Rico and Dominican Republic, the low relief Florida karst

was represented by broad, shallow depressions.

Sinkhole density maps for old and new sinkholes were created by Bahtijarevic

(1989) for a highly karstified area in central Florida. In order to classify areas based on

their sinkhole densities she performed a terrane evaluation on the overall sinkhole

population.

Wilson (1995) studied the surficial and buried sinkhole densities and new sinkhole

frequencies in Peninsular Florida. Based on data obtained by ground penetrating radar

studies, derived from topographic maps and from the database of new sinkholes he








provided statistical information on old, new and buried sinkhole densities. He stressed the

importance of buried sinkhole density information on assessing the vulnerability of the

Floridan aquifer to ground water contamination.

Methodology


Morphometric and spatial distribution parameters of karstic depressions in the

Suwannee River basin were evaluated in Geographical Information Systems (GIS)

environment, using ArcInfo 7.0. In this study, morphometric analysis of 24,757 karstic

depressions in an area covered by twenty-four 7.5 minute standard USGS topographic

quads were made possible by the powerful and rapid analytical capabilities of the GIS. The

procedure followed in obtaining morphometric data discussed in this chapter consists of

extensive manual and computer work.

The first stage of the study involves detailed topographic map analysis. All the

natural depressions depicted by hachured closed contours on topographic maps along the

Suwannee River were delineated on transparent papers. The following topographic

quadrangles were analyzed in this study: Lee, Ellaville, Fort Union, Hillcoat, Madison SE,

Fallmouth, Live Oak West, Day, Dowling Park, Mayo, Mayo SE, Mayo NE, O'Brien,

O'Brien SE, Branford, Hildreth, Hatchbend, Bell, Wannee, Fourmile Lake, Suwannee

River, Trenton, Manatee Springs, and Chiefland (Figure 2.1).

All the depressions delineated on topographic quads were transferred one by one

to Arclnfo (GIS) environment. Digital maps (coverages) of depressions were produced. In

order to utilize GRID functions of Arclnfo, vector coverages of depressions were









































4 0 4 8 Kilometers


+


Figure 2.1. Names and location of topographic quadrangles within the study area.








converted to raster files. GRID is a cell-based module of Arclnfo. It manipulates rater files

in which a regular "mesh" is draped overthe landscape. The predominant phenemonon

within a cell becomes the value of that cell. Morphometric measurements were performed

on all of the depressions using zonal functions of GRID. Compound depressions were

taken as single depressions for measurements of length, width, depth, and major axis

orientation.

Nearest neighbor analysis was performed for each quadrangle. Compound

depressions were divided into their components and each depression was taken into

account individually. Nearest neighbor distances were calculated by using Fragstats, a

powerful spatial analysis software originally created for landscape pattern analysis of

digital data. Depressions were represented by their centroids. Ideally, karstic depressions

are represented by their deepest points, i.e. swallow holes, for an accurate spatial

description which is of paramount importance in spatial analysis. Williams (1971, 1972)

was able to locate the swallow holes and drainage patterns in the polygonal karst of New

Guinea. However, they are not detectable in the low-relief karst of Florida, which is

covered by at least several meters of soil and thick vegetation. Therefore, in order to

maintain consistency and avoid subjectivity in assessing sinkhole locations, karstic

depressions were represented by their centroids that were readily and precisely determined

by GRID functions in GIS.

Two GIS databases for karstic depressions were created: The compound

depressions database includes morphometric parameters such as depression area,

perimeter, length, width, mean diameter, length/width ratio, circularity index, major axis

orientation, and approximate depth. The database for individual depressions includes








centroid locations of depressions, nearest neighbor distances, and azimuth values for

nearest neighbor vectors.

Among a great number of morphometric parameters proposed by various karst

geomorphologists, only those that are possible to be measured or calculated by GIS based

on the available data were included in this study:

-Length: length of the major axis of an ellipsoid representing the depression.

-Width: length of the minor axis of an ellipsoid representing the depression.

-Orientation: orientation of the major axis. Originally expressed by GRID as the

angle from the east, counter-clockwise, it was converted to regular azimuth values.

-Area: area of each depression measured in sq.m.

-Pitting index: (Area of karst)/(Total depression area).

-Circularity index: the measure of the circularity of a depression.

Depth information was obtained by counting the number of closed contours for

each depression.




Results


Depression Density

Depression densities were calculated for each topographic quad by dividing the

number of depressions by area (Table 2.1). A total of 25,157 depressions were located in

the 4,063.138 k2 area examined. Mean depression density for the whole area is 6.06/

km2. It ranges from 2.43 (Madison SE) to 15.801 (Mayo NE) with a standard deviation of

3.15.

















Table 2.1. Morphometric and spatial distribution parameters of depression for topographic quadrangles.

QUADRANGLE NUMBER OF DEPRESSION TOTAL DEPRESSION MEAN DEPRESSION MEAN NEAREST NEAREST NEIGHBOR DISTRIBUTION PITTING LENGTH!
NAME DEPRESSIONS DENSITY AREA (sq.kin) AREA (sq.kn) NEIGHBOR DISTANCE (m) INDEX (R) PATTERN INDEX WIDTH
WANNEE 904 5.3 10.7 0.012 186.2 0.859 Tending to clustei 15.83 1.70
FOURMILELAKE 588 3.4 8.5 0.014 212.6 0.789 Tending to cluster 20.15 1.81
SUWANNEE RIVER 953 5.6 9.8 0.010 164.4 0.778 Tending to cluster 17.35 1.72
TRENTON 1283 7.5 20.5 0.016 174.8 0.956 Near random 8.35 1.74
MVLANATEE SPRINGS 953 5.6 12.4 0.013 170.8 0.809 Clustered 13.72 1.82
CHIEFLAND 1742 10.2 8.5 0,005 129.7 0.827 Tending to cluster 20.23 1.62
LEE 593 3.5 18.0 0.030 235.4 0,881 Tendingto cluster 9.41 1.70
MIADISON SE 405 2.4 13.0 0.032 245.4 0.765 Tending to cluster 12.84 1.99
ELLAVILLE 622 3.6 14.1 0.023 214.7 0.818 Tending to cluster 12.19 1.89
FORT UNION 482 2.9 16.6 0.034 255.5 0.864 Tending to cluster 10.20 1.80
FALLMOUTH 1002 5.8 11.4 0.011 174.9 0.843 Tending to cluster 15.06 1.80
LIVEOAK WEST 1389 8,1 19,1 0.014 161.3 0.919 Near random 8.97 1.69
HILLCOAT 444 2.6 4.2 0.009 178.8 0.578 Clustered 40.91 1.90
DAY 388 2.7 7.3 0.019 241.4 0.796 Tending to cluste: 19.49 1.84
DOWLING PARK 1514 8.9 21.7 0.014 159.6 0.950 Near random 7.89 1.76
MAYONE 2721 15.8 23.5 0.009 128.9 1.025 Near random 7.32 1.61
MAYO 1273 7.5 13.2 0.010 160.9 0.881 Tendingto cluster 12.86 1.77
ILAYO SE 1187 6.9 11.6 0.010 179.7 0.945 Near randorn 14.81 1.72
O'BRIEN 1724 10.1 16.6 0.010 160.3 1.019 Near random 10.27 1.72
O'BRIEN SE 1663 9.8 25.5 0.015 158.5 0.990 Near random 6.70 1.74
BRANFORD 766 4.5 8.4 0.011 191.0 0.814 Tending to cluster 20.15 1.81
HILDRETH 971 5.7 11.7 0.012 189.1 0.905 Near random 14.47 1.80
HATCHBEND 753 4.4 8.8 0.012 205.3 0.860 Tending to cluster 19.60 1.82
BELL 837 4.9 16.4 0.020 187.0 0.829 Tending to cluster 10.37 1.89








Distribution of depression density is also presented as a GIS layer with a spatial

resolution of 10 m. It was created by calculating the number of depression centroid points

in a one km2 window moving 10 m at each step. Nearly half of the study area (-1,900

kin2) is represented by a depression density between 1 to 5 / km2 (Figures 2.2 and 2.3).


60

, 50
.u
40

30

20

S10 -

0


1 -!
5-10 10-15
Depression Density (#Isq.km)


15-20


Figure 2.2. Histogram of depression density distribution.



The area of each depression was calculated by GIS functions. Based on the

calculations for each quad (Table 1), total depression area varies from 4.15 km2 (Hillcoat)

to 25.4 km2 (O'Brien SE) with a mean of 13.80 km2 and a standard deviation of 5.38 km2.

Total depression area is 331.390 km2, corresponding to 8.15% of the study area (4,063.14

km2).




















































60612


4I


Kilometers


Figure 2.3. Distribution of depression density.


/ r Major rivers
Depression density (#/sq.km)
10.1 -2
12-5
M5-10
10-15
15-20
20 25
25 30
30-35
35 42
No Data


i~i ;iiii l


ri i







As a measure of surficial karst development, the pitting index (total area of karst /

total depression area) provides information about the extent of karstification. For the

polygonal karst landscapes covered by tightly spaced depressions, the pitting index

approaches unity. Within the study area, the pitting index varies from 6.69 (O'Brien SE)

to 40.91 (Day) with a mean of 14.5 and a standard deviation of 6.98 (Table 2.1).

Spatial distribution of depressions treated as points located at their centroids was

analyzed using the simple nearest neighbor technique (Williams, 1972a, 1972b). The mean

nearest neighbor distance (La) for karstic depressions within each quad was calculated

using FRAGSTATS (Tablel). It ranges from 128.95 m (Mayo NE) to 255.53 m (Fort

Union) with an overall mean and standard deviation of 186.09 m and 33.3 m, respectively.

The expected mean nearest neighbor distance (Le) in an infinitely large, randomly located

population with same depression density, D, is given by Le = 1/(2,iD) and ranges from

125.78 m (Mayo NE) to 320.98 m (Madison SE) for the study area.

The nearest neighbor index R, described as the ratio of La / Le, was derived for

corresponding quads and presented in Table 1. It ranges from 0.578 (Hillcoat) with a

clustered distribution to 1.025 (Mayo NE), which represents randomly distributed

depressions. Mean nearest neighbor index for the whole area is 0.854. Most of the

topographic quadrangles reveal a depression distribution significantly different from

random expectation (R=I) at the 0.05 level. All the depressions except those in Trenton,

Live Oak West, Dowling Park, Mayo NE, Mayo SE, O'Brien, O'Brien SE and Hildreth

indicate a tendency towards clustering.

Nearest neighbor analysis was also performed for the whole area by using map

algebra functions of the GIS. Based on the centroid locations of depressions, layers of









measured nearest neighbor distance, La, and expected nearest neighbor distance, L were


created. The distribution of nearest neighbor index (R) within the study area was obtained


by the ratio of two GIS layers (La / L,). A greater part of the study area is represented by a


nearest neighbor index less than one (Figures 2.4 and 2.5) indicating a trend towards


clustering.


40

35

w 30
25
0
W
O 20--
z 15-
W
0
n 10


0~ N---AR ES,, T N IGBOR' IND, I

0.04- 0.27- 0.51- 0.74- 0.97- 1.21- 1,44- 1.67- 1.90-
0.27 0.51 0.74 0.97 1.21 1.44 1.67 1.90 2.14
NEAREST NEIGBOR INDEX


Figure 2.4. Histogram of nearest neighbor index.




Models explaining spatial evolution of sinkholes such as that proposed by Drake


and Ford (1972), and Kemmerly (1986) describe a contagious evolutionary pattern in


which primary or first generation sinkholes are initiated on major fractures and clustered


by secondary sinkholes forming around them. The evolution of karstic depressions within


the study area is discussed in Chapter 5. Only the descriptive parameters of depression


morphometry and spatial distribution are addressed within the content of this chapter.























V Major rivers
Nearest Neighbor Index
0.5
0.5 0.7
0.7 1.1
1.1 -1.7
1.7 2.1
No Data


5 0 Kilometers


R
) N
Im llm:W-L


Figure 2.5. Distribution of nearest neighbor index.










A histogram of nearest neighbor distance data set for depression centroid points is


shown in Figure 2.6. The mean distance to nearest neighbor is 188.097 m. The values


range from 20 m to 2956.0 m with a standard deviation of 116.47 m. Cumulative


percentage graph of depression nearest neighbor distance values shows that 90% of


depressions are situated closer than 500 m to their nearest neighbor (Figure 2.7).


12.0 -


10.0


z 8.0

LLI
,, 6.0
I-


xr 4.0
II H l

2.0-


0.0 -
20-40 80-100 140- 160 200-220 260- 280 320-340 380-400 440-460 500-520

NEAREST NEIGHBOR DISTANCE (m)


Figure 2.6. Percent frequency of nearest neighbor distance.





10000


1000
U0~
F-w
ci)H




10 _10
z

1
0 20 40 60 80 100
CUMULATIVE PERCENTAGE

Figure 2.7. Cumulative percentage of nearest neighbor distance.








Planimetric Shape

As an index of planimetric shape, the length/width ratios and circularity index of

depressions were investigated. Length and width values were taken as major and minor

axis values of depressions calculated by the "zonalgeometry" function of GIS. The ratio

varies between 1 for equidimensional depressions to 14.1 for extremely elongated

depressions representing ancient valley floors of sinking streams. Mean L/W value and the

standard deviation is 1.74 and 0.68, respectively (Figure 2.8).

Most dolines are near equidimensional in plan with a modal ratio class between 1 -

1.5. Ninety percent of the total population has a L/W ratio smaller than 2.5 (Figure 2.9).

Plots of area vs. LiW for each topographic quad (Appendix A) do not show a strong

tendency of elongation (high L/W) for larger sinkholes, which may be expected in time for

depressions located on major fracture lines.

The relationship between length and width of depressions was analyzed for each

quad. Linear plots of length vs. width are presented in Appendix B.

The planimetric shapes of depressions are also described by the circularity index,

1, expressed as:

Ic Am/Ae

where Am = measured depression area, and A = effective depression area.

The effective depression area is calculated by

S=* r2o

The effective radius, re, is estimated by

re = 2(A/P.)

where Pm is measured perimeter of the depression.














50

45

40

>- 35
0
z
n 30
a

S25-
z
w20

w
n_ 15

10-

5-

0 -.-
1-1.5 1.5-2 2-2.5 2.5-3
LENGTH



Figure 2.8. ]Percent frequency of L/W ratio.





5.0

4.5

4.0
0
3.5

3.0
2.5

2.0
0
z 1.5
1.0

0.5

0.0 -- _... ...... ___....i"_


3-3.5 3.5-4 4-4.5 4.5-5 >5
/WIDTH RATIO


0 10 20 30 40 50 60 70 80 90 100
CUMULATIVE PERCENTAGE


Figure 2.9. Cumulative percentage of L/W ratio.










For a perfectly circular depression, the index of circularity would be unity. The


greater it departs from one, the less circular is the depression. Elongated features have


values smaller than one whereas convoluted shapes present values greater than one.


Most depressions are near circular in plan with the major modal ratio class


between 1-1.2 (Figure 2.10). Mean circularity index value is 1.33 with a standard


deviation of 0.45. Approximately 85% of the total population have a circularity index


smaller than 1.5 (Figure 2.11).


60

-50
z
D40
o
u-30-
z
W 20-



0 I- F-1
0-1 1-1.2 1.2-1.4 1.4-1.6 1.6-2.0 2.0-3.0 3.0-4.0
CIRCULARITY INDEX

Figure 2.10. Percent frequency of circularity index.





4.0
X3.5
03.0
z
>2.5
I-
E2.0
-J1.5
01.0

00.5
0 0 0 30 4 50 6


>4.0


70 80 90 100


0 10 20 30 40 50 60
CUMULATIVE PERCENTAGE

Figure 2.11. Cumulative percentage of circularity index.


11I










Mean Diameter


Mean depression diameters were estimated by obtaining the mean of length and


width for each depression. In karst morphometry, the use of mean diameter is more helpful


than major and minor axes, since these measures represent maximum rather than average


dimensions. Mean diameter distribution for the total population is shown in Figure 2.12.


The values range from 10.1 to 2,785.9 m with a standard deviation of 106.6 and a mean of


98.6 m. The major modal class is 5 to 50 meters. Frequency distribution of mean diameter


(Figure 2.13) indicates that some 95% of the total population have a mean diameter of less


than 300 m


40
35-
(930
25

Wi20
a -
W15


5 to 50 100 to 150 200 to 250 300) to 350 40 to 450 500 to 550 600 to 650
MEAN DIAMETER (m)


Figure 2.12. Percent frequency of mean diameter.





1600
Ep1400 -
1200
W
1000
800
5600
~400-
200--
0
0 10 20 30 40 50 60 70 80 90 100
CUMULATIVE PERCENTAGE

Figure 2.13. Cumulative percentage of mean diameter.








The frequency distribution of depression areas for the total measured population is

given in Figure 2.14. The mean value is 14,570.31 in2, the standard deviation 71,349.08

in2, and the range of values from 25.0 to 5,804,750 in2. Within the total measured

population, 50% of depressions are smaller than approximately 5,000 M2.


1000000 -

1003

10000
E
100
10


0 10 20 30 40 50 60
CUMULATIVE PERCENTAGE

Figure 2.14. Cumulative percentage of depression area.


70 80 90 100


Depth of Karstic Depressions


Depression depth is considered as an important parameter in distinguishing

landforms of temperate and tropical karst terranes. In their comparative study on various

karst regions, Troester et al. (1984) concluded that depression depth distributions on

temperate and tropical karst regions could be explained by an exponential equation as

follows:

(number of sinkholes) = Noe-Kd








where N and K are constants, d is depth. The No coefficient is affected by the number of

depressions, whereas the K coefficient varies within ranges corresponding to temperate

and tropical karst areas. Comparing the internal relief within various karst regions, they

found that the Florida karst is very flat, represented by broad, shallow depressions with a

K value of 0.362 ft1.

The depth frequency distribution for 23,031 depressions in the study area is

shown in Figure 2.15. Depth values were estimated by counting the number of contours,

representing the maximum depth values for depressions. The depth frequency

distribution was found to change exponentially with depth, expressed as:

number of depressions = 137292e357'. It should be noted that the analysis of a much

larger population of depressions in this study gives a K value of 0.357 that is similar to

0.362 calculated by Troester et al. (1984) for the Florida karst.




100000

10000* -o
_y = 137292e'-'
R7 2 0.9895
1000
00-
0)
-0
E
z 10

1 - ... I
0 5 10 15 20 25 30
Depth (ft)

Figure 2.15. Depth-frequency distribution for the depressions.








Discussion


The landforms of the temperate Suwannee River karst area can be compared to

karst areas of different morphoclimatic settings. The application of Geographic

Information Systems allows the capture, storage, analysis, and presentation of

morphometric and spatial distribution parameters of karstic depressions. Despite the

accurate, rapid, and objective data processing of GIS, it should be kept in mind that all the

analyses performed on karstic depressions are based on the resolution and precision of

standard 7.5-minute USGS topographic maps with a contour interval of 5 feet. Karstic

depressions in the study area may have been underrepresented because of the limited map

resolution and the thick vegetation and sediment cover greatly reducing the overall relief

of the terrain. Nevertheless, this study provides general information on the morphometry

and spatial distribution of karst depressions on a regional scale.

Overall spatial distribution parameters of the Suwannee River basin karst area are

summarized as follows:

Total area: 4063.138 sq.km.

Total depression area: 331.39 sq.km.

Number of depressions: 25157

Average depression density: 6.07 depressions/sq.km.

Average nearest neighbor distance: 186.09 m

Average nearest neighbor distance index: 0.854

Average pitting index: 12.2








A statistical summary of the morphometric parameters of karstic depressions is

given in Table 2.2.


Table 2.2. Statistical summary of morphometric parameters
Area Mean Diameter L/W Circularity
(sq.m) (m) Index
Mean 14571.4 98.6 1.74 1.33
Maximum 5,804,750.0 2785.9 14.1 8.86
Minimum 75.0 10.1 1.0 0.006
Standard Deviation 71355.2 106.6 0.68 0.455



Table 2.3 compares the average spatial distribution parameters obtained in this

study with those of various karst regions from diverse morphoclimatic settings. This study,

with the largest number of depressions examined, provides an accurate representation of

the karst landforms of the subtropical Florida karst. Mean depression density is

significantly lower than those calculated for other karst regions such as Papua New

Guinea, New Zealand, Guatemala, Jamaica, and Puerto Rico. Despite comparable

tectonic, lithologic and climatic conditions with the Yucatan karst, the Florida karst is

represented by much higher depression densities. This may be resulted from the chemical

aggressiveness of recharge waters enhanced by the siliciclastic overburden in Florida.

It should also be noted that the values for depression density and the nearest

neighbor index calculated for the study area show great variations. Depression densities

calculated for individual quadrangles range from 2.43 (Madison SE) to 15.8 (Mayo SE)

overlapping with values representing both temperate and tropical karst areas. Similarly,

the nearest neighbor index varying from 0.578 (Hillcoat) to 1.025 (Mayo NE) indicates a








wide range of spatial distribution patterns of depressions in the study area ranging from

clustered to uniform. These significant variations in spatial distribution parameters along

with the morphometry of depressions in a single morphoclimatic zone implicate the

existence of some other local factors (e.g. thickness of the overburden above the Floridan

aquifer, potentiometric level fluctuation of the Floridan aquifer) controlling the

development of karst infrastructure.



Table 2.3. Depression density and nearest neighbor statistics for various karst areas
(modified after Ford and Williams, 1989).
AREA NUMBER OF DENSITY NEAREST PATTERN
DEPRESSIONS NEIGHBOR
INDEX
Papua New Guinea 1128 10-22.1 1.091 1.404 Near randomto
approaching uniform
Waitomo, New Zealand 1930 55.3 1.1236 Near random
Yucatan (Carrillo 100 3.52 1.362 Approaching uniform
Puerto Formation)
Yucatan (Chicken Itza 25 3.15 0.987 Near random
Formation)
Barbados 360 3.5 13.9 0.874 Tending to cluster
Antigua 45 0.39 0.533 Clustered
Guatemala 524 13.1 1.217 Approaching uniform
Belize 203 9.7 1.193 Approaching uniform
Guadelope 123 11.2 1.154 Near random
Jamaica (Browns 301 12.5 1.246 Approaching uniform
Town-Waldeston
Formation)
Jamaica (Swanswick 273 12.4 1.275 Approaching uniform
Formation)
Puerto Rico (Lares 459 15.3 1.141 Near random
Formation)
Puerto Rico (Aguada 122 8.7 1.124 Near random
Formation)
Guangxi, China 566 1.96 6.51 1.60 1.67 Approaching uniform
Spain, Sierra de Segura 817 18 80 1.66 2.14 Near uniform
Florida, Suwannee 25,157 6.06 0.854 Tending to cluster













CHAPTER 3
FACTORS AFFECTING DEPRESSION MORPHOMETRY AND
DISTRIBUTION


Prospectus


Morphometic features and spatial distribution of karstic depressions may be

controlled not only by global variables such as climate, but also by local factors involved

with the hydrology, hydrogeology, structural geology, and the stratigraphy of the karst

terrain. These factors, brought about by the geologic setting of the area, are collectively

responsible for the unique evolutionary patterns of individual karst terrains. In most

cases, their influence predominates over global variables and may lead to karst landform

developments distinct from those at nearby areas of the same climatic belt. Therefore,

investigation of their effects on the morphometry and spatial distribution of karstic

depressions is imperative in understanding karst landform development in a given area.

Ford (1964) analyzed the spatial distribution of dolines in the Mendip Hills karst

area and found a relationship between the doline occurrence and topographic setting. He

reported that eighty per cent of the dolines are located within narrow valley floors. He

also showed that there was not a significant relationship between doline development and

cave passages. La Valle (1967, 1968) related doline development in the Central Kentucky

karst to geologic variables using multiple regression analysis. He found that the

elongation of 25% of the dolines could be controlled by structural and topographic








factors. Palmer and Palmer (1975) reported correlation between doline development and

underlying cave formation in the southern Indiana karst.

Kemmerly (1976) statistically compared two random samples of dolines in

Tennessee based on the agreement of their long axis orientation with systematic joint

sets. He found polymodal preferred doline long-axis orientations correlating with the

three systematic joint sets in the study area. In a later study on the temporal distribution

of doline occurrence, Kemmerly (1980) explained a positive correlation between monthly

collapse rates and deviation of rainfall from historical means.

Structural control on the orientations of dolines in the karst of West Virginia was

reported by both Ogden and Reger (1977) and Wigal (1978) who also related the doline

density to lithologic factors. They found the highest doline occurrence on pure,

cavernous, thick-bedded limestones.

White and White (1979) used factor analysis to relate hydrologic and geologic

measures to doline development in the Appalachian Highlands. They found rock type as

an important factor controlling the dimensions and frequency of dolines.

Performing a regression analysis on the spatial distribution of dolines in the

glacial-mantled karst of Iowa, Palmquist (1979) was able to relate doline occurrence to

topographic factors such as relief, gradient and elevation; hydrologic factors such as

proximity to a drainage way; and to the thickness of soil cover. He concluded that the

doline initiation and enlargement in the study area were controlled by regolith thickness

and by the amount of groundwater recharge and secondary porosity, respectively.

In an attempt to determine the role of material properties in the development of

tropical karst landforms, Day (1982) analyzed purity, petrographic character, and








hardness of rocks representing 13 areas of tropical karst in the Caribbean and Central

America. He reported a negative correlation between limestone purity and rock hardness

determined by in situ measurements of compressive strength using a Schmidt Test

Hammer. In their study on the collapse sinkholes in Puerto Rico, Soto and Morales

(1984) noted a possible structural control on the development of sinkholes whose long

axis orientations showed a significant concentration in northeast direction.

Karstic depressions on the flat-lying dolomitic rocks of the southeastern

Minnesota was systematically mapped and analyzed by Dalgleish and Alexander (1984).

They found the bedrock stratigraphy to be the primary control on the distribution of

sinkholes. Aunong the secondary controls were slope of the land surface, and composition

of surficial materials.

Based on an inventory of modern sinkhole occurrences in west central Florida,

Littlefield et al. (1984) reported that sinkhole development commonly occurred along

linear patterns especially along the largest photolinear features. In a similar study by

Beggs and Ruth (1984), modern sinkhole occurrences documented by the Florida

Department of Transportation were correlated with actual rainfall data and geologic

controls in the study area. They found no relationship between the rainfall data and

sinkhole occurrences that seemed to be controlled by the depth to limerock and phreatic

surface.

The spatial distribution of modern sinkholes in west central Florida was

statistically compared by Upchurch and Littlefield (1987) with that of ancient karstic

depressions identified on topographic maps and aerial photographs. They concluded that








the distribution of ancient karstic depressions in areas of uncovered karst significantly

predicted the locations of modern sinkholes.

Schmidt and Scott (1984) indicated the importance of structural features such as

the Ocala Uplift and Sanford High on the spatial distribution and morphometric features

of karstic depressions in Florida. In a detailed study based on the reported sinkhole

inventory of the former Florida Sinkhole Research Institute, Wilson and Beck (1992)

attempted to relate new sinkhole occurrence in Orlando area to a variety of hydrologic

and geologic data. They reported new sinkholes occurring in high recharge areas because

the potentiometric level of the Floridan aquifer decreased below its mode value. No

relationship existed between photolinears and spatial distribution of sinkholes, mainly

because the thick mantle deposits prevented most of the fractures from being detected as

photolinears.

This chapter explains and discusses the procedure and the results of an attempt to

investigate the effect of hydrogeologic factors on the karst geomorphology of the

Suwannee River basin. Since the entire analysis was performed in a GIS environment, the

selection of factors addressed in this discussion is constrained by the availability of data

in digital format. Yet, the digital data layers utilized in this study cover a variety of

hydrologic and geologic factors that may have played significant roles in the regional

karst landform evolution. It should be noted, however, the precision of GIS analysis

performed here depends upon the accuracy of the digital data layers, which is a direct

function of their spatial resolution.

Factors considered in this regional analysis include; thickness of the overburden

material above the karstic limestones of the Floridan aquifer, depth to water table, soil








type, potentiometric level fluctuation, lithology of the vadose zone, and fracture traces.

Spatial and morphometric parameters of depressions include depression density, length,

width, length-width ratio, average diameter, area and perimeter.

Methodology


The procedure followed in this analysis consists of a series of GIS operations on

the karstic depressions database and the depression density layer described in Chapter 2.

These two GIS layers were compared with rasterized data layers representing the

geologic and hydrologic controls within the study area. The morphometric data

summarized within various categories of control layers were transferred to spreadsheet

format in which statistical summaries of morphometric parameters were obtained. Data

on the spatial distribution of depressions were utilized by summarizing the depression

density layer (Figure 2.3) within various categories of controlling raster layers. Statistical

summaries of the depression density layer were expressed by the histograms prepared

within the Spatial Analyst extension of the ArcView 3.0 software.

The thickness of the overburden layer was created by applying map algebra

functions of GRID, a module of ARC/INFO for the analysis of rasterized data. Depth to

bedrock (limestone) layer was subtracted from the topographic elevation layer in order to

obtain the thickness values of the overburden material covering the karstic limestone

surface within the study area. The resultant layer has a spatial resolution of 62.8 m (side

length of the square-shaped cells of GRID layers).

The fluctuation of the Floridan aquifer potentiometric level is represented by a

GRID layer created from a digitized contour map prepared by the Suwannee River Water

Management District. It represents the total fluctuation between the maximum and








minimum levels of the Floridan aquifer potentiometric level as of 12/1988. The layer has

a cell size of 200 m

The rest of the GRID layers representing the depth to water table, soil types, and

the vadose zone lithology were converted from the vector layers created by the Suwannee

River Water Management District as sublayers of the DRASTIC system. DRASTIC is a

standardized system for evaluating ground water pollution potential using hydrogeologic

settings (EPA, 1985). The word "DRASTIC" is an acronym representing the most

important mappable factors that control the ground water pollution potential; Depth to

water, net Recharge, Aquifer media, Soil media, Topography (slope), Impact of the

vadose zone, Conductivity (hydraulic) of the aquifer. Among these factors only the depth

to water, impact of the vadose zone, and the soil media sublayers show significant

variation in the study area so as to affect the distribution of morphometry and density of

karstic depressions. These layers were converted to GRIDS with 50 m cell size and their

relationships with spatial and morphometric distribution of karstic depressions was

analyzed by using the Spatial Analyst extension of the ArcView 3.0 program.

Tectonic control on the development of karst landforms was investigated

comparing fracture traces with the major axis orientations and nearest neighbor vectors of

karstic depressions compiled in two different GIS databases presented in Chapter 2.

In addition to the karstic depressions depicted on the 1/24,000 scale topographic

quads, reported sinkholes compiled in the sinkhole database of the Florida Geological

Survey were also analyzed in this study. Despite its bias towards the inhabited areas, the

reported sinkhole database provides insight to the current karst development within the








study area. Their distribution with regard to the above mentioned geologic controls was

also analyzed within the GIS environment.

Results


Lithology of the Vadose Zone

With a cell size of 50 m, the layer representing the vadose zone lithology displays

the distribution of silt/clay, shale, sand and gravel, and karstic limestone as the lithologies

comprising the vadose zone in the study area (Figure 3.1). The predominant vadose zone

lithology within the Suwannee River basin is comprised of silt/clay as indicated by the

DRASTIC GIS layer. However, most of the karstic depressions selected for this study are

underlain by sand and gravel. The distribution of depression density within vadose zone

lithologies is given in Figure 3.2. Percentages of areas corresponding to depression

density ranges are in Table 3.1.


Table 3.1. Percentages of areas corresponding to depression density ranges for the vadose
zone lithologies.
Depression Silt/Clay Shale Sand and gravel Sand and Karstic
Density (1) (4) with significant silt gravel limestone
(#/sq.km) and clay (5&6) (8) (9)
0-5 73.2% 20.8% 44.5% 33.1% 19.2%
5-10 24.4% 41.6% 44.5% 37.3% 57.7%
10-15 2.4% 37.5% 11% 20.8% 19.2%
15-20 7.4% 3.8%
20-25 1 I I 1 1.4%


















































9 0 9 18 Kilometers


' Major rivers
Itology of the vadose zone
=Silt/Clay
7Shale
Sand and gravel with significant slit and clay
Sand and gravel
Karstlc limestone


Figure 3. 1. Distribution of vadose zone lithology within the study area.











12 T
Karst limestone


-4Sand and gravel
SShale

w Sand and gravel
0
Z 6 with significant silt
0 an jca
0
w Silt/clay
a-4
z
w 2



1 4 5&6 8 9
VADOSE ZONE LITHOLOGY

Figure 3.2. Distribution of depression density within the vadose zone lithologies



Thickness of the Overburden Material

Thickness of the overburden material ranges from zero within the Gulf Coastal

Lowlands to 156.48 m at the Northern Highlands where thick deposits of the Hawthorn

Group overlie the Floridan aquifer (Figure 3.3). It tends to increase with the topographic

elevation. Mean thickness value is 19.32 m, with a standard deviation of 21.5.

A greater part of the study area is overlain by a thin soil layer of 0-6 m in

thickness. In this area, the Floridan aquifer is unconfined and receives significant

recharge through diffuse infiltration of the surface runoff. Semiconfined hydrogeologic

conditions occur where the overlying material gets thicker and contains clay layers of the

Hawthorn Group. These local semiconfined conditions are mostly encountered close to

the Cody Scarp, where impermeable layers within the outliers of the retreating Hawthorn
































/\I Major rivers
Thickness
_0-20 ft (0-6.1 m)
20-30 ft (6.1-9.1 m)
S30-40 ft (9.1-12.2 m)
40-50 ft (12.2-15.2 m)
M 50-60 ft (15.2-18.3 m)
60-70 ft (18.3-21.3 m)
S70-100 ft (21.3-30.5 m)
>100 ft (>30.5 m)


8 0 8 16 Kilometers


Figure 3.3. Distribution of overburden thickness within the study area.










Group sediments cover the Floridan aquifer. At the Northern Highlands physiographic


region, thickness of the overburden material increases significantly to the northeast from


30 m to approximately 155 m.


The spatial distribution of karstic depressions within the zones of overburden


thickness is given in Figure 3.4. Depression density tends to remain between 7 to 8


depressions per sq. km. within the 0-24.4m (0-80 ft) overburden thickness range. A


significant decrease in the number of depressions is observed for greater thickness values


of the overburden material. Mean depression density drastically goes down to 2.3


depressions per sq.km where thickness of the overlying material is greater than


approximately 35 m (120 ft). This results from the confined conditions brought about by


the increased silt/clay content of the overburden material.



f 0-20 ft

8-20-50 ft 50-8 ft


6
z
WL 5
0
Z ~80-120Oft

(D)
,,,
wj 3-
ry 120-170 ft
W 2
Z

0 L
0-6.1 6.1-15.2 15.2-24.4 24.4-36.6 36.6-51.8
OVERBURDEN THICKNESS (m)



Figure 3.4. Changes in depression density within the zones of the overburden thickness.








The distribution of depression density within the zones of the overburden

thickness values is also expressed as percentages of areas corresponding to depression

density intervals in Table 3.2.


Table 3.2. Percentages of areas corresponding to depression density ranges for the zones
of the overburden thickness.
Depression Overburden Overburden Overburden Overburden
Density Thickness Thickness Thickness Thickness
(#/sq.kmn) 0-6.1 m 6.1 15.2m 15.2- 24.4 m 24.4- 36.5 m
(0-20 ft) (20-50 ft) (50-80 ft) (80-120 ft)
0-5 36.5% 36.7% 37.2% 75.5%
5-10 36.5% 38.3% 33.3% 18.9%
10-15 18.2% 18.2% 24.4% 5.6%
15-20 7.3% 5.8% 5.1%
20-25 1.5% 1.0%


Distribution of morphometric parameters of karstic depressions can not be

explained solely by the variation of the overburden thickness. A collective interpretation

of the overburden thickness variation together with the type of vadose zone lithology

provides a better understanding of the morphometric distribution of depressions. Figure

3.5 summarizes the lithological variation within the zones of the overburden thickness.

Mean depression area varies only between 13179.9 to 15763.9 sq.m. for the values of

overburden thickness up to 21 m (Figure 3.6). With further increase in overburden

thickness, depression area is controlled by the lithologic properties of the overlying unit.

Mean depression area reaches a maximum of approximately 23,500 sq.m., for the

overburden thickness zone of 21 to 30 m corresponding to the outliers of the

semipermeable Hawthorn Group within the Marginal Zone. It then significantly

decreases for thickness values greater than 30 m owing to the elevated silt and clay

contents of the overlying unit.











100% -- ----

W
it 80%k X

n: U. Karstic lmst.
ccX
80% -E Sand/gravel

0 M- Sand/gravel w/silt
B Shale

0 Silt/clay
w
a 20% -

0%


0-6.1 6.1-9.1 9.1- 12.2- 15.2- 18.3- 21.3- >30.5
12.2 15.2 18.3 21.3 30.5
THICKNESS OF THE OVERBURDEN MATERIAL (m)




Figure 3.5. Changes in vadose zone lithology with the thickness of the overburden.






25000


,ajooo71 ft10
2000 01Cf
50-60 ft
W 20-ft 60-70 ft
< 15000 0-20 ft 30-40 ft 40-50 ft
0
UJ)
w 1=0o >100 ft
w
z
w



0-61 6.1- 9.1- 12.2- 15.2- 18.3- 21.3- >30.5
9.1 12.2 15.2 18.3 21.3 30.5
OVERBURDEN THICKNESS (in)

Figure 3.6. Changes in mean depression area with the overburden thickness.






59



Mean values of depression diameter and length/width ratio show even less


variation within the zones of overburden thickness (Figures 3.7 & 3.8). Smaller


length/width ratios observed for thickness values grater than 30 m indicate a tendency for


elongation of depressions as the thickness and cohesiveness of the overburden material


increase.


140 .. .. . .
12070-100 ft
wu 120--
20-30 ft 30-0 ft 6 0-70 f
10C 1 0-20 ft t 640-50fft> an


z 80 i
0
60 "
w
(n 40

Z 20 -

2 0 i - -
0-6.1 6.1- 9.1- 12.2- 15.2- 18.3- 21.3- >30.5
9.1 12,2 15.2 18.3 21.3 30.5
OVERBURDEN THICKNESS (M)


Figure 3.7. Changes in mean depression diameter with the overburden thickness.




S1.8
0-20 ft
1.75- 050 ft
40 ft 50-60 ft 60-70 ft 70-100 ft
I- 1.7
z
w 1,65
-0
S1.6

1.55- >100 ft
a- 1,5
w
0
z 1.45
Uj 1.4 -. -. I H -I- --A I
0-6.1 6.1- 9.1- 12.2- 15.2- 18.3- 21.3- >30.5
9.1 12.2 15.2 18.3 21.3 30.5
OVERBURDEN THICKNESS (m)


Figure 3.8. Changes in mean length/width ratio with the overburden thickness.








Floridan Aquifer Potentiometric Level Fluctuation

Potentiometric level fluctuation values of the Floridan aquifer range from 1.25 m

at the western part of the study area to 11.77 m at north, with a mean value of 3.65 m.

The standard deviation is 2.08 (Figure 3.9). A complicated surface water ground water

interaction through karstic conduits explained in Chapter 1 is also revealed by the high

potentiometric level fluctuation along the Suwannee River. Regions of minimum

potentiometric level fluctuations west of the Suwannee River correspond to the discharge

area of the Floridan aquifer.

The spatial distribution of karstic depressions within the zones of potentiometric

level fluctuation approaches a normal distribution curve (Figure 3.10). Depression

density values generally increasing with the potentiometric level fluctuation reach their

maximum value for the fluctuation interval of 4.3 5.2 m (14-17 ft) and gradually

declines for greater fluctuation rates.

The distribution of depression density ranges within the zones of the

potentiometric level fluctuation is expressed as percentages in Table 3.3.


Table 3.3. Percentages of areas corresponding to depression density ranges for the zones
of Floridan aquifer potentiometric level fluctuations.
Depression Flue. Flue. Flue. Flue. Flue. Flue. Flue. Flue. Flue.
Density 1.2-2.1 m 2.1-3.0 m 3.0-4.3 m 4.3-5.2 m 5.2-6.4 m 6.4-7.3 m 7.3-8.5 m 8.5-9.4m 9.4-10.6 m
(#/sq.km) (4-71) (7-10 It) (10-14 ft) (14-17 ft) (17-21 ft) (21-24t) (24-28 ft) (28-31 ft) (31-35 ft)
0-5 64.3% 33.4% 30-5% 21.5% 31.5% 64.8% 71.2% 100% 81.6%
5-10 33.3% 35.5% 38.2% 39.3% 44.3% 32.4% 25.4% 18.4%
10-15 2.4% 20.7% 22.0% 27.7% 10.8% 2.7% 2.8% -
15-20 7.40% 8.2% 9.1% 3.3% 0
20-25 2. 1 % 1. 1 % 2.5% --
25-30 0.5% -o
30-35 0.3%






































. ~'Rivers
Pot Level Fluctuation
4- 8 ft (1.2-2.4 m)
m8 12 t (2.4-3.6 m)
12 17 ft(3.6-5.2 m)
ME 17- 21 ft (5.2-6.4 m)
21 26 ft (6.4-7.9 m)
26 30 ft (7.9-9.1 m)
30 35 ft (9.1-9.4 m)


7 0 7 14 Kilometers


3.9. Potentiometric level fluctuation of the Floridan aquifer.





62











10
14-17 ft
Cr 7-10ft 10-14ft
_V,
8 -17-21 ft
: 7 -
w 6
a 21-24 ft
Z 5- 4-7 ft
0 24-28 ft
4-2-
w 31-35 ft
CTUTUTO 28-31 ft
0 2-
Z

0 .. ....i - , i ,1 ... I ...... i ..
1.2-2.1 2.1-3.0 3.0-4.3 4.3-5.2 5.2-6.4 6.4-7.3 7.3-8.5 8.5-9.4 9.4-10.6
POTENTIOMETRIC LEVEL FLUCTUATION (m)


Figure 3.10. Changes in mean depression density with the potentiometric fluctuation.



Distribution of mean depression area within the zones of potentiometric level

fluctuation shows a unimodal pattern with a major mode for potentiometric level

fluctuations between 6.4 to 9.1 m (Figure 3.11). For fluctuations greater than 9.1 m, mean

depression area decreases significantly because of the confined hydrogeologic conditions.

Mean depression diameter reaches its maximum of 127.9 m within the fluctuation

interval of 7.9 9.1 m (Figure 3.12). Variation of mean length/width ratio indicates an

increase in elongation of depressions with increasing fluctuation values (Figure 3.13). It

tends to decrease for fluctuation values greater than 7.9 m.

















30000


25000


20000


15000


10000


500


0


26-30 ft


1.2-2.4 2.4-3.6 3.6-5.2 5.2-6.4 6.4-7.9 7.9-9.1 9.1-9.4
POTENTIOMETRIC LEVEL FLUCTUATION (M)


Figure 3.11. Changes in mean depression area with potentiometric fluctation.


1 2-2.4 2.4-3.6 3.6-5.2 5.2-6.4 6.4-7.9 7.9-9.1 9.1-9.4
POTENTIOMETRIC LEVEL FLUCTUATION (M)


Figure 3.12. Changes in mean depression diameter with potentiometric fluctation.

















1.9
21-26 ft
o 1.85 26-30 ft


S 1.30-35 ft
z 17-21 ft
0 1.75 8-12ft1
n 1.7 4-8ft

,
o_ 1.65
0<
a
z
1.6

1.55 -.- -- -i i r- -F -
1.2-2.4 2.4-3.6 3.6-5.2 5.2-6.4 6.4-7,9 7.9-9.1 9,1-9.4
POTENTIOMETRIC LEVEL FLUCTUATION (M)

Figure 3.13. Changes in mean length/width ratio with potentiometric fluctation.




Soil Type


The soil media layer with a cell size of 50 m displays the distribution of a variety


of soil types such as nonshrinking and/or nonaggregated clay, peat, muck, (soil type = 1);


sandy loam (soil type = 6); shrinking and/or aggregated clay (soil type = 7); and, sand


(soil type = 9)(Figure 3.14). Greater part of the Suwannee River basin is covered by

"nonshrinking and nonaggregated clay, peat, and muck". However, within the study area,


"sand" and "shrinking and/or aggregated clay" categories occur as dominant soil types.


Values of depression density are summarized within the zones of soil media in


Figure 3.15. In soil type 7 depicted as "shrinking and/or aggregated clay", mean karstic


depression density peaks at 8.6 depressions per sq.km. The "sand" category is also


represented by a high density of depressions (-7.5 dep./sq.km.).















2. ~ ... .:-:







:7X.



x..
=. .=. ..=. . . . . .. .


...... . ...... ....... .... .. i...- i!:. ii;,i /










//Rivers
::::..:% : i.-i.,- .........






Soil Type
Nonshrinking and nonaggregated clay, peat, muck
SSilty sandy loam
~ Sandy loam
II Shrinking and/or aggregated clay
Sand


Figure 3.14. Distribution of soil types within the study area.





66






9 Shrinking and/or
ag regate c y
8" 8 Sand
7-
Sandy loam
I-6
z
WU Nonshrinking and
z nonaggregated clay,
0 peat, muck
C) 4
LU
Q- 3

< 2
Lu

10
0 I-- - H --}. .....
1 6 SOIL TYPES 7 9

Figure 3.15. Changes in mean depression density with soil types.



The distribution of depression density categories within the zones of soil types is

summarized by the percentage of corresponding area in Table 3.4.

Table 3.4. Percentages of areas corresponding to depression density ranges for the soil
types.
Depression Soil type: I Soil type: 6 Soil type: 7 Soil type: 9
Density (#/sq.km)
0-5 75.0% 69.0% 25.7% 37.1%
5-10 23.1% 31.0% 42.6% 35.8%
10-15 1.9% 22.3% 20%
15-20 8.1% 6.1%
20-25 1.3% 1.0%


Evaluation of morphometric features of karstic depressions for each soil category

reveals that the depression area and mean depression diameter are greatest for the soil

type 1 (nonshrinking and nonaggregated clay, peat, muck) (Figures 3.16, 3.17 & 3.18).


















25000




0
U,





55 15000-
w,

0 10000-
z

15ODO







0-
16 7
SOIL TYPE


Figure 3.16. Changes in mean depression area with soil types.


1 6 7
SOIL TYPE


Figure 3.17. Changes in mean depression diameter with soil types.






68






Areas with soil types 9 and 7 (sand, and shrinking and/or aggregated clay) have


the most irregular depression shapes, with the length/width ratios getting as high as 1.75.





0 2--

-i 2
--

---

o 2
z
-J
z 2
0

wa 2

0 2-
w
"o 2

2
1 6
SOIL TYPE


Figure 3.18. Changes in mean length/width ratio with soil types.




Depth to Water Table


The GRID layer of "depth to water table" has a cell size of 50 m as well. Ranges


of depth displayed in the map include; 0-4.5m (0-15 ft), 4.5-9.1 m (15-30 It), 9.1-15.2 m


(30-50 ft), 15.2-22.9 m (50-75 ft), 2.9-30.5 m (75-100 ft) and >30.5m (Figure 3.19).


Depth to water table within the Suwannee River basin varies between 0 to 4.5 m along


the coastal zone and the Suwannee River, and increases with the topographic elevation of


the area. The Northern Highlands physiographic province is represented by the deepest


water table conditions.





























+


a Rivers
pth to water table
0-15 ft (0-4.5 m)
15-30 ft (4.5-9.1 m)
30-50 ft (9.1-15.2 m)
50-75 ft (15.2-22.9 m)
M75-100 ft (22.9-30.5 m)
M >100 ft (>30.5 m)


10 0 10 20 Kilometers


Figure 3.19. Distribution of depth to water table within the study area.





70


Depressions are most densely located where depth to water table is between 9.1 to

15 m. (Figure 3.20). Depression density gradually decreases from around 9 to less than 3

depressions per sq.km for water table configurations deeper than 15 m. This trend is

correlated with the increase in silt/clay content of the vadose zone above the water table

(Figure 3.21). Low densities for water depth values greater than 20 m roughly correspond

to semiconfined and confined conditions of the Floridan aquifer where interstratal

dissolution and collapse sinkholes predominate over solution sinkholes.



1030-50 ft
9
LU 0-15 ft
7-- 15-30 ft 50-75ft
0
ul) c 5-
5 75-100 ft
Wi 3 --I
o2 III
z 2


0-4.6 4.6-9.1 9.1-15.2 15.2-22.7 22.7-30.5 >30.5
DEPTH TO WATER TABLE (M)

Figure 3.20. Changes in mean depression density with depth to water table.




100%-
I- 0 Karstic Imst.
,<, 80% I
"- 8 'Sand/gravel
< 0 < 60%
IJ 0 Sand/gravel w/silt,clay
U 40%- ISShale
0 >
0 20%,- 0 Silt/clay
0%.

0-4.6 4.6- 9.1- 15.2- 22.7- >30.5
9.1 15.2 22.7 30.5
DEPTH TO WATER TABLE (M)

Figure 3.21. Changes in vadose zone lithology with depth to water table.












Table 3.5 shows the percent distribution of areas corresponding to depression

density intervals within the zones of the depth to water table.


Table 3.5. Percentages of areas corresponding to depression density ranges for the zones
of the depth to water table.
Depression Depth to Depth to Depth to Depth to Depth to Depth to
Density water: water: water: water: water: water:
(#/sq.km) 0-4.6 m 4.6-9.1 m 9.1-15.2 m 15.2-22.8 m 22.8-30.5 m >30.5 m
(0-15 t) (15-30 ft) (30-50 t) (50-75 ft) (75-100 ft) (>100 ift)
0-5 43.0 40.0 24.4 44.3 87.5 80
5-10 39.3 40.2 35.1 36.1 12.5 20
10-15 14.2 16.8 26.9 19.4 -
15-20 4.9 2.9 11.1 3.3
20-25 1.12 0.4 2.6 -
25-30 0.4 -
30-35 0.2 -


In areas where water table is less than 30.4 m deep, mean depression area

increases with depth to water table (Figure 3.22). The decrease in mean depression area

observed for deeper water table depths is caused by the thick and cohesive overburden

layer within the Northern Highlands physiographic region. The variation of mean

depression diameter within the zones of depth to water table tends to increase with depth

(Figure 3.23), whereas mean length/width ratio of depressions approaches unity as the

depth to water table increases (Figure 3.24). In other words, depressions become more

equidimensional as the water table gets deeper.














25000

20000

15000

10000


5000 -


+ ..........---... .... -L -


50-100 ft


15-30 ft


0-15 ft


30-50 ft


0-4.6 4.6-9.1 9.1-15.2 15.2-30.5
DEPTH TO WATER (m)

Figure 3.22. Changes in mean depression area with depth to water table.


140

120

100

80

60

40

20
n


50-100 ft


15-30 ft


O-15ft


30-50 ft


>100 ft


>30.5


>100ft


0-4.6 4.6-9.1 9.1-15.2 15.2-30.5
DEPTH TO WATER (m)

Figure 3.23. Changes in mean depression diameter with depth to water table.


>30.5











1.80
1.78
1.76 15-3o ft
z 1.74
uJO
z 1.72 500 ft
r3o 3D-50 ft
,, 1.70 >100 ft
,' 1.68
< 1.66
Lu
1.64
0-4.6 4.6-9.1 9.1-15.2 15.2-30.5 >30.5
DEPTH TO WATER (m)

Figure 3.24. Changes in mean depression length/width ratio with depth to water table.





Photolineaments and Orientation of Depressions

The influence of fractures on karstic depression development was investigated by

overlaying the map of fracture traces detected as photolinears on the layer of depression

density (Figure 3.25). The digitized map of fracture traces compiled by the Florida

Geological Survey shows regional photolineaments with preferred orientations in NE-SW

and NW-SE directions.

Orientation of depressions was analyzed in two phases. The long axis orientations

of 24,757 depressions were measured in GIS and classified into eighteen 100 classes.

Alignment of depressions with regards to each other (nearest neighbor direction vector)

was investigated using Euclidian distance functions of GIS based on the centroid

locations of depressions.












































8 0 8 16 Kilometers


Figure 3 .25. Map of photolinears and depression density within the study area.










The distribution of long axis orientations of depressions does not indicate a


preferred direction (Figure 3.26). No attempts were made to statistically infer a difference


of one subclass of azimuths from others because of the apparent homogenous distribution


of their percent frequencies. The lack of dominant long axis orientation may be related to


the absence of controlling regional structural trends.


7.0



6.0


5.0


4.0
W

LL
I-
Z 3.0
W
0
n,-
W
2.0


u .u f I I I i 1 1 1 [
0-10 20-30 40-50 60-70 80-90 100-110 120-130 140-150 160-170
MAJOR AXIS AZIMUTH

Figure 3.26. Histogram of depression major axis orientations.




The distribution of nearest neighbor direction vectors, however, indicates a


preponderance of orientation for the azimuth range between 80 -100' (Figure 3.27).


Approximately 18% of all depressions have a nearest neighbor vector between 80 -100'.


The second most common nearest neighbor vector azimuth range occurs between 160 -


1800 with a percentage of about 15%.










18.0

16.0

14.0

Z 12.0
UJ
,,W 10.0
LL
t- 8.0 --
Z
U.JI
6.0
LU
4.0 --

2.0

0.0 L ____ __
0-20 40-60 80-100 120-140 160-180
NEAREST NEIGHBOR AZIMUTH

Figure 3.27. Histogram of nearest neighbor directions.




Collective interpretation and comparison of major axis and nearest neighbor


vector orientations with regional photolineaments do not reveal an apparent pattern of


preferred orientation common to both data sets. However, the implications of dominant


nearest neighbor vector orientations can be explained by considering the aspect of


topographic slope within the study area.


Topographic Slope


A majority of karstic depressions analyzed in this study occur at the relatively


flat-lying karst terranes of the Gulf Coastal Lowlands physiographic province (Figure


1.2). However, most karstic activity takes place within the Marginal Zone along the Cody


escarpment in which topographic relief is much more pronounced (Figure 3.28). The


importance of this area in current karst landform development can be explained by


addressing the chemical aggressiveness of allogenic surface waters originating from the


Northern Highlands. It should also be noted that increased topographic relief has been



















































/NdPM,8,ICr -vi"
-0.75 -0.50 Std. Dev.
-0.50 -0-.25 Std. Dev.
-0.25 0.00 Std. Dev.
Mean
0.00 0.25 Sd. Dev.
0.25 0.50 Std. Dev.
0.50 0.75 Sid. Dev.
0.75 1.00 Std. Dev.
1.00 1.25 Std. Dev.
1.25 1.50 Sid. Dev.
1.50 1.75 Std. Dev.
1.75-2.00 Std. 0ev.
100 2.25 Std. Dev.
125 2.50 Std. Dev.
.50 2.75 Sid. Dev.
175 3.00 Std. Dev.
> 3 Mfd. lDev.


10 0 10 20 Kilometers


Figure 3.28. Map of topographic slope within the study area.









instrumental in sustaining surface runoff essential for the formation and development of

solution sinkholes.

The preferred orientation of nearest neighbor vectors in an azimuth interval of 80-

100 degrees does not correspond to the regional fracture pattern. However, its close

agreement with the predominant E-W aspect of topographic slope might indicate that the

alignment of karstic depressions is controlled by the relief of local topography (Figure


3.29).


50
45-
w 40
< 35
I-
Z30-
X 25
UA
o 20
1 15
w
< 10-


FLAT N NE E SE S SW W NW
TOPOGRAPHIC SLOPE

Figure 3.29. Histogram of topographic slope direction distribution within the study area.




Reported Sinkholes


Based on the database compiled by the Florida Geological Survey, the distribution

of sinkholes reported in the study area between 1948 and 1992 is given in Figure 3.30.

Approximately 95% of all reported sinkholes located in the study area occur where the

Floridan aquifer is unconfined or semiconfined. However, since a great number of new

sinkholes may have been unnoticed or unreported especially at the rural parts of the study






































0 Reported sinkholes (between 1948-1992)
Cody escarpment
lCounty boundaries
SPhotolinears


9 0 9 18 Kilometers


Figure 3.30. Distribution of reported sinkholes within the study area.










area, actual population of new sinkholes can be much higher than is suggested by the

database. In order to eliminate the problem of under-reporting, Wilson (1995), based on

the types of land-use, recommends various adjustment factors for reported sinkhole

frequencies.

In this study, an overall morphometric analysis of reported sinkholes is attempted

as they are described in the database. Because of the problem of under-reporting,

frequency and density distribution of new sinkhole formation is not addressed. From the

morphometric standpoint, new sinkholes are strikingly different from closed depressions

depicted in topographic maps. They tend to be much smaller and circular in shape than

those karstic depressions whose dimensions are constrained by the map scale and

changed over time due to erosion and coalescing sinkholes. Structural control on new

sinkhole development is clearly seen especially at the south and southeast of the area

where sinkholes are aligned along NW-SE photolineaments (Figure 3.30). The influence

of E-W topographic slope is also suggested especially close to high relief areas.

Distribution of reported sinkholes with regards to the vadose zone reveals that

almost 80% of the 163 sinkholes occur within sand and gravel (Figure 3.31). Despite its

most wide-spread coverage within the study area, the silt/clay vadose zone lithology

hosts only 6.1% of the total reported sinkholes. The mean diameter shows a relatively

homogenous distribution except for the shale lithology in which the mean diameter is

close to 10 meters within a population of only three sinkholes (Figure 3.32) Sinkholes

occurring within karstic limestone, i.e. with a little or no overburden, have the smallest

mean diameter.
















60

50

40


20

10


1: silt/clay
4: shale
6: sand and gravel with silt/clay
8: sand and gravel
9: karstic limestone


6
TYPE OF VADOSE ZONE


Figure 3.3 1. Changes in reported sinkhole frequency with vadose zone lithology.


10 m


1: silt/clay
4: shale
6: sand and gravel
8: sand and gravel
9: karstic limestonE


vith silt/4lay


--I


6
TYPE OF VADOSE ZONE


Figure 3.32.
sinkholes.


Changes in mean diameter with vadose zone lithology for reported










Similarly, sinkhole length/width ratio approaches the unity for karstic limestones

and increases with the sand and gravel content of the vadose zone material (Figure 3.33).

Mean depth values for different vadose zone types range from 1.88 to 6.79 (Figure 3.34).

Approximately 51% of the reported sinkholes occur within the 3.0-4.2 m

fluctuation range of the Floridan aquifer potentiometric level (Figure 3.35). The mean

diameter shows a little variation around 4 meters and reaches 11 meters for the 7.6-8.5 m

fluctuation range (Figure 3.36). The length/width ratio of reported sinkholes is normally

distributed within the zones of the potentiometric level fluctuation (Figure 3.37).

Sinkholes occurring within the 6.7-7.6 m fluctuation interval represent the maximum

length/width ratio in the study area. The ratio approaches unity for higher values of

fluctuation. Similarly, mean depth values slightly varying around 3 meters reach a

maximum of 7.7 for the 6.7-7.6 m fluctuation interval (Figure 3.38).

The distribution of reported sinkholes with regard to the depth to water table is

given in Figure 3.39. More than half of the sinkholes occur within the water table depth

range of 4.5 to 9.1 m. Areas with a depth to water table more than 30.4 meters have the

largest sinkholes (Figure 3.40), whereas water table depth range of 4.5 to 9.1 m

represents the deepest sinkholes (Figure 3.41). Length/width ratio within the zones of the

depth to water table does not show significant variation (Figure 3.42). Sinkholes with a

water table depth of 4.5 to 15.24 m are relatively irregular.











7

ZE6
-J
0
5

C,)
0


nL
EL

--
0
2-

0
Z
<


0-


4 6 8 9
TYPE OF VADOSE ZONE


Figure 3.33. Changes in mean depth with vadose zone lithology for reported sinkholes.


1: silt/clay
4: shale
6: sand and gravel
8: sand and gravel
9, karstic limestone


with silt4lay


4 6 89
TYPE OF VADOSE ZONE


Figure 3.34.
sinkholes.


Changes in mean length/width ratio with vadose zone lithology for reported


1: silt/clay
4: shale
6: sand and gravel
8: sand and gravel
9: karstic limestone


1.15


1.1
UI
0
V.05






0.95






































1.2-2.1 m 3.0-4.2 m 4.2-5.2 m 5.4-6.7 m 6.7-7.6 m 7.6-8.5 m 8.5-10.6 m
POTENTIOMETRIC LEVEL FLUCTUATION

Figure 3.35. Changes in reported sinkhole frequency with potentiomeric level fluctuation
of the Floridan aquifer.


u-j
7



<





2-

1.1


1.2-2.1 m 3.0-4.2 m


4.2-5.2 m 5.4-6.7 m 6.7-7.6 m
POTENTIOMETRIC LEVEL FLUCTUATION


__ __L


7.6-8.5 m 8.5-10.6 m


Figure 3.36. Changes in mean diameter with potentiomeric level fluctuation of the
Floridan aquifer for reported sinkholes.














1.3


1.25

',
01.2

z
~1.15



0L
I-

1.15

1.05-


1.2-2.1 m


3.0-4.2 m 4.2-5.2 m 5.4-6.7 m 6.7-7.6 m
POTENTIOMETRIC LEVEL FLUCTUATION


7.6-8.5 m


r-7


8.5-10.6 m


Figure 3.37. Changes in mean length/width ratio with potentiometric level fluctuation of
the Floridan aquifer for reported sinkholes.


7
LU
I-
r 6
0J j



02

z 3.
uJ


1.2-2.1 m 3,0-4.2 m 4.2-5.2 m 5.4-6.7 m 6.7-7.6 m
POTENTIOMETRIC LEVEL FLUCTUATION


__L__ __L


7.6-8.5 m 8.5-10.6 m


Figure 3.38. Changes in mean depth with potentiometric level fluctuation of the Floridan
aquifer for reported sinkholes.


n7


I I I ....-L. I m I ..... I I I I


































0-1.5 1.5-4.5 4.5-9.1 9.1-15.24 15.24-22.8 >30.4
DEPTH TO WATER TABLE (M)

Figure 3.39. Changes in reported sinkhole frequency with depth to water table.


5-
4.5
4-

3 5
w3-
w
0 2.5-
S2-
1.5-
1-
0.5-
0-


-~ -p-- -&


0-1.5 1.5-4.5 4.5-9.1 9.1-15.24 15-24-22.8 >30,4
DEPTH TO WATER TABLE (m)


Figure 3.40. Changes in mean diameter with depth to water table for reported sinkholes.














1.12
1.1

1.08-

1.06-

1.04-

1.02-

1-

0.98-

0.96-
r) O.A


0-1.5 1.5-4.5 4.5-9.1 9.1-15.24 15.24-22.8 >30.4
DEPTH TO WATER TABLE (M)


Figure 3.41. Changes in mean length/width ratio with depth to water table for reported
sinkholes.


4-

3.5-

3-

2.5 -
co
_ 2
z 1.5-
Uf)
1-

0.5-

0


-~


0-1.5 1.5-4.5 4.5-9.1 9.1-15.24 15.24-22.8 >30.4
DEPTH TO WATER TABLE (M)


Figure 3.42. Changes in mean depth with depth to water table for reported sinkholes.