Title Page
 Front Matter
 Table of Contents

Karst in Florida ( FGS: Special publication 29 )
Full Citation
Permanent Link: http://ufdc.ufl.edu/UF00000144/00001
 Material Information
Title: Karst in Florida ( FGS: Special publication 29 )
Series Title: ( FGS: Special publication 29 )
Physical Description: Book
 Record Information
Source Institution: University of Florida
Holding Location: University of Florida
Rights Management:
The author dedicated the work to the public domain by waiving all of his or her rights to the work worldwide under copyright law and all related or neighboring legal rights he or she had in the work, to the extent allowable by law.
Resource Identifier: ltqf - AAA0295
ltuf - AKM4746
alephbibnum - 002036986
System ID: UF00000144:00001


This item has the following downloads:

Binder1 ( PDF )

Table of Contents
    Title Page
        Page i
    Front Matter
        Page ii
        Page iii
        Page iv
    Table of Contents
        Page v
        Page vi
        Page vii
        Page viii
        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
        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
        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
        Page 97
        Page 98
        Page 99
        Page 100
        Page 101
Full Text

Elton J. Gissendanner, Executive Director

Art Wilde, Director

Walter Schmidt, Chief




Ed Lane

Published for the


LIBR4j ,


Secretary of State


Commissioner of Education

Attorney General


Commissioner of Agriculture

Executive Director


Bureau of Geology
Tallahassee, Florida
August, 1986

Governor Bob Graham, Chairman
Florida Department of Natural Resources
Tallahassee, Florida 32304

Dear Governor Graham:

The Bureau of Geology, Division of Resource Management, Depart-
ment of Natural Resources, is publishing as its Special Publication No.
29, Karst In Florida. This report explains the origins of Florida karst, gives
examples of its occurrence throughout the State, and discusses benefits,
hazards, and what can be done about it. This publication will be useful to
professionals in earth-science related fields, teachers, governmental
agencies, and the citizens of Florida.

Respectfully yours,

Walter Schmidt, Chief
Bureau of Geology

Printed for the
Florida Geological Survey


ISSN No. 0085-0640




Introduction .. .. .. .. . .... .. .... .. .. ... . .. .. 1
Metric Conversion Factors ............................. 2
Porosity and Permeability .............................. 2
Hydrologic Cycle and Aquifers .......................... 7
Evolution of Karst Terrain .............................. 11
Chemical Weathering of Carbonate Rocks .............. 12
Lowering of Land Surface .......................... 15
Karst Features in Florida ............................... 20
S inkholes ...................................... 2 1
W inter Park Sinkhole .............................. 25
Lakes .......................................... 28
W oodville Karst Plain .............................. 31
Big Dism al Sink ........................... 32
Lost Lake ................................ 32
Natural Bridge .............. ..... ....... . 32
W akulla Springs .......................... 36
Underground Rivers ..................... 40
C aves ............................... 4 1
Florida Caverns State Park .......... 41
Falling Waters State Recreation Area .. 53
S springs ........................................ 57
Submarine Springs ............................ 60
Q uicksand .................................. 63
W windows to the Past ........................... 65
Little Salt and Warm Mineral Springs ..... 65
Hornsby and Darby Springs ............ 70
Recreational Use ................................. 71
Deep Zones of High Transmissivity ................. ..... 72
Boulder Zone .................................... 76
Uses of Deep Zones of High Transmissivity ............. 80
Fresh-W ater Storage ........................ 80
Salt-W ater Injection ........................ 82
Acid Rain and Its Effect on Karst Formation ................ 82
Detection and Prediction ........................... ... 85
Direct Methods-Exploration and Drilling ............... 85
Indirect Methods-Remote Sensing ................... 86
Geophysical Methods ........................... 86
G ravity .................................. 86
Seismic Refraction ......................... 87
Ground Penetrating Radar .................... 87
Resistivity ............................... 92

Aerial Photography ............................ 94
Satellite Im agery .............................. 94
Summary .......................................... 95
References ........................................ 97


Figure Page

1 Karst limestone surface showing solution pipes ......... 3
2 House and car damaged by sinkhole ......... ........ 4
3 House falling into a sinkhole ................. ...... 5
4 Diagram illustrating porosity and permeability ........... 6
5 Hydrologic cycle ................................ 8
6 Stratigraphic sequence showing types of aquifers common
to Florida ....................................... 10
7 Evolution of karst landscape ................. 12,13,14,15
8 Limestone surface showing karst weathering ........... 16
9 Close-up of solution pipes ......................... 17
10 Map of Florida showing areas prone to sinkhole
develop ent ................................... 22
11 Diagram showing sinkhole formation ................. 23
12 Drill rig in sinkhole .............................. 24
13 W inter Park sinkhole ............................. 26
14 Map of Orlando area showing karst lakes .............. 27
15 Dry bed of Lake lamonia, 1981 ..................... 29
16 Dry bed of Lake lamonia, 1981 ..................... 30
17 Map of W oodville Karst Plain ....................... 31
18 Big Dismal Sink, location map ...................... 33
19 Big Dismal Sink, plan and cross section ............... 34
20 Big Dism al Sink ................................. 35
21 Lost Lake ..................................... 36
22 Natural Bridge, location map ....................... 37
23 Wakulla Springs, cross section ..................... 39
24 Florida Caverns State Park, location map .............. 42
25 Florida Caverns State Park, map of geologic features ..... 43
26 Stalactites .................................... 44
27 Cave columns .................................. 44
28 Dripstone column ............................... 45
29 Cave drapery formation ........................... 46

30 Calcite crystals ................................. 47
31 Newly discovered cave room, Florida Caverns State Park .48
32 Cave grotto with stream .......................... 49
33 Newly discovered cave room, Florida Caverns State Park .50
34 Cave colum n .................................. 51
35 Newly discovered cave room, Florida Caverns State Park 52
36 Waterfall at Falling Waters State Recreation Area ........ 53
37 Map of cave system associated with waterfall .......... 54
38 Location map of Florida's 27 first-magnitude springs ..... 55
39 Drainage basins of the Oklawaha River, and Silver and
Rainbow Springs ................................ 59
40 Graph showing interrelation of rainfall, water levels in the
Floridan Aquifer, and discharge of Silver Springs ........ 61
41 Submarine springs on the Floridan Plateau ............. 62
42 Quicksand-forming conditions ...................... 64
43 Quicksand in a stream bed and on a beach ............ 66
44 Little Salt and Warm Mineral Springs, location map ...... 68
45 Little Salt Spring, cross section ..................... 69
46 Hornsby and Darby Springs, location map ............. 70
47 Floridan Plateau with submarine karst features .......... 73
48 Geologic cross section across the Floridan-Bahama
Platform ...................................... 74
49 Boulder Zone, photograph showing porous dolomite ..... 77
50 Boulder Zone, down-hole photograph in W-6929 ........ 78
51 Cross section showing concept of geothermal convection
currents ...................................... 79
52 Map showing Boulder Zone injection wells ............. 81
53 Cross section showing oil production and brine injection in
south Florida .................................. 83
54 Field layout of a 12-channel seismograph ............. 88
55 Recording from a 12-channel seismograph ............. 89
56 Diagram of a ground penetrating radar system .......... 90
57 Ground penetrating radar traverse across a paleokarst site 91
58 Resistivity measuring configuration .................. 92
59 Map constructed from a resistivity survey ............. 93


Table Page

1 Ten of Florida's first-magnitude springs and discharges . 18
2 Calculated rates of surface lowering due to solution ...... 19
3 The 27 first-magnitude springs of Florida .............. 56
4 Water quality data for Little Salt
and W arm Mineral Springs ........................ 72


Ed Lane


Familiar features of surface drainage systems are streams, rivers, and
lakes (all interconnected) which cross the land and eventually discharge
into an ocean. In contrast, karst terrains have drainage systems that are
distinctly different from these surface drainage systems. Karst terrains
develop in areas underlain by carbonate rocks, primarily limestone and
dolomite, and have drainage which is manifested by sinkholes, springs,
caves, disappearing streams and underground drainage channels. Karst
topography is usually irregular due to the solution activity of acidic sur-
face and groundwaters, which dissolve the carbonate rocks, forming
cavities and allowing surficial sediments to collapse or subside.
Carbonates are a large group of minerals which have as a common
constituent the carbonate ion (CO3). When combined with other ele-
ments these carbonate ions form various carbonate minerals, of which
the three most common are calcite and aragonite (CaCO3), and dolomite
(CaMg(CO3)2). Calcite is by far the most abundant carbonate mineral. It
occurs as enormous and widespread sedimentary deposits in which it is
the predominant mineral. In pure limestones, some of which occur in
Florida, calcite makes up 98 to 100 percent of the rock. Practically all
carbonate rocks in Florida are limestone or dolomite, with limestone pre-
It has been estimated that limestones and dolomites constitute about
20 percent of all sedimentary rocks (Gilluly, et al., 1959), and that 5 to
10 percent of earth's land surface is karstic (Jackson, 1982). Because
carbonate rocks comprise such a large proportion of the rocks on or near
the earth's surface, karst terrains occur in many parts of the world.
The classic karst area, from which the name is derived, is the Karst
district of Yugoslavia, near the eastern shore of the Adriatic Sea. It is
nearly 100-miles wide in places, with the entire district approximating
the area of New York State. There, the limestone rocks are honeycombed
by tunnels and caverns, so that most of the drainage is underground.
Large sinkholes are abundant, some as deep as 600 feet. Streamless
valleys are common since streams often disappear into swallow holes
(American Geological Institute, 1962, p. 271). Streams tend to have
intermittent surface flow for short times after rain or snow melt. The
hummocky terrain is characterized by deeply eroded, isolated valleys and
steep-sided hills. These geomorphic features are so characteristic of the
Yugoslavian Karst district that the generic term karst terrain has been
universally applied to them, meaning terrain that has been shaped by
dissolution of the underlying carbonate rocks.
Karst regions in the United States include the caves and sinks region of


southern Indiana; central Kentucky and Tennessee, which has the Mam-
moth Cave system; the Carlsbad Cavern region of southern New Mexico;
the Appalachian Mountain's Great Valley limestone belt, which has Natu-
ral Bridge and Luray Caverns; and Florida's extensive karst plains, sink-
hole lakes and caves.
Why study karst? Figures 1, 2, and 3 show the significance of karst in
Florida and why concern is justified. An understanding of karst is impor-
tant to Floridians because Florida is almost entirely underlain by carbon-
ate rocks. Karst is more than an academic problem when one considers
that the surface of much of Florida's bedrock limestone probably resem-
bles Figure 1, if one could strip off all overburden. Florida's karst means
special problems which necessitate special considerations and precau-
tions. Planners at all levels need to be familiar with karst, from the private
citizen who plans to build a home to architects and engineers who design
and site buildings and government officials who issue permits for con-
struction or waste disposal. Florida's rapid population growth results in
more construction of roads, houses, and other facilities, increased need
for the safe disposal of all kinds of wastes, and increased demands on
the State's water resources for consumptive and non-consumptive uses.
All of these human activities place continually increasing stresses on the
environment, which poses a need to understand Florida's karst. This
publication will explore Florida karst: what causes it, specific examples
of it, benefits and hazards associated with it, and what can be done
about it.


The Florida Bureau of Geology, in order to prevent duplication of paren-
thetical conversions, inserts a tabular listing of conversion factors to
obtain metric units.
feet 0.03048 meters
square feet 0.092 square meters
miles 1.609 kilometers
square miles 2.590 square kilometers
cubic feet 2.8311 cubic meters
gallons 3.785 liters
tons 0.907 metric tons


Karst formation involves primarily the chemical weathering and ero-
sion of carbonate rocks. It is appropriate, therefore, to discuss factors
relating to and controlling the movement of underground water. The two
properties that are common to all rocks, and which control the move-
ment of underground water, are porosity and permeability.
Porosity and permeability are intimately related. A porous rock con-

.9 ~

i p


1 ii M- '' ,

: ,.* r
'- B -*-~**" 4' -"' *
~ ~ ~ ."F A- "''*-Jl*^ *,
-" --^--, ^^^^

1 *-i "- ^" "
i ~ ~ *i-flH '-*-i ^ .

Figure 1. Karst limestone surface showing honeycomb of round solu-
tion pipes. This surface was exposed when the overburden
was scraped off and the sands and clays plugging the pipes
were removed by water jets. Bedrock is Ocala limestone of
Eocene age in the abandoned Buda limerock mine off Route
41 between Newberry and High Springs. Picture taken about
1972 and used by permission of William A. Wisner, geologist,
Florida Department of Transportation.

-t ""

__ pi '~ 'V 0~

*. W,

A", ,M

96 fo

Fgr 2. H e a. cr d..

Figure 2. House and car damaged by sinkhole formation. Florida Geological Survey photo.

.::t,;* -^--wwnr~r ~ Wy"-
- fn # *: Y '?*< y
* -* ",' ^ 1


Figure 3. House going into a sinkhole. Florida Geological Survey photo.


as sand. In Figure A the porous and permeable sand is clean with open, interconnected voids
that allow water to move freely. In Figure B the same well-sorted, porous sand is rendered
impermeable to water flow due to the retarding effect of the interstitial material, such as clay.


tains voids; clean, well-sorted sand (having all grains approximately the
same size) or gravel are good examples. Permeability is a measure of a
rock's ability to allow fluids to move through its pore spaces. By defini-
tion, permeability implies that a rock's pore spaces must be intercon-
nected to allow fluids to move freely. A clean, well-sorted sand is said,
therefore, to be permeable; water can migrate through it (Figure 4a).
Porous rocks are not always permeable, however. A similar, well-sorted
sand may have its interstices filled with clay, small grains of organic
matter, or some other fine-grained material, which effectively blocks the
free passage of water (Figure 4b). In this case, the sand would be classi-
fied as impermeable (if no water could be transmitted) or as having low
permeability (if only insignificant quantities of water could pass through
Limestone, though usually thought of as being "solid" rock, often has
a granular texture and considerable porosity and permeability, either pri-
mary (developed when the limestone was deposited) or secondary
(developed after deposition). Groundwater flow through granular and
porous limestone is, therefore, similar to flow through sand. This is an
important concept to keep in mind during the following discussions of
aquifers and chemical weathering of limestone.


Only two things are necessary to create karst terrain: carbonate rocks
and slightly acidic water to attack them. Florida has an abundance of
both. Any discussion of karst revolves around water, its movement and
its interaction with carbonate rocks. An explanation of the hydrologic
cycle will allow a better understanding of the movement of water above,
on, and under the earth's surface. The hydrologic cycle is the name given
to the complex, ever-changing migration of groundwater, atmospheric
and surface waters.
The hydrologic cycle is driven by the elemental forces of sunshine and
gravity. Figure 5 shows the paths water may take as it moves through
the hydrologic cycle. Starting at the ocean, the sun's radiation heats the
ocean's surface and evaporates water, which is carried aloft by rising
convection currents of air, eventually forming clouds of water vapor. The
clouds are carried over the land by winds, where they drop their moisture
as rain, snow, or hail. Under usual atmospheric conditions some of the
precipitation evaporates before it reaches the ground. After precipitation
reaches the ground, three things can happen to it: some will evaporate
directly from soil, plants, and free bodies of water; some may infiltrate
the soil or rocks; and some may run off across the land surface. The
runoff may contribute to normal surface drainage in the form of streams
or lakes, eventually returning to the ocean to begin the hydrologic cycle
Water that finds its way underground, however, will have a more circu-
itous route before it returns to the sea. Some water may percolate down-



u." T

i I' LAKE ;

Figure 5 Hydrologic cycle: the constant movement of groundwater' : surface and atmospheric waters. The dia-; .,
gram is h y s ACE d.
*.,, DRAINAGE .,,,,.,' .' .' .. . .. .- .. . . ... ,
^^ ^ ::: A-:::.;rirri : <;ARlNf.E.r:1:^;^0 .. ,.. : : :.....

4.'. "..'.':. .. .. . . . . . . . . . .. . .

Figure 5. Hydrologic cycle: the constant movement of groundwater, surface and atmospheric waters. The dia-
gram is highly simplified.


ward to recharge groundwater aquifers, then move laterally until being
discharged to stream beds or in surface or submarine springs. Some
underground water may be taken up by plants and evapotranspired to
the atmosphere; some may be withdrawn by wells for human use.
All groundwater occurs in open spaces within the rock materials of the
earth's crust. Aquifers are subsurface zones of rocks or sediments that
yield water in sufficient quantities to be economically useful for man's
activities. Aquifers are classified as either unconfined, semi-confined, or
confined. Figure 6 illustrates several situations commonly encountered in
Florida sediments and rocks.
Water that is in direct contact with the atmosphere through the pores
or voids in sands, gravels, or rocks is called unconfined water, and the
zone of sediments or rocks saturated with water is an unconfined aqui-
fer, sometimes referred to as the surficial or watertable aquifer. The top
of the watertable can be visualized as being the upper surface of the zone
of saturation. The elevation of the watertable is also represented by the
water level in wells. The watertable surface is usually a subdued replica
of surface topography, with the watertable lying at shallow depths in
much of Florida.
Semi-confined or confined water is separated from direct contact with
the atmosphere by impermeable materials, such as clay beds or consoli-
dated rocks. Confinement may impose pressures on the contained water
that are higher than atmospheric, creating artesian conditions. Artesian
conditions originally meant that a well produced flowing water at the
surface because of the pressure in the penetrated aquifer. The term now
refers to any condition where water is under greater-than-atmospheric
pressure and will rise some distance up a well that penetrates a confined
aquifer; however, the well need not flow at land surface.
In nature, the distinction between unconfined and confined ground-
water is not so clear-cut, but is usually gradational due to the physical
characteristics of most rocks. At one extreme are loose materials, such
as sand, gravel, and many soils, which have relatively high permeabili-
ties. At the other extreme are so-called tight, solid, or impermeable rocks
or clays. While most clays, or sediments with significant amounts of
clay, do have some permeability, it is usually so low that, for water-
yielding purposes, they are classified as being "impermeable" and are
considered to act as confining beds to more permeable rocks. Some-
where between these extremes lie countless combinations of rock types
with varying degrees of permeability which are classified as semiperme-
able, i.e., they may transmit enough water to allow recharge to contigu-
ous strata, but they cannot provide useful quantities of water to a well.
Aquifers bounded by semipermeable units may be classified as semi-
confined, depending on the water-yielding abilities of the rocks. Figure 6
illustrates this situation by showing surface water being recharged
directly to the confined limestone aquifer through a sinkhole that has
breached the confining bed, and indirect recharge of water from the






Ii .- .
S.C .C.ER G-....,....V.
.. .. ... ..
..............i..... MAAT:iANJ A411A: *:*::*. : . . ...


............. . ....... X.:- AX .:.j -X:I::'::~':''.''>::;::r:::::


Figure 6. Unconfined and confined aquifer in a simplified stratigraphic sequence that is common in Florida
All materials below the watertable are saturated. Recharge to the watertable is by rain. Recharge to
the confined aquifer is by water moving downward through the confining beds or through karst
features that breach confining beds, such as sinkholes.
~::P~::::::..... :::::~:.. I::.. ......: ..:::: I
:.:i I~~j:;i;~;;~;:.. .......... ... ....... 0::: :.: ;::::::::::::I:::::::i
... ...... ...... ............... M
M A.: . . . . ..::':::~ : :: ::: ,:
TRF -<::..:-:; 2:;:;
X. 4q!! :~.~.;::::::;:iiW~iiiifiiii~ij::::
..............:::: I~' I::~.
k : : : : ~ : ::t ::: :i : 5::~ ::: ::... . . . . . . . . . . . . . ..:t~ ~ . ~ : : :
...... ........... ... ............:~
i:~B l';:':::. ..... --------:- G

......... .. .. .........':.~~
REAM ~ ~ ~ ~ .. .. .:.. .. ...iI~iii:~t :::::':::?:::
X.,~ ~~b ?:l~::: ; ~~~::::::: ,i:::~::::iI~::~S''::''''':::'
.............- . ..... ...:.
Figure 6. Unconfined and confined aquifers '.'"'. in a simplified stratigraphic sequence that is common in Florida.:::
All materials below the watertable are saturated. Recharge to the watertable is by rain. Recharge t
th cnfne auierisbywaermoin dwnar troghth cnfnig ed o troghkas
features that breach confining beds, such as sinkholes.::::;:;;::::::::::


unconfined watertable aquifer by slow, downward seepage through the
semi-permeable clay.
Unconfined and confined groundwaters move in response to gravity,
the same as surface water, from higher to lower elevations. Confined
groundwater also moves in response to pressure gradients, similar to the
movement of water in pressurized pipes. As shown in Figure 6, water
will migrate downward through the confining clay if the pressure created
by the weight of the water in the overlying watertable aquifer is higher
than the pressure in the confined aquifer. Conversely, if pressure in the
confined aquifer were high enough to overcome the pressure of the
water in the unconfined aquifer, then water would move from the lime-
stone through the clay to the sand. Both of these situations commonly
occur in Florida due to inhomogeneities of strata, karst features, and
pressure gradients.
Springs are an expression of leakage from a watertable, semi-
confined, or confined aquifer. In Figure 6, for example, the surface spring
occurs because the watertable aquifer occurs on top of a confining bed
that impedes the downward percolation of recharge water. This situation
forces the water to move laterally, downslope, and discharge where the
permeable sand and the less permeable clay bed intersect land surface.
After prolonged periods of no rain the aquifer may become so depleted
that the spring ceases to flow. This type of spring is frequently seen in
the steep-walled stream valleys of north Florida.
Similarly in Figure 6, the subaqueous spring that discharges into the
stream bed from the confined aquifer does so because of higher water
pressure in the aquifer. In this situation, however, if the pressure in the
aquifer falls too low due to depletion of water, the spring may reverse its
flow, taking water back into the aquifer or recharging it. This, too, is a
well-documented occurrence in some Florida streams. Ceryak, et al.
(1983) found that many sinkholes in the bed of the Alapaha River near
Jennings, in northwest Hamilton County, Florida, have recharged as
much as 770 cubic feet per second (497,420,000 gallons per day) to the
adjacent aquifer. Miller, et al. (1978) found that, at times of high stages
of the Suwannee River, river water was recharged to the limestone aqui-
fer through karst features in its channel.
It is these driving forces in the hydrologic cycle that move underground
water through Florida's carbonate rocks. In transit, the water dissolves
and carries away in solution the chemical components of the rocks,
leaving behind caves, solution pipes, and other voids that result in a karst


The evolution of any terrain into characteristic landforms involves
weathering and erosional processes: wind, water, frost heaving, slump-
ing, or wave activity, to name a few. In most areas, the predominant


Figure 7a. Relatively young karst landscape showing underlying
limestone beds and sandy overburden with normal, inte-
grated surface drainage. Solution features are just
beginning to develop in the limestone.

weathering, erosional, and transporting agent is water, either falling,
flowing across the land, or circulating through subsurface rocks.

Chemical Weathering of Carbonate Rocks

Since the genesis of karst involves the development of underground
drainage systems, it is necessary to study such systems to understand
the formation of karst. Karst processes tend to be secretive and imper-
ceptible because most development occurs underground over long peri-
ods of time. The results of these persistent processes will be manifested,
sooner or later, in the subsidence of surficial sediments to form swales,
the formation of a new sinkhole, a sudden influx of muddy water in a
water-well after a heavy rain, or some other karst phenomenon that may
disturb or disrupt man's activities. Figure 7 illustrates the evolution of
karst terrain, as described below.
Chemical weathering is the predominant erosive process that forms
karst terrain. Chemical weathering of limestone removes rock-mass
through solution activity. As rain falls through the atmosphere, some
carbon dioxide and nitrogen gases dissolve in it, forming a weak acidic
solution. When the water comes into contact with decaying organic
matter in the soil, it becomes more acidic. Upon contact with limestone,


'- |

Figure 7b. Detail of Figure 7a showing early stages of karst forma-
tion. Limestone is relatively competent and uneroded.
Chemical weathering is just beginning, with little internal
circulation of water through the limestone. Swales,
forming incipient sinkholes, act to concentrate recharge.

a chemical reaction takes place that dissolves some of the rock. All rocks
and minerals are soluble in water to some extent, but limestone is espe-
cially susceptible to dissolution by acidic water. Limestones, by nature,
tend to be fractured, jointed, laminated, and to have units of differing
texture, all characteristics which, from the standpoint of percolating
groundwater, are potential zones of weakness. These zones of weakness
in the limestone are avenues of attack that, in time, the acidic waters will
enlarge and extend. Given geologic time, conduits will permeate the rock
that allow water to flow relatively unimpeded for long distances.
During the chemical process of dissolving the limestone, the water
takes into solution some of the minerals. The water containing the dis-
solved minerals moves to some point of discharge, which may be a
spring, a stream bed, the ocean, or a well, and another tiny volume of
Florida's rock substrate has been removed.
Removal of the rock, with the continuing formation or enlargement of
cavities, can ultimately lead to the collapse of overlying rocks or sedi-
ments. If the collapse is sudden and complete, an open sinkhole will



Figure 7c. Advanced karst landscape. Original surface has been
lowered by solution and erosion. Only major streams
flow in surface channels and they may cease to flow in
dry seasons. Swales and sinkholes capture most of the
surface water and shunt it tb the underground drainage
system. Cavernous zones are well-developed in the

result, sometimes revealing the cavity in the rock (Figures 8 and 9). More
often, though, debris or water covers the entrance to subterranean drain-
age. Partial subsidence of the overburden into cavities will form swales
at the surface, producing hummocky, undulating topography. By this
slow, persistent process of dissolution of limestone and subsequent col-
lapse of overburden, the land is worn down to form a karst terrain.
At some point in this process of dissolution of underground rocks, a
normal surface drainage system will begin to be transformed into a dry or
disappearing stream system. Continuing dissolution of the limestone will
create more swales and sinkholes, which will divert more of the surface
water into the underground drainage. Eventually, all of the surface drain-
age may by diverted underground, leaving dry stream channels that flow
only during floods, or disappearing streams that flow down swallow
holes (sinkholes in stream beds) and reappear at distant points to flow as
springs or resurgent streams.


Figure 7d. Detail of Figure 7c showing advanced stage of karst
formation. Limestone has well-developed intercon-
nected passages that form an underground drainage
system, which captures much or all of prior surface
drainage. Overburden has collapsed into cavities form-
ing swales or sinkholes. Caves may form. Land surface
has been lowered due to loss of sand into the lime-
stone's voids. Wakulla Springs and Silver Springs are
examples of cavernous underwater springs.

Lowering of Land Surface

Inherent in the formation of karst terrain is the lowering of land surface
on a regional scale, in contrast to the very localized lowering at a sink-
hole. Regional lowering of the land surface takes place through the
cumulative effects of thousands of individual, localized events, and
through the continual removal of carbonate rock by dissolution. Several
investigations have been made to determine an "average" rate of sur-
face lowering in Florida, the results of which are discussed below and
shown in Tables 1 and 2.
Table 1 gives comparative data for the ten largest first-magnitude
springs in Florida. The amounts of solids removed from the land by these
springs' flows range from 59 to 541 tons per day. These figures are
impressive, but they do not indicate how rapidly the land surface may be
being lowered. More meaningful are the amounts of material that are
carried off per year per square mile of land surface, which can be calcu-

a ,1


Figure 8. Karst limestone surface exposed by a flash-flood in the city of Ocala in 1982. Note similarity
of karst weathering to that in Figure 1. Florida Geological Survey photo.

Figure 9. Close-up of solution pipes in same area as Figure 8. Pipes are about two feet in diameter.
Sinkholes and other karst surface expressions can appear when such pipes become
unplugged. Florida Geological Survey photo.


, w *i~,*\

Table 1. Ten first-magnitude springs, ranked according to average discharge for period of record. See Figure 38
for locations. A first-magnitude spring's average flow must be 100 cubic feet per second (cfs) or more.
Data from Rosenau, et al., 1977, p. 7.
Spring DISSOLVED SOLIDS in solution,
(County) Range Average (mg/L = ppm) in tons per day.*
Spring Creek only one 2,003 2,400**
Springs (Wakulla) measurement
Crystal River *** 916 144 355
Springs (Citrus)
SilrSrings 539-1,290 820 245 541 C

St. Marks Spring 310-950 519 154 215
Wakulla Springs 25-1,910 390 153 160 0
(Wakulla) 25-1,910 390 153 160
Wacissa Springs 280-605 389 150 157
Group (Jefferson)
Itchetucknee Springs 241 -578 361 170 165
Blue Springs 56-287 190 116 59

cfs X 62.4 Ibs/ft3 X 86,400 s/day X TDS
*Calculated by: = tons per day in solution.
2 X 109
*A submarine spring system that may have as many as 14 outlets. Discharge affected by tides, and fresh groundwater is probably being
mixed with Gulf water, accounting for high TDS.
***A system of some 30 springs, which may be tidally affected.


Table 2. Calculated rates of surface lowering due to solution showing
possible range of rates depending on variable densities of lime-
stone. Rates of limestone lowering in inches per thousand

LOCATION REFERENCE:RATE (inches/1,000 yrs.)

1. Rainbow Springs Fennell, 1969: 0.5 Bishop, 1982: 0.7
(pers. comm.)

2. Itchetucknee Springs Fennell, 1969: 0.8 Bishop, 1982: 1.2
(pers. comm.)

3. Silver Springs Fennell, 1969: 1.4 Bishop, 1982: 2.1
(pers. comm.)

4. central peninsula Sellards, 1908:

5. Suwannee River
drainage basin Brooks, 1967:

6. near Tampa well
field Sinclair, 1982:
approx. 7.0

7. northern peninsula Opdyke, et al., 1984:
approx. 1.0

lated when the area of a spring's karst drainage basin is known. The
karst drainage basins, which includes surface and underground drainage
systems, have been determined for only three of the springs in Table 1:
Silver, Rainbow, and Itchetucknee.
Faulkner (1970, Fig. 22), as part of a study of the Cross-Florida Barge
Canal, prepared potentiometric maps in the vicinity of Silver and Rain-
bow springs. The maps delineated the underground drainage areas that
supply water to the springs, and showed each spring's drainage area to
be approximately 730 square miles. Using data from Table 1, the
amounts of solids carried away in solution by Silver and Rainbow springs
were calculated to be 270 and 95 tons per square mile per year, respec-
Similarly, using potentiometric maps constructed as part of a hydro-
logic study of the Itchetucknee Springs area, Hunn (1982, personal com-
munication, U.S. Geologic Survey, Tallahassee Subdistrict Office) calcu-
lated a drainage area of approximately 400 square miles. These data
indicate that Itchetucknee Springs may carry away as much as 155 tons
of dissolved rock per square mile per year.
Using the results of two separate studies (Fennell, 1969; Bishop, 1982


personal communication), the author calculated the rates at which Silver,
Rainbow, and Itchetucknee springs may be lowering land surface. Fen-
nell (1969, p. 47) analyzed 10 samples of limestone from the Tallahassee
area and obtained an average density of 166.6 Ibs./cu. ft. Ernest Bishop
(personal communication, 1982) analyzed 19 core samples of limestone
from three geological formations near Tallahassee and obtained an aver-
age density of 111.1 Ibs./cu. ft. Within their respective drainage basins
Silver, Rainbow, and Itchetucknee springs may be lowering ground sur-
face by solution from 0.5 to 2.1 inches per thousand years, as shown in
Table 2.
Using scant data, Sellards (1908, p. 47) calculated the amount of
material being carried away in solution by the waters from eight springs
in central Florida. The amounts ranged from 29-600 tons per day. Sel-
lards (1908, p. 48) assumed that most of the rock material being
removed was limestone, and stated, "From these estimates it would
appear that the surface level of the central peninsular section of Florida is
being lowered by solution at the rate of a foot in five or six thousand
years." These estimates yield a rate of solutional degradation of the
surface of 2 to 2.4 inches per thousand years.
In a study of the Suwannee and Waccasassa rivers' drainage basins,
Brooks (1967) found in various parts of the basins that the amount of
carbonate rock being carried away in solution ranged from 172 to 262
tons per square mile per year. He calculated that the rate of solutional
degradation of the karst terrain ranged from 1.2 to 2 inches per thousand
years, for an overall rate of 1.5 inches per thousand years.
The above studies were of regional scale and represent the range of
land surface lowering that could be expected under long-term natural
conditions. Man's activities, particularly on a local scale, disturb the
natural environment, often times drastically within short spans of time.
Sinclair (1982), as part of a study on pumpage-induced sinkhole collapse
at a municipal well field in the Tampa area, calculated a rate of surface
lowering of the limestone of about one foot per 1,700 years (7 inches/
1,000 years). This rate of lowering would only be relevant to the area of
a few square miles immediately adjacent to the well field that is affected
by the pumpage. However, in comparison with the natural rates shown in
Table 2, it gives an idea of how man's activities can affect karst proc-


A good way to illustrate the variety of physiographic features associ-
ated with a karst region is to cite examples. Some of the case histories
presented in this section have been selected, admittedly, because of their
spectacular nature. In a way, their spectacular nature can be used to
advantage to show the evolution of karst terrain which, until now, has
been discussed from a theoretical viewpoint. They demonstrate the


cumulative results of millenia of solution activity, which usually operates
at microscopic scales on a day-to-day basis.
Selected examples will be discussed in detail to illustrate the generic
types of karst features, with other examples noted which are in various
parts of Florida. Most of these examples are accessible to the public, so
they can be visited, explored, and photographed.


In any karst terrain, sinkholes are the most common feature, as well as
one of the most easily recognized. Figure 10 shows, in a general way, the
areas of Florida that are prone to sinkhole development. However, this
should not be strictly interpreted as a sinkhole-risk map. Local factors
will govern whether sinkholes actually do form; under certain conditions
they could form in an area that ordinarily would be at low risk.
Chemical weathering of limestone is the ultimate cause of sinkhole
development, but localized stress triggers overburden collapse into pre-
existing cavities, which may have taken eons to dissolve out of the
limestone. The stress can be natural or a result of man's activities.
In Florida, the most common natural sources of stress that trigger
sinkhole formation are fluctuations of water levels or water pressures
resulting from torrential rains and flooding, or the opposite, severe
drought with lowered water levels. Figure 11 illustrates how this can
happen. Water, and its relationship to the overburden and bedrock, are
the controlling factors under flood or drought conditions. Objects
immersed in water weigh less because of the effect of buoyancy. So it is
with sand, clay, or limestone beneath the watertable in the zone of satu-
ration. Sand and clay, for example, weigh 40 percent less when
immersed in water (Sinclair, 1982). Saturated materials below the
watertable, therefore, have significant supportive forces, which tend to
forestall collapse of overburden into cavities.
By definition, a karst terrain has underground solution cavities, any
one of which may have sinkhole-forming potential under the right condi-
tions. Figure 1 la shows normal, stable conditions. In a drought the
watertable will fall, decreasing support for the overburden above the
cavity (Figure 11b). At some point the overburden will be unable to
support its own weight, resulting in surface subsidence or a sinkhole,
depending on the degree of collapse.
Under conditions of torrential rains or flooding, the role of water can
change-it can become an additional burden to the soil column. Another
force comes into play under flood conditions; a downward-acting one
that increases the stress on the overburden. The increasing height of the
water in the overburden creates an increased flow downward and away
from the cavity. This increased downward force may be enough to trig-
ger collapse of overburden into a cavity.
Man's activities also impose stresses on the environment, which, in a
karst terrain, pose special concerns. In Florida, two categories account


I: .

I :..- i r, 7 t ," ,

S LGOVER Il MORtE lAN 21uu r i in iAi.. culnsiiL ui
cohesive sediments interlayered with discontinuous
carbonate beds Sinkholes are very few, but several large
diameter, deep sinkholes occur Cover collapse sinkholes

0 50 miles

80 km

I. ., .-
,. ." -

-' '-

"J *' : / t
I"- -\ :

"-' ,,F

/I- -/ 00

I mdoq - 1

Figure 10. Map of Florida showing areas prone to sinkhole devel-
opment. Modified after Sinclair and Stewart, 1985.

1 -



-CLAY--- .

CAVITY 0 ". .


A _Bl Cl
Figure 11. A) Cavity has formed in limestone due to circulation of groundwater. Cavity grows upward by
stopping spellingg of material from ceiling of a cavity.)
B) Enough loose sand has been piped into cavities to cause subsidence at surface. The watertable
may be lowered due to unrestricted flow into underground drainage. A lowered watertable may
be manifested locally by plants stressed for lack of water, or wells may go dry.
C) Continual enlargement of underground drainage and removal of overburden results in the
typical, cone-shaped sinkhole. A sudden collapse may swallow trees or buildings. The depression
may intersect the watertable, forming a pond.



Figure 12. Drill rig in a sinkhole that formed
when the drilling operation upset
the equilibrium of the soil column
above an underground karst cavity.
Hernando County, Weeki Wachee
area, 1981. Photo by Tony Gilboy,

Southwest Florida Water Manage-
ment District.

for most of man-induced sinkhole formation: pumpage from wells and
construction activities. The mechanisms responsible for the actual col-
lapse of overburden are the same as explained above; however, man's

activities create the instability.
Referring again to Figure 11b, if a well had dewatered the sediments,
collapse could have been triggered. This has been a frequent, though
unexpected occurrence in Florida. For example, in one documented
instance more than 30 small sinkholes occurred near a large-capacity

municipal well-field north of Tampa, all within a year after beginning

pumpage (Sinclair, 1982).
A related activity, well drilling, can also pose problems-for the driller

as well as for the environment. Larger, deeper wells are usually drilled
using pressurized, circulating water to lubricate the drill bit and to
remove the cuttings from the hole. Visualize a large, heavy drill rig drilling
into the cavity in Figure 11 a. Simultaneously, the driller has disturbed the

*' ""
* '* '
"- *'
I**""" -
'' '~"
- ''*'' -^
'' ~*' """

-~ -"

. -f




cohesion of the overburden by drilling through it, breached the cavity,
and imposed the extra load of the drill rig on the soil. Within seconds, or
minutes, the situation goes from 11 a to 11c, with the drill rig in the
sinkhole. There have been several instances of this in Florida (Figure 12).
Construction activities, such as excavating and dewatering for founda-
tions, heavy equipment traffic, blasting, and altering natural drainage
patterns, all can trigger sinkhole collapse. The weight of buildings and
loads from reservoirs can also case sinkholes to form.
Although the general notion is that sinkholes collapse suddenly with-
out warning, one or more precursors may occur (from Sinclair, 1982):

1. Slumping or sagging. Canting of fence posts or other objects from
vertical, and doors and windows that fail to open or close properly
may be early warnings of subsidence.
2. Structural failure. Cracks along mortar joints in walls and in pave-
ments, however small, may be indications of subsidence.
3. Ponding. The ponding of rainfall may serve as a first indication of
land subsidence.
4. Vegetative stress. One of the earliest effects at an incipient sink-
hole is lowering of the watertable through percolation to the under-
lying aquifer. The lowered watertable may result in visible stress to
a small area of vegetation.
5. Turbidity in well water. Water sometimes becomes turbid during
the early stages of development of a nearby sinkhole.

Sinkhole is a word that has acquired bad connotations, such as danger,
tragedy, threats to property, people or animals (Figures 2, 3, and 13).
There are other aspects of sinkholes, however, that are beneficial. A
discussion of a recently formed sinkhole will be useful to illustrate the
pros and cons of sinkholes.

Winter Park Sinkhole

The most famous sinkhole in the United States in recent years is the
one that formed in May 1981, at Winter Park, near Orlando, Florida. It
captured national headlines in all the media with magazine articles still
appearing in March, 1982. An aerial view of the sinkhole is shown in
Figure 13. It is roughly circular but elongated, approximately 300 feet by
350 feet. The water level in the photograph is about 45 feet below the
ground surface. The sinkhole is hydraulically connected to the local aqui-
fers since its water level fluctuates in response to rainfall and changes in
the surrounding watertable, sometimes rising to within 15 feet of ground
The following statistics can give an idea of the extent of damage which
may be associated with sinkhole formation. The Winter Park sinkhole
swallowed: one house and shed, half of the municipal swimming pool, a
Porsche sports car, several large oak trees, a section of street that


Figure 13. Winter Park sinkhole, formed in May, 1981. Photo by
Richard Deuerling.

crossed the sinkhole and part of another street that adjoined it, and an
estimated 4-million cubic feet of soil. Additional damages that can be
seen in the picture are: three other Porsche sports cars and a pick-up
camper which slid into the crater, the rear of an auto-repair shop that
cracked open, and various utility lines were exposed or damaged.
As spectacular as this event was, and granted that the total cost of
damages was great, it needs to viewed in perspective of the environment
in which it formed. Figure 14 is a map of part of the Winter Park-Orlando
area that shows all lakes and closed depressions in the vicinity of the
new sinkhole. The U.S. Geological Survey has estimated that 95 percent
of such features are former sinkholes (Brainard, 1982). It is obvious that
sinkhole formation has been common in the past, and, judging from the
size of many of the lakes, there probably have been more spectacular
collapses. The new sinkhole is one of the smaller karst features of the
It is not possible to predict the exact evolutionary course of this or any
other particular sinkhole, but some generalizations can be made based on
the geological history of the area. This sinkhole may be unstable for an
indefinite time. It may become temporarily plugged with debris, organic


(/ ,^-O= 0

Figure 14. Map of the Orlando-Winter Park area in the vicinity of the
new sinkhole, showing karst features that hold water
some or all of the time.

matter, silt, or clay and thus retain water. The plug may break and drain
the water occasionally. During its inactive periods, erosive processes will
wear down its sides and fill the sinkhole's bottom. Eventually, the sink-
hole will probably become plugged to such an extent that it will be stabi-
lized, for all practical purposes. At that time it will be just another local
lake, which will be considered a community asset.



The foregoing discussion of the possible evolution of the Winter Park
sinkhole into a pond or lake is an appropriate introduction to the general
subject of lakes, one of Florida's most ubiquitous natural resources.
Lakes occur in a profusion of shapes and sizes throughout Florida. There
are more than 7,700 named lakes and ponds that are larger than 10
acres, totalling more than 2,860,000 acres (over 4,400 square miles)
(Fla. Bd. of Conservation, 1969). There are probably as many more that
are smaller than 10 acres. These thousands of lakes, most formed by
karst processes, provide inestimable benefits to Florida citizens and visi-
Bishop (1967) described several origins for lakes in Florida. Some of
the larger lakes occupy depressions that were in the sea bottom when
the ocean covered parts of Florida during periods of the Ice Age. These
are lakes Okeechobee, Istokpoga, and probably some of the larger lakes
in the Kissimmee and St. Johns river valleys. There are also a few lakes
formed in low areas between sand dunes; some created by man-made
dams, and some as a result of reclamation projects following open-pit
phosphate, limestone and dolomite, or sand-gravel mining. However,
most of Florida's lakes are solution-based lakes created by groundwater
solution of underlying limestone and subsequent lowering of local land
surface, by karst processes already discussed.
It is not surprising, then, that most of Florida's karst-origin lakes have
physical characteristics of sinkholes, such as relatively steep sloping
sides, no surface streams into or out of them, and circular outlines. Many
of these lakes are dependent on rain for sustenance and large fluctua-
tions from local average rainfall can cause significant, even drastic
effects on water levels. Some have connections with aquifers. A few
show a pattern of disappearing when their underground karst drainage
systems become "unplugged." For example, two lakes near Tallahassee,
lakes Jackson and lamonia, drain periodically. Lake lamonia drained in
1910, 1917, 1934 and 1981 (Figures 15 and 16). Lake Jackson has
drained about every 25 years since 1881; its latest disappearance was in
1982. Since then it has refilled.
These two lakes, both quite shallow, are underlain by carbonate rocks,
pinnacles of which can be seen in their beds after they drain dry. Severe
weather conditions, such as droughts or flooding, impose stresses on
the local hydrological regime that can trigger sinkhole collapse. If the
material bridging a karst cavity happens to be in the bed of a lake-and
hydraulic pressure causes the bridge to collapse-the result is the same
as pulling the stopper out of a bathtub drain (Figures 15 and 16).

:.. .-.. ^ ^ "
rrt IH t'' ; ^ f," 1^- ^
C I'

.*1 ^^' J ~ ^1w g^^.^' 1
4M s*:e1R~r
*51~ I,.



Figure 15. Dry bed of Lake lamonia, 1981. The main sinkhole drain is near the person at lower left.
Limestone is exposed in several circular pits throughout the lake bed, which also act as
drains. Florida Geological Survey photo.

4... -


low,' - -
1.'. ." .. -

. . .-
'". ';-," . :h 0

... .. .. .. 0 -


Figure 16. Dry bed of Lake lamonia, 1981. Unconsolidated lake-bed deposits showing bedding. The
meandering, incised stream channel is an artifact of many past drainings of the lake; each
successive draining etches it deeper. Florida Geological Survey photo.


0 5

A 1 Wakulla Spring
O 2 Big Dismal Sink
3 Lost Lake
4 Natural Bridge
Woodville Karst Plain

SLake lamconi

271A r)Lake Jackson




i. -l90


Figure 17. Map showing extent of the Woodville Karst Plain in Leon
and Wakulla counties. After Hendry and Sproul, 1966.

Woodville Karst Plain

Sinkholes, springs, swales, hummocky terrain, disappearing streams,
natural bridges, and cavernous openings, all with their associated under-
ground drainage, are manifestations of karst processes and are prevalent
in areas underlain by limestone or other carbonate rocks. The Woodville
Karst Plain (Hendry and Sproul, 1966, pp. 25, 29-33), which lies south
of Tallahassee in Leon and Wakulla counties (Figure 17), exhibits all of
these features.
The Woodville Karst Plain has a flat to gently undulating surface of


sand that overlies carbonate rocks. The carbonates, which lie at shallow
depths of 30 feet or less, have undergone extensive solution by ground-
water. This plain exhibits karst features that are still evolving, for exam-
ple: many old, well developed sinkholes that are either permanently or
intermittently flooded (Big Dismal Sink), disappearing streams and natu-
ral bridges (Natural Bridge), Wakulla Springs, and new sinkholes reported


Big Dismal Sink is a circular sinkhole located about 10 miles south of
Tallahassee and is accessible in good weather by sand trails from State
Road 319 (Figure 18). Figures 18, 19, and 20 illustrate its size and shape.
The sinkhole intercepts the local water table, and it is about 65 feet to
the water surface, which fluctuates according to seasonal rains. Approx-
imately 30 feet of limestone is exposed above the water. The limestone
is overlain by a sandy, clayey layer, which in turn is covered by about 20
to 25 feet of loose sand. Springs occur in the clayey sand and limestone
of the sinkhole's walls. The limestone walls form overhangs just above
the water. The sides are steep and treacherous, although it is used as a
local swimming hole. Divers have not reported an exact depth, but it
would not be an unusual configuration if it extended as far below the
water as above. One sounding taken from the southeast rim by Florida
Geological Survey staff members showed a depth of 61 feet below water
In comparison with the 1981 Winter Park sinkhole, Big Dismal Sink has
consumed over three-million cubic feet of rock and soil, which places it in
the same magnitude. Figure 18 shows that Big Dismal is only one of
countless sinks and other karst features that make up the Wakulla Karst


Lost Lake is about seven miles south of Tallahassee on State Road 373
and is a designated recreation area in the Apalachicola National Forest
(Figures 17 and 21). It is a slightly elongated, circular swale that was
created by solution of limestone and subsequent subsidence of sandy
soil. Because the swale has subsided enough to intercept the local
watertable it contains a perennial lake that is about 600 to 650-feet
across and 10 to 12-feet deep. The dimensions and depth of the lake
varies in accordance with wetness or dryness of the seasons.


Natural Bridge is located about six miles east of Woodville (Figures 17
and 22). For approximately one-half mile along this part of its channel the
entire flow of the St. Marks River disappears under limestone "bridges"
several times and reappears as springs short distances downstream. The



\ */ S o
t 1 17




K~ _:_qoc

0 0.5 mi
H~r,LII rLrAi







Figure 18. Map of part of the Woodville Karst Plain showing Big Dis-
mal Sink. This small area of approximately seven square
miles has over 100 sinkholes and flooded swales, which
are indicated by the irregularly rounded, closed elevation
contour lines. From the U.S. Geological Survey 7-1/2 min-
ute Lake Munson quadrangle.




Not to scare.
Vertical exaggeration about 2X.







Rim elevation
75' above
sea level

Water elevation 10'
above sea level

Figure 19. Big Dismal Sink, plan and cross section.




I,. .-
xI ~~raiso


ic..; '
'"*'*' "-n.*^1 A .*

-^ '*.e

-i" t --tn


Figure 20. Big Dismal Sink, looking from north rim to south wall. Note
the person at top center for scale. Florida Geological Survey

county road uses one of the bridges to cross the stream. The bridges
were formed by dissolution of weaker limestone, leaving more resistant
portions of the rock in place. The surface stream has been captured by
underground drainage for those reaches where it goes under the bridges.
About one-half mile south of the bridges the river continues to the Gulf in
a well-defined channel in the limestone.
The bridges could eventually collapse if erosion and solution activity
persist. However, if the river should create drainage channels at still
deeper levels in the bedrock, the bridges could be left isolated and rela-
tively free from the river's erosion.

* S


A *-"^ "^ '"'
iii S miii i*" ** 1cFfln3e i


"%. .


k1 P.

.y 4

Figure 21. Lost Lake, view to west. Stake in right center of picture
marks approximate center of lake. Florida Geological Sur-
vey photo.


Wakulla Springs is a major north Florida tourist attraction that is
located approximately 14 miles south of Tallahassee on State Road 61
(Figure 17).
Wakulla Springs, with its rich history of local lore, has been commer-
cially developed since 1844. In that year, Mr. P. Randall ran an ad in the
Tallahassee newspaper offering lodging to the public and tours of the
springs and river in a glass-bottomed boat (personal communication,
Charles Daniels, Wakulla Springs, 1984). Present-day visitors to Wakulla
Springs can still take entertaining and educational rides on glass-
bottomed boats, where they can look down through more than 100 feet
of crystal-clear water into the main vent of the springs.
Wakulla Springs is classified as a first magnitude spring. It has an
average discharge of about 390 cfs (251 million gallons per day); its
maximum measured discharge was 1,910 cfs (1,233 million gallons per
day) (Rosenau, et al., 1977). It is the main source of the Wakulla River;

-* lims ,-.UDX~ .*.


Figure 22. Natural Bridge area, east of Woodville. From the U.S. Geo-
logical Survey 7-1/2 minute Woodville quadrangle.


there are several smaller springs near the main spring that contribute to
the river.
Figure 23 is a cross section of the spring and 1,000 feet of the main
spring's conduit. Divers have explored over 2,000 feet into the cavern,
with no end in sight. Visitors to the springs can see a color movie of their
exploration of this fascinating natural wonder.
Several sets of mastodon and mammoth bones have been recovered
from the springs (Sellards, 1916). In 1930 the Florida Geological Survey
recovered a nearly complete mastodon skeleton from the main spring. It
is on display in the R. A. Gray State Museum in Tallahassee. Other fossils
or artifacts recovered from the main spring include a giant sloth's tooth,
bones of tapir, deer, armadillo, and charred wood and more than 600
bone spear points, (Mohr, 1964; Boyles, 1965). These finds indicate that
the Spring was probably used as a watering hole by these prehistoric
beasts, and possibly as a shelter by early man.
An obvious question is, "How did prehistoric animals or man get into
the spring's inner recesses against such a rapidly flowing current?"
Reconstruction of a scenario of Wakulla Springs' prehistoric environment
will explain the general natural history of the springs and help to resolve
the seeming paradoxical question.
One explanation may lie in the possibility that the springs haven't
always been "springs." During the Pleistocene (also called the Ice Ages,
the period between 2,000,000 and 10,000 years before present) world-
wide sea levels were lowered several times by as much as 300 feet. The
lowering of the nearby Gulf of Mexico by several hundred feet undoubt-
edly would have caused a lowering of local inland watertables, such as at
Wakulla Springs. Referring to Figure 23, if the water level fell from its
present elevation (Point A) to some prehistoric elevation (Point B), then it
would be, in fact, Wakulla cave-not spring.
Reflecting on previous discussions of the Winter Park Sinkhole and Big
Dismal Sink, one cannot help but notice striking similarities between
them and Wakulla Springs: the sinkholes are 250 to 350-feet across and
their visible pits are about 65 to 100-feet deep. From a hydrological
viewpoint, the only difference between the Winter Park Sinkhole, and Big
Dismal and Wakulla Springs is the direction the local water flows, either
into or out of the underground drainage. Another part of an explanation
to the question is that geological evidence indicates that Florida was
probably semi-arid and savanna-like during periods of the Pleistocene
(Simpson, 1929, p. 242). All the parts of the prehistoric scenario are
now in place: large cohorts of animals and associated human hunters
living on a semi-arid Florida savanna, similar to an African veldt. Water-
tables are lower than present and water is scarce much of the year. The
deep Wakulla cave-sink is a more dependable water hole during dry sea-
son, but a steep slope must be negotiated-a feat that some apparently
did not successfully complete-their bones attest to their failure. In a
similar fashion, prehistoric man could have used the cave for shelter,



- Point A

- Point B

100 200

Figure 23. Wakulla Springs, cross section. After Boyles, 1965.


possibly bringing killed animals into the cave and leaving their bones for
later explorers to find and wonder about.

Underground Rivers

The preceding discussion of the Wakulla Springs and its cavernous
underground drainage system leads to the subject of so-called under-
ground rivers. The idea of underground rivers is usually put to geologists
in the form of a question, "Is it true that underground rivers flow from
the mountains of Alabama or Georgia, under Florida, and to the Gulf of
Mexico?" The idea of underground rivers that run for hundreds of miles is
a fascinating concept that occurs frequently in discussions of ground-
water, aquifers, springs, and underground drainage systems. In this con-
text, however, the idea is wrong-underground rivers do not run from the
Appalachians to the Gulf.
Because springs obviously are fed by water that is flowing out of the
ground, there is a certain logic to the thought that water can flow in
underground streams, as it does in surface channels, sometimes for hun-
dreds of miles. There have been rare confirmed cases of underground
rivers (more accurately called drainage systems) that flow for tens of
miles in other parts of the world (Jackson, 1982). However, no such
lengthy underground channel has been documented for any spring's
source in Florida. All of the flow of springs of Florida for which measured
discharges are available, including the major ones of Wakulla, Silver,
Rainbow, and Itchetucknee, can be readily accounted for on the basis of
rainfall within relatively small surface drainage basins. It is not necessary
to imagine the existence of huge underground rivers to supply their
In two hydrologic investigations discussed earlier in the section titled
"Lowering of Land Surface," it was shown that the flows for three of
Florida's largest springs represented the collection potential for small,
local drainage basins. Rainbow and Silver springs each had drainage
basins of approximately 730 square miles (Faulkner, 1970), and Itche-
tucknee Springs had about 400 square miles (Hunn, pers. comm.,
1982). On a regional basis, these are relatively small catchment areas;
for example, an area of 730 square miles can be visualized as a square
that is 27 miles long and wide, while 400 square miles is 20 miles on
each side.
This is not to imply, though, that extensive underground drainage sys-
tems do not exist in Florida-they do. Discussions in previous sections
have shown how they originate. Parts of some systems that have been
explored are impressive. Divers have followed the main conduit of
Wakulla Springs for over 2,000 feet and have found chambers as large as
100-feet high and 150-feet wide (Figure 23). Fossil skeletons of several
Ice Age mastodons found in the recesses of the spring indicates that it
has been of large size since prehistoric times.




Florida Caverns State Park, near Marianna, Jackson County, has the
best example of a karst cave in Florida (Figures 24 and 25). Karst proc-
esses, and how they form caverns, have already been discussed. Up to
now, all discussions of karst processes have concerned their corrosive
character; that is, groundwater dissolving limestone, carrying it away in
solution, and creating voids. Here, at the Caverns, one can see that the
chemical processes are reversible-they can be depositional as well as
erosive. Visitors can walk through the cave and examine closely the
myriad cave deposits formed by karst processes and underground drain-
Stalactites (Figure 26) grow downward from cracks or tiny openings in
a cave's ceiling due to seepage of ground water from the limestone.
When the groundwater, which has entrained carbonate ions and carbon
dioxide gas, encounters the open air of the cave, some of the water
evaporates and some of the gas escapes to the atmosphere. This creates
a chemical disequilibrium which causes some calcium carbonate to pre-
cipitate and form calcite stalactites.
Usually, some of the excess water then drips from the stalactite to the
cave floor, and, in turn, evaporates and precipitates calcite to create a
stalagmite, growing upward from the floor (Figure 27). The stalagmite
may eventually join the stalactite to form a column (Figure 27), if favor-
able depositional conditions exist for a long enough time, possibly hun-
dreds of thousands of years. Once joined, columns can become quite
thick as the water continues to deposit successive thin layers of calcite.
Dripstone and flowstone (Figure 28) are descriptive terms for the decora-
tive deposits that cover or cascade over and around a cave's walls, floor,
ceiling, stalactites, stalagmites, columns, and boulders.
Another type of deposit seen in the Caverns is a drapery (Figure 29).
This forms similarly to a stalactite, except that the water seeps from a
narrow crack in the stone; its precipitate forms a thin, elongated forma-
tion, often wavy and color-banded.
Crystalline structures are common in caves or in voids in carbonate
rocks. Crystals of calcite, dolomite, or other mineral dissolved into solu-
tion by groundwater may be precipitated when the water is released
from confinement in the rock. The flat crystal faces reflect light, as do
the facets of a cut gemstone, often producing kaleidoscopic sparkles in
the dim recesses. Figure 30 shows an example of calcite crystals found
in caves and voids in limestones.
The larger caverns that are open to the public represent only a part of
this area's underground drainage system. Tantalizing clues to its much
larger extent were provided in 1981 and 1984 when some new cave
rooms were discovered by following a narrow "squeezeway" from one

Marianna --- A C j




Figure 24. Location of Florida Caverns State Park.


River Swamp



0 2000 ft 0)

500 meters 0


S/ Ri Station

Figure 25. Florida Caverns State Park, showing geological features of interest: 1) Caverns entrance and visitor
center, 2) Cave tunnel, a walk-through karst feature, 3) Flood plain river bluffs eroded by the
Chipola River, 4) Limestone outcrops occur throughout the park, 5) Cave openings are exposed in
several places, 6) Sinkholes are present throughout the park, 7) Natural bridge, where the Chipola
River goes underground, 8) Blue Hole Spring flows to the surface from depths within the limestone.
From Schmidt, 1982.
From Schmidt, 1982. WJ


Figure 26.

Thin, straw-like stalactites can be seen on
the cave ceiling. Eons of dripping water
build them thicker and longer. From Sch-
midt, 1982.


Figure 27. Columns form when stalactites meet stalag-
mites. From Schmidt, 1982.



The rock formations take many shapes.
Here a dripstone-decorated column
appears as a wedding cake in the well-
lighted underground trail. From Schmidt,

Figure 28.


crack in the rock. Prolonged seepage adds
more deposits that form a banded drapery.
From Schmidt, 1982.


Figure 30. Calcite crystals formed in a cavity in limestone from a
Florida quarry. Photo by T. Scott, Florida Geological

of the older known passages (Figures 31 to 35). Continued exploration
by spelunkerss" may discover more caverns.

i A

a. 1 .

*1 a
4 C
'. .. .. -

Figure 31. A pristine cave room-never seen by humans before. Florida State
Parks photo.
~"C' jC

g jl; 3Pr~Sm
~ili~ -
~~E "ilE, )

Parks photo.




~3 ~ ~"* CZm
*V C

I.." "- ; . O,


I h)

Figure 32. One of the newly discovered rooms at Florida Caverns has a grotto with a stream.
Florida State Parks photo.

.' .li
j****^ i 1

Figure 33. A recently discovered cave room at Florida Caverns.

Florida State Parks photo.


'> r t^ -' l-a'
j ri~ n ^B^ '". jiii


C:, ,~aJS CDs~r:


Figure i3o room is.u F. L"
Figure 34. A large column in one of the new cave rooms. Florida State Parks photo, 01

Figure 35. A newly discovered

is if 1 g '; *^ j

SjI Jlsi 11



cave room at Florida Caverns. Florida State Parks photo.



-. 1 -a ,

k .4

*i Kjb. Jlq"'.. s.(''w



Figure 36. Waterfall at Falling Waters State Recreational Area.
Florida State Parks photo.


Falling Waters State Recreation Area's main attraction is a sinkhole
with a waterfall (Figure 36). The sinkhole is cylindrical with sheer walls
that allows the waterfall to plummet 100 feet to its bottom. Associated
with the sink is a cave system which is part of a local underground
drainage system, connected to several adjacent sinkholes. Segments of
the cave system have been explored and mapped (Figure 37). Most of
the underground drainage network consists of a tortuous maze of
crawlways and squeezeways only a few inches wide and high, and

Figure 37. Map of cave system associated with waterfall.

2 / LEON -1 ISO 1
L T \ .......^^ ^ ..--.. s

r Y-1 i ,X" \
1u1 --l N -*N" 1 j"
BA -- A ,, A

4 I
1--- 20,o'. >
1 03
,CITRUn.. "A z
3---- 1 z

Figure 38. Locations of Florida's 27 first-magnitude springs or groups of springs having
average flows of 100 cubic feet per second or more. Modified from Rosenau,
et al., 1977, Figure 11, p. 39.


Table 3. The 27 first-magnitude springs and spring groups of Florida
with discharges, representative temperatures, and dissolved
solids. A first-magnitude spring discharges over 100 cubic
feet per second (cfs), or more than 64.6 million gallons per
day (mgd). Modified from Rosenau, et al., 1977, Table 2,
p. 7).

Average Range
(cfs) (cfs)

Spring and number
by county
(refer to figure 38)
Hornsby Springs
Gainer Springs
Chassahowitzka Springs
Cyrstal River Springs
Homosassa Springs
Itchetucknee Springs
Alapaha Rise
Holton Spring
Weeki Wachee Springs
Blue Springs
Wacissa Springs Group
Troy Spring
Alexander Springs
Natural Bridge Spring
St. Marks Spring
Fannin Springs
Manatee Spring
Blue Spring
Rainbow Springs
Silver Glen Springs
Silver Springs

Water Dissolved
temperature solids
C F (mg/L)

22.5 73 230

22.0 72 60

23.5 74 740
25.0 75 144
23.0 73 1,800

22.5 73 170

19.0 66 130





241 -578


101 -275








487 1,230










22. Falmouth Springs

158 60-220*** 21.0

70 190











Table 3. Continued.
Discharge* Average
Spring and number Average Range Water Dissolved
by county (cfs) (cfs) temperature solids
(refer to figure 38) C F (mg/L)
23. Blue Springs 162 63-214 23.0 73 826
24. Kini Spring 176 20.0 68 110
25. River Sink Spring 164 102-215 20.0 68 110
26. Wakulla Springs 390 25-1,190 21.0 70 153
27. Spring Creek Springs 2,003 ** 19.5 67 2,400
*Cubic feet per second (cfs): Multiply (cfs) x 0.646 = million gallons per day.
**Affected by tides.
***Reverse flow of 365 cfs measured on 2-10-33.

which takes many sharp-angled jogs. The many sharp changes of direc-
tion are the result of the preferential manner in which ground water has
dissolved the limestone along bedding planes, joints, and fractures in the


Throughout history humans have been fascinated by springs. And jus-
tifiably so, since they are sources of water, the necessity of all life. As
such, they also provide predators, including man, with convenient places
for stalking and ambushing prey. For primitive hunter-gatherer societies,
springs often became the nucleus of communities, a tendency that still
occurs today. Because of their importance to primitive people, some
springs were imbued with mystical, magical, or sacred status, such as
the sacred spring that is the origin of the Seine River. While such super-
stitions may be scoffed at today, many springs still enjoy favored status
as health spas, water recreation and fishing areas, tourist attractions,
local water supplies, or just for their serene, idyllic settings. Florida's
karst springs provide all of these benefits.
Florida has 300 known springs: 27 of first magnitude, 70 of second
magnitude, and 190 of third magnitude or less (Figure 38 and Table 3).
Combined estimated discharge from all 300 known springs is 12,600 cfs
or eight billion gallons per day. Although a world-inventory of springs is
not available, it appears that the major springs of Florida exceed in both
number and in quantity of water discharged those of other states or
nations. These springs are invaluable resources and assets. To date,
none has been found to be contaminated with pesticides, herbicides, or
metals (Rosenau, et al., 1977). Their generic nature, however, makes
them extremely susceptible to pollution and care must be exercised in
any activities planned near them. Rosenau, et al. (1977) described.in


detail, with photos, the known springs of Florida, so only brief descrip-
tions of selected examples will be given here.
Florida owes its national, and possibly world, leadership in spring
activity to karst. All of it major springs discharge from solution-riddled
limestones associated with the Floridan Aquifer, which underlies the
entire State of Florida (Rosenau, et al., 1977).
Springs can occur in any kind of rock. Florida's surficial and subsurface
rocks that contain springs, however, are of four types: sand, clay, lime-
stone, dolomite or some combination of them. Each spring occurs as the
result of a favorable combination of geological, hydrological, and topo-
graphical factors, and, in the case of limestones, the degree of solutional
weathering. Since this report deals with karst and, since all of Florida's
major springs occur in limestone, the small springs in sand or clay sedi-
ments will not be discussed.
Geological factors that influence the location and discharge of a spring
include the stratigraphy, structure, porosity, and permeability of rocks,
which were introduced in earlier sections on the hydrologic cycle, aqui-
fers, and evolution of karst terrains (see Figures 4, 5, 6 and 7). Strati-
graphy refers to the layered nature of rocks of differing types. Although
Florida stratigraphy is generally as shown in Figure 5, countless varia-
tions occur in thicknesses, sequences of layering, areal extent, and litho-
logic characteristics. Structure refers to the attitude of rock layers, e.g.,
flat, level, dipping, folded or distorted. Florida's rock strata are usually
relatively flat and level to gently dipping, as shown in Figure 5. Again,
deviations from this rule-of-thumb occurs. The structure and strati-
graphy of the rocks influences surface drainage which, in turn, erodes
the rocks to produce an area's topography.
Hydrological factors affecting the locations and discharges of springs
in Florida are the sizes and types of recharge areas and drainage basins,
topography, and rainfall. The most significant influence on a spring's
discharge is its local drainage basin. In those parts of Florida that have
little prominent relief, many streams' surface drainage divides are diffi-
cult to define and may not be very meaningful.
The presence of karst can further complicate the delineation of drain-
age basins. For example, Figure 39 shows the Oklawaha River drainage
basin, a tributary of the larger St. Johns River drainage basin. Also
shown are the drainage basins for Silver and Rainbow springs, that lie
mostly within it. Note, however, that the springs' basins do not coincide
with the river basin's boundary; both springs' basins lie partly within and
partly outside. This anomalous situation is the result of karst, which is
responsible for the springs' locations and the creation of their under-
ground drainage. The river's drainage basin boundary was drawn along
topographic highs, as is customary. However, the springs' drainage basin
boundaries could only be determined by analyzing surface and ground-
water levels throughout the basins (Faulkner, 1970).
This example demonstrates the contradictory problems that can occur
when dealing with surface or groundwater in karst terrain. For instance,



- Sprnng drainage area boundary

SOklawaha River drainage b 1in boundary, 2.870 square mles

Figure 39. Drainage basin of the Oklawaha River with drainage
basins of Silver and Rainbow springs superimposed.
Modified after Faulkner, 1970, Figure 22.

if one were trying to trace pollutant travel in surface waters, the drainage
system would be accessible to test or monitor. The same problem in
karst terrain may be impossible to solve because of a hidden, tortuous
maze of underground drainage, which may divert part of the surface
water in unknown directions in three dimensions.
Seasonal variations in flow is the norm for Florida springs, as well as
for most springs in the world. Each spring exhibits unique flow character-
istics, varying between extreme high and low flows, which are controlled
by local hydrological factors. Because rain is the ultimate source of all,


water for springs, the seasonal rainfall pattern for any area will greatly
influence the spring flows of that area.
The extent to which spring flow is dependent on rainfall is shown in
Figure 40. For the three years shown, rainfall was greatest from May
through August, with marked seasonality. Note the lag-time between
peak rainfall and peak spring flow, indicated by the vertical dashed lines.
From its seasonal low flow, Silver Springs' discharge only begins to
gradually increase after one or more months of high rainfall; the springs'
high flow occurs after the wet season has waned. Also note that the
configuration of the graph showing water level in the Floridan Aquifer
well closely duplicates the spring's flow graph; both gradually rise, peak,
then gradually decline. This demonstrates another point-the springs's
water is released slowly from storage in the limestone aquifer's voids.
This example illustrates a sequence of events that is important in
understanding the groundwater phase of the hydrologic cycle in Florida.
To recapitulate: rain percolates downward to replenish the groundwater
aquifers, raising their water levels. This takes time, often measured in
weeks or months. Increased volume of stored groundwater causes the
normal flow through the aquifers to increase. Further delays of weeks or
months due to the travel time from recharge areas to points of discharge
(a spring in this example) are manifested in the lag-time shown on Figure
40. After the seasonal highs, the spring's flow gradually decreases as
the groundwater reservoir is depleted. Protracted droughts can cause
groundwater depletion to the point where springs and surface streams
cease to flow.
Some springflows vary from nothing, or near zero, to flood propor-
tions. For example, Wakulla Springs has been recorded as varying
between a low of 25 cubic foot per second (cfs) (a relative trickle) to a
flood of 1,910 cfs (16 million to 1,234 million gallons per day) (Table 3).
This means the high flow was more than 76 times as large as the low
flow. These happen to be the extremes of flow measured between 1904
and 1974; longer periods of record will undoubtedly show an even wider


Florida has 16 known submarine springs, all originating in limestone,
some of which have multiple vents (Figure 41). Most are within a mile of
shore; some are in bays or tidal stream channels. However, four springs
lie farther offshore. Crescent Beach submarine spring lies 2.5 miles off-
shore in about 55 feet of water; Mud Hole submarine spring lies 13.8
miles south of Sanibel Island lighthouse in 43 feet of water; Bay Hole
Spring is about 20 miles offshore in 38 feet of water; Red Snapper Sink is
located about 25 miles east of Crescent Beach in water approximately
88-feet deep.
Most of the springs' waters have been analyzed, and all but one dis-
charge relatively fresh-to-brackish water (Rosenau, et al., 1977). The


r r I,, r r rl rI rI ,III I rI I I 11I I l u l l I I II

MZ Discharge of Silver Springs
O: C near Ocala
'Z J 1000

~a.. 800

"0 il- < I l1

600 I
3 : 5 5 .0 . . . . I i, , ' ' ' ' 1
Water level in an Ocala well that
Staps the Floridan Aquifer r

L 57.5 -

l-z 60.0 _

a 62.5 ______
"'J _i 0

65.0 I1
6 5. I I I I I I I I
_j I0 Monthly rainfall at Ocala -
I 1-- -

I (No Record) I
J F M A M J J A S 0 N D J F M AM J J A S 0 N D J F M A M J J A S 0 N D
1966 1967 1968
Figure 40. Interrelation of rainfall, water levels in the Floridan Aquifer, and discharge of Silver Springs. Note
lag-time between maximum rainfall and peak discharge of the springs (dashed lines). Modified
after Rosenau, et al., 1977, Figure 10, p. 28.


Figure 41. Floridan Plateau with submarine springs. The Floridan Pla-
teau is encompassed by the 300-feet depth contour line.
Modified from Uchupi, 1967. Submarine springs from Rose-
nau, et al., 1977.

one exception is Red Snapper Sink, The spring is about 160 feet in
diameter at the sea bottom and is probably more than 465-feet deep
(Rosenau, et al., 1977). A dye-dispersion test indicated a slight down-
ward velocity of seawater, suggesting that this vent may be a point of
seawater intrusion into the Floridan Aquifer (Kohout, et al., 1975).
Similar phenomena from other studies suggest that such karst features
on the Floridan Plateau may be common; more of them just haven't been
discovered yet. Benjamin (1970) reported on his scuba exploration of 54
"blue holes" offshore on Andros Island. Andros Island is the largest of
the Bahama Islands, lying 140 miles southeast of Miami. It is the emer-


gent portion of the Great Bahama Bank, a carbonate platform of more
than 14,000 feet of limestones, that rise precipitously from depths of
over 12,000 feet in the Atlantic Ocean. Besides the 54 he explored,
Benjamin charted several hundred other locations that he felt might
prove to be blue holes. The graphic name "blue hole" describes the
appearance of the deep submarine pits when seen from above: dark blue
against the crystal clear, light green, shallow water, from three to about
20-feet deep. He explored one hole to 230-feet in depth before his flood-
light went out. Flash pictures revealed what appeared to be massive cave
Benjamin (1970) theorized that these blue holes formed in Andros
Island's limestone at times of low water, during the last great Ice Age,
probably as caves or sinkholes on what was then dry land. The melting
glaciers raised sea level, flooding the karst features, forming the blue
holes. The island's fresh watertable also rises and falls in cycles through
the island's karst drainage.
Benjamin also collaborated with Jacques Cousteau (1971) in exploring
blue holes near Andros Island. Dye and current meter tests proved that
some of the blue holes, at least, have direct connections with similar
karst features on the nearby islands.
Jacques Cousteau (1971) also explored an enormous blue hole that is
1,000 feet in diameter and 412-feet deep, located on Lighthouse Reef,
British Honduras. Cousteau's divers encountered vast caves 120 feet
below the surface that were filled with large stalactites, some more than
20-feet long. They found caves at the bottom of the blue hole that had
small stalactites, indicating that sea level had once been more than 412
feet lower than present, since stalactites can only form in air. A stalactite
that had fallen from the ceiling of the caves at the 120-feet depth was
laboratory age-dated to be about 12,000 years old, providing evidence
that sea level has risen more than 120 feet in 12,000 years.


Quicksand is a hydrological phenomenon that may be found where
there are springs. Quicksand is, as defined by Webster's New Collegiate
Dictionary: "Sand readily yielding to pressure; esp., a deep mass of loose
sand mixed with water, into which a person or heavy object sinks."
Quicksand is a phenomenon which, fortunately, many people never
encounter. However, when it is encountered it can be an unnerving expe-
rience at best, and at worst, potentially dangerous. An examination of
the facts will dispel the mystery and the ideas of exaggerated danger that
have become synonymous with quicksand.
Quicksand can only occur under certain circumstances. First, there
must be loose sand. Sand composed of well-sorted (all grains about the
same size), clean, fine-sized, rounded grains tend to become "quick"
more easily than coarser types. There must be a source of upward flow-
ing water, such as a spring. These conditions are more usual along


Figure 42a. Under ordinary conditions sand grains are packed
against each other, forming a mutually self-
supporting structure. Sand in this configuration
can support a load.

B |*,r- | ,; -. -Y:l'. ., .
B |J -: I L^: -I(7 IQ
Figure 42b. Water (blue arrows) moving upward through the
sand fills the pores, buoys the grains and tends to
"float" them apart. Under these conditions the
sand cannot support a load-any heavy object
placed on its surface will gradually sink through
the sand-it has become "quick." Separation of
grains is exaggerated for clarity.


streams and beaches, and it is in these places that most quicksand
Figure 42a shows sand grains under ordinary conditions, either
exposed at the surface level or underwater. The grains form an interlock-
ing structure which can support a load on its surface. Under these condi-
tions, if a load is placed on the sand, such as a person walking, the grains
may shift position slightly to accommodate the load, but there will not be
any significant change in the density of the sand-column.
If water moves upward through the sand with sufficient velocity, as
from an underlying spring, quicksand conditions may be created, Figure
42b. The sand grains may be forced apart, partially suspended, and
"floated" or buoyed up. In a sense, the sand column becomes "fluid,"
and it will not support weight. Any heavy object placed on the sand will
sink. The object's sinking speed will be determined by the local condi-
tions, e.g., the quantity and velocity of the water flow, size and shape of
the sand grains, and the size, shape, and weight of the object. Generally
speaking, though, if a person places their foot upon an area of quicksand,
the sand gives way instantly. Most such quicksand occurs as small,
shallow pockets, and the plummet is only a few inches before firm bot-
tom is reached. It would be a surprise, but unless the person fell or
twisted a leg, it would not be a dangerous experience.
Quicksand is usually restricted to small areas where an underlying
spring maintains an upward flow. If the upward flow decreases below a
certain critical amount, the "quick" sand will gradually revert to firm-
ness. Local factors will, therefore, control the occurrence, the duration,
and the extent of quicksand. For example, during dry seasons springs'
discharges and watertables may fall to such low levels that they cannot
maintain quicksand conditions at their points of discharge.
Quicksand cannot be detected simply by looking for it. Under ideal
quicksand-forming conditions, the upward flow is too slow to disturb the
surface or the water may seep into the surrounding sand before reaching
ground surface. In either circumstance, the quicksand area is not notice-
ably different from the surrounding ground-until weight is placed upon
it-and it gives way. If upward seepage is too fast, the sand will "boil,"
or create wash-outs, which will be noticeable.
Figure 43 illustrates two generalized situations that favor the forma-
tion of quicksand. It must be emphasized, however, that specialized local
factors will determine whether quicksand actually will form.


Little Salt Spring and Warm Mineral Springs

Several Florida springs have proven to be windows into the past. Their
waters have acted to preserve fossils and archeological artifacts, which
continue to provide remarkable clues to the State's natural and cultural
histories. One, Wakulla Springs in the north has already been discussed.


A i
Figure 43a.

Figure 43b.

Quicksand may be formed in the bed or along the banks of
a stream by the upwelling springs that discharge from the
porous limestone or sand. Depending upon local condi-
tions, quicksand may form anywhere in the stream's bed.
Blue arrows denote water moving through the limestone
and sand. All material below the watertable is saturated.

Quicksand may form along a beach due to local conditions
that create springs. In a possible situation shown here,
groundwater (blue arrows) migrating through loose sand is
diverted upward by the denser sand or clay units. If upward
seepage is fast enough, areas of quicksand may form in the
area with the darker pattern.


Two others lie near the southwest Gulf coast in Sarasota County; they
are Warm Mineral Springs and Little Salt Spring (Figure 44).
Of any Florida spring, Little Salt Spring has been the most intensively
studied and the most prolific repository of fossils and artifacts. Until the
late 1950's, Little Salt Spring was thought to be just another shallow
pond, when divers discovered it to be a flooded sinkhole of impressive
dimensions (Figure 45). In 1974, a non-profit research organization was
founded to provide support for the professional exploration and study of
the site. Some of the more spectacular discoveries from the spring and
nearby bog to date include (Biggs, 1976; Clausen, et al., 1979):

1. The shell of an extinct, giant land tortoise, 12 feet in diameter, with
a sharp-pointed wooden stake embedded in it, which apparently
was used to kill it. The tortoise shell and wooden stake provided
radioactive dates of 13,450 and 12,030 years before present (BP),
respectively. The 1,000 years difference between the dates can be
accounted for by the possible age of the tortoise. The life span for
such a giant species could have been several centuries, before it
was killed by a paleo-lndian. This fossil and the weapon are evi-
dence for the earliest known human activity in Florida.
2. An Archaic Period cemetery with an estimated 1,000 burials in an
adjoining slough. Radiocarbon dates of three bone samples gave
dates from 6,180 to 5,220 years BP.
3. Parts of more than 30 human skeletons have been removed from
the spring.
4. Fossils of bison, camels, horses, a mastodon, and giant sloth, along
with bones of smaller animals.
5. Boomerangs, throwing sticks, atl-atl (a spear-throwing device), and
stone projectile points. One of the non-returning boomerangs may
be the oldest specimen of this kind of weapon in the world and is
the first found in the Western Hemisphere.
6. Portions of a well-preserved brain were found in the skull of one
skeleton from the cemetery. Another human brain preserved in a
skull was recovered from Warm Mineral Springs; its enclosing sedi-
mentary sequence was radioactive carbon dated at 10,630 to
8,500 years BP.

Based on such geological and archeological evidence from the spring,
it has been inferred that the climate of south Florida was much drier
12,000 years ago than it is today. World temperatures were cooler and

I I i




11 6 mies


Figure 44. Location map of Little Salt and Warm Mineral springs. Warm
Mineral Springs is a resort development that is accessible
from Route 41. Little Salt Spring is not accessible to the




.1 -

awA IIk wA t llb l


- -70,

-. 210' P UNKNOWN --

Figure 45. Cross section of Little Salt Spring. The circular upper pool
has a diameter of approximately 250 feet, sloping gently
down to the circular, 80-feet wide orifice at 40-feet depth.
The stepped-back overhanging ledges at 40 and 70 feet
have stalactites that formed when water levels were lower
than their respective depths below ground. Illustration
from Science, February 16, 1979, cover, copyright 1979
by AAAS. Used with permission.



Figure 46. Location map of Hornsby and Darby springs.

sea level was much lower (Clausen, et al., 1979). Under these conditions
most of Florida's sinkholes were probably cenotes, which are large, cav-
ernous sinkholes with water at depth. At that time the water level in
Little Salt Spring cenote was near the 90-foot ledge (Figure 45), where
the giant tortoise had apparently been killed, overturned, and cooked in
its shell by paleo-lndians. The Little Salt Spring archeological site is off-
limits to the public.

Hornsby and Darby Springs

The findings at Little Salt and Warm Mineral springs corroborate earlier
archeological finds in north Florida at Hornsby and Darby springs, Ala-
chua County (Figure 46). Investigations at Hornsby and Darby springs
were sponsored by the Florida Geological Survey in 1951 and 1952. One
sedimentary sequence that contained several paleo-lndian chert tools
associated with mastodon bones was carbon-dated as being 9,880
( 270) years BP (Dolan and Allen, 1961), which gives a minimum date
Tor human habitation at this north Florida site, and which closely agrees
with dated artifacts from Little Salt and Warm Mineral springs. The lime-
stone near the springs contain chert seams that provided the Indians


with the raw materials for stone tools. Judging from the numbers and
associations of the stone tools, projectile points, and pottery, they con-
ducted an extensive lithic industry at the springs, from about 10,000 to
3,000 years BP (Dolan and Allen, 1961). Hornsby Spring is inside the
grounds of a private camp and is not open to the public.

Recreational Use

Some 49 of Florida's 300 known springs (about 15 percent) are pres-
ently developed for recreational use. Several springs form the nucleus of
state and private parks or campgrounds, which have pools, bath houses,
concessions, and other facilities, where visitors can picnic, swim, boat,
and snorkel or scuba dive.
The water temperature of most springs varies only slightly throughout
the year, because their sources are underground waters which experi-
ence little, if any seasonal temperature variations. The range of tempera-
tures is between 70 to 79 OF (21 to 25.5 oC), with the majority averag-
ing closer to 75 OF (24 OC); these temperatures provide a cool,
refreshing swim.
The mention of "hot springs" brings to mind visions of steaming
springs where nobility and wealthy gentry congregate to "take the
waters" in splendid surroundings. The lore and lure of some hot springs
of the world has reached near-legendary proportions. "Spa" is the name
of a famous mineral-springs resort in Belgium, whose name has become
synonymous with any commercially developed spring which has water
that is hot or highly mineralized; odorous, sulfurous water seems to be
especially beneficial to both owners and patrons.
Florida has two known hot springs, Little Salt Spring and Warm Mineral
Springs. "Hot" is used here to signify that they have discharges of 80 OF
(26.5 OC) or hotter. Both springs are in Sarasota County (Figure 44). The
archeological significance of Little Salt Spring has been discussed in a
preceding section.
Warm Mineral Springs is about 13 miles southeast of Venice on U.S.
Highway 41, one mile north of the road. The spring is privately owned
and is developed into a public swimming and recreation area. The
spring's physical shape is the same as Little Salt Spring, as shown on
Figure 45. A comparison of dimensions shows Warm Mineral Springs to
be larger: its main orifice, at a depth of 43 feet below water level, is
approximately 170 feet in diameter (Little Salt Spring's is 80 to 100
feet); depth to the floor of the main chamber is about 240 feet (versus
200 feet). Comparative water quality data are given in Table 4.
These two springs may be closely related in age, genesis, and sources
of water. The source of their warm, saline, sulfurous water may be from
the 2,000 to 3,000-feet deep Boulder Zone (Rosenau, et al., 1977). The
route of the water from the Boulder Zone is not known, but Sproul, et al.
(1972) presented evidence of faulting near Ft. Meyers, 40 miles to the
southeast, which could provide avenues of upward migration. They stud-


Table 4. Water quality data for Little Salt Spring and Warm Mineral
Springs. Units are in milligrams per liter unless otherwise indi-
cated (Rosenau, et al., 1977).
(sampled 10/30/72) (sampled 4/24/72)
Calcium 180 500
Magnesium 130 580
Sodium 750 5,200
Chloride 1,300 9,500
Carbonate 0 0
Sulfate 510 1,700
Fluoride 1.8 1.9
Nitrate .53 -
Strontium 28 31
Dissolved solids (calculated) 3,000 18,000
pH (units) 8.0 7.3
Temperature (OF/C) 81/27 85/29.5
81 to 88 OF 73 to 99 OF

ied old, abandoned irrigation wells in the area, 1,000 to 1,500-feet deep,
that were flowing saline, sulfurous water. Regarding chloride concentra-
tions, it was determined that they could range from 15,000 to 20,000
mg/L in aquifers below 1,400 feet in Lee County. Although they could
not pin-point the exact source of the mineralized water that was contam-
inating the shallow aquifers, their evidence led them to conclude that
upward leakage may be occurring along the faults that connect the deep
strata with shallow strata. A well in Charlotte County, 23 miles south-
east of Little Salt Spring, produced 96 OF (35 OC) water from about
1,600 feet (Sproul, et al., 1972). These findings tend to corroborate
Kohout's (1965) theory as to the source of some of the springs' water.
It may be, therefore, that Little Salt and Warm Mineral springs occur as
hot springs as the result of a combination of geological circumstances
that allows anomalously hot and mineralized water to migrate upward.
The springs' karst drainage systems probably extend deep enough to
intercept aquifers whose confining beds have been breached by faults,
allowing the invasion of deep, mineralized water. Since the deeper aqui-
fers are under artesian pressure, some of the water is then diverted
through the springs. The springwater is not as mineralized because it is
diluted by fresher water from shallower zones that also supply the


The Floridan Plateau is a relatively flat platform that forms the eastern
side of the Gulf of Mexico basin (Figure 47). The emergent part of this
plateau is peninsular Florida. The plateau's boundaries are placed at
water depths of 300 feet. At Ft. Myers, its edge lies some 100 miles


Figure 47. Floridan Plateau with submarine karst features. The Floridan
Plateau is encompassed by the 300-feet depth contour line
(modified from Uchupi, 1967, Figure 1). Cross section
below shows the broad "platform," of which the Florida
peninsula is the part presently above sea level.


Figure 48. Generalized cross section across the
Floridan-Bahama Platform. Modified after
Banks, 1964, Figure 1.

offshore; at southeast Florida near Miami, it only lies two to three miles
offshore. Geologically, this platform is a layer-cake of rock units. Older,
deeper rocks are predominantly carbonates, while younger, shallower
units are mixtures of clastics (sands, clays, gravels) and carbonates (Fig-
ure 48).
Available evidence indicates that the floor, or "basement," upon
which these thousands-of-feet of sedimentary strata have been depos-
ited are granites, basalts, or similar types of volcanic or metamorphic
rocks (Applin, 1951), some of which have been dated as old as 634
million years old (Milton, 1972). At some time during that ancient era,
carbonate rocks began to accumulate on top of the basement rocks that
formed the floor of a shallow sea, much like the reef environment of


south Florida today. While these carbonates were being deposited the
basement rocks of this proto-Florida were slowly subsiding. Conditions
similar to these persisted for millions of years, because all of the carbon-
ate strata originated under relatively shallow marine conditions, indica-
ting that depositional rates approximated subsidence rates (Banks,
1967). These rocks and their included marine fossils also show evidence
for many rises and falls of global sea levels, between then and now.
The most recent episodes of major sea level fluctuation occurred dur-
ing the past million years or so, when the Pleistocene glaciers and conti-
nental icecaps removed so much water from the hydrologic cycle that
sea level dropped hundreds of feet, exposing broad expanses of the
continental shelves. Whenever sea level dropped low enough to expose
the rocks on the platform, weathering and solution of the carbonate
could begin to form karst. Given long enough periods of low stands of
sea level, extensive karst drainage systems could have formed. Then, as
now, the karst drainage would have extended outward beneath the parts
of the plateau above sea level, which could have been 200-miles wide in
places. Successive rises in sea level, combined with the gradual subsid-
ence of the plateau, have covered these older surficial karst features
(springs, sinkholes) with sea water or sediments. Though buried, they
should not be forgotten, for recent evidence indicates that some parts of
this ancient, underground karst system are still active components of
Florida's hydrologic system.
Oceanographic surveys have, by chance, found several sinkholes on
the submerged plateau. Jordan (1954) reported the discovery of three
large sinkholes 14-miles offshore on the Pourtales Terrace, south of Key
West, at 900-feet depths (Figure 47). Each was more than one-half-mile
wide at the top, from 450 to 540-feet deep, and appeared to be free from
sediment-fill. Jordan (1954) thought these karst features formed when
the terrace was emergent and later submerged. Malloy and Hurley
(1970) recognized karst-like topography on the Miami Terrace (Figure
47), which they interpreted to be solution and collapse features formed,
or at least maintained, by water from the Floridan Aquifer discharging
subaqueously into the Straits of Florida.
Similar discoveries indicate that this type of ancient karst may not be
unique to the Floridan Plateau-indeed, it probably exists in many coastal
regions around the world-wherever ancient carbonate rocks crop out
beneath the sea. Robb (1982) reported evidence that undersea discharge
of groundwater during periods of lower sea level may have eroded cliffs
and terraced valley walls on part of the lower continental slope, about 70
miles offshore New Jersey, at depths of over 6,000 feet. Some karst-like
features of the submarine topography suggest that solution of the car-
bonate rocks also took place.


The Boulder Zone

Generally lying at depths of greater than 1,200 feet beneath peninsular
Florida is an extensive zone of solution cavities (Vernon, 1970). This
zone is distinct from the surficial and shallow karst features that pres-
ently concern citizens of the State. This zone of cavernous porosity is
called the Boulder Zone. It is an erroneous name given by early oil-
exploration well drillers; the Boulder Zone does not contain any boulders.
Rather, it is a zone of very hard and dense dolomite (Vernon, 1970). As
drills encountered the roofs of cavities, large chunks of rock (boulders)
broke off. By getting between the drill bits and walls of the bore hole
these fragments created drilling problems, including severe vibration and
damage as the rotating drills bounced and rolled over them. Many holes
had to be abandoned because the drillers could not make progress once
they encountered the Boulder Zone. Figure 49 shows the porous nature
of the dolomite. Figure 50, taken with a special down-hole camera,
shows vugs (small cavities) and tunnels. Some of these deep oil test
wells have encountered caverns, the two largest being 90 and 100 feet
from ceiling to floor (Puri and Winston, 1974). This zone extends to
depths greater than 5,530 feet in south Florida (Vernon, 1970). The
horizontal and vertical distribution of cavities and caverns seems to be
random; drillers cannot predict where they will encounter them. Vernon
(1970) reported that this "Boulder" zone of cavernous, high transmis-
sivity appears to respond as a part of the Floridan Aquifer; i.e., they are
hydraulically connected.
Kohout (1965) postulated geothermally heated convection currents as
a driving mechanism to circulate sea water through the deep zones of
high transmissivity in south Florida, depicted in Figure 51. Referring to
Figure 51: geothermal heat from deep within the Earth's mantle is trans-
ferred through the low-permeability Cedar Keys anhydrite (calcium sul-
fate) beds, (1, on figure), raising the temperature of the water in the
lower Floridan Aquifer system and generating thermal convection circu-
lation. The warmed, less-dense aquifer water rises, pulling in cold, dense
sea water that has access to the aquifer through karstic, cavernous
dolomite that crops out in the Straits of Florida (2). The rising convection
flow of sea water eventually contacts seaward-moving fresh water,
which was recharged to the aquifer from the karst region of cental Flor-
ida (3). After contact and mixing, the diluted sea water moves seaward
and may discharge through the upper part of the aquifer either by upward
leakage through confining beds into shallow aquifers, or by discharge
through submarine springs and seeps on the continental shelf and slope
in the Straits of Florida (4). More recent data from deep oiltest wells in
the Florida Keys and Gulf of Mexico show that the Boulder Zone extends
some distance westward from the peninsula (Vernon, 1970). If Kohout's
theory is valid, then it is possible that his convective circulation model
shown in Figure 51 may have a mirror-image to the west, with hydraulic
connection to the Gulf.


. .-.



Figure 50. Photograph of 2436-feet level in Sun 32-3 Red Cattle well
(W-6929, Fl. Bureau of Geology), Hendry County. Right
side, upper hole: 8" cavity; lower hole: 14" X 14" tunnel.
Scattered vugs in walls. The oblate shape of the hole is
caused by the drill bit being deflected when it encoun-
tered cavities and "boulders," making it wobble. After
Puri and Winston, 1974, Figure 65.

Other investigations lend weight to Kohout's model. It is normal to find
increasing temperatures of rocks with depth. A temperature survey run
on an oil-test well, Sun Oil Co. Red Cattle No. 32-3, Hendry County,
T45S, R29E, section 32, showed a temperature gradient of one degree
Fahrenheit increase for each 122 feet of depth, between 900 to 2,100
feet. From 2,100 to 3,300 feet, through the Boulder Zone, the gradient
was less steep, being only one degree for each 480 feet of increasing
depth, with a bottom-hole temperature of 100 oF (37 oC).
Meyer (1974) reported on a well near Miami that was drilled 2,927-
feet deep, into the Boulder Zone. This well was used to inject secondarily
treated sewage effluent into the Boulder Zone. This investigation indi-
cated that locally the Boulder Zone water has the chemistry of sea water
with a temperature of 60.4 oF (15 OC). The water's chemistry and tem-
perature, along with the regional geology, suggest that the Boulder Zone

.* -0o
BEDS r 1000 -
BEDS .---------- F---- '-10
.. 2000


___- soo a

Figure 51. Idealized cross section through Miami showing concept of geothermally induced convec-
tion currents. Modified from Kohout, 1965, Figure 10.


crops out in the Straits of Florida and is hydraulically connected to the

Uses of Deep Zones of High Transmissivity

With explosive population growth and urban development comes
increasing demand for potable water and the attendant problem of waste
disposal. Deep zones of high transmissivity may prove helpful in solving
both problems.
Florida has sufficient fresh water to supply all of its needs. Water
supply problems arise because population centers are distant from
sources of fresh water. Also, Florida's rain occurs in seasonal patterns,
which do not always coincide with peak demands. The solution to the
water supply problem, then, is to get the water to the population centers
in sufficient quantity and when it is needed. Relevant to these problems,
several research projects have studied possible uses for these deep
The water in the Floridan Aquifer system at these depths is brackish to
saline (Klein, 1975). Although the water in deep zones may be non-
potable, the zones themselves may prove to be exploitable, beneficial
natural resources-their cavernous porosity represents an enormous
potential storage reservoir.


Five projects in south Florida have tested the possibility of using deep
carbonate aquifers that contain non-potable water as storage zones for
surplus fresh water (Figure 52) (Merritt, et al., 1983). In theory, during
periods of surplus water supplies, freshwater is injected into non-potable
aquifers, where it is stored for a period of time, then pumped out for use
when local water supplies are deficient. Theoretical studies and results
of test projects indicate that efficient recovery of stored potable water
improves in the first few inject-store-recover cycles, provided only pota-
ble water is recovered in each cycle (Merritt, et al., 1983).
Advantages of this subsurface storage concept are: (1) storage space
is cost-free, (2) it can be conveniently located directly under a water-
treatment plant or near a water-transport system, (3) no water is lost by
evaporation or evapotranspiration.
Some disadvantages are: (1) cost of testing to ascertain if local, favor-
able hydrogeological conditions exist, (2) construction of deep wells, (3)
cost of operating a system of cyclic injection wells.
These five projects have begun what must be a long research program
to determine if this concept may be one factor in cost-effective solutions
to south Florida's increasingly complex and stressful water needs. Some
of the tests have shown limited, local success. Similar projects need to
be done in other areas of south Florida to determine local hydrogeological




02 Fresh-water injection test well (Merritt
et al., 1983).

- Salt-water disposal wells into the Boulder
Zone (from Florida Bureau of Geology
records). Some locations have more than
one well.

0 50 MILES

Figure 52. Map showing Boulder Zone injection wells. Modified from
Merritt, et al., 1983. Lines delineate the primary study area
of Merritt's report.


conditions. Only when such basic information is available can govern-
ment officials make strategic plans to alleviate water-supply problems.


Thirteen salt-water injection wells into the Boulder Zone are in use in
south Florida (Figure 52). These are used in conjunction with oil produc-
tion wells.
Popular terminology ascribes the source of oil and natural gas to under-
ground "pools." In reality, liquid petroleum (crude oil or crude) is con-
tained in the pores and voids in certain rocks. In many cases, such as in
south Florida, crude oil, gas, and salt water occur in the same rocks
(Figure 53). Through time the oil, being immiscible in water, floated
upward, forming separate layers. Gas, being lightest, migrated to the top
of the reservoir rocks.
In October 1985, there were 84 producing oil wells in south Florida.
The production zone is from 11,322 to 11,892 feet below sea level. For
each barrel of oil produced (42 U.S. gallons), the reservoirs yield 100
standard cubic feet of gas and about six barrels of salt water (Florida
Bureau of Geology records). Oil and gas are valuable commodities, but
the salt water is a waste product that must be disposed of. Disposal of
the salt water present problems since its chloride content of approxi-
mately 160,000 parts per million (ppm) make it some six-times "saltier"
than Gulf water, of 20,000 ppm chlorides.
The Boulder Zone, which causes so many problems for oil well drillers,
now is used to solve a problem when oil is found. Many exploratory oil
wells do not find oil; they are "dry." However, when oil is found, then
some of these dry holes are used to inject the waste brines into the
Boulder Zone, which, with its cavernous porosity can accept practically
limitless quantities.


As explained earlier, the primary causative agent of karst formation is
the chemical dissolution of carbonate rocks by acidic water. It was
pointed out that most rain is naturally somewhat acidic because it com-
bines with carbon dioxide in the air. But other, man-made, sources of acid
in the atmosphere have caused concern in recent years.
Air pollution-in the form of acid rain or the deposition of particulates
capable of forming acids-has become a major, world-wide environmen-
tal problem. Some natural processes, such as volcanic eruptions, forest
fires, and decomposition of plants, contribute considerable amounts of
acid-forming pollutants to the atmosphere. Present concern, however, is
with the ever-increasing amounts of acid rain resulting from industrial
pollution, especially emissions from electric generating plants that burn
fossil fuels. Some particulate and gaseous by-products of burning fuels
interact with constituents in the atmosphere to become dilute sulfuric

I I _I -- -- exist in the pores. Producing
- - -- -wells bring up a mixture of m
S --- oil, brine, and dissolved gas, 0
SBOULDER -^ which are separated near the
ZONE wellhead.


1\ Mi

Curvature of beds ZONgreatly exaggerated.
::................ . . . . .

W i....:.. ............ *


Not to scale

Figure 53. Generalized cross section showing oil production and brine injection in south Florida.
Curvature of beds is greatly exaggerated.


and nitric acids, eventually returning to earth as acid rain or acid snow.
During periods of intensive air pollution the acidity of rain can increase 10
to 30 times above normal (LaBastille, 1981).
Today, most of the United States, including Florida, has to contend
with air pollution. If air pollution is causing rain to be more acidic and,
since acidic water dissolves carbonate rocks to create karst, does that
imply that Florida might conceivably experience increased rates of sink-
hole formation?
Pertinent to this question, a recent report by the Florida Department of
Environmental Regulation (FDER, 1984) presented the following points
with respect to the state-of-knowledge of acid rain in Florida:
1. Florida is experiencing acid rain, with 1981-82 pH values ranging
from 4.3 in the panhandle to 4.7 in south Florida. So-called "pure-
rain" has a pH of about 5.6 (de Pena, 1982).
2. There is evidence that the acidity of Florida rain has increased over
the past 20 years, but the trend is not clear.
3. The sources of air pollutants contributing to Florida's acid rain are
not well-defined. However, it seems likely that much of the pollu-
tants are from in-state utility power plants.
4. There is little evidence of damage to crops in Florida from acid rain.

Because there are so many unknowns about acid deposition and its
effects in Florida, the Florida Congressional Delegation initiated the Flor-
ida Acid Deposition Study in 1981 to attempt to address unresolved
issues. Scheduled for completion in 1985, the study is focusing on
(FDER, 1984):
1. the collection of high quality precipitation chemistry data through a
statewide monitoring system;
2. source attribution analysis through a variety of modeling
approaches, including mass balances and back trajectory analysis,
3. an analysis of potential ecological effects.

One of the questions that the FDER study believed should be
addressed in future research projects was: Does acid deposition affect
structural stability of soils for road beds and does it influence sinkhole
When the primary factors involved in karst formation are examined and
placed in perspective, there does not seem to be much reason to expect
that Florida will experience a measurable increase in the frequency of
sinkhole formation due to foreseeable increases in air pollution or acid
Time-geological time-on the order of hundreds-of-thousands or mil-
lions of years, is on our side. Consider Florida's karst today and be aware
that it has taken hundreds-of-thousands of years to produce the karst
features we see. If the acidity of groundwater were increased by 10
times, and this became "normal," then instead of 100,000 or more


years to produce a sinkhole it might take only a few tens-of-thousands of
years. Experience from other parts of the United States, Canada, and
Europe has shown that, long before we have to worry about increased
karst hazards, we will have to contend with sterile lakes, streams, and
possibly extensive damage to forests and crops. And these effects will
take place in the course of a few tens-of-years; a person's lifetime. The
greatest danger to mankind's short-term future from acid rain, then, is
the potential decrease in food supplies, as well as other associated health
hazards-not an increase in karst formation.


Karst areas present enigmatic problems relative to certain basic human
activities, such as construction, and the provision of water supplies and
waste disposal, which are critical to any modern, technological society.
Property owners, planners, government officials, and engineers require
information or quantitative data that relate to potential problems in order
to assess the chances for success for projects or plans. While most
karst-forming processes take place underground and out of sight, it is
still possible to study them. Techniques that are used can be classified as
direct or indirect.

Direct Methods

Until recently it was only possible to use direct methods to study karst.
Scientists and adventurers studied karst first-hand-they climbed and
crawled into caves and sinkholes. This method of study is limited by the
size of openings in the caves and by water which often floods the under-
ground drainage. While this direct, exploratory approach provides impor-
tant background information which has been useful in formulating theo-
ries to explain how and why karst forms, it is a hindsight approach. One
can only study what has happened after a sinkhole collapses, for exam-
One of the oldest, most straightforward approaches to evaluating a
specific site for karst hazards is drilling. In this approach, holes are
drilled, just as for water wells, except the object is to locate cavities. This
is still the accepted engineering standard for assessing a site's founda-
tion qualifications. The number, locations, and depths of test holes are
specified depending on the expected types and depths of soil and
bedrock, size of building, and type of proposed foundation.
If cavities are found, they can sometimes be sealed by pumping them
full of cement, or they may be bypassed by setting pilings into solid
bedrock to support the structure. Economics and other engineering con-
siderations will dictate the method used to secure the foundation. Occa-
sionally, though, drilling into a cavity produces more immediate prob-
lems, such as when the drill rig begins to sink into the subsiding hole
(Figure 12).


These approaches have limited predictive capability. One cannot say
with any degree of certainty where or when the next karst feature will
appear at the surface, or, for that matter, the extent or location of karst
drainage channels.

Indirect Methods-Remote Sensing

Indirect approaches to study karst have been used with the hope that a
technique would be found to evaluate an area with respect to detecting
and predicting potential danger from karst. No perfect technique has
been found. Some techniques, however, have shown promise, either
alone or in combination with other methods. At present, detection and
prediction of karst hazards is mostly state-of-the-art, with new technol-
ogy and techniques promising to increase the success rate.
As the name implies, remote sensing refers to techniques that indi-
rectly obtain information about an object or place some distance from a
station that is remotely located. Six remote sensing techniques that are
most commonly used today are discussed below. The field of remote
sensing is ever-expanding because of rapid advances in technologies of
electronics, optics, satellites, and sensors.


Some geophysical methods operate on the principle of detecting and
monitoring changes in natural properties of the earth, such as gravity.
Other geophysical methods rely on monitoring changes in man-induced
signals into the soil or bedrock, such as ground penetrating radar, sound
waves, and electrical currents. Regardless of the method, the main
objective is to try and detect and quantify any anomalies in the subsur-
face, which may signify and locate underground cavities. Four of the
more common geophysical methods that have been used successfully to
identify and locate underground karst features are gravity surveys, seis-
mic refraction, ground penetrating radar, and electrical resistivity.


In this method of geophysical investigation, extremely sensitive instru-
ments, gravity meters, are used to measure the strength of local gravity.
The force of earth's gravity is not the same at every place on its surface.
The force of gravity varies because of (1) difference in elevation, which
changes the distance from the center of the earth; (2) effects of the
earth's speed of rotation, which changes with latitude; and (3) variations
in density of under-lying rocks. Usually, the effect of each of these fac-
tors is very small in going short distances from point-to-point in local
areas, such as investigating a site for a building foundation.
This technique can be limited by the fact that cavities may be so small,
relatively speaking, that the instrument cannot detect the tiny, local dif-


ference in density between rock with cavities and rock without cavities.
Also, the instruments are very sensitive to vibrations, a factor that can-
not be overlooked when working near highways or urban areas.

Seismic Refraction

In this geophysical technique seismic energy waves are transmitted
into the subsurface to determine the depth to and thickness of rock
strata, depth to the water table, and anomalous features such as cavities
(Figure 54). Seismic energy waves travel at different velocities in materi-
als that have different densities and hardness. The waves are refracted,
or bent, at the interfaces between strata of differing composition. An
array of microphone sensors, called geophones, that are implanted in the
topsoil detect the reflected and refracted seismic waves. Electronic pro-
cessing of these signals produce patterns that indicate subsurface fea-
tures (Figure 55).
Portable seismic systems can use a sledgehammer to generate low
energy seismic waves for very local, shallow investigations. Automated
portable seismic sources can be used for deeper or production work. The
depths to which seismic techniques are effective are generally limited
only by the amount of transmitted energy; the more energy sent, the
deeper the strata penetrated. An inherent weakness of this technique is
its sensitivity to extraneous vibrations, particularly if used near noisy
urban areas.

Ground Penetrating Radar

Ground penetrating radar (GPR) is a reflection technique using high
frequency electromagnetic radiation (Figure 56). GPR surveys produce
graphic profiles of subsurface conditions that resemble the side walls of
trench cuts. The reflections shown on the radar record are produced as a
result of contrasts in the complex dielectric constant of individual, sub-
surface materials. Electronic processing of the reflected energy waves
reveals the configuration of strata, the watertable, cavities, buried pipes,
and other subsurface features. With ideal conditions GPR can penetrate
up to 100 feet or more and produce a very graphic representation of
strata (Benson and Glaccum, 1979). Figure 57 shows an old sinkhole
that is filled with clean quartz sand with only a subtle depression at the
surface. The cone-shaped profile indicates the extent of sinkhole activity
and the zone of potential hazard.
Distinct advantages of GPR over most other geophysical techniques
are: (1) its continuous nature of operation as the antenna is moved
slowly across the ground surface; (2) mobility of equipment can make
total site coverage economically feasible, not just spot-checked as with
other methods; and (3) on-the-spot analysis of data because of the
picture-like radar presentation.
Since it uses electrical energy, GPR can be severely limited by some

/\ .PLATE >




Figure 54. Field layout of a 12-channel seismograph showing the path of direct and refracted seismic waves in a
two-layer soil/rock system. Hammer blow sends a trigger pulse that starts recording of signals picked
up by geophones. Used with permission of Benson and Glaccum, 1979.

. . . . .i"
z 5 F ._ ;i

0:1 N... 'iii "
8 i
0u 9


50 100 150 200 250 300 350 400 450
Figure 55. Recording from a 12-channel seismograph. All 12 channels were recorded simultaneously from
a single hammer impact. Used with permission of Benson and Glaccum, 1979.





5 -300
0 D 0 D







Figure 56. Block diagram of ground penetrating radar system. Radar
waves are reflected from soil/rock interface, electronically
processed, then recorded on tape and graphically. Used
with permission of Benson and Glaccum, 1979.

local conditions. The radar waves are attenuated by layers of fine-
grained materials, such as silts or clays. Groundwater with high concen-
trations of dissolved minerals, such as salt water, which is a better elec-
trical conductor than potable water, absorbs the radar's energy instead
of reflecting it to the sensors. Under these conditions radar penetration
can be limited to about three feet. While man-made electronic sources
can interfere with the radar signal, this is not usually a serious problem.



*r. ~ Si U lL* It

"I t

.; ...,
___ '** ~ TABLE I

*~ 2 0 I t .


-1 *, "
1 '

Figure 57. Ground penetrating radar traverse across a paleokarst site.
Used with permission of Benson and Glaccum, 1979.


Figure 58. Basic resistivity measuring configuration. Note distor-
tion of the current field near the anomaly, which has
different resistivity than the surrounding earth. Used
with permission of Benson and Glaccum, 1979.


Resistivity techniques measure the natural resistivity of soil, rocks,
and groundwater, and can be used to assess vertical sections and lateral
changes in subsurface materials.
The method uses arrays of electrodes (usually steel stakes) in different
configurations, which are set out along traverse lines across the site
being investigated (Figure 58). One pair of electrodes, the two C-
electrodes in Figure 58, injects an electrical current into the ground, and
the generated voltage field is measured between a second pair of elec-
trodes, the two P-electrodes in Figure 58. The injected current's distribu-
tion in the earth is determined by the relative high and low resistivities of
subsurface materials. Note in Figure 58 that the lines of current are
distorted near the anomalous feature, which may be a cavity.
Resistivity is calculated at each station using the measured voltage,
the applied current, and the geometry of the electrodes' placement. In
this manner a point-by-point subsurface profile is constructed. By com-
bining data from a grid of traverses a contour map can be made of
subsurface features. Figure 59 shows what such a map might look like,