FLRD GEOLOSk ( IC SUfRiW


COPYRIGHT NOTICE
[year of publication as printed] Florida Geological Survey [source text]


The Florida Geological Survey holds all rights to the source text of
this electronic resource on behalf of the State of Florida. The
Florida Geological Survey shall be considered the copyright holder
for the text of this publication.

Under the Statutes of the State of Florida (FS 257.05; 257.105, and
377.075), the Florida Geologic Survey (Tallahassee, FL), publisher of
the Florida Geologic Survey, as a division of state government,
makes its documents public (i.e., published) and extends to the
state's official agencies and libraries, including the University of
Florida's Smathers Libraries, rights of reproduction.

The Florida Geological Survey has made its publications available to
the University of Florida, on behalf of the State University System of
Florida, for the purpose of digitization and Internet distribution.

The Florida Geological Survey reserves all rights to its publications.
All uses, excluding those made under "fair use" provisions of U.S.
copyright legislation (U.S. Code, Title 17, Section 107), are
restricted. Contact the Florida Geological Survey for additional
information and permissions.












4~tr-~Yt


c.. -- ---- -.-..-
--vs4

r--- -


C-




'-1

r~rrm
:Ti~
r`"









.r= r ~~ni q~


-,,~-. I--M--~- -----~i;
"r..
: F~--plLri
.-~~.~ ~~E



"-`i~-.T
--R
`lr;


.+=;vr
----
---~i~~-r -i.-.



~-el-r"'--~f;ss+i~

~L ~db~_LP~CliL~i












je



1. r --..;;-- ~u
-- -r-. ~--.-. .. 3~1
i-- -- --


~- ~-Y;L~~C~
-- .
Ir~L~
---s


~-n6a~


.--.-;-- -- -~ '~-



w-~-

-Y; -i~-_~-i
C ---- -----r-~- -~r-;-
~





i~--- -;;-;;-c~;.--,l -LiiL


~c~------ '-~F~-~-





.c~~

~- ~nf=1












s;3 ~
IT~L~

'-1





'jl


---i -- --=--;r



r--
;( ---..I

.1 .





c-~
~"-"r ;.-=: ~*


~FCC"'I






STATE OF FLORIDA
DEPARTMENT OF NATURAL RESOURCES
Harmon Shields, Executive Director




DIVISION OF INTERIOR RESOURCES
Charles M. Sanders, Director




BUREAU OF GEOLOGY
Charles W. Hendry, Jr., Chief



Report of Investigations No. 74



HYDROLOGIC CONSEQUENCES OF USING GROUND WATER TO
MAINTAIN LAKE LEVELS AFFECTED BY WATER
WELLS NEAR TAMPA, FLORIDA




by
J. W. Stewart and G. H. Hughes
U. S. Geological Survey




Prepared by the
UNITED STATES GEOLOGICAL SURVEY
in cooperation with
SOUTHWEST FLORIDA WATER MANAGEMENT DISTRICT
and the
BUREAU OF GEOLOGY
FLORIDA DEPARTMENT OF NATURAL RESOURCES


Tallahassee, Florida


1974





DEPARTMENT
OF
NATURAL RESOURCES






REUBIN O'D. ASKEW
Governor


DOROTHY W. GLISSON
Secretary of State






THOMAS D. O'MALLEY
Treasurer






RALPH D. TURLINGTON
Commissioner of Education


ROBERT L. SHEVIN
Attorney General





FRED O. DICKINSON, JR.
Comptroller






DOYLE CONNER
Commissioner ofAgriculture


HARMON W. SHIELDS
Executive Director






LETTER OF TRANSMITTAL


Bureau of Geology
Tallahassee
October 10, 1974


Honorable Reubin O'D. Askew, Chairman
Department of Natural Resources
Tallahassee, Florida

Dear Governor Askew:

We are pleased to make available the report, "Hydrologic Consequences of
Using Ground-water to Maintain Lake Levels Affected by Water Wells Near
Tampa, Florida" by J. W. Stewart and G. H. Hughes. This report should be of
substantial importance as it relates to the problem of urbanization around lakes
in Florida and the affects of urbanization on lowering of lake levels. It is
possible, under certain circumstances, that the lake levels may be artificially
maintained to offset the effects of pumping or below normal rainfall without
substantially increasing the loss of water by evaporation. This may well represent
an appropriate compromise in order to develop the ground-water resources and
also to maintain the surface environment associated with the lakes of Florida.

Respectfully yours,



Charles W. Hendry, Jr., Chief
Bureau of Geology



















































Completed manuscript received
July 29, 1974
Printed for the
Florida Department of Natural Resources
Division of Interior Resources
Bureau of Geology

Tallahassee
1974



iv








CONTENTS

Abstract ............................................................ 1
Introduction ........................................................ 2
Hydrologic setting .................................................... 5
Northwest Hillsborough County lakes ................................ 5
Section 21 well field ............................................. 13
History of pumping from wells into lakes .................................. 17
Round Lake ....... ............... .. .... ..... ......... ..... 17
Saddleback Lake ..................................... .......... 19
Lake Charles ................................................... 23
Maintaining lake stages ............................................... 23
Movement and distribution of water pumped into lakes ....................... 26
Hydrologic consequences .............................................. 31
Summary .......................................................... 38
References cited ..................................................... 41






ILLUSTRATIONS


Figure Page

1. Graph showing yearly rainfall at Tampa, Florida, 1947-71 (National
Weather Service Office) ........................ .................... 3

2- Map of northwest Hillsborough County showing location of selected lakes .. 5

3. Map of Round Lake showing depths ................................ 6

4. Graph showing relation of lake level to lake storage ................... 7

5. Map of Saddleback Lake showing depths ........................... 8

6- Map of Lake Charles showing depths .............................. 10

7. Oblique photograph of Section 21 well-field area taken July 1971 from
elevation of 2,000 feet................ ......................... 11

8. Graphs showing pumpage at Section 21 well field, water levels in shallow
and deep wells, and level of Starvation Lake ......................... 12

9. Section showing water table and potentiometric surface of Floridan Aquifer
on May 21, 1971, depth of lakes, and thickness of surficial sediments in the
Section 21 well-field area................................... ... 14

10- Diagrammatic section showing general relation between lake level, water
table of shallow aquifer, and potentiometric surface of Floridan Aquifer in
the Section 21 well field before and after the development of the well-field 15

II. Graphs showing pumpage at Section 21 well field, pumpage into Round
Lake, rainfall, and level of Round Lake ............................ 18

12- Photographs showing level of Round Lake at 50.0 feet and 53.5 feet, before
and after adding ground water ................................... 20

13. Graphs showing pumpage at Section 21 well field, pumpage into Saddleback
Lake, rainfall, and level of Saddleback Lake ......................... 21

14. Photographs showing level of Saddleback Lake before and after adding
ground water ................................................. 22

15. Graphs showing pumpage at Section 21 well field, pumpage into Lake
Charles, rainfall, and level of Lake Charles ........................... 24

16. Graphs showing annual rainfall and pumpage into selected lakes in the
Section 21 well-field area ......................................... 25

17. Map of Section 21 well-field area and water table of the shallow aquifer,
May 21, 1971 ............................................... 28







ILLUSTRATIONS Continued

18. Map of Section 21 well-field area and potentiometric surface of Floridan
Aquifer, May 21, 1971 ......................................... 30

19. Diagrammatic sketch showing how lake area varies with lake level ........ 32

20. Graph showing relation between lake area and increase in lake evaporation
attributable to the arrest of the decline in lake level ................... 34




TABLES

Table Page

1. Water-budget data for Round, Saddleback, and Charles Lakes, May 1971. 29

2. Selected chemical constituents of water from lakes, Floridan Aquifer,
and mixtures of waters from lakes and Floridan Aquifer ............ 36










HYDROLOGIC CONSEQUENCES OF USING GROUND WATER TO
MAINTAIN LAKE LEVELS AFFECTED BY WATER
WELLS NEAR TAMPA, FLORIDA


By
J. W. Stewart and G. H. Hughes
U. S. Geological Survey


ABSTRACT

Levels of several homesite lakes about 12 miles north of Tampa were lower
than usual during the last several years. Rainfall was substantially less than
normal, changes in the drainage system diverted storm runoff away from some
lakes, and ground-water levels and therefore lake levels, because these lakes
are hydraulically connected with the aquifer system were lowered locally by
pumping at a municipal well field.

Extreme declines in lake levels were forestalled since 1966 at Round Lake
and since 1968 at Charles and Saddleback Lakes by pumping water from the
Floridan Aquifer into the lakes. Levels were maintained at a nearly constant
stage slightly below the altitude at which the lakes begin to overflow and
generally above the altitude that the lake levels would have attained naturally;
that is, an attempt was made to compensate for the effects of (1) pumping for
public water supply and (2) below normal rainfall. Combined pumpage into the
three lakes averaged 765 gpm (gallons per minute) or 1.1 mgd (million gallons
per day) during 1971. During 1971, rainfall was about equal to lake evaporation
so the water pumped into the lakes was used entirely to replace water that
leaked from the lakes into the local aquifer system. Almost all leakage eventually
returned to the Floridan Aquifer. Less pumpage would have been required if the
lake levels had been permitted to fluctuate through a range about equal to the
natural range in stage for lakes in the same general area.

Evaporation from the three lakes was somewhat greater than would have
occurred if the lake levels had continued to decline without the addition of
ground water. Assuming that without the addition of ground water lake levels
would have declined until the lake areas were reduced by 50 percent, and that
annual evapotranspiration from the exposed lake bottoms would have been 18
inches less than lake evaporation, the estimated increase in lake
evaporation attributable to maintaining the lake levels against the effects of
pumping for public water supply and below normal rainfall would have been
25 gpm for the three lakes or 3 percent of the ground-water input for 1971. If
lake levels had been maintained at a variable stage compensating only for
i well-field pumping but not for below normal rainfall, only part of the increase in







BUREAU OF GEOLOGY


lake evaporation that part associated with maintaining lake levels against the
stress of less than normal rainfall would be considered a reduction of the local
water resource. Thus, in the hypothetical example above, perhaps less than half
the 25 gpm can be considered a reduction of the local water resource.

Although maintaining levels of lakes at constant and relatively high stages
increases the potential for lakes to overflow, because of sudden storms of high
magnitude, Round, Charles, and Saddleback Lakes through 1971 had not lost
any appreciable quantity of water by overflow since their levels have been so
maintained. During this time, of course, rainfall was generally less than normal.
A larger quantity of water may be expected to overflow when rainfall is greater
than normal, but during such times the need for water will not be critical.
Adding ground water to the lakes caused a distinct change in the chemical
quality of the lake water, generally characterized by increases in calcium,
bicarbonate, hardness, and alkalinity. However, many lakes in Florida are
naturally supplied with the same kind of ground water; hence, any long-term
ecological changes are not expected to be drastic or harmful. Further study will
be needed to document whether any long-term ecological changes are likely to
occur.

Maintaining levels of lakes in the well-field area at higher than natural
levels causes an increase in leakage from the lakes which in turn may lead to an
increase in the permeability of materials under the lakes if dissolution of
limestone is accelerated. Hence, for the same climatic and hydrologic conditions,
the pumpage required to maintain lake levels at a given elevation may increase
with time. The increase in required pumpage to date (1971) doubtless has been
small. Continued monitoring of pumping rates will assist in determining whether
required pumpage will increase markedly for times when climatic and hydrologic
conditions are nearly the same.

INTRODUCTION

Lake levels in the vicinity of Tampa were lower than usual during the last
decade, largely because of deficient rainfall. From 1961 through 1971, the
yearly rainfall at the National Weather Service Office in Tampa averaged more
than 8 inches below normal and was above normal only in 1964 and 1969, as
shown in figure 1. The problem of low lake levels was worsened in some
instances' by changes in the natural drainage system caused by highway
construction, land reclamation, and urban development. Such changes diverted
part or all of the storm, runoff away from some lakes. The problem was also
aggravated locally by the lowering of ground-water levels caused by pumping at
the Section 21 well field of the city of St. Petersburg. The lakes in this area are
hydraulically connected with the aquifer system.










80
RAINFALL
Norm I (1930 60), 51.57 inches
0 70 -


z 60


- --- ---- -- -
- 50 -

z

2 -

30 _- 2


20 J- -----.------J----
o 0 o o o c o -
it 0 CD t Wp U) D 0
0) m m m 00) O) ) 0)
_=~~~6 6 6 = _=-= _


Figure 1 Graph showing yearly rainfall at Tampa, Florida, 1947 71 (National Weather Service Office).






BUREAU OF GEOLOGY


As early as 1963, in the Section 21 well field area in northwest
Hillsborough County, lake-side residents complained about excessive declines in
take levels. Between 1965 and 1968 the levels of some lakes declined as much as
10 feet. The decline in level of lakes near the Section 21 well field was
considerably greater than in those at a distance from the field. Therefore,
well-field pumping must be considered an important cause of the excessive
decline in lake levels in that area. The extent to which changes in the natural
drainage system may have affected lake levels in the Section 21 well field area
has not been ascertained.

In order to stop the progressive decline in lake levels, residents at each of
several lakes constructed wells near the lakes to pump water into them. The
levels of these lakes were maintained at a nearly constant stage slightly below the
elevation at which the lakes begin to overflow and generally above the elevation
that the lakes would have attained naturally.

The words "natural" and "naturally", as they pertain in this report to lake
levels and hydrologic conditions in general, mean the levels and conditions that
would prevail as a result of actual rainfall but without any effect of well-field
pumping and (or) of pumping water into the lakes.

The purpose of this report is to describe the hydrologic and geologic
conditions in the general area of the Section 21 well field where the lake-level
maintenance program is being carried on and to evaluate the hydrologic effects
of using water from the Floridan Aquifer to maintain the levels of lakes.

The investigation was done by the U. S. Geological Survey in cooperation
with the Southwest Florida Water Management District, as part of the District's
overall program of evaluating the water resources of west-central Florida, and
with the Bureau of Geology, Florida Department of Natural Resources, as part
of the statewide program to develop hydrologic relations.

For the convenience of those readers who prefer to use International
System (metric) units rather than English units, the conversion factors for terms
used in this report are listed below:

Multiply English unit By To obtain metric unit
inches 25.4 millimeters
feet .3048 meters
miles 1.609 kilometers
acres 4,047 square meters
gallons 3.785 liters
acre-feet 1,233 cubic meters
gallons per minute 6.309 x 10-5 cubic meters per second
million gallons per day .04381 cubic meters per second
square miles 2.590 square kilometers








REPORT OF INVESTIGATIONS NO. 74 5


HYDROLOGIC SETTING


NORTHWEST HILLSBOROUGH COUNTY LAKES


Round, Saddleback, Charles, and Crenshaw Lakes are privately-owned
homesite lakes in northwest Hillsborough County about 12 miles north of
downtown Tampa (fig. 2). The lakes are 500 to 6,500 feet east of the city of St.
Petersburg's Section 21 well field. Four lakes are owned by St. Petersburg within
the Section 21 well field: Jackson and Starvation Lakes in the northern part, and
Crum and Simmons Lakes in the southern part (fig. 2). The largest declines in

82'31' 82027'

Van Dy e Road
Crenhoaw
Do 'on Le0 amp?
.A /. 1 1 cZ=. LakLak
Ltake *ae S oddishoc **L -k
Laken Lake
28007' SECTION 21L/ Charl e s
WELL FIELD -

CAnorLon














Municipal supply well FLETCHER










REPORT I Le



< I I I I
Floridon oquifor obser-
vation wIll --











Figure 2 Map of northwest Hilsborouh County showing location of selected
LIMTlakes.
0 I MILE

Figure 2 Map of northwest Hillsborough County showing location of selected
lakes.


































0
-N-








EXPLANATION
-10- LINI OFr lMA. EPTH To mrM,
ELOw LAE IrrA1Ol R O 5.5 Pfr
WSOVS M.AN WA LV ~,JANUARY IIRIll
INTERVAL 6 PET
It ISDLATIO DPTH IAPYN I MINT IN FIT


Figure 3 Map of Round Lake showing depths.







REPORT OF INVESTIGATIONS NO. 74


Slake levels in the area occurred in Crum, Simmons, and Starvation Lakes. The
levels of Crum and Starvation Lakes declined so low that the lakes divided into
separate pools. Records were not available to determine the amount of decline in
Jackson Lake.

Of the privately owned lakes whose levels are being artificially maintained,
pumpage records and other data are available for only three, Round, Saddleback,
and Charles; the discussion in this report will be concerned chiefly with these.

Round Lake, about 500 feet east of the east edge of Section 21 well field,
has a surface area of 11.3 acres when its level is 53.3 feet above msl (mean sea
level). At this level the lake is 6 to 8 feet deep except for a dredged channel
around all but the south side of the lake (fig. 3). The channel, generally 10 to 20
feet deep, has three depressions 22 to 24 feet deep (29 to 31 feet above msl).





56
EXPLANATION
0 Saddleback Lake
Round Lake
A Lake Charles
-J 52

-.J

4
48 48

4
w

w 44


I-

S 40




36


to5 10

10 106 107 10
LAKE STORAGE, GALLONS


Figure 4 Graph showing relation of lake level to lake storage.


























EXPLANATION
LINE OF EQUAL DEPTH TO BOTTOM
BELOW LAKE STAGE OF 5.S PEET
ABOVE MEAN SEA LEVEL, AML IS, 0w
INTERVAL FEET
ISOLATED DEP.YH MEASUREMENT, IN FEET


0 600 FEET
I I I i I


Figure 5 Map of Saddleback Lake showing depths.






REPORT OF INVESTIGATIONS NO. 74


These depressions probably are sinkhole depressions caused by subsidence of
surficial sediments into solution openings that have developed in the limestone.


Round Lake has no surface outlet and virtually no surface inflow because
its drainage area is less than 0.01 sq mi (square mile). A drainage canal
constructed in June 1965 from Saddleback Lake to Round Lake carries overflow
from Saddleback Lake when the level of Saddleback Lake rises to 54.2 feet
above msl; however, this level is seldom reached.


At elevation 53.3 feet above msl Round Lake contains 31.5 million gallons
or about 96.7 acre-feet (fig. 4). A change in lake level of 0.1 foot at this
elevation represents a volume of about 370,000 gallons.

Saddleback Lake, about 1,000 feet east of Section 21 well field, has a
surface area of 31.2 acres when its level is 53.6 feet above msl. The drainage area
is about 1.5 sq mi. The lake is hour-glass shaped; at a level of 52.3 feet above msl
(fig. 5), the lake averages 8 to 10 feet deep near the centers of the two large
sections. In the narrow neck, the lake is 4 to 5 feet deep. Several sinkhole
depressions in the south part of the lake are 20 to 24 feet deep (28 to 32 feet
above msl). The south part of the lake was dredged in 1961 and again in 1970.
Saddleback Lake receives some water from Crenshaw Lake through a canal that
connects the two. Crenshaw Lake overflows into the canal at a stage of 55.1 feet
above msl.


The volume of water stored in Saddleback Lake at elevation 53.6 feet
above msl is about 60 million gallons or 184.1 acre-feet (fig. 4). At this level a
change in lake level of 0.1 foot represents a volume of about 1 million gallons of
water.


Lake Charles, about 1.2 miles east of Section 21 well field (fig. 2), has a
surface area of 12.6 acres at a level of 53.2 feet above msl. Lake Charles receives
an undetermined amount of surface inflow from a drainage area of about 0.08 sq
mi during wet periods. The lake averages 9 to 10 feet deep at a level of 52.5 feet
(fig. 6). Several sinkhole depressions in the lake are 16 to 19 feet deep. The lake
has not been dredged.


The capacity of the lake at a level of 53.2 feet is 30.0 million gallons or
about 120 acre-feet (fig. 4). A change in lake level of 0.1 foot at this level
represents a volume of about 410,000 gallons.






BUREAU OF GEOLOGY


p3;i
rr~g



) 9
REV


Figure 6 Map of Lake Charles showing depths.


Starvation Lake has a surface drainage area of about 0.3 sq mi and a
water-surface area of less than 25 acres at a lake level of about 41 feet above msl.
The lake has not been dredged. Throughout most of 1971 the lake level was so
low that the lake formed two separate pools. In contrast, the lake had a surface
area of about 50 acres at a stage of 53 feet above msl in 1956. A photograph of
the Section 21 well field in July 1971 (fig. 7) shows that both Starvation and
Crum Lakes consist of two separate pools. The stage of Round Lake is
maintained at near the tree line shown in the photograph by pumping ground
water into the lake.







































Figure 7 Oblique photograph of Section 21 well-field area taken July 1971 from elevation of 2,000 feet.










SLAME LEVEL, FEET ABOVE MEAN SEA LEVEL
uD


MONTHLY PUMPAG,
GIAUNO-WATER LEVEL, FEET ABOVE MEAN SEA LEVEL MILLION GALLONS PER
[ s 8 t f 8 a o












Le0
Or 0






REPORT OF INVESTIGATIONS NO. 74


SECTION 21 WELL FIELD

Six production wells are in the Section 21 well field. The wells are in two
lines that form a right angle. The wells are from 411 to 601 feet deep, cased to
depths ranging from 70 to 116 feet, and they produce water from the Tampa
Limestone and the Suwannee Limestone of the Floridan Aquifer. Pumping from
the well field began in February 1963. Pumpage averaged 12.6 mgd (million
gallons per day) in 1964, the first complete year of pumping. Since 1966
pumpage has averaged about 17 mgd, but has ranged from 4 to 24 mgd (fig. 8).
Stewart (1968) showed that pumping of the Section 21 wells affected the levels
of Round, Starvation, and Saddleback Lakes.


A section through the well field and the lakes illustrates the relation
between the water table and the potentiometric surface of the Floridan Aquifer
on May 21, 1971, and the depth of the lakes, and the thickness of the surficial
sediments in the area (fig. 9). The water table is above the potentiometric
surface of the Floridan Aquifer throughout the Section 21 well-field area.

Head difference between the water table and the potentiometric surface of
the Floridan Aquifer was greatest west of the Section 21 well field, and least
north of Lake Charles. Lake depths ranged from less than a foot to about 10
feet, however the sinkhole depressions of several lakes were more than 20 feet
deep. The thickness of the surficial sediments ranged from about 25 feet near
Lake Charles to more than 50 feet near Van Dyke Road in the northwestern parn
of the well field. However, the thickness of the surficial sediments from the
deepest parts of the lakes to the top of the limestone ranged from about 12 feet
at Lake Charles to about 40 feet at Starvation Lake.

The hydrology of the area and the overall effects of pumping at the
Section 21 well field were described in detail by Stewart (1968). Basically, two
aquifers are involved in the hydrologic system, a water-table aquifer surficiall
sediments) and the Floridan Aquifer (fig. 10).

The water-table aquifer consists predominantly of fine to medium silty
sand having an average saturated thickness of 35 feet. The depth to water ranges
from less than 5 to as much as 15 feet below land surface. Undisturbed core
samples collected at depths of 8 to 32 feet below land surface west of Round
Lake had an average hydraulic conductivity of 13 feet per day. An aquifer test
of the shallow water-table aquifer in the Section 21 well field also gave a
hydraulic conductivity of 13 feet per day, (Oral commun., Sinclair, W. C.,
1972).













A A
7d0. IF


0 2000 4000 6000 8000 10,000 12,000
Vertical exaggeration 80X DISTANCE, FEET'


Figure 9 Section showing water table and potentiometric surface of Floridan Aquifer on May 21, 1971, depth of
lakes, and thickness of surficial sediments in the Section 21 well-field area.


SEA







REPORT OF INVESTIGATIONS NO. 74


A. Before


B. After

Figure 10 Diagrammatic section showing general relation between lake level,
water table of shallow aquifer, and potentiometric surface of Floridan
Aquifer in the Section 21 well field before and after development of
the well field.


Water G I .-:-
Lake leveltj'~n~T~; ~'-




;..-.urf ici dim~nr ''




I I I 1 -1Floridan Aquiferr


..:..:: ..... :r". :
,.....

:.:::. : -.:::: -':;~:~:I :.: :-. :'.:
..c::.'.:.'.'..:::: : ~;. ...: ,. ...

.; /... -:: ~~.:,.-:?,:.
S*1*.*.*
:.mhhI.'.


I 1 I I I I I I I I I I I I


u~~T;T;;~F~T~T


A. Before


m


- I -I r I 1 r 1 1 -r


Fl o'rii do'n MUif or


~


I_ I I I -1 1 1 1 1 1 I I I I


I I I I 1 .1 1 q L I 1 1, I I I I I I --T


J-1~-ii? .~ Water ab


p-: ot~entiometric sur acei~
elcof~~r~~:
,,.,, Z) S Urficial sedimints `f,


I I I I I I I I I I I I L I I I I






BUREAU OF GEOLOGY


The principal aquifer in the area is the Floridan, an artesian limestone
aquifer interbedded with sand and clay in the upper part and becoming
dolomitic at depth. This report is concerned only with the upper part of the
Floridan Aquifer consisting of the Tampa Limestone and the Suwannee
Limestone because all known large producing wells in northwest Hillsborough
County are developed in these formations. The Tampa Limestone is about 100
feet thick in the Section 21 well-field area. The top of the Tampa Limestone
ranges from slightly below sea level to. about 20 feet above. The Suwannee
Limestone, which underlies the Tampa Limestone, is about 400 feet thick in the
area-

Water in the Floridan Aquifer is to some degree confined by overlying
materials which have a lower hydraulic conductivity than the limestone of the
Floridan Aquifer. The potentiometric surface an imaginary surface that
represents the level to which water will rise in tightly cased wells that penetrate
the Floridan Aquifer is above the top of the Floridan. Because the water table
of the shallow aquifer is above the potentiometric surface of the Floridan, water
moves downward from the shallow aquifer to the Floridan Aquifer primarily in
areas where the impermeable confining materials are discontinuous or are absent.

The loss of any substantial quantity of water from the lakes will depend
upon the following conditions: (1) large differences between the levels of the
lakes and the potentiometric surface of the Floridan Aquifer in the area; (2)
occurrence of permeable zones in the surficial sediments that will provide routes
for movement of water from the lakes to the Floridan Aquifer; and (3) a high
transmissivity of the Floridan Aquifer that will provide for rapid transmission of
water out of the area.

The above conditions are fulfilled in the Section 21 well-field area. First,
the potentiometric surface of the Floridan Aquifer is well below the water table.
The withdrawal of large quantities of water from the Floridan Aquifer from a
relatively small area results in large local drawdowns, thereby increasing the
differences in head between the lakes and the potentiometric surface of the
Floridan Aquifer. In May 1971, the potentiometric surface of the Floridan
Aquifer at the well field averaged 13 feet below the water table. Maximum
water-level differences occurred at pumped wells and ranged from 15 to more
than 60 feet. Second, both Round and Saddleback Lakes have been dredged, and
dredging of these lakes may have exposed permeable zones in the lake bottoms.
In parts of Round Lake, a dredged channel has several sinkhole depressions that
are 29 to 31 feet above msl. The south half of Saddleback Lake has several
depressions at 28 to 32 feet above msl or less than 10 feet above the limestone.
Lake Charles has several sinkhole depressions 35 to 37 feet above msl. Third, the
Floridan Aquifer has a transmissivity of 300,000 to 500,000 gpd (gallons per







REPORT OF INVESTIGATIONS NO. 74


day) per foot as determined from aquifer tests of wells in the Section 21 well
field. These high values of transmissivity indicate that the Floridan Aquifer is
capable of transmitting large quantities of water out of the area.

The rate at which water moves downward from the water-table aquifer
varies directly with the distance between the water table and the potentiometric
surface of the Floridan Aquifer. Pumping from the Floridan Aquifer at the
Section 21 well field lowers its potentiometric surface as indicated in figure 10.
This decline in water level temporarily increases the distance between the water
table and the potentiometric surface of the Floridan Aquifer and causes an
increase in leakage from the shallow aquifer to the Floridan Aquifer;
consequently, the water table declines below its natural level for a given climatic
condition.

In the vicinity of the well field, a direct hydraulic connection commonly
exists between the lakes and the water-table aquifer. Depending on climatic
conditions, water moves naturally from the water-table aquifer to the lake or
from the lake to the aquifer. When pumping from the Floridan Aquifer takes
place, the potentiometric surface is lowered; flow from the water-table aquifer
to the Floridan Aquifer increases, and the water table is lowered below its
natural level and more water than normal moves from the lakes into the shallow
aquifer (fig. 10). In addition, if a direct hydraulic connection exists between the
lake and the Floridan Aquifer, the lowering of the potentiometric surface of the
Floridan Aquifer induces more water to leak from the lake to the Floridan
Aquifer than otherwise would occur. As a consequence of increased flow from
the lakes to one or both aquifers, lakes within the area of drawdown must
decline below the level that would result naturally from a given climatic
condition.

HISTORY OF PUMPING FROM WELLS INTO LAKES

Ground-water pumpage into Round, Charles, and Saddleback Lakes was
used to maintain lake levels at a constant stage generally above the level that
would occur naturally. Considerably less water would have been required had
the lake levels been permitted to fluctuate through a range about equal to the
natural range in stage of lakes in the same general area.

ROUND LAKE

Round Lake is the smallest of three privately owned lakes in northwest
Hillsborough County into which water from the Floridan Aquifer is pumped to
stabilize lake levels. Lake-level records are available for Round Lake beginning
January 1965, when the lake level was 51.5 feet above msl. In June 1965 the








LAKE LEVEL,
FEET ABOVE MEAN SEA LEVEL








(0





0








B 0 2 I
2I'



5 -j
? *


MONTHLY
RAINFALL, INCHES
0o








a


a

a

0


MONTHLY PUMPAGE, MILLION GALLONS


p


0*
0


PER DAY


5 0







REPORT OF INVESTIGATIONS NO. 74


lake level declined to 49.1 feet and in June 1966 it declined to 48.2 feet (fig.
11). Water was first pumped into Round Lake on a trial basis for a few days in
June, July, and August 1965 when the city of St. Petersburg filled the lake with
water from one of the wells in the Section 21 well field. This trial demonstrated
the feasibility of using Floridan Aquifer water to raise the level of the lake.

In order to pump water from the Floridan Aquifer into the lake, residents
of the area constructed a well on the northeast side of the lake in June 1966.
The well is 6 inches in diameter, 395 feet deep, cased to 74 feet, and is equipped
with a 7 1/2 horsepower electric motor with a submersible pump that delivers
about 400 gpm. The well is.about 200 feet from the edge of the lake, and water
is pumped into a 15-inch corrugated iron pipe that slopes to the edge of the lake.
Pumping begins when the lake level declines to 53.3 feet above msl. The
water-level control is set for a fluctuation range of about 0.2 foot. In the absence
of natural inflow, the pump operates continuously for about 4 days, then shuts
down for a like period while the lake level declines. Routine pumping into
Round Lake began in June 1966. Since early 1967 the lake level has been
maintained at a stage of about 53.3 feet above msl ( fig. 11). The surface area of
the lake in June 1965 was about 9 acres, and early in 1967 it was about 11 acres.
A decline in lake level of 2 to 3 feet caused a drastic change in the esthetic
qualities of the lake (fig. 12).

From June 25, 1966 through December 21, 1971 (5 1/2 years), about
252.4 million gallons of water were pumped into Round Lake from the 6-inch
well. In 1971, pumpage averaged 143,000 gpd or about 100 gpm. In that year,
52.3 million gallons of water were pumped into the lake enough to fill the
lake 1.7 times from a completely empty condition to stage 53.3 feet above msl.

SADDLEBACK LAKE

Saddleback Lake is the largest of three lakes studied in northwest
Hillsborough County into which water from the Floridan is pumped to stabilize
lake levels. Lake-level records are available for Saddleback Lake beginning in the
latter part of April 1965 when its stage was 51.7 feet. Because of the downward
trend of the annual minimum lake levels in 1966 68 (fig. 13), residents around
the lake constructed, in June 1968, an 8-inch diameter well 445 feet deep (cased
to 79 feet), tapping the Floridan Aquifer. The well is between Round and
Saddleback Lakes and is about 300 feet west of Saddleback Lake and 400 feet
east of Round Lake. A turbine pump actuated by a 76-HP electric motor
delivers about 450 gpm.

Since the start of pumping on July 6, 1968, the lake level has fluctuated
between 52.4 and 54.4 feet above msl, and has averaged about 53.6 feet. This






BUREAU OF GEOLOGY


A. Before pumping ground water into lake, July 15,
1965. Lake level 50.0 feet above ,msl.


B. After pumping ground water into lake, September 22,
1965. Lake level 53.5 feet above msl.



Fiure 12 Photographs showing level of Round Lake at 50.0 feet and 53.5 feet,
before and after adding ground water.











LAKE LEVEL, FEET ABOVE MEAN SEA LEVEL


MONTHLY
RAINFALL, INCHES MONTHLY PUMPAGE, MILLION
0 o 0o h 0
L -


GALLONS PER DAY
0 5.


Li






0'


%a









p


0~
I0
'T


=- I-

L_
C I






BUREAU OF GEOLOGY


A. Before pumping ground water into lake, June 5,
1968. Lake level 47.0 feet above msl.


B. After pumping ground water into lake, May 11,
1972. Lake level 53.6 feet above msl.




Figue 14 Photographs showing level of Saddleback Lake before and after
adding ground water.






REPORT OF INVESTIGATIONS NO. 74


wide range in fluctuation was due partly to a rise in level caused by heavy rains
in 1968 and partly to a decline in level in 1970 when the pump was inoperative
for about 1 month. Otherwise, the lake level would have fluctuated within a
range of a few tenths of.a foot.

From July 6, 1968 through December 31, 1971 about 410.2 million
gallons of water were pumped into the lake from the Floridan Aquifer well. In
1971, pumpage into the lake averaged 468,000 gpd or about 325 gpm. This
volume of water was sufficient to fill the lake, assuming it was empty, nearly
three times to an elevation of 53.6 feet above msl. Figure 14 shows how the
addition of ground water has changed the lake's appearance.


LAKE CHARLES

Lake-level records are available for Lake Charles beginning in September
1965, when the lake level was at 52 feet. The annual minimum lake level trended
downward in 1966 1968 (fig. 15), reaching a stage of 43.6 feet in June 1968.

Lake-side residents drilled a 485-foot well (cased to 48 feet) in June 1968
to control the lake level by pumping water from the Floridan Aquifer into the
lake. The well is about 100 feet northwest of Lake Charles. A turbine pump
actuated by a 7%-HP electric motor delivers about 500 gpm.

Pumping into the lake began August 3, 1968. Since that time the level of
the lake has fluctuated between 51.2 and 53.7 feet above msl and has averaged
53.2 feet above msl (fig. 15). Through December 31, 1971, about 410 million
gallons of water were pumped from the Floridan Aquifer into Lake Charles. In
1971, pumpage into the lake averaged 492,000 gpd or about 340 gpm. The
volume of water pumped in 1971 was sufficient to fill the lake 4.5 times to an
elevation of 53.2 feet above msl.


MAINTAINING LAKE STAGES

The rate at which water must be pumped from the Floridan Aquifer to
maintain the levels of Round, Saddleback, and Charles Lakes at controlled
elevations will not be the same each year but rather will vary inversely with
rainfall (fig. 16). The quantity of water required to maintain the level of the
lakes in 1969 1971 ranged from 9,000 to 13,000 gallons per day per acre of
lake surface for Round Lake; 7,000 to 14,000 for Saddleback Lake; and 22,000
to 39,000 for Lake Charles. Total pumpage into the three lakes ranged from 217
million gallons in 1969 to 401 million gallons in 1971.







MONTHLY
LAKE LEVEL, FEET ABOVE MEAN SEA LEVEL RAINFALL, INCHES MONTHLY PUMPAGE, MILLION GALLONS PER DAY





8.I

i4 e *a o
goo

& *i
r
5 I I
.B Io :






REPORT OF INVESTIGATIONS NO. 74


80



260
-5
-J

z40


20


40


30


20


10


1967 1968 1969 1970 1971


Figure 16 Graphs showing annual rainfall and pumpage into selected lakes in the
Section 21 well-field area.

Even though the difference between lake levels and the potentiometric
surface of the Floridan Aqufier (fig. 9) was much smaller for Lake Charles than
for Round and Saddleback Lakes, ground-water pumpage into Lake Charles (per
unit area) was 2 to 3 times greater than for the other two lakes. The hydraulic
conductivity of the material between the lakes and the Floridan Aquifer through


LAKE CHARLES



ROUND LAKE
-




SADDLEBACK
LAKE
I I I







BUREAU OF GEOLOGY


one or more sinkholes is much more permeable for Lake Charles than for Round
and Saddleback Lakes.

During 1967 69, rainfall in the area generally increased and pumpage into
Round Lake decreased. Since 1969 pumpage into the three lakes has trended
upward, but the trend has been greater for Charles and Saddleback Lakes than
for Round Lake. For example, from 1969 through 1971, pumpage increased 45
percent at Round Lake, 80 percent at Lakes Charles, and 110 percent at
Saddleback.

Pumpage of ground water into the lakes was substantially greater in 1971
than in 1969 because in 1971 ground-water levels and lake levels were lower
throughout the area. An analysis of water-level data for Lake Crenshaw and
several shallow and deep wells indicates that in 1971 water levels in the general
area not only were well below those of 1969, but also were below the levels of
1967. It would be expected, therefore, that pumpage into Round Lake would be
significantly greater in 1971 than in 1967. However, this was not the case
because about 52 million gallons of water was pumped into Round Lake in 1971
as compared to 58 million gallons in 1967. The decrease in pumpage into Round
Lake is explained by the fact that in 1971 the level of Saddleback Lake was
controlled by pumping whereas in 1967 it was not. In 1967 the level of
Saddleback Lake averaged about 2 feet lower than the controlled level of Round
Lake but in 1971 the level of Saddleback Lake averaged about 0.3 foot higher
than the level of Round Lake. Thus, in 1971 eastward ground-water flow from
Round Lake was blocked by ground-water movement from Saddleback Lake as a
result of the higher lake level at Saddleback Lake. If Round Lake and
Saddleback Lake are considered as one lake, the increase in pumpage from 1969
to 1971 was 90 percent versus 80 percent for Lake Charles.

Thus, the trend of pumpage from 1969 to 1971 in itself does not
necessarily indicate that a progressive increase in pumpage will be required to
maintain the lake levels. If water had been pumped into Saddleback and Charles
Lakes in 1967, pumpage would have been substantially greater in 1967 than in
1969; hence, between 1967 and 1969 the trend in pumpage would have been
downward just as was indicated for Round Lake. On the assumption that the
pumpage from Section 21 well field has leveled off at an average of 18 to 20
mgd, pumpage into the lakes will decrease once rainfall in the general area
returns to normal.

MOVEMENT AND DISTRIBUTION OF WATER PUMPED INTO LAKES

In 1971 aggregate pumpage from all three wells was 401 million gallons
(1.1 mgd or 765 gpm). This quantity of water is equivalent to 22.5 days of







REPORT OF INVESTIGATIONS NO. 74


pumping from Section 21 well field, based on the 1971 average withdrawal rate
of 17.8 mgd. Rainfall at Starvation Lake gage was about 50 inches for the year
and presumably rainfall added about as much water to the lakes as was lost by
evaporation (estimated as 50. inches per year). Since none of the lakes
overflowed during the year, the ground water that was pumped into the lakes
only replaced water that leaked through the lake bottoms.


The general direction of movement of the ground water in the Section 21
well-field area is indicated by the flow lines (arrows) drawn perpendicular to the
contours in figure 17. Regional ground-water flow in the water-table aquifer into
Section 21 well-field area is from areas north and northeast of the well field. The
water table was highest (56 feet above msl) about a mile northeast of the well
field, and lowest (38 feet above msl) south of Crum Lake near the south-central
part of the well field. On the assumption that depth and permeability of the
surficial materials are uniform, flow is greatest where the contours are closest
together and least where the contours are farthest apart.

On May 21, 1971, water-table gradients in and near the Section 21 well
field were steepest on the south and west sides of Round Lake, and ranged from
about 0.01 to 0.005 foot per foot. Most of the lateral ground-water movement
from Round Lake was west toward Starvation and Crum Lakes, where the
ground water recharged the Floridian Aquifer.


From Saddleback Lake water moved laterally in all directions from the
lake into the water-table aquifer, but gradients were steepest to the south, and
most of the flow was south toward a water-table depression centered about
3,000 feet south-southeast of the lake. Water that moved eastward and
northeastward from Saddleback Lake eventually reached this same water-table
depression. Some of the northerly flow from Saddleback Lake moved
northwestward into Section 21 well field.

Lateral ground-water movement from Lake Charles was generally to the
southwest into the water-table depression south-southeast of Saddleback Lake.

Localized cones of depression are produced by wells pumping into the
lakes. Therefore, part of the water that moves into the Floridan Aquifer
probably is intercepted by the wells during these pumping periods.

The quantities of water moving into and through the water-table aquifer
from Round, Saddleback, and Charles Lakes were computed by use of Darcy's
law expressed as Q = KIA,



















I






.n/ S0 1000 2000 FEET

EXPLANATION
Q Municipal supply well Observation well
a Lake gage and stage
-48 Water-table contour. Contour Interval:2 feet.
Datum is mean sea level
Flow lines Indicate direction of ground-water
movement
Figure 17 Map of Section 21 well-field area and water table of the shallow aquifer, May 21, 1971.







REPORT OF INVESTIGATIONS NO. 74


where

Q = quantity of water, in cubic feet per day
K = hydraulic conductivity of the aquifer, in feet per day,
I = hydraulic gradient, in feet per foot, and
A = cross-sectional area through which water moves,
in square feet.

The hydraulic conductivity of the aquifer is 13 feet per day and the aquifer's
thickness is 35 feet. Corresponding values of the hydraulic gradient and width of
cross-section were estimated for each lake from the contours on figure 17. The
computations show that on May 21, 1971, 86,000 gallons moved laterally from
Round Lake into and through the water-table aquifer, 89,000 gallons from
Saddleback Lake, and 77,000 gallons from Lake Charles.

Based on the assumption that the computed rates at which water moves
laterally from the lakes to the water-table aquifer represent the average flow
rates from May 1971, the rate at which water moves vertically from the lakes
through the shallow aquifer to the Floridan Aquifer can be determined by the
water-budget method. For example, during May 1971 total pumpage into
Round, Saddleback, and Charles Lakes was 1,410,000 gpd. The change in lake
levels was virtually negligible for the month; hence, the change in volume of
water stored in the lakes was negligible. The lakes did not overflow nor did they
receive any surface-water inflow. Estimated lake evaporation for May was 0.33
foot greater than the monthly rainfall as measured at Starvation Lake;
multiplying this value by the total area of the three lakes sets the net loss of
water to the atmosphere at 190,000 gpd. The total amount of water that moved
from the lakes to the water-table aquifer was computed to be 252,000 gpd. The
only unknown item in the water budget for the lakes was the quantity of water
that moved vertically from the lakes to the Floridan Aquifer. For May 1971, this
quantity was 968,000 gpd. Water-budget data for individual lakes are given in
table 1.

TABLE 1
Water-budget data for Round, Saddleback, and Charles Lakes, May 1971
(Quantities in thousands of gallons per day)
Net loss Lateral flow Vertical flow
Pumpage Evaporation from lakes to from lakes to
into minus water-table Floridan
Lake lakes rainfall aquifer Aquifer
Round 194 39 86 69
Saddleback 574 106 A 89 379
Charles 642 45 77 520
TOTAL 1,410 190 252 968

























0 2








0 1000 2000FEET EXPLANATION
I i J Munlc!pql supply well o Observation well
4- Flow lines Indloate direction of ground-water movement
36"- Potentlometric contour. Contour Interval: 2 feet, Datum
Is mealon ea level

Figure 18 Map of Section 21 well-field area and potentiometric surface of Floridan Aquifer, May 21, 1971.







REPORT OF INVESTIGATIONS NO. 74


During May 1971, about 69 percent of the water pumped into Round,
Saddleback, and Charles Lakes returned directly to the Floridan Aquifer. An
additional 6 percent of the water flowed through the water-table aquifer from
Round Lake to reach the Floridan Aquifer in Section 21 well field. About 13
percent of the water was lost by evaporation from the lakes. The remaining 12
percent of the water moved from Saddleback and Charles Lakes into the
water-table depression south-southeast of Saddleback Lake and probably
returned to the Floridan Aquifer. The water-table depression south-southeast of
Saddleback Lake apparently is caused by leakage from the water-table aquifer to
the Floridan Aquifer (Oral comm., Sinclair, W. C., 1972). In 1972 several test
holes were drilled to map the extent of this depression. The test-hole data
indicated a good hydraulic connection between the water-table aquifer and the
Floridan aquifer. Detailed maps of the water table indicated that the flood
control ditch (Channel F) is not responsible for the depression.

The distribution of the pumpage from 1971 (765 gpm) probably was
about the same as that for May 1971 except thatthe net loss to the atmosphere
(evaporation minus rainfall) was negligible for the year; therefore, all the
pumpage must be distributed between leakage to the Floridan Aquifer and
leakage to the water-table aquifer. On the assumption that the leakage for 1971
was proportional to the leakage for the different lakes and aquifers in May 1971
(table 1), about 79 percent of the yearly pumpage leaked directly to the
Floridan Aquifer and 21 percent moved laterally into the water-table aquifer.
Almost all of the water that moved laterally to the water-table aquifer eventually
reached the Floridan Aquifer.

Figure 18 shows the potentiometric surface of the Floridan Aquifer in the
Section 21 well-field area on May 21, 1971 when the wells there were pumped at
about 20.4 mgd. Most of the water moved into the well-field area from the north
and northeast from the same general areas shown for the water-table aquifer
on figure 1-7. The direction of movement of water into the well field proper was
toward a low in the potentiometric surface (26 ft above msl). The location of
this low generally coincided with the low water levels of the water-table aquifer,
indicating a good hydraulic connection between the water table and the
potentiometric surface of the Floridan Aquifer.

HYDROLOGIC CONSEQUENCES

If water is pumped into a lake, the lake level is higher and the lake area is
larger than either would have been had water not been added, and the lake level
is also higher in relation to ground-water levels in the adjacent and subjacent
aquifers than it otherwise would have been. Consequently, (a) leakage from the
lake increases; (b) lake evaporation increases; and (c) the potential for the lake






BUREAU OF GEOLOGY


to overflow during wet periods increases. Any of the above may cause a
reduction of the local water resource.

Whether the increase in leakage causes a reduction of the local water
resource depends on whether the water is locally retrievable for beneficial use. In
1971 almost all of the water pumped into Round, Charles, and Saddleback
Lakes eventually returned to the Floridan Aquifer at or generally east of Section
21 well field where the water was retrievable. Because the water is retrievable,
leakage from these three lakes constitutes a negligible reduction of the local
water resource.







REPORT OF INVESTIGATIONS NO. 74


Evaporation from a lake is increased by pumping of ground water into the
lake because the resulting lake area is greater than it would have been had water
not been pumped into the lake. Data are lacking for an accurate appraisal of the
increase in evaporation but the relative magnitude of quantity of evaporated
water can be reasonably estimated. The quantity of water that evaporates from a
lake is proportional to the lake area, which, because of the basin-like shape of
the lake bottom, decreases as the lake level declines (fig. 19). If the input of
ground water maintains the lake level at Ho when it otherwise would decline to
H1, the resulting increase in evaporation is expressed as follows:

V = (Ao A) (E E') (1)

wherein

V = increase in volume of water evaporated from lake (L),
Ao = lake area at level H (L2)
Al = lake area at level H1 (L2),
E = depth of water evaporated from lake surface (L),
and E' = depth of water evaporated from lake bottom that would
be exposed if lake declined to level H1 (L).

According to estimates of lake evaporation by Kohler, Nordenson, and
Baker (1959), yearly lake evaporation in the Tampa area averages about 50
inches. According to Cherry, Stewart, and Mann (1970), evapotranspiration in
the Middle Gulf area north of Tampa is about 38 inches per year. For a
condition where the water table is lowered by pumping, evapotranspiration from
the bare and sandy exposed lake bottoms is presumably less than it would be for
the Middle Gulf area as a whole; hence, the value of E' is arbitrarily assumed to
be 32 inches per year.

On the basis of equation 1 and an estimate of 18 inches per year for the
factor E-E', the yearly increase in lake evaporation that results from arresting the
decline in-lake level is determined as indicated in figure 20. For example, if the
decline of a lake level is arrested at a given level and if at that level it has an area
of 10 acres, and if the continued decline in level would have reduced the lake
area by 50 percent, on the average, the increase in lake evaporation attributable
to the act of arresting the decline in level would average about 4.6 gpm. If the
continued decline in level would have dried up the lake, the increase in
evaporation would average about 9.3 gpm.

On the assumption that the areas of Charles, Round, and Saddleback
Lakes would have been reduced by an average of 50 percent had water not been
pumped into these lakes, the increase in evaporation attributable to the arrest of









1000
J
0
cl-

IW 100-



Percentages represent the
IW r reduction in lake area that
e n l ee.- >
4c 'W 10 would result from continued
<1 _1 decline in lake level S
W

_J

10 100 1000

INCREASE IN LAKE EVAPORATION, GALLONS PER MINUTE

Figure 20 Graph showing relation between lake area and increase in lake evaporation attributable to the arrest of the
decline in lake level.








REPORT OF INVESTIGATIONS NO. 74


the decline in level would average about 25 gpm for the three lakes compared to
the combined pumpage of 765 gpm into the three lakes in 1971.

Whether the entire 25 'gpm (which, of course, is intended only to indicate
the approximate magnitude of the increase in evaporation) can be considered a
reduction of the local water resource is open to question. The decline in level
forestalled in the foregoing hypothetical example was presumed to result from
the effects of: (1) pumping for public water supply; and (2) less-than-normal
rainfall, the first probably being more important than the second for the lakes
involved in this investigation. To the extent that the lake levels were maintained
at a constant stage against the stress imposed by less-than-normal rainfall, the
increase in evaporation represents a reduction of the local water resource. On the
other hand, if lake levels had been maintained at a variable stage compensating
only for well-field pumping but not for below-normal rainfall, the increase in
evaporation that results from maintaining lake levels against the stress imposed
by well-field pumping alone would not be considered a reduction of the local
water resource. In other words, if the lake levels are maintained so as to
compensate only for the effect of well-field pumping, lake evaporation would be
exactly the same as would occur in the absence of well-field pumping, and,
hence, no tendency to reduce the local water resource could result. Thus, in this
view, perhaps less than half of the 25 gpm increase in lake evaporation can be
considered as a reduction of the local water resource.

If the decline in level of a lake is arrested by pumping water into the lake,
the lake has less capacity remaining to store future natural inflows, and the
potential for the lake to overflow is increased. Any water that spills from the
lake in excess of the amount that would have spilled naturally represents a
reduction of the local water resource.

The loss of water due to overflow can be minimized by maintaining the
lake level some distance below the level at which the lake overflows, preferably
the level that would have obtained had there been no pumping for public water
supply; however, little water would spill from direct rainfall on the lake if the
lake were maintained at least 1 foot below the level of the lowest surface outlet.
For a lake that normally declines 0.02 foot per day from evaporation and
leakage, for example, direct rainfall would be stored without loss in amounts as
great as 1.02 feet in 1 day, 1.2 feet in 10 days, and 1.6 feet in 30 days. Rainfall
is seldom in excess of these rates.
Some additional freeboard is required to minimize the overflow of lake
water derived from storm runoff. The quantity of runoff generated by a given
rainfall varies with the size and surficial characteristics of the contributing
drainage area; however, these elements differ so widely for individual lakes that
few generalizations can be made that are useful.







BUREAU OF GEOLOGY


The natural drainage basins of Charles, Round, and Saddleback Lakes are
flat, sandy, and well vegetated. These lakes apparently receive little surface-water
inflow. Charles and Round Lakes did not overflow during 1969, for example,
when the yearly rainfall was greater than average. Saddleback Lake apparently
spilled a small quantity of water during a few days in December 1969. The levels
of Charles, Round, and Saddleback Lakes apparently are maintained low enough
so that the overflow of water attributable to the input of ground water is small
and does not constitute a significant reduction of the water resource.

The addition of Floridan Aquifer water to Charles, Round, and
Saddleback Lakes has caused a distinct change in the chemical quality of water
in the lakes. As Floridan Aquifer water is continually discharged into the lakes,
the chemical quality of the lake water becomes increasingly similar to Floridan
Aquifer water. Some variation in the chemical quality results as the discharge
rate varies from wet to dry periods. Table 2 lists concentrations of selected
chemical constituents of water from natural lakes, from wells tapping the
Floridan Aquifer, and from mixtures of natural lake water and Floridan Aquifer
water in northwest Hillsborough County.

TABLE 2
Selected chemical constituents of water from lakes,
Floridan Aquifer, and mixtures of water from lakes and
Floridan Aquifer, northwest Hilsborough County

(All values, exceph pH, in milligrams per liter)


Source of
water
Crenshaw Lake 1

Round Lake /

Lake Charles 2

Saddleback Lake 2
North end
South end

Wells near Section
21 tapping
Floridan Aquifer 2/

SLake water
Mixture of lake


Dissolved


Alkalinity
as


pH solids Hardness HCO CaCO3 Sodium
5.4 22 12 2.0 2.0 4


Calcium
3


7.9 150 130 168 110 6 50

8.1 105 140 180 148 5 55


7.9 30 96 133 103 5 39
7.9 30 76 102 84 5 30



8.1 220 200 225 180 10 65


water and Floridan Aquifer water


3 Floridan Aquifer water







REPORT OF INVESTIGATIONS NO. 74


Charles, Round, and Saddleback Lakes were not studied for possible
ecological changes that might result from the change in chemical quality of the
lake water by addition of ground, water or from holding the lake level constant.
These subjects are being investigated by scientists of the Southwest Florida
Water Management District. Several residents reported. that the growth of
aquatic plants increased after the input of ground water into the lakes. However,
many lakes in Florida are naturally supplied with ground water that is similar in
chemical quality to the Floridan Aquifer water added to Round, Charles, and
Saddleback Lakes, so whatever ecological changes that might occur over the long
run are not expected to be either drastic or harmful. Further study is underway,
as noted above, to determine whether possible long-term ecological changes are
attributable to the addition of Floridan Aquifer water to the lakes.

Under natural conditions that is, conditions that were free of the effects
of pumping for public water supply and pumping of water to maintain lake
levels water leaked from Round, Charles, and Saddle back Lakes into the
Floridan Aquifer and to some extent contributed to the dissolution of limestone
in the Floridan Aquifer. This may have led to a gradual increase in the
permeability of limestone beneath the lakes generally and in some instances to
the eventual collapse of the materials that overlie the Floridan, as is evident from
the sinkhole depressions in the lake bottoms. Leakage from the lakes continues,
of course, after pumping for public water supply takes place. When the lake
levels are maintained by pumping of ground water into the lakes, however,
leakage from the lakes to the Floridan Aquifer is greater than occurred under
natural conditions because the hydraulic gradient between the lakes and the
Floridan Aquifer is increased. To the extent that leakage from the lakes is so
increased, the rate of dissolution of limestone tends to increase; however, the
increased leakage consists of pumped ground water that has passed through the
limestone of the Floridan Aquifer and, hence, has already used up much of its
capacity for dissolving limestone.

Thus, if the rate of dissolution of limestone is increased, as a result of
maintaining levels of lakes by pumping ground water into the lakes, the increase
probably is small. The effect of such an increase in the rate of dissolution of
limestone on the permeability of materials under the lakes and, hence, on the
pumpage rates required to maintain lake levels at a given stage under the same
climatic and hydrologic conditions probably is not great enough to be
detectable after only a few years of pumping. Over the long term, however, the
cumulative effect of a small increase in the rate of dissolution of limestone may
be appreciable. Resolution of this important matter will require comparison of
present and future pumpage rates for times when climatic and hydrologic
conditions are virtually the same.






BUREAU OF GEOLOGY


SUMMARY

Lake levels in the Section 21 well-field area north of Tampa, Florida, have
been lower than normal in recent years owing to less-than-normal rainfall and a
lowering of ground-water levels by large withdrawals of water from the Floridan
Aquifer in Section 21 well field.

By pumping Floridan Aquifer water into the lakes, local residents have
stabilized the levels of Round Lake since mid-1966 and of Saddleback and
Charles Lakes since mid-1968. Total pumpage into the lakes ranged from 217
million gallons in 1969 to 401 million gallons in 1971; the large increase in
pumpage is attributed to weather conditions which were drier in 1971 than in
1969. In 1971 pumpage averaged 143,000 gpd (100 gpm) at Round Lake;
468,000 gpd (325 gpm) at Saddleback Lake; and 492,000 gpd (340 gpm) at
Lake Charles. These values represent the water required to maintain lake levels at
a nearly constant level which generally was above the level that would have
occurred naturally. Considerably less water would have been required had the
lake levels been permitted to fluctuate through a range about equal to the
natural range in stage of lakes in the same general area.

The distribution of water pumped into the lakes was estimated for May
1971 when the combined pumpage averaged 1,410,000 gpd. Of this amount,
968,000 gpd (69 percent) returned directly to the Floridan Aquifer; 252,000
gpd (18 percent) moved laterally into the water-table aquifer; and 190,000 gpd
(13 percent) was lost by evaporation. Almost all of the water that moved
laterally into the water-table aquifer eventually reached the Floridan Aquifer at
or generally east of Section 21 well field.

The distribution of the yearly pumpage for 1971 probably was about the
same as indicated for May 1971 except that the net loss to the atmosphere
(evaporation minus rainfall) was negligible for the year. Thus, the entire
pumpage for 1971, averaging 765 gpm, was distributed between leakage to the
water-table aquifer (21 percent) and leakage to the Floridan Aquifer (79
percent).

The total leakage from Charles, Round, and Saddleback Lakes returned to
the Floridan Aquifer at or slightly east of Section 21 well field where the water
was locally retrievable for beneficial use. Thus, little reduction of the local water
resource resulted from increased leakage because of the maintaining of lake
levels in 1971.

Mntaintaining levels of lakes at a constant and higher than natural stage
caused some reduction of the local water resource because of an increase in the







REPORT OF INVESTIGATIONS NO. 74


volume of water that evaporated from the lakes. Lake areas were considerably
larger than they would have been had water not been added to the lakes. The
combined increase in water evaporated from Round, Charles, and Saddleback
Lakes was estimated to be 25 gpm on the assumption that lake areas were
maintained twice as large as they would have been had ground water not been
added to the lakes. The rate of evaporation from a water surface is greater than
the rate of evapotranspiration from an exposed lake bottom. The difference
between the rates of evaporation and evapotranspiration was assumed to be 18
inches per year. The estimate of 25 gpm which is intended only to represent
the relative magnitude of the increase in lake evaporation attributable to the
maintaining of lake levels equals about 3 percent of the water pumped into
the lakes in 1971. The lake levels were maintained to compensate for effects of:
(1) pumping for public water supply; and (2) less-than-normal rainfall. If lake
levels had been maintained at a variable stage compensating only for well-field
pumping but not for below-normal rainfall, then only part of the increase in lake
evaporation that is, the part associated with maintaining lake levels against the
stress of less-than-normal rainfall would be considered a reduction of the local
water resource. Thus, with this view in mind, perhaps less than half the 25 gpm
estimated in the hypothetical example above can be considered as a reduction of
the local water resource.

The levels of Round, Charles, and Saddleback Lakes seem to have been
maintained low enough that no appreciable loss of water resulted from overflow
of the lakes during years of normal or below normal rainfall.

The addition of Floridan Aquifer water causes a distinct change in the
chemical quality of water in Round, Charles, and Saddleback Lakes generally
characterized by increases in calcium, bicarbonate, hardness, and alkalinity.
Many lakes in Florida are naturally supplied with ground water that is similar in
chemical quality to the Floridan Aquifer for water being added to these three
lakes.

Maintaining lake levels in Section 21 well-field area causes an increase in
leakage through the lake bottoms because the hydraulic gradient between the
lakes and the Floridan Aquifer is increased. This may increase the dissolution of
limestone which in turn would increase the permeability of materials beneath
the lakes and, hence, increase the pumpage required to maintain levels of lakes at
a given elevation under the same climatic and hydrologic conditions. However,
the increased leakage from the lakes consists of ground water that has passed
through the limestone of the Floridan Aquifer and, consequently, has already
used much of its capacity for dissolving limestone. Thus, any resultant increase
in pumpage required to maintain lake levels under the same given conditions
likely would be small and probably would be undetectable after only a few years






40 BUREAU OF GEOLOGY

of pumping. Resolution of this important matter possible effects of increased
dissolution of limestone will require careful comparison of present and future
pumpage rates for times when climatic and hydrologic conditions are the same.







REPORT OF INVESTIGATIONS NO. 74 41

REFERENCES CITED

Cherry, R. N., Stewart. J. W., and Mann, J. A.
1970 General hydrology of the middle gulf area, Florida: Florida Dept Nat.
Resources, Bur. Geology, Rept. of Inv. 56, 96 p.

Kohler, M. A., Nordenson, T. J., and Baker, D. R.
1959 Evaporation maps for the United States: U. S. Weather Bur. Tech. Paper 37, 13
p.m. 5 pl.

Stewart, J. W.
1968 Hydrologic effects of pumping from the Floridan aquifer in northwest
Hillsborough, northeast Pinellas, and southwest Pasco Counties, Florida: U. S.
Geol. Survey, open-file rept. 241 p., 60 figs.