wri02_4033_lee

7/29/2019 wri02_4033_lee

1/61

Lake

IntermediateConfining Unit

Breaches inIntermediate

Confining Unit

UndifferentiatedSurficial Deposits

LakeSediment

GROUND-WATER

CATCHMENTWATERTABLE

Factors Affecting Ground-Water

Exchange and Catchment Size for FloridaLakes in Mantled Karst Terrain

U.S. Geological Survey

Prepared in cooperation with the

Southwest Florida Water Management District

and the

St. Johns River Water Management District

Water-Resources Investigations

Report 02-4033

7/29/2019 wri02_4033_lee

2/61

Factors Affecting Ground-Water

Exchange and Catchment Size for FloridaLakes in Mantled Karst Terrain

U.S. GEOLOGICAL SURVEYWater-Resources Investigations Report 02-4033

Prepared in cooperation with theSOUTHWEST FLORIDA WATER MANAGEMENT DISTRICT

and theST. JOHNS RIVER WATER MANAGEMENT DISTRICT

Tallahassee, Florida2002

ByT.M. Lee

7/29/2019 wri02_4033_lee

3/61

Copies of this report can bepurchased from:

U.S. Geological SurveyBranch of Information ServicesBox 25286Denver, CO 80225-0286888-ASK-USGS

The use of firm, trade, and brand names in this report is for identification purposes only

and does not constitute endorsement by the U.S. Geological Survey.

For additional informationwrite to:

District ChiefU.S. Geological SurveySuite 3015227 N. Bronough StreetTallahassee, FL 32301

Additional information about water resources in Florida is available on the WorldWide Web at http://fl.water.usgs.gov

U.S. DEPARTMENT OF THE INTERIORGALE A. NORTON, Secretary

U.S. GEOLOGICAL SURVEYCHARLES G. GROAT, Director

7/29/2019 wri02_4033_lee

4/61

Contents III

CONTENTS

Abstract.................................................................................................................................................................................. 1

Introduction ........................................................................................................................................................................... 1

Purpose and Scope....................................................................................................................................................... 2

Acknowledgments ....................................................................................................................................................... 2

Background.................................................................................................................................................................. 3

Physical Characterization of Lake Basins ............................................................................................................................. 5

Methods ....................................................................................................................................................................... 7

Physical Characteristics ............................................................................................................................................... 8

Topography and Ground-Water Flow Patterns.................................................................................................. 20

Hydrogeologic Framework................................................................................................................................ 21

Numerical Modeling of Ground-Water Flow ........................................................................................................................ 28

Methods ....................................................................................................................................................................... 28

Radial Models .................................................................................................................................................... 29

Model Boundaries.............................................................................................................................................. 32Limits to Hypothetical Steady-State Simulations.............................................................................................. 32

Limits to Hypothetical Transient Simulations ................................................................................................... 33

Steady-State Simulation Results.................................................................................................................................. 33

Initial Lake Basin Model ................................................................................................................................... 33

Effect of Recharge Rate..................................................................................................................................... 35

Effect of Basin Size ........................................................................................................................................... 36

Effect of Surficial Aquifer Conductivity ........................................................................................................... 37

Effect of Intermediate Confining Unit............................................................................................................... 38

Effect of Lake Sediment .................................................................................................................................... 41

Effect of Upper Floridan Aquifer Boundary Condition..................................................................................... 41

Effect of Lake Stage .......................................................................................................................................... 42

Effect of Lake Depth.......................................................................................................................................... 43

Transient Simulation Results ....................................................................................................................................... 46Factors Affecting Ground-Water Exchange and Catchment Size ......................................................................................... 48

Catchment Size in Mantled Karst Terrain.................................................................................................................... 49

Hydrogeologic Controls on Ground-Water Exchange................................................................................................. 50

Summary and Conclusions.................................................................................................................................................... 50

References ............................................................................................................................................................................. 52

FIGURES

1. Schematic showing generalized hydrogeologic section through a Florida ridge lake in a

flow-through setting................................................................................................................................................. 3

2-14. Maps showing:

2. Locations of the study lakes in Florida ........................................................................................................... 6

3. Topographic setting and general direction of ground-water flow in the surficial aquifer aroundLake Annie...................................................................................................................................................... 9

4. Topographic setting of Lake Barco, general direction of ground-water flow in the surrounding

surficial aquifer, and the simulated steady-state ground-water catchment ..................................................... 10

5. Topographic setting of Lake Five-O, general direction of ground-water flow in the surrounding

surficial aquifer, and the simulated steady-state ground-water catchment ..................................................... 11

6. Topographic setting and general direction of ground-water flow in the surficial aquifer around

Lake George and Grassy Lake ........................................................................................................................ 12
http://text.pdf/http://text.pdf/http://text.pdf/http://text.pdf/http://text.pdf/http://text.pdf/http://text.pdf/http://text.pdf/http://text.pdf/http://text.pdf/http://text.pdf/http://text.pdf/http://text.pdf/http://text.pdf/http://text.pdf/http://text.pdf/http://text.pdf/http://text.pdf/http://text.pdf/http://text.pdf/http://text.pdf/http://text.pdf/http://text.pdf/http://text.pdf/http://text.pdf/http://text.pdf/http://text.pdf/http://text.pdf/http://text.pdf/http://text.pdf/http://text.pdf/http://text.pdf/http://text.pdf/http://text.pdf/http://text.pdf/http://text.pdf/http://text.pdf/http://text.pdf/http://text.pdf/http://text.pdf/http://text.pdf/http://text.pdf/http://text.pdf/http://text.pdf/http://text.pdf/http://text.pdf/http://text.pdf/http://text.pdf/http://text.pdf/http://text.pdf/http://text.pdf/http://text.pdf/http://text.pdf/http://text.pdf/http://text.pdf/http://text.pdf/http://text.pdf/http://text.pdf/http://text.pdf/http://text.pdf/http://text.pdf/http://text.pdf/http://text.pdf/http://text.pdf/http://text.pdf/http://text.pdf/http://text.pdf/http://text.pdf/http://text.pdf/http://text.pdf/http://text.pdf/http://text.pdf/http://text.pdf/http://text.pdf/http://text.pdf/http://text.pdf/http://text.pdf/http://text.pdf/http://text.pdf/http://text.pdf/

7/29/2019 wri02_4033_lee

5/61

IV Contents


Halfmoon Lake............................................................................................................................................... 13


Lake Hollingsworth ........................................................................................................................................ 14


Lake Isis.......................................................................................................................................................... 15


Lake Lucerne .................................................................................................................................................. 16

11. Topographic setting and general direction of ground-water flow in the surficial aquifer aroundLake Olivia ..................................................................................................................................................... 17


Round Lake and Saddle Blanket Lakes.......................................................................................................... 18

13. Topographic setting of Lake Starr, general direction of ground-water flow in the surrounding

surficial aquifer, and the simulated steady-state ground-water catchment..................................................... 19


Swim Lake...................................................................................................................................................... 20

15-20. Cross sections showing simplified hydrogeologic sections along hillsides in the topographic basins of:

15. Lake Annie and Lake Barco........................................................................................................................... 22

16. Lake Five-O, Lake George, Grassy Lake, and Halfmoon Lake ..................................................................... 23

17. Lake Hollingsworth and Lake Isis.................................................................................................................. 24

18. Lake Lucerne and Lake Olivia ....................................................................................................................... 25

19. Round Lake and Saddle Blanket Lakes.......................................................................................................... 2620. Lake Starr and Swim Lake ............................................................................................................................. 27

21. Graph showing relation of the head difference between the lake and Upper Floridan aquifer to mantle

thickness near lake and to lake elevation ................................................................................................................ 28

22-24. Schematics of:

22. The initial lake basin model .................................................................................................................................... 29

23. The hillsides used for transient simulations ............................................................................................................ 31

24. Simulation results for the initial lake basin model: areal view of basin showing the location of the

ground-water catchment to the lake, and radial view showing the simulated water table and location

of the ground-water catchment................................................................................................................................ 34

25. Graph showing ground-water flow distribution along the lakebed for the initial lake basin model ....................... 35

26-29. Bar charts showing simulated ground-water inflow and leakage rates, and catchment sizes for modeled

lakes with different:

26. Recharge rates................................................................................................................................................. 3527. Size basins ...................................................................................................................................................... 36

28. Horizontal hydraulic conductivity (Kh) values in the surficial aquifer........................................................... 37

29. Vertical hydraulic conductivity (Kv) values in the intermediate confining unit surrounding

the lake............................................................................................................................................................ 38

30. Graph showing ground-water flow distribution along the lakebed for different vertical hydraulic

conductivity (Kv) values in the intermediate confining unit surrounding the lake ................................................. 39

31. Schematic showing different size collapse features beneath lake and various arrangements of lake

sediment used in ground-water flow simulations.................................................................................................... 39

32-36. Bar charts showing:

32. Simulated ground-water inflow and leakage rates for modeled lakes with different size collapse

features beneath the lake................................................................................................................................. 40

33. Simulated ground-water inflow and leakage rates and catchment sizes for modeled lakes with

breaches in the intermediate confining unit surrounding the lake.................................................................. 40

34. Simulated ground-water inflow and leakage rates for modeled lakes with different arrangements

of lake sediment.............................................................................................................................................. 41

35. Simulated ground-water inflow and leakage rates and catchment sizes for modeled lakes with

different head values in the Upper Floridan aquifer ....................................................................................... 42

36. Simulated ground-water inflow and leakage rates and catchment sizes for modeled lakes

maintained at different lake stages ................................................................................................................. 43

37. Schematic showing different lake depths and surficial aquifer thicknesses used in model simulations........ ......... 44
http://text.pdf/http://text.pdf/http://text.pdf/http://text.pdf/http://text.pdf/http://text.pdf/http://text.pdf/http://text.pdf/http://text.pdf/http://text.pdf/http://text.pdf/http://text.pdf/http://text.pdf/http://text.pdf/http://text.pdf/http://text.pdf/http://text.pdf/http://text.pdf/http://text.pdf/http://text.pdf/http://text.pdf/http://text.pdf/http://text.pdf/http://text.pdf/http://text.pdf/http://text.pdf/http://text.pdf/http://text.pdf/http://text.pdf/http://text.pdf/http://text.pdf/http://text.pdf/http://text.pdf/http://text.pdf/http://text.pdf/http://text.pdf/http://text.pdf/http://text.pdf/http://text.pdf/http://text.pdf/http://text.pdf/http://text.pdf/http://text.pdf/http://text.pdf/http://text.pdf/http://text.pdf/http://text.pdf/http://text.pdf/http://text.pdf/http://text.pdf/http://text.pdf/http://text.pdf/http://text.pdf/http://text.pdf/http://text.pdf/http://text.pdf/http://text.pdf/http://text.pdf/http://text.pdf/http://text.pdf/http://text.pdf/http://text.pdf/http://text.pdf/http://text.pdf/http://text.pdf/http://text.pdf/http://text.pdf/http://text.pdf/http://text.pdf/http://text.pdf/http://text.pdf/http://text.pdf/http://text.pdf/http://text.pdf/http://text.pdf/http://text.pdf/http://text.pdf/http://text.pdf/http://text.pdf/http://text.pdf/http://text.pdf/http://text.pdf/

7/29/2019 wri02_4033_lee

6/61

Contents V

38. Bar chart showing simulated ground-water inflow and leakage rates, and catchment sizes for

modeled lakes of different depths with the surficial aquifer thickness near lake equal to 41 feet

and 57.4 feet ................................................................................................................................................... 45

39-40. Graphs showing:

39. Transient ground-water inflow to modeled lakes for basins with different unsaturated zone thicknesses..... 47

40. Relation between lake size, the distance offshore that ground-water inflow occurs, and the percentage

of the lakebed receiving ground-water inflow................................................................................................ 49

TABLES1. Relation of stratigraphic and hydrogeologic units in central Florida .............. ............. ........................................ 42. Physical characteristics of selected lake basins in Florida ..................................................................................... 7

3. Hydrogeologic characteristics of selected lake basins in Florida........................................................................... 8

4. Hydraulic parameters used in the initial lake basin model..................................................................................... 30

5. Simulated water-table slopes and catchment sizes resulting from varying horizontal hydraulic

conductivity (Kh) in the surficial aquifer ................................................................................................................ 37

6. Simulated ground-water inflow to lakes adjacent to level and steep hillsides........................................................ 46

Conversion Factors and Vertical Datum

Sea level: In this report, sea level refers to the National Geodetic Vertical Datum of 1929 (NGVD of 1929) -- a geodetic

datum derived from a general adjustment of the first-order level nets of the United States and Canada, formerly called SeaLevel Datum of 1929.

ACRONYMS AND ABBREVIATIONS

Multiply By To obtain

inch (in.) 2.54 centimeter

inch per year (in/yr) 25.4 millimeter per year

foot (ft) 0.3048 meter (m)

foot per day (ft/d) 0.3048 meter per day

mile (mi) 1.609 kilometer

acre 4,047 square meter

acre 0.4047 hectare

cubic foot per day (ft3/d) 0.02832 cubic meter per day

R Radius

SJRWMD St. Johns River Water Management District

SWFWMD Southwest Florida Water Management District

USGS U.S. Geological Survey
http://text.pdf/http://text.pdf/http://text.pdf/http://text.pdf/http://text.pdf/http://text.pdf/http://text.pdf/http://text.pdf/http://text.pdf/http://text.pdf/http://text.pdf/http://text.pdf/http://text.pdf/http://text.pdf/http://text.pdf/http://text.pdf/http://text.pdf/http://text.pdf/http://text.pdf/http://text.pdf/http://text.pdf/http://text.pdf/http://text.pdf/http://text.pdf/http://text.pdf/http://text.pdf/

7/29/2019 wri02_4033_lee

7/61

VI Contents

7/29/2019 wri02_4033_lee

8/61

Abstract 1

Factors Affecting Ground-Water Exchange and

Catchment Size for Florida Lakes in MantledKarst Terrain

ByT.M. Lee

In the mantled karst terrain of Florida, thesize of the catchment delivering ground-water

inflow to lakes is often considerably smaller thanthe topographically defined drainage basin. Thesize is determined by a balance of factors that actindividually to enhance or diminish the hydraulicconnection between the lake and the adjacentsurficial aquifer, as well as the hydraulic connec-tion between the surficial aquifer and the deeperlimestone aquifer. Factors affecting ground-waterexchange and the size of the ground-watercatchment for lakes in mantled karst terrain wereexamined by: (1) reviewing the physical and

hydrogeological characteristics of 14 Florida lakebasins with available ground-water inflow esti-mates, and (2) simulating ground-water flow inhypothetical lake basins. Variably-saturated flowmodeling was used to simulate a range of physicaland hydrogeologic factors observed at the 14 lakebasins. These factors included: recharge rate to thesurficial aquifer, thickness of the unsaturatedzone, size of the topographically defined basin,depth of the lake, thickness of the surficial aquifer,hydraulic conductivity of the geologic units, thelocation and size of karst subsidence featuresbeneath and onshore of the lake, and the head inthe Upper Floridan aquifer.

Catchment size and the magnitude ofground-water inflow increased with increases inrecharge rate to the surficial aquifer, the size ofthe topographically defined basin, hydraulic

conductivity in the surficial aquifer, the degree ofconfinement of the deeper Upper Floridan aquifer,and the head in the Upper Floridan aquifer. Thecatchment size and magnitude of ground-waterinflow increased with decreases in the number and

size of karst subsidence features in the basin, andthe thickness of the unsaturated zone near the lake.Model results, although qualitative, providedinsights into: (1) the types of lake basins in man-tled karst terrain that have the potential to generatesmall and large amounts of ground-water inflow,and (2) the location of ground-water catchmentsthat could be managed to safeguard lake waterquality. Knowledge of how ground-water catch-ments are related to lakes could be used by water-resource managers to recommend setback dis-

tances for septic tank drain fields, agricultural landuses, and other land-use practices that contributenutrients and major ions to lakes.

INTRODUCTION

Ground-water interactions with lakes in man-tled karst terrain are a fundamental concern to watermanagers interested in protecting and developingwater resources. In Florida, more than 7,000 lakes aresituated in a layer or mantle of sand and clay that blan-

kets an extensive and highly productive limestoneaquifer, the Upper Floridan aquifer. Many are seepagelakes with basins that lack natural streams, and all relyto varying degrees on ground water to convey inflowand outflow between the lake and surrounding aqui-fers. Understanding the location and size of the catch-ment contributing ground-water inflow to lakes isparticularly important because all of the ground water

ABSTRACT

7/29/2019 wri02_4033_lee

9/61

2 Factors Affecting Ground-Water Exchange and Catchment Size for Florida Lakes in Mantled Karst Terrain

within the catchment eventually flows into the lake.Ground water from the surficial aquifer system flowsinto the lake and helps to sustain lake stage by replac-ing the lake water that leaks to the deeper limestoneaquifer. Ground-water inflow also helps offset lakeevaporation losses, which can exceed rainfall in west-central Florida lakes (Swancar and others, 2000). Fer-tilizers and other solutes applied within catchmentareas can enter the lakes through ground-water inflowand have a measurable effect on the water quality oflakes in Florida (Eilers and others, 1988; Lee andSacks, 1991; Pollman and others, 1991; Stauffer,1991; Sacks and others, 1998).

Results of water, solute, and isotope mass bal-ance studies of 14 lakes in Florida suggest that theamount of annual ground-water inflow to lakes can varywidely, even for lakes of similar size and apparentphysical setting (Pollman and others, 1991; Sacks andothers, 1998). In addition, the size of ground-water

catchments can vary substantially from lake to lake(Grubbs, 1995; Lee, 1996; Amy Swancar and T.M. Lee,USGS, written commun., 2002). The effects of physicalsetting on ground-water exchange have been describedin detail in a few basin-scale lake studies that combinehydrogeologic descriptions with water budgets andground-water flow modeling (Lee and others, 1991;Grubbs, 1995; Lee and Swancar, 1997, Swancar andothers, 2000). However, a more general quantitativeframework is needed to anticipate the ground-waterinteractions in the larger population of lakes.

Modeling the distinctive settings of Floridalakes can aid in characterizing the principle types offlow regimes within the overall lake population.Numerical modeling studies of hypothetical lakebasins have been used to characterize the ground-water flow regimes of lakes in the glacial terrain ofNorth America (Winter, 1976, 1978; Anderson andMunter, 1981; Winter and Pfannkuch, 1984), and thecoastal plain near Perth, Western Australia (Nield andothers, 1994; Townley and Trefry, 2000). Because ofessential differences in geology and climate, however,it can be difficult or impossible to extrapolate these

results to Florida lakes in mantled karst terrain. TheU.S. Geological Survey began a study in 1998 in coop-eration with the Southwest Florida Water ManagementDistrict and the St. Johns River Water ManagementDistrict to examine the effect of recharge and basinhydrogeologic characteristics on the magnitude ofground-water exchange and catchment sizes for lakesin the mantled karst terrain of Florida.

Purpose and Scope

This report presents the results of numericalground-water flow modeling of hypothetical lakebasins with a range of basin characteristics typical ofmantled karst terrain. In this report, steady-state andtransient ground-water flow modeling is used to simu-

late how recharge, hydrogeologic setting, and basingeometry affect the size of the ground-water catch-ment and the magnitude of ground-water inflow. Mod-eling results are summarized and used to makegeneralizations about the potential for different arche-typal lake basins to generate ground-water inflow toFlorida lakes.

Ground-water flow is simulated using a variablysaturated flow model. Most of the model simulationsassume steady-state conditions, but transient simula-tions are used to explore how the timing of recharge

fluxes may affect ground-water interactions with thelake. The idealized lake basin geometries and hydro-geologic data used in the modeling relied on physicalcharacteristics summarized for 14 lake basins in ridgeareas of Florida. Estimates of the ground-water inflowto these lakes and data on their hydrogeologic settingsare available in published reports. Model simulationsof hypothetical lakes examine the effect of the follow-ing characteristics on the ground-water exchange andcatchment size: recharge rate, topographically definedbasin size, surficial aquifer conductivity, intermediate

confining unit integrity, lake sediment, Upper Floridanaquifer boundary condition, lake stage, and lake depth.

Acknowledgments

This study was conducted by the U.S. Geologi-cal Survey (USGS) in cooperation with the SouthwestFlorida Water Management District (SWFWMD) andthe St. Johns River Water Management District(SJRWMD), two agencies committed to providing abetter understanding of lake hydrology in Florida. Theauthor is grateful to Kathleen Hammett (USGS) forher steadfast support during the study; Angel Martin,Laura Sacks, and Amy Swancar (USGS) for helpfuldiscussions on Florida lakes and preliminary reviewsof the manuscript; and Richard Schultz (SWFWMD)for perspectives on water management in west-centralFlorida.

7/29/2019 wri02_4033_lee

10/61

Introduction 3

Background

Previous modeling investigations have exploredthe effect of physical and hydrological settings onground-water interactions in hypothetical lakes (Nieldand others, 1994; Townley and Davidson, 1988; Town-ley and Trefry, 2000; Winter 1976, 1978, 1983; Winterand Pfannkuch, 1984). In these modeling studies, the

lake and underlying aquifers were underlain by animpervious no-flow boundary. For this reason, all ofthe recharge to the model (or influx from a lateralmodel boundary) eventually discharged to one or morelakes, or exited the model as lateral flow. The consid-erable vertical ground-water flow occurring in manyFlorida lake basins was not represented. Winter proba-bly made the earliest applicable simulations of hypo-thetical Florida lake basins to aid discussions with aFlorida colleague (T.C. Winter, USGS, Denver, Colo.,written commun. to G. H. Hughes, USGS, Tallahassee,FL, 1977). Although not published, the conceptual

framework contained in these simulations provided theUSGS a departure point for modeling ground-water andlake interactions in Florida.

In the mantled karst terrain of central Florida, flowin the surficial aquifer is predominantly downward andmassive amounts of recharge flow vertically to the deeperUpper Floridan aquifer across an intermediate confiningunit. The comparatively smaller amount of lateralground-water flow intercepted by lakes depends upon

boundary fluxes and head conditions and the hydrogeo-logic framework. Small-scale features of the geologicframework within lake basins are important. The interme-diate confining unit below the lake typically differs fromthat in the surrounding basin due to the sinkhole pro-cesses that formed the lake. Further, the head differencebetween the two aquifers that causes the downward flowcan be highly dynamic, subject to changes due to the sea-sonal climate and withdrawals from the Upper Floridanaquifer. Thus, modeling assumptions about boundaryconditions and the hydrogeologic framework should bespecific to each lake basin simulated.

Many lakes in Florida are situated within sandhills and ridges along the central peninsula, referred toas the Central Lake District (Brooks, 1981). Otherlakes are concentrated in smaller ridge areas in thepanhandle and west-central part of the peninsula(Griffith and others, 1997). The general hydrogeologicsetting of many of the ridge lake basins is similar

(fig. 1), although the exact geometry and conductivityof the hydrogeologic units can vary widely (Geraghtyand Miller, Inc., 1980; Tihansky and others, 1996;Schiffer, 1998).

Most ridge lakes occupy topographic depres-sions resulting from the piping and subsidence of surf-icial sand and clay deposits into solution cavities in theunderlying limestone (Tihansky, 1999). In peninsularFlorida, clay-rich beds of the Hawthorn Group make

CITRUS

GROVES

IRRIGATIONWELL

RECHARGE

RAINFALL

EVAPORATIONSEPTICTANK

WATERTABLE

UPPER FLORIDANAQUIFER

INTERMEDIATE CONFININGUNIT

SURFICIAL AQUIFERSYSTEM

NOT TO SCALE

Figure 1. Generalized hydrogeologic section through a Florida ridge lake in a flow-throughsetting (modified from Tihansky and Sacks, 1997).

7/29/2019 wri02_4033_lee

11/61


up the intermediate confining unit that separates thelimestone Upper Floridan aquifer system from theoverlying surficial aquifer system (fig. 1 and table 1).In the basin surrounding a lake, the intermediate con-fining unit may be relatively intact. Beneath the lake,however, the confining unit has been disrupted to vary-ing degrees by sinkhole formation. Because the inter-mediate confining unit only slows down vertical flow,

ground water in the surficial aquifer flows laterally anddownward. Near the lake, ground water flowing in a pre-dominantly lateral direction enters the shallow lake bot-tom. Lake water can leak out laterally near the shorelinewhen the water table in the surficial aquifer slopes awayfrom the lake. Along the deeper lake bottom, lake waterleaks vertically downward. Ultimately, all of the lakeleakage and ground water in the deeper surficial aquiferflow downward across the intermediate confining unit torecharge the Upper Florida aquifer (fig. 1).

The subsidence structure beneath sinkhole lakes

substantially affects vertical leakage losses. High-resolution seismic reflection surveys have providedinsights into the shape and geologic structure of col-lapse features beneath lakes in north-central Florida(Subsurface Detection Investigations, Inc., 1992;

Kindinger and others, 1994) and west-central Florida(Tihansky and others, 1996). The hydraulic conductiv-ity distribution beneath these lakes, however, cannotbe similarly inferred.

The hydraulic conductivity below lakes is notknown but has been estimated from water budget stud-ies. For example, by assuming that all lake leakagederived from lake water budgets was vertical, averageleakance values (Kv/b) were derived for the column ofmaterial between the lake bottom and the Upper Flori-dan aquifer (Motz, 1998). In modeling studies, indi-vidual hydraulic conductivity values were assigned toorganic lake sediment, surficial sediments, and remnantsof the intermediate confining unit filling the collapse fea-tures below lakes (Lee and Swancar, 1997; AmySwancar and T.M. Lee, USGS, written commun.,2002). For example, in simulations of Lake Starr inPolk County, sediments collapsed into the sinkholebeneath the lake were 12.5 times more conductive than

the intermediate confining unit they replaced (AmySwancar and T.M. Lee, USGS, written commun., 2002).Lake leakage also depends on the potentiometric

surface of the Upper Floridan aquifer, which isaffected by pumping and recharge rates (Yobbi, 1996).

LITHOSTRATIGRAPHICUNIT

HYDROSTRATIGRAPHICUNIT

Hawthorn

Group

Arcadia

Formation

Peace River

Formation

SuwanneeLimestone

Ocala Limestone

Surficial

Aquifer System

(SAS)

(IAS - ICU)

Intermediate

Aquifer

System

and/or

Intermediate

Confining Unit

Upper

Floridan

Aquifer

(UFA)

Highly variable lithology ranging fromunconsolidated sands to clay beds with variableamounts of shell fragments, gravel-sized quartz

grains and reworked phosphate

Interbedded sands, clays and carbonates withsiliciclastic component being dominant andvariably mixed; moderate to high phosphatesand/gravel content

Arcadia Formation is a fine-grained carbonatewith low to moderate phosphate andquartz sand, variably dolomitic

Suwannee Limestone is a fine- to medium-grainedpackstone to grainstone with trace organics and

variable dolomite and clay content

Ocala Limestone is a chalky, very fine- to fine-grained wackestone/packstone varying with depthto a biogenic medium- to coarse-grainedpackstone grainstone; trace amounts of organicmaterial, clay, and variable amounts of dolomite

GENERALIZED LITHOLOGY

Undifferentiated

sand, shell,

and clay

(UDSC)

FloridanAquiferS

ystem

Table 1. Relation of stratigraphic and hydrogeologic units in central Florida

7/29/2019 wri02_4033_lee

12/61

Physical Characterization of Lake Basins 5

As the head in the underlying aquifer drops, the down-ward head gradient controlling lake leakage increases.For example, the estimated monthly leakage fromLake Lucerne was 1.8 inches (in.) in May 1986, whenpumping increased the (daily average) downward headdifference between the lake and the Upper Floridanaquifer head to about 13 feet (ft). During September1986, when this head difference was only 5 ft, lakeleakage was 0.7 in. (Lee and Swancar, 1997). TheUpper Floridan aquifer head also has a large effect onthe size of the ground-water catchment and the magni-tude of ground-water inflow; however, few previousstudies have examined this relation.

Annual recharge to the surficial aquifer affectsthe annual ground-water inflow to lakes. For example,when rainfall was about 25 percent above average in1989 at Lake Five-O near Panama City, Fla., ground-water inflow appreciably exceeded lake leakage. In1990, however, rainfall was about 25 percent below

average, and lake leakage greatly exceeded inflow(Grubbs, 1995).The timing of rainfall (winter or summer)

affects how much of the rainfall becomes recharge.Annual rainfall averages about 50 inches/year in cen-tral Florida (National Oceanic and AtmosphericAdministration, 1996); however, it varies widelybetween years and can differ substantially at differentlake basins within a geographic area for the same year(Sacks and others, 1998). The annual rainfall was sim-ilar in two consecutive years of study at Lake Starrnear Lake Wales, Fla. Yet recharge to the surficialaquifer and ground-water inflow to the lake were sub-stantially greater in the second year (August 1997 -July 1998), because most of the annual rainfall fell inthe winter and spring when evapotranspiration waslow. Annual recharge to the surficial aquifer in ridgeareas of central Florida has been estimated to rangefrom 30 to 53 percent of average annual rainfall(Knowles, 1996; Sumner, 1996).

Topographic relief in lake basins also affectsground-water inflow to lakes. Near the lake shoreline,where the unsaturated zone is typically thinnest,

recharge to the water table can be rapid and cause theformation of transient water-table mounds that canincrease the ground-water inflow (Winter, 1983; Lee,2000). Higher land-surface elevation increases thethickness of the unsaturated zone above the water tableand increases the time between rainfall and aquiferrecharge. As a result of the slower, more prolongedrecharge process, elevation gradients in the water table

are lower, resulting in less lateral flow. Variably satu-rated flow modeling was first used by Lee (2000) tosimulate the effects of topography and transientrecharge on the magnitude of ground-water inflow to alake in mantled karst terrain. Saturated ground-waterflow models typically overlook these processes.Recently, the results of one-dimensional unsaturatedground-water flow modeling were combined with athree-dimensional saturated flow model to improve theestimates of ground-water flow to Lake Starr, incentral Florida (Amy Swancar and T.M. Lee, USGS,written commun., 2002).

PHYSICAL CHARACTERIZATION OFLAKE BASINS

The physical settings of 14 lakes helped todefine a range of basin characteristics that potentially

affect ground-water interactions with lakes in Florida(fig. 2). These physical settings provided concepts forthe numerical modeling of hypothetical lakes. Elevenof the lakes are located in Polk and Highlands Coun-ties (Sacks and others, 1998; Lee and others, 1991).Lake Five-O, in Bay County, is in the panhandle ofFlorida and its hydrogeologic setting is described inAndrews and others (1990). Lake Barco is in PutnamCounty in north-central Florida, and is described bySacks and others (1992). Halfmoon Lake, in Hillsbor-ough County, is in a physiographic region called theNorthern Gulf Coast Lowlands (White, 1970). The

hydrogeologic setting and water budget of HalfmoonLake are described by Metz and Sacks (2002).

The annual ground-water inflows to all 14 lakeswere estimated in previous studies. Sacks and others(1998) estimated the steady-state ground-water inflowrates to 10 of the lakes using a combination of water-balance and chemical mass-balance approaches. Five ofthe 14 lakes (including one of the 10 lakes describedabove) were the subjects of detailed basin-scale model-ing studies. Ground-water inflow and lake-water leakageat Lakes Lucerne, Barco, Five-O, Starr, and Halfmoon

were estimated using a combination of water-balanceand numerical modeling approaches (Grubbs, 1995;Lee, 1996; Lee and Swancar, 1997; Swancar and others,2000; Metz and Sacks, 2002; Amy Swancar andT.M. Lee, USGS, written commun., 2002; and R.Yager,USGS, Ithaca N.Y., personal commun., July 2001). Theinformation in the following sections is taken fromthese sources, unless another reference is cited.

7/29/2019 wri02_4033_lee

13/61


Saint Cloud

LakeHollingsworth

GrassyLake

RoundLake Lake

StarrSwim

Lake

LakeGeorge

LakeLucerne

Saddle BlanketLakes

LakeOlivia

LakeIsis

LakeAnnie

Peace

Riv

er

Horse

Cre

ek

Ridge Areas (after White, 1970)

EXPLANATION

2815

2800

45

30

2715

8200 45 30 8115

HIGHLANDS COUNTY

OSCEOLA COUNTY

OSCEOLA COUNTY

HARDEE COUNTY

DE SOTO COUNTY

POLK COUNTYHILLSBOROUGH

COUNTY

MANATEECO

UNTY

SARASOTACOUNTY

Base from Southwest Florida Water Management District digital data, 1:250,000, 1992Universal Transverse Mercator projection,Zone 17

Kissim

meeRiver

Winter Haven

Ridge

LakelandRidge

Lake Henry

Ridge

Lake Wales

Ridge

Bombing Range

Ridge

0 5 10 MILES

0 5 10 KILOMETERS

LOCATION

OF DETAILED

AREA

Lake Five-O

Lake Barco

Halfmoon Lake

Figure 2. Locations of the study lakes in Florida (modified from Sacks and others, 1998).

7/29/2019 wri02_4033_lee

14/61


Methods

The physical basin characteristics compiled dur-ing the previous studies included: the size and depthof each lake, the thickness of the surficial aquiferencompassing the lake, the slope of the water table,and the thickness of the intermediate confining unitseparating the surficial aquifer from the underlying

Upper Floridan aquifer (table 2). In addition, the thick-ness of the unsaturated zone and the size of the topo-graphically defined basin were described for each lake.

Detailed hydrogeologic information is availablefor the five lakes that were the subject of basin-scalemodeling studies (Lucerne, Barco, Five-O, Starr, andHalfmoon Lakes). Generalized hydrogeologic infor-mation is available for the remaining nine lake basinsand was taken from regionalized lithologic maps(Buono and Rutledge, 1978; Buono and others, 1979;Tihansky and others, 1996). Information on lake

bathymetry, basin topography, and geology was avail-able on all 14 of the lakes. With the exception of SwimLake and Lake Five-O, aerial photos (scale 1 in. =200 ft) with topographic contours at 1-ft intervals wereused during this study to describe elevations alonghillsides in the lake basins. Topographic maps (scale

1 in. = 24,000 in.) were used to describe the landelevation around Swim Lake and Lake Five-O.

Ground-water elevations were available in thebasins of all 14 lakes. Monthly water levels measuredduring the previous studies of each lake were used tocharacterize the direction of ground-water flow in thebasin, and high and low water-table conditions. Dailylake-stage data and monthly or biweekly water-table

measurements typically were available. The slope ofthe water table was computed between adjacent moni-tor wells located along a head gradient into or out ofthe lake.

The potentiometric level of the Upper Flori-dan aquifer was measured in the nearest availableobservation well or from wells drilled onsite for thestudy. The head difference between the lake and theUpper Floridan aquifer was the mean of wet- anddry-season observations (typically made duringMay and September, respectively) (table 3).

Because ground-water pumping effects are minimalin the Lake Five-O and Lake Barco basins, periodicmeasurements of head in the Upper Floridan aquifergenerally were representative of monthly condi-tions. In contrast, periodic measurements typicallymade a poor surrogate for the average monthly head

Table 2. Physical characteristics of selected lake basins in Florida

[R, principal radius; r, secondary radius; ft, msl, feet above mean sea level; max, maximum; min, minimum; GW Inflow (in/yr), estimated annualground-water inflow to lake expressed as inches per year above the lake surface area]

Lakename

Surface

area(acres)

Lake stage

(ft msl)

Estimated

lake radius

(feet)

Basin

dimension

Distance

to closest

lake

Lake bed

slope

(percent)

Water table1

wet season

slope(percent)

1Negative values indicate water table sloping away from lake.

Water table

dry season

slope(percent)

GW

inflow(in/yr)

Wet

season

Dry

seasonr R Min Max R Mean Max Min Max Min Max

Annie 92 110.5 109.8 500 1,000 0.2R 6.0R 4.5R 6 11 -0.1 0.5 -0.1 0.3 240

Barco 29 88.0 83.6 640 0.7R 4.0R 1.2R 3 10 -1.1 2.3 -0.7 2.2 8

Five-O 27 50.2 45.1 650 1.0R 7.0R 1.2R 7 14 0.3 1.5 0.3 1.4 237

George 59 130.5 130.0 800 0.1R 2.0R 2.0R 2 5 -0.5 1.1 -0.6 0.5 54

Grassy 76 130.9 129.7 1,000 0.8R 2.2R 1.5R 2 6 0.1 0.3 0.3 0.2 30

Halfmoon 33 42.0 40.7 300 500 0.5R 2.0R 2.0R 9 11 0.0 2.6 0.0 0.7 232

2Average of 3 years, including year with rainfall 34 inches above normal.

Hollingsworth 356 131.3 130.5 2,250 0.3R 1.3R 0.5R 0.2 0.8 0.5 1.8 0.5 1.8 110

Isis 50 108.7 109.5 600 650 0.5R 7.0R 2.5R 9 14 -0.6 1.8 -1.2 1.4 100

Lucerne 38 126.0 124.8 860 1.0R 2.0R 0.8R 2 5 0.5 0.6 0.1 0.3 24

Olivia 86 114.5 114.7 800 0.3R 3.5R 1.0R 5 13 -0.4 0.7 -0.4 0.6 33

Round 31 131.4 130.8 650 0.7R 1.7R 1.2R 4 11 0.4 1.0 0.1 0.3 6

Saddle Blanket 6 116.9 117.7 330 1.0R 7.0R 1.0R 3 7 -0.2 0.7 -0.6 0.4 30

Starr 134 103.9 104.2 650 1,000 1.0R 2.0R 2.2R 3 9 -0.2 0.2 -0.2 0.2 23

Swim 5 96.8 96.2 240 1.0R 3.0R 2.0R 12 20 -0.1 0.8 -0.1 0.9 180

mean 1.1 0.8

median 0.9 0.6

7/29/2019 wri02_4033_lee

15/61


conditions for basins in Polk and Highlands Counties,where ground-water pumping could cause consider-able hour-to-hour variation in head. For this reason,heads presented in table 3 provide only ageneral indi-cation of the aquifer conditions near these lakes.

Physical Characteristics

Although the 14 lakes surveyed represent a tinyproportion of the thousands of lakes located in theCentral Lake District, they exhibit a wide range ofphysical characteristics (tables 2 and 3). The lakesrange in size from 5 (Lake Swim) to 356 acres (LakeHollingsworth). Most of the lakes are less than100 acres in size (table 2), and are roughly circular in

shape (figs. 3-14), except Halfmoon Lake in Hillsbor-ough County which is least circular and more seg-mented than the other 13 lakes (fig. 7). Thetopographically defined basins surrounding the lakescan be roughly circular (for example, George (fig. 6),Grassy (fig. 6), Hollingsworth (fig. 8), and RoundLakes (fig. 12)). More commonly, the drainage divideextends a considerable distance from the lake alongone axis, where as in other areas, the divide is rela-tively close to the lake margin (Barco (fig. 4), Five-O(fig. 5), Isis (fig. 9), Olivia (fig. 11), Saddle Blanket(fig. 12), and Swim Lakes (fig. 14)). Following theconvention of Sacks and others (1998), the SaddleBlanket Lake referred to in this report is the larger ofthe two lakes named together as Saddle Blanket Lakes(fig. 12).

Table 3. Hydrogeologic characteristics of selected lake basins in Florida

[USD, undifferentiated surficial deposits; UFA, Upper Floridan aquifer; Mean vertical head difference equal to lake stage minus the mean head in the UFA;ft, msl, feet above mean sea level]

Lake name County

Refer-

ence

stage1

(ft msl)

1Reference stage used to calculate the vertical head difference with the Upper Floridan aquifer.

Lake

depth

maximum

(feet)

Thick-

ness

USD

(feet)

Thickness

Hawthorn2

formation

(feet)

2Thickness of the Hawthorn formation is assumed equivalent to the thickness of the intermediate confining unit.

Thick-

ness

mantle

total

(feet)

Thick-

ness

mantle

sublake

(feet)

Hydraulic head

Upper Floridan

aquifer

(ft msl)

Mean vertical

head difference

lake to UFA

(feet)Wet season Dry season

Annie Highlands 110.0 65 300 250 550 485 52.6 49.8 59

Barco Putnam 88.0 22 40 60 100 78 83.2 78.4 7

Five-O Bay 47.0 48 23 37 60 12 44.7 39.8 5

George Polk 130.0 15 55 75 130 115 122.9 119.1 9

Grassy Polk 130.0 23 80 100 180 157 107.5 101.0 26

Halfmoon Hillsborough 41.4 22 35 7 42 20 30.2 27.3 13

Hollingsworth Polk 131.0 6 10 210 220 214 81.2 73.3 54Isis Highlands 109.0 64 230 200 430 366 88.3 83.1 23

Lucerne Polk 125.0 22 50 90 140 118 120.2 112.9 9

Olivia Highlands 114.5 47 230 200 430 383 75.3 74.8 39

Round Polk 131.0 28 55 100 155 127 112.3 105.1 21

Saddle Blanket Polk 117.0 11 150 200 350 339 79.6 72.2 41

Starr Polk 104.0 33 50 40 90 57 103.0 100.5 2

Swim Polk 96.5 30 100 100 200 170 88.6 84.8 10

Crooked Lake3

3Data for the nine additional lakes at the bottom of table 3 are from Geraghty and Miller (1980) and Barcelo and others (1990), and are used infigure 21a.

Polk 108.5 13 100 150 250 237 49

Clinch Polk 104.0 29 140 160 300 271 33

Lotela Highlands 101.8 25 140 290 430 405 43

Jackson Highlands 101.4 26 120 380 500 474 49

Letta Highlands 91.0 9 140 360 500 491 37

Placid Highlands 90.8 57 200 400 600 543 44

Reedy Polk 78.0 no data 180 160 340 no data 11

June-in-Winter Highlands 73.7 35 130 420 550 515 24

Hamilton Polk 120.0 12.5 90 100 190 177.5 11

7/29/2019 wri02_4033_lee

16/61


RADIUSOFCIRCULAR

LAKEAREA

LO

CATIONOF

H

ILLSIDE

SECTION

S

HOWNIN

FIGURE15

TOP

OGRAPHICDRAINAGE

DIVIDE

LAND

-SURFACEELEVATION--

Infe

etabovesealevel.

Con

tourintervalis5feet

GR

OUND-WATERFLOW

D

IRECTION

MOST

F

REQUENTLY

O

BSERVEDIN

S

URFICIALAQUIFER

R

EXPLANATION

135

BasefromU.S.GeologicalSu

rveydigitaldata,1:24,000,1972

UniversalTransverseMercato

rprojection,

Zone17

Childsquadrangle1953(1972

)

0 0

300

1,000

2,000FEET

600METERS

R

r

N

Figure

3.

Topographic

settingandgeneraldirectionofground

-waterflow

inthesurficialaquiferarou

ndLakeAnnie.

7/29/2019 wri02_4033_lee

17/61


0

0 300

1,000 2,000 FEET

600 METERS

Melrose quadrangle 1966 (1981)

SHORELINE WHERE GROUND-WATER FLOWDIRECTION PERIODICALLY REVERSES

RADIUS OF CIRCULAR LAKE AREA

LOCATION OF HILLSIDE SECTION SHOWNIN FIGURE 15

TOPOGRAPHIC DRAINAGE DIVIDE

LAND-SURFACE CONTOUR- Elevation in feetabove sea level. Contour interval is 5 feet

SIMULATED GROUND-WATER CATCHMENT

GROUND-WATER FLOW DIRECTION MOSTFREQUENTLY OBSERVED IN SURFICIALAQUIFER

R

N

EXPLANATION

R

Base from U.S. Geological Survey digital data, 1:24,000,1981Universal Transverse Mercator projection,Zone 17

Lake Barco

DIRECTION OF

GROUND-WATER FLOW

100

Figure 4. Topographic setting of Lake Barco, generaldirection of ground-water flow in the surroundingsurficial aquifer, and the simulated steady-state ground-water catchment (modified from Lee, 1996).

7/29/2019 wri02_4033_lee

18/61


RADIUSOFCIRCULARLAKEAREA

LOCATIONOFHILLSIDESECTIONSHOWNINFIGURE16

TOPOGRAPHICDRAINAGEDIVIDE

LAN

D-SURFACEELEVATION--Inmetersabovesealevel.

Co

ntourintervalistwometers

SIM

ULATEDGROUND-WATERCATCHMENT

GROUND-WATERFLOWD

IRECTIONMOSTFREQUENTLY

OBSERVEDINSURFICIALAQUIFER

R

E

XPLANATION

R

N

Lake

Five-O

Lake

Five-O

0 0

300

1,000

2,000FEET

600METERS

26

CrystalLakequadrangle1982

Basefrom

U.S.GeologicalSurveydigitaldata,1:24,000,1987

UniversalT

ransverseMercatorprojection,

Zone17

DIRECTIONOF

GROUND-WATERFLOW

Figure5

.

TopographicsettingofLakeFive-O,

generaldirectionofground-waterflow

inthesurroundingsurficialaquifer,an

dthe

simulate

dsteady-stateground-watercatchmen

t(modifiedfromG

rubbs,

1995).Note

thesimulatedground-watercatchment

extends

beyondthetopographicdrainagedivideeastofthelake.

7/29/2019 wri02_4033_lee

19/61


R

134

133

R



LOCATION OF HILLSIDE SECTIONSHOWN IN FIGURE 16


LAND-SURFACE ELEVATION--In feet above sealevel. Contour interval is 5 feet at Lake Georgeand 10 feet at Grassy Lake

R

EXPLANATION

N

N


Base from U.S. Geological Survey digital data, 1:24,000, 1980, 1988TUniversal ransverse Mercator projection,

Zone 17

130

.Base from U.S Geological Survey digital data, 1:24,000, 1987Universal Transverse Mercator projection,Zone 17

Winter Haven quadrangle 1959 (1980)

Bartow quadrangle 1949 (1987)

0

0 300

1,000 2,000 FEET

600 METERS

Figure 6. Topographic setting and general direction of ground-water flow in the surficial aquifer around Lake Georgeand Grassy Lake.

7/29/2019 wri02_4033_lee

20/61






LAND-SURFACE ELEVATION-- In feet above sealevel. Contour interval is 5 feet


R

EXPLANATION

R

r

N

Base from U.S. Geological Survey digital data,1:24,000, 1987Universl Transverse Mercator projection,Zone 17

50

Citrus Park quadrangle 1956 (1987)

0 1,000 2,000 FEET

0 600 METERS300

Figure 7. Topographic setting and general direction ofground-water flow in the surficial aquifer aroundHalfmoon Lake.

7/29/2019 wri02_4033_lee

21/61


Lakeland quadrangle 1975 (1987)

R

N


LOCATION OF HILLSIDE SECTION SHOWN IN FIGURE 17


LAND-SURFACE ELEVATION-- In feet above sea level.Contour interval is 5 feet

GROUND-WATER FLOW DIRECTION MOST FREQUENTLY OBSERVEDIN SURFICIAL AQUIFER

R

EXPLANATION

Base from U.S. Geological Survey digital data, 1:24,000, 1987

Universal Transverse Mercator projection,Zone 17

200

0

0 300

1,000 2,000 FEET

600 METERS

Figure 8. Topographic setting and general direction of ground-water flow in the surficial aquiferaround Lake Hollingsworth.

7/29/2019 wri02_4033_lee

22/61


R

r

Lake Byrd108

Lake Brentwood104

N




GROUND-WATER FLOW DIRECTION MOST FREQUENTLY

OBSERVED IN SURFICIAL AQUIFER

R

EXPLANATION

Avon Park quadrangle 1953 (1987)

Base from U.S. Geological Survey digital data, 1:24,000, 1987Universal Transverse Mercator projection,Zone 17

0

0 300

1,000 2,000 FEET

600 METERS

150


Figure 9. Topographic setting and general direction of ground-waterflow in the surficial aquifer around Lake Isis.

7/29/2019 wri02_4033_lee

23/61


R

Winter Haven quadrangle 1959 (1980)

N





GROUND-WATER FLOW DIRECTION MOST FREQUENTLY OBSERVEDIN SURFICIAL AQUIFER

R

EXPLANATION


150

0

0 300

1,000 2,000 FEET

600 METERS

Figure 10. Topographic setting and general direction of ground-water flow in the

surficial aquifer around Lake Lucerne.

7/29/2019 wri02_4033_lee

24/61


Avon P ark quadrangle 1953 (1987)Frostproof quadrangle 1953 (1972)

R

N






GROUND-WATER FLOW DIRECTION MOST FREQUENTLYOBSERVED IN SURFICIAL AQUIFER

R

EXPLANATION

Base from U.S. Geological Survey digital data, 1:24,000, 1972, 1987Universal Transverse Mercator projection,Zone 17

100

0

0 300

1,000 2,000 FEET

600 METERS

Figure 11. Topographic setting and general direction of ground-water flow in the surficial aquiferaround Lake Olivia.

7/29/2019 wri02_4033_lee

25/61


R

105Lake Streety

Frostproof quadrangle 1953 (1972)

LakeWinterset

131

125

Eloise quadrangle 1955 (1987)

N

N



LOCATION OF HILLSIDE SECTIONSHOWN IN FIGURE 19


LAND-SURFACE ELEVATION-- In feet above sea level.Contour interval is 10 feet at Round Lake and 5 feetat Saddle Blanket Lake

GROUND-WATER FLOW DIRECTION MOSTFREQUENTLY OBSERVED IN SURFICIAL AQUIFER

R

EXPLANATION

R

125



0

0 300

1,000 2,000 FEET

600 METERS

Figure 12. Topographic setting and general direction of ground-water flow in the surficial aquifer around Round Lake andSaddle Blanket Lakes.

7/29/2019 wri02_4033_lee

26/61


LakeWalesquadrangle

1987

N

RADIUSOFCIRCULARLAKEA

REA

LOCATIONOFHILLSIDESECTIONSHOWNIN

FIGURE20

TOPOGRAPHICDRAINAGEDIV

IDE

LAND-SURFACEELEVATION--Infeetabovesealevel.

Contourintervalis5feet

SIMULATEDGROUND-WATERCATCHMENT

GROUND-WATERFLOWD

IREC

TIONMOST

FREQUENTLYOBSERVEDINSURFICIALAQUIFER

R

EXPLANATION

LakeStarr

R

r

200

BasefromU.S.Geologic

alSurveydigitaldata,1:24,000,1987

UniversalTransverseMe

rcatorprojection,

Zone17

0 0

300

1,000

2,000FEET

600METERS

DIRECTIONOF

GROUND-WATERFLOW

Figure

13.

TopographicsettingofLakeStarr,generaldire

ctionofground-waterflow

inthesurro

undingsurficialaquifer,andthesimula

tedsteady-state

ground-watercatchment.

7/29/2019 wri02_4033_lee

27/61


As an approximation, the surface areas of thelakes can be represented as circular in shape (figs. 3-14). Representing the lake geometry in this mannerprovides a conceptual framework used in this reportfor modeling lake basins in radial section. Theassumptions of a nearly circular lake and radial sym-metry in the surrounding basin were used by Lee(2000) to simulate lake and ground-water interactions

using a two-dimensional variably saturated flowmodel. The principle radius (R) describing the largestcircular area of each lake ranged from about 240 ft inlength for Swim Lake to about 1,000 ft for three of thelarger lakes (Lakes Annie, Grassy, and Starr) (figs. 3-14 and table 2). The lake with the largest radius wasLake Hollingsworth (R = 2,250 ft). The distance fromeach lake margin to another point in the basin can be

expressed as a multiple of the radius of the lake. Forexample, the longest hillside in the Lake Olivia basinis about 3.5R (2,800 ft) in length, the shortest is about0.3R (240 ft) (fig. 11).

Topography and Ground-Water Flow Patterns

The areas of a lake basin capable of generating

ground-water inflow from a surficial aquifer cannot beaccurately determined by topography alone. When thetypical direction of ground-water flow in the basins ofthe 14 lakes was superimposed onto the basin topogra-phy, however, some general tendencies became appar-ent (figs. 3-14).

Most of the lakes were in a flow-through set-ting during some part of the year, with ground waterflowing into the lake along part of the perimeter andlake water flowing out along another. Flow reversals,when the typical ground-water flow direction tempo-

rarily reverses, were routinely observed in the basinsof Lakes Barco, Grassy, Halfmoon, Olivia, and Round.Often, during rainy months, recharge was sufficient toraise the water table above lake stage along much ofthe shoreline, increasing the inflow. For example, thenorthern perimeter of Lake Olivia (fig. 11) receivedground-water inflow only during June, July, andAugust 1996. After the rainy season ended, the watertable again sloped away from the lake in this area, andinflow decreased substantially (Sacks and others,1998). In other basins, drought conditions coupledwith poor confinement of the Upper Floridan aquifer

cause the adjacent water table to fall below lake stage,resulting in lateral leakage along parts of the perimeterthat typically receive inflow. The Lake Barco basinprovides an example of this effect (Sacks and others,1992).

In general, sustained ground-water inflowarrived from the areas of the basins with the highestridges. Lateral leakage and flow reversals typicallyoccurred on the side of the lake with the lowest basinelevations (figs. 3-14). Lateral lake leakage appearedmore likely where the drainage divide was close to the

lake margin, for example, south of Lake Barco (fig. 4),southeast of Lake George (fig. 6), west of HalfmoonLake (fig. 7), north of Lake Isis (fig. 9), and northwestof Lake Olivia (fig. 11). Lake Starr was somewhatatypical as it leaked toward the southeast, in the direc-tion of the highest topographic ridge. However, thisalso was the area of the basin characterized by con-spicuous sinkhole depressions (fig. 13).

SwimLake

R

N




LAND-SURFACE ELEVATION-- In feet abovesea level. Contour interval is 5 feet


R

EXPLANATION

Hesperides quadrangle 1952 (1972)


110

0

0 300

1,000 2,000 FEET

600 METERS

Figure 14. Topographic setting and general direction of

ground-water flow in the surficial aquifer around SwimLake.

7/29/2019 wri02_4033_lee

28/61


Proximity to other lakes also may affect the pat-tern of ground-water flow around the 14 lakes. Thirteenof the 14 lakes were within 2.5R of one or more lakes,and 8 were within 1.5R (table 2). Lake Annie was far-thest from any lake (about 4.5R). The tendency forground water from the intervening basin to enter thelake is affected by whether the adjacent lake is higher orlower. For example, lake water typically leaked out lat-

erally through the eastern shoreline of Round Laketoward a region of the eastern basin with low topogra-phy and a short hillside (fig. 12). Although the basintopography is similar on the western side of the lake, thelake consistently received ground-water inflow in thisarea. This may be because nearby Lake Winterset,which had a slightly higher stage than Round Lake, keptthe adjacent water table elevated (fig. 12).

Ground-water catchments were simulated inprevious studies for Lakes Barco, Five-O, and Starrusing three-dimensional, finite-difference, saturated

ground-water flow modeling combined with particletracking. The models resolved steady-state flow-fieldvelocities to determine the pathlines of ground waterentering the lake. The extent of these pathlinesdefined the catchment. All of the ground water withinthe simulated ground-water catchment eventuallyflows into the lake. The sizes of the ground-watercatchments for Lakes Barco, Five-O, and Starr differedsubstantially. The ground-water catchments to LakesBarco and Starr reached about 0.3R and 1.0R fromtheir respective lake margins, but both catchmentswere much smaller than their topographically defined

basins (figs. 4 and 13). In contrast, at Lake Five-O, theground-water catchment encompassed most of thetopographic basin (fig. 5).

Hydrogeologic Framework

The size of ground-water catchments dependsupon the hydrogeologic framework of the lake basin aswell as basin topography. Hillside views of the basinsreveal differences and similarities in the physical andgeologic settings of the 14 lakes (see figs. 15-20).Hillside views are the perspective used in the two-

dimensional modeling in the next section. The locationof each hillside is shown on basin maps (figs. 3-14).Hillsides begin in the lake center at the lowest land-surface elevation (lake bottom), and end at the hilltopof the topographic drainage divide. Hillside viewsplotted at the same horizontal and vertical scale showthe depth and slope of the lakebeds, basin radius andelevation, the thickness of the surficial aquifer and the

intermediate confining unit, as well as the thickness ofthe unsaturated zone (figs. 15-20). The bottom eleva-tion of each cross section is the approximate top of theUpper Floridan aquifer in the basin.

The thickness of the surficial deposits within thebasins ranged widely from about 10 ft at Lake Holling-sworth to about 300 ft at Lake Annie (table 3).Although the hydraulic conductivity of the surficialaquifer was not known for all of the 14 basins, in the5 intensively studied basins, the surficial aquifer con-ductivity ranged over an order of magnitude. The rep-resentative horizontal hydraulic conductivity (Kh) inthe shallow surficial aquifer was about 60 feet per day(ft/d) at Five-O, 30 ft/day at Lake Starr, 8 ft/d at LakeLucerne, 5 ft/d at Halfmoon Lake, and 3 ft/day at LakeBarco. Anisotropy was not directly determined in anyof the basins, but the surficial deposits were heteroge-neous and Kh typically decreased with depth in thesurficial aquifer.

The depth of the water table below land surfacealso varied considerably among basins. Along thesouthern extent of the Lake Starr basin, the water tablewas overlain by nearly 120 ft of unsaturated material(fig. 20). At Lake Annie, the maximum unsaturatedthickness was about 10 ft (fig. 15). Unsaturated zonethickness was affected more by topography than by theslope of the water table, which by comparison was rel-atively flat. At the end of a wet season, the maximumwater-table slope found on the inflow side of the lakesranged from 0.2 to 2.6 percent (2.6 ft vertical changeper 100 ft horizontal distance). The maximum slopemeasured was less than 1 percent at Lakes Annie,Grassy, Lucerne, Olivia, Saddle Blanket, Starr, andSwim. The maximum water-table slope was greaterthan 1 percent at Lakes Barco, Five-O, George, Half-moon, Hollingsworth and Isis. Lakes Barco and Isis,which were flow-through lakes, also had the largestnegative water-table slopes on their outflow sides(table 2).

The lakes ranged in depth from 6 ft (Holling-sworth) to 65 ft (Annie) (figs. 15 and 17, and table 3).Four of the 14 lakes were considered deep, with a

maximum depth greater than 45 ft (Annie, Five-O,Isis, and Olivia). The beds of these lakes also hadsteep slopes (maximum bed slope greater than 10 per-cent). Swim Lake, with a maximum depth of 30 ft buta small surface area (5 acres) had the steepest bedslope (maximum about 20 percent). Of the five lakesestimated to receive the highest ground-water inflow,four also had a steep bed slope (Annie, Five-O, Isis,

7/29/2019 wri02_4033_lee

29/61


Lake Annie

INTERMEDIATECONFININGUNIT

UNDIFFERENTIATEDSURFICIAL DEPOSITS

440

400

360

320

280

240

200

160

120

80

40

40

80

120

160

SEA LEVEL Lake center

Topographic drainage divide

4,0000 1,000 2,000 3,000

Lake Barco

10

20

60

100

140

180

SEA LEVEL

EXPLANATION

WATER TABLE

LAND SURFACE

LAKE SURFACE

ELEVATIONABOVE

ORBELOWS

EALEVEL,INFEET

TOP OF UPPER FLORIDAN AQUIFER

DISTANCE FROM LAKE CENTER, IN FEET

Figure 15. Simplified hydrogeologic sections along hillsides in thetopographic basins of Lake Annie and Lake Barco. Section locations areshown in figures 3 and 4.

and Swim Lakes (table 2)). The largeground-water inflow estimated for LakeHollingsworth, with its flat bottom, isprobably attributable to other basin char-acteristics. Three of the four deep lakeswere in Highlands County where the surf-icial deposits were greater than 200 ftthick (Lakes Annie, Isis, and Olivia)(figs. 15, 17, and 18). The fourth, LakeFive-O in the Florida panhandle, pene-trated the entire thickness of a compara-tively thin surficial aquifer (about 27 ftthick) (fig. 16).

The Hawthorn Group comprisesthe intermediate aquifer/confining unitbelow all of the lakes except Lake Five-O(Buono and others, 1979). The HawthornGroup is composed of several formationsthat confine the Upper Floridan aquifer to

varying degrees. For example, at LakeLucerne and Lake Starr in Polk County,detailed lithology and hydraulic head dataindicated that the lower part of the Haw-thorn Group (Arcadia Formation) is con-nected hydraulically to the UpperFloridan aquifer. The upper part of theHawthorn Group (the Peace River Forma-tion) provides the confinement. Similarly,at Lake Five-O, in the panhandle of Flor-ida, the Jackson Bluff Formation is con-sidered the intermediate confining unit,and is about 40-ft thick in the basin(fig. 16). A dense, black, shelly clay lessthan 10 ft thick present at the base of theJackson Bluff Formation, however, pro-vides most of the confinement (Andrewsand others, 1990). Because detailed lithol-ogy is not available for all of the basins,the convention of Buono and others(1979) is used in this report, wherein theentire thickness of the Hawthorn Group(or Jackson Bluff Formation) is referred

to as the intermediate confining unit.In the lake basins surveyed, theintermediate confining unit ranged inthickness from a minimum of about 5-10 ft at Halfmoon Lake to a maximum ofabout 250 ft at Lake Annie. The interme-diate confining unit was greater than200 ft thick in the southern range of the

7/29/2019 wri02_4033_lee

30/61


Central Lake District (southern Polk and High-lands Counties), and typically less than 150 ftthick in central and northern Polk County. Sub-sidence features breaching the intermediateconfining unit were identified beneath all fiveintensively studied lakes based on seismicreflection surveys (Lakes Barco, Five-O, Half-moon, Lucerne, and Starr). Breaches in the con-fining unit were also documented in the basinssurrounding Lakes Barco and Starr. Althoughrepresented as being flat in the hillside figures,the intermediate confining unit surface andthickness is uneven with variability in altitudeof tens of feet over distances of hundreds of feet(Tihansky and others, 1996). The contour inter-val used to map the thickness of this unit inregional maps was 50 ft (Buono and others,1979).

Soft lake sediment was described in the

intensively studied lake basins. Sediment cov-ered over two-thirds of the bottom of LakeBarco and was about 12 ft thick at the center(Sacks and others, 1992, fig. 8). Soft sedimentswere thinner at Lake Starr, (maximum thicknessabout 4 ft) and occupied less than half of thelake bottom area (Swancar and others, 2000,fig. 10). Thin and discontinuous sediments werereported for Lake Five-O by scuba divers(Roger Sweets, University of Kentucky, writtencommun., 1996). Sediment cores indicated thatsediment was present mostly in the deepestareas of Lake Lucerne and generally was lessthan 3 ft thick (Lee and Swancar, 1997). Simi-larly, soft sediment was zero to 4 ft thick in thecenter of Halfmoon Lake (Metz and Sacks,2002).

The downward head difference betweenthe lake and the Upper Floridan aquifer variedfrom about 2 to 59 ft in the 14 lakes surveyed inthis study (table 3). Due to differences in hydro-geology, the head difference typically wasgreater in lake basins in southern Polk County

and Highlands County than in central andnorthern Polk County (table 3). The limestoneof the Upper Floridan aquifer becomes deeperand the head in the Upper Floridan aquiferdecreases toward the south, whereas the overly-ing surficial deposits, surficial aquifer, andintermediate aquifer/confining unit thicken(Tihansky and others, 1996).


ELEVATIONABOVEORBELOWS

EALEVEL,INFEET

0 1,000 2,000 3,000



Grassy Lake150

100

50

SEALEVEL

50

Lake center


WATER TABLE

LAND SURFACE

LAKE SURFACE

EXPLANATION

Lake Five-O

40

20

80

120

SEALEVEL

180

140

100

60

20

Lake George

Halfmoon Lake80

40

SEA

LEVEL

Topographic drainagedivide

Figure 16. Simplified hydrogeologic sections along hillsides inthe topographic basins of Lake Five-O, Lake George, GrassyLake, and Halfmoon Lake. Section locations are shown infigures 5, 6, and 7.

7/29/2019 wri02_4033_lee

31/61


Lake Hollingsworth250

200

150

100

50

50

100

Lake Isis

320

280

240

200

160

120

80

40

40

80

120

160

200



SEALEVEL


Lake center

Lake Morton

WATER TABLE

LAND SURFACE

LAKE SURFACE

EXPLANATION

4,000 5,0000 1,000 2,000 3,000

SEALEVEL

ELEVATIONABOVE

ORBELOWS

EALEVEL,INFEET

INTERMEDIATECONFINING

UNIT


Figure 17. Simplified hydrogeologic sections along hillsides in the topographic basins ofLake Hollingsworth and Lake Isis. Section locations are shown in figures 8 and 9.

7/29/2019 wri02_4033_lee

32/61




Lake Olivia


Lake Lucerne


Lake center

4,000 5,0000 1,000 2,000 3,000

WATER TABLE (? WHERE ESTIMATED)

LAND SURFACE

LAKE SURFACE

EXPLANATION

320

280

240

200

160

120

80

40

40

80

120

160

200

SEALEVEL

40

80

120

160

200

SEALEVEL

ELEVATIONABOV

EORBELOWS

EALEVEL,INFE

ET

Topographic drainage divide

? ? ?

Figure 18. Simplified hydrogeologic sections along hillsides in the topographic basins ofLake Lucerne and Lake Olivia. Section locations are shown in figures 10 and 11.

7/29/2019 wri02_4033_lee

33/61


20

40

80

120

160

200Round Lake

200

160

120

80

40

40

80

120

160




SEA LEVEL

Saddle Blanket Lake


Lake center

4,0000 1,000 2,000 3,000

SEA LEVEL


EALEVEL,INFEE

T

WATER TABLE

LAND SURFACE

LAKE SURFACE

EXPLANATION


Figure 19. Simplified hydrogeologic sections along hillsides in thetopographic basins of Round Lake and Saddle Blanket Lakes. Section

locations are shown in figure 12.

7/29/2019 wri02_4033_lee

34/61


20

20

60

60

100

140

180

220

SEA LEVEL



100

50

50

100

150

SEA LEVEL

Lake Starr North

Swim Lake

INTERMEDIATE

CONFINING UNIT


4,0000 1,000 2,000 3,000

20

20

60

60

100

140

180

220

SEA LEVEL

Lake center

Breaches in the


Lake Starr South

WATER TABLE

LAND SURFACE

LAKE SURFACE

EXPLANATION



EALEVEL,INFEET

Figure 20. Simplified hydrogeologic sections along hillsides in thetopographic basins of Lake Starr and Swim Lake. Section locations areshown in figures 13 and 14.

7/29/2019 wri02_4033_lee

35/61


Regressions were used to examine relationsbetween head difference and physical setting. Ninelakes described in two earlier investigations of lakes inthe Highlands Ridge were included with the 14 lakespresented in this report (Geraghty and Miller, 1980;Barcelo and others, 1990) (table 3). In all cases, headdifferences were assumed to reflect steady conditions(no change in head with time). The head differencesbelow these 23 lakes were significantly correlated to themantle thickness (sum of surficial deposits and interme-diate confining unit) in the basin (fig. 21a), but not tothe elevation of the lakes, as suggested by Geraghty andMiller (1980) (fig. 21b). Most of the variation in headdifference was explained by the thickness of the inter-mediate confining unit (R2=0.43 for intermediate con-fining unit thickness alone, compared with R2=0.27 forsurficial aquifer thickness alone).

NUMERICAL MODELING OFGROUND-WATER FLOW

Numerical modeling was used to simulate thepotential effects of the physical setting of lake basinson the ground-water inflow to lakes. Ground-waterflow beneath a series of conceptualized lake basinswas simulated using a two-dimensional variably satu-

rated flow model. The model was applied along aradial transect (hillside) through a series of hypotheti-cal, circular lake basins. None of the lakes surveyeddisplayed complete radial symmetry in their interac-tions with the adjacent surficial aquifer, althoughground-water interactions can be symmetric alongsome arc of a lake perimeter. Assuming radial symme-try for the entire lake basin is an oversimplificationmade for purposes of numerical modeling.

Methods

Variably saturated flow was simulated using theHYDRUS-2D flow modeling software (imunek andothers, 1996). At the core of HYDRUS-2D is the com-puter program SWMS_2D, a two-dimensional, finite-element model code that numerically solves the Rich-ards equation for unsaturated-saturated water flow inporous media and the Fickian-based convection-dis-persion equation for solute transport (imunek, et al.,1992). HYDRUS-2D unites this modeling code with asophisticated graphical user interface. The model sim-ulates two-dimensional, isothermal, Darcian flow ofwater in a variably saturated rigid porous media. Airflow is assumed to be insignificant. The governingflow equation for variably saturated flow is Richardsequation as applied in imunek and others (1996):

, (1)

where: is the volumetric water content [L3 L-3],h is the pressure head [L],S is a sink term [T-1],

xi (i=1,2) are the spatial coordinates [L],t is time [T], and

are the components of a dimensionlessanisotropy tensor KA used to account foran anisotropic medium.

(B)

0

10

20

30

40

50

60

70

0 50 100 150

LAKE ELEVATION, IN FEET ABOVESEA LEVEL

(A)R = 0.4632

0

10

20

30

40

50

60

70

0 200 400 600 800

MANTLE THICKNESS, IN FEET

HEADDIFFERENCE,

INFEET

HEADDIFFERENCE,

INFEET

R = 0.09112

Figure 21. Relation of head difference betweenthe lake and Upper Floridan aquifer to (A) mantlethickness near lake and (B) lake elevation.(Mantle thickness is the combined thickness ofthe surficial deposits and intermediate confiningunit.)

t------

xi

K K

ijA h

xj

------- K

izA+

S=

KijA

7/29/2019 wri02_4033_lee

36/61

Numerical Modeling of Ground-Water Flow 29

K is the unsaturated hydraulic conductivity func-tion [LT-1] given by

, (2)

where:Ks is saturated hydraulic conductivity and

Kr is relative hydraulic conductivity [L T-1

].If equation (1) is applied to planar flow in a ver-

tical cross section,x1 =xis the horizontal dimensionand x2 =z is the vertical dimension, assumed positiveupward. For flow in an axisymmetric system,x1= r,where ris the radial distance from the center line.Model simulations were performed using metric units,but are expressed in English units for this report.

Radial Models

Ground-water flow was simulated for a radialsection that represents a hillside taken from a circularbasin. The catchment size was delineated using the

simulated ground-water velocity vectors. The catch-ment size also was derived using the modeled ground-water inflow rate and with the recharge flux along thehillside. Both methods gave comparable estimates ofthe length of the catchment along the hillside.

The model described in figure 22, without thebreaches in confining unit onshore of the lake, was the

starting point for most of the steady-state simulations.Variations on this original case were made for succes-sive simulations. The lake had a radius of 656 ft(200 m), and a maximum depth of 24.6 ft (7.5 m). Thehillside extended 1,968 ft (600 m) or 3R from the edgeof the lake to the topographic drainage divide. The ver-tical relief from the deepest point on the lake bottom tothe hilltop was 39 ft (12 m). The conceptual basin wasanalogous to a circular lake with a surface area of31 acres surrounded by a 465-acre basin with lowtopographic relief. The hypothetical lake is similar insize to Lakes Lucerne, Barco, and Round (table 2).

The model domain represented the lake and themantle deposits overlying the Upper Floridan aquifer.

( ) ( ) ( )tzxhKzxKtzxhK rs ,,,,,,, =

UndifferentiatedSurficial DepositsLake

LakeSediment

0 328 656 984 1,312 1,640 1,968 2,296 2,624

40

120

140

60

20

100

80

0

RADIAL DISTANCE FROM MODEL ORIGIN, IN FEET

ELEVATIONABOVEMODELORIGIN,INFEET

R0.5R

1R

2R

3R

Distance onshore

Breaches inIntermediateConfining Unit


Figure 22. Schematic of the initial lake basin model. Breaches in the intermediate confining unit around the lake werenot included in the initial lake basin model, but were used in subsequent simulations.

7/29/2019 wri02_4033_lee

37/61


The intermediate confining unit, in the lower part ofthe model, was 65 ft thick. Directly below the lake, theintermediate confining unit has been breached by asinkhole structure and replaced with surficial deposits.Surficial sand and clay deposits ranging in thicknessfrom 40 to 60 ft overlie the intermediate confiningunit. Lake sediment underlies the lake, and this sedi-ment has a much lower hydraulic conductivity than theencompassing surficial deposits.

The hydraulic characteristics of the hydrogeo-logic units used in the model are shown in table 4 andare loosely adapted from the Lake Barco basin. Theunconsolidated surficial deposits were assumed to bepredominantly sand with a hydraulic conductivity of3 ft/d. The daily flux of water entering the hillside wasbased on 35 percent of the average annual rainfall rateof 52 in/yr. The head in the Upper Floridan aquiferwas 4.9 ft below the lake stage. With these hydraulicparameters set, the Kh of the intermediate confinin

wri02_4033_lee

Documents

Transcript of wri02_4033_lee