THE RELATIONSHIP BETWEEN URBANIZATION AND STREAYFLOW
IN THE HILILSBOROUGH RIVER BASIN, 1940-1970
PETER SALVATORE SEGRETTO
A DISSERTATION PRESENTED TO THE GRADUATE COUNCIL OF
THE UNIVERSITY OF FLORIDA
IN T'ARTIAL FULFILLMENT OF TFE REQUIREMENTS FOR THE
DEGREE OF DOCTOR OF PHILOSOPHY
UNIVERSITY OF FLORIDA
Copyright 1975 by Peter Salvatore Segretto
All rights reserved. No part of this
document may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopying,
recording or by any information storage and retrieval system, without permission in writing from the author. However, permission is given to cite short textual references provided that proper
citation is given to the author.
This dissertation could not have been completed
without the assistance of many people. I am especially indebted to my advisor, Dr. Clark I. Cross, for his constructive criticisms and suggestions during the preparation of this research along with his continued encouragement and guidance throughout my program of study. Special appreciation is also extended to the members of my supervisory committee who have added to the depth of my geographic and geologic knowledge necessary for completion of this study. Gratitude is expressed to the numerous professional people in the Tampa area who assisted in the collection of the necessary data and information for this research.
In addition to these people, deep appreciation is expressed to my wife, Ellen, to whom I dedicate this dissertation, for her many years of encouragement and personal sacrifice.
TABLE OF CONTENTS
ACKNOWLEDGEMENTS .............. .......... ...... iv
LIST OF TABLES ................................. vii
LIST OF FIGURES ................................ x
ABSTRACT .............................. . . .. xiv
I. INTRODUCTION ........ ......... . .......... 1
Location of the Area .................... 3
Urban Effects on Hydrologic Systems ...... 6
Urban Trends ........................ 6
Possible Relationships .............. 8
The Runoff Process .................. 10
Research ..... . .... ........ ....... 17
Scope of the Study --..... .............. .. 26
Subregions ......................... . 26
Method of Analysis .................. 28
II.PHYSICAL SETTING ........... .............. 30
Geology of the Study Area ............... 31
Topography ..........-.. ....-. . -...... .... 35
Physiographic Regions ................,... 42
The Lime-Sink Region ................ 45
The Flatwoods ....................... 47
The Polk Uplands ............ o ....... 48
Climate .......,......... o............ o..... 49
Temperature and Precipitation ....... 52
Climate According to Thornthwaite ... 58
Vegetation and Soils ..................... 71
Pine Flatwoods ................... . ... 72
Upland Pine and Scrub Oak Forests ... 74
Swamp Forests ....................... 74
Soils --........ ............. ..... o... 75
Hydrology .... ............................ 79
Ground Water and the Artesian System. 79
The Water Balance ................... 82
III. HISTORICAL DEVELOPMENT AND LAND USE ........ 85
Settlement and Development to 1940 .... 87
Development Patterns 1940-1970 ..,... o.. 92 Population Patterns .................. 97
Land Use ........ ....... .............. 102
Subregion Land Use ......................... 106
Blackwater Creek ............. .... 106
Alligator Creek ....................... 107
IV. FLOW EXTREMES AND THE F.R.B. PROJECT ....... 112
Soil Flooding ... ...... .............. ..... 114
Channel Flooding and Prediction ............ 119
Floods of Record...................... 119
Frequency Estimates ................... 120
Flood Profiles ................. ......... 124
The Four River Basins Project .............. 131
General Plan of Development ........... 136
The 1962 Proposed Plan for the
Hillsborough River Basin ........... 139
Present Status of the F.R.B. Plan ..... 148
V. IMPACT OF INCREASED URBANIZATION ON THE
HYDROLOGIC CHARACTERISTICS OF THE
HILLSBOROUGH RIVER BASIN ................ 153
Base Flow and the Ground Water Table ....... 155
The Floridan Aquifer .................. 155
Ground Water Depletion in Northwest
Hillsborough County ................ 162
Changes in Surface Runoff .................. 173
Combined Effects of Urbanization on
-Streamflow . ..... ... ... .... ...... 181
Hydrograph Analysis ............... 186
Cumulative Curve Analysis ............. 191
Regression Analysis ................... 199
Conclusions .... . . ......... o ........ o ..... 209
I. PROCEDURES USED TO CALCULATE POTENTIAL
EVAPOTRANSPIRATION IN THE HILLSBOROUGH
RIVER BASIN .. ...... ... . ... .. .. . 217
II. AIR PHOTOGRAPHS: ALLIGATOR CREEK BASIN, 1974 223
III. STATISTICAL CALCULATIONS ................. 249
REFERENCE NOTES ................... ........ ...... 257
REFERENCES ........................................ 258
BIOGRAPHICAL SKETCH ........................ ....... 265
LIST OF TABLES
1. Population of Florida and the Tampa Bay Region: 1940-1970 ........................... 2
2. Population Growth Within Standard Metropolitan Statistical Areas: 1960-1970 ........ 6
3. Values of Constants in Horton's Interception
Equation .................................... 12
4. Climatic Variance Over Urban Areas .......... 19
5. General Characteristics of Physiographic
Regions ..................................... 46
6. Climatological Data for Tampa, Florida ...... 50
7. Translation of Koeppen Cfa Climatic Classification for Tampa, Florida ................. 51
8. Comparison of Koeppen and Thornthwaite Climatic Types ................................. 51
9. Precipitation Data for Seven Recording Stations - Hillsborough River Basin ............ 56
10. Areas of Thiessen Polygons .................. 58
11. Comparison of Mean Precipitation Values ..... 59
12. 1931 Thornthwaite Classification of Climate . 61
13. Monthly Water Balance Table - Hillsborough River Basin ................................. 65
14. Limits of Thornthwaite's Moisture Index ..... 69
15. Limits of Thornthwaite's Index of Thermal Efficiency .................................. 69
16. Seasonal Variation of Effective Moisture .... 71
17. Limits for Summer Concentration of Thermal Efficiency Classification ................... 72
18. Associated Vegetation and Soil Types Hillsborough River Basin .................... 76
19. Description of Soil Profiles
Hillsborough River Basin .................... 77
20. Runoff Coefficients for West Central Florida. 87 21. Hillsborough River Basin, Land Use - 1970 ... 105 22. Blackwater Creek, Land Use - 1963 ........... 109
23. Alligator Creek, Land Use - 1970 ............ 110
24. Ground Water Flood Hazard of Specific Soil
Types in the Hillsborough River Basin ....... 118
25. Hillsborough River Basin - March 1960 Flood
Data ........................................ 121
26. Recurrence Intervals for Annual Maximum Flows
at the Tampa Dam ............................ 125
27. Flood Profile Data for the Hillsborough
River Basin - Tampa Dam to Fletcher Avenue .. 129
28. Public Water Systems of Pinellas County ..... 163
29. Average Ground Water Withdrawal from Well
Fields of Northwest Hillsborough County for
Various Years ........................... 169
30. Projected Urban Land Use and Population
Hillsborough County, Florida -............... 213
31. Monthly Values of i Corresponding to Monthly
Mean Temperature ............... ............ . 219
32. Monthly Values of Unadjusted Potential 0Evapotranspiration at Temperatures Above 80 F .... 220
33. Mean Possible Duration of Sunlight in the
Northern Hemisphere Expressed in Units of
30 Days of 12 Hours Each ................... 221
34. Determination of Adjusted Potential Evapotranspiration for the Hillsborough River
Basin --------- ....-- ..- -.. -. ------.. ------- 250
35. Streamflow and Precipitation Data in the
Blackwater Creek Basin ...................... 251
36. Regression Analysis of Blackwater Creek..... 252
37. Streamflow and Precipitation Data in the
Alligator Creek Basin ...................... 253
38. Regression Analysis of Alligator Creek ..... 254
LIST OF FIGURES
1. Hillsborough River Basin and Surrounding Area ................................ ..... 5
2. Components of Base Flow ..................... 16
3. Storm Hydrograph - Assunpink Creek .......... 22
4. Hydrograph Comparison ....................... 23
5. Geology ..................................... 32
6. Columnar Sections ........................... 37
7. Florida Topographic Regions ................. 38
8. Florida Marine Terraces ..................... 40
9. Physiographic Regions - Hillsborough River Basin ............ ...... ...... ............... ... ... ... ... ... ... ... 44
10. Thiessen Polygons - Hillsborough River
Basin ....................................... 54
11. Thornthwaite Comparison of Potential Evapotranspiration and Precipitation - Hillsborough River Basin .............. ........... 67
12. North Hillsborough County - River Basins .... 88
13. North Hillsborough County - Development, 1940 93
14. North Hillsborough County - Development, 1948 94
15. North Hillsborough County - Development, 1970 96
16. North Hillsborough County - New Subdivisions,
196o-1970 ............ *..........*.......... . 98
17. North Hillsborough County - Population
Growth, 1950-1960 .............. 0 ...... .... . 100
18. North Hillsborough County - Population
Growth, 1960-1970 ................*.......... 101
19. North Hillsborough County - Land Use, 1970 .. 103
20. Alligator Creek Subregion - Changing Development 1954-1970 ......................... 108
21. Soil Flooding ..--.. -... ... --. .............. 116
22. Gumble Frequency Distribution - Hilisborough
River at Tampa Dam .........-... ......... 127
23. Flood Profiles of the Lower Hillsborough
River - ... -...........- ... . - -............ . 130
24. Tampa Reservoir . .. ..... -.. -................ 129
25. Tampa Dam ................................... 130
26. Lower Hillsborough River .................... 130
27. Stage Recorder -------........----.. ....... 132
28. Tampa Reservoir - Low Level ................. 132
29. Intake Structure ................................... .. 133
30. New Intake Structure .............--...-.-.- 134
31. Four River Basins, Florida - Project Area ... 137
32. 1962 F.R.B. Plan - Hillsborough River Basin . 140
33. Green Swamp Reservoir - Area-Volume Curves .. 141
34. Lower Hillsborough Reservoir - Area-Volume
Curves ...................................... 144
35. Big Cypress Reservoir - Area-Volume Curves .. 146
36. Completed Tampa Bypass Canal ................ 151
37. Tampa Bypass Construction ........-........ 151
38. Crystal and Sulphur Springs - Locaticn ...... 156
39. Changes in Level of the Floridan Aquifer
West Central Florida, 1949-1969 ............. 158
40. Potentiometric Surface of the Floridan
Aquifer - Hillsborough County, Florida,
1949-1969 ....--- . .. .. -- - - - ......------... -.. 159
41. Changes in Level of the Floridan Aquifer
West Central Florida, 1964-1969 ..-.......... 161
42. lowered lake Level ...................--...- 166
43. Lake Rogers ... .. ......---- --- ----. -------- 167
44. St. Petersburg Facilities .........---------- 170
45. St. Petersburg Well Head ------------ . 170
46. Lake Rogers ............... ..................171
47. Lake Rogers Well Head ....................... 171
48. Shady Oaks Plaza .-.-..................... . 175
49. New Office Complex ....---------------------- 175
50. Kash and Karry Plaza .......----------------- 276
51. New Landfill on North Florida Avenue ........ 176
52. North Florida Avenue Drainage Culvert ...,... 177
53. Siobhan Avenue Drainage Easement ............ 178
54. Siobhan Avenue Flooding h..n...e..e...oo ...n . 178
55. Alligator Creek Basin - Urban Area, 1974 ... 184
56. Alligator Creek - Storm Hydrograph, Less Than 10% Urban .............................. 187
57. Blackwater Creek - Storm Hydrograph ......... 188
58. Alligator Creek - Storm Hydrograph, 74% Urban t Curv-es-..-- - B.-a... r C k - ...... 190
59. Cumulative Curves - Blackwater Creek ........ 193
60. Cumulative Curves - Alligator Creek .-------- 195
61. Alligator Creek Basin - 1965 ................ 198
62. Regression Graph ....... ..................... 202
63. Regression Graph - Blackwater Creek ......... 205
64. Regression Graph - Alligator Creek .......... 207
65. North Hillsborough County - 1990 Projected Increase in Urban Land Use ............... 212
- Hillsborough River Basin Creek Basin - Section A,
Nomogram Alligator Alligator Alligator Alligator Alligator Alligator Alligator Alligator Alligator Alligator Alligator Alligator
Creek Creek Creek Creek Creek Creek Creek Creek Creek Creek Creek
Basin Basin Basin Basin Basin Basin Basin Basin Basin Basin Basin -
Section Section Section Section Section Section Section Section Section
Figure 79 ..
Figure Figure Figure Figure Figure Figure
Figure Figure Figure Figure Figure
79. Location of Air Photos ........................
79 79 79 79
79 79 79 79
225 227 229 231 233
241 243 245 247
Abstract of Dissertation Presented to the
Graduate Council of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy
THE RELATIONSHIP BETWEEN URBANIZATION AND STREAMFLOW IN
THE HILLSBOROUGH RIVER BASIN, 1940-1970 By
Peter Salvatore Segretto
Chairman: Clark I. Cross
Major Department: Geography
The major objective of this research was to determine the impact of urbanization on the hydrologic component of streamflow in the Hillsborough River Basin between 1940 and 1970. A brief look at the runoff process within the Hydrologic Cycle along with a general discussion of similar research in other regions was helpful in determining several possible relationships in the study area. Final results were based on field observations and multiple methods of analysis of precipitation and streamnflow data.
Before determination of the basic relationship could be made, it was necessary to establish the hydrologic character of the region. The climatological and geological settings were therefore determined. Because of the variable nature of precipitation on the 690 square mile study area, average rainfall was calculated by using data from seven recording stations and constructing the respective Thiessen polygons. In addition, the relative
distribution of available water within the region was based on temperature relationships along with adjusted precipitation figures according to the Thornthwaite water balance.
Several limiting factors such as large basin size, relative extent and location of urban area, and tidal influence on the streamflow recording gauge, necessitated the selection of two subareas for analysis. The Blackwater Creek Basin was used as a rural control because of the noticeable lack of urban development in the region. The Alligator Creek Basin, although located fifteen miles west of the major study area was selected as an urban subregion because of its physical and developmental similarities to the urban portion of the Hillsborough River Basin.
The historical developmental patterns of the major study area and the subregions were examined to determine the relative extent of urbanization. In addition, the recurrence of flood and drought periods was also examined and related to the major water problems of the region. Analysis of the 1962 Army Corps of Engineers' Four River Basins Project, along with its major revisions, revealed the relative usefulness of the provisions of this program.
The streamflow response to precipitation was exa-yiined in both the Blackwater Creek and Alligator Creek Basins by the use of three methods of analysis. These
were: 1) Storm Hydrograph Analysis, 2) Cumulative Curve Analysis, and 3) Regression Analysis. Investigations revealed that increased urbanization in the study area has increased streamflow in the Hillsborough River Basin. In addition, comparative results from regression analysis determined that over 33 percent of the variation in streamflow in the urban basin was explained by urban development.
The results from this study have shown that increased urbanization has generally increased streamflow. However, localized field observations have also shown that increased surface runoff has decreased ground water recharge, thus decreasing base flow during low flow periods. Overall, increased urbanization in the study area has intensified the regional water problem. General recommendations for the regional water program include implementation of the revised F.R.B. Plan along with local modifications to provide for increased surface water retention and local ground water recharge.
Recently, Florida's population haa been one of the fastest growing in the country. However, the pattern of population is not uniform as some regions have witnessed much greater increases than others. Since 1960, population growth within the tri-county area of the Tampa Bay Region (Hillsborough, Pinellas, and Pasco) has accounted for a large portion of the overall increase in population; thus resulting in Florida's present status. Analysis of the statistics in Table 1 will indicate the rate and basic distribution of population growth within the region since 1940.
Population in the Tampa-St. Petersburg Standard
Metropolitan Statistical Area has increased considerably within the period of record.* The 1970 census data show that between 1960 and 1970, population increased 31.1 percent in this area. This figure, however, does not present the true spatial distribution of growth within the region. Further inspection of the data will reveal
*SMSA was not used until the 1960 Census. For convenience of comparison, S"SA for 1940 and 1950 includes summation of data for Hillsborough and Pinellas Counties and Tampa-St. Petersburg Inner Cities.
Table 1: Population of Florida and the Tampa Bay Region: 1940-1970
St. Petersburg 60,812
Hillsborough County 180,148
Pinellas County 91,852
Pasco County 13,981
(Hillsborough,Pinellas) 272,000 SMSA - Inner City 169,203
SMSA - Outside Inner 102,797
Source: Bureau of the Census, 1971
249,894 159,249 20,529
409,143 221,419 187,724
274,970 181,298 397,788 374,665 36,785 772,453 456,268 316,185
490,265 522,329 74,955
Period Trend Percent Inc. 1940-1970 258%
157% 256% 172%
443% 272% 192%
Recent Trend Percent Inc. 1960-1970 37.1% 1.0%
39.4% 106.5o 31.1% 8. Y
that during this period, the inner cities increased only 8.3 percent while the population of the fringe areas or counties increased at the rate of 64 percent (Bureau of the Census, 1971). It is doubtful that this 64 percent increase resulted from a purely rural population. Rather, it has been caused by suburban development.
Recent urban development within the Tampa Bay Region, therefore, has transcended traditional boundaries. Accommodations for this new population since 1940 have changed the basic land use pattern of the area. The effect that these changes have had on the hydrological systems of the Hillsborough River Basin is the subject of this paper.
Location of the Area
The Hillsborough River Basin is located in the west central portion of the State of Florida (Figure 1). The river rises in the Green Swamp just north of the city of Lakeland where it is separated from the Withlacoochee River by a low topographic saddle. During periods of high water, both rivers share a common source. From the Green Swamp, the Hillsborough River flows predominantly to the scuthwest for fifty-four miles until it enters Hillsborough Bay at Tampa. The river drains a watershed of approximately 690 square miles, including portions of Hillsborough, Polk and Pasco Counties.
The major tributaries to the Hillsborough River include Blackwater Creek, Pemberton Creek and Flint Creek
on the south and Cypress Creek, New River and Trout Creek on the north. Blackwater Creek rises west of Lakeland
and flows to the river just east of the Hillsborough River State Park. This tributary drains a watershed of approximately eighty-six square miles. Pemberton Creek rises just north of Plant City and empties into Lake Thonotossassa. Flint Creek is an outlet to this lake and flows northward to the Hillsborough River. Together, Pemberton Creek and Flint Creek drain a sixty-five square mile basin.
Of the northern tributaries, Cypress Creek is the largest draining about 165 square miles. It rises in Pasco County northwest of San Antonio and flows southward to the Hillsborough. New River begins near St. Leo and joins the main river just west of the state park. Along with Trout Creek which rises near Wesley Chapel, these two tributaries drain a basin of about 135 square miles (Watson and Company, 1967).
Figure 1 shows the spatial distribution of urban
centers within and around the Hillsborough River Basin. In addition, this map shows the location of Alligator Creek near Safety Harbor, Florida. It should be noted that Alligator Creek is outside the study area. It is treated in this study because it is believed that to do so will clarify hydrologic relationships. This Creek drains a relatively small, but highly urbanized area of about nine square miles on the Pinellas Peninsula.
.PLANT TA MPA'
* LAKELAND CITY
UNDA R Y
Further discussion of the use of Alligator Creek will be found later in this chapter.
Urban Effects on Hydrological Systems Urban Trends
Since 1900, the population of the United States has tended to concentrate in cities and around urbanized areas (Murphy, 1966). At one time, it was necessary to live in a city in order to enjoy the conveniences and services of the central business district. Recently, however, the rate at which our major cities have grown, has declined in comparison with their related urban fringes. Table 2 compares urban growth with urban fringe growth and illustrates the trend of population movement from 1960 to 1970.
Table 2: Population Growth Within Standard Metropolitan
Statistical Areas: 1960 to 1970
Location 1970 1960 Increase 1960-1970
In SMSA's 139,418,811 119,594,754 19,824,057 16.6 In Cent.
Cities 63,796,943 59,947,129 3,849,814 6.4
Out of Cent.
Cities 75,621,868 59,647,625 15,974,243 26.8
Source: Bureau of the Census, 1971
Movement to the suburbs in the last ten to fifteen years has increased rapidly for several reasons. The
development of better transportation systems and the shopping center concept have made suburban living more convenient. Outside of corporate boundaries, taxes are considerably lower and most people find open space a pleasant change from the crowded city. Many of the city's advantages are now found in the suburbs. Shopping centers provide services and convenience, replacing the function of the C.B.D. without the related problems of traffic and congestion.
Although there are many advantages to living outside
of central cities, there are also several social and economic disadvantages.* For this reason, many geographers, sociologists and other social scientists have examined the rural-urban fringe (Wehrwein, 1942) in great detail. However, traditional studies of this area have been limited to delineation of the region and discussion of the social and economic implications of urban encrouchment on agricultural land.
This conventional approach has been exemplified by
several authors such as Wehrwein (1942), Blizzard (1954), and Murphy (1964). Until recently, geographers have neglected studying the effects of rural-urban land use changes on physical and hydrological systems. Most of
*It is beyond the scope of this paper to discuss the problems of urban fringe areas in any detail. However, it should be mentioned at -this point that the major
disadvantages of urban fringe living are usually found in the lack of zoning regulations and the often insufficient provision of county fire and police protection (Murphy, 1966).
this research has been done by civil engineers, geologists (urban hydrologists), and to a lesser extent, climatologists. Consideration of the "Hydrologic Cycle" as an integral part of Physical Geography and the recognition of Urban Geography as a viable segment of the discipline, justify classifying this type of study as geographic in nature.
Many geographers consider urbanization of a region
to include the development of residential, commercial and industrial areas. Such land use changes in transforming
a formerly natural or agricultural area would appear to modify considerably the natural hydrologic characteristics of that region or watershed. Direct runoff and streamflow are also noticeably affected. Possible Relationships
Construction of roads, parking lots and structures related to urbanization can decrease the surface area of the basin capable of absorbing precipitation. Increases in streamflow can result from this additional impervious surface area. Also, less surface water is lost to infiltration and ground water recharge, thus creating an additional increase in direct runoff.
Increased urbanization may on the other hand, decrease streamflow in a river basin. This relationship can result from an actual decrease in surface area drained by a particular river. Unnatural drainage patterns are frequently created by man-made drainage channels.
Sometimes these divert a portion of direct runoff to a neighboring basin. This decrease in catchment area can subsequently decrease streamflow. There is at the same time a likelihood of artificial increases.
A third relationship between urbanization and streamflow may be possible. Urban development may have no net effect on direct runoff within a given watershed. If the basin area was sufficiently large and the urban development was limited to a relatively small portion of the total area, the hydrologic response to urbanization might be insignificant. Also, net increases in direct runoff due to land use change may be negated by factors which tend to decrease runoff. This relationship would also result in a negligible change in streamflow.
Basin alterations due to increased urbanization may in effect create an overall change in the regime of flow of a river. For the most part, this effect would be best seen in the hydrologic response of streamflow to storm runoff. Increased impervious surface area tends to increase the rate at which peak flows are attained and decrease their relative duration. A more detailed discussion of possible regime changes will be found later in this chapter and in Chapter V. First, a brief analysis of the runoff process is necessary for a better understanding of the effects of urban development on streamflow.
The Runoff Process
Direct runoff is that portion of precipitation that flows over the surface of the earth in thin sheets (overland flow) or in systematized channels. The amount of direct runoff from any given storm is dependent upon several factors such as precipitation intensity and duration along with surface configuration of the basin. (The latter may be considered a function of physical makeup, climatic situation and cultural activity.) When precipitation reaches the earth, it may either be stored on the surface, infiltrate through the soil, or flow overland. The first aspect of the runoff process to be considered will be surface storage.
Surface detention is the net water loss to runoff created by interception and depression storage. In the initial stage of a storm, a large portion of the rainfall is intercepted by vegetation and cultural structures. Some of this moisture is quickly evaporated and never reaches the ground. The amount of rainfall originally lost to interception in a natural watershed is dependent upon the type and density of vegetation present. Horton (1919) developed a method to calculate the amount of water intercepted during any given storm. This quantity may be determined according to the formula:
I = a + b pn
where I is equal to Interception, P is equal to total
precipitation in inches and a, b, n are constants derived from differing types of vegetation (Horton, 1919). Table
3 lists the constants for several selected types of vegetative cover.
Kittredge observed that between 2 and 40 percent of the total rainfall from a storm may be initially lost to interception within a well developed forest. He found, in his 1937 study, that within a forest of Australian Eucalyptus, only 2 to 3 percent of precipitation was intercepted, while in an area covered by California Hemlock and Douglas Fir up to 40 percent of the moisture was initially prevented from reaching the ground (Bruce and Clark, 1969; Horton, 1919). From these studies, it is evident that the role of interception of rainfall by natural vegetation is significant in the runoff process.
As precipitation continues, the storage capacity of the natural cover is satisfied and stem flow and drip occurs, allowing a greater proportion of moisture to reach ground level. Water then begins to infiltrate, adding to soil moisture and subsequently to ground water recharge. Depression storage does not occur until the soil has become saturated (satisfying soil moisture capacity) or until the rainfall intensity exceeds the infiltration rate of the basin surface. At that point, the smallest depressions within the basin begin to fill, coalesce and spill over into larger depressions.
Table 3: Values of Constants in Horton's Interception
Vegetation a b n
Orchards 0.04 0.18 1.00
Oak Woods 0.05 0.18 1.00
Maple Woods 0.04 0.18 1.00
Willow Shrubs 0.02 0.40 1.00
Hemlock and Fine Woods 0.05 0.20 0.50
Clover and Meadowgrass 0.005h* 0.08h 1.00
Small Grains, Rye, Wheet, Barley 0.05h 0.05h 1.00
*h - height of plants (in feet) Source: Bruce and Clark, 1969; Horton, 1919
Ultimately, all of the depressions fill and direct runoff in the form of sheetwash and streamflow begins. It should be considered, however, that water held in depression storage does not go directly to runoff. Water stored in this method is held in place (such as in large flats or marsh areas) until such a time that it evaporates or infiltrates, replenishing soil moisture or the ground water table. The amount of water held in depressions is inversely proportional to the infiltration rate and slope of the basin. In most areas, water will readily flow over a surface with significant relief, thus decreasing the overall amount of surface water retention and total infiltration within the basin. In karst areas, however, this relationship between relative slope and infiltration is altered somewhat by the porous characteristics of surficial material and the bedrock complex. Therefore, with
the exception of many karst regions, slope is an important factor governing the infiltration rate.
Infiltration is the movement of surface water into the pore spaces and openings of the soil. The amount of precipitation allowed to infiltrate during a particular storm can be measured quantitatively (Horton, 1939). However, this study will be concerned only with empirical increases or decreases in the infiltration rate caused by changes in land use. In addition to slope, the other factors affecting this rate are; frequency and intensity of precipitation along with the related soil moisture content, physical characteristics of the soil and vegetative cover.
Soil moisture is interrelated with the other factors governing the infiltration rate. It is considered to be the amount of water held in the soil, equal to the total of hydroscopic and capillary water. Major depletions to soil moisture are caused by percolation to ground water, use by vegetation (photosynthesis and transpiration) and evaporation caused by capillary movement of water to the surface during dry periods. If sufficient time elapses between storms for these processes to operate effectively, soil moisture capacity will be increased, thus increasing potential infiltration. If the time period is short, however, soil moisture capacity will be satisfied quickly and the infiltration rate will decrease. In this circumstance, the effect is to shorten the lag time before runoff occurs.
Soil moisture and infiltration capacities are also dependent upon the physical characteristics of the soil. Coarsely textured soils (sand) have a greater percentage of pore space than fine textured soils (clay). For this reason, they have a greater ability to absorb moisture and their occurrence increases the amount of infiltration within a watershed. In addition to their fine texture, clay soils usually contain a relatively high percentage of colloidal material. This causes the individual particles of clay to swell during periods of wetting, further decreasing their pore space and ability to infiltrate moisture. Besides texture, infiltration is affected by the compactive nature of the soil. Tightly packed soil created by rain striking barren ground or by the activity of man and animals, can greatly decrease pore space in the soil, thus decreasing its ability to soak up moisture. The impact of this condition, however, may be minimized by the presence of vegetation.
It has been found in several cases, that the type of vegetation present in an area is more significant in determining infiltration capacity than soil type (Wisler and Brater, 1967). A dense cover of vegetation is known to rapidly increase the infiltration rate. The protective cover of vegetation intercepts large volumes of water during a storm, thus decreasing the effective intensity of precipitation and preventing soil compaction. In addition, vegetation tends to separate soil particles with
its roots and provides an environment where the action of animals and insects can, in effect, aerate the soil. In each instance, additional pore space is created, thus increasing the infiltration capabilities of the basin surface. It should be remembered that each factor which tends to increase infiltration also tends to decrease direct runoff and streamflow within the basin.
The final aspect of the runoff process to be considered is that of systematized channel flow or streamflow. Streamflow may be divided into two major components; direct runoff and base flow. During storm periods, the leading contributor to streamflow is direct runoff which was discussed earlier with respect to infiltration and surface storage. Base flow, however, is that portion of water supplied to a river during periods when direct runoff is lacking. It consists of indirect runoff (subsurface flow) and ground water seepage (Figure 2).
Indirect runoff is the net movement of water beneath
the basin surface. It is the result of two primary factors; gravity movement of soil moisture and the movement of ground water. (In karst regions, the movement of water through subsurface streams must also be considered.) In each case, the amount of indirect runoff is directly proportional to the infiltration capabilities of the basin surface.
In a similar sense, the level of the ground water
table is a function of infiltration. Reduced percolation of surface water will ultimately result in extensive ground
Sub- Zone of
surface flow saturation
Level of ground Level of ground
water table water table
Movement of Ground Water
Figure 2: Components of Base Flow
= m .
water depletion, thus decreasing this component of base flow. During long dry periods, streamflow may be sustained solely by ground water seepage. If the ground water table is lowered sufficiently by reduced recharge during wet periods, or by overuse of the resource, streamflow during droughts may cease completely. Notably, subsurface water is an important segment of the streamflow process in an effluent stream (Hillsborough River) and is directly affected by land use changes in the basin.
This brief discussion of the runoff process illustrates the delicate balance of the hydrologic cycle. When urbanization takes place, several factors of this cycle are altered, thus disturbing this natural balance. Students of hydrology have become aware of the implications of rural-urban land use changes. Most texts in the discipline include a general discussion of the effects of land use changes on the hydrologic cycle. Only recently, however, have studies focused on the exact relationship of urbanization and streamflow within specific hydrologic regions. Research
Urbanization is known to have a considerable impact on local hydrologic systems. Much of the literature on this subject has been concerned with the urban effects on climate as well as surface and ground water hydrology. Because of the importance of evapotranspiration and the nature of precipitation on the runoff process, climatic modification or alteration due to urban expansion must
be considered in a study of this type. Landsberg (1956) concluded that increased urbanization has affected the microclimatology of several European regions. His findings from additional investigations which are summarized in Table 4, reveal that there is a significant variance in precipitation, temperature, humidity and wind speed over urban, compared to rural areas.
There is little question that precipitation changes
have a direct effect on runoff. Increases in precipitation will also add to soil moisture, ground water recharge and subsequently to the components of streamflow. However, if the characteristics of total rainfall are changed, producing frequent storms of light intensity and short duration, total runoff may decrease. The portion of water lost to evaporation, infiltration and percolation from a storm of this type is dependent upon the surface characteristics of the basin.
Most of the literature concerning the relationship
between surface characteristics and runoff has been written in the last ten to fifteen years. In 1961, the United States Geological Survey published the first chapter of a five chapter water supply paper entitled, Hydrologic Effects of Urban Growth.* In Chapter A, Savini and Kammerer
*The five chapters of this water supply paper were published under separate covers and are listed under U.S.G.S. Water Supply Papers 1591A-E.
Table 4: Climatic Variance Over Urban Areas
Climatic Component Urban Variance
Annual Mean Winter Min. Relative Humidity Wind Speed
+5 to 10%
+1.0 - 1.50F +2 to 30F
-20 to 30%
Increase in air pollution - particulates and condensation nuclei
Decrease in evapotranspiration
Increased surface friction
Source: Bruce and Clark, 1969; Landsberg, 1961
discussed the general effects urban growth had on the water regimen. They reported that urbanization had a tendency to significantly change local characteristics of evaporation, transpiration, infiltration and ground water recharge. The authors credited these changes to three basic alterations of the natural environment: 1) decreased opportunity for infiltration, 2) reduced vegetation and 3) change in surficial cover.
Although the general effect was characterized by a
decrease in these processes, Savini and Kammerer also reported possible increases in these processes due to urban expansion. In most cases, these increases resulted from the addition of imported water to the basin. This water,
used for household consumption, landscape maintenance, and waste water disposal, acted as an artificial source to the hydrologic system. The increased use of water also tended to deplete the ground water table and increase the quantity, but decrease the quality of streamflow.
Waananen (1969) refers to an Illinois study in which two nearby drainage basins (one rural, one urban) were compared for sustained flow during dry periods. During wet years, streamflow was essentially equal in both situations. In dry periods, however, the urban basin maintained a much higher level of flow than the rural basin. This reaction resulted from two hydrologic differences in the sample areas. The rural basin experienced a greater portion of water loss to evapotranspiration due to the dominance of vegetation. Also, streamflow in the urban basin was influenced by decreased infiltration and the addition of urban waste water to the system.
Additional studies have been concerned with the influence that urban environments have on peak and low flows. Research has shown that in the Salt Creek Basin within the metropolitan area of Chicago, urbanization has increased the intensity, but decreased the duration of peak flows (Spieker, 1970). This hydrologic response has been attributed to an increase in the amount of impervious surface area and storm drainage systems within the basin. The amount of water during low flows has also increased, but for the most part consists of sewage effluent. In an
earlier study of Assunpink Creek in central New Jersey (Miller, 1966), sewage disposal accounted for over 60 percent of streamflow during dry periods.
Urban development, it has been demonstrated, affects the timing and distribution of runoff and peak flows. Assunpink Creek is located in an 89 square mile basin with a considerable urban area found downstream from rural land. The recording gauge on this stream is located downstream from the developed portion of the basin. Therefore, streamflow measurements reflect both the urban and rural response to storm runoff. Figure 3 represents a storm hydrograph developed by Miller in his study of this region. Inspection of the hydrograph shows two distinct peaks caused by a short high intensity storm in 1950. The first peak (530 cfs), which was reached very quickly, but had a short duration, illustrates the hydrologic response to urban runoff. The second peak (270 cfs), which was reached much slower and maintained for a longer period of time, is characteristic of rural hydrologic response and was created from runoff in the upper reaches of the basin.
Seaburn (1969) also used hydrograph analysis as one method of determining the effect of urbanization on the timing and distribution of direct runoff in East Meadow Brook, Long Island. Figure 4 illustrates two one-hour unit hydrographs taken from runoff measurements recorded during two distinct periods of urban development in the study area. For this study, Seaburn used total acres served
Peak Discharge (530 cfs)
Peak Discharge (270 cfs)
Figure 3: Storm hydrograph - Assunpink
Creek (Waananen, 1969; Miller, 1966)
' 5 9 13 17 21 25 29 33
Time in Hours
A. Storm hydrograph, 1939, 10% urban
SI li i l (i iii If L.
1 5 9 13 17 21 25 29 33 Time in Hours
B. Storm hydrograph, 1962, 56.2% urban
Figure 4: Hydrograph Comparison - Two periods of urban development in East Meadow Brook Basin (Seaburn, 1969).
- Peak Discharge
(50 fs )
i 1 1 1 1 1 T-7-T-
by storm sewers as the total urban area of the region. Hydrograph A was developed from a storm in 1939 when only ten percent of the area was sewered. Hydrograph B, however, shows the distribution of direct runoff in 1962 when 56.2 percent of the area was served by storm sewers. Comparison of the hydrographs illustrates that as urbanization increased in East Meadow Brook Basin, peak flows from storm runoff increased in intensity, but decreased in duration.
Because most natural channels are adjusted to accommodate average flow, rapid development of unusually high peak flows increases the possibility of flooding in urban areas. For this reason, the primary concern of many urban hydrologists has been related to the increase in storm and flood flows caused by urbanization (Waananen, 1969). Most of the literature concerned with the effects of urban development on streamflow has reported marked increases in peak flows and reductions in low flows along with their related impact on annual water yields in urban areas (Bigwood and Thomas, 1955). Wilson (1967), however, was not merely concerned with the realization that increased development increased the frequency of river flooding. Rather, he was interested in the quantitative relationship between flood stage and the degree of urbanization within the basin. In his study near Jackson, Mississippi, he found that within a totally urbanized basin, the mean annual flood increased by a factor of 4.5 over a similar
rural situation while the magnitude of a fifty year flood increased only three times. Wilson therefore determined that the effect of urbanization on floods decreased with increasing flood magnitude.
Numerous computer models have been developed to predict what effect changes in specific components of the hydrologic cycle will have on streamflow. The School of Engineering at the University of South Florida has recently completed the first step in construction of such a model. Further additions and refinements, however, are necessary before it will be useful as a regional planning tool. James (1965) used the Stanford Watershed Model in researching the effects of urbanization on Morrison Creek in Sacramento County, California. His test results have shown the model to be reasonably accurate when they were compared to observed data and frequencies.
In almost every case study, current research in urban hydrology has shown increases in direct runoff caused by expanded urban development. It should be remembered, however, that this common result should not be considered the only possible result in all situations. As previously stated, urban modifications to the physical landscape or a river basin may result in decreased or unchanged streamflow. It is the purpose of this study to determine the hydrologic impact of expanded urbanization in the Hillsborough River Basin since 1940.
Scope of the Study
The previous discussion of the population growth in the Tampa Bay Region has concluded that since 1940 there has been significant urban growth in the Hillsborough River Basin. Alterations of the natural environment have been associated with the expansion of urbanized area. The objectives of this study are to: 1) Analyze the physical and hydrological systems of the region, 2) Examine the rate and distribution of urban growth within the region and 3) Determine the net effect of urban development on the hydrologic characteristics of the Hillsborough River System (primarily the component of streamflow). Subregions
Examination of the hydrologic response of the entire
Hillsborough River Basin to increased urbanization has been hampered by three interrelated characteristics of the region. These are: 1) size, 2) percent and location of urban area and 3) tidal influence. The Hillsborough River drains an area in excess of 690 square miles. Precipitation over the basin varies considerably because of its spatial extent and climatic situation. Also, the hydrographic response to storm runoff within most large basins is characterized by low broad "peaks" which at times are difficult to analyze on the unit hydrograph.
Although urban development has increased to a great
extent in the area, the vast majority of land in the basin is still rural in nature. Therefore, it is contended that
consideration of the entire water regime would lead to results common to those of a rural or agricultural basin. The major urban development is found near the mouth of the river and historical expansion due to the urban sprawl of the city of Tampa has proceeded upstream. The total urbanized area of the basin, however, only accounts for a small portion of the total 690 square miles. Initial observation of the distribution of land use categories within the study area is somewhat similar to that of Miller's Assunpink Creek Study in central New Jersey. Viable streamflow data from the urban environment, however, is not available because mcst development has taken place below the Hillsborough River dam where the river is tidally influenced.
Because of these considerations, two subareas within the Tampa Bay Region have been selected for analysis and comparison. These are Blackwater Creek and Alligator Creek near Safety Harbor, Florida. Blackwater Creek was chosen as a rural sub-basin and will be used as a comparative control in the study. Urban development in this 89 square mile basin has been non-existent throughout the period of record. Therefore, streamflow fluctuations should be controlled by variance in precipitation and potentiometric surface, rather than by land use change.
Selection of the second sub-basin was more difficult because of the specific requirements necessary for the study. This area had to be sufficiently small for analysis, possess useful streamflow data for a minimum of twenty years
and have significant urban growth within the period of record. An area meeting all of these requirements and within the boundaries of the Hillsborough River Basin was not available. Alligator Creek was chosen because, in addition to meeting each of these criteria, it was within fifteen miles of the major study area and located in a region similar physiographically to the urbanized portion of the Hillsborough River Basin. Method of Analysis
Precipitation and streamflow data have been collected for both subareas along with statistics for urban development within the Alligator Creek Basin for the period of record. Three basic methods of analysis will be used to determine the relationship between urbanization and streamflow in the Hillsborough River Basin. (Because of their proximity and physical similarity, Alligator Creek will be equated with the urbanized area of the major study region and the results obtained will be expected to be characteristic to both.)
The first method of analysis will involve a two-fold storm hydrograph comparison. First, hydrographs from the urban basin during different periods of development will be analyzed for significant variation in streamflow. Secondly, storm hydrographs for Alligator Creek will be compared to a hydrograph from Blackwater Creek. In each case, peak flows will be examined to determine the rate at
which they occurred along with their relative intensity and duration.
The second method of analysis will include a graphic comparison of the urban and rural streamflow regimes with respect to precipitation. This presentation will involve plotting cumulative precipitation against cumulative stream discharge for both subareas for the period of record. Comparison of the slopes of each plot will determine if any significant variance exists between the rural control and the urban basin.
The last method of analysis to be used in this study includes several basic statistical tests. These processes will not only reveal the existance or non-existence of a relationship, but also the extent to which the tested components of the hydrologic cycle affect streamflow, within a given level of confidence. It is expected that this last method of analysis will further define any relationship determined by the first two procedures.
Although Florida is the second largest state east of the Mississippi River, peninsular Florida exposes less than one half of the continental margin known as the Floridan Plateau. The basement complex of this physiographic feature is composed predominantly of Precambrian crystallines, similar to those of the Appalachian Provinces to the north. The presence of this ancient core, covered by more recent sedimentaries, suggests that the plateau has long been a part of the North American Continent with, however, only recent emergence of the Florida Peninsula (Thornbury, 1967).
In particular locations, sedimentary formations are
found at depths in excess of 18,000 feet, containing rocks which data to Cambrian, Ordivician and Silurian time. Paleozoic rocks of these types, however, are rare and only found at extreme depths in the state. The major portion of the Florida Peninsula is underlain by more recent Cenezoic and Mesozoic sedimentaries. The oldest rock outcrop in the state is the Eocene Avon Park formation. Other limestones of Oligocene, Miocene and Pliocene age outcrop more frequently.
The dominant structural feature of the peninsula is the Ocala Arch. Upwarping of the Tertiary strata began prior to late Eocene time and continued through the Pliocene. Because of the present position of former Pleistocene shorelines, upwarping is believed to have stopped prior to the Pleistocene Epoch (Cooke, 1945).
Deformation of the Tertiary strata during the Ocala uplift has greatly influenced the formation and operation of the Floridan Aquifer. Ocala limestone which is the most important component of the aquifer system, is found 150 feet above sea level in the central part of the state. However, at Miami, 250 miles to the south, the formation is found at 1200 feet below sea level. Movement of groundwater through this formation and other parts of the aquifer has greatly influenced the physical and cultural patterns of present day Southern Florida. At more detailed discussion of the components and influence of the Floridan Aquifer will be found later in this chapter.
Geology of the Study Area
The Hillsborough River Basin is underlain by three major limestone formations of Miocene and Oligocene age. Figure 5 illustrates the spatial distribution of these formations in West Central Florida and lists them in descending chronological order. The oldest series found in the study area is Late Oligocene in age, represented by Suwannee limestone.
GEOLOGY A A Hawthorne-miocen
St. marks-miocen Suwannee-Oligoce x Location of Columns in B Figure 6
x 2 xI
Nearly all of Pasco County is underlain by Suwannee
limestone. Although this formation is covered in the northeastern and southwestern portions of the county by the younger Hawthorne and Tampa limestones,* it continues to dominate the subsurface geology of the upper Hillsborough River Basin. This limestone is commonly yellow to cream in color and is characterized as a soft, granular, thin bedded rock which, in particular areas, is highly dolomitized. Cooke (1945) described the unaltered formation as 91 to 98 percent calcium carbonate (CaCO 3). However, surface exposures of this rock are usually heavily silicified, resulting in the occurrence of Suwannee flint rock. Suwannee limestone is a part of the principal artesian aquifer in the basin and provides an adequate supply of water for local use. There is significant utilization of this formation within the area. Suwannee limestone is used as a source of road metal and aggregate for concrete. Both are economically important to the region.
The northern tributaries of the Hillsborough River
(Cypress Creek, New River and Trout Creek) are influenced by the St. Marks formation. This Early Miocene formation underlies the landscape of Northwest Hillsborough, North Pinellas and Southwest Pasco Counties. The limestone,
*According to the Florida Board of Conservation, Bureau of Geology, the Tampa Stage within the Miocene Series consists of the St. Marks and Chattahoochee formations. Therefore, for the purpose of this study, Tampa limestone will be equated with the St. Marks formation.
commonly referred to as Tampa limestone, appears as a sandy rock, usually light yellow in color. Cooke (1945) described its varying chemical composition as containing approximately 74 percent calcium carbonate (CaCO3) and 24 percent silica (SiO2 ).
Tampa limestone constitutes a major portion of the principal artesian system in this part of West Central Florida. This formation is of little use as a source of lime or aggregate agent when compared to the purer Suwannee or Ocala groups because of its considerable silica content. However, due to its sandy texture, the St. Marks formation is extremely porous and serves as a major source of fresh water for the area. Pinellas County and the City of St. Petersburg derive their entire source of fresh water from well fields which tap this formation.
The porosity of the Tampa limestone has resulted in a system of interior drainage within this portion of the study area. Comparison of the basin location may (Figure 1) with the local geologic map (Figure 5) show that surface water drainage is minimal within the northwestern limits of the Hillsborough Basin underlain by this formation. A further discussion of the effects of the St. Marks formation on the surface water hydrology of the region will be found later in this chapter.
The youngest formation in the study area is the Hawthorne. In addition to forming the bedrock complex of the southern tributaries of the Hillsborough (Blackwater,
Pemberton, and Flint Creeks), this formation also underlies the Alligator Creek subarea. According to the U.S. Geological Survey (Vernon and Puri, 1964), the beds of this formation are composed of phosphoric marine sands, clays and marls and may be described as a yellow sandy limestone. Although phosphate production from this series is important in portions of Hillsborough and Polk Counties, it is of less significance within the limits of the study area.
The numerous clay and marl impurities associated with this limestone have decreased the porosity and ground water yield of the Hawthorne formation. Because of these characteristics, surface water runoff from precipitation responds at a more rapid rate in this area than in portions of the basin underlain by the St. Marks and Suwannee formations.
In Hillsborough County, the Hawthorne is considered a
shallow surface aquifer yielding a low quantity and quality of water. Although the formation is not a totally impervious layer, it does tend to hamper ground water movement to the underlying formations of the Floridan Aouifer. The relationship of the Hawthorne to other strata in the basin may be seen in Figure 6, along with other columnar sections selected in Hillsborough County. The locations of each of these stratigraphic sequences are shown in Figure 5.
C. W. Cooke (1939) described five topographic regions within the State of Florida (Figure 7). The Hillsborough
Figure 6: Columnar Sections - Hillsborough River Basin
Numbers within these columnar sections represent the following strata.
1. Hawthorne Limestone
2. Tampa Limestone
3. Suwannee Limestone
4. The Ocala Group
5. Avon Park Limestone 6. Lake City Limestone
4 5 6
- 200' .
- 600' .
COL. 4 2,3
.WESTERN HIGHLANDS 2.MARIANNA LOWLANDS &TALLAHASSEE HILLS 4.CENTRAL HIGHLANDS CENTRAL LOWLANDS
SOURCE: COOKE, 1 939
River rises in the southern portion of the Central Highlands and flows to the southwest across the Coastal Lowlands to Tampa Bay. The topography of the Central Highlands in this part of West Central Florida is dominated by the irregular rolling hills of the Brooksville Ridge. The Hillsborough breaches this ridge through its oldest and southernmost gap at Zephyrhills (White, 1970).
The division of these two topographic regions is
generally accepted to be found at the 100 foot contour line. Near Tampa, the Coastal Lowlands extend inland for approximately forty miles. The landscape of the basin across this lowland plain is influenced by the presence of four marine terraces which were formed during Pleistocene fluctuations in the sea level. These terraces, as described by Cooke, are the Wicomico Terrace (100 feet), Penholoway Terrace (70 feet), Talbot Terrace (42 feet), and Pamlico Terrace (25 feet). The spatial extent of the Wicomico and Pamlico stands of sea level may be seen in Figure 8.
In addition to fluctuations in sea level, other elements have helped develop the character of Florida's topography. These are: 1) erosion by running water, 2) ground water solution, 3) coastal processes and 4) wind. Within the Hillsborough Basin, however, erosion by running water and ground water solution have been the most active agents shaping the landscape while coastal processes and wind have been more active outside of the study area.
- PAMLICO TERRACE
SOURCE: COOKE, 1939
Because of the character of the soils and underlying bedrock, fluvial erosion has not been as dominant in Florida or the study area as might be expected in other similarly humid regions (Marcus, 1964). Local drainage patterns and textures within the Hillsborough River Basin exemplify this relationship. Comparison of the drainage textures of the southern tributaries with those of the northern tributaries and the northwestern portion of the basin, illustrates this relationship between fluvial erosion and bedrock porosity and permeability.
The landforms and topography in the vicinity of the trunk stream and major tributaries underlain by the Hawthorne formation are similar to Davis' "Old Age" erosional classification. Drainage divides within the area are difficult to locate and extensive floodplains are common to the landscape.
Removed from this region of fluvial topography is an area of interior drainage located in the northern and western parts of the basin. This area is underlain by the more porous Tampa and Suwannee formations and is characterized by a noticeable lack of extensive surface water drainage. Rather, the area is dotted with numerous sinkholes and lime-sink lakes found in conjunction with other features of subsidence and limestone solution.
There will be a further discussion of these topographic differences in the next section concerning physiographic division..
In an explanation of the physical setting of the
Hillsborough River Basin, the physiographic region serves as a base for a systematic method of studying the local characteristics and differences within the study area. The purpose of physiography as a science is explanation rather than pure physical description (Fenneman, 1938). For this reason, detailed analysis of the physical characteristics of the basin will lead to a better understanding of how man's past cultural activity has affected the area and, more important, of what his future influence may be. Before actual delineation and discussion of the physiographic divisions takes place, it seems appropriate to establish a usable regional definition.
The regional concept is used by geographers to study the spatial characteristics and interactions of places on the earth. In this respect, however, the study of place does not restrict the geographer to mere site and situation of specific localities. Rather, it allows him to study larger areas (regions) along with their internal structures and relationship to other regions.
Many regional definitions have been proposed, but with little success for universal acceptance. Because of the diversity of their discipline, geographers have different needs for establishing regional boundaries. For example, Fenneman's Great Plains Region differs considerably from the Great Plains delineated by Baker in 1923 as an
agricultural region (Broek and Webb, 1968). Thus, the formulation of a regional law containing specific criteria for all regions is impractical. Generically, there are as many types of regions as there are reasons for delimiting them. Although most regions may be classified as either formal or functional in nature (Hartshorne, 1959), the possibilities for further definition are numerous.
In a regional definition, specific criteria should be used to satisfy the requirements of the individual. Just as a cartographer uses different pen sizes for visual impact, the regional geographer should use different regional criteria to establish his specific regions. If in geography, the region must be universally defined, the definition should be flexible enough so as to be useful to geographers. Broek and Webb suggested such a definition when they said, "A region, then, is a part of the earth that is alike in specific criteria chosen to delimit it from other regions " (Broek and Webb, 1968, pp. 14, 16). In this way three physiographic regions have been established in the Hillsborough River Basin.
The study area, as defined by the U.S. Geological Survey, includes the land area drained both internally and externally by the Hillsborough River System. Although the area is ultimately drained by the main stream, differences in drainage characteristics have led to the formulation of three separate physiographic subregions within the basin(Figure 9). These areas include; the Limesink
Physiographic Regio Hillsborough River Bas Lime-sink
Lime- s -,sink olk Uplands
Region, the Flatwoods and the Tolk Uplands. Each of these physical divisions are significantly different from each other in topography and local drainage to justify separation. A general comparison of the physiographic characteristics of each is found in Table 5.
The Lime-sink Region
The lime-sink region of the Hillsborough River Basin is characterized by predominantly well-drained soils containing relatively few extensive surface streams. The area is dotted with numerous small sinkhole lakes which serve as reservoirs for the surface water of the area. Most of this physiographic division is typically karst in appearance.
This portion of the river basin is underlain by the Tampa and Suwannee formations. The combination of a relatively porous underpinning covered by a layer of very porous sand has resulted in wide-ranged limestone solution in the region. Although relief is suitable to induce runoff from precipitation throughout most of the area, the porosity of the surface material and substratum have limited the development of an extensive surface water drainage system. Most surface drainage in this division consists of small, sometimes intermittent streams which flow into many of the small karst lakes.
One such stream is Curiosity Creek located in the southwestern portion of the basin. This stream follows
Table 5: General Characteristics of Physiographic Regions
Region Bedrock Relief Dominate Soil
Lime-Sink Tampa and Suwannee Little to Interior Good
Flatwoods Tampa and Suwannee Little Surface Poor to
Limestones with Fair
underlying pan; Hawthorne Limestone
Polk Uplands Hawthorne and Moderate to Surface Fair to
Bone Valley* Hilly Good
*The Bone Valley formation is found throughout much of the Polk Uplands described by White (1970), but is almost totally absent in the study area.
a course of a few miles connecting several sinkhole ponds and ultimately drains into Blue Sink near North Tampa. From this sink, water from Curiosity Creek flows underground until it reappears at the surface at Sulphur Springs and thereby becomes a fresh-water source to the lower Hillsborough River. Examples of this type of drainage are not uncommon in the lime-sink region of the river basin.
The Flatwoods region is the most extensive of the three physical divisions in the Hillsborough River Basin. Most of the main river and its tributaries flow through this area, shaping its landscape. Within the low broad interfluves, surface drainage is relatively poor resulting in numerous swamp and marsh areas throughout. Although the higher portions of the Flatwoods are better drained, the occurrence of low lying areas with standing water is more frequent here than in the other physiographic regions.
This region is underlain by the clayey, marl sediments of the Hawthorne formation and the porous limestones of the Tampa and Suwannee groups. The sedimentaries of the Hawthorne are less porous and less susceptible to subsurface solution than the purer Tampa and Suwannee limestones. Although examples of subsidence are found in the Flatwoods, they are more subtle than those found in the lime-sink region. The most widespread evidence of solution in this area is found with relation to the numerous circular stands
of cypress which dot the higher grasslands of the region. Swamps and poorly drained bottom lands characterize most of the river's course.
Although sands compose much of the surficial coverage of the Flatwoods, the drainage is fairly poor. Many areas of well drained grasslands are found, but they are intermixed with numerous low lying poorly drained patches. Primarily, the poor drainage within this region may be partially explained by the relationship between fairly low elevation and the presence of a high water table. Also, the soils found above the porous Tampa and Suwannee limestones contain a relatively impervious organic pan which decreases the rate and extent of percolation (U.S.D.A., 1958). Therefore, the combination of a high regional water table, low relief and elevation, and the presence of an underlying aquiclude results in the typically poor soil drainage of much of the Flatwoods.
The Polk Uplands
Only a small portion of the Polk Uplands described by White (1970) are found in the Hillsborough River Basin. Although the Bone Valley formation dominates the underlying geology of this physiographic region outside of the study area, it is almost totally absent within the limits of the Hillsborough Basin. Here, similar to the Flatwoods, the relatively impervious clay-marls of the Hawthorne formation are found beneath a surface covering of sand or alluvial soils.
Although both regions are underlain by the same formation, the greater relief of the Central Highlands has increased surface water runoff and soil drainage in the Polk Uplands over that of the Flatwoods. Consequently, there is less evidence of solution of the linestones in the uplands. Solution of the Hawthorne in each of these regions, however, is less extensive than solution in the lime-sink area.
Traditionally, the study of climate has been concerned with the comparison of average temperature and precipitation relationships. At this level, the Koeppen Climatic Classification is suitable for worldwide, continental and large national comparisons. However, its limitations in the microclimatological study of the Hillsborough River Basin are numerous.
Analysis of the climatological data listed in Table
6 results in the subsequent classification of Tampa, Florida as a Cfa climate in the Koeppen system. Translation of these symbols (Table 7) indicates the general range of temperature and distribution of annual rainfall within the region, but reveals little if any information concerning evapotranspiration or precipitation effectiveness. Study of these latter components of climate is critical for research on the relationship between precipitation, urbanization and runoff.
Table 6: Climatological Data for Tampa, Florida
Jan. Feb. Mar. Apr. May June July Aug. Sept. Oct. Nov. Dec.
Temp. F 61.2 62.7 66.0 71.4 76.8 80.6 81.6 82.0 80.5 74.7 66.8 62.3
Precip. In. 2.13 2.84 3.75 2.84 2.85 7.28 8.62 8.24 6.89 2.78 1.46 1.89
Average Annual Temperature = 72.20F; Total Annual Precipitation = 51.57" Source: Bradley, 1972
Table 7: Translation of Koeppen Ofa Climatic Classification
for Tampa, Florida
Symbol Requirements for Classification
C Average Temperature
1) the warmest month is greater than 50OF 2) the coldest month 6s greater than 32 F
but less than 64.4 F
Precipitation not meeting conditions of summer or winter drought (s or w classification respectively)
Average temperatuSe of warmest summer month is greater than 71.6 F
Source: Critchfield, 1966 Table 8: Comparison of Koeppen and Thornthwaite Climatic Types
Wet & Dry Dry
AA'r,BA'r BA'w,CA'w CA'd,DA'd CA'r DA'w EA'd
Cs Cw, Dw,BSk
Mid-latitude -------------------------------AB'r,BC'r BB's,CC's CB'd,DC'd AC'r,CB'r BB'w,DB's CC'd,EB'd BB'r,CC'r BC's,DA'w DB'd,EC'd CB's,DB'w
Source: Hidore, 1969
Most climatic classifications serve as systematic
methods of evaluating particular climatic data. The system which best satisfies the requirements of this study was developed by C. W. Thornthwaite in 1931 with further revision in 1948. A comparison of the Koeppen and Thornthwaite Climatic Types may be found in Table 8. Before Thornthwaite's climatic formulas are applied to the Hillsborough River Basin, it is necessary to discuss the characteristics of temperature and precipitation in the study area.
Temperature and Precipitation
There is little, regional variance in average temperature throughout most of the study area. According to Bradley (1972), average annual temperatures in the basin vary about 0.3 F with Tampa and Plant City reporting 72.20F and 71.90F respectively. Mean monthly temperatures range from 800 to 820 in the summer and 610 to 660 in the winter, giving the region an average reading of about 720F.
The greatest discrepancies within the Tampa Bay region exist between coastal and inland areas and reflect the moderating effects of the Gulf of Mexico upon coastal locations. However, within the inland river basin, the smallest diurnal ranges are related to slope and lake proximity. This fact greatly influences citrus production, but has little effect on the monthly or annual average temperature and thus is not incorporated in Koeppen's System.
Although applications of Koeppen formulas and the 1931 Thornthwaite Classification indicate an equitable annual distribution of rainfall in the Tampa Bay region, there is a notable difference between the relatively dry winters and wet summers. These seasonal fluctuations, however, are not great enough to warrant a Koeppen Cwa Classification for the area.
Unlike the stable climatic component of temperature,
precipitation varies considerably on the local scale throughout most of the Hillsborough Basin. This fact may be explained by the numerous localized thunderstorms which are common to much of central Florida during the summer. Furthermore, land and sea breezes contribute to a thunderstorm pattern which plays a part in local irregularities in precipitation. For this reason, precipitation data from one recording gauge does not present an accurate account of the rainfall characteristics of the region.
An example of this may be seen from the comparison of daily precipitation records of two fire towers in the river basin. The first of these is Hamner Tower (Station 1, Figure 10) located in the western portion of the study area. The second is the Hillsborough River State Park Tower (Station 3, Figure 10) which is located approximately fifteen miles east of Hamner. On July 8, 1959, records indicate that 2.4 inches of precipitation from a thunderstorm fell at Station 1. However, no measurable rainfall was recorded at the state park. It is interesting that the
Thiessen Polygons Hillsborough River Basin
xLocation of recording 4 x precipitation station
monthly record for July, 1959 only varied 0.17 inches between the stations (Hamner, 1.1.15"; Park, 10.98"). For the year, total annual precipitation varied 26.29 inches (Hamner, 107.6"; Park, 81.31"). Although 1959 was a considerably wet year for Florida, Hamner Tower more than doubled the 51.57 inches considered normal for the Tampa area.
With variable precipitation of this magnitude, it is difficult to establish reliable mean values for rainfall over the basin area. In order to obtain valid figures for average monthly and annual precipitation, three stations were selected fcr analysis within the study area. When available, fire tower data was selected over urban or airport recordings in order to achieve a more natural situation for measuring rainfall. Use of a simple arithmetic mean from these stations would improve the validity of the statistic over that of the mean Tampa value. However, the size of the available sample was not sufficient. Although the three stations were scattered across the study area, additional data were necessary to overcome the precipitation variability factor.
Four other stations were selected on the periphery of the watershed (Table 9, Figure 10). Again, a simple arithmetic mean could be computed from the data of these seven stations. However, this method of establishing mean precipitation values would give equal weight to each of the reporting stations, regardless of their position and total
Table 9: Precipitation Data for Seven Recording Stations - Hillsborough RiverfBasin
Sta.# Jan. Feb. Mar. Apr. May June July Aug. Sept. Oct. Nov. Dec. Ann.
1 3.38 4.53 5.94 2.20 3.27 7.93 12.14 12.00 7.53 3.07 1.77 2.81 66.94
2 3.52 3.72 4.77' 2.09 4.06 10.10 9.08 9.53 6.58 3.14 1.44 2.12 60.11
3 3.36 4.39 5.40 2.04 '4.45 10.02 10.24 11.21 8.00 2.88 1.84 2.72 66.26
4 2.30 2.90 4.10 3.20 4.20 8.30 8.80 7.90 7.10 3.40 2.00 2.30 56.30
5 2.13 2.84 3.75 .2.84 2.85 7.28 8.62 8.24 6.89 2.78 1.46 1.89 51.57
6 2.20 2.60 3.90 3.00 4.30 7.30 8.00 7.30 6.30 2.70 1.70 2.00 51.20
7 2.40 2.90 4.00 3.30 3.90 7.80 9.70 8.60 7.00 3.30 2.10 2.60 57.50
Sources: Stations 1-3: Fire Tower Records provided by Dr. Dewey Stowers, U.S.F.
Station 5: Bradley, 1972
Stations 4,6,7: Florida Board of Conservation, 1966
Note: It should be noted at this point that the fire tower statistics appear to be significantly higher than the other stations recorded. This fact may be partially attributed to the period of record used for computation of averages. Fire tower data was calculated over an eleven year period of record (1958-1968) while values for the other stations used a fifty year period. Although tower data increased the ultimate mean value considerably, it should be remembered that it was recorded d-aring the period of record for this study. For this reason, its inclusion and influence are extremely important.
influence on the basin. A method developed by A. H. Thiessen in 1911 (Wisler and Brater, 1967) was used in order to compute the most representative precipitation averages for the study area.
The Thiessen polygons (Figure 10) are constructed
so that the area within each figure is closer to the enclosed station than to any of the surrounding stations. Therefore, figures from each recording gauge are considered to be the most representative of that area's precipitation. The basin average is calculated according to the formula
Pb P1A1 + P2A2 +...+ PnAn
where P is equal to the average precipitation of station 1 (etc.), A1 is equal to the area of the respective polygon and T is equal to the total basin area. By using this method, the resulting basin average (Pb) is weighted proportionally to the area represented by each station.
The areas of the Thiessen polygons listed in Table 10 were determined by use of a simple dot grid. These values along with the mean monthly total for each station were used to determine the monthly Thiessen mean. These values were then totaled to determine the average annual precipitation for the basin.
Table 11 lists the several possible sets of precipitation data available for use. Inspection of this data reveals a maximum difference of 11.78 inches between these averages.
Table 10: Areas of Thiessen Polygons
Station Name Dot Grid % of
1. Hamner Fire Tower 35 15.4
2. Valrico Fire Tower 20 8.8
3. State Park Tower 69 30.4
4. St. Leo 52 22.9
5. Tampa 7 3.1
6. Lakeland 35 15.4
7. Brooksville 9 4.0
Total 227 100.0
This significant variance mandates separation of one set of data which appears to be most representative of the basin's actual mean precipitation. Because the Thiessen method provides a proportional average of numerous stations, it is considered the most accurate representation of precipitation in the Hillsborough River Basin. Therefore, this set of climatological data accompanied by mean temperature values for Tampa will be used in application of the Thornthwaite models.
Climate According to Thornthwaite
In an article published in the Geographic Review of Oct ., 1931, C. Warren Thornthwaite introduced a "new" system for analyzing the climates of North America. This system was expanded in 1933 as a basic world wide classification of climate. Similar to Koeppen's previous system,
Table 11: Comparison of Mean Precipitation Values
Jan. Feb. Mar. Apr. May June July Aug. Sept. O'ct. Nov. Dec. Ann. Tampa 2.13 2.84 3.75 2.84 Z.85 7.28 8.62 8.24 6.89 2.78 1.46 1.89 51.57
Basin 3.01 3.94 5.15 '2.48 4.12 8.75 10.39 10.37 7.54 3.12 .1.87 2.61 63.35
Arith.Mean 2.76 3.41 4.55 2.67 3.93 8.39 9.51 9.25 7.06 3.04 1.76 2.35 58.68
Thiessen 2.88 3.63 4.79 2.56 4.15 8.72 9.73 9.61 7.26 3.04 1.81 2.44 60.62
Thornthwaite established climatic formulas, derived climatic boundaries through emperical analysis of vegetation, soil and drainage types and thus, developed a tri-symbol system for climatic classification (Critchfield, 1966). His primary deviation from the Koeppen system was found in the quantitative derivation of two new climatic components; precipitation effectiveness and temperature efficiency. The original 1931 classification is summerized in Table 12.
Thornthwaite's reasoning for a new system evolved
from the need of a more detailed comparative analysis of the temperature-precipitation relationship than the Koeppen System supplied. Koeppen suggested the idea of precipitation effectiveness by differentiating the seasonal distribution of rainfall (f, s, and w). Thornthwaite's classification, however, established quantitative limits for a P-E Index which more clearly described effective precipitation with relation to temperature.
According to the 1931 Thornthwaite boundaries, the Hillsborough River Basin is classified as a BB'r to CB'r climatic type. Translation of these symbols describes the area as a humid to subhumid mesothermal climate with no drought period. Simple comparison of the Koeppen and Thornthwaite translations reveals little difference in description. However, recognition of the P-E and T-E Indices with the latter classification, increases its value for analysis of the area's water budget.
Table 12: 1931 Thornthwaite Classification of Climate Precipitation Effectiveness
P-E Index I = sum of twelve monthly values of 115(P/T
-10)10/9 where P = mean monthly precipitation in inches and T = mean monthly temperature in 0F.
Humidity Province Vegetation P-E Index
A Wet Rain Forest 128 and above
B Humid Forest 64 to 127
C Subhumid Grassland 32 to 63
D Semiarid Steppe 16 to 31
E Arid Desert less than 16
T-E Index I' = sum of twelve monthly values of T-32/4
where T = mean monthly temperature in 0F.
Temperature Province T-E Index
A' Tropical 128 and above
B' Mesothermal 64 to 127
C' Microthermal 32 to 63
D' Taiga 16 to 31
E' Tundra 1 to 15
F' Frost 0
Seasonal Distribution of Precipitation
r Rainfall adequate in all seasons
8 Rainfall deficient in summer w Rainfall deficient in winter
d Rainfall deficient in all seasons
Source: Thornthwaite, 1931
In 1948, Thornthwaite modified this 1931 system
significantly. Cursory inspection of the revised classification merely reveals the addition of a fourth criterion and a finer breakdown of precipitation effectiveness and thermal efficiency limits. However, a more detailed comparison of the two, shows that the modified climatic divisions are not distinguished by precipitation and temperature data alone. Rather, the component of potential evapotranspiration has been added. With this added dimension, the role of vegetation in climatic evaluation was changed. In the 1931 system, Koeppen's consideration of vegetation as an expression of climate was used, whereas, in the revised classification, vegetation was considered a mechanism of moisture transfer from the soil to the atmosphere (Thornthwaite, 1948).
The importance of evapotranspiration as a process of water transfer in the hydrologic cycle has been recognized by geographers and hydrologists. Potential evapotranspiration is the maximum amount of water possibly transferred from a given region over a particular period of time. It is a function of temperature, number of daylight hours, vegetation, and soil characteristics. Consideration of this climatic component is necessary for the study of the microclimatology and water balance of the Hillsborough River Basin because of its influence on atmospheric moisture and subsurface water retention.
Potential evapotranspiration may be considered the amount of water necessary to satis-Cy the requirements of temperature related evaporation, and vegetation related transpiration within a particular area. When this water need is satisfied by local precipitation and soil moisture supply, potential evapotranspiration will equal actual evapotranspiration for the region. When available water from these two sources is in excess of water demand, there will be a water surplus to runoff or recharge the ground water table. However, when local need surpasses local supply, there is a water deficit for the region and drought conditions exist.
The importance of soil moisture is evident when considering the occurrence of water surplus or water deficiency. The maximum amount of water possibly stored in the soil at a given time (soil moisture capacity) and available for vegetative use is dependent upon two factors. These are; 1) porosity of the soil and 2) the average depth of the vegetative root zone. Thornthwaite and Mather (1957) developed a set of tables from which maximum soil moisture storage can be determined for various vegetation and soil combinations. Since neither soil nor vegetation are uniform across the watershed, no single association could be used to determine the soil moisture capacity of the study area. Therefore several combinations were averaged to determine a mean storage capacity of five inches for
the Hillsborough River Basin. This figure represents an average for the region which will be used in determining the moisture characteristics of the study area.
Table 13 lists the data necessary to implement the revised Thornthwaite system of climatic classification. The basic temperature and precipitation statistics used were explained in the previous section. Potential evapotranspiration figures were calculated according to the method described by Thornthwaite in his 1948 modification. Computation of these figures along with the necessary graphs and tables may be found in Appendix I.
Calculation of the moisture data for the classification of climate began in the March column. This was selected as the starting point because of the sufficiently wet period that preceded it. During March, soil moisture storage was assumed to be at its maximum of five inches. For that month, precipitation exceeded the water need and soil moisture was at full capacity. Therefore, there was a water surplus of 2.36 inches for the month. Similar inspections were made for the remainder of the year. Whenever precipitation was less than potential evapotranspiration, there was a net loss recorded in the storage change row. This amount was then subtracted from soil moisture storage figure and the remainder was entered as available soil moisture for the following month. Since there was no period when water need was not satisfied, there was no water
Table 13: Monthly Water Balance Table - Hillsborough River Basin Jan. Feb. Mar. Apr. May June July Aug. Sept. Oct. Nov. Dec. Ann.
Temperature OF Precipitation (In) I = sum of i's Unadjusted PE Adjusted PE Precip. - PE Acc.Pot.Wat. Loss Storage Storage Change Actual Evapotrans. Deficit Surplus Runoff
G'water Storage Base Flow Surface Runoff
6.41 1.77 1.56 2.07
4.49 5.00 +0.51
1.04 0.65 1.05 0.53
2.34 1.17 0.19
66.0 71.4 4.79 2.56 7.48 9.35 2.36 3.38 2.43 3.62 2.36. -1.06
2.36 1.86 2.95
0.74 0.37 0.56
0.19 0.10 0.37
80.6 8.72 12.85 5.50 6.38
0.23 0.05 0.23 0.00.
5.69 6.71 3.02
4.91 +0.09 6.71
82.0 80.5 74.7
9.61 7.26 3.04 13.41 12.81 10.56 5.76 5.48 4.09 6.51 5.64 4.01 3.10 1.62 -0.97
5.00 5.00 5.00
6.51 5.64 4.01
2.34 3.87 1.93
1.62 1.98 2.59 1.29
0.99 0.65 0.33 0.66
66.8 1.81 7.75 2.52 2.27
0.49 0.16 0.08
2.44 6.22 1.69 1.52 0.92
3.57 +0.92 1.52
0.24 0.04 0.02 0.22
72.20 60.62" 117.39
deficiency for the year. Thus, actual evapotranspiration always equaled the potential figure.
Inspection of Table 13 and Figure 11 points out two periods when water from the soil moisture reservoir is necessary to prevent drought conditions. Maximum water depletion from this source occurs in May with a total loss of 2.43 inches through evapotranspiration. Using these figures, Thornthwaite's formulas will be applied to determine the climatic classification of the study area.
The first component of the classification is that of
the Moisture Index (I m). It is determined by the relationship between the relative humidity and aridity of the overall climatic situation. Thornthwaite first established the indices of humidity and aridity as percentages. Their formulas are:
Ih = lOOs/n and Ia = 100d/n
where Ih and Ia are the indices of humidity and aridity respectively, s is the water surplus, d is the water deficiency and n is water need (potential evapotranspiration). It is known that water surpluses and deficiencies occur in many areas during different parts of the year. To some extent, a water surplus in one season can partially compensate for or decrease the overall effect of a water deficit during another season through the recharge of soil moisture. Therefore, in the moisture index, the index of humidity has greater weight than the index of aridity.
ED Soil Moisture
-~~ ~ ~ ~ O~~J ~~ IU ru re a e
s orL soi moiLsure rechargar
restricted by a five inch soil moisture capacity.
J F NM
A M i S
A M J J A S
I i I I
0 N D J
Figure 11: Thornthwaite Comparison of Potential Evapotranspiration and Precipitation - Hillsborough River Basin
Note: Limit f
Thornthwaite calculated that in the total moisture index, the Index of Aridity had only six-tenths the impact of the Index of Humidity. Therefore, the Moisture Index was stated as follows:
100s - 60d
When the appropriate values for the study area are substituted in this formula, the results are: 100 (13.12) - 60 (0)
Im = 1312 / 47.5
Im = 27.62
When compared to the limits listed in Table 14, the Hillsborough River Basin is said to be a B1 (humid) climatic type.
The second component of the system is the Index of
Thermal Efficiency. Because potential evapotranspiration
(PE) expresses precipitation effectiveness with relationship to temperature and length of day, Thornthwaite believed that this figure would best represent the climatic characteristic of thermal efficiency. The potential evapotranspiration calculated for the study area is 47.5 inches per year. When compared to the limits established in Table 15, the thermal efficiency of the region is classified as an A' (Megathermal) climate.
Table 14: Limits of Thornthwaite's Moisture Index
Climatic Type A Perhumid B4 Humid B3 Hurid B2 Humid B1 Humid C2 Moist sub C1 Dry subhu D Semiarid E Arid
Moisture Index 100 and above 80 to 100 60 to 80 40 to 60 20 to 40 0 to 20
-20 to 0
-40 to -20
-60 to -40
Source: Thornthwaite, 1948
Table 15: Limits of Thornthwaite's Index of Thermal Efficiency
TE Index (PE)
cm. in. Climatic Type
14.2 5.61 28.5 11.22 42.7 16.83
57.0 22.44 --71.2 28.05 --85.5 33.66 --99.7 39.27 --114.0 44.88 --Source: Thornthwaite, 1948
------------ DT--------.-------------- gT----------------------- CTI Microthermal
------------ 2-------------------- BT------------N2 Mesothermal
--------.T i----- --ma
Thornthwaite's third criterion is that of seasonal
variation of effective moisture. Although this component seems to have appeared in his former classification, present derivation of this climatic boundary is based on the indices of humidity and aridity which are functions of water need. In moist climates, the period.and magnitude of I determine this segment of the classification. Because the study area has no seasonal water deficiency, Ia = 0 and the symbol r results (Table 16).
Thornthwaite's last climatic component in his 1948
classification concerns the summer concentration of thermal efficiency. The symbol representing this characteristic is determined by calculating the percentage of the total annual potential evapotranspiration that occurs in the summer quarter. Values listed in Table 17 represent the boundaries established for each of these climatic symbols. The total PE for the study area is 47.5 inches. Of this total, 25.24 inches or 53.1 percent occurs in the summer. Thus, the final symbol for the regional classification is b' .
According to Thornthwaite's 1948 classification, the Hillsborough River Basin is established as a B1A'rb'3 climatic type. Mere verbal translation of these symbols is less significant to the hydrologist than the quantitative limits represented by each. Further use and explanation
of the data listed in Table 13 will take place in the discussion of the regional water budget later in this chapter.
Table 16: Seasonal Variation of Effective Mfoisture
Moist Climates (A, B, C 2)
r little or no water deficiency 0 - 16.7
s moderate summer water deficiency 16.7 - 33.3
w moderate winter water deficiency 16.7 - 33.3
s2 large summer water deficiency 33.3 +
w2 large winter water deficiency 33.3+
Dry Climates (C1, D, E) Humidity Index
d little or no water surplus 0 - 10
s moderate winter water surplus 10 - 20
w moderate summer water surplus 10 - 20
s, large winter water surplus 20 +
w2 large summer water surplus 20 +
Source: Thornthwaite, 1948
Vegetation and Soils
The previous discussion of Thornthwaite's climatic classification and its methods of determining moisture data point out the important relationships between soil, vegetation and the hydrologic characteristics of the Hillsborough River Basin. The interrelationships of precipitation, storage, ground water recharge, and runoff have already been mentioned. However, it should be noted once more, that soil moisture capacity is solely determined by the soil-vegetation association present in the region (Thornthwaite and Mather, 1957). It is beyond the scope
Table 17: Limits for Summer Concentration of Thermal
Total Potential Evapotranspiratior
39.27 ----- 99.7
33.66 ----- 85.5
28.05 ----- 71.2
22.44 ----- 57.0
16.83 ----- 42.7
11.22 ----- 28.5
5.61 ----- 14.2
------------- 48.0 -----b'
------------- 51.9 -------------- A-----b' 2
------------- 56.3 -------------------------------- 61.6 -------------- -----b'-
------------- 68.0 --------------1--------------- 76.3 -------------- ------------------ 88.0 -------------1-----d'
of this study to discuss at great length the numerous botanical and pedologic characteristics of the study area. Rather, a general description of the vegetative and soil types will be given so as to clarify the soil moisture relationships and illustrate the overall features of the landscape.
Of the thirteen native vegetation types listed by
Davis, three are present in the Hillsborough River Basin (Marcus, 1964). These are: 1) Pine Flatwoods, 2) Upland Pine and Scrub Oak Open Forests, and 3) Swamp Forests.
The Pine Flatwoods cover a major portion of the
watershed. The name and distribution of this vegetation
type coincide with that of the Flatwoods phyciographic region established previously in this chapter. This similarity in terms, however, should not confuse the reader as to the validity of the Flatwoods physiographic division. Rather, the term flatwoods refers not only to a vegetative type, but also to an area of particular soil, topography, and drainage type; the specific criteria used for delineation and separation of the physical area.
The vegetative region resembles an open pine woodland with many varieties of short grasses and palmettoes acting as ground cover. On the periphery of these forests, prairies of saw-palmettoes, short grasses and widely scattered longleaf pines are not unusual. Upon initial observation, the observer may mistake this association for a savannah type vegetation. However, as Davis (1943) pointed out, this comparison is incorrect because tall grasses dominate the vegetation of the tropical savannah.
In addition to open forests and palmetto prairies,
there are many areas of poorly drained sands which support intermixed stands of bald cypress, water oak and hickory,
red maple and cabbage pa.lm. Throughout much of the study area covered by this thick forest, island hammocks of deciduous trees are common (Marcus, 1964).
Today, man has significantly changed the character of the natural vegetation of the flatwoods. Within the study area, many stands of pine have been completely removed for lumber and saw-palmetto prairies have been replaced
by pastures. This fact was considered when the soil moisture capacity figure was established for the basin. Upland Pine and Scrub Oak Forests
These forests are naturally found within the Central Highlands and Lime-sink regions of the Hilisborough River Basin, associated with well to moderately well drained sands. The high pine forests are usually open stands of long-leaf pine intermixed with Turkey and Blue-jack Oaks. The normal ground cover in these areas consists of short grasses.
The scrub forest is dominated by the presence of
several types of scrub oaks. Unlike the upland pine areas, this forest association consists of many low trees and shrubs intermixed with sand pines. Because of the thick tree cover, few grasses are found within this type of forest.
These forests are native to the Polk Uplands and Limesink physiographic regions of the study area. Today, much of the land covered by these forests is planted in citrus. The well drained characteristics of the associated soils and rolling topography make these areas ideal for citrus production. When calculating the soil moisture capacity for the study area, the Thornthwaite combination of "orchards on fine sand" was used to represent this area. Swamp Forests
The third and least extensive type of vegetation found in the Hillsborough River Basin is swamp forest land. For
the most part, these forests are restricted to the poorly
drained bottom lands found lining the course of the main river and major tributaries. These dense forests are mostly composed of several varieties of water oaks, red maple, cabbage palm and bald cypress. Because these areas are under water for much of the year, there has been little modification to their natural distribution. Table 18 illustrates the vegetation-soil associations found in the study area.
According to the U.S. Department of Agriculture (1958), there are four broad classes of soils present in the Hillsborough River Basin. These are: 1) Somewhat poorly drained sands with organic pan (Leon and Immokalee), 2) Bottom lands, swamps and ponds (undifferentiated), 3) Somewhat poorly drained sands over calcareous substratum (Ruskin, Sunniland and Adamsville), and 4) Well drained deep sands (Blanton, Lakeland and Eustis). The general description of the soil profiles for each soil type may be found in Table 19.
The soils of the flatwoods predominantly consist of those types listed in the first three classifications. Because of the water content and dense forest cover of the bottom land and swamp soils, detailed classification of these types has been difficult. For this reason, description of their profiles is not included in Table 19.
Table 18: Associated Vegetation
General Type of Vegetation
1. Pine flatwoods with sawpalmetto and wire grasses common
2. Scrub vegetation composed
of sand pine, scrub oaks and sam-palmetto with few grasses
3. Sandhill or high pine forests with oaks common. Some with scrubby tree growth, grasses and sawpalmetto
4. Swamps dominated by cypress trees and cypress heads
and Soil Types - Hillsborough
Soil Types Generally Associated with this Type of Vegetation
1. Mostly ground water podzols as Leon and Immokalee
2. Characteristically deep, dry well drained sands such as Eustis, Lakeland, St. Lucie and Lakewood fine sands
3. Blanton fine sand and in adjacent areas Norfolk fine sand, all well drained
4. Plummer fine sands and other unclassified soils (alluvial bottom land and swamp soils)
Source: Davis, 1943
The better drainage of the Lime-sink and Polk Uplands regions is partially explained by soil type. Well drained deep sands of the Blanton, Lakeland, and Eustis varieties dominate the character of the lime-sink area. Patches of poorly drained alluvial soil can be found in the region, but their occurrence is relatively rare when compared to that of the well drained sands.
The soils of the Polk Uplands are similar to those of
the flatwoods. However, the increased elevations and relief associated with a somewhat lower water table has resulted
- mi s'
Description of Soil Profiles - Hillsborough River Basin
Soil Type Depth (In.) Description
Leon fine sand Immokalee fine
Ruskin fine sand
0-5 Dark gray nearly lose fine
sand salt-and-pepper appearance due to small amount of
5-20 Light gray loose fine sand. 20-24 Very dark grayish-brown fine
sand; cemented with organic
24-30 Dark brown fine sand, weakly
30-42+ Yellowish-brown loose fine
sand in upper parts grading
to lighter colors with depth.
0-6 Dark gray loose fine sand
with salt-and-pepper appearance.
6-12 Gray loose fine sand. 12-32 Light gray loose fine sand
with few light brownish gray
32-38 Very dark brown fine sand with
slightly cemented organic pan. 38-42+ Dark grayish brown loose fine
sand grading lighter with
0-6 Dark gray nearly loose fine
sand with small to moderate amount of organic material. 6-24 Light gray loose fine sand.
24-36 Brownish-yellow fine sandy
clay loam mottled with light
36-42+ White and pale yellow shell
0-4 Dark gray nearly loose fine
sand with small amount of
organic matter, strongly acid. 4-20 Light-gray loose fine sand with
yellow streaks in lower part,
20-40 Mottled yellowish-brown strong
brown to light gray fine sandy
clay to fine sandy clay loam,
strongly acid to neutral to
mild alkaline with depth.
Table 19 - Continued
Soil Type Depth (In.) Description
Blanton fine sand Lakeland fine
Eustis fine sand
20-40 Mottled yellowish-brown strongbrown to light gray fine sandy
clay or fine sandy clay loam,
strongly acid to neutral to
mild alkaline with depth. 40-48+ White marl, sandy clay to
sandy clay loam texture, few
0-4 Dark gray to gray loose fine
4-12 Light gray loose fine sand.
12-24 Light yellowish-brown loose
24-34 Yellowish-brown loose fine
34-42+ Brownish-yellow loose fine
0-6 Dark gray to gray nearly loose
6-18 Grayish-brown to light brownish-gray loose fine sand.
18-42+ Very pale brown or light gray
loose fine sand, splotched, with pale yellow or yellow.
0-5 Dark gray loose fine sand;
low in organic matter.
5-12 Grayish-brown loose fine sand.
12-30 Yellowish-brown loose fine
30-48+ Brownish-yellow loose fine
0-6 Dark grayish-brown to grayishbrown loose fine sand; contains a small amount of organic matter.
6-12 Yellowish-brown to brownishyellow loose fine sand
12-48+ Strong brown to reddishyellow or yellowish-red loose
Source: U.S.D.A., 1958
in better surface drainage for this region. In addition, the well drained deep sands are less common to this area of the Central Highlands than they are in the Flatwoods.
Most of the aspects of the surface water hydrology of the Hillsborough River Basin, have been discussed with some detail previously in this paper. For this reason, further discussion of the hydrologic character of the study area will include the general characteristics of the groundwater and artesian system. Also, additional analysis of the monthly water balance (Table 13) will be considered. Ground Water and the Artesian System
Because of the nature of the soils and underlying formations, much of the precipitation that falls on the Hillsborough River Basin goes directly to ground water recharge. The aquifer system underlying the study area consists of three individual aquifers of differing characteristics and depths (Briley, Wild and Associates, 1970). These are in descending order; the water table aquifer, the shallow artesian aquifer and the principal Floridan Aquifer.
The water table aquifer consists of undifferentiated sands, clays and marls of Pleistocene and Pliocene age and is found to depths of 150 feet. Although the clay and marl materials do not yield significant amounts of water, the sands have been known to discharge up to 200 gpm. (Briley et al., 1970). The level of the water table is approximately
ten feet from the surface and follows the topography of the area in a subdued manner. In the Flatwoods, the fairly high water table associated with typically low relief and elevations, results in generally poor soil drainage.
The shallow artesian aquifer is predominantly composed of the Hawthorne formation and is found in some places from the surface to depths of 250 feet. This Miocene formation consists of sands, clays and some porous limestones which yield up to 200 gpm. The significant clay impurities associated with the Hawthorne act as a subsurface aquiclude in many areas of the river basin.
The principal Floridan Aquifer in the study area is made up of numerous differentiated layers of limestone from Miocene to Eocene age. The younger strata of this system are the Tampa and Suwannee limestones which were described previously. Together, they range in depth from 80 to 400 feet. These rock layers yield water up to the rate of 1000 gpm and supply most of the commercial and domestic wells of the area.
Beneath the Oligocene Suwannee limestone lies the
Eocene formations of the Ocala group. This group is composed of the Crystal River formation, the Williston formation and Inglis limestone which are found at varying depths from 90 to 300 feet. The rocks of the Ocala group may be described as yellow-gray and brown soft, almost pure limestones. The Crystal River and Williston formations are rarely used as a source of water because of their low
transmissibility. The Inglis limestone, however, in association with the Avon Park and Lake City limestones supply a major source of ground water for the region.
The Eocene Avon Park and Lake City limestones are found at depths of from 200 to 500 feet. They are soft, chalky and cream to brown in color, containing zones of coquina and crystalline dolomite limestone. Together with the Inglis limestone of the Ocala group, these layers generally yield over 500 gpm and have been known to exceed yields of over 5000 gpm-in some wells. The deeper limestones of the Floridan Aquifer are not presently used as a source of water in the Hillsborough River Basin. However, the layers of the Eocene Oldsmar limestone at depths in excess of 500 feet do provide a potential source of fresh water for the region (Briley et al., 1970).
Recharge to the surface and shallow aquifers of the basin usually occurs after precipitation. Replenishment of the deeper artesian system, however, is more complex. Recharge at greater depths is hampered by the overlying confining beds of the Hawthorne and, in some instances, the Crystal River and Williston formations. Some percolation of water through these beds does occur, but major recharge to these layers is from surface waterz filtering down through sinkholes and other conduits formed from solution. Reference will be made to this relationship when considering methods of decreasing regional water loss from the hydrologic system.
The Water Balance
Explanation of Thornthwaite's monthly water balance system (Table 13) was begun during the regional classification of climate according to his 1948 method. The theory of the water balance was first introduced by Thornthwaite in 1944. It has been used in this paper as a systematic method of describing the climatic and physical relationships occurring within the study area. Although refinement of methods for calculating certain components of the system is necessary (eg. potential evapotranspiration), this system provides a satisfactory framework for comparison. The figures listed in Table 13 represent theoretical averages and are not meant to portray the exact relationship for each water year. Rather, they have been constructed to demonstrate the possible relationships of the components of the water budget within the study area.
The amount of water available for runoff in any given month is proportional to the water surplus for that month. In watersheds larger than 100 square miles, fifty percent of available moisture surplus runs off (Thornthwaite, 1948). This percentage may be decreased in a smaller basin area. The remaining half of the moisture surplus is detained in the basin until the next month when it is combined with that month's surplus to equal the total available water for runoff. Again, fifty percent of that figure is entered as runoff for the respective month. Using this method
runoff figures were calculated for the study area beginning in July; the first month of water surplus after soil moisture demand was satisfied.
The runoff figures listed in the water balance table represent the total runoff for the study area. Included in these figures are the monthly totals of surface runoff (SRO) and base flow (BF), the two components of runoff. Much of the water surplus in the Hillsborough River Basin goes directly to ground water storage (GWS) and enters the streams as base flow. Base flow is the amount of water contributed to streamflow from the ground water table. Therefore, this component of runoff is directly proportional to the amount of ground water held in ground water storage. The amount of water held in this fashion for any given month, however, is partially determined by the amount of water lost to base flow in the previous month.
Again, it is assumed that in large watersheds, fifty percent of the total ground water storage is released to streamflow and the remaining fifty percent available is held over into the next month. Therefore, ground water storage for any month is equal to the carry-over from the previous month (equal to the base flow from the previous month) plus the present recharge due to soil moisture surplus. The figures for ground water storage were calculated according to the formula:
GWS = 0.5 BFp + S
where BFp is equal to base flow for the previous month and S is the current moisture surplus. Base flow for any given month is a function of the amount of water held in storage. Therefore:
BF = 0.5 GWS
Because these components are dependent upon each other, an arbitrary starting point with an assumed value for ground water storage must be selected for their computation. The month of July was chosen as the point of origin because it was the month after the longest period without ground water recharge. Therefore, during this month, ground water storage was assumed to be equal to the moisture surplus of 2.93 inches. Calculation according to the above formulas was carried out for several annual cycles; each additional cycle was believed to more closely approximate the real value.
The surface runoff figures in the table were determined by subtracting the base flow amounts from the total runoff
(RO) for each month.
SRO = RO - BF
Therefore, in each month, total runoff is equal to combined base flow and surface water runoff. Also, inspection of the annual regional runoff and surplus figures demonstrates that all water in excess of regional demand is accounted for and the system is in balance.