- Collection:
- Georgia Government Publications
- Title:
- Hydrology and model evaluation of the principal artesian aquifer, Dougherty Plain, southwest Georgia
- Creator:
- Hayes, Larry R
- Contributor to Resource:
- Maslia, Morris L
Meeks, Wanda C
Geological Survey (U.S.) - Publisher:
- Atlanta, Ga. : Georgia Geologic Survey
- Date of Original:
- 1983
- Subject:
- Groundwater--Georgia
Water quality--Georgia - Location:
- United States, Georgia, 32.75042, -83.50018
- Medium:
- state government records
- Type:
- Text
- Format:
- application/pdf
- Description:
- Includes bibliographical references (p. 91-93)
- External Identifiers:
- OCLC 10740397
NZ MMS ID 9910374837802931
Call Number N200.G3 B1 no. 97
Call Number GB1199.3.G4 - Metadata URL:
- https://dlg.galileo.usg.edu/id:dlg_ggpd_s-ga-bn200-pg3-bb1-bno-p-b97
- Digital Object URL:
- https://dlg.galileo.usg.edu/do:dlg_ggpd_s-ga-bn200-pg3-bb1-bno-p-b97
- Language:
- eng
- Extent:
- ix, 93 p. : ill., maps (some col.) ; 28 cm.
- Holding Institution:
- University of Georgia. Map and Government Information Library
- Rights:
-
HYDROLOGY AND MODEL EVALUATION
OF THE PRINCIPAL ARTESIAN AQUIFER,
DOUGHERTY PLAIN, SOUTHWEST GEORGIA
by
Larry R. Bayes, Morris L. Maslia, and Wanda C. Meeks
NORTH
Potentiometric surface of the principal artesian aquifer
SOUTH
R.esiduum
leak age)
Semiconfined aquifer Regional flow
Lisbon confining bed
(no vertical leakage)
Department of Natural Resources Environmental Protection Division Georgia Geologic Survey
Prepared in cooperation with the U.S. Geological Survey
BULLETIN 97
HYDROLOGY AND MODEL EVALUATION OF THE PRINCIPAL ARTESIAN AQUIFER,
,
DOUGHERTY PLAIN, SOUTHWEST GEORGIA By
Larry R. Hayes, Morris L. Maslia, and \Ianda C. Meeks
Prepared in cooperation with the u.s. Geological Survey
DEPARTMENT OF NATURAL RESOURCES Joe D. Tanner, Commissioner
ENVIRONMENTAL PROTECTION DIVISION J. Leonard Ledbetter, Director GEORGIA GEOLOGIC SURVEY
William H. tkLemore, State Geologist
Atlanta, Georgia 1983
CONTENTS
Abstract Introduction
Previous investigations
Purpose and scope Data collection and methods
Well and surface-water station numbering systems Test-well drilling Sources and use of hydrologic data Acknowledgments Geography
Geology Residuum Ocala Limestone Lisbon Formation
. ........... ............ ... . ...... .. .... . .. . . . The hydrologic system Rainfall Surface water Drainage description Stream low Flow duration
Low-flow frequencY Average runoff
Base flow Ground water
Residuum Hydraulic properties Water levels
Principal artesian aquifer
..................................... Hydraulic properties
Water levels Lis bon Formation Recharge, discharge, and flow characteristics Water budget
Ground-water qualitY Pesticides
Ground-water flow model Model de script ion
System concepts Ground-water flow analysis Finite-difference grid and boundary conditions Data requirements Hydraulic properties Initial conditions
Model calibration
Calibration procedures November 1979 steady-state simulation May~November 1980 transient simulation
Page 1 1 3 3 3 3 4 4
7
7
7 10 10
10 10 10 16 16 20 20 26 29 29 34 34 34 41 41 41
47 51 51 56
57
58 58 58
60
64 64 66
66
68 68 70 70
75
iii
CONTENTS Page
Simulated effects of pumpage during a hypothetical drought and during normal recharge conditions................................ 77 Effects of irrigation pumpage during a hypothetical 3-year drought..... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 81 Pumpage of 113 billion gallons per year................... 81 Pumpage of 408 billion gallons per year................... 86 Effects of pumping 287 billion gallons per year with normal recharge.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 86
Summary and conclusions.................................................. 86 Selected references...................................................... 91
iv
Figures 1-8.
Figure
9.
10.
11. 12. 13. 14. 15. 16. 17. 18.
19.
20. 21.
Figures 22-27.
ILLUSTRATIONS
Map showing: 1. Area of investigation 2. Locations of test wells 3. Locations of water-level observation wells open to
the principal artesian aquifer 4. Approximate thickness of the residuum 5. Altitude of top of the Ocala Limestone 6. Approximate thickness of the Ocala Limestone 7. Generalized altitude of the top of the Lisbon
Formation. . . . . . . . . . . 8. Average annual rainfall in the Dougherty Plain area,
1941-70........................................... Graphs showing monthly and annual precipitation at Albany
and monthly and annual runoff of Flint River between Montezuma 3 and Albany 24............................. Graph showing difference in monthly streamflow, precipitation, and principal artesian aquifer water levels near Albany. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Map showing locations of streamflow gaging stations...... Graph showing duration of daily flow at selected stations for eight major streams............................... Graph showing duration of daily flow at selected stations for nine minor streams....................... Map showing distribution of 7-day, 10-year minimum annual flows. . . . . . . . . . . . . . . . . . . . . . . . . Map showing distribution and range of annual mean and
seasonal runoff.......................................
Graph showing relation between base flow estimated from hydrograph separation and median flow.................
Map showing distribution and range of annual mean and seasonal base flows...................................
Stratigraphic section, geophysical logs, and waterbearing characteristic of geohydrologic units near Newton, test well 205-37..............................
lfup showing distribution of estimated vertical and horizontal hydraulic conductivity and transmissivity of
the residuum.. . . . . . . . . . . . . . . . . . . Map showing distribution of estimated leakance based on
test-well data and digital modeling analyses.......... Graphs showing water levels in residuum wells 087-44
and 201-16 and rainfall at Bainbridge and Colquitt
for 1980..............................................
Map showing: 22. Generalized altitude of the water table in the
residuum for mean yearly hydrologic contitions... 23. Distribution of point and regional values of trans-
missivity in the principal artesian aquifer...... 24. Distribution of point values of storage coeffici-
ents of the principal artesian aquifer........... 25. Potentiometric surface of the principal artesian
aquifer, May 1980................................
Page
2 5 8 11
12 13 14 15
17
18 19 23 24 32
33
36 37
38
40 42
43
44 45 48 49
v
Figures 22-27.
Figure
28.
29.
30. 31.
32. 33. 34.
35. 36.
37.
38.
39. 40. 41. 42.
43.
ILLUSTRATIONS
Map showing:--Continued 26. Seasonal water-level declines in the principal
artesian aquifer between May and November 1980 27. Difference in principal artesian aquifer water
levels between May 1980 and April 1981 Hydrographs showing fluctuations of mean monthly water
levels in the principal artesian aquifer at wells 087-23 and 095-68..................................... Hydrographs showing fluctuations of mean daily water levels in the principal artesian aquifer at wells 095-59 and 205-16 and 5-day rainfall totals at ~bany and Camilla.................................... Diagram showing conceptual flow model of the principal artesian aquifer system............................... Diagram showing conceptual flow model of hydraulic connection between the principal artesian aquifer and the Flint River................................... Map showing the model area with finite-difference grid and boundary conditions ~ Map showing measured stream discharge for August 1980 and January 1981...................................... Map showing measured water levels and simulated potentiometric surface of the principal artesian aquifer,
November 1979.........................................
Graph showing distribution of head error for the November 1979 calibration of steady-state simulation...........
Map showing areal distribution of difference between the November 1979 simulated potentiometric surface and the potentiometric surface constructed from measured water levels.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
tfup showing locations and capacities of agricultural irrigation systems in the Dougherty Plain area as of spring 1980...........................................
Map showing measured water levels and simulated potentiometric surface of the principal artesian aquifer, November 1980. . . . . . . . . . . . . . . . . . . .
Hydrographs showing measured and simulated water levels in wells 087-10, 087-23, 087-43, and 095-68, 1980.....
Hydrographs showing measured and simulated water levels in wells 201-05, 205-16, 253-08, and 253-26, 1980.....
Map showing locations and capacities of projected potential irrigation systems in the Dougherty Plain area...
Map showing simulated water-level declines in the principal artesian aquifer after pumping 113 billion gallons per year for 3 years during a hypothetical hydrologic drought....................................
Map showing simulated water-level declines below the top of the principal artesian aquifer after pumping 113 billion gallons per year for 3 years during a hypothetical hydrologic drought.......................
Page 50 52 53
54 63 65 67 69
71
72
73 76 78 79 80 82
83
84
vi
Figure 44. 45. 46.
ILLUSTRATIONS
Hydrographs showing measured and simulated water levels in the principal artesian aquifer in wells 087-23, 095-68, 201-05, 205-01, and 253-12................................
~~p showing simulated water-level declines in the principal artesian aquifer after pumping 408 billion gallons per year for 3 years during a hypothetical hydrologic drought . . . . . . . . . . .
~p showing simulated water-level declines below the top of the principal artesian aquifer after pumping 408 billion gallons per year for 3 years during a hypothetical hydrologic drought . ................ . .
Page 85 87 88
Table 1. 2.
3. 4. 5. 6.
7.
8.
9. 10. 11.
12.
13. 14.
15. 16.
17.
18.
TABLES
Page
Summary of test-well data 6 Generalized stratigraphy, water-bearing properties, and
water-quality characteristics of formations underlying
the Albany area ..............................
9
Continuous-record streamflow gag.ing stations 21
Base-flow discharge measurements 22
Summary of flow-duration data 25
Flow duration for individual months at selected streamflow
gaging stations 27-28
Low-flow characteristics at selected streamflow gaging
stations .................................................. 30-31
Base flow estimated from hydrograph separation and median
flow....................................................... 35
Hydraulic and water-level data for residuum test wells........ 39
Transmissivities and storage coefficients for the principal artesian aquifer ~... 46
Specific-capacity data and estimated transmissivities for
the principal artesian aquifer............................. 47
Estimated mean annual hydrologic budget factors for the
principal artesian aquifer system.......................... 56
Recommended and maximum concentrations of selected constitu-
ents in public drinking water supplies..................... 57
Selected water-quality data for wells from which water was
analyzed for major inorganic constituents and pesticides... 59
Agricultural pesticides commonly used in southwest Georgia,
1976-7 7. . . . . . . . . . . . . . . . 60
Statistical summary of water-quality data pertinent to the
residuum (RSDH) and the principal artesian aquifer (PCPA) 61-62
Measured and simulated ground-water discharge to selected
streams.................................................... 74
Sensitivity of aquifer transmissivity (T), confining zone
leakance (L), and riverhead (R) on the calibrated model
for November 1979 74
vii
CONVERSION FACTORS
For those readers who may prefer to use metric units or the International System of Units (SI) rather than inch-pound units, conversion factors for the terms used in this report are listed below:
Multiply inch-pound unit
To obtain metric (SI) unit
Length
inch (in.) foot (ft) mile (mi)
25.40 0.3048 1.609
millimeter (rom) meter (m) kilometer (km)
Area
acre square mile (mi2)
0.4047 2.590
hectare (ha) square kilometers (km2)
Volume
gallon (gal)
million gallons (Mgal) inch per acre (in./acre)
3.785
-3
3. 785 X 10
3,785
62.76
liter (L) cubic meter (m3) cubic meter (m3) millimeter per hectare
(rom/ha)
Flow
gallon per minute (gal/min)
million gallons per day (Mgal/d)
inch per year (in./yr) cubic foot per second
(ft 3js)
[(ft3/s)/mi2]
0.06309 6. 309 X 10-5
0.04381
25.40
-2
2.832 X 10
liter per second (L/s) cubic meter per second
(m3/s) cubic meter per second
(m3 /s) millimeter per year (rom/a) cubic meter per second
(m3/s) [(m3/s)/km2]
Transmissivity
foot squared per day (ft2/d)
0.09290
meter squared per day
(m 2/d)
Hydraulic conductivity
foot per day (ft/d)
0.3048
meter per day (m/d)
viii
Multiply inch~pound unit
gallon per day per cubic foot [(gal/d)/ft3]
foot per day per foot [(ft/d)/ft]
CONVERSION FACTORS
To obtain metric (SI) unit
Leak.ance
0.1337 1.000
meter per day per meter [(m/d)/m]
meter per day per meter
[ (m/ d) /m]
Ground-water term Transmissivity, T
Hydraulic conductivity, K
EXPLANATION OF UNITS
Original form
(m3/d)/m
=
(ft3/d)/ft
=
(gal/d)/ft
(m2/d)/m
=
=
(ft2/d)/ft
=
=
(gal/d)/ft
Reduced form m2/d ft2/d
m/d ft/d
National Geodetic Vertical Datum of 1929 (NGVD of 1929).--A geodetic datum derived from a general adjustment of the first-order level nets of both the United States and Canada formerly called mean sea level. NGVD of 1929 is referred to as sea level in the text of this report.
ix
HYDROLOGY AND MODEL EVALUATION OF THE PRINCIPAL ARTESIAN AQUIFER,
DOUGHERTY PLAIN, SOUTHWEST GEORGIA
Larry R. Hayes, Morris L. Maslia, and Wanda C. Meeks
ABSTRACT
Use of ground water for irrigation in the Dougherty Plain area of southwest Georgia increased from about 47 billion gallons in 1977 to about 76 billion gallons in 1980, and to 107 billion gallons in 1981. Most ground-water withdrawals are from a limestone aquifer, which is referred to locally as the Ocala aquifer but is more widely known in Georgia as the principal artesian aquifer. The aquifer in the Dougherty Plain area is overlain by about 25 to 125 feet of sandy clay residuum derived from chemical weathering of the Ocala Limestone.
Transmissivities of the principal artesian aquifer range from 2,000 to 1,300,000 feet squared per day. Storage coefficients range from 2 X 1o-4 to 3 X 10-2. Measured yields of wells in the principal artesian aquifer range from about 40 to 1,600 gallons per minute and commonly exceed 1,000 gallons per minute where transmissivity exceeds 50,000 feet squared per day.
Annual rainfall in the Dougherty Plain area averages about 53 inches. The annual mean, spring high, and late-summer low runoffs are, respectively, 5,200, 9,200, and 2,700 cubic feet per second. Average annual and summer mean base flows are, respectively, 4,000 and 2,300 cubic feet per second.
Under average hydrologic conditions, mean annual water levels in the principal artesian aquifer remain constant (recharge equals discharge). Annual mean recharge to the aquifer in the Dougherty Plain area is about 2,200 million gallons per day. About 90 percent of annual mean recharge is discharged to streams.
Water from the principal artesian aquifer is generally suitable for publicsupply, industrial, and irrigation purposes. Pesticides were detected in water
1
from 11 residuum wells and four principal artesian aquifer wells. None of the water samples from the principal artesian aquifer contained pesticide concentrations exceeding the recommended limits for public drinking supplies.
A two-dimensional finite difference model was used to simulate flow in the principal artesian aquifer. Simulation of a 3-year drought with pumpage of 113 billion gallons per year resulted in a mean water-level decline of 26 feet. Increasing pumpage to 408 billion gallons per year resulted in a mean decline of 33 feet. During the drought simulations, ground-water discharge to major streams was severely reduced and smaller streams ceased flowing. A 10-year simulation using average recharge and pumpage of 287 billion gallons per year resulted in a mean water-level decline of 4 feet and a 30-percent reduction in discharge to streaiils.
During drought conditions, present pumpage demands combined with reduced recharge could result in water-levels declining below the top of the aquifer and cause dry wells, well collapses, or possibly sinkhole development. Increased pumpage could increase the extent and magnitude of these problems.
INTRODUCTION
The principal artesian aquifer, which underlies parts of Alabama, South Carolina, Georgia, and all of Florida, is one of the most productive aquifers in the country. Large withdrawals of water from this aquifer for supplemental irrigation in the Dougherty Plain area of southwest Georgia (fig. 1); the potential withdrawal from the aquifer in other
85
84
83
- - + 35'!..,- -- l - - - - - - 1 - - -
35
\
' \
, (
' \ ..
34o::.....'1I....
34
~~'-:~... 330
.....
I....,
IV
"'
)
----~1
. Sovon ~' 320
-..J
c~.-,..."'
(
I
r
~
/J
~
I~ ~
85o- --'- --
a"'
(.)
0 -- -
84
..)_31 t.
+ (f-.. II
830- -
-~.,
_
j 8
0
2
--
"~t~~i:::
31
0
50
100 MILES
II III
J xc:r~
'
~~~ - ~
\ ~v ,
~
__
=_
___ ..,
I
l_
_
_
_
J I
s s'
0
10
20
30
40
50 MILES
I
Figure I. -Area of investigation.
areas of the Coastal Plain for irrigation; and declining water levels in the aquifer throughout the Coastal Plain are of concern to State and local officials. The Environmental Protection Division of the Georgia Department of Natural Resources, which has the responsibility of administering the Ground-Water Use Act (No. 1478, as amended through 1973), is especially concerned.
A mild climate, an abundant supply of good-quality ground water, a flat to gently rolling terrain, and the introduction of center-pivot irrigation systems have spurred a remarkable increase in agricultural irrigation in southwest Georgia. Ground-water use for irrigation in the Dougherty Plain between 1977 and 1980 increased from about 47 to 76 billion gallons per year (H. E. Gill, U.S. Geological Survey, written commun., 1981), with most of the water being pumped from the principal artesian aquifer.
Information regarding the hydrologic character of the principal artesian aquifer in the Dougherty Plain area is limited. Consequently, it was not known if the aquifer would be capable of supplying the increasing, long-term water needs of municipalities, industry, and agriculture, especially during hydrologic droughts such as occurred in 1954 and in 1980-81.
Previous Investigations
The general geology and ground-water resources of the Coastal Plain of Georgia have been previously discussed in McCallie (1898), Stephenson and Veatch (1915), Cooke (1943), and Herrick (1961). Geohydrologic reports primarily concerned with the Dougherty Plain include those by Wait (1963), Sever (1965a and 1965b), Pollard and others (1978), and Hicks and others (1981).
Purpose and Scope
The primary objectives of this investigation, which was carried out in co-
operation with the Georgia Geologic Survey, were to (1) define the geohydrology and hydraulic characteristics of the principal artesian aquifer system within the Dougherty Plain, largely through an extensive test-well drilling program; (2) develop a hydrologic budget in which total streamflow, base streamflow, and ground-water recharge or discharge are defined and quantified; and (3) develop a digital hydrologic model that can be used to simulate water-level changes in the principal artesian aquifer resulting from real or hypothetical pumpage increases.
Secondary objectives of the investigation were to (1) verify, expand, and add new hydrologic data to the existing data base; (2) evaluate present waterlevel and water-quality networks and to modify and expand these networks where necessary; and (3) analyze ground-water samples for pesticides, herbicides, and major inorganic dissolved constituents.
The Dougherty Plain investigation concentrated on delineating the hydrogeology of middle Eocene and younger rocks in a 15-county area of southwest Georgia (fig. 1). Twelve of these counties lie wholly or partially in the Dougherty Plain, which is the main area of interest. The total investigation covers an area of about 4,400 mi2 and lies within the High Irrigation Water-Use Zone, as defined by the Georgia Geologic Survey (W. H. McLemore, Georgia Geologic Survey, written commun., 1979).
Data Collection and Methods
Well and Surface-Water Station Numbering Systems
Data from 403 privately owned wells were entered into the computerized GWSI (Ground Water Site Inventory) system of
the u.s. Geological Survey. A listing of
these wells, with well construction and other pertinent information, and a location map are presented in a basic-data report prepared as part of the Dougherty Plain investigation (Mitchell, 1981, table 1 and plate 1).
3
The numbering system used to identify wells in this report follows that of Mitchell (1981) and consists of a 3-digit number that identifies the county in which a well is located, followed by a hyphen and a 2-digit number that is the serial number of the well in that county. For example, well 007-05 is in Baker County and has a serial number of 5. The table below lists the counties and their reference numbers:
Baker
007
Calhoun 037
Crisp
081
Decatur 087
Dooly
093
Dougherty 095
Early
099
Grady
131
Lee
177
Miller 201
Mitchell 205
Seminole 253
Sumter 261
Terrell 273
Worth
321
The 3-digit county number has been omitted in figures and tables that include county names.
Since October 1, 1950, the order of listing surface-water stations in U.S. Geological Survey reports is in a downstream direction along the main stream. All stations on a tributary entering upstream from a main-stream station are listed before that station. A station on a tributary that enters between two mainstream stations is listed between them. A similar order is followed in listing stations on first rank, second rank, and lower ranks of tributaries.
As an added means of identification, each surface-water hydrologic station and partial-record station has been assigned a station number. In assigning station numbers, no distinction is made b~t ween partial-record stations and other stations; therefore, the station number for a partial-record station indicates downstream-order position in a list made up of both types of stations. Gaps are left in the series of numbers to allow for new stations that may be established; hence, the numbers are not consecutive. The complete 8-digit number for each station such as 02349500 includes the 2-digit part number "02" plus the 6-digit downstream order number "349500". In this
report, the 3-digit sequence "023", which is common to all stations in the study area, has been omitted. Also for the reader's convenience, stations shown in all figures and referred to in the text are identified by a 1- or 2-digit map identification number. This number is keyed to the appropriate station number in the tables.
Test-Well Drilling
A test drilling program was necessary to obtain geophysical logs, lithologic samples, water samples, and hydraulic data where no wells existed or where existing data were inadequate. Thirtyfive wells were drilled under private contract and 15 wells were drilled by the Georgia Geologic Survey. Twelve wells were drilled in 1979 and the remaining 38 wells were drilled in 1980. (See fig. 2 and table 1.)
Spontaneous-potential, electricresistivity, gamma-gamma, neutron, caliper, and gamma logs were run in the 50 Dougherty Plain investigation test wells and in 18 privately owned wells. Drill cuttings from the test wells were collected, examined, and described lithologica 11y (Mit c he 11 , 1 9 8 1 , tab 1 e s 3- 4 6 ) These geophysical and lithologic logs and data from previous investigations were used as an aid in delineating and correlating stratigraphic and geohydrologic units. Watson (1981) presents this information diagrammatically in maps showing altitudes of the tops and thicknesses of the principal artesian aquifer and associated confining beds and a generalized geohydrologic section in a geohydrologic atlas prepared as part of the Dougherty Plain investigation.
Sources and Use of Hydrologic Data
The main ~ources of temperature, precipitation, and other climatological data are monthly bulletins and other reports published by the National Weather Service, National Oceanographic and Atmospheric Administration. Data concerning
4
30'
32oo'
COUNTY CODES
Baker
007
Calhoun
037
Crisp
081
Decatur
087
Dooly
093
Dougherty 095
Early
099
Grady
131
Lee
177
Miller
201
Milchell 206
Seminole 253
Sumter 261
Terrell
273
Worth
321
ft-4~uo'
-~
30'
EXPLANATION
c:::::=J AREA OF DOUGHERTY PLAIN
TEST WELL AND IDENTIFICATION NUMBER-Number refers to well listed in table 1
Bau lro.m U.S t o aQicql Survey 1 ~ 250 1 000 fjUOrlrDII!jjhtS
A
,.
ID
20
25
30 MILES
Figure 2:-- Locations of test wells .
5
Table I.--Summary of test-well data
[Geohydrologic unit: PCPA, principal artesian aquifer; RSDtt, residuum; TLLT, Tallahatta aquifer. Lithology (number in parenthesis is clay percentage): LHST, limestone; CS, sandy clay; SC, clayey
sand; SD, clean sand; SP, poorly sorted sand; SS 1 sand containing silt and clay]
\/ell No.
Well name
Geohydrologic unit
Altitude of
land surface (ft above
NGVD)
\/ell depth ( ft)
Casing depth
(ft)
Geohydrologic unit characteristics
Thickness (ft)
Lithology
06 Jo-Su-Li TW 29 T. Rentz TW 38 T. Rentz RW 39 Jo-Su-Li RW
PCPA PCPA RSDH RSDH
Baker County
160
IUO
158
liZ
!55
16
160
20
76
160
LHST
70
75
LMST
6
21
55(15)
10
29
CS( 57)
24 B. Jordan TW 25 B. Jordan T\1
Calhoun County
RSDtl
192
32
22
37
SC(30)
PCPA
195
145
60
U!ST
09
A. Newton 1 South TW
10
A. Newton, North TW
33 J. Hall TW I
42 .YoP 4
43 DP 5
44 DP 6
45 J, Hall TW 2
46 G. Bolton TW
47
A. Newton
Decatur County
PCPA -
115
145
60
250
UIST
PCPA
120
185
76
250
U!ST
PCPA
142
160
88
325
LI!ST
TLLT
145
455
382
so
PCPA
145
90
54
264
LMST
RSDU
145
40
30
54
SC( 30)
RSDU
135
35
25
40
SC( 35)
RSDtl
128
27
17
32
SC(32)
RSDH
112
39
29
53
SP(9)
14 Nilo, South TW
IS Nilo, North T\1
69 School Bus Road T\1
70 Game and Fish TW l
71 Nilo TW 3
72
USIIC Supply TW I
PCPA PCPA RSDH RSDtt RSDtl RSDH
Dougherty County
203
!50
201
150
195
29
215
15
202
40
227
45
60
!50
LHST
63
165
LHST
19
35
CS(54)
6
19
SC( 37)
30
50
SS(ll)
35
107
SS( 17)
39 I. Newberry T\J 45 I. Newberry TW 46 V. Evans TW 1
Early County
PCPA
230
125
61
70
LHST
RSDH
230
30
20
40
CS(50)
RSDH
178
40
30
46
SC(40)
15 U. Hoorman TW 1
PCPA
40
Piedmont Plant farm T\1 I
RSDtl
41 S. Stocks TW 1
RSml
42 B. King TW 1
RSD!l
43 II. Usry TW 1
RSIJH
44 S, Stocks TW
PCPA
Lee County
240
190
245
40
238
40
306
19
300
28
23U
64
140
LMST
30
47
SC( 37)
30
50
SP( 6)
9
24
SC( 49)
18
34
CS(68)
135
U!ST
15 DP 2 16 DP 3 33 J. Fleet Til 2
fUller County
PCPA
180
75
RSIJH
180
40
RSDtl
152
36
64
120
Lf!ST
30
55
CS(65)
26
41
SC( 31)
16 C. llol ton TW 1 34 H. l1einders TW 35 C. Holton Til 2 36 II, Davis TW I 37 DP 10 38 DP II 39 DP 12
Uitchell County
PCPA
150
190
50
250
Ll1ST
RSD!l
145
40
30
59
SS( 22)
RSDH
160
50
40
60
SD
RSDU
14 7
35
25
TLLT
165
417
397
40
SC(25)
so
PCPA
165
225
RSllH
165
37
62
252
U!ST
21
40
sc
08 Roddenberry TW 26 -21 0, Harvey TW I 27 Roddenberry TW 28 -21 D. Harvey TW 2
Seminole County
PCPA
115
150
PCPA
152
125
RSDtl
115
33
RSDtl
151
39
63
225
Ll1ST
58
75
U!ST
23
39
CS(SS)
30
54
SC(48)
22 E. Stephens TW
Sumter County
RSDH
290
27
17
34
SC(37)
14 A. Vann TW 1
Terrell County
RSU~l
263
20
10
20
C5(73)
03 llP 7
04
DP 8
OS DP 9
09 C. Odom TW I
1/orth County
TLLT
230
330
315
so
PCPA
230
120
63
162
Lf!ST
RSDtl
230
28
10
40
SS(20)
RSDtt
275
34
24
43
CS( 55)
};_/ DP indicates that the well is one of three test wells at the same site: DP 4, 7, and 10 are Tallahatta wells; DP 2, 5, 8, and 11 are principal artesian aquifer wells; and DP J, 6, 9, and 12 are residuum wells.
II Well is actually just across county line in Early County; however, to avoid changes in the
numbering system devised by llitchell (1981), the well is listed in Seminole County.
6
streamflow and stage measurements in and
adjacent to the study area are available
from the files and publications on
surface-water supply by the U.S. Geologi-
cal Survey. Additional streamflow meas-
urements were made in major streams in
the Dougherty Plain during August 1980
and January 1981. Ground-water contribu-
tions to streamflow were estimated by
using hydrograph separation techniques
and baseflow recession and flow-duration
curves. Water levels were measured twice
a year in about 200 wells that are open
only to the principal artesian aquifer
(fig. 3).
Periodic water-level measurements
also were made in wells open to the con-
fining beds immediately overlying and
underlying the principal artesian aqui-
fer. Water from selected wells open to
either the principal artesian aquifer or
the overlying residuum was analyzed for
organic and inorganic constituents.
Additional hydraulic data were ob-
ctauitnt1e.ndgsfrofmroma
quifer test
tests, wells,
core and ge
ospamhypsliecsal'
logs. Water-level drawdown and recovery
measurements were made in pumping and ob-
servation well(s) and used to compute
transmissivity and storage coefficients
of the principal artesian aquifer. Re-
sults of digital ground-water flow model-
ing were used to aid in defining the
aquifer flow system and to simulate re-
sults of hypothetical pumping situations.
Acknowledgments
Appreciation is extended to the following for allowing test drilling on their properties and for their continued cooperation throughout the study: Alvin Newton, I . M. Newberry, Jr., M. Moorman, Douglas Harvey, J. Hall, T. Rentz, Randall Newberry, Gerome Wells, Clyde Bradley of the Roddenberr~ Co., Clayton Holton of the Reba Corp., Robert Webber of AG-CON, Inc., and L. Johnson and Ralph Thompson of Jo-Su-Li Farms. The courtesies and help extended by T. Brogden, F. Thompson, and Kendall Bradley, and by John Flatt of Layne-Atlantic Co., are sincerely appreciated.
GEOGRAPHY
The Dougherty Plain, which receives its name from Dougherty County, is a nearly level area consisting of a series of level units. The plain is bounded on the west by the Chattahoochee River, on the east by the Pelham Escarpment, and lies roughly southward of the updip limit of the principal artesian aquifer (fig. 1). The plain slopes southeastward or southward from about 300 ft above sea level along the northern border to about 150 ft above sea level along the foot of the Pelham Escarpment and to about 50 ft above sea level below the confluence of the Flint and Chattahoochee Rivers. The average land-surface elevation is about 160 ft above sea level.
The Dougherty Plain is characterized by karst topography having numerous shallow, nearly circular, depressions (filled-in sinkholes) ranging in size from a few tens of square feet to many acres. Most of the older sink-hole bottoms are filled with silt and clay. As a result of the inability of water to move through these low permeability sediments, the older sinkholes form ponds that may hold water year round (Hendricks and Goodwin, 1952). The younger sinkholes normally do not hold water because their bottoms are not filled with lowpermeability materials. Consequently, water can move easily from them or into them from the underlylng limestone aquifer, depending upon head differential.
The Dougherty Plain is drained by the Chattahoochee and Flint Rivers and their tributaries. The drainage system will be discussed in more detail later in the report.
GEOLOGY
The area of investigation is underlain by a succession of sand, clay, and carbonate rocks to a depth of more than 5,000 ft (table 2). This report, however, is concerned with only the uppermost geologic units consisting of the residuum, the Ocala Limestone, and the
7
o'
3200' -
COUNTY CODES
Baker
007
Calhoun
037
Crisp
081
Decatur
087
Dooly
093
Dougherty 095
Early
099
Grady
131
Lee
177
Miller
201
Mitchell 205
Seminole 253
Sumter 261
Terrell
273
Worth
321
oo' -
EXPLANATION AREA OF DOUGHERTY PLAIN WELL-Number is county sequential well number
un n:;r
A
10
15
20
25
30 MILES
Figure 3.- Locations of water-level observation wells open to the principal artesian aquifer.
8
Table 2.-Generalh.ed stratigraphy, water-bearing properties, and water-qualit:y characteristics of formations underlying the Al))any area
[Froc:~ Hicks and others, 1981]
Era I System
I
Series
Quaternary
Pleist:ocene
I
I Gulf Coast: Stage I
Group and
formation
---
Oune sand
Terrace deposits
Thic1<.nes10 (feet)
0-35 ()-20
11t:hology
fine to coarse, ~oell sorted, angular to subangular quartz sand Poorly sorted gravel. sand, and clay
Wacer-bearing properties Not ~t~ater bearing rlot water bearing
\.later-quality characteristics
Oligocene
I VicKsburgian
Flint River Formation
Light-gray, cherty l.i.meston~
Propert:ies unkno..,n
Quality WI known
J~ckSonian
Ocala
15()-200 White to light:-pink, fossiliferous
Ocala aquifer is a very productive u'a:r:er-bearing
Water is generally a hard calcium bicarbonate
Lime!'ltone
limestone
unit throughout the Dougherty Plain. Reported well r:ype that meets all State drinKing water stanyields of more than 2,000 gal/min.. Yields decrease dards (1977)
--
~I
Lisbon
For~Dation
north and west of .Uhany
I Slightly glauconitic. fine, calcart!:ous Limited '-'ater-bearing potent:ial-used only in uul-
sand, clay, and interbedded limestones tiaquifer wells where other aquifers are tapped
Eocene
I
" Claibornian! e~
~I
;:;
Tallahatta Formation
235-340
Fine to medium sand, clayey sand, and interbedded limestone layers thar: are very fossilif~rous at the top of the formation
Tallahatta aquifer is a major aquifer in the Albany area; used for munictpal, agricultural, and industrial supplies. Reported well yields of as (iluch as 1,400 gal/min
Wgter 'la- " hurd edc.Jwu blctbon.ece cyp.e that meets all State drinking 'Jat~r a.ta.mbrdti ( 1977)
and .i.s suitable for most uses
Tertiary
..1
Sabinian lg
Hatchetigbee Formation
Very fine, green-stained quartz sand, locally calcareous and glauconitic
Aquifer is tapped by many multiaquifer ~o~ells; ho..,ever, water-bearing properr:ies unknown
Upper
I; ~
Tuscahoma
I llQ-120
Fine to r~~edium, micaceous, clay-rich
I Used i.n some multiaquifer wells; lo'ater-bearing
Paleocene
~ c
Sand and olanafalia Formation
sand. Glauconite is abundant tllrough- properties unknown out. Lower part is nonfossiliferous,
Quality unknown
~ undifferentiated
clay-rich sand (occasionally greater
~
than 50 percent: clay)
Clayton
I fine to medium. calcareous quart.: sand Used in soae r.mltiaqui.fer ~oiells; water-bearing
Fortaa~ion
40-120 and interbedded th.in limestones
g.
(upper unit)
1
properties un~mo~o~n
Lower
Pal~ocene
~ H i d w a y a n
~
F~~:~~~~n
(limestone unit)
=
Clayton
Formation
(lo~o~er unit)
7o-12S
15-40
~i::!~:~e~ig~~;~r~~~s~~~~~;~~!l ~~e~he
top of the unit
Fine to mediuc., arkosic SAnd, locally glauconitic >:~nd !:>tlty
~!:!~on~~~i!~r ~~a~/~~~;~ra~~~~!~r1!"a t~~o~:;:~u-
;~~id:o::v:~~a~o a~h~~~~~\:~~m~~r~:s~e~e;!ported
'Water-bearing properties unknown
The Clayton aquifer produces war:er that is suitable for municipal, agricultural. and industrial supply. It is generally a so.ft sodium bicarbaL\at:e type that meets all State drinKing water standards (1977)
Navarroan
Providence Sand
Ripley Formation
>2, 500
Upper part of unit is a dense, gray,
I Providence aquifec is used in the Alb!iny area for
clayey sand. Hiddle part is generaHy municipal and industrial supply. Yields range
a coquina. Lower part is sand con-
from less than 25 to about sao gal/min
tainiag varying amounts of silt
Fine to medium, calcareous sand and fossil i.ferous claystone
Not water bearing
Water from tt->is aquifer is a soft sodium bicarbonate type that is suitable for tiiOSt uses and 111eets State drinking water standards (1977)
Cretaceous
Gulf ian
Tayloran
Cusseta Sand
BLuff town Formation
Fine, micaceous, calca.reous sand containing varying &IUOunts of silt and clay
Not used as an aquifer in the Albany area; however in other areas of Georgia yields as great as 500 gal/min have been reportt:d
\.later is a soft sodilllil bicarbonate type that has concentrations of chloride and dissolved solids that exceed State drinking water standards ( 1977)
Comanchean
Austinian
Eaglefordian Waodbinian
Washitan, Fn de,ieiCo: - ' burgian, ~tltd Trinitian
Eutaw .rormation
Tuscaloosa forru.ation
UnJ. if feren.t i.ated
Alternating layer:> of sand, sandy clay. and clay
Not used i.n. the Alb11ny area
Water ~uality is abour: r:he same as that in th~ Cusseta and does not significantly change through the Tuscaloosa. Below the Tuscaloosa the concentration of $0diuc chloride is reported to incr<!ase ~1J;;n1ficancly
Lisbon Formation. The reader is referred to Hicks and others (1981) and Wait (1963) for a discussion of the lower units.
Residuum
The surficial geology of the Dougherty Plain consists of a residual layer of sand and clay derived from chemical weathering of the Ocala Limestone. The ratio of sand to clay in this residuum varies throughout the study area. Testdrilling data indicate that the residuum consists mainly of brown to red, mottled, clayey sand to slightly-sandy clay (Mitchell, 1981, tables 3-46). Clay content ranges from approximately 10 to 70 percent, with samples from 45 of 50 test wells consisting of more than 25 percent clay.
The residual layer ranges in thickness from a few feet to slightly more than 125 ft, and has an average thickness of approximately 50 ft (fig. 4).
Ocala Limestone
The Ocala Limestone is light colored and fossiliferous. The upper surface dips generally southeastward and occurs from about 300 ft above sea level in the northern part of the study area to about sea level in the southern part, but is highly irregular because of differential weathering (fig. 5).
The Ocala ranges in thickness from a few feet at the updip limit to about 350 ft in the southeastern part of the Dougherty Plain (fig. 6). The limestone is exposed along sections of major streams such as the Chattahoochee and Flint Rivers and Spring Creek, where erosion has removed the residuum. The Ocala is reduced in thickness at these exposures and near the updip limit may be entirely removed by a deeply incised stream.
The irregular surface of the top of the Ocala Limestone reflects solution that probably occurred during advances and retreats of Pleistocene seas. Numerous circular depressions in the topogra-
phy of the study area seem to be the result of settling of sediment-filled sinkholes.
Sinkholes formed by recent collapse and erosion are also common. Collapse sinks, which are normally steep sided and a few feet to tens of feet deep, can develop without warning and be fully developed in a short time. Erosion sinks occur where large volumes of residuum migrate downward into solution openings in the limestone and are carried away by moving water, creating a large cavity in the overlying residuum (Newton, 1976). When the cavity becomes so large that the strength of the overlying material is insufficient to maintain a cavity roof, collapse takes place, forming a sink. Most erosion sinks are shallow and have gently sloping sides.
Lisbon Formation
The Ocala Limestone is underlain by the Lisbon Formation, which consists of hard, well-cemented, sandy, clayey limestone of middle Eocene age. The Lisbon dips generally southwestward at about 12 ft per mile. The top surface occurs at altitudes ranging from nearly 300 ft above sea level in the northwestern part of the report area to about 350 ft below sea level in the southeastern part (fig. 7). Because of its distinctly lower water-yielding capability compared to the Ocala Limestone, the top of the Lisbon is considered to be the base of the principal artesian aquifer in the Dougherty Plain area.
THE HYDROLOGIC SYSTEM
Rainfall
Annual ~ainfall in the Dougherty Plain area averages about 53 inches and ranges from about 46 to 56 inches (fig. 8). Average monthly rainfall varies from 2 inches in October to 5 inches in March and 6 inches in July. Rainfall during the periods January through March and
10
;o'
3~ o.o ~ -
EXPLANATION
C J AREA OF DOUGHERTY PLAIN
-100 -- LINE DOaFshEeOdU:hLe:-eHICKNESS OF RESIDUUMInterval 25 leetapproxlmately Jocated
DATA POINT
01_.......~--10'---___I ~_[___.2l0______"1~_]30 MILES
Figure 4.-Approx imate thickness of the residuum. From Watson (1981) .
11
30' 3zoo'
3.0'
EXPLANATION
c=J AREA OF DOUGHERlY PLAIN
-100-- STRUCTURE CONTOUR-Shows altitude of the top of the Ocala Limestone. Dashed where approximately located. Contour interval 25 feet. National Geodetic Vertical Datum of 1929
DATA POINT
BQ!.-11 Pf4ttl U , e OQ ca Sutl.re' r-zso,ooo ll"Qdt(lnQiu
F L 0 R 1. D A
'"
.
20
26
30 MILES
Figure 5.-Aititude of top of the Ocala Limestone. From Watson (1981 )_
12
.A!S"oo'
30'
>D'
EXPLANATION
c=J AREA OF DOUGHERTY PLAIN
-325- LINE OF APPROXIMATE EQUAL THICKNESS OF THE OCALA LIMESTONEInterval 25 feet
DATA POINT
F e os_l! lrom U.S (1'0 OQI:Q "'"'r"l:ll 1; 2_50,000 Q 'll r;Jd lQl~ l
10
15
20
~'
30 MILES
Figure 6.-Approximate thickness of the Ocala Limestone. From Watson (1981) .
13
s5"oo'
30'
30' -' 1'"00"-
EXPLANATION
U
AREA OF DOUGHERTY PLAIN
--250-
STRUCTURE CONTOUR-Shows approximate altitude of the top of the Lisbon Formation, Contour interval 50 feet. National Geodetic Vertical Datum of 1929
DATA POINT
Bo!l! frarn ,S It'll o c;~ .: n l S ut on~~ H250,000 qillgd ~ an Q\m
......._..._..__1.0.__ _15.__.2.0l.--._25--'---3'0 MILES
Figure 7.-Generalized altitude of the top of the Lisbon Formation. From Watson (1981) .
14
EXPLANATION
D AREA OF DOUGHERTY PLAIN
-50- LINE OF EQUAL MEAN ANNUAL
RAINFALL-Interval 2 inches
10
15
20
,~s
3C'I MILES
~~--~~~
I
Figure a.-Average annual rainfall in the Dougherty Plain area, 1941-70. From Carter and Stiles (1982)
15
June through August is about equal in magnitude (15 inches), but differs greatly in duration and distribution. Rainfall in the winter months is usually of long duration and moderate intensity; rainfall in the summer months is usually of short duration and high intensity.
Rainfall varies considerably from year to year and from month to month. For example, annual rainfall at Albany varied from 35 inches in 1968 to 73 inches in 1964 (fig. 9). Monthly rainfall varied from 0.4 inch in October 1979 to 10 inches in February 1979 and from 0.8 inch in November 1980 to 12 inches in March 1980. Rainfall data collected at eight other stations in the Dougherty Plain indicate that spatial variation of rainfall is considerable and may vary from half to twice as much as that recorded at Albany for the same month.
As shown in figure 10, during September through May there is usually a direct correlation among precipitation, streamflow, and water levels in the principal artesian aquifer. Streamflow peaks occur soon after rainfall peaks as a result of direct runoff and precipitation falling directly into the stream channel. Ground-water peaks shown in figure 10 generally occur about 1 month after major precipitation peaks. This lag occurs because the precipitation moves slowly
downward through the low-permeability residuum and takes some time to show up as recharge to the principal artesian aquifer. Ground-water recharge resulting from rainfall will be discussed in more detail later in the report.
Rainfall seems to have little effect on streamflow and water levels from June through September (figs. 9 and 10). This is because evaporation-transpiration is extremely high during these months in the Dougherty Plain area, and almost all rainfall is lost to the evaporationtranspiration process. Consequently, rainfall is ineffective in recharging the ground-water system during summer months, and ground-water discharge is the primary source of streamflow.
Surface Water
Drainage Description
Streams draining the Dougherty Plain are of two types: (1) through-flowing streams that originate outside the area, including the Chattahoochee and Flint Rivers, and (2) streams that originate within the area, such as Spring, Kinchafoonee, Muckalee, and Turkey Creeks. (See fig. 11 for stream locations.)
The Flint River, which receives its name from large boulders of flint and silicified limestone, drains an area of about 6,000 mi2 within the Coastal Plain. Major tributaries to the Flint River in the Dougherty Plain include Cooleewahee, Ichawaynochaway, and Spring Creeks, all of which originate in the Dougherty Plain. Muckafoonee Creek, which enters the Flint River upstream from Albany, is formed by Muckalee and Kinchafoonee Creeks, which rise near the western edge of the Dougherty Plain. Coo leewahee Creek flows southward from its origin west of Albany through a shallow, swampy valley to the Flint River at Newton. Ichawaynochaway Creek and its tributary, Chickasawhatchee Creek, drain shallow, swampy valleys and flow southward from their origin in Terrell County to the Flint River south of Newton. Spring Creek rises north of Colquitt and flows southward into Lake Seminole, about 3 miles northeast of the junction of the Flint and Chattahoochee Rivers. No large streams enter the Flint from the east. The Pelham Escarpment to the east of the Flint River forms both a surface-water and a ground-water di~ide. Numerous small streams on the west side of the divide flow westward to the Flint River.
The Chattahoochee River is longer and larger than the Flint River but drains only about 1,800 mi2 within the Coastal Plain, or less than one-third of the area drained by the Flint. The Chattahoochee, like the Flint, is deeply incised within its flood plain and cuts i~to the underlying limestone a~uifer.
16
15
Precipitation
14
Q Runoll
15
~ Preolpltallon
14
bSJ Runoff
13
(/)
w J: 12
0 z
z 11
- z 10
0
1-
< 9
1-
a.
0 B w
- 0:
a. 7
...... Average precipitation lor 1935 - 80
'
-
13
. 12 -
11
10
9
8 -
--Average precipitation lor 1936-80
.
0
z <
6 ~
LL
. '
.
LL 5
.
0 z
::::l 0:
-
.
r
l
4
.
.
1r- . '
3
~ - - 2
~ I- r-
-.
\ ~l-
~ ~ ~ "~ . ~ 0
~[\.
"' z
<C
w
.' ~I:\ " " ,. w
< .. < z
,5 .. ~~"f1! \. ;0 ;:~s:
['\
> 0
:\
" W
6
I- . '
. ri 5 . .
-~
! - ~t-;:1-
r .,
i' '' I'
'
r-
l,.- ~ 1-
r -'
: i
3
-- ~' ~ 2
~~ i
- ~
.. 'I
!
~~. ~.,'\ ... t. "'"' ~Rr~,,
" ~ "'"' "' " "' ~ 0 z ::!; w < " ""' ...
<
,_ zw =>
0
<
" 0
0 z ~
I
w "
0
1979
TIME
1980
75~--------------------------------------~
0 Puu:~lltll ll o r\
m Precipilalion
70
tsSJ Runofl
7
Q Runoff
Average preclpltallon 51 ,66 Averago tunoff '15. 13
Av erage precipitation 4.28 1962- 80 Averag e runo lf 1.25 1961 - 80
60
6
(/)
w J: 50 0
z
z
z 40
0
1-
<
1-
a. 30
0 w 0:
a.
0z 20 <
LL LL
0 z :a::::l 10
TIME
Figure 9.-Monthly and annual precipitation at Albany and monthly and annual runoff of Flint River between Montezuma 3 and Albany 24.
17
15
w
(.)
I
< 20
IaL:
:::;)
(/)
-w-' o 25 >w__,_z<_,
"~'0;: 30
;<:-wm-' 35
fww-
IL
40
~
45
Well 095-068
20
0 z
18
0
- Dilference n streamllow
Station 3- Station 24
(.)
w
(/)
.....
00
16
(/)
w
r
(.)
~ 14
~
zQ 12
I~ h
f-
<
fa-: 10
0aw:
0..
8
>
-r-'
fz -
6
~~~~w
(
~
~~
t,
/\
Precipita lion
~
' 'II
A \
J
"
v
~
\
~
~
A
M'\J LJ
a:
w
0..
wfw
I
IL
(.)
iii
:::;)
A
I
(.)
~
oi
0 --'
IL
J \; ' :; <wa: f-
0
(/)
:; 4
~
w
2 ~I
j
~ :ml
:
a(zw.:)
w
'
IL IL
0 1 981
1962
1a-&3
1964
1965
1966
1967
1968
1969
19 70
1971
1972 I 1973
1974
:M
1975
1976
1977
1978
1.979
1880
0
Figure 10.-Difference in monthly streamflow, precipitation, and principal artesian aquifer water levels near Albany .
30'
EXPLANATION
c::J AREA OF DOUGHERTY PLAIN
A 34
CONTINUOUS-RECORD GAGING STATION AND IDENTIFICATION NUMBER
~ 45
MEASUREMENT SITE WITHOUT A GAGE, AND IDENTIFICATION NUMBER
10
115
20
25
~0 MILES
Figure II. -Locations of streamflow gaging stations.
19
Swamps occur only along the lower reach of the Chattahoochee, primarily in Seminole and southern Early Counties. No large tributaries to the Chattahoochee River occur in the Dougherty Plain.
Streamflow
An important characteristic of streamflow is its variability with time and location. In order to measure and record streamflow on a systematic basis, continuous-record gaging stations have been operated in southwestern Georgia since the early 1900's. (See table 3 for a listing and figure ll for locations of continuous-record stations referred to in this report.) Additional streamflow data used in this investigation include measurements of discharge at partial-record measurement stations made during August 4-7, 1980, and January 5-7, 1981 (table 4). Because streams in the Dougherty Plain area are utilized appreciably for both irrigation and power generation, collection and analysis of streamflow records are necessary to evaluate streamflow characteristics that may be used by planners, designers, and farmers in deciding how to best utilize the stream resources.
Flow duration
The flow-duration curve is a cumulative frequency curve that shows the percentage of time during which specified discharges were equaled or exceeded in a given period (Searcy, 1959). A flowduration curve simply provides a means of representing in one curve streamflow characteristics throughout the range of discharge. It is important, however, to note that the flow-duration curve does not show the chronological sequence of flows and therefore is not a reliable method for predicting the dependability of flow. It is also not generally applicable to flood studies. If the curve is based on a sufficiently long period of stream-discharge data, the curve may be used to predict the distribution of
future flows for water-power, watersupply, and pollution-load studies. The flow-duration curve also may be used for studying and comparing watershed characteristics.
Flow-duration curves were developed for stations in the Dougherty Plain area having more than 7 years of daily record by using standard computer programs
developed by the u.s. Geological Survey
(figs. 12 and 13). Selected streamflow station data and computer-generated coordinates used in plotting the curves are listed in table 5.
Except in watersheds where soils are highly permeable, the distribution of high flows is governed mainly by the climate, watershed physiography, and plant cover. Low-flow distribution is controlled mainly by basin geology. Consequently, the high end of the flowduration curve is an indicator of direct runoff characteristics and the low end is an indicator of base runoff or groundwater contribution to streamflow.
The moderately steep slopes of the upper halves of the flow-duration curves in figures 12 and 13 indicate that direct runoff significantly contributes to the higher flows; the relatively flat slopes, particularly at the lower end, indicate that low flows are maintained by groundwater discharge or that a large amount of ground- or surface-water storage occurs in the watershed.
Regulation of streamflows resulting from changes in storage in reservoirs or lakes has a significant effect on the flow regime at a specific stream site and is reflected in the stage and character of the flow-duration curve. Normally high flows are reduced in magnitude and low flows are augmented. Flow-duration curves for the Apalachicola River at Chattahoochee, Fla. (station 61), before and after regulation illustrate a moderate change in duration resulting from regulation by the Jim Woodruff Dam (fig. 12).
The low discharge parts of the flowduration curves at stations 5, 35, and 40 (fig. 13) show distinctly steeper slopes than do the other curves shown in figures 12 and 13. During extended periods of
20
Table 3.--Continuous-record streamflow gaging stations
Station No.
Station name
Drainage area (mi2)
Period of
record used
Average annual Median runoff discharge (in.) (ft3/s)
Average discharge
Max. discharge (ft3/s)
Hin,
discharge (ft3/s)
43500 Chattahoochee River at Columbia, Ala. 8,050 1928-60 17.80 7,300 10,540
44000 Chattahoochee River at Alaga, Ala.
8,340 1939-70 18.87 8,500 11,590
49500 Flint River at Montezuma
2,900 1930-80 16.92 2,500
3,613
49500
do.
2,900 1980 17.46 2,280
3, 719
49900 Turkey Creek at Byromville
45 1958-80 14.97
19
50
49900 50500
do.
Flint River at Oakfield ll
45 1980 11.77 3,860 1929-58 15.46
3,100
39 4,397
50600 52500
Kinchafoonee Creek at Preston
Flint River at Albany ll
197 1951-77 14.82 5,310 1901-80 16.21
150 4,300
215 6,338
52500 53000
do.
Flint River at Newton l l
5,310 1980 16.03 5,740 1938-80 16.77
3,800 5,000
6,251 7,090
53000
do.
5,740 1980 15.60
6, 579
53400 Pachitla Creek near Edison
188 1959-69 17.70
170
245
53500 Ichawaynochaway Creek at Hil ford
620 1939-80 17.57
560
802
53500
do.
620 1980 16.13
734
54000 Alligator Creek near l!ilford
14 1942-50 11.63
6.9
12
54500 Chickasawhatchee Creek at Elmodel
320 1940-50 16.08
180
379
55000 Ichawaynochaway Creek near Newton
1,020 1938-47 15.80
880
1, 187
55500 56000
Big Cypress Creek near Hilford
Flint River at Bainbridge ll
1942-49 7,570 1908-71 15.68
6 6,400
3.3 8,730
56000
do,
7,570 1958-71 16.31 6,500
9,093
56500 Long Branch near Damascus
18 1945-49 17.35
27
23
57000 Spring Creek near Iron City
485 1938-70 13.36
230
477
57000
do.
485 19.58-70 14.28
230
510
58000 58000
Apalachicola River at Chattahoochee, Fla.
do. 3/
17,200 1929-80 17.86 17,000 17,200 1958-80 19.31 17,000
22,570 24,400
58000
do.
17,200 1980 20.12 20,000 25,420
1.31 1.40 1. 24 1. 28 1.10
.86 l. 14 1.09 1. 19 l. 18 1. 24 1.15 1.3 1.29 1.18 .86 1.18 1. 16
1.15 1. 20 1. 27
.98 1.05
1. 31 1. 42 1. 48
6,812 203,000 1,210
7,490 112,000 1,230
2,335 68,900 585
2,403 28,700 845
32
3,940
.1
25
679
2.5
2,842 60,500 152
139
8, 200
18
4,096 77' 000 372
4,040 39,100 1,220
4,582 66,600 790
4,252 34,500 1,610
158
9,060
35
518 11 '900 116
474
7,240 122
7.8
84
0
245
3,630
5
767 10,300 205
2.1
105
0
5,648 83,200 1,340
5,877 67' 500 1'340
15
787
0
308 12,600
9
330
8,260
11
14,586 15,769 16,428
291,000 5,010 165' 000 6, 730 103,000 8,790
Discharge affected by powerplant operation, but normal operation of powerplant does not materially affect average monthly figures of runoff.
3/ After construction and filling of Jim Woodruff Dam (1954-57).
21
Table 4.-Base-flow discharge measurements
Station No.
Station name
Drainage
area (mi2)
Date
Measurements
Discharge (ft3 /s)
Date
Discharge (ft3 /s)
49800
Flint River near Methvins
3,200
8-4-80
1, 07 0
1-5-81
1,600
49910
Turkey Creek near Drayton
76.0
8-6-80
10
1-5-81
21.1
49980
Pennahatchee Creek near Drayton
102
8-6-80
1.5
1-5-81
5.24
50070
Lime Creek near DeSoto
35.9
8-5-80
21
50220
Gum Creek at Coney
73.0
8-5-80
14
1-5-81
9. 29
50360
Swift Creek near Warwick
40.0
8-5-80
15
1-5-81
11.0
50509
Jones Creek near Oakfield
50.5
8-5-80
7.1
1-5-81
4.39
50524
Abrams Creek near Oakfield
80.2
8-4-80
12
1-5-81
10.7
50543
Piney Woods Creek
above Albany
N
N
50860
Kinchafoonee Creek
near Smithville
60.4 485
8-4-80 8-5-80
0
1-6-81
0
66
1-6-81
210
51000
Kinchafoonee Creek
586
8-5-80
105
1-6-81
230
near Leesburg
51700
Muckalee Creek near
265
8-5-80
Smithville
39
1-5-81
163
51780
Muckaloochee Creek near Americus
27.1
8-7-80
8.9
1-6-81
16.8
51800
Muckaloochee Creek at Smithville
47
8-5-80
16
1-5-81
30.4
51920
't-1uckalee Creek below
416
Leesburg
8-5-80
77
1-6-81
222
52760
Dry Creek near Putney
68.1
8-4-80
0
1-6-81
52920
Raccoon Creek near
92.9
8-4-80
0
1-6-81
0
Baconton
52980
Cooleewahee Creek at Newton
151
8-5-80
7.8
1-6-81
1. 84
53100
lchawaynochaway Creek
118
near Dawson
8-4-80
28
1-5-81
66.5
53265
Ichawaynochaway Creek
301
near Morgan
8-5-80
70
1-5-81
158
53460
lchawaynochaway Creek
570
8-5-80
130
1-6-81
280
near Leary
Station No.
Table 4.--Base-flow discharge measurements-Continued
Station name
Drainage
area (mi2)
Date
Measurements
Discharge (ft3 /s)
Date
Discharge (ft3/s)
54350
54410
54440 55000 55350 55600 55785 55830 55880 55950 56100 56220 56290 56600 56640 56860 56970 57025 57050 57310
Chickasawhatchee
118
Creek near
Albany
Chickasawhatchee
157
Creek near
Leary
Kiokee Creek near Pretoria
67.0
Ichawaynochaway Creek 1, 020 near Newton
Ichawaynochaway Creek 1, 040 below Newton
Big Cypress Creek near Newton
Big Slough near
105
Camilla
Big Slough below
157
Camilla
Big Slough near
214
Pelham
Big Slough near
315
Bainbridge
Spring Creek near
49
Arlington
Spring Creek at Damascus
99.8
Dry Creek near Blakely
45.5
Long Branch near Colquitt
Spring Creek at
281
Colquitt
Big Drain Creek near Boykin
Aycocks Creek below Colquitt
Dry Creek near Iron City
Spring Creek at
560
Brinson
Fishpond Drain near Donalsonville
8-4-80
8-5-80
8-5-80 8-5-80 8-5-80 8-5-80 8-4-80 8-4-80 8-4-80 8-4-80 8-5-80 8-4-80 8-4-80 8-4-80 8-4-80 8-5-80 8-5-80 8-5-80 8-5-80 8-5-80
.62 1-19-81
1. 3
1-7-81
0 260 268
0 1.1 0 0 0 3.0 11 6. 9 0 42 0 0 0 150 0
1-7-81 1-6-81 1-)-81 1-5-81 1-5-81 1-5-81 1-5-81 1-5-81 1-6-81 1-5-81 1-5-81 1-5-81 1-5-81 1-6-81 1-6-81 1-6-81 1-6-81 1-6-81
28
37.1
0 374 380
0 4.6
28 0 0 2.40 4.95 8. 78 0 36.6 0 0 0 82.7 0
Cz l
0
(.)
w
(/)
w0:::
a_ 10,000 I-
w w
lL
(.)
(])
:::>
(.)
z
w
(!)
0:::
<{
:::c
(.) (/)
Cl
1000
REGULATION
Map identification numbe 7 0 6/
PERCENTAGE OF TIME INDICATED DISCHARGE WAS EQUALED OR EXCEEDED
Figure 12-.-Duration of daily flow at selected stations for eight major streams.
23
0
z
0
(.)
w
(/)
a::
w
CL
1w w u..
(.)
m ::::>
(.)
z
w
(!)
a::
<(
r
(.) (/)
0
10
/ Mopidentification number
"o'
_.._
/ Period of record
'11>".'.., /
(water year)
~
PERCENTAGE OF TIME INDICATED DISCHARGE WAS EQUALED OR EXCEEDED
99.99
Figure 13.-Duration of daily flow at selected stations for nine minor streams.
24
Table 5.--Sumnary of flow-duratif~n data
Sta tion No .
Station name
Drainage Period
Percentage of time flow, in ft3/s, was equaled or exceeded
Variability
area
of
(miZ) record
0.1
o. 5
I
z
5
10
zo
30
50
70
80
90
I 95
98
99
99.5
99. 9
index q50/q90
43500 Chattahoochee River at Columbia, Ala.
8,040 1930-60 90,000 70,000 57,000 44,000 Z8,000 zo,ooo 13,000 11,000 7 ,zoo
5,000
4, zoo 3,ZOO z, sao Z,OOO I, 700 I, 500 I,ZOO
z. zs
44000 L"hattahoochee River at Alaga. Ala.
8, 340 194()-70 100,000 78,000 64,000 48,000 30,000 22,000 15,000 IZ,OOO 8,400
5, 800
4,800 3,800 3, 200 2, 500 Z, 200 1, 900 1,500
2. 21
49500 Flint River at Montezuma.
Z,900 193Z-80 36,000 Z5, 000 20,000 15,000 10,000 7,400 5,000
3,800 Z,500
1,700
1,400 I,ZOO 1,000
840
780
no
640
Z. 08
49900 Turkey Creek at Byromville
45 196()-80 I, 100
640
450
310
180
120
68
42
18
10
7. 6
5.8
4. 8
3.8
3.1
Z.4
1.1 3.10
50500 Flint River at Oakfield
3,860 1931-58 39,000 Z7 ,000 23,000 18,000 13,000 8,800 5,800
4,400 3,000
Z, 100
1, 700 1,400 1,000
660
480
380
zso
Z.l 4
50600 Kinchafoonee Creek at Pres t on
197 1953-77 Z,400 1,500 1,200
880
580
410
290
220
ISO
100
80
58
46
35
30
27
2Z
z. 59
5Z500 Flint River at Albany
5, 310 1903-80 49,000 36,000 31,000 Z5,000 18,000 13,000 8,600
6,600 4,600
3,000
2, 500 1,900 1,600 1,ZOO 1,000
880
700
Z. 4Z
53000 Flint River at Newton
5, 740 1939-30 49,000 39,000 33,000 27,000 19,000 14,000 9,200
7,000 4,900
3, 700
3,100 2, 500 Z, 100 1,800 1, 600 1, 500 I,ZOO
1.96
53400 Pachitla Creek
N
near Edison
VI
53500 Ichawaynochaway Creek
at Milford
188 1961-69 Z, 700 1, 700 1,300
980
640
460
3!0
Z40
170
130
100
80
64
50
45
42
39
Z.1 2
6ZO 1941-80 8,400 4, 900 3,800 3,000 Z, 100 1, 600 1,100
szo
560
400
330
260
220
180
ISO
140
120
Z.I S
54000 Alligator Creek near Mil ford
14 1943-50
240
140
98
62
38
27
18
13
6.8
3.1
- 1.5
-
-
-- -
-
-
54500 Chickasawhatchee Creek
at Elmodel
320 1941-49 3,800 2,800 2, 300 1, 900 1,400 1,000
620
400
180
90
62
34
21
11
7.8
6.2
5.2 s. 29
55000 Ichawaynochaway Creek
near Newton
1,0ZO 1939-47
-
7,800 6,200 4,800 3, 300 2, 500 1, 700
1,300
880
600
490
370
310
260
240
230
210
Z.38
55500 Big Cypress Creek near Mil ford
-
1943-49
62
39
32
23
15
10
5.4
3 -
- - -- - - - - - --
56000 Flint River at Bainbridge
7, 570 1909-71 64,000 44,000 36,000 30,000 22,000 16,000 12,000
9,000 6,400
4,800
4,200 3,400 5,000 2,600 2,400 2,200 2,000
1.88
57000 Spring Creek near Iron City
485 1939-70 6,400 4, 300 3, 500 2, 600 1, 700 1,100
700
470
230
130
94
60
40
25
19
15
10
3.83
58000 58000
Apalachicola River near Chattahoochee, Fla.
do .JJ
17,200 17,200
1930-30 130,000 1930-53 140,000
99,000 90,000 74,000 57,000 44,000 31,000 98.000 84,000 70,000 54,000 40,000 28,000
24,000 17,000 22,000 16,000
12,000 12,000
10,000 10,000
9,000 8,800
7. 700 7, BOO
6,500 6,800
5,900 6,400
5,600 6,000
5,000 5 , 600
1.89 1. 82
58000
do.!:/
17,200
1959-80 130,000 110,000 94,000 32.000 64,000 50,000 35,000
27.000 18,000
13,000
11,{)00
9, 800
9,000
8, 200
7 ,BOO
7,600
7' 200
1.84
!f Prioc to construction and filling of Jim Woodruff Oal.il. !:_/ After construction and filling of Jim Woodruff Dam.
little or no rainfall, these streams receive little base runoff from the underlying principal artesian aquifer, and usually cease flowing. The sharp downward steepening of the curve below the 98-percent duration of flow at Turkey Creek at Byromville (station 5) may be partly due to irrigation withdrawals from the stream upstream of the gage.
Normally the flow-duration curve for a particular station is based on all the observations of flow throughout the year for the available period of record; and, as indicated before, a curve computed in this manner fails to take into account time and seasonal effects. But, the seasonal nature of streamflow can be defined from a partial duration curve based on daily mean discharges from the historical records of individual months. For example, all the daily mean discharges for all the January months for which records are available, are used to define a January curve.
Table 6 summarizes individual monthly flow-duration data for five stations and indicates seasonal variability in expected streamflows. If a graphical presentation is desired, the data can be plotted on log-probability paper which would give curves similar to those in figures 12 and 13. Because of the seasonal importance of low flows and to allow the low-water season to be considered as a unit, the climatic year
(April 1 to March 31) was used as a basis for period of record for monthly flowduration data and for low-flow frequency data.
Flow-duration values defined from all flows or from only individual months for similar periods of record vary considerably. For example, at station 24 (Flint River at Albany) for the same period of record, the 50-percent duration flow based on all March data is 10,000 ft3/s, while the 50-percent duration flow based on all October data is 2,300 ft3/s (table 6).
The seasonal nature of streamflow is of particular importance to those who use streamflow for supplemental irrigation. Those who use streamflow for irrigation are primarily concerned with the stream-
flow available from May through September. Data in table 6 can be used to estimate probable streamflows of selected streams for those months. This applies, however, only if the historical period of record from which the data were derived can be considered representative of the predictive period. As will be discussed later, decline of ground-water levels associated with irrigation pumpage may result in some streams becoming influent, i.e., supplying water to the ground-water system whereas before ground water discharged to the streams, and to some extent "drying up." Where this occurs, streamflow will be less than discussed above.
Low-flow frequency
Information on low-flow recurrence is particularly important in the design of water-supply and waste-treatment facilities, because the lowest discharge commonly establishes the limit of supply without storage or the expected minimum dilution level for treatment operations during critical low-flow periods. For design purposes, the 7-day, 10-year low flow is the most commonly used value. It is based on annual 1111.n1.mum flows and indicates the lowest average flow during 7 consecutive days that is likely to be equaled or exceeded in severity on the average of 10 times in 100 years. This is not a common, nor is it an extremely rare flow.
The low-flow frequency data given in this report are for two types of stations: (1) daily-record stations having 10 consecutive years or more of daily record, and (2) low-flow partial-record stations. The data for the long-term daily-record stations were developed using the log-Pearson Type III method of analysis. At partial-record gaging stations, flow measurements made usually once a year during a time of base flow are related to concurrent flows at a nearby index continuous-record gaging station. The relation between these concurrent flows is used along with a frequency curve for the continuous-record
26
Table 6.--Flow duration for individual months at selected streamflow gaging stations [Period of record, climatic years 1959-70]
Month
Percentage of time flow, in ft3/s, was equaled or exceeded
10
25
50
75
90
Percentage of time flow, in ft3/s, was equaled or exceeded
10
25
50
75
90
Percentage of time flow, in ft3/s, was equaled or exceeded
10
25
50
75
90
Station 44000 Jan. 29,000 23,000 14,000 8,800 4,600 Feb. 33,000 23,000 14,000 9,300 5,100 Mar. 40,000 26,000 17,000 11,000 5,700 Apr. 44,000 24,000 14,000 9,400 3,900 May 23,000 14,000 9,400 5,900 3,600 June 15,000 11,000 8,000 5,600 3,500 July 13,000 10,000 7,700 4,700 2,700 Aug. 13,000 10,000 7,300 4,800 3,300 Sept. 11,000 8,800 6,600 4,600 3,600 Oct. 12,000 9,100 5,700 3,900 2,400 Nov. 15,000 11,000 6,900 4,900 3,600 Dec. 25,000 15,000 9,700 6,000 3,800
Station 49900
Station 50500
160
84
41
18
8.3 15,000 7,500 5,600 3,700 2,500
220
140
76
36
14
14,000 9,700 5,400 4,000 2,500
180
120
71
39
20
19,000 11,000 6,900 5,500 3,300
210
89
42
21
11
15,000 8,500 5,800 3,900 2,500
68
28
15
9.3
5
8,500 5,100 3,600 2,600 1,400
68
28
14
8.3
5.4 5,800 3,800 3,000 2,100 1,000
100
29
14
7.6
4.5 6,900 4,300 3,300 2,400 1,400
71
31
11
7. 1
4.9 5,700 3,800 2,900 2,100 1,100
27
16
8.2
5.7
3.4 3,500 2,700 2,100 1,600 520
30
14
7.5
5.6
2.7 3,400 2,600 2,000 1,600 620
24
16
8.9
6.5
4.5 5,600 3,300 2,400 1,600 980
58
25
13
8.5
5.3 12,000 5,300 3,500 2,600 1,500
Station 50600
Station 52500
Station 53000
Jan.
510
350
250
170 110 19,000 11,000 6,700 4,400 3,100
19,000 11,000 7,400 4,700 3,700
Feb.
650
470
290
180 130 21,000 15,000 9,100 5,300 3,900
21,000 16,000 9,700 5,900 4,500
Mar.
650
430
300
200 140 23,000 15,000 10,000 6,700 5,000
24,000 16,000 11,000 7,600 5,700
Apr.
690
380
250
150
99 22,000 14,000 7,100 4,800 3,700
23,000 15,000 8,500 5,700 4,400
May
360
210
130
88
56 11,000 6,300 4,300 3,200 2,500
12,000 7,400 5,100 4,000 3,300
June
280
170
110
77
37 9,300 5,400 3,600 2,700 2,100
9,800 6,200 4,300 3,600 3,000
July
260
170
110
81
42 7,100 4,900 3,500 2,600 2,000
7,600 5,600 4,300 3,300 2,600
Aug.
260
160
93
67
36 6,900 4,900 3,000 2,200 1, 500
7,600 5,700 3,800 2,800 2,200
Sept.
170
120
81
54
28 4,400 3,200 2,300 1,700 1,100
5,200 4,000 3,000 2,300 1,800
Oct.
240
120
80
60
36 6,300 3,400 2,300 1,600
940
6,600 3,900 2,900 2,200 1,600
Nov.
220
160
110
90
53 5,300 3,900 2,700 1,700 1,200
5,800 4,400 3,100 2,200 1,700
Dec.
350
230
160
120
98 8,700 6,000 3,900 2,800 2,100
8,700 6,200 4,200 3,200 2,700
27
Table 6.--Flow duration for individual months at selected streamflow gsging stations--Continued [Period of record, climatic years 19S9-70]
Month
Percentage of time flow, in ft3/s, was equaled or exceeded
10
2S
so 7S 90
Percentage of time flow, in ft3/s, was equaled or exceeded
10
25
so
7S
90
Percentage of time flow, in ft3/s, was equaled or exceeded
10
2S
so
7S
90
Jan. Feb. Mar. Apr. May June July Aug. Sept. Oct. Nov. Dec.
Station S3SOO 2,000 1,200 7SO S70 4SO
2,400 1,600 1,100 660 soo
2,300 1,600 1,100 790 630
2,100 1,300 860 sso 390
1,200 690 480 3SO 280 1,200 720 440 330 210 1,000 680 480 320 2SO
860 S70 420 320 240 S90 440 330 240 180 840 490 330 2SO 200 690 S30 370 300 2SO 1,100 720 S20 400 3SO
1,300 1,200 1,900 1,700
800 380 480 S20 230 230 2SO 1,300
Station S4SOO
810
400
2SO
8SO
S10
310
1,200
680
460
810
S10
320
440
210
110
170
92
42
260
120
78
290
170
7S
140
7S
34
120
63
33
120
70
36
400
120
76
110 200 230 120
4S 16 41 24 12 6.3 12 41
Station S6000 21,000 13,000 8,900 6,100 4,800 24,000 19,000 12,000 7,200 S,700 26,000 21,000 14,000 10,000 7,400 27,000 18,000 12,000 7,SOO S,900 14,000 10,000 7,100 S,400 4,SOO 13,000 8,700 6,100 4,700 3,800
9,800 7,700 S,900 4,600 3,SOO 9,800 7,200 S,400 4.,000 3,200 6,700 S,400 4,300 3,300 2,600 8,900 S,700 4,200 3,300 2,700 7,SOO 6,200 4,300 3,400 2,700 11,000 7,900 S,800 4,SOO 3,800
Jan. Feb. Mar. Apr. May June July Aug. Sept. Oct. Nov. Dec.
Station S7000 1, 600 800 410 220 140 2,100 1, 400 810 320 220 2,000 1,400 900 S90 370 1, 900 1, 000 610 370 220
680 430 260 180 130 790 330 190 130 90 610 370 180 100 6S 4SO 290 200 110 73 330 160 110 68 46 470 230 120 68 36 390 170 110 68 39 430 2SO 170 97 7S
Station S8000 S3,000 36,000 2S,OOO 16,000 12,000 62,000 47,000 31,000 20,000 14,000 6S,OOO S2,000 38,000 24,000 17,000
73,000 so,ooo 31,000 19,000 13,000
38,000 26,000 17,000 13,000 11,000 30,000 21,000 1S,OOO 13,000 11,000 2S,OOO 19,000 14,000 12,000 10,000 22,000 17,000 13,000 11,000 9,800 19,000 14,000 12,000 9,800 8,600 22,000 13,000 11,000 8,800 7,SOO 21,000 16,000 11,000 9,200 7,400 34,000 22,000 14,000 12,000 10,000
28
site to approximate flow-frequency data for the partial-record site. All lowflow data prior to 1977 used in this report are from Carter and Putnam (1978). Low-flow data extending beyond 1977 have been analyzed in accordance with their methods. (See table 7 for a listing of these data.) The areal distribution of 7-day, 10-year minimum annual flows is shown in figure 14.
Figure 14 and table 7 may be helpful to those interested in the use of streamflow for irrigation during periods of less than normal rainfall. Since the minimum flows often coincide with peak irrigation demands, the low-flow data give an indication of sustained streamflow available at that critical time. Expected low flows for gaged streams can be estimated from table 7, while expected low flows for ungaged streams can be estimated from figure 14 by using the data pertinent to that specific stream or watershed. Note, however, that their suggested use requires one to make an assessment of the severity of the hydrologic drought condition in order to choose the appropriate set of low-flow data.
Smaller streams in the Dougherty Plain generally have very low flows during dry seasons, with most having 7day, 10-year flows of zero. Only the large streams are incised deeply enough to remain below the potentiometric surface of the principal artesian aquifer and receive ground-water discharge during extended dry periods. Contribution to 7-day, 10-year flow derived from drainage within the Dougherty Plain study area is about 1,600 ft3/s.
Average Runoff
Runoff in the Dougherty Plain area varies from year to year and from month to month (fig. 9). Average runoff for selected basins was estimated by using runoff data from eight stations operated during a common period when average climatic conditions are believed to have been similar to long-term average climatic conditions. The common period selec-
ted was water years 1959 through 1970. Average annual rainfall for seven precipitation stations in the area of study during this period was 52.8 inches; longterm average annual precipitation (generally 1935 to 1979) for the same stations was 53.1 inches. Average runoff for the base period, 1959-70, is considered to be representative of expected basin runoff for periods of average hydrologic and climatic conditions.
The annual mean, early-spring high (Feb.-Apr.), and late-summer low (Sept.Nov.) runoffs for the eight watersheds in the Dougherty Plain were determined for water years 1959-70. The runoff for each watershed (fig. 15) is that runoff measured at the most upstream part of the watershed subtracted from that runoff measured at the most downstream part. Runoff is much less during the summer months, primarily because of extremely high evapotranspiration losses and secondarily because of less rainfall than in the spring months.
Runoff for that part of each watershed within the Dougherty Plain was estimated. The sum (rounded) of these runoffs gives an annual mean runoff of 5,200 ft3/s; a spring high of 9,200 ft 3/s; and a late-summer low of 2, 700 f t 3; s. These quantities are the approximate total annual, spring high, and late-summer lowwater yields of the Dougherty Plain area under average climatic and hydrologic conditions.
Base flow
The base flow of streams in the Dougherty Plain consists mostly of ground water discharged from aquifers hydraulically connected to the streams. Therefore, base flow can provide an estimate of the perennial ground-water yield of watersheds in the Dougherty Plain.
The base flow of a stream can be estimated by separating the overland flow from total discharge on a streamflow hydrograph by using techniques described by Riggs ( 1963) ._ The base flows of nine streams in the Dougherty Plain were estimated by hydrograph separation techniques
29
Table 7.--Low-flow charac teristics s t selec ted streamflow gaging stations
Sta tion No.
Station name
Period of
Drainage record, Recurrence
area climatic interval
(mi2)
year
(years)
I
Annual low flow, in ft3/s, for indicated consecutive days
7
14
30
60
90
120
!83
43500
Chattahoochee River 8,040 at Columbia, Ala.
44000
Cha ttahoochee Riy7r 8,340 a t Alaga, Ala. -
495 00
Flint River at Montezuma
2,900
49900 Turkey Creek
45
at Byromville
50500
Flint Rive r a t Oakfield
~/
3,860
50600 Kic haf oonee Creek
197
at Preston
50900 Kinc hafoonee Creek
527
near Dawson
51700 Uuc kalee Creek
265
near Smithville
51900 Uuc kalee Creek
405
near Leesburg
52500
Flint River at Albany -11
5,310
53000
Flint River I/ a t Newton -
5,740
53200 Nochaway Creek
52
near Shellman
1930-60
2
5
10
20
30
50
1940-70
2
5
10
20
1932-80
2
5
10
20
30
so
1960-80
2
5
10
20
30
1931-58
2
5
10
20
30
1953-77
2
5
10
20
30
Partial Z/
2
record -
5
10
20
30
Pa r t i a l re c o rd -3
2 5
10
20
30
Partial A
2
record -
5
10
20
30
1903-80
2
5
10
20
30
50
1939-80
2
5
10
20
30
50
p~:~!~! 2
2 5
10
20
30
2, 54 0 1., 89 0 i , 600 1, 3 90 J, 200 1 , 1 80
2,860 2,070 L, 730 1,480 I ,300 L, 230
2,990 2,160 1, 800 1 , 540 1, 360 1,280
3,330 2,360 I, 940 1,640 1,370 1,340
3,780 2, 660 2,170 1,820 1,500 1,480
4,300 3,060 2,510 2,110 1,800 I , 720
4,660 3,390 2,840 2,440 2,100 2,,050
2, 42 0 1, 760 1, 450 1, 200
3 , 500 2, 300 1,820 l , 470
3,800 2,660 2,190 1,860
4,230 3,070 2,570 2,500
4,780 3,460 2,900 2,500
5,130 3,800 3,260 2,890
5,600 4,230 3,650 3,230
950
99 0
1, 050 1,160 .I ,330 1,470 1,620
75 0
800
830
910 l ,030 1,140 1,280
670
720
750
800
910 1,000 1,130
603
650
680
730
810
890 1,020
600
630
620
660
740
800
920
540
590
620
650
720
90
910
5.2
5.4
5.4
6. 0
7.0
7.9
8.6
3.3
4.1
4.3
5. 0
6.2
6.3
7.4
2.0
2.8
3.5
4. 2
5.1
5.2
6.2
.86
1. 6
2.4
3.2
3.6
4.2
4.8
.52
.88
1.9
2. 0
2.8
3.5
4.0
430
1,300
1,400 1,500 l '700 1,800 2,000
260
910
1, 050 1 , 100 1,200 1,400 1,600
200
760
920
960 1,000 1,200 1,400
170
660
840
870
910 1,000 1,200
ISO
610
800
830
850
950 I, 100
48
52
57
65
77
89
98
32
35
39
45
54
64
71
25
28
31
37
45
53
59
21
23
26
31
39
46
51
19
21
24
20
38
41
47
------
170 110
90 84
----
--
190 130 110 97
210 160 130 110
250 170 140 120
-----
77
--
90
100
110
--
------
-
73 46 38
----
32
--
29
-
88 60 50 45 42
110 72 58 48 44
120 86 74 68 67
------
-----
-
160
-
110
-
89
-
80
-
76
-
170 130
200 160
250 165
--
110
130
135
-
95 89
110 97
120 110
--
1,080 720 590 500 450 410
1,670 1,260 1,090
960 920 840
1, 820 l ,400 1, 220 ..1,090 1,000
96 0
2,000 1,550 1,370 1,240 1,200 I, 110
2,310 1,790 1,580 1,430 1,400 1,280
2,580 2, <l00 J, 770 1,610 1, 500 1, 460
2,850 2,210 I, 960 1,780 1,700 1,600
1,650 1,330 1,180 1,070 1,000
950
--
-
--
2,160 1,780 1,600 1,470 1,360 1,340
26 19 16 15 14
2,310 1,910 1,740 1,610 1,500 1,480
29 22 20 19 18
2,440 2,020 1,850 1,730 1,700 1,620
----
---
2, 690 2, 250 2, 080 1, 960 1,860 1, 850
33 26 22 20 19
2,910 2,440 2,270 2,150 2,100 2,055
35 29 26 25 24
3,160 2,670 2,480 2,350 2,300 2,230
----
-
5,480 4,070 3,460 3,020 2,600 2, 590
b,640 5,290 3, 730 4,340
1,870 1,500 1,330 1,210 1,100 1,090
15 8.2 7.0 5.0 4. 2
2, 400 1, 700 1, 500 1,300 1,250
120 84 70 60 56
---------------------------
3,460 2,560 2,270 2,070 1,950 1,870
3,660 3,130 2,930 2,790 2,700 2,650
------
--
30
Table 7.--Low-flow characteristics at selected streamflow gaging stations--Continued
Station No.
Station name
Drainage
area (mi2)
Period of record,
climatic year
Recurrence interval (years)
Annual low flow, in ft3/s, for indicated consecutive days
1
7
14
30
60
90
120
183
53400
Pachltla Creek near Edison
53500
Ichawaynochaway Creek at Milford
188
1961-69
620
1941-80
54500 Chickasawhatchee
320
Creek at Montezuma
55000
Ichawaynochaway Creek near NeWton
1,020
56000
Flint River at Bainbridge
~/
7,570
1940-49 1939-47 1909-71
57000
Spring Creek near Iron City
485
1938-71
58000
Apalachicola River at Chattahoochee,
Fla.~/, if
17' 200
1930-53
58000
do. ~1. !../ 17' 200
1959-80
2 5 10
75 53 43
82 58 47
88 61 49
94 --
66 55
--
---
---
-
--
2
220
240
560
290
330
370
390
450
5
170
180
190
210
250
270
290
340
10
140
150
165
180
210
230
250
290
20
130
140
150
160
180
200
220
260
30
120
120
140
150
160
180
200
230
50
110
118
130
140
155
170
190
220
2
14
16
17
22
25
39
45
61
5
5.6
6.6
7.8
11
15
16
20
30
10
3.7
4.6
4.6
7.2
10
11
14
21
2
320
330
350
370
430
500
550
640
5
240
250
270
290
340
380
400
460
10
210
230
240
260
300
320
340
390
2
2,840
3,270
3,420 3,600 3,930 4,220 4,490 4,970
5
2,250
2,620
2,740 2,850 3,110 3,320 3,540 3,900
10
2, 010
2, 340
2,450 2,550 2,780 2,970 3,150 3,460
20
1, 840
2, 140
2,240 2,340 2,550 2, 720 2,880 3,150
30
1,800
2,100
2,200 2,200 2,450 2,500 2,700 2,900
50
1, 680
1,950
2,040 2,130 2,320 2,490 2,600 2,850
2
52
58
62
67
80
100
120
150
5
27
29
30
34
41
46
56
82
10
17
18
19
21
25
30
35
52
20
11
12
13
14
16
20
23
33
30
8.4
9.0
10
11
12
15
18
26
50
6.0
6.5
7.5
8.0
8.8
11
l3
19
2
7,940
8,420
8,630 9,030 9,860 10,.700 11,400 12,700
5
6,570
6,860
6,970 7, 200 7,800 8,350 9,030 10,100
10
5,960
6,160
6,230 6,400 6,900 7,330 8,050 8,990
20
5,500
5,620
5,680 5, 800 6,240 6,580 7,340 8,200
30
5,200
5,400
5,600 5, 700 5,850 5,900 6,400 7,700
2
9,470
9,800 10,000 10,400 10,800 11,400 12,200 13,600
5
8,030
8,380
8,540 8, 860 9,190 9,640 10,300 ll '600
10
7,320
7,735
7,900 8,200 8,600 9,040 9,540 11,000
20
6,760
7,250
7,430 7. 730 8,230 8 , 670 9,070 10,500
30
6,000
6,600
6,800 7,000 7,400 8 , 000 8,400 10,000
1/ Affected by regulation. 2/ Based on correlation of 14 independent base-flow measurements with concurrent base flows at gaging station 53500. 3/ Based on correlation of 10 independent base-flow measurements with concurrent base flows at gaging station 50600. 4/ Based on correlation of 12 independent base-flow measurements with concurrent base flows at gaging station 53500. 51 Based on correlation of 10 independent base-flow measurements with concurrent base flows at gaging station 50600. 6! Prior to construction and filling of Jim Woodruff Dam.
!} After construction and filling of Jim Woodruff Dam.
31
30
Figure
0
5
~L.!
10
-~ -------~---
'IS -
2:0
30MilEs
-
--
14 -Distribut ion of 7-day, 10-y- ea'r mm 1mum annual flows
32
AREA OF DOUGHERTY PLAIN
WATERSHED BOUNDARY
""28
CONTINUOUS STATION AND
-I~EE~T~~~~
GAGING
I I ~TAL (540)(720)(450)
ATION NUMBER SLaplrein-RsgUuhmNigOmhFeFr -lIonwm1. lllon gall ons par day
Annual mean
-
w............
"30 MILES
Figure 15.-Distribution and range of annual mean and seasonal runoff.
33
and compared with discharges at the 50percent flow duration or median flow value. (See table 8.) Figure 16 illustrates the results of this comparison. The upper curve is a plot of median flow for a year of record plotted against base flow for that particular year determined from hydrograph separation techniques. The lower curve shows monthly median flow for a month of record plotted against monthly base runoff for that particular month determined from hydrograph separation techniques. The points on both curves plot on or close to the line of equality, indicating that there is a reasonable agreement between base flow determined from hydrograph separation techniques and base flow determined from median flow. The yearly plot, however, more closely approximates the line of equality than does the monthly plot. Consequently, estimation of yearly mean base flow from yearly median flow is probably more valid than estimating monthly base flow from monthly median flow.
The relation between base flow and median flow illustrated in figure 16 provides a means of estimating base flow in the Dougherty Plain without using timeconsuming hydrograph separation methods.
Median flow at gaging stations can easily be calculated by standard U.S. Geological Survey programs, using daily-flow values.
Using the above calculated relation between base flow and median flow, annual mean base flow and seasonal ranges of base flow were estimated for the eight watersheds in the Dougherty Plain area (fig. 17).
Weighting the above data on the basis of watershed area within the Dougherty Plain area gives an annual mean base flow of 4,000 ft3/s; a late-summer (Sept.-Nov.) mean base flow of 2,300 ft3fs; and an early-spring (Feb.-Apr.) mean base flow of 7,400 ft3/s. The method used herein to estimate base flow is considered subject to greater error for high flows than for low flows and the actual early-spring base flow may be much lower than estimates made from either hydrograph spearation techniques or
median flow (T. W. Hale, u.s. Geological
Survey, oral commun., 1981).
Ground Water
The primary geohydrologic units of interest in this investigation are, in decending vertical order, (1) the residuum; (2) the Ocala Limestone, called the principal artesian aquifer; and (3) the Lisbon Formation, which hydraulically separates the principal artesian aquifer from underlying sediments. Figure 18 shows the stratigraphic position and thickness of these units, selected geophysical logs, and a summary of waterbearing and lithologic characteristics determined from test well 205-37 in Mitchell County east of Newton.
Residuum
Hydraulic properties
The hydraulic conductivity of the residuum has been estimated from sieve analyses of drill cuttings collected at 5-foot intervals, geophysical logs, and aquifer tests. (See table 9.) The areal distribution of estimated hydraulic conductivity and transmissivity (estimated from hydraulic conductivity and average saturated thickness) are shown in figure 19.
Estimated vertical hydraulic conductivity varies from 0.0001 ft/d to 9 ft/d, with the median being 0.003 ft/d. Estimated horizontal hyaraulic conductivity varies from 0.0004 ft/d to 30 ft/d, with the median being 0.02 ft/d.
Estimates of transmissivity values range from 0.002 ft2/d to 1,000 ft2/d, with the median being 0. 3 ft 2jd (table
9). An average value of saturated thick-
ness based on observed seasonal waterlevel changes at each well was used to calculate transmissivity. Consequently, the estimated transmissivity values represent average conditions only.
The predominant lithologic factor determining transmissivity, however, is the presence or absence of permeable sand lenses within the saturated residuum thickness. Test drilling indicates that such sand lenses occur more commonly in the upper half of the residuum than in the lower half. Consequently, transmis-
34
Table 8 .. --Base flow estimated from hydrograph separation and median flow [A, hydrograph separation technique of Riggs (1963); B, median flow]
Station Water
No.
year llethod
Oct.
Nov.
Dec.
Estimated monthly mean base runoff, in ft3/s
Jan .
Feb .
Har .
Apr .
Hay
June
July
Aug . Sept .
Est. annual
base runoff (ft3/s)
Annual mean runoff
(ft3/s)
49500 1978
A
1, 200 2,030
2,180
3, 570 4, 260 3,670 2, 960 2, 820 I, 360 1 ,030 I, 490
919
B
1,280 2, 250
2,540
4,810 3, 700 4, 340 3,010 3, 980 I , 940 1,140 1,830
982
1979
A
839
945
I, 620
2,860 3,041 5,025 4,850 2,630 1, 1,1,0 1, 320
966 1,160
B
318
986
I, 8 10
3,160 4,270 5,050 9,180 3, 260 1, 520 L, 51,0 1,040 1,860
1980
A
1, 380 1,610
1, 950
2, 290 3,100 6, 790 6,330 2,100 1,860 I , 300
985
955
B
1,410 1,840
2, 020
2, 360 3, 870 9,820 7,040 3, 240 2 , 160 I , 270 1,070 1,040
2,290 2, 530 2,220 2,070 2,550 2,280
3,608 3,428 3, 719
49900 1978
A
B
1979
A
B
1970
A
B
6.2
7.0
13
66
84
66
33
33
10
7. 4
7.9
6. 0
6.3
7.6
20
60
84
85
36
42
8.3
8.3
10
6. 3
5. 6
6.0
8 . 1,
23
39
58
40
38
9.6
9. 8
5. 2
5. 7
5.5
6.0
9. 0
17
57
67
57
40
9. 5
13
5. 2
7. 2
8 .4
11
18
23
32
96
77
15
6. 3
4. 6
4.0
5.1
8.1
15
20
19
35
131
83
16
7. 6
4.6
6.9
5.4
28
l7
so. 9
21
13
42.8
25
17
38.7
50600 1956
A
37
68
80
80
156
!56
122
50
36
52
32
20
B
42
69
83
77
225
233
ISO
58
40
66
34
38
1957
A
69
74
122
129
116
153
188
167 116
57
44
48
B
86
77
145
ISO
130
194
242
260 145
77
50
68
1958
A
104
139
187
168
165
268
259
108
68
119
107
62
B
96
210
210
217
202
290
300
120
89
180
135
66
74
74
121
107
125
177
146
187
219
52500 1978
A
1, 890 2,140
2, 860
7,700 7,810 6, 270 4,470 4,670 3,090 1,830 2, 760 1,330
B
2, 230 4,240
4, 560
8,010 7. 910 7,920 5, 960 7, 350 3,1,70 1, 990 3,610 1,1,60
1979
A
1,370 1, 240
2,130
3,130 5. 780 8,600 7. 470 4, 390 1,890 2, 270 1,520 1, 880
B
I, 750 I, 750
2, 800
5, 500 8, 740 9,720 15,800 5,120 2, 710 3,450 2, 070 2,880
1980
A
2, 510 2, 210
2, 250
3,140 5,600 11,600 10,900 4,050 2,600 I, 760 I, 360 I, 300
B
2,740 3, 270
3,150
4, 220 6,460 17.400 14,000 5, 370 3, 210 1,970 1, 660 1, 430
3,900 4 , 550 3,470 3,510 1,,110 3,800
6,1,83 5, 81,4 6,251
53000 1978
A
1, 850 2, 590
3,130
6, 530 6,790 8 ,1 00 5, 200 1,,100 3,890 2, 500 2, 970 I, 760
B
2, 630 4,320
5,060
7. 300 1.1,550 8, 740 6, 760 7. 990 4,440 2,890 ' ,1 50 I , 970
1979
A
I,, 670 I, 650
2, 420
3, 570 4, 910 9,140 7,690 5, 340 2, 780 3,120 2 , 130 2,080
B
Z, 040 2, 060
3, 020
a, 5, 380
s7o 10,000 14,300 6,650 3, 690
4,230
2 ,680
3,3 50
1980
A
2, 980 2,800
3,030
3, 710 5,480 11, 300 10,800 4,470 3,000 2, 260 1,890 1, 710
B
3, 380 3,620
3, 790
4, 640 6,830 17.900 14,900 6, 390 4,280 2,680 2,180 1,880
4,120 5,110 3,880 4, 230 4,450 4,360
6,693 6,077 6,579
53500 1978
A
37 2
479
631
861 1,150 1, 470
825
650 446
326
385
254
B
398
544
751
784 I, 150 1, 480
836
823 472
360
441
261
1979
A
245
280
400
632
786
900
621
450 261
356
232
260
B
248
298
450
788
964
998
938
615 348
504
269
333
1980
A
320
334
433
507
554 1, 130 1,190
540 281
234
146
184
B
321
390
475
530
618 2,130 1, 420
677 390
362
212
224
654
653
964
452
485
666
488
485
734
55000 1940
A
B
1941
A
B
1942
A
B
635
500
585
515
310
525
334
655
351,
325
428
353
610
876 1,820 1,220 I, 380
660 603 1,190
610
397
58 5
I, 300 2, 970 I, 650 I, 650
700 700 1,1,70
820
375
607
913
834 1,010
860
370 257
453
284
233
695
1,000 1,000 1,470
865
410 277
518
372
277
680
I, 500 I, 500 1,830 1,140
581 543
748
789
405
728
1,no 1,460 3,080 1,420
636 747
785 1,020
544
875
865 I, 211
555
55l
690
866
785 1,230
56000 1908
A
6,100 4,500 12,500 18,400 17,600 13,000 12,400 15,000 7,000 6,570 5, 280 6,030
B
6,300 4, 970 19,100 23,400 26,100 13 , 700 16,500 13,700 8,030 7,790 6,400 6,100
1909
A
4, 700 5,000
5,200
5, 700 7,21,0 8, 890 9,062 7. 784 5, 300 5, 590 5, 230 3, 870
B
4,~90
5, 290
5,290
5, 900 10, BOO l b ,OOO 10,100 8,670 6, 300 5,810 6,400 3, 900
1910
A
3, 600 3, 500
4,000
4, 500 5, 560 7,000 4, 790 1,,460 4,000 5,380 4,270 3,640
B
3,790 3, 660
4,240
4, 520 6,400 7,910 5,050 5,400 4,800 6,730 4,810 1,,300
10,400 10,900 6,130
6,000 4,560 4,800
14,055 8,128 5, 655
57000 1940
A
229
16 5
120
334 I, 110
646
489
180 161
300
143
58
u
346
177
!59
475 I, 680
810
574
213 213
1114
135
55
1941
A
34
51
86
22 8
253
395
362
97
61
102
96
55
II
38
82
89
30 9
286
607
376
125
57
92
108
57
1942
A
76
62
64
59 8
855 I, 360
67 4
266 448
281
300
!57
ll
96
68
100
835
635 1,880
775
334 695
408
408
201,
"
328
264
480
152
108
191
428
394
603
35
0 z u0w 10 -
(Jl
wCl:: a...
ww1-
lL
u
[D
:u:J
lL
0
(Jl
0cz:r
(Jl
::J 0 I 1-
z
~ 0
_J lL
z c:r
0 w
:2:
Line of equality ~
Yearly median flow versus yearly base flow determined from hyd rograph -separation techniques
0
z
u0w
(Jl
Cl::
w
a...
1w w
lL
u
[D
u::J
lL 0
(Jl
0 z c:r
(Jl
::J 0 I 1-
z
~
Line of equall ly
~
0
_J
lL
zc:r
Monthly median flow versus monthly base flow determined from hydrographseparation techniques
0 w
:2:
O.OOf.~__JL_L_J_LUU~--_j__L_J_LLULL-__JL_L_J_LLU~--_L_L_LLU~L-__JL_L_J_LLU~
0.001
0.01
0.1
I
10
100
BASE FLOW FROM HYDROGAPH SEPARATION, IN THOUSANDS OF FEET PER SECOND
Figure 16.-Relation between base flow estimated from hydrograph separation and median flow.
36
Figure 17.-Distribution and range of annual mean an d seasonal base flows .
37
GULF GROUP AND
SERIES COAST FORMATION
STAGE
~~ -c ~
.2
-~
~
t.~3
Residuum
CL
GAMMA-RAY
NEUTRON -POROSITY
O I
LITHOLOGY
WATER-BEARING CHARACTERISTICS
Variegated brownish-red to creanrcolored
clay containing poorly sorted coarse to
fine-grained subangular to subrounded
quartz sand. Contains some residual limestone and chert nodules
Not generally used as an aquifer
"c '
"u0'
UJ
a;
Q Q
c
"o:'
0en
.""u.".,'
:::>
"c '
0 (ij
E"'
::::i
s= ~
u"'
0
~---::::f.---------- 200'
I..W C1J
- -r ,......,.
!
White to light pink, very fossiliferous
Principal artesian aquifer in
limestone , saccharoidal and extensively southwest Georgia. Well yields
calcitized. Upper zone fractured and
range from less than 500
cavernous, becoming more dense at depth. gallons per minute to more than
Contains lenses of sand and clay.
1000 gallons per minute
Dolomitization common in some areas.
depending on the thickness of
I : ~~~;_:J
the limestone section.
c
.2
""uc0''
UJ
'"0'
"0
~
c
o":'
0 D;;; 0
0. :::> 0
c'5
E"'
,0e
0"'
~
0 LJ._
c
0 Den
::::i
Tallahatta Formation
3001
Radiation increases - - - - ---'""
Porosity decreases -----7
I "=1 Medium gray, moderately indurated, very sandy, fossiliferous limestone interbedded- Limited water-bearing potential-
wfth calcitic sandstone. Sand grains are used only in multiaquifer wells
medium to tine-grained, subangular to
where other aquifers are tapped.
subrounded quartz.
A major aquifer in the Albany area
Fine-grained well-sorted unindurated quartz
Little used and comparatively
sand. Contains phosphate and glauconite. unknown in other areas of
southwest Georgia.
Figure lB.-Stratigraphic section, geophysical logs, and water-bearing characteristic of geohydrologic units near Newton,
test well 205-37.
Table 9. --Hydraulic and water-level data for residuum test wells [Water levels measured January 1980-September 1981]
Well No.
Well name
Estimated average hydraulic
conductivity (ft/d)
Vertical Horizontal
Ratio of average horizontal
to vertical hydraulic
conductivity
Estimated trans-
missivity (ft2/d)
Residuum water levels, in ft
below land surface
Msx. Min. Average
Residuum thicknes s
(ft)
Saturated thickness
(ft)
t1ax. Hin. Average
007-38 T. Rentz RW
0.08
007-39 Jo-Su-Li RW
.0003
037-24 B. Jordan TW1 087-44 DP 6
.0005
.oos
087-45 J. Hall TW2
.002
087-46 G. Bolton TW2
.006
087-47 A. Newton
.2
095-69 School Bus
Road TW1
.001
095-70 Game and Fish TW1
.004
095-71 Nilo TW3
.2
095-72 USMC Supply TW1 .004
099-45 I. Newberry TW2 .001
099-46 v. Evans TW1
.003
177-40 Piedmont Plant Farm TW1
177-41 s. Stocks TW1
.003 9
177-42 B. King TW1
.0009
177-43 H. Usry TW1
.0002
201-16 DP 3
.0005
201-33 J. Fleet TW2
.01
205-34 H. Meinders TW2 .01
205-35 C. Holton TW2
.003
205-36 H. Davis TW1
.002
20S-39 DP 12
.0005
253-27 Roddenberry TW2 .0005
253-28 D. Harvey TW2
.002
261-22 E. Stephens TW1 .0009
273-14 A. Vann TW1
.0001
321-05 DP 9
.OS
321-09 C. Odom TW1
.0007
20
. oos
7
6 .I 10
.002
.009 4
s
. 006 .006
. 02 30
01 .0006
. 02 5 .4 .02
.7 .003 .003 .0004
. 0006
300 20
10,000
200
.I
100
Dry Dry 23.4
Dry Dry 20.8
Dry Dry 21.6
300 20
so
2
2 20 1,000
6 2
3 10
3
2
so
100 10
10
Dry Dry
Dry
24.9 19. s 22.7
300
Dry Dry
Dry
. 02 29.5 17.S 26.7
.t 50 400
.2 .2
12.0 38.4 28.0 21.8
9.0
10.1 35.S 23.3
7.6 6.2
11.4 36.4 25.2 14.3
7.7
.3 1, 000
.02
.3 10 10
.3
34.5 13.4 16.6 11.5 22.8 32.2 Dry 32.8 32.6
31.4 11.7 8.3
2.3 1.2 22.3 Dry 28.8 23.0
32.9 13.0 11.4
s.o
12.0 26.4 Dry 30.8 27.1
1,000 2 3 4
10
27.4
t 28.1
07 20.1
.002 16.8
19.5 18.4 6.6 11.3
22.8 22.3 11.6 15.0
9
.2
8. 2 6. 4
7. 2
21
29
37
16. 2 13 . 6 1S.4
54
40
32
12.S 7.1 9.3
53
35
17.S s.s 8.3
19
8.9 7.0 7.6
50
14.5 11.6 13.6
107
83.7 79.0 81.8
40
32.4 18.2 25.7
46
39.8 37.0 38.3
47
15.6 12.5 14.1
50
38.3 36.6 37.0
24
15.7 7.4 12.6
34
31.7 22.5 29.0
55
41
13.7 8.8 14.6
59
60
21.2 27.2 29.2
40
17.0 7.4 12.9
40
39
19.5 11.6 16.2
54
35.6 25.9 31.7
34
27.4 13.9 22.4
20
8.7 3.2 5.0
40
43
36.6 34.8 35.8
39
84 ..00'
>t>'
D
EXPLANATION AREA OF DOUGHERTY PLAIN
0s.2(o4)
DATA POINT- Upper lelt number is
estimated vertical hydraulic conductivity, in feet per day; number in parenthesis is estimated ~orizontal hydraulic conductivity, 1n feet per day. Lower number is estimated transmissivity, in feet squared per day, using_ average saturated thickness given in table 9
rrcam us 8CIIUI
f!,(l OQieol Sat.,.;
f-: 2~0,000 qaocf, On'i!lat
RI0 A
0
~
10
15
20
25
30 MILES
~~~--~--~--_J
Figure 19 -Distribution of estimated vertical and horizontal hydraulic conductivity and transmissivity of the residuum .
40
sivities may increase greatly during periods of high water levels as the permeable sand lenses in the upper half of the residuum become saturated.
Figure 20 illustrates a generalization of the areal range of leakance coefficients of the residuum. Preliminary leakance values were calculated by dividing estimated residuum vertical hydraulic conductivity (k') by residuum confining bed thickness (b'), which is considered to be equivalent to the bottom half of the residuum. The point values presented in figure 20 are considered to be accurate within an order of magnitude. But, because of the paucity and variability of leakance data, the regionalization is highly generalized. Regionalization of the data has, in part, been estimated using digital modeling techniques that will be discussed later in this report.
Small quantities of water are obtained from some residuum wells throughout the study area. As expected, yields are highly variable, ranging from generally less than 1 gal/min to, in a few places, as much as 50 gal/min. During drought conditions or toward the end of periods of low rainfall, residuum wells may go dry as the water table falls below the bottom of the well.
Water levels
Continuous water-level recorders were installed on four of the 29 residuum test wells drilled for the Dougherty Plain investigation, and water levels were measured 2 or 3 times monthly in the remaining 25 wells. The data available indicate that water levels resp,ond in a subdued manner to rainfall and are highest during March-April and decline to their lowest values during NovemberJanuary (fig. 21). Late spring and summer rains seem to have little effect on residuum water levels, probably because most of this rainfall either replaces soil moisture in the unsaturated zone or is lost to evapotranspiration before the water can percolate down through the sandy clay to the saturated zone.
Water levels in 21 wells ranged from about 1 to 38 ft below land surface from January 1980 to September 1981 (table 9). Water-level fluctuations in individual wells ranged from about 2 to 14 ft, with an average fluctuation for all wells of 6 ft.
A generalized map of the altitude of the water table in the residuum for estimated average yearly levels is shown in figure 22. Where the residuum is relatively thick and impermeable, the water table is believed to be a subdued replica of the topography; where the residuum is relatively thin and permeable, the water table is believed to be a subdued but higher replica of the potentiometric surface of the principal artesian aquifer. Relatively steep water-table gradients are believed to adjoin the major stream courses, and relatively low water-table gradients occur in the interstream areas.
Principal Artesian Aquifer
Within the study area, the principal artesian aquifer consists primarily of the Ocala Limestone of late Eocene age. In other parts of Georgia and in Florida and South Carolina, rocks of younger age comprise the upper part of the principal artesian aquifer (Stringfield, 1966). The principal artesian aquifer is the primary source of water for domestic, irrigation, and public supply use in southwest Georgia.
Hydraulic properties
The capacity of the principal artesian aquifer to store and transmit large quantities of water is due largely to the fractured nature of the Ocala Limestone. Water moving through small fractures or cracks in the limestone has slowly enlarged these fractures, through solution, forming an interconnected labyrinth of subterranean channels, giving the rock a high permeability.
Figure 23 shows the distribution of transmissivity in the principal artesian aquifer. The control points are trans-
41
30'-
EXPLANATION
C J AREA OF DOUGHERTY PLAIN
REGIONALIZED BOUNDARY
.0003
REGIONALIZED LEAKANCE-Number ts estimated average leakance in feet per day per foot, per bounded area
.0002 DATA POINT-Number is leakance in leet
per day per tooL To convert to gallons per day per cubic feel multiply by 7.48
Oau ro.tn
11 z..so.ooo
\,1 s II01oglool
qvodronQf~t.
Sl.ur,>Jey
10
Ul
::!0
!! !I
.\0 MILES
Figure 20 .-Distribution of estimated leakance based on test -well data and digital modeling analyses.
42
0
5.0
Water level in well 201-16
10
4.0
20
w
(.)
<u.. a:
;::)
en
30
0 z
<
...J
Rainfall at Colquitt
:1:
0
w...J
III 40
1.0
1wwu.-.
z
...J
~
50WUUUUU~~~~LLLLdaaa~~~~~UL~LLLL~~----~~~~~~~L-UD~~~~
W
2Sr-----~----.-----,-----,------.-----.-----.------.-----.-----~----.---~
...J
aw :
1-
< :1:
Water level in well 087-44
1980
Figure 21.-Water levels in residuum wells 087-44 and 201-16 and rainfall at Bainbridge and Colquitt for 1980.
43
JQ'
32"oo'
Jll ' -
E XPL ANATIO N
DOUGH ERTY PL A IN B OUNDAR Y
WATER-TAB LE ALTITUD E- Shows app roximate allitude o f wa ter table, in l ee t above Nationa l Geodet ic Vertical Datum of 1929
-
D 7Dt o 100
200 to 260
c::::J t=J 10o to 150
250 lo 400
0
150 l o 200
120
OAT A POINT -Numbe r is a ltitude o l m easured
wa ter level for 1980-81
R
0 A
OQJ.o rroll'l U.S G~ok:l; r ..:o l S1.1rvty I 2 :10~ 000 l.hJfttl rftiH~.I M
0t._,_,_ , ,_5,__
_
,10_
_ _
J
_ _
20 _ ,
30MILES
Figure 22.-Generalized altitude of water table in the residuum for mean yearly hydrologic conditions .
44
,..
, - , , - -....r~.~-:....J
i
30'
EXPLANATION
---DOUGHERTY PLAIN BOUNDARY
REGIONALIZED TRANSMISSIVITYIn thousands of feet squared per day
~ Ototo ~ 10to75
c.:.J 75 to 150
. . 150 to 300
370(E) DATA POINT- Number is transmissivity in thousands of feet squared per day. (E) indicates estimated value
B9tijG' rrorn U.S. GeoloV!CGI Sur ~e-; l ~ (l'50,000 (11.1MtGf!Qle5
F L0 R
0 A
0
I-
. tQ
..
30MILES
.1-1
Figure 23 -Distribution of point and regional values of transmissivity in the principal artesian aquifer.
45
missivity data obtained from aquifer tests or estimated from specific-capacity tests. (See tables 10 and 11.)
The transmissivity data have been extrapolated to cover the entire study region. The regionalization is based on aquifer thickness, data average and trends, regional hydraulic gradients, and digital modeling results. The variability of the data is large; thus the point value, while locally accurate, may not be representative of regional transmissivity.
Large interconnected solution channels may account for only a small part of the cross-sectional flow area, but they carry a major part of the flow. Consequently, where wells do not penetrate these solution channels, aquifer tests may not indicate point values as high as the "effective" regional transmissivity. Conversely, a well that penetrates exceptionally large solution channels may not be stressed enough during aquifer testing to yield results representative of regional aquifer characteristics and may indicate a point value that is greater than the regional transmissivity.
Computed point values of transmissivity of the principal artesian aquifer range from 2,000 to 1,300,000 ft 2jd, whereas effective regional values range from 3,000 to 300,000 ft2/d (fig. 23). Transmissivity is lowest in the northern part of the report area, where the aquifer is relatively thin, and increases to the south where the aquifer is thicker. Transmissivity is high near the Chattahoochee and Flint Rivers and Spring Creek, because water moving between the surface-water system and the ground-water system adjacent to these major rechargedischarge areas has accelerated the development of solution channels.
Figure 24 shows the areal distribution of storag~ coefficients computed from aquifer test data. Storage coefficients range from 2 X 1o-4 to 3 X 10-2 , but generally range from 10-3 to 10-4. The storage values indicate that the principal artesian aquifer generally can be considered confined to semiconfined.
Well yields are largely dependent on hydraulic conductivity, length of well
Table 10.--Transmissivitles and storage coefficients for the principal artesian aquifer
Well No .
Casing depth
(ft)
Length of open
hole ( ft)
Aquifer thickness
(ft)
Method of
analysis
Transrnissivity Storage
(ft2/d) coefficient
007-06
os7-16 .Y
79
101
144
325
087-33
88
72
087-48 3_/. }I 120
350
o87-49 .Y
168
408
095-15
095-39 !!./
095-73 2_/
099-26 !!./
63
87
60
240
50
105
099-39
61
64
177-15
64
126
201-05
130
95
205-16
50
140
205-22 205-30 !!._/
77
131
110
140
253-08
63
87
253-12
118
107
253-26
58
67
160
Theis
42,000
0.02
325
do.
80,000
.003
325
Hantush-Jacob
43,000
.001
350
Theis
1,300,000
.002
235
do.
330,000
.001
165
llelayed yield
12,000
.004
260
Uantush-Jacob
130,000
.0004
205
Theis
130,000
.03
80
Han tush-Jacob
41,000
.003
70
Theis
24,000
.0004
140
Han tush-Jacob
43,000
.01
165
Theis
21,000
.001
250
Hantush-Jacob
90,000
.003
260
do.
75,000
.001
235
do.
112,000
.0003
225
Theis
112,000
.001
180
do.
41,000
.0002
75
Delayed yield
27,000
.003
1/ From Sever, 1965a. 2/ From Sever, 1965b.
"J.! Open-hole section extends below base of principal artesian aquifer
and penetrates the Lisbon Formation.
4/ From P. E. Lamoreaux and Associates, written COmiilun., 1980.
"I_! From R. L. Wait, u.s. Geological Survey, written commun., 1957.
open to the aquifer, well efficiency, well diameter, and pump capacity. Measured well yields in the Dougherty Plain area range from about 40 to 1,600 gal/min (table 11). Many wells in the area do not penetrate the full thickness of the aquifer and, consequently, yield less than the maximum possible rate. Yields of 1,000 to 2,000 gal/min, however, are common in areas where transmissivity exceeds 50,000 ft2/d, and yields of more than 2,000 gal/min may be expected where transmissivity exceeds 75,000 ft2/d.
A commonly used measure of well yield is specific capacity. Specific cap~city is defined as the yield per unit of drawdown. Because specific capacity will generally decrease with time as the drawdown increases, the time of pumping prior to the time the drawdown is meas-
46
Table !I.--Specific-capacity data and estimated transmissivities for the principal artesian aquifer
[R, reported]
Well No.
Diameter of well
(in.)
Length of
open hole (ft)
Aquifer thickness
(ft)
Static water level
(ft)
Drawdown (ft)
Duration of
pumping (hrs)
Discharge (gal/min)
Specific capacity [(gal/min)/ft]
Estimated trans-
missivitv ( 1, 000 ft2/d)
007-34
16
108
150
19
ll
8R
1,500
140
45
037-07
12
82
82
32
20
8
570
28
081-13
6
52
110
36
350
50
27
081-17
10
90
150
IS
24
4R
400
17
6
087-25
12
140
277
46
4
8R
800
200
98
087-28
12
100
325
35.97
4.09
017
700
170
86
087-35
12
100
330
22.22
2.78
.43
960
340
87
093-11
10
28
50
115
17
6R
90
5
093-12
10
38
50
97
20
230
12
095-17
10
168
208
41.33
2. 5
!!R
1,000
400
140
095~35
12
124
230
57
6
8R
1,500
250
120
095-45
16
109
180
32.50
1. 45 48
1,400
960
440
095-47
16
100
202
55
1.0
1,000
1,000
460
095-62
16
55
150
17
54
8
210
4
2
095-66
12
64
64
21
71
144
400
6
099-29
16
41
60
17
8R
1,500
300
98
177-14
6
60
138
42
18
12
150
8
4
177-20
4
97
125
26
18
4R
225
12
4
177-44
8
26
75
25
18
4R
210
12
6
205-11
20
100
302
55
3
24
1,600
530
370
205-12
12
186
302
44
3
6
1,500
500
220
205-32
16
156
260
40
8R
1,500
250
100
321-04
4
97
120
.26
3.6
40
11
5
ured should be given. Measured specific capacities of wells in the principal artesian aquifer range from 4 to 1,040 (gal/min)/ft, with respective pumping durations of 8 and 1 hours (table 11).
Water levels
Water-level measurements were made four times in the Dougherty Plain test wells and in about 200 privately owned wells between November 1979 and April 1981. (See Mitchell, 1981, table 47, for a listing of the 1979 and May 1980 measurements.) Maps showing the potentio-
metric surface of the principal artesian aquifer for November 1-5, 1979 (Mitchell, 1981, pl. 2), and May 12-16, 1980 (fig. 25), November 3-7, 1980 (Watson, 1980, p. 2), and March 30-April 3, 1981 (unpublished), were constructed from these measurements. Figure 25 shows the May 1980 potentiometric surface of the principal artesian aquifer, which reflects generally seasonal high levels following late-winter through early-spring recharge. November water levels are generally about 10 ft lower than in May because of seasonal summer declines (fig. 26). Because of a drought beginning in June 1980 and lasting through the summer
47
30'
lC1 v
EXPLANATION
AREA OF DOUGHERTY PLAIN DATA POINT-Number is storage coeff icient
ID A
10
!0
1:0
25
30 MILES
Figure 24. -Distribution of point values of storage coefficients of the principal artesian aquifer.
48
32" oo'
30'
EXPLANATION
[:=:! AREA OF DOUGHERTY PLAIN
- 140-- POTENTIOMETRIC CONTOUR-Show s altitude al which water level would ha ve stood in tightly cased wells Dashed where approximately lo c ated . Contour int erval 10 feet. National Geodetic Verti cal Datum of 1929
DATA POINT
6 13-U fro m u. . tto o ql~ a1 s urvey 1: 25 :0 ~0 00 QUOdi UJ10ifil
I0 A
,.
10
20
!0 MILES
Figure 25.-Potentiometric surface of the principal artesian aquifer, May 1980. From Mitchell (1981) .
49
! () '
EXPLANATION
- - - DOUGHERTY PLAIN BOUNDARY
v
WATER-LEVEL DECLINE BETWEEN MAY AND
NOVEMBER 1980, IN FEET
CJ 010 5
c:.=:=:J 5 10 10
[=:J 10 to 20
. . Greater than 20
F
SaEe lt'orn U.S. Q.~o(I\J,i{jreal SurYCry 11250.000 qu o runn()fe.s
0l,..u..LJ.J~_ _1,0 __ _1.5.__ _2,0__-'-25-----3'0 !.IlLES
Figure 26.- Seasonal water-level declines in the principal artesian aquifer between May and November 1980.
50
of 1981, water levels for March 30-April 3, 1981, which should be about the same as Hay 1980, are generally about 10 ft lower than the May 1980 water levels (fig. 27). The small amount of rainfall between June 1980 and April 1981 was not enough to recharge the aquifer to its normal seasonal high. Water levels remained at about the November 1980 seasonal low.
Figure 28 shows long-term cyclic fluctuations of water levels in two wells open to the principal artesian aquifer. The water level in well 087-23, which is in an area of very high transmissivity, normally fluctuates about 5 ft. The water level in well 095-68, which is in an area of moderate transmissivity, normally fluctuates about 10 ft. The actual fluctuation for any particular year at a particular site depends primarily upon the timing and amount of spring rainfall that recharges the aquifer and the amount of subsequent summer decline. Summer declines are dependent upon natural discharge to streams, evapotranspiration rates, and, to a lesser degree, summer rainfall and pumpage.
Lisbon Formation
Because it has relatively low transmissivity compared with the principal artesian aquifer (Watson, 1981), the top of the Lisbon is considered to be the base of the principal artesian aquifer in the area of investigation. Although domestic supplies of water may be obtainable from the Lisbon south and east of the study area, no wells within the Dougherty Plain are known to yield more than a few gallons per minute.
Recharge, Discharge, and Flow Characteristics
Annual mean recharge of about 2,800 Mgal/d to the residuum occurs chiefly from rainfall during January through May (fig. 21). Rainfall that is not evaporated, transpired, retained in the unsaturated zone as soil moisture, or dis-
charged to streams, moves downward through the residuum to recharge the principal artesian aquifer. Most rainfall occurring during the summer months is lost to evapotranspiration or is retained as soil moisture in the unsaturated zone of the residuum. Consequently, little, if any, summer rainfall infiltrates to the water table or the principal artesian aquifer (fig. 29).
The vertical hydraulic conductivity of the residuum confining zone is generally low--about 0.003 ft/d (table 9). Within the Dougherty Plain, however, the cross-sectional area of flow in the vertical direction is large--about 4,400 mi2--and consequently large quantities of water are transmitted through the residuum confining zone to the principal artesian aguifer. Digital modeling results indicate that annual mean recharge to the artesian aquifer is about 2,200 Mgal/d (10 in.), whereas late-summer recharge is 1,400 Hgal/d (6 in.).
Recharge to the principal artesian aquifer varies considerably with location because of the highly variable leakance of the residuum (fig. 20). For example, digital modeling results indicate that recharge varies from about 0.1 to 2 (Mgal/d)/mi2.
The principal artesian aquifer transmits water from interstream areas of recharge to natural areas of discharge and to wells. Natural outlets include springs, streams, and the overlying residuum or underlying Lisbon Formation, where hydrostatic pressure in them is less than in the principal artesian aquifer.
Hydrograph separation techniques (discussed previously) indicate that annual mean ground-water discharge to streams from the residuum and the principal artesian aquifer is about 2,600 Mgal/d, and late-summer mean discharge is about 1,500 Hgal/d. Additionally, annual mean discharge to wells from the principal artesian aquifer is about 225 Mgal/d (210 Mgal/d for irrigation and 15 Mgal/d for all other). As with recharge, discharge varies considerably with both areal location in the Dougherty Plain and time of year. (See section on Base Flow.)
51
30'
Hl-
EXPLANATION
- - - DOUGHERTY PLAIN BOUNDARY
v
WATER LEVEL DECLINE BETWEEN MAY
u1980 AND APRIL 1981 , IN FEET OloS
~ 5to10
CJ 10 to 20
. . Grea1er than 20
0 R
D A
l) gu hom U. S. C~ e~ l o_gl c;o l Sur..,e)' 1 : 2~0; o go quo dtotiQI I!I:I
10 I
" I
20
30MILES
Figure 27-Difference in principal artesian aquifer water levels between May 1980 and April
1981.
52
35
'
I
Well 087-23
40
w u
45~~ <{
LL
a:
:::) (j)
r\
I \ (\
I
\ rNo .... ........r..e..c._ord
z 0
<{
_J
s
0
_J
w co
55
wfw-
LL
z
Uw 1
- 25
_J
w
> w
_J
aw: 30
f-
s<(
35
40
I
0 w
0
ci
<t :2
z w =>
~
Cw L
J (f)
I
0 w
0
1971
.
a: <t
=zw>
1-'
CL
w
:2 "") (f)
I,
0
w
0
a: Wz <t =>
:2 J
1-' I
CL 0
w w
(f) 0
1972
1973
a: Wz <t =>
:2 J
f-'
Cw L
(f)
I
wo'
0
a:' <t 2
1974
w t-'
Z
=>
J
Cw L
(f)
1,
0 w
0
.
a:
z w
<t =>
2 J
t-' Cw L
(f)
I
o'
w
0
1975
1976
a:
<t
w=z>
1-' CL
w
I
o'
w
:2 J (f) 0
1977
a:
:<2t
=zw->,
t-'
0...
w
(f)
'
w 0
0
1978
~ @ ci
~
w
z =>
t-'
CL
w
o' '
w
:2 J (f) 0
w
z
~
CL
=->,
w
(f)
0
1979
1980
Figure 28.- Fluctuations of mean month I y water levels in the principal artesian aquifer at wells 087-23 arld 095-68.
8
I
I
1w-W o
We( LL.u.
"'~:-:aI: 20
.w. Jzo >~..<J 24
a::=
~ ~ <w
28
3:m
32
36 6
1.1'1
~
5
"w '
:r
4
0
~
~ 3
.i .<.J
LL.
~ 2
<
a:
1979 total rainfall 54.28 inches
1980 total rainfall 46.07 inches
1981 total rainfall 43.07 inches
Rainfall at Albany
/
.. o fJ1?.~_-!:I P..~ QP:L.!~.~--~pl111_fjd_~.tcr.n.!!f!__ ['!g_ ~~~_o_ ~J~#fM!:.!] ~ ~ ~tl!J1J~..:!11: ,:::11 ~.!1?1'?-~ ,.~ ...kP.P..f:l'l~L~ , ~ ~- I@1
Figure 2 9.-Fluctuations of mean daily water levels in the principal artesian aquifer at wells 095-59 and 205-16 and 5-day rainfall totals at Albany and Camilla.
Long-term water-level records of the principal artesian aquifer indicate that except for cyclic seasonal fluctuations and hydrologic extremes, the potentiometric surface of the principal artesian aquifer has remained fairly constant. This implies that over the long term, aquifer storage changes have been minimal and recharge approximately equaled discharge.
The potentiometric map of the principal artesian aquifer (fig. 25) illustrates several hydrogeologic characteristics of the Dougherty Plain area. Regional ground-water flow direction within the principal artesian aquifer is from the northern part of the area southward toward Lake Seminole. The shape of the potentiometric contours indicates, however, that major streams are principal areas of ground-water discharge. Baserunoff analyses and digital modeling results indicate that about 90 percent of the annual ground-water discharge occurs
as discharge to streams and springs.
Whereas the potentiometric contour map may be used to estimate the general direction of ground-water flow, the actual movement of a single water molecule may be very complex and may differ greatly from that which is implied by the two-dimensional potentiometric map. The actual flow of ground water is three dimensional and is affected not only by hydraulic gradient, but also by changes in aquifer properties (such as permeability, porosity, and thickness). Furthermore, aquifer properties of a limestone vary widely, depending upon the hydrogeologic characteristics of the aquifer.
White (1969) proposed a three-part classification of carbonate aquifers based upon recognizable physical features: (1) a diffuse-flow solutional modification; (2) a free-flow aquifer in which ground-water flow paths have been localized by solutional modification into well integrated systems of conduits; and (3) a confined-flow aquifer in which geologic boundaries rather than hydraulics are the flow-limiting factors. Groundwater movement through the diffuse--flow system is analagous to flow in a homogeneous aquifer and more nearly follows the
"basic" assumptions upon which groundwater flow equations are based. In a free-flow system, flow occurs in distinct conduits or channels, while nearby rock may have little porosity or permeability. Flows in these conduits often have high velocities and may be turbulent.
The principal artesian aquifer generally functions as a free-flow system. However; it may function as a confined diffuse-flow aquifer where the surface water-ground water interaction is slight. Consequently, flow equations that assume laminar flow in an isotropic and homogeneous medium cannot be rigorously applied to the principal artesian aquifer. Nevertheless, if the limitations of basic flow equations as regards a particular set of geohydrologic conditions are considered, these flow equations used in conjunction with potentiometric maps may be used to indicate the general direction and average velocity of ground-water flow.
The average velocity of ground-water flow may be computed by the following equation:
--K- -d'h-/d-l' Q
(Lohman, 1979)
where
v = average velocity, in feet per
day
K lateral hydraulic conductivity,
in feet per day,
dh/dl change in head with respect to
change in distance, in feet
per foot,
Q porosity, as a decimal fraction,
and
the minus sign indicates that
flow is in the direction of
decreasing head.
It must be stressed that the solu-
tion of this equation is the average
velocity, and may not correspond to the
actual velocity of a discrete unit of
water between any two points in the aqui-
fer. Actual ground-water velocity may be
more or less than this average value,
depending upon the flow path followed and
local geohydrologic conditions.
Because the principal artesian aqui-
fer acts as both a free-flow and a
diffuse-flow system, average velocities
55
of ground-water flow vary greatly. The effective hydraulic conductivity of the upper part of the principal artesian aquifer adjacent to the Flint River above Bainbridge is believed to exceed 1,000 ft/d as a result of secondary solution (free-flow system). The effective porosity of this part of the aquifer is assumed to be about 20 percent, and the hydraulic gradient is about 3 ft/mi (fig. 25). Using the preceeding equation, the average velocity of ground-water flow is about 3 ft/d.
The effective hydraulic conductivity of the aquifer in the northern part of the study area, away from streams, is about 100 ft/d, based on data shown in figures 6 and 23 (diffuse-flow system). The effective porosity is estimated to be about 20 percent, and the hydraulic gradient is about. 2 ft/mi (fig. 25). Thus, the average velocity of groundwater flow in this area is about 0.2 ft/d.
Water Budget
Very little ground-water development from the principal artesian aquifer has taken place in the Dougherty Plain except for irrigation purposes. During average hydrologic conditions, pumpage from the aquifer accounts for about 10 percent of its total ground-water discharge. Conse-
quently, under average hydrologic conditions the principal artesian aquifer has not been significantly stressed, and over the long term, it is in a state of hydrologic equilibrium. Equilibrium, or steady state, implies that the rate of discharge from the hydrogeologic system is equal to the rate of recharge and that no change in ground-water storage takes place. Obviously, there are seasonal changes in recharge, discharge, and ground-water storage (fig. 28), but on an annual average basis the water budget becomes balanced--that is, there are no long-term or permanent changes in storage.
The ground-water budget can be estimated by assuming the steady-state condition that inflow is equal to outflow. Table 12 presents a simplified water budget for the residuum and the principal artesian aquifer hydrogeologic system in the Dougherty Plain area in which precipitation is equal to the sum of overland runoff, base runoff, evapotranspiration, and ground-water pumpage. Assuming steady-state and average conditions, water is estimated to circulate through the hydrogeologic system at the rate of 4,300 ft3/s (2,800 Mgal/d), plus or minus
about 860 ft3fs (560 Hgal/d).
The volume of ground water in storage in the principal artesian aquifer within the study area cannot be accurately determined, because sufficient
Table 12.--Estimated mean annual hydrologic budget factors for the principal artesian aquifer system
Factor
Estimated quantity
(in.)
Accuracy of estimate
(percent)
Variability of factor (in.)
Precipitation
53
+ 5
Overland runoff
4
+20
Base runoff ];_/
12
+20
Evaporation-transpiration
36
+17
Pumpage ]:_/
.+20
+2 . 6 + .8 +2.4 +6. 0 + .2
~/ Primarily ground-water discharge to streams from principal artesian aquifer, but also includes some contribution from the residuum.
]:_/ Total 1900 pumpage from principal artesian aquifer . No significant pumpage occurs from residuum.
56
specific-yield and storage-coefficient data are not available. However, estimates of average specific yield (0.15), aquifer thickness (200ft), storage coefficient (0.003), and potentiometric head (20ft) indicate that about 3,700 billion cubic feet (28,000 billion gallons) of water is presently (1981) in storage.
Water in storage in the principal artesian aquifer could, in theory, supply the entire present (1981) pumpage requirements of the Dougherty Plain for a number of years. In practice, however, reducing water levels below the top of the principal artesian aquifer for any period of time could increase the possibility of sinkhole collapse, reduce or eliminate base flow to streams, and increase well construction and pumping costs. In fact, most existing wells could not be pumped if water levels declined more than 10 to 20 feet below the top of the aquifer. Therefore, the desirable limit to water available from storage alone is about 50 billion gallons, or about half of the 1981 total pumpage of 113 billion gallons.
GROUND-WATER QUALITY
All ground water contains inorganic and, in some instances, organic constituents in solution. The type and concentration of constituents depend upon the surface and subsurface environment through which the water moves, the rate of movement, and the acidity of the water. Concentrations of naturally occurring dissolved constituents gener~ ally increase with depth and distance from the area where the water entered the subsurface.
Excess concentrations of dissolved solids may affect the suitability of ground water for various uses. Waterquality criteria or standards have been established for various uses and serve as a basis for assessing the chemical suitability of water for its intended use. The most important water-quality standards are public health standards which have been established for drinking water.
Selected chemical constituents of interest to this study are those for which recommended and mandatory drinking water standards have been established by the U.S. Environmental Protection Agency (1975, 1977, and 1979). (See table 13.)
Table 13.--Recommended and maximum concentrations of selected constituents in public drinking water supplie s
[From U.S. Environmental Protection Agency, 1979; 1980]
Constituent
Recommended concentration limit in milligrams per liter, except where noted
Inorganic Total dissolved solids . . .. . . ... . Chloride ( Cl)., , , , , , .... , , .. , . Sulfate (S04) ................. ........ .. ........... .
Nitrate (NOrN) ...................... .. .. .. .. Iron (Fe) , , , ..... .. , ...... , Manganese (Mn), .. ,,., Copper (Cu) ... .. .... . . ... ... ... .............. ... ... .... . . Zinc (Zn) ............ , ......... , . . . . ..... . .... . ...... , .. . Hydrogen sulfide (H2S) ...................... .. ....
500 250 250
10
.3
.05 1.0 5. 0
05
Maximum permissible concen tration
Arsenic (As),,., ,,, .............. Barium (Ba) ............................................ .. Cadmiurn (Cd) ............... . ................ ,, ...... .... . Chromium (Cr + 6).,,,, .... . ,.,,., .. Lead (Pb) .. , . ... ....... .. . , .... , . .... . .................. . Mercury (llg) ,,, ... , , ........ ,, Fluoride (F) .. ,, ..... , .. , ............ , . , . ... ......... ,.
0.05 1.0
.01 .05 .05 .002 (See comments.)
Organic
Cyanide ,, .. .. .. .. ..
Endrtne , . . .
Lindane .. , , .....
Methoxychlor . . . ~. 0 I
a I .
Toxaphene , .. a
2, 4-D.......... , ............ .. .... . ....... ..... ........ .
2, 4, 5-TP silvex ............ ... ....................... ..
Phenols ,,, , , .......... , .. , , .. .
Carbon chloroform extract . ....... .
Synthetic detergents .. ........... .. . ............. . ... ... . .
0.05 .0002 .004 .1 .005 .1 .01 .001 .2 5
Fluoride: When the annual average of the maximum daily air temperatures for the location in which the water-supply syste111 is located is the following, the maximum permissible concentration (MPC) for fluoride is:
Temperature, in degrees Fahrenheit
Temperature, in degrees Celsius
HPC, in milligrams per liter
53.7 and below
12.0 and below
2. 4
53.8 to 58.3
12.1 to 14.6
2. 2
58.4 to 63.8
14.7 to 17.6
2.0
63.9 to 70.6
17.7 to 21.4
1. 8
70.7 to 79.2
21.5 to 26.2
1. 6
79.3 to 90.5
26. 3 to 32. 5
1. 4
Recommended standards apply to those constituents which may adversely affect public health, the taste, color, or turbidity of the water, or may impart some other undesirable characteristic to the water. Consumption of water having con-
57
centrations somewhat above the recommended limits is generally not harmful to humans. Mandatory limits, however, establish maximum permissable concentrations in drinking water, and human consumption of water having concentrations above these limits may produce specific toxic or adverse physiological effects.
Water samples were collected from 16 residuum and 14 principal artesian aquifer wells (table 14). These samples were analyzed for concentrations of major inorganic constituents and for agricultural pesticides, herbicides, insecticides, and fungicides commonly used in southwest Georgia (table 15). Analyses for the trace elements arsenic, lead, mercury, copper, and zinc were not made because previous studies (Radtke and others, 1980; Pollard and others, 1978) found none of these elements in concentrations exceeding the recommended or maximum limits in either surface or ground water in the Dougherty Plain.
The chemical quality of ground water in the Dougherty Plain area varies both within and among the separate geohydrologic units. Statistical analyses of concentrations of selected constituents in water from the residuum and principal artesian aquifer are listed in table 16. These data indicate that water from these units usually meet U.S. Environmental Protection Agency recommended or mandatory standards.
Pesticides
Rapid growth in agricultural use of large-acreage irrigation systems has resulted in increases in the use of fertilizers and pesticides, some of which are toxic to humans, long-lasting, and tend to accumulate in the hydrogeologic system.
Pesticides were detected in water from 11 residuum wells and 4 principal artesian aquifer wells. Total pesticide concentrations were usually greater in water from the residuum than in water from the principal artesian aquifer. Water from two of the principal artesian aquifer wells contained pesticide concentrations only slightly above detection
limits, whereas water from the other two principal artesian aquifer wells contained concentrations of pesticides within detection limits.
The presence or ab~ence of pesticides in water from principal artesian aquifer wells as reported herein is valid only for the times that the samples were taken. Concentrations of pesticides could be greater or less in samples from these same wells at other times. As discussed previously, water flow in the aquifer may range from about 0.2 ft/d to 3 ft/d. Consequently, pesticides can quickly move through the aquifer in areas where flow velocities are relatively high. Thus, pesticide detection is strongly time dependent.
The areal extent, severity, and the long-term affects of pesticides upon quality of water from the principal artesian aquifer cannot be determined from
the available data. The u.s. Geological
Survey and the U.S. Environmental Protection Agency currently are conducting further investigations and analyses ~f pesticide movement in the Dougherty Plain area.
GROUND-WATER FLOW MODEL
Model Description
A two-dimensional numerical model developed by Trescott and others (1976) was used to simulate water levels in the principal artesian aquifer. Water levels in the aquifer are affected by pumpage and variations in natural recharge and leakage to and from streams. The digital model utilizes a central finite-difference scheme to evaluate the partial differential ground-water flow equations in which the head is the dependent variable.
Three reasons underlie the choice of a two-dimensional ground-water model to simulate ground-water conditions in the Dougherty Plain: (1) the flow system in and around the Dougherty Plain area can be conceptualized (without significant simulation error) as a two-dimensional flow system; (2) during a drought period in which substantial amounts of agricultural pumping is occurring, the aquifer
58
Table 14.--Selected water-quality data for wells from which water was analyzed for major inorganic constituents and pesticides
[Ceo hydrologic unit: PCPA, principal artesian aquifer; RSDH, residuum]
Well No.
Well name
Geohydrologic
unit
Date sampled
Water level (ft below land surface)
Temperature {"C)
Specific conduc-
tance pH at 25C
Alkalinity as CaC03 (mg/L) Unfiltered Filtered
Hydrogen sulfide
(mg/L)
29 T. Rentz TW 1
Baker County
PCPA 04-23-81 17. 29
20.5 7.66 280
118
118
24 B. Jordan TW 25 B. Jordan TW
Calhoun County
RSDH 04-14-81 26.30
21.0 6. 40
60
26
21
PCPA 04-14-81 20.98
20.4 7.74 270
180
107
0.4
Decatur County
10 A. Newton, North TW
PCPA 04-22-81 43.81
20 . 5 7.78 220
95
93
.4
43 DP 5
PCPA 04-22-81 54.17
20.6 7.79 210
107
104
44 DP 6
RSDH 08-20-80
5.60
50
15 Nilo, North TW 2 71 Nilo, South TW 3 72 USMC Supply TW 1
Dougherty County
PCPA 04-14-81 31. 25
20.6 7.60 230
RSDM 04-15-81 36.13
20.5 6.60 125
RSDH 04-13-81 23.47
5.75
34
138
52
44
Early County
39 I. Newberry TW 1
PCPA 04-23-81
20.3 7. 44 265
116
116
.o
45 I. Newberry TW 2
46 v. Evans TW 1
RSD~I
RSDH
04-15-81 25.72 04-23-81 8. 97
21.0 7.30 160 18.7 6.36 110
75
72
43
43
Lee County
15 M. lloorman TW 1
PCPA 04-20-61 27. 51
20.1 7.60 200
95
85
.o
40 Piedmont Plant Farm TW 1 RSmt 04-20-81 34.00
21. 5 6.80
83
20
21
41 s. Stocks TW 1
RSDII 04-20-81 14.07
20.5 6. 80
77
7
7
43 II. Usry TW1
RSDII 04-16-81 23.80
19.0 6.10
63
16
16
15 DP 2 16 DP 3 33 J. Fleet TW 2
Hiller County
PCPA 04-22-51 20.79
21.0 7. 61 270
184
135
2
RSDH 04-22-81 34.94
21. 1 6.90 320
740
7
RSDH .04-23-81 32.37
6.78 320
148
148
16 c. Holton TW 1 35 c. Holton TW 2
36 H. Davis TW 1 38 DP 11 40 H. Davis TW 2
l1itchell County
PCPA 04-15-81 28. 21
20.7 7.95 185
RSDII 04-15-81 43.50
20.0 6.30
55
RSDM 04-21-81 31.78
20.0 6. 10 160
PCPA 04-15-81 44.92
20.3 7.90 195
PCPA 04-21-81 45.35
20.7 7.80 210
103
87
20
18
11
90
80
103
102
Seminole County
08 Roddenberry TW 1
PCPA 04-22-81 28. 51
20.8 7.60 265
129
126
.8
26 D. Harvey TW 1
PCPA 04-23-81 57.07
21.1 7. 72 230
107
107
.o
22 E. Stephens TW
Sumter County
RSDH 04-16-81 14.95
19.0 6.00
70
14
15
04 DP 8 05 DP 9
09 c. Odom TW
Worth County
PCPA 04-16-61 12. 53
20.6 7.75 215
RSDII 04-16-81 22.68
20.0 7.90 155
RSDH 04-16-81 24.90
19.5 5. 20
60
117
113
1,070
71
14
15
becomes partly unconfined and the twodimensional model is capable of simulating an aquifer which is changing states (from confined to unconfined); and (3) the amount of data available could not justify the use of a three-dimensional model.
Heads simulated by the model were compared with measured heads from wells in the project area. The comparisons were made to evaluate the concepts used
in the calibration process and to measure the accuracy of the model in response to hypothetical changes in the geohydrologic system. The calibrated model was used to predict the effects of pumpage ranging from 113 to 408 billion gallons per year urtder hypothetical hydrologic drought conditions and 287 billion gallons per year under long-term, average hydrologic conditions.
59
Table 15.--Agricultural pesticides commonly used in southwest Georgia, 1976-77 [From Radtke and others, 1980]
Chemical name
Class
Crop
Pounds of active ingredients/acre
Residual
HERBICIDES
Translated (systemic) herbicides
2,4-D 2,4-DB
Atrazine Proazine Simazine
Chloroxuron Linuron
Butylate Vernolate
Alachlor Bene fin Trif luralin
Contact herbicides
Dinoseb Paraquat
Phenoxy acid do .
Triazine do . do.
Substituted urea do.
Carbamate do.
Substituted aniline do. do.
Phenol Pyridylium
Corn, grain sorghum Peanuts
Corn, grain sorghum Grain sorghum Corn
Soybeans Grain sorghum, soybeans
Corn Peanuts
Peanuts, corn, soybeans Peanuts Soybeans, vegetables
Peanuts, soybeans Corn
0.5 . 25
2-3 2
2-3
1-1.5 1
3-6 2-2.25
3 1-1. 5
5-1
2 . 25
week week
3-12 weeks 2~3 weeks 2-3 weeks
1-2 weeks 3 weeks
3-8 weeks 3-8 weeks
3 weeks 2-4 months
2 weeks
none
INSI.lCTIC IDES
Dicofol Carbofuran Diazinion Malathion Disulfoton
Chloronated hydrocarbon Carbamate Organophosphate
do. do.
Peanuts, soybeans Peanuts Peanuts, soybeans Peanuts, tobacco Peanuts
.8 1.5 1. 5 1. 0
.75
NEMATOCIDJ.lS
Dibromochloro pro pane Ethoprop Carbofuran
Fumigant Nonfumigant organophosphate Nonfumigant carbamate
Peanuts Peanuts, corn, soybeans Peanuts, corn
6 qts. 2 1. 5
FUNGICIDES
Benomyl Chlorothalonil Quinrozene
Carbamate Chloronated hydrocarbon Chloronated benzene
Vegetables, peanuts Peanuts Peanuts
5 1.0 10.0
System Concepts
To numerically model the principal artesian aquifer flow system, a conceptual flow model of the aquifer flow system in southwest Georgia was developed. The aquifer may be conceptualized as being confined from above by the residuum (described previously in this
report) and from below by the Lisbon Formation. Furthermore, the aquifer is assumed to be homogeneous and isotropic. Water recharges the aquifer by moving vertically downward through the residuum and discharges from the aquifer to pumping wells and to streams that are hydraulically connected to the aquifer. This conceptual flow model is illustrated in figure 30.
60
Table 16.--Statistical summary of water-quality data pertinent to the residuum (RSDM) and the principal artesian aquifer (PCPA)
Source
Constituent
Number of
samples Minimum
Statistical analysis
Maximum Mean
Standard deviation Median
Mode
RSDM PCPA
RSDM PCPA
RSDM PCPA
RSDM PCPA
RSDM PCPA
RSDM PCPA
RSDM PCPA
RSDM PCPA
RSDM PCPA
RSDM PCPA
RSDM PCPA
RSDM PCPA
RSDM PCPA
RSDM PCPA
RSDM PCPA
RSDM PCPA
RSDM PCPA
RSDM PCPA
Temperature (degree C)
14
do.
20
Specific conductance (umhos)
19
do.
20
pH (units)
18
do.
16
Alkalinity field (mg/L, as CaC03)
15
do.
16
Alkalinity (mg/L, as CaC03)
13
do.
13
Hardness (mg/L, as CaC03)
18
do.
20
Calcium, dissolved (mg/L, as Ca)
18
do.
20
Magnesium, dissolved (mg/L, as Mg)
18
do.
20
Sodium, dissolved (mg/L, as Na)
18
do.
20
Sodium adsorption ratio
18
do.
20
Potassium, dissolved (mg/L, as K)
18
do .
20
Chloride, dissolved (mg/L, as Cl)
18
do.
20
Sulfate, dissolved (mg/L, as S04)
18
do.
20
Fluoride, dissolved (mg/L, as F)
18
do.
20
Silica, dissolved (mg/L, as Si02)
18
do.
20
Iron, suspended recoverable (mg/L, as Fe) 10
do.
12
Iron, total recoverable (mg/L, as Fe)
16
do.
16
Iron, dissolved (mg/L, as Fe)
17
do.
16
18.7 17.0
34.0 150.
5.2 7.4
7.0 9.5
7.0 90
11 89
3.4 35.0
.3 .2
1.3 1.3
.1 1
. 2 .1
2.0 1.5
0 0
.1 0
.1 5.2
.160 .08
.180 .100
.010 .000
21.5 21. 1
490.0 280
7.9 8.0
210 184
1070 184
240 140.0
87.0 53.0
4.3 3.3
6.8 6.0
.3 .3
5.9 1.1
14.0 6.2
6.1 8.8
.5 2
12.0 20.0
73.0 15.0
73.0 15.0
3.20 .510
20.1 19.9
145 220
61 112
159 119
61 105 . 0
22.0 42.1
!.2 9
2.8 2.2
.2 .1
.9 . 3
4.3 3.2
2.5 1.8
.2 1
6.07 8.2
20.9 31.8
18.2 3 . 55
.289 .051
0.9 1.3
118 35
64 39
338 30
60 15.3
22.0 6.0
1.0 .6
1.5 l. 1
.l .1
1. 3 .2
2.7 1.3
2.0 2.2
1 .03
3.1 3.6
23.7 53.7
20.1 5.08
.779 .123
20.0 20.5
llO 212
19.0 20.5
60 200
7.7
26 108
20 107
38 105.0
14.0 40.0
.9 .8
2.4 1.9
.2 1
.4 .3
3.7 3.0
1.8 1.0
.1 .l
6.3 7.0
12. 1 .71
12.5 .755
.020 .020
7.6
14 103
14 95
17 100.0
7.9 40.0
1.0 .5
2.6 2.1
.2 1
.3 .3
3.0 1.6
1.3 1
.1 1
3.0 5.2
.160 .80
2.80 .100
.010 .010
61
Table 16.--Statistical summary of water-quality data pertinent to the residuum (RSDM) and the principal artesian aquifer (PCPA)--Continued
Source
Constituent
Number of
samples Minimum
Statistical analysis
Maximum Mean
Standard deviation Median
Mode
RSDM
PCPA
RSDM PCPA
RSDM PCPA
RSDM PCPA
RSDM PCPA
RSDM PCPA
RSDM PCPA
RSDM PCPA
RSDM
PCPA
RSDM PCPA
RSDM PCPA
RSDM PCPA
RSDM PCPA
RSDM PCPA
RSDM PCPA
RSDM PCPA
Manganese, suspended recoverable
(mg/L, as Mn)
13
do.
9
Manganese, total recoverable (mg/L, as Mn) 14
do.
14
Manganese, dissolved (mg/L, as Mn)
14
do.
14
Nitrogen, dissolved (mg/L, as N)
3
do.
10
Nitrogen, organic dissolved (mg/L, as N)
3
do.
10
Nitrogen ammonia, dissolved (mg/L, as N)
16
do.
16
Nitrogen nitrate, dissolved (mg/L, as N)
13
do.
14
Nitrogen nitrite, dissolved (mg/L, as N)
5
do .
2
Nitrogen ammonia + organic, dissolved
(mg/L, as N)
3
do.
10
Nitrogen, N03 + N02, dissolved (mg/L as N) 16
do.
16
Phosphorus, dissolved (mg/L, as P)
17
do.
16
Carbon, organic, total (mg/L as C)
14
do.
14
Sulfide, total (mg/L, as S)
1
do.
9
Nitrogen ammonia, dissolved (mg/L, as NH4) 16
do.
16
Nitrogen nitrate, dissolved (mg/L, as N03) 5
do.
2
Nitr.ogen nitrite, dissolved (mg/L, as N02) 5
do.
2
.ooo
.000
.030 .010
.001 1.0
.3 .1
1 0
.02 .01
.01 .01
0.08 2.5
9.90 .770
30.0 1.10
33.0 30.0
1.7 3.6
.5 .3
1.5 .3
.02 01
0.6 3.3
2.13 .146
4.56 .189
2.80 7.30
.8 2. 0
2 . 1
.2 .09
.01 .01
0.4 2.9
2 .3
.02 .02
.01 .01
1.6 8
7 0
.03 .01
.4 11.0
.07 .03
.6 .4
3.1 3.3
.03 .07
120 4.3
.7
.a
1.9 .4
2.7 15.0
.07 .03
.3 .2
.8 1.5
.02 .02
15.2 2.3
7
2
. 2 1
1.6 13.0
.06 . 03
3.24 .247
7.95 .330
8.73 10.2
1.3
.1 .4 .08 .005 0
.I
9 !. 2
. 008 . 02 30.1 1. 2
. 3 .4 1
.500 .030
1.42 .030
.135 2.0
1.9
.l
.06 .06
.01 .01
.ooo
.030
.030 .030
.280 1.0
1.9
0
.05 .04
01 .01
.2
.4 1.4
.01 .02
5.8 2.1
0
.08 .07
.03
.02 .04
.01 .02
1.6 1.0
0
.06 .05
62
NORTH
Potentiometric surface of the principal artesian aquifer
SOUTH
Residuum
(vertical leakage)
Local flow
Semiconfined aquifer Regional flow
Lisbon confining bed
(no vertical leakage)
Figure 30.-Conceptual flow model of the principal artesian aquifer system.
63
Ground-Water Flow Analysis
Two-dimensional transient groundwater flow in a confined aquifer can be described by
__1_(Txxah)+ 2(Tyy.ah) =
( 1)
ax ax ay ay
in which
Txx Tyy are the principal components of the transmissivity tensor, in the x and y di,ections, respectively (L2cl),
h is the hydraulic head (1), S is the storage coefficient
(dimensionless), and
W(x,y,t) is the volumetric flux of recharge or withdrawal per unit surface area of the aquifer system (1t-1).
For steady~state conditions, equation (1) can be reduced to
YYaY _a_x1_(T~ ah)+_aly_(T ah) = W(x,y)
(2)
Equations (1) and (2) are approximated by the use of a finite-difference scheme which is described in detail by Trescott and others (1976).
The numerical model utilizes the following equation to calculate leakage into and out of the aquifer
Q = K'v
w
(hr-ha )
A '
( 3)
in which
Q is the leakage (13t-1),
k'v is the vertical hydraulic conductivity of the residuum confining layer (Lt-1),
M' is the thickness of the residuum confining layer (1),
A is the unit surface area (12), hr is the head in the residuum
(water table, 1), and
ha is the head in the principal artesian aquifer (1).
For a stream that cuts into the principal artesian aquifer, leakage was estimated in a slightly different manner. Where the stream cuts into the aqu.ifer, water flows laterally toward the stream
and, in the immediate vicinity of the stream, flows vertically upward into the stream as shown in figure 30. Assuming that the aquifer has homogeneous geologic properties, its average head occurs onehalf the distance from the streambed to the bottom of the aquifer (Johnston, 1977, p. 12). Thus, the discharge into a stream may be calculated by conceptualizing the upper half of the aquifer as a confining unit and the head in the confining unit as the head in the aquifer. However, since a stream occupies a much smaller surface area than does a cell block (described in the following section and illustrated in figure 32), and since the leakage computation is for an entire cell area, the quantity of leakage must be reduced in accordance with the ratio of stream surface area to cell block surface area. This is accomplished by reducing the values of vertical hydraulic conductivity by the ratio of stream area to cell block area. Therefore, discharge into the stream is calculated by
. q1 = As - ~ Chr- hz )
(4)
where An
b/ 2
q1 = leakage (13t-1), As = surface area of stream
(12),
= surface area of cell block
(12),
= vertical hydraulic conductiv-
ity (12t-1),
= head in the confining layer
(water table) (1),
h2 = head of the aquifer at the
stream (1),
and
b/2 = 1/2 thickness of the aquifer
(thickness of the artificial con-
fining bed, 1).
Equation (4) is described in detail by
figure 31.
Finite-Difference Grid and Boundary Conditions
The principal artesian aquifer was idealized by using a 78-row by 105-column finite-difference grid, shown in figure 32. Each cell block of the grid.occupies
64
UNIFORM RECHARGE RATE
t
t
h2 hR
ARTIFICIAL CONFINING BED
AQUIFER
AQUIFER
AQUIFER
INTERSTREAM NODE
(Kv = o)
STREAM NODE
INTERSTREAM NODE
(Kv = 0)
Where q1= Leakage, As= Surface Area of Stream, AN= Surface Area of
Node, Kv =Vertical Hydraulic Conductivity, hR =River Head (constant), h1, h2 , h3 = Heads in Aquifer, and b/ z =Thickness of Artificial
Confining Bed ( 1; 2 Aquifer Thickness).
Figure 31. --Conceptual flow model of hydraulic connection between the principal artesian aquifer and the Flint River. Modified from Johnston (1977) .
65
a 1-square mile area throughout the grid with the node located at the center of the cell block. The following boundary conditions (shown in figure 32) were imposed on the model: (1) Constant-head boundary: Along the
Chattahoochee River to the west and below Lake Seminole, the aquifer head at a specified cell block was held constant for a specific simulation. Along the Chattahoochee River, the aquifer head ranged from 75 ft above sea level near Lake Seminole to 144 ft above sea level at the model's northern boundary. (2) Constant-flux boundary: The Dougherty Plain is separated from the Tifton Upland on the east by a topographic and ground-water divide across which no water is assumed to flow. Furthermore, the updip limit of the principal artesian aquifer generally coincides with the northern physiographic boundary of the Dougherty Plain. Thus, the eastern and northern boundaries of the model were assigned a constant flux value of zero for all simulations.
These boundary conditions are realistic for both steady-state and transient simulations and are factual representations of existing field geologic and hydrologic conditions. Presently, there are no centers of major pumping along the zero flux boundaries.
Data Requirements
The data requirements of the model are aquifer and confining bed hydraulic parameters and initial conditions. Aquifer transmissivity (T) and vertical hydraulic conductivity (K'v) and thickness (b') of the residuum confining unit are required for each cell block. Aquifer storage coefficient (S) is required
only for transient analysis. Furthermore, initial heads of the residuum and the aquifer, and stages of streams must be specified for each block in the aquifer for both steady-state and transient simulations.
Hydraulic properties
Transmissivity values for the principal artesian aquifer used in the model ranged from 3,000 ft2Jd to 300,000 ft2/d, and are within the limits of the field data, as previously discussed.
Storage coefficients for the principal artesian aquifer calculated from 18 aquifer tests range from 2 x 1o-4 to 3 x 10-2. A storage coefficient of 5 x 10-4 was assumed for confined conditions throughout the modeled area. Where water levels declined below the top of the principal artesian aquifer (a water-table condition) as a result of pumping or other stress, a specific yield value of 0. 2 was used.
The vertical hydraulic conductivity and thickness of the residuum confining unit were combined to form the parameter known as leakance. Thus
v L' = K' v,
(5)
where L' is the leakance (t-1), K'v is the vertical hydraulic conductivity of the residuum confining layer (Lt-1),
and b' is the thickness of the residuum confining layer (L). Figure 20 illustrates the areal
range of leakance (K'v/b') of the residuum used in the calibrated model. Preliminary valu~s of leakance were estimated by dividing estimated vertical hydraulic conductivity values by residuum confining layer thickness, which is considered to be equivalent to the bottom half of the residuum. An initial leakance map based on these results was modified during the steady-state calibration of the digital model. Consequently, the resulting map presents estimated ranges of leakance based on test-drilling data and digital modeling analyses. The values presented in figure 20 are considered realistic and within the range of accuracy of the field data.
66
EXPLANATION
DOUGHERTY PLAIN BOUNDARY
- - - CONSTANT-FLUX (NO FLOW) MODEL BOUNDARY
-
CONSTANT-HEAD MODEL BOUNDARY
Bose from u.s. Geological Survey 1:250,000 quadrangles
Figure 32:---The model area with finite-difference grid end boundary conditions. No flow where boundary end grid lines coincide; constant heed where boundary line bisects cell block.
67
The method used in calculating the vertical hydraulic conductivity of nodes dedicated to streams is described below. The thickness of confining beds underlying all streams was assumed to be 10 ft. Measurements were made to estimate the amount of ground water discharged to the largest (perennial) streams (fig. 33). This process yielded four different classifications of streams:
(1) For streams that gained ground water at a rate of less than 0.1 (ft3/s)/mi2, the confining layer between the streams and the aquifer was assigned a vertical hydraulic conductivity value of 0.005 ft/d. This was the median value of hydraulic conductivities obtained from field tests. These streams were considered to be minor streams that had little or no effect on the aquifer. (2) The confining layer under streams that had gains ranging from 0.1 to 0. 2 5 (f t 3j s ) / mi 2 were as s i g ned a vertical hydraulic conductivty value of 0.01 ft/d and a stream node ratio of 0.02. (3) The confining layer under streams that had gains greater than 0.25 (ft3/s)/mi2 were assigned a vertical hydraulic conductivity value of 0.05 ft/d and a stream node ratio of 0.04. (4) Streams that cut into the aquifer (such as the Flint and Chattahoochee Rivers) were assigned vertical hydraulic conductivity values of 40 ft/d. Initial model runs indicated that these values were generally acceptable. Leakance values were changed only slightly to achieve a calibrated model.
Another required hydraulic input parameter is the altitude of the water table in the residuum (or riverhead) (fig. 22). The water-table altitude varies, depending on climatic conditions and the time of year. Data from 29 residuum wells were used as control points. Regionalization of these data is based on topography, lithologic character of the residuum, stream and surface drainage features, and data trends. For purposes of steady-state model calibration, the November 1979 water-table altitudes
(riverheads) were used. Furthermore, water-table altitudes assigned to stream
nodes were average stream surface altitudes obtained from 1:24,000-scale topographic maps and from gaging-station data.
The digital model used in these simulations has the ability to simulate unconfined and confined aquifers. Unconfined or water-table conditions occur in the principal artesian aquifer where the potentiometric surface in the aquifer falls below the base of the residuum confining layer. In order to simulate an unconfined aquifer, three additional hydraulic parameters are required:
(1) Horizontal hydraulic conductivity of the principal artesian aquifer. This value was determined by dividing the transmissivity (fig. 23) of the aquifer by its thickness for each cell block (fig. 6). ( 2) Altitudes of the top and base of the aquifer. These values were determined fromstructure contour maps of the tops of the principal artesian aquifer (fig. 5) and the Lisbon Formation (fig. 7). (3) Specific yield of the principal artesian aquifer. A constant value of 0.2 was used in the model. While this value may seem high for a limestone, it was believed that the large solution channels occurring as secondary porosity justified the use of the large value.
Initial conditions
Initial potentiometric values were assigned to each cell block of the principal artesian aquifer being modeled. Measured water levels were used to construct the potentiometric surface map for November 1979 (Mitchell, 1981, pl. 2), and values for each cell block were derived from this map.
Model Calibration
The purpose of a calibration procedure is to represent natural ground-water flow conditions with a digital model as accurately as possible, within existing
68
30'
Jli -
EXPLANATION
[=:J AREA OF DOUGHERTY PLAIN
0 .01(004)
STREAMFLOW GAGING STATION-
Left number is streamflow for August
4-7,1980. Number in parenthesis is
streamflow for January 5-7,1981.
Streamflow is in cubic feet per second
per square mile
Otifl hom U.S Geo Q(Ji~;:o! S11twey 112:.)Q,OOO quodrongles
10
ID
20
25
30 MilES
I
Figure 33.-Measured stream discharge for August 1980 and January 1981.
69
limits of available data. The process of adjusting model input parameters until realistic results are obtained is termed calibration. The digital model was calibrated for November 1979 steady-state conditions. The model calibration was then tested by simulating transient conditions during May to November 1980.
Calibration Procedures
The error criterion selected for calibration required the mean error between simulated and derived cell-block values to approach zero for all cell blocks and the standard deviation to be less than + 5.0 ft. Assuming a normal error distribution, this would assure that 95 percent of all simulated heads would be within + 10 ft of derived heads which themselves-are considered accurate to generally+ 10 ft. Since input head values are derived from potentiometric maps based on measured heads, the errors computed by the model (drawdown) are the input heads minus the simulated heads.
During the calibration procedure, aquifer transmissivity, leakance of the residuum confining unit, and water-table altitudes (riverheads) were varied. Calibrated transmissivity values ranged from about 3,000 to 300,000 ft2/d. These compared well with measured values, and none were more than 2 times the measured values.
Leakance (K'v/b') was varied more than the transmissivity during the calibration process; however, care was taken to assure that the final calibrated values agreed, in general, with values determined from test drilling and general data trends (fig. 20).
Because few water-table altitudes were available, this parameter was least accurately known and was the most varied. However, in all areas the data were checked to assure that water-table altitudes were above the top of the artesian aquifer and below land surface.
November 1979 Steady-State Simulation
The simulated steady-state potentiometric surface and measured water levels
(heads) for November 1979 are shown in figure 34. From the figure it is apparent that the simulated values compare favorably with the measured data. Average simulation error was 0.6 foot with a standard deviation of error of 4.6 ft. This was within the desired criterion that 95 percent of all the simulated heads be within +10 ft of input data.
The distribution of the head error (difference between cell-block values derived from measured heads and simulated heads) is shown in figure 35. The error in the heads approximates a normal distribution at a class interval of 4.0 ft. The difference between the simulated potentiometric surface and the potentiometric surface constructed from measured water levels is shown in figure 36. The areas of greatest difference usually occur along or near the streams. This is probably due to the required application of stream leakage over an entire cell block, as discussed previously. The authors considered it necessary to simulate quantities of water discharged to and received from streams that would approach values obtained from field measurements. Consequently, at several stream nodes the drawdown required to leak this discharge exceeded the desired calibration error criterion. However, since these nodes were few in number, they did not affect the overall calibration significantly (fig. 35).
In addition to requiring the simulated heads to meet the error criterion, the simulated ground-water discharge to streams must also compare acceptably with measured ground-water discharge. Although no seepage-run measurements were available for November 1979, two sets of measurements were made at selected sites during August 4-7, 1980, and January 5-7, 1981 (fig. 33). Comparison of these measured flows with simulated flows for November 1979 indicates that the simulated values are reasonable because stream baseflow and ground-water conditions were
si~ilar for November 1979 and August 1980. The field measurements for these months are compared with simulated results for November 1979 in table 17. Allowing for the changes in ground-water levels and slight climatic differences between November 1979 and August 1980,
70
es"oo'
30'
EXPLANATION
C J AREA OF DOUGHERT Y PLAIN
- 120- SIMULATED POTENTIOMETRIC CONTOUR -
Shows altitude a t which water level would have siood in tightly cased wells. Contour Inte rv al 20 feet. National Geodetic Vertical Datum of 1929
00 '
DATA POINT - Number is altitude of
measu red water level.
.s IJou h om
ll!!o a gl~:ot Sur ~e~
l t 2~0,000 quadrangles
F L0 R
0 A
0
$
10
15
20
'!.l'!ll
.lO MILES
~~ ----~---L--~----L-~
Figure 3 4.-- Measured water levels and simulated potentiometric surface of the principal artesian aquifer, November 1979.
71
1600~------~--------,---------r-------~---------r--------~-------.---------r--------.-------~~-------.--------~
CwIJ 1000
0
0
z
LL.
0
a:
" N
w
Ill
:2:
::l
z
Average error: X=0.6 foot Standard deviation: S =4.6 4,474 Active nodes
0 I -23.4
I? J Q? >:E"? -&z--z:-V / / / 6 / / / / Y / / / / ' 1 / / / / ,..t / / / / V / / / /lz=z--2 z,t., - q, ? I
Q
I
-19.4
-15.4
-11.4
-7.4
-3.4
0.6
4.6
8.6
12.6
16.6
20.6
24.6
AVERAGE ERROR, IN FEET
Figure 35.- Distribution of head error for the November 1979 ca Ii brat ion of steady-state si mu lotion .
...
31 41 00'
EXPLANATION
- - - DOUGHERTY PLAIN BOUNDARY
DIFFERENCE IN FEET
CJ 0 to 5
. . 5to l 0 . . 10to15
F Oooo ftQfft l,l. S GooiGQlt:Gl S\H 'li!'f II~!Jo;ooo qvo.dr6.nqln
A
.
25
lO MILES
Figure 36 -Areal distribution of difference between the November 1979 simulated potentiometric surface and the potentiometric surface constructed from measured water levels.
73
conducted by varying these parameters.
Table 17.--Ueaaured and simulated ground-water discharge to s~lected streams
Transmissivity and leakance values were
Stl'earu
Flow, in ft3/s Meusured Aug. 4-7, 19tW Jan. 5-7, 1981
Simulated Nov. 1979
Stream n~ach
Upstream Downstream
station
station
varied from 25 percent to 400 percent of the calibrated value. Water-table altitudes were varied from 80 percent to 120
Dry Creek Spring Cre~k Ichawaynoch.away Creek Chickasawhatchee Creek Kinchafoonee Cr~ek Huckaloochee Creek Lime Creek Turkey Creek Pennnhatchee Creer.. Huckalee Creek Jones Cl'eek Hint River
l32 IS 39 7.1
1.5 22
7.1
J:J 1,200
12
14
Headwater
56290
percent of the values used in the cali-
71
55
56100
57050
bration. Ten different computer runs
53
80
53266
55350
were made varying the calibrated parame-
38
29
Headwater
54500
ters. The average error, standard devi-
20
20
50860
51000
ation, and simulated ground-water dis-
14
7. 2
51780
51800
charge for each run are given in table
16
15
Headwater
50100
18.
15
12
49900
49910
Several conclusions may be drawn
5. 2
3. 7
Headwater
49980
from table 18.
29
65
51700
51920
(1) By varying transmissivity and
4.4
6. 2
Headwatcr
50509
leakance, acceptable average simulated
1,300
49500
53000
head errors and standard deviations
!/ Net gain in Hint River flow betwe(!n Hontezurna (station 49500) and Newton (station 53000),
aftt:!t:' subtracting tributary inflow tu Flint River between t(ontezun!a and Newton.
could be achieved. However, these new parameters could not simulate an
acceptable water budget.
(2) Even though the altitude of the
water table in the residuum (river-
head) was varied by only + 20 percent,
it produced drastic changes in simula-
ted heads--values that would be unac-
ceptable if they were used in a cali-
brated model.
and the effects of irrigation pumpage from the stream, the agreement between measured flows and simulated flows was acceptable.
Another factor to consider in evalu-
(3) The model is most sensitive to changes in water-table altitude. Thus, the accuracy of the calibrated model could be most improved by additional field data that better define
ating calibration validity is the total
the water table in the residuum.
ground-water budget for the Dougherty
Plain area. Base-flow analysis was used
for eight watershed areas comprising the
Dougherty Plain. (Refer to section on Base Flow.) Over a 10-year period (water
Table lB.--Sensitivity of aquifer transmissivity (T), confining zone leakance (L), and riverhead (R) on the calibrated model for Nov e mber 1979
years 1959-70) late-summer (Sept., Oct., and Nov.) mean base runoff derived primarily from the principal artesian aquifer was calculated to be about 2,300
Run No.
c -1/
Parameters T, L, R
Average error (ft)
+0.6
Standard deviation
4.6
Water budget [ft3/s (in./yr)]
2, 207 ( 6 . 4)
ft3/s. Because the simulation was pertinent only to November 1979 and because November base flow is slightly less than the average of September through November
0. 25T, L, R 0.50T, L, k 2.0T, L, R
-2.0 -. 9
+2.6
4. 4
934 (2 . 7)
4. 2
I, 458 ( 4. 2)
5. 7
3, 213 ( 9. 3)
flows, the simulated ground-water flow
4
4.0T, L, R
+5.4
7.7
4,491 (13 .0)
should be slightly less than estimated
T, 0.251, k
+5.4
7.7
1' 127 ( 3. 3)
late-summer values. In fact, the simula-
T, O.SOL, R
+2. 7
5. 7
1,610 ( 4. 6 )
ted ground-water discharge was 2,200 ft3/s, which was considered an acceptable comparison with the hydrograph analysis.
To measure the sensitivity of the
T, 2.01, R T, 4.01, R T, L, O.BR
-.8 -2.0 +38.2
4.2 4.4 18.0
2,817 (8.1) 3, 700 (10. 7) 3,056 (8. B)
calibrated model to changing transmissiv-
10
T, L, I. 2R
-37.0
18.0
3, 943 ( 11.4)
ity, leakance, and water-table altitudes (riverheads), a sensitivity analysis was
!/ Calibrated model run for November 1979 .
74
May-November 1980 Transient Simulation
Because of the increasing demand on ground-water supplies for agricultural irrigation in the Dougherty Plain, the utility of the model would be considerably enhanced if it were capable of accurately (within the established error criterion) reproducing ground-water conditions during a given irrigation season. With measurements of the altitude of the water surface in the principal artesian aquifer available during May and November 1980, ground-water conditions were simulated by the model for the period of May 15 to November 5, 1980.
This total period was simulated in stages by using 3 time periods. During the first period of 17 days (May 15-31, 1980), only municipal pumpage of 24 ft3/s was considered. Starting heads were those obtained from measured values for May 1980. Water-table altitudes in the residuum (riverhead) were obtained in a manner described below.
Because an areal distribution of measured water-table altitudes was not available for May 1980, the authors calibrated a steady-state model using available May 1980 potentiometric surface measurements to obtain water-table riverhead) values. The May 1980 steadystate calibration utilized aquifer transmissivity and leakance values determined from the November 1979 steady-state calibration. The ~fuy 1980 steady-state simulation produced water-table altitudes (riverheads) that were greater in magnitude than those used in the November steady-state simulation. This was in agreement with existing hydrologic conditions at the simulation time, since an increase in precipitation had occurred during the wi~ter months (Dec.-Apr. 1980). The calibrated steady-state model for May 1980 (using the simulated watertable values) met the calibration error criterion required of all model calibrations (average simulation err.or of 1 ft and standard deviation of error of 4.7 ft). Furthermore, all water-table data were checked to assure that values were above the top of the principal artesian aquifer and below land-surface altitudes.
The second time period of 107 days began on June 1, 1980, and ended on September 15, 1980. This period included
most of the 1980 growing season when
agricultural pumping reached a very high level in the Dougherty Plain (H. E. Gill, U.S. Geological Survey, oral commun., Nov. 1981). In terms of an annualized rate, ground water was used for agricultural irrigation at the rate of about 1,100 ft3/s during 1980. For modeling purposes, all irrigation systems within a 1-square mile grid-block area were summed and assumed to be at the center of the block. Data on the number and capacity of irrigation systems in the Dougherty Plain were obtained through a field survey of existing irrigation systems during the spring of 1980 (R. R. Pierce, U.S. Geological Survey, oral commun., 1981). The locations of agricultural irrigation systems in the modeled area, as of the spring of 1980, are shown in figure 37. In addition to agricultural use, municipal pumpage of 24 ft3/s was also included during the second time period.
The simulated head values calculated at the end of the first time period were used as the starting head values for the second time period. However, constanthead nodes (Chattahoochee River and Lake Seminole) were assigned a value between the measured potentiometric water-levels of May 15 and November 5, 1980. The values assigned to the c.onst.ant head nodes for September 15, 1980,' were estimated by inspection of ground-water hydrographs from wells located near the Chattahoochee River and Lake Seminole. The assumption was made that all constant-head nodes would have similar ground-water-level declines as did the wells where measurements were available. Therefore, the amount of water-level decline occurring from ~fuy 15 to September 15, 1980, in the measured water levels was applied to all constant-head nodes.
The third time period, September 16 to November 5, 1980 (51 days), was simulated by using only the 24 ft3/s of municipal pumpage. The simulated potentiometric heads computed at the end of the second time period were used as the starting heads for this period. Constant head nodes were assigned the paten-
75
B4 "oo'
v
v
,
EXPLANATION
- - - AREA OF DOUGHERTY PLAIN
IRRIGATION PUMPING CAPACITIES, IN GALLONS PER MINUTE PER SQUARE MILE
0
No systems
~~~~ 1000 to 3000
a
Less than 500
3000 to 5000
F
1!!1!!1 500 to 1000
B~;~se from U.S. Gel)lo..-to SV!'Yf 112.'50,000 q~adrono l e--~;
10
lo
20
I
30 Plllill
Figure 37- Locations and capacities of agricultural irrigation systems in the Dougherty Plain area as of spring 1980.
76
tiometric surface values at those locations for November 1980.
Areal measurements of water-table altitudes in the residuum (riverheads) were also not available for the second and third pumping periods (June 1 to Nov. 5, 1980). However, based upon a review of hydrographs, water-table values were set equal to the starting potentiometric head values of the principal artesian aquifer for pumping periods two and three. For grid nodes identified as river nodes, the water-table values were set equal to what was believed to be a reliable surface-water altitude of the stream or river at the start of the pumping period.
The storage coefficient of 5 x 10-3 was assumed not to vary significantly throughout the Dougherty Plain. Therefore, this value was used throughout the modeling area for all three pumping periods.
Measured water levels in the principal artesian aquifer for November 1980 and those simulated at the end of the transient simulation are presented in figure 38. The simulation error for November 1980 averaged 0.2 ft with a standard deviation of 3.4 ft. This was well within the desired criterion that 95 percent of all simulated heads be within
+ 10 ft of the derived data.
The simulated water levels for eight wells in the Dougherty Plain during May 15 to November 5, 1980, compare satisfactorily with measured water levels for this time period. Measured and simulated water levels in wells 087-10, 087-23, and 095-68 are shown in figure 39, and measured and simulated water levels in wells 201-05, 205-16, 253-08, and 253-26 are shown in figure 40.
It should be noted that the measured water levels represent a point value; whereas, the simulated water levels represent an average value for a 1 mi2 block. Therefore, while the fluctuation with time of simulated and measured values should be similar, actual simulated and measured values may differ considerably.
Simu lated Effects of Pumpage During A Hypothe t i c a l Dr oug ht a nd
During No r mal Recharge Condi tions
Transient model analyses were used to simulate changes in the potentiometric surface of the principal artesian aquifer and discharge to or recharge from overlying streams resulting from three sets of hydrologic conditions: (1) 1981 pumpage (municipal, industrial, and irrigation) during a hypothetical 3-year hydrologic drought, (2) 1981 pumpage plus the projected potential increase in irrigation pumpage during a hypothetical 3-year hydrologic drought, and (3) 1980 pumpage plus the projected potential increase in irrigation pumpage during a 10-year period of long-term average recharge conditions.
The transient model was used as previously calibrated for all predictive simulations with the exception of watertable altitude (riverhead) and storage coefficient, as explained below. Model results are presented as a series of maps showing ranges of water-level declines resulting from drought conditions (reduced recharge) and increased irrigation pumpage.
All simulations were made by using the water-table conversion option of the model (Trescott and others, 1976, p. 1112). As simulated water levels in the aquifer drop below the top of the aquifer, the initial artesian storage coefficient (0.005) converts to a predetermined water-table specific yield value (0.2). Also, to treat leakage more realistically, if parts of an artesian aquifer change to water-table conditions, the maximum head difference across the confining bed is limited to the difference between the altitude of the water table in the residuum (riverhead) and the top of the aquifer.
In the modeled area, ground-water withdrawals fro~ the principal artesian aquifer for irrigation use increased from 47 billion gallons per year in 1977 to about 76 billion gallons in 1980. Partly because of constantly increasing irriga-
77
3D'
v
v
31 "00'
EXPLANATION
D
AREA OF DOUGHERTY PLAIN
-120 -
SIMULATED POTENTIOMETRIC CONTOURShows altitude at which water level would ha ve s tood in tightly cased wells. Contour interval 20 feet. National Geodetic Vertical Datum of 1929
DATA POINT - Number is altitude of measured water level.
So~oe fro m U.S tolo9lc: al Su'n)' 1: 250,000 quod1ongles
F L
D A
10
15
20
25
30 MILES
Figure 38--Measured water levels and simulated potentiometric surface of the principal artesian aquifer, November 1980.
78
:).5
35
087-10 40
~ ~
' ..............,._
Simulated/ ''-...........,_ _ _ .-o
I
50
w
0
< IaL: ::l ss r-
"c z '
<
...J
""1
55
:;:
0
.w..J
eo
J4N.
FEB. M4R. 4PR. M4 Y
JUNE
JULY
4UG. SEPT. OCT. NOV,
DEC .
60 J4N.
FEB .
MAR .
APR. M4Y JUNE JULY
AUG. SEPT. OCT .
NOV . DEC.
.ID..
w w
.......,
IL
35
20
\0
~
_j
w
~f >w
...J
a:
.w..
:<;:
087-43
AA
451-
I
\
so ~
I \ }\(
>-.
so ,_
Simu lated'
-- \..- ......... ~~ .........
""c.....
............... .......
l "f I 09568
~'-/
Simulated
-1
SO r
I
l 351
./ '"- ' ' ' ' ' '
'
....,
-1
40
eo !J A N .
FEB . MAR .. 4PR . M4Y JUNE JULY AUG SEPT. OCT . NOV . DEC.
4S JAN.
FEB .
MAR. APR. M4Y JUNE JULY
o\ UG . SEPT , OCT . NO,V . DE C.
Figure 39.- Measured and simulated water levels in wells 087-10, 087-23, 087-43, and 095-68, 1980.
o,----.----.-----.----r----.-----.----.-----.----.----,-----,----,
201-05
10
253-08
10
,/'',,
20
' ' , , , / i mulated
15
........................
' ' ' , / s1mulated
w
<0...
~
::l
"c z '
o
<
..J
......,._ _ _ _ _ _ -o 20
/ Measured
25
"' ' ' ' ' ' 'c.._------o
:;::
0
.w.J
50 L_J_A_N-.~~F~E~B-.~~M-A-R-.-L-A-P-R-.-.L-M~A~Y~LJ~U~N~E--L-J-U_L_Y~--A-U_G_._LS~E~P=T -.L_o~c=T-. ~-N-O-V-.-L-D-E-C-.~
30 L_J~A~N~.~~F~E~B~.~7.M~A~R~.-L-A~P~=R.~~M~A~Y~~JU~N~E_L~J~U~L~Y~~A~U~G~._L~S~E~P~T~.~~O~C~T~._L~N7.0=v~._L~o=EC~-.J
m
....
00 0
w.w..
!
0,----.----.----.----.----.----.----,----,----.-----.----.----,
20 r-----.----.r-----r-----r-----r----.r-----.-----.-----.-----.-----.-----,
_;
w
> w
205-16
..J
253-26
.w~... 10
:<;::
20
30
\
\
\
40
').._
......... ....... ........ ........
~
Simulated/ '-......' ,"'o--_..---.....0
~ -....., / s1mulated
""' '
40
60
SO JAN. FEB. MAR. APR. MAY JUNE JUL Y AUG. SEPT. OCT. NOV. DEC.
rol-J7.A~N~.~~F~E~B~.~~M~A~R=._L-A~P=R-.~~M~A~Y~~J~U7.N=E_L_J~U~L~Y~~A~U~G=.-L~S~E~P~T=.~~O~C~T~. _L~N~O~V.-i_~D~E~C~.~
Figure 40.-Measured and simulated water levels in wells 201-05, 205-16, 253-08, and 253-26, 1980.
tion use, and partly because of a hydrologic drought that occurred from the summer of 1980 through the summer of 1981, irrigation withdrawals are estimated to have increased to about 107 billion gallons in 1981. Ground-water use for irrigation is expected to continue to increase throughout the area as additional land is converted to farm use and farmers become more and more dependent on supplemental irrigation to insure successful growth of two or three crops yearly. Average yearly water use for all purposes, other than irrigation, is about 6 billion gallons.
Projected potential increase in agricultural land within the Dougherty Plain area was estimated from county land-use maps prepared by the Soil Conservation Service (R. R. Pierce, U.S. Geological Survey, written commun., 1981). Projected pumpage was assigned to each square-mile block in the model based on the number of acres of potential agricultural land still available in that node for new or additional irrigation and an average application rate per acre (fig. 41). Potential additional irrigation pumpage was not assigned to urban or urbanizing areas, areas not suitable for irrigation by center-pivot systems, or in those counties that lie mostly outside of the Dougherty Plain. The potential additional projected irrigation pumpage under normal recharge conditions is estimated to be about 205 billion gallons per year and, under drought conditions, is estimated to be about 295 billion gallons per year.
Effects of Irrigation Pumpage During a Hypothetical 3-Year Drought
Model runs were made simulating water-level declines from initial low water levels (Nov. 1979) resulting from reduced recharge and from increased irrigation pumpage during two 3-year drought periods. A single irrigation season of 154 days (May-Sept.) was simulated for each year.
Recharge used in the model for the two drought simulations was estimated as follows: estimated mean recharge for
1981 (a drought year) was used as recharge for the year-1 simulation; 80 percent and 60 percent of the 1981 recharge were input as year-2 and year-3 recharge, respectively. The assumption was made that as the drought continued, water levels in the residuum would decline and, consequently, recharge (leakage from the residuum into the principal artesian aquifer) would decline accordingly. Actual recharge for years 2 and 3 of a 3year drought is unknown. The figures given here are, however, believed to be reasonable, based on the limited residuum water-level data available before and during the first year of the 1980-81 drought. Nevertheless, it must be emphasized that the simulation results to be discussed ?re valid only for the given set of recharge conditions. If recharge conditions during a concurrent 3-year drought are significantly different from those described above, simulation results would also be significantly different.
Pumpage of 113 billion gallons per year
The simulated mean declines in the potentiometric surface of the principal artesian aquif~r for drought years 1, 2, and 3 were, respectively, 18, 22, and 26 ft below the starting potentiometric surface. Simulated declines at the end of the hypothetical 3-year drought were generally less than 43 ft, but ranged from 43 to 61 ft in about 15 percent of the modeled area (fig. 42). In some areas, water levels fell from a few feet to 10 ft below the top of the aquifer (fig. 43). Figure 44 shows hydrographs of actual water-level declines due to the 1980-81 drought and projected waterlevel declines resulting from the simulated 3-year drought.
During the hypothetical 3-year drought, about half of the total pumpage of 339 billion gallons (321 billion gallons for irrigation and 18 billion gallons for all other) was derived from aquifer storage and half from recharge. Aquifer discharge to streams was considerably reduced, and all streams originating within the Dougherty Plain stopped flowing. Simulated flow of the Flint
81
;o'
30'
,
,
EXPLANATION
- - - AREA OF DOUGHERTY PLAIN
0Citl lto.m u.s. GII!OIOI'OIIc.al SW''WI<IIJ'
J 2.,0.000 CUJGdHrl"'9tiS'
R
0 A
10 ,. o
IRRIGATION PUMPING CAPACITIES, IN GALLONS PER MINUTE PER SQUARE MILE
CJ f No systems
! 2200
llillllll '100
-
3300
Figure 41.-Locations and capacities of projected potential irrigation systems in the Dougherty Plain area.
82
es"oo'
10'
32"00'
30' -
EXPLANATION
- - - DOUGHERTY PLAIN BOUNDARY
31"0 0'
WATER -L EVEL DECLINE, IN FEET
CJ Less than 10
..[JJ 10 10 25
C J 2.S to 50
Greater than 50
I0 A
Bou ftom U.S. Gaoloylcol 5ur\le)l 1 :2~0,000 quodronQlU
0
~
10
1.$
:0
U
30MILES
~lu~~~~--~~. ~----L---~--~
Figure 42 .- Simulated water-level declines in the principal artesian aquifer after pumping 113 billion gallons per
year for 3 years during a hypothetical hydrologic drought .
83
,. ,.
EXPLANATION
DOUGHERTY PLAIN BOUNDARY
WATER-LEVEL DECLINE BELOW TOP OF PRINCIPAL ARTESIAN AQUIFER, IN FEET
CJ 1!010
Figure 43.-Simulated water-level declines below the top of the principal artesian aquifer after pumping 113 billion gallons per year for 3 years during a hypothetical hydrologic drought.
84
Well 087-23
0 z <
....1
~ 0
.wm...1
1w- 50 w
LL
z 60
_j
w > w
....1
70 40
a:
w
1< -
50
~
60
70
Well 201-05
' \ \ \ -- \__ _,__- -_, -o.-__ --o
80 10
20
Well 2~3-12
30
40
60
60 1972
1973
1974
1976
EXPLANATION
- - - MEASURED WATER LEVEL - - - - SIMULATED WATER LEVEL, RESULTING FROM HYPOTHETICAL HYDROLOGIC DROUGHT
1983
Figure 44.- Measured and simulated water levels in the principal artesian aquifer in wells 087-23, 095-68, 201-05, 205-01, and 253-12.
85
River at t he end of t h e 3-year dro ught declined to about 8 00 f t 3/ s . Simulate d f l ows of Ichawaynoch a wa y, Ki n c haio one e, a nd Muckal ee Cre e ks were a bo ut 5 0, 1 00 , and 40. f t3/s , res pec tivel y , with mo st of t he fl ow be i ng de r i ved fro m o utsid e the Doughe rty Pl ai n. In compa r is on , measured f lows of t he Ic ha way nochaway , Kinc haf oonee, and Muckalee Creeks were 268, 105, and 77 ft~s, respectively, in August 1980, and were 144, 83, and 37 ft3/s, respectively, in July 1981. The effects upon streamflow of direct withdrawal of water from rivers for irrigation were not modeled. Consequently, quantitative comparisons of streamflow measurements with simulated streamflows should not be made.
Pumpage of 408 billion gallons per year
The simulated mean declines in the potentiometri c s urfa c e of t he princ i pal arte s i a n aquifer f or dr oHght ye ar s 1, 2, and 3 were , res~ec tiv ely , 25, 2 9 , an d 33 ft below t he s t art i ng po t entiome t r i c s ur f ace . Si mula t ed de clines a t t he en d of the hypothetical 3-year drought were generally less than 53 ft, but ranged from 53 to 73 ft in about 15 percent of the modeled area (fig. 45). Water levels declined from a few feet to 10 ft below the top of the aquifer in about 30 percent of the modeled area and more than 10 ft in some places (fig. 46).
During the hypothetical 3-year dr9ught simulation, the total pumpage of 1,224 billion gallons (1,206 billion gallons for irrigation and 18 billion gallons for all other) was supplied by aquifer storage (634 billion gallons), induced recharge from surface water (410 billion gallons), and recharge from the residuum (180 billion gallons). Most of the surface-water input to the aquifer was from the Flint River (water entering the Flint River upstream of the Dougherty Plain) and Lake Seminole (from lake storage and input from the Chattahoochee River). Mean flows of the Chattahoochee and Flint Rivers and Kinchafoonee Creek were severely reduced. All other streams stopped flowing.
Effects of Pumping 287 Billion Gallons Per Year with Normal Recharge
A 10-year transient simulation using mean annual hydrologic conditions and previously calibrated hydraulic parameters was made to determine the effects of long-term irrigation pumpage on water levels. Water-level declines are the difference between simulated water levels at the end of the 10-year simulation and yearly average water levels, as determined from November 1979 (low levels) and May 1980 (high levels) potentiometric maps. Pumpage input to the model consisted of 1980 pumpage (76 billion gallons for irrigation and 6 billion for other pumpage) plus projected potential irrigation pumpage (205 billion gallons per year).
The mean decline in the potentiometric surface at the end of the 10-year period was 4 ft, with the general range of decline being 0 to 9 ft. Maximum declines of 9 to 15 ft occurred in less than 15 percent of the modeled area. On a yearly mean basis, 2 billion gallons was removed from storage--less than 1 percent of the 287 billion gallons pumped. The remaining ?85 billion gallons came primarily from intercepted discharge to streams. Consequently, the main result of increased irrigation pumpage from the principal artesian aquifer would be slightly lowered water levels and about a 30-percent reduction in aquifer discharge to streams resulting in significantly reducing streamflow throughout the Dougherty Plain area.
SUMMARY AND CONCLUSIONS
The hydrologic character of the principal artesian aquifer in the Dougherty Plain, an area of about 4,400 mi2 in southwest Georgia, was investigated to determine if this aquifer is capable of supplying expected future increases in agricultural pumpage, especially during hydrologic droughts.
The Dougherty Plain, part of the Georgia Coastal Plain, is underlain by
86
30'
EXPLANATION
---DOUGHERTY PLAIN BOUNDARY
WATER-LEVEL DECLINE, IN FEET
C:::J Less than 10
QJ 10 to 25
CJ 25 to 50
. . Greater than 50
RI0 A
8CSP from U.S. .GeoiOQt OI S~f\li")'
IJ .2 5"0 ;o.oo quadrangles
c
10
"
Figure 45_- Simulated water-level declines in the principal artesian aquifer after pumping 408 billion gallons
per year for 3 years during a hypothetical hydrologic drought .
87
EXPLANATION
- - - DOUGHERTY PLAIN BOUNDARY
WATER-LEVEL DECLINE BELOW TOP OF PRINCIPAL ARTESIAN AOUIFER.IN FEET
I:==J 1 1o 10
10 to so
RI0 A
ou hort~ U.S, Geola r;~rc:o t Sutvcw 1: 250,000 quodron9les
0
~
t0
15
20
t.e.
;JQ MILES
Lwwu~~--L---~----L----~~
Figure 46 --Simulated water-/eve/ declines below the top of the principal artesian aquifer after pumping 408 billion gallons per year for 3 years during a hypothetical hydrologic drought.
88
sediments ranging in age from Upper Cretaceous to Holocene. The sediments consist of sand, clay, and carbonate rocks to thicknesses of more than 5,000 ft. The Dougherty Plain slopes gently to the south or southeast and averages about 160 ft above sea level. The plain is characterized by karst topography marked by numerous depressions or sinkholes, and is covered by about 25 to 125 ft of sandy clay residuum that contains silicified boulders. The plain is drained by the Chattahoochee and Flint Rivers.
Annual rainfall in the Dougherty Plain averages about 53 in. Rainfall for January through March and for June through August is about equal in magnitude (15 in.), but differs greatly in duration and distribution. Rainfall in the winter months is generally of long duration and moderate intensity; rainfall in the summer months is usually of short duration and high intensity. Most ground-water recharge from precipitation occurs from January through March. Rainfall during the summer months is generally lost to overland runoff to streams or to evapotranspiration.
Average annual runoff of eight watersheds was weighted by the basin drainage area within. the study area. The sum of the weighted produc-ts give an annual mean runoff rate for the Dougherty Plain area of 5,200 ft3/s; a spring high of 9,200 ft3/s; and a late-summer lowof 2,700 ft3/s. These quantities are the approximate average total annual, spring high, and late-summer low water yields of the Dougherty Plain area under average climatic and hydrologic conditions.
The base flow of streams in the Dougherty Plain is primarily groundwater discharge from the principal artesian aquifer. Therefore, base flow is a measure of the perennial ground-water yield of the principal artesian aquifer. Average annual mean base flow in the area of investigation was calculated as 4,000 ft3js. Average late-summer (Sept., Oct., and Nov.) mean base flow is considerably less--about 2,300 ft3/s. Total ~tream base flow during 7-day, 10-year minimum annual flows occurring within the Dougherty Plain area is about 1,600 ft3/s and is, probably, almost entirely discharge from the principal artesian aquifer.
89
Estimated vertical hydraulic conductivity of the residuum, which is generally sandy clay or clayey sand, varies from as low as 0.0001 ft/d to a high of 9 ft/d, the median being 0.003 ft/d. Estimated horizontal hydraulic conductivity varies from a low of 0.0004 ft/d to a high of 30 ft/d, with the median being 0.02 ft/d.
Small quantities of water are obtained from residuum wells throughout the study area. As to be expected, yields are highly variable and range from generally less than 1 gal/min to, in a few places, as much as 50 gal/min. Water levels ranged from about 1 to 38 ft below land surface from January to September 1981. Water-level fluctuations in individual wells ranged from about 2 to 14 ft, with an average fluctuation for all wells of about 6 ft.
Within the study area, the principal artesian aquifer consists primarily of the Ocala Limestone of late Eocene age. The Ocala, which is a light-colored, fossiliferous limestone, is a wedge-shaped formation trending from northe?st to southwest across Georgia, thickening to the southeast. The Ocala ranges in
thicKness from a few feet at the updip limit to about 350 ft in the southeastern part of the Dougherty Plain. The limestone is exposed along sections of major streams such as the Chattahoochee and Flint Rivers and Spring Creek, where erosion has removed the residuum.
Transmissivity of the principal artesian aquifer calculated from aquifer tests and ~stimated from specific capacities of wells ranges from 2,000 to 1,300,000 ft2/d. Transmissivity is lowest in the northern part of the report area where the aquifer is relatively thin, and increases to the south where the aquifer is thicker. Transmissivity is high near the Chattahoochee and Flint Rivers and Spring Creek, because water moving between the surface-water system and the ground-water system adjacent to these major recharge-discharge areas has accelerated the solution of ground-water conduits.
Storage coefficients for the aquifer
range from 2 x 10-4 ~o 3 x 12-2 , but are
generally in the 10- to 10- range. The storage values indicate that the princi-
pal artesian aquifer generally can b~ considered semiconfined: water in the aquifer is confined by a leaky confining bed (residuum) that allows significant vertical movement of water into or out of the aquifer.
Measured well yields in the Dougherty Plain area range from about 40 to 1,600 gal/min. Many wells in the area do not penetrate the full thickness of the aquifer and, consequently, yield less than the maximum possible rate. Yields of 1,000 to 2,000 gal/min, however, are common in areas where transmissivity exceeds 50,000 ft2/d, and where transmissivity exceeds 75,000 ft2/d yields of more than 2,000 gal/d may be expected. Measured specific capacities of wells, which are common measures of well yield, range from 4 to 1,000 (gal/min)/ft.
Recharge occurs chiefly from rainfall that leaks downward through the residuum during January through May. Most rainfall occurring during the summer months is lost to evapotranspiration or to soil moisture in the unsaturated zone of the residuum. Although the vertical
hydraulic conductivity of the residuum confining zone is generally low (about 0.003 ft/d), the cross-sectional area of flow in the vertical direction is large (about 4,400 mi ), and consequently large volumes of water can be transmitted through the residuum confining zone. Annual mean recharge to the principal artesian aquifer is estimated to be about 2,200Mgal/d (10 in.). Late-summer recharge is considerably less--about 1,400 Hgal/d (6 in.).
Because of a hydrologic drought that began in June 1980 and lasted through the summer of 1981, water levels in April 1981 were generally about 10 feet lower ~han water levels in May 1980. The small amount of rain that fell between June 1980 and April 1981 was not enough to recharge the aquifer to its normal spring level, and water levels remained about the same as in November.
The principal artesian aquifer transmits water from areas of recharge to natural areas of discharge and to wells. Natural outlets include springs, streams, and the overlying residuum or underlying Lisbon Formation, where hydro-
static pressure in them is less than in the principal artesian aquifer. About 90 percent of the annual ground-water discharge is to streams and springs.
Very little ground-water development has taken place in the Dougherty Plain except for irrigation. Under normal hydrologic conditions the ground-water system has not been significantly stressed, and over the long term is in a state of hydrologic equilibrium. Water is estimated to circulate through the steadystate hydrologic system at the rate of 4,300 ft3/s (2,800 Mgal/d) plus or minus about 860 ft3/s (560 Hgal/d).
Water in storage in the principal artesian aquifer could, in theory, supply the present pumpage requirements of the Dougherty Plain for a number of years. In practice, however, it would be unwise
to reduce water levels below the top of the principal artesian aquifer for any period of time because of increased possibility of sinkhole collapse, reduction or elimination of base flow to streams,
and increased well construction and pumping costs (most existing wells could not be pumped if water levels declined more than 10 to 20 ft below the top of the aquifer). Therefore, the desirable limit to water available from storage alone is about 50 billion gallons, or about half of the 1981 pumpage of 113 billion gallons.
Water samples from 16 residuum and 14 principal artesian aquifer wells were analyzed for concentrations of major inorganic constituents and for agricultural pesticides, herbicides, insecticides, and fungicides commonly used in southwest Georgia. While overall quality of water from the residuum and principal artesian aquifer is good, pesticides were detected in water from 11 residuum and 4 principal artesian aquifer wells. None of the water samples from the principal artesian aquifer, however, contained pesticide concentrations that exceeded the recommended limits for public drinking water.
A two-dimensional numerical model was constructed and calibrated to simulate water levels in the principal artesian aquifer. The digital model utilizes a finite-difference scheme to evaluate the partial differential ground-water
90
flow equations in which the head is the dependent variable.
The digital model was initially calibrated for steady-state water levels occurring as of November 1979. Model calibration was verified by a transient simulation of the period May to November 1980. Host heads simulated by the calibrated model were within 5 ft of measured heads. Simulated ground-water discharges were within 10 percent of discharges estimated from base-flow analyses.
Transient model analyses were used to simulate changes in the potentiometric surface of the principal artesian aquifer and discharge to or recharge from overlying streams resulting from applying three sets of hydrologic conditions: (1) 1981
pumpage of 113 billion gallons per year during a hypothetical 3-year hydrologic drought, (2) 1981 pumpage plus the projected potential increase in irrigation pumpage of 295 billion gallons per year during a hypothetical 3-year drought, and
(3) 1980 pumpage of 82 billion gallons per year plus the projected potential increase in irrigation pumpage of 205 billion gallons per year during a 10-year period of normal recharge conditions.
Simulated declines at the end of a 3-year drought with present pumpage of 113 billion gallons per year averaged about 26 ft and were generally less than 43 ft. Declines of from 43 to 61 ft occurred in about 15 percent of the modeled area. In some areas, water levels fell from a few feet to 10 ft below the top of the aquifer. Aquifer discharge to streams was considerably reduced, and all streams originating within the Dougherty Plain stopped flowing.
Simulated declines at the end of a 3-year drought with pumpage of 408 billion gallons per year averaged about 33 ft and were generally less than 53 ft. Declines of from 53 to 73 ft occurred in about 15 percent of the modeled area. Water levels declined from a few feet to 10 ft below the top of the aquifer in about 30 percent of the modeled area and more than 10ft in some places. Stream discharge to the aquifer exceeded aquifer discha~ge to streams. About half of the
pumpage came from aquifer storage and half came from surface-water discharge to the aquifer and leakage through the residuum.
A 10-year transient simulation using mean annual residuum water-table levels and previously calibrated hydraulic parameters was made to determine the effects of increasing irrigation pumpage under normal hydrologic conditions. Pumpage input to the model consisted of 1980 pumpage (82 billion gallons) plus projected potential irrigation pumpage (205 billion gallons per year).
The mean decline in the potentiometric surface at the end of the 10-year period was 4 ft, with the general range of decline being 0 to 9 ft. Maximum declines of 9 to 15 ft occurred in less than 15 percent of the modeled area. Over the 10-year simulation period, only 6 Hgal/d was removed from storage--about 1 percent of the amount pumped. Net discharge to streams, however, was reduced by 30 percent. Consequently, a major effect of increased pumping would be to reduce the base flow of streams in the Dougherty Plain area during the irrigation season.
SELECTED REFERENCES
Carter, R. F., and Putnam, s. A., 1978,
Low-flow frequency of Georgia streams: U.S. Geological Survey Water-Resources Investigations 77-127, 104 p. Carter, R. F., and Stiles, H. R., 1982, Average annual rainfall and runoff in Georgia, 1941-70: Georgia Geologic Survey Hydrologic Atlas 9, 1 sheet.
Clark, W. z., and Zisa, A. C., 1976,
Physiographic map of Georgia: Georgia Geological Survey. Cooke, C. W., 1943, Geology of the Coas-
tal Plain of Georgia: u.s. Geological
Survey Bulletin 941, 121 p. Hendricks, E. L., 1963, Compilation of
records of surface water of the United States, October 1950 to September 1960: U.S. Geological Survey WaterSupply Paper 1724, 458 p.
91
Hendricks, E. L., and Goodwin, lf. H., Jr., 1952, Water-level fluctuations in limestone sinks in southwestern Georgia: U.S. Geological Survey WaterSupply Paper 1110-E, 246 p.
Herrick, S. M. , 1961, Well logs of the Coastal Plain of Georgia: Georgia Geological Survey Bulletin 70, 462 P
Herrick, S. M., and Vorhis, R. C., 1963, Subsurrace geology of the Georgia Coastal Plain: Georgia Geological Survey Information Circular 25, 78 p.
Hicks, D. W., Krause, R. E., and Clarke,
J. s., 1981, Geohydrology of the
Albany area, Georgia: Georgia Geologic Survey Information Circular 57, 31 p. Johnston, R. H., 1977, Digital model of the unconfined aquifer in central and southeastern Delaware: Newark, Del., Delaware Geological Survey Bulletin 15, 47 p.
Johnston, R. H., Healy, H. G., and Hayes, L. R., 1981, Potentiometric surface of the Tertiary limestone aquifer system, Southeastern United States, May 1980: U.S. Geological Survey Open-File Report 81-486.
Lohman, S. W., 1979, Ground-water hydrau-
lics: u.s. Geological Survey Profes-
sional Paper 708, 70 p.
McCallie, S. w., 1898, A preliminary
report on the artesian well system of Georgia: Georgia Geological Survey Bulletin 7, 214 p. 1908, A preliminary report on the ---underground waters of Georgia: Georgia Geological Survey Bulletin 15, 376 p. Mitchell, G. D., 1981, Hydrogeologic data of the Dougherty Plain and adjacent areas, southwest Georgia: Georgia Geologic Survey Information Circular 58, 124 p. National Academy of Science, 1977, Drinking water and health: Washington,
D.C., p. 796. Newton, J. G., 1976, Early detection and
correction of sinkhole problems in Alabama, with a preliminary evaluation of remote sensing applications: Montgomery, Ala., State of Alabama Highway Department, Bureau of Materials and Tests, HPR Report No. 76, 83 p.
Owen, Vaux, Jr., 1958, Summary of groundwater resources of Lee County, Georgia: Georgia Geological Survey Mineral Newsletter, v. 11, no. 4, p. 118121. 1963a; Geology and ground-water re-
---sources of Lee and Sumter Counties, southwest Georgia: U.S. Geological Survey Water-Supply Paper 1666, 70 p. 1963b, Geology and ground-water re-
---sources of Mitchell County, Georgia: Georgia Geological Survey Information Circular 24, 40 p.
Pollard, L. D., Grant ham, R. G. , and Blanchard, H. E., Jr., 1978, A preliminary appraisal of the impact of agriculture on ground-water availability in southwest Georgia: U.S. Geological Survey Water-Resources Investigations 79-7, 22 p.
Radtke, D. B., McConnell, J. B., and Carey, W. P., 1980, A preliminary appraisal of the effects of agriculture on stream quality in southwest Georgia: U.S. Geological Survey Water-Resources Investigations 80-771,
40 P Riggs, H. C., 1963, The base flow reces-
sion curve as an indicator of ground water: Extract of Publication No. 63, International Association of Scientific Hydrology, p. 352-363. Rushton, K. R., and Tomlinson, L. M., 1971, Digital computer solutions of ground-water flow: Journal of Hydrology, v. 12, p. 339-362. Searcy, J. K., 1959, Flow-duration curves: U.S. Geological Survey WaterSupply Paper 1542-A, 33 p. Sever, C. W., 1965a, Ground-water resources and geology of Seminole , Decatur, and Grady Counties, Georgia: U.S. Geological Survey Water-Supply Paper 1809-Q, 30 p. 1965b, Ground-water resources of Bain---bridge, Georgia: Georgia Geological Survey Information Circular 32, 10 p. 1966, Reconnaissance of the ground ---water and geology of Thomas County, Georgia: Georgia Geological Survey Information Circular 34, 14 p. Spencer, J. W., 1891, First report of progress, 1890-91: Georgia Geological Survey Administrative Report, p. 5-10.
92
Stephenson, L. W., and Veatch, J. 0., 1915, Underground waters of the Coastal Plain of Georgia, and a discussion of the Quality of the water, by R. B. Dole: U.S. Geological Survey WaterSupply Paper 341, 539 p.
Stringfield, v. T., 1966, Artesian water
in Tertiary limestone in the Southeastern States: U.S. Geological Survey Professional Paper 517, 226 P Trescott, P. C., Pinder, G. F., and Larson, S. P., 1976, Finite-difference model for aquifer simulation in two dimensions with results of numerical experiments: U.S. Geological Survey Techniques of Water-Resources Investigations, Book 7, Chapter C1, 116 p.
U.S. Environmental Protection Agency, 1975, National interim primary drinking water regulations: Federal Register, v. 4, part IV, no. 248, December 24, 1975, P 59566-59588. 1976, National interim primary drink-
---ing water regulations: EPA-570/9-76-
003, 159 p. 1979, Water quality criteria: Federal ---Register , v. 44 , no 52 , p. 15 9 2 6-
15981. 1980, Water quality criteria documents ---availability: Federal Register,
v. 45, no. 231, p. 78318-78379. U.S. Geological Survey, 1970, Surface-
water supply of the United States, 196 1-6 5 : U S Ge o 1 o g i c a 1 Survey Water-Supply Paper 1906, 774 P
Wait, R. L., 1960, Source and quality of
ground water in southwestern Georgia:
Georgia Geological Survey Information
Circular 18, 74 p.
1963, Geology and ground-water re-
---sources of Dougherty County, Georgia:
U.S. Geological Survey Water-Supply
Paper 1539-P, 102 p.
Walton, W. C., 1970, Ground-water resour-
ces evaluation: New York, McGraw-Hill
Book Co., 664 p.
Watson, T. w., 1976, The geohydrology of
Ben Hill, Irwin, Tift, Turner, and
Worth Counties, Georgia: Georgia Geo-
logical Survey Hydrologic Atlas 2.
___1981, Geohydrology of the Dougherty
Plain and adjacent area, southwest
Georgia: Georgia Geologic Survey
Hydrologic Atlas 5.
Wells, J. v. B., 1960, Compilation of
records of surface water of the United
States through September 1950: u.s.
Geological Sur~ey Water-Supply Paper
1304, 399 p.
White, W. B., 1969, Conceptual models for
carbonate aquifers: Ground Water, v.
7, no. 3,
p.
Zimmerman, E.~, 1977, Ground-water re-
sources and geology of Colquitt
County, Georgia: u.s. Geological Sur-
vey Open-File Report 77-56, 70 p.
93
For convenience in selecting our reports from your bookshelves, they are color-keyed across the spine by subject as follows:
Red
Dk. Purple
Maroon
Lt. Green Lt. Blue Dk. Green
Olive
Dk. Blue Yellow
Dk. Orange
Brown Black Dk. Brown
Valley and Ridge mapping and structural geology
Piedmont and Blue Ridge mapping and structural geology
Coastal Plain mapping and stratigraphy
Paleontology Coast11l Zone studies Geochemical and geophysical
studies Economic geology Mining directory Hydrology Environmental studies Engineering studies Bibliographies and lists of
publications Petroleum and natural gas Field trip guidebooks Collections of papers
Colors have been selected at random and will be augmented as new subjects are publiched.
The Department of Natural Resources is an Equal Opportunity employer and employs without regard to race or color, sex, religion, and national origin.
Front Cover: Conceptual flow model of the principal artesian aquifer system.