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 incrD' 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 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 \; ' :; (.) 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, "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 (.) w _J aw: 30 f- s<( 35 40 I 0 w 0 ci ~ Cw L J (f) I 0 w 0 1971 . a: 1-' CL w :2 "") (f) I, 0 w 0 a: Wz :2 J 1-' I CL 0 w w (f) 0 1972 1973 a: Wz :2 J f-' Cw L (f) I wo' 0 a:' J Cw L (f) 1, 0 w 0 . a: z w 2 J t-' Cw L (f) I o' w 0 1975 1976 a: 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 >~..: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 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.