– 48 feet mean low water (MLW), a Stakeholders Evaluation Group (SEG) was formed. The Aquifer Committee was formed by the SEG in response to concerns that were raised regarding the hydraulic integrity of the Upper Floridan confining unit. These concerns were raised by a co-author of a published report that identified outcrop exposures of Middle Eocene and older deposits having joint and fracture orientations reported to correspond with that of the regional tectonic stress field (Bartholomew et al., 2000).
The mission of the Aquifer Committee is “to identify the concerns regarding the potential effect upon the Floridan Aquifer of dredging the Savannah River navigation channel to the maximum depth required to maintain a project depth of - 48 feet MLW and to recommend to the SEG the scope of scientific or engineering investigation(s) and analysis (es) to address these concerns.” The Aquifer Committee met on several occasions during 2000 and 2001 and an outcome of these meetings was the establishment of a Working Group comprised of individuals having technical backgrounds in hydrogeology, geology, stratigraphy, geochemistry, geophysics, numerical modeling or other relevant scientific disciplines, as well as having direct experience in investigations of the hydrogeology of coastal Georgia; especially within the Savannah, Georgia-Hilton Head Island, South Carolina area.
The Working Group was charged with developing a conceptual Study Plan to be recommended to the Aquifer Committee and, in turn, to the SEG. As a first step in the process of developing the conceptual plan the Working Group first met during March 2001 to determine the relevant technical concerns the Plan of Study should address. The following specific technical concerns were decided upon for recommendation to the Aquifer Committee.
These five technical concerns were recommended to the Aquifer Committee, and were accepted by the Aquifer Committee without revision. The Aquifer Committee in turn recommended these technical concerns to the SEG.
The Working Group next met during July 2001 to determine what actions should be taken to address the technical concerns. Ten study tasks were discussed for recommendation to the Aquifer Committee. A Draft Plan of Study, based upon those tasks was prepared and distributed to the Working Group for review and comment.
The Working Group reconvened on December 6, 2001 to discuss and make a final decision upon the study tasks to be recommended to the Aquifer Committee. This conceptual study plan is based upon the tasks decided upon, and is designed to upon its implementation, provide the technical information necessary to determine the magnitude of the impact that deepening of the Savannah River navigation channel to a maximum authorized project depth of – 48 feet is likely to have upon the Upper Floridan aquifer.
Concerns regarding saltwater encroachment and intrusion into the Upper Floridan aquifer (previously referred to as the principal artesian aquifer and the Tertiary Limestone aquifer) in the Savannah Georgia-Low Country area of South Carolina date back to the early 1900s. As early as 1903, two water-supply wells at Parris Island South Carolina were taken out of service because of elevated chloride concentration (Landmeyer and Belval, 1996).
During the late 1930s, the U.S. Geological Survey began a systematic study of regional ground-water flow within the aquifer in coastal Georgia. The results of this study provided the first estimated pre-development potentiometric surface of the Upper Floridan aquifer, as well as maps of the 1943 potentiometric surface and the predicted potentiometric surface if pumpage at Savannah were to increase to 60 million gallons per day (mgd). This study (Warren, 1944) noted that water levels within the principal artesian aquifer had been lowered by from 70 to 100 feet below the pre-development piezometric level of approximately 37 feet above mean sea level. The report concluded that the significantly lowered head centered at Savannah could cause future saltwater contamination of the aquifer through inflow from areas where the aquifer may contain saltwater (Warren, 1944).
Counts and Donsky (1963) discussed three potential scenarios by which saltwater contamination into the principal artesian aquifer might occur. The first scenario identified was lateral migration of saltwater through the aquifer toward the center of the lowered piezometric level (Savannah) from areas where saltwater recharge may be occurring, such as Port Royal Sound, Beaufort River, and other areas where the upper confining unit might be thin or absent. Counts and Donsky (1963) estimated that at the 1963 pumping rate (approximately 60 mgd), less than a third of the total volume of water pumped at Savannah was being replaced by encroachment of salty water moving toward the pumping center. Two thirds or more of the recharge was estimated to be moving from areas where the principal artesian aquifer contained freshwater. The authors considered lateral saltwater encroachment through the aquifer as the most likely source of saltwater contamination of the principal artesian aquifer within the Savannah area and should receive first attention in future studies that might be undertaken.
The second mechanism of saltwater contamination considered by Counts and Donsky was vertical upward migration of salty water through the lower confining unit (saltwater intrusion). This scenario was considered unlikely because of the thickness of low permeability sediments between the bottom of the principal artesian aquifer and upper-most aquifer containing saltwater.
Lastly, Counts and Donsky discussed downward leakage from surface saltwater bodies through the Upper Floridan confining unit. They estimated that downward leakage through the Upper Floridan confining unit at the pumping center was about 37,000 gallons per day per square mile or about 58 gallons per day per acre. At distances of eight to ten miles from the pumping center, they estimated that downward leakage was about 9,300 gallons per day square mile or about 14.5 gallons per day per acre.
McCollum (1964) reported that saltwater encroachment into the principal artesian aquifer was a direct result of large ground-water withdrawals from a small area centered at Savannah. He identified two sources of saltwater contamination; the upper water-bearing zones were being contaminated in the Parris Island area by downward vertical leakage of modern-day saltwater into the aquifer at Port Royal Sound, and older “connate” saltwater in the lower water-bearing zones was moving toward Savannah. McCollum concluded that saltwater contamination in the Savannah area was minimized because only about 20 percent of the ground-water moving toward the center of the cone of depression was coming from areas where saltwater encroachment was taking place.
McCollum and Counts (1964) identified five major water-yielding zones within the principal artesian aquifer that are separated by relatively low permeability hydrostratigraphic units. In general, the upper water-yielding zones were reported to be thicker than the lower zones. Chloride concentrations were reported to increase in each water-yielding zone toward the east and northeast and with increasing depth. The authors also reported that saltwater was present in the upper permeable zones of the principal artesian aquifer at Parris Island, and the Beaufort River and Port Royal Sound areas. They estimated that under the 1964 pumping rate of approximately 62 mgd and the 1964 head decline of 160 feet at the pumping center, about 400 years would be required for saltwater in the upper permeable zone to reach the ground-water withdrawal points at the center of the cone of depression. The hydraulic gradient within the lower permeable zones were reported to be steeper and it was estimated that saltwater might reach the pumping center in about 90 years under the prevailing 1964 hydraulic conditions.
The Georgia Geologic Survey evaluated the potential for phosphate mining in eastern Chatham County during the late 1960s to determine whether mining would compromise the principal artesian aquifer and the extent, depth, grade, volume, and economic value of the phosphate matrix (Furlow, 1969). The study area consisted of that part of Chatham County south of the South Carolina-Georgia state line, north of Ossabaw Sound, east of most residential development, and within the three-mile oceanic limit. The phosphate matrix was reported to be primarily present at the base of the Duplin Marl (upper Miocene) that unconformably overlies the Hawthorn Formation in the Savannah area (Furlow, 1969). The Hawthorn Formation was described as consisting of tough sandy clays having low permeability and having an average thickness of about 40 feet throughout the study area. The Duplin Marl was reported to consist of olive-green sand, sandy clay and clayey sand and to be difficult to distinguish visually from the upper Hawthorn Formation. However, the phosphate content of the two formations was reported to differ significantly, with the Duplin Marl having the higher concentrations.
Furlow observed that the Duplin Marl and the Hawthorn Formation, in combination, form the upper confining unit of the principal artesian aquifer. He also reported that were mining to occur, the Duplin Marl would be excavated and removed since it forms much of the phosphate matrix. If the overburden and phosphate matrix were removed, only the Hawthorn Formation would remain in place as the upper confining unit within the mined areas. To determine the effectiveness of the Hawthorn Formation as a confining unit, 52 core samples were collected and analyzed to determine estimates of vertical permeability. The average vertical permeability reported for the Hawthorn Formation samples was 0.0096 gallons per day per square foot under one foot of hydraulic head (Furlow, 1969). Using the reported average value for vertical permeability of 0.0096 gallons per day per square foot, a vertical hydraulic gradient of 0.375, and an area of one square foot, it was estimated that about 0.0036 gallons of saltwater per day per square foot would leak through the Hawthorn Formation into the underlying principal artesian aquifer (about 160 gallons of saltwater per day per acre) if mining operations removed the overburden material and the part of the confining unit comprised of the Duplin Marl. Furlow further estimated that even without mining operations, saltwater was probably leaking into the principal artesian aquifer at a rate of similar magnitude.
Spigner and Ransom (1979) identified several actual and potential ground-water problems within the Low Country Capacity Use Area. The major problems noted were: regional water-level declines throughout large areas of the South Carolina Low Country and adjacent counties in Georgia; saltwater contamination of the Tertiary Limestone Aquifer (primarily Beaufort County); local well interferences; interaquifer transfer; inadequate requirements relating to well location, spacing, construction and abandonment; and lack of requirements for water-well reporting.
Krause, et al. (1984) reported that the development of an extensive cone of depression within the potentiometric surface of the principal artesian aquifer might be causing seawater encroachment from Port Royal Sound and adjacent areas. However, the authors also reported that the water-level and water-quality monitoring networks in the Savannah area were distributed so as to emphasize those parts of the principal artesian aquifer where potential saltwater encroachment problems may occur. The authors reported that although brackish-water zones underlying the principal artesian aquifer had chloride concentrations as high as 13,000 milligrams per liter, the water-level decline in the aquifer had not caused significant chloride concentration increases in the monitored wells during the period 1964 to 1984.
In 1970 the U. S. Geological Survey, in cooperation with the City of Savannah, Chatham County, and the State of Georgia began working on a digital model representing the principal artesian aquifer in the Savannah area. The resulting publication (Counts and Krause, 1976) reported the results of water-level changes resulting from computer simulations of redistribution of pumpage, increases and decreases in pumpage, or combinations of these changes. The two-dimensional finite-difference model was calibrated against Warren’s (1944) pre-development potentiometric surface of the principal artesian aquifer and the model was verified by comparing computed water-levels with water-level measurements made in 1956, 1960, and 1970. The model verification indicated reasonably good matches between computed and measured ground-water levels within the principal artesian aquifer.
This modeling effort was refined, updated, and enhanced by Randolph and Krause (1984). The steady-state model indicated that the ground-water flow system prior to development was sluggish and flow through the modeled area was only about 65 mgd. Simulation of the 1984 ground-water flow system with pumping stresses of about 85 mgd in the Savannah-Hilton Head area indicated flow through the modeled area of about 250 mgd. Vertical recharge from the overlying surficial aquifer more than doubled that of the pre-development flow system. A 25% increase in hypothetical pumpage within the model area was simulated to approximate future industrial and municipal withdrawal over the next 20 to 30 years. The simulation indicated a water level elevation of -165 feet mean sea level at the center of the cone of depression with a corresponding five-foot water level decline at a radius of 20 miles from the pumping center.
Other hypothetical pumping increases that were evaluated included doubling pumpage on Hilton Head Island and the introduction of agricultural pumping north of Savannah. On Hilton Head Island, the increase was about 9.7 mgd, distributed throughout the island. The simulation indicated a maximum drawdown of about eight feet on Hilton Head Island and about two feet in the Savannah area. The agricultural pumpage was represented by an increase of 29.7 mgd; the resulting cone of depression was centered at the northeast corner of Bulloch County, Georgia with a maximum water-level decline of about 29 feet.
The hydraulic conductivity value used for the upper confining unit in the various simulations was 8.6 x 10-5 foot per day was based on laboratory analysis and is representative of low-permeability clays and silts (Randolph and Krause, 1984).
Davis (1987) evaluated re-leveling surveys throughout the Atlantic Coastal Plain and reported that land subsidence at Savannah resulting from pumping the Floridan aquifer was insufficient to be considered a serious engineering concern, but was of more interest as a consideration in the adjustment of precise leveling and as a complicating factor in modeling the ground-water system. Artesian head declines showed marked acceleration during the mid 1930s, leveling off during the mid-1940s, acceleration from 1953 to 1963, and stability thereafter. Precise leveling completed in 1918, 1933, 1935, and 1955 indicated subsidence of as much as 100 millimeters had occurred, mostly since 1933. By 1933, an area of more than 130 square kilometers had subsided at least 20 millimeters. The observed subsidence was postulated to have been caused by declines in the head of the Floridan aquifer, which by 1955 had reached 50 meters, of which 40 meters of decline had occurred between 1933 and 1955. It was concluded that the limestone matrix of the Floridan aquifer was not compacting, but that the observed subsidence was more likely caused by compaction within the clays, marls, and silts of the overlying confining unit (Davis, 1987).
Bush and Johnston (1988) completed a Regional Aquifer System Analysis (RASA) model of the Floridan aquifer throughout the four states where it occurs. The regional flow system was found not to be significantly altered by development; however, long-term water-level declines of greater than 10 feet were reported within coastal Georgia and South Carolina among other areas. The authors classified the Upper Floridan aquifer in Savannah, Georgia-Beaufort, South Carolina area as being semi-confined based on the thickness of the upper confining unit (generally less than 100 feet), breaching of the confining unit, or both. The reported transmissivity of the Upper Floridan aquifer in this area was 10,000 to 50,000 feet squared per day. The leakage coefficient estimated from simulations for the confining unit in the Savannah area was reported to be less than 0.01 inches per year per foot of confining material.
The estimated potentiometric surface of the Upper Floridan aquifer in the Savannah area was from about 30 to 40 feet above mean sea level prior to development. By 1980, the reported potentiometric surface at Savannah was about –90 feet mean sea level and the zero foot contour extended well offshore (Bush and Johnston, 1988).
The ground-water flow system within the Floridan aquifer throughout southeastern Georgia and adjacent areas of Florida and South Carolina was simulated using a three-dimensional finite-difference digital model by Krause and Randolph (1989). The authors reported that the flow system in most of the area down-gradient of the Gulf Trough was characterized by slow lateral movement and that throughout the study area, almost all ground-water circulation was within the Upper Floridan aquifer. Ground-water withdrawal of approximately 625 mgd in areas concentrated down-gradient of the Gulf Trough was reported to have altered the flow system substantially. The estimated leakance through the upper confining unit in the Savannah area was reported to range from 10-5 to 10-4 foot per day per foot.
Clarke, et al. (1990) reported on the ground-water resources of coastal Georgia. The authors identified five aquifers that provide ground-water to consumers in the coastal area. In descending order these aquifers are: the surficial, upper Brunswick, lower Brunswick, Upper Floridan, and Lower Floridan. The upper and lower Brunswick aquifers were delineated and named during the course of the study. Wells open to the Upper Floridan aquifer may yield up to 10,000 gallons per minute and transmissivities of up to 500,000 feet squared per day in the southern part of the study area were reported. The transmissivity of the upper Brunswick aquifer was reported to be about 680 feet squared per day at Skidaway Island.
The Miocene sediments in coastal Georgia were described as three similar depositional sequences (in descending order Miocene units A, B, and C) of downward fining sediments that are indicative of three cycles of transgression/regression. Each of the units is comprised of a basal carbonate layer, a middle clay layer, and an upper sand. In the Savannah area the middle clay and upper sand layers are reported to be absent from Miocene unit C. The total thickness of the Miocene sediments is reported to be about 65 feet at Fort Pulaski (Clarke, et al., 1990)
In the Savannah area, chloride concentration in ground-water was reported to increase with depth, and possible saltwater encroachment from the ocean or intrusion through vertically upward leakage from deeper, saline zones into the Upper Floridan was reported as a concern. However, no elevated chloride concentration has been reported in the Upper Floridan in the Savannah area during the previous twenty years of ground-water monitoring. Downward leakage from the surficial aquifer into the Upper Floridan was reported to be most likely in Bulloch County; however, a potential for downward leakage exists in other inland areas within the western part of the study area and near large pumping centers at Savannah and Brunswick (Clarke, et al., 1990).
The Floridan aquifer system was simulated in detail in the Savannah-Hilton Head Island area by Garza and Krause (1996). The model was “telescoped” from the original regional RASA model of Krause and Randolph (1989) by using ground-water fluxes computed by the RASA model as boundary inputs to the new Savannah area model. This model, which simulates ground-water flow in both the Upper and Lower Floridan aquifers and leakage between aquifers and from above and below the aquifers, superseded that of Randolph and Krause (1984). The model stands as the model of use for the Floridan aquifer system in the area by the USGS, Georgia EPD, and others.
As a part of the Georgia Geologic Survey Evaluation of Miocene Aquifers in Coastal Georgia, an aquifer test was conducted at Tybee Island during March 1997 (Sharp et al., 1997). The pumping well was reported to have been open to the upper Brunswick aquifer; the reported transmissivity for the pumping well was 21,500 feet squared per day and the storativity of the aquifer tested was reported to be 0.0001. Water-levels in a nearby observation well open to the surficial aquifer did not respond to pumping, indicating that no measurable leakage through the confining unit separating the aquifers had occurred. It should be noted that the reported transmissivity of the water-yielding unit tested was well above that reported by Clarke et al. (1990) for the upper Brunswick aquifer, but is within the reported range of transmissivity for the Upper Floridan aquifer in the Savannah area, suggesting that the test well was open to the Upper Floridan aquifer rather than the upper Brunswick aquifer.
Jordan, Jones, and Goulding, Inc. (2001) at the request of Georgia EPD, conducted an assessment of the feasibility of conducting in-situ tests to determine estimates of the vertical hydraulic conductivity of the Upper Floridan confining unit using data from the Tybee Island aquifer test (1997) and nearby vicinity. Five scenarios were evaluated for the Tybee Island test site using different vertical hydraulic conductivities derived from laboratory and ground-water modeling leakance data and the thickness of the confining unit. Each scenario was evaluated twice using different values of specific storage for the Upper Floridan confining unit. The evaluation concluded that an aquitard test of 30 days duration or longer might yield useful data from which an estimate of the vertical hydraulic conductivity of the confining unit could be derived based on the results of 5 of the 10 scenarios calculated. It should be noted that this study is unpublished and has not been technically reviewed.
The Savannah Harbor Expansion Feasibility Study prepared for the U.S. Army Corps of Engineers-Savannah District was a multi-faceted study to determine the feasibility of expanding and deepening the existing Savannah Harbor navigation channel. The study was the first step in determining the engineering, environmental, and economic feasibility of the proposed project. A study of potential ground-water impacts was prepared as a part of a separate technical study and was an attachment to the Engineering Appendix of the feasibility study (U.S. Army Corps of Engineers, 1998). The ground-water study included geophysical investigations, drilling of core holes, determinations of vertical hydraulic conductivity, grain-size distribution and other geotechnical parameters, borehole geophysical logging, installation of observation wells, and a survey of water wells open to the surficial aquifer and deeper Miocene age hydrostratigraphic units.
As part of the U.S. Army Corps of Engineers-Savannah District study, the USGS conducted an evaluation of water chemistry and water levels as a means of evaluating the interconnection between the various units. These studies demonstrated the separation between the various surficial and Miocene zones, and the underlying Upper Floridan aquifer.
The range of estimated vertical hydraulic conductivity for 22 undisturbed samples collected from six boreholes within Miocene units A and B was 6.0 x 10-5 foot per day to 4.3 x 10-2 foot per day; with the average being 5.7 x 10-3 foot per day (2.0 x 10-6 centimeter per second). Four undisturbed samples were also collected from the in-fill material within relict stream channels incised into the Miocene units (paleochannels). The reported vertical hydraulic conductivities of these samples ranged from 8.5 x 10-4 foot per day to 2.3 x 10-2 foot per day, with the average being 7.7 x 10-3 foot per day
(2.7 x 10-6 centimeter per day). The reported vertical hydraulic conductivities were interpreted as being indicative of material having more than 15 percent clay and nearly 30 percent total fines (material passing a number 200 sieve. A vertically downward increase in clay content within the Miocene unit was reported (U.S Army Corps of Engineers, 1998). The study concluded that the proposed Savannah Harbor and channel deepening would have no significant impact upon the Upper Floridan aquifer.
HydroVision, Inc. (2000) conducted a review and evaluation of the 1998 Army Corps of Engineers report on behalf of the City of Savannah and suggested that insufficient study had been completed to definitively conclude that no significant impact to the Upper Floridan aquifer would occur as a result of harbor expansion. Specifically, HydroVision, Inc. commented: the range of transmissivity values reported for the Upper Floridan aquifer was too high, estimates of vertical hydraulic conductivity were based upon laboratory analyses rather than in-situ data, an arithmetic mean of the 22 estimated values of vertical hydraulic conductivity was used instead of a geometric mean, estimates of the thickness of the upper confining unit and hydraulic gradient values were questioned, the analysis was based on equilibrium conditions and did not consider temporal effects, and although the data appeared to be aerially well distributed, they were reduced to a single location for the purpose of calculating estimates of downward leakage through the confining unit.
Peck et al. (1999) reported that the water-level elevation near the center of the Savannah cone of depression was –97 feet mean sea level in May 1998. The authors also noted that the Savannah cone of depression is larger and deeper than other coastal Georgia cones of depression (Jesup-Doctortown, Brunswick, and St. Marys, Georgia-Fernandina, Florida) although the ground-water withdrawal rates are similar. This phenomenon was attributed to the generally lower transmissivity of the Upper Floridan aquifer in the Savannah area.
A field examination and evaluation of the potential hydrologic characteristics of fractures in Cenozoic sediments of the Georgia and South Carolina coastal plains was conducted by Law Engineering and Environmental Services, Inc. (2000) for Georgia EPD. Fractures in coastal plain sediments were reported to probably have some effect upon the hydraulic properties of the sedimentary material; however, the fractures in combination with the hydraulic properties of the material matrix control the hydrologic properties of a given hydrostratigraphic unit. The report indicated that fractures would have a limited role in the hydrogeology of the area.
The report suggested that rather than relying on large-scale aquifer testing, since such testing would be affected by both fracture and matrix permeabilities, an evaluation of regional ground-water flow should be conducted to evaluate the potential hydrologic significance of the combination of fracture and matrix hydraulic conductivities. For example, significant drawdown in the potentiometric surface of the Upper Floridan is well documented in the Savannah area; however, no corresponding decrease in hydraulic head of the overlying surficial aquifer has been observed (Law Engineering Law Environmental Services, Inc., 2000). It should be noted that this is an unpublished report that has not undergone technical review.
3.0 Recommended Technical Approach
Although a significant body of technical literature regarding the hydrogeology of the Savannah area exists, no study has been conducted that specifically addresses the technical concerns identified and recommended to the Aquifer Committee by the Working Group. The technical elements discussed during the July 2001 Working Group meeting served as the basis for a preliminary draft Plan of Study that was distributed to the Working Group for review and comment. The Working Group re-convened during early December 2001 to discuss and refine the specific tasks that should be included within the Study Plan to be recommended to the Aquifer Committee for consideration.
The technical tasks were discussed and voted on by the Working Group for recommendation to the Aquifer Committee during the December 2001 meeting. Some of the tasks were unanimously accepted for recommendation; others were not. In instances where there was not unanimous acceptance, the majority position prevailed. For example, the majority opinion of the Working Group regarding in-situ aquitard testing was that it should not be included for recommendation to the Aquifer Committee. Conversely, the minority opinion with regard to numerical modeling was that it should not be included for recommendation; however, since the majority opinion was in favor of numerical modeling as a task, it was retained for inclusion in the Draft Study Plan to be recommended to the Aquifer Committee. The December 6, 2001 meeting minutes provide discussion of majority and minority positions.
The tasks that are recommended to the Aquifer Committee are as follows:
1. Complete additional seismic reflection survey.
2. Complete a geologic transect through the radius of the cone of depression.
3. Construct monitoring well clusters.
4. Complete in-channel exploratory drilling.
5. Compile historical data research and develop database.
6. Develop a numerical flow and contaminant transport model of the hydrologic system, including and underlying the navigation channel.
The first five tasks are intended to provide the geologic and hydrologic data necessary to gain a better understanding of the nature and character of the Upper Floridan confining unit and to provide input data for the numerical flow and transport model (Task 6). The following paragraphs discuss the recommended technical tasks.
3.1 Task 1: Seismic reflection survey.
Previous riverine and offshore seismic reflection surveys indicated the presence of buried, sediment-filled, relict stream channels (paleochannels), that were probably down-cut during periods of lowered sea level that occurred during Pleistocene time and subsequently in-filled as sea level rose to its present level. The paleochannels are reported to be especially prevalent in the area of a structural high (variously referred to as the Tybee High or Beaufort Arch), from Fields Cut to several miles offshore of Tybee Island (U.S. Army Corps of Engineers, 1998). Inasmuch as the Upper Floridan confining unit is thinned over this structural high, down-cutting into or through it would potentially offer a preferential downward vertical pathway for seawater to enter the Upper Floridan aquifer in these locations.
As the initial task to address the concern that the hydraulic properties and geometry of the paleochannels are poorly understood, a seismic reflection survey is recommended to be completed along two parallel lines on either side of the navigation channel from about Fields Cut to as much as three miles offshore of Tybee Island. The South Channel should also be surveyed as well as several tie lines between the longitudinal survey lines along the navigation channel.
The data obtained from the recommended seismic survey would be used to more accurately determine the thickness of the Upper Floridan confining unit, gain better resolution of the geometry and orientation of paleochannels with respect to the navigation channel, and identify potential exploratory drilling locations along the navigation channel.
3.2 Task 2: Geologic transect.
A geologic transect from Tybee Island to the north end of Hutchinson Island is recommended to be completed by drilling exploratory borings at Tybee Island, Fort Pulaski, across the Savannah River from the northern end of Elba Island, and the north end of Hutchinson Island. An exploratory boring was drilled at the U.S. Highway 80 Bridge over Bull River (Bull River site) during March, 2001 and would be included as part of the transect; however, additional exploratory borings would not be necessary at this site. In addition to the boring locations within the transect, it is recommended that an exploratory boring be drilled near Ridgeland South Carolina as a control site (a location of similar lowered head in the Upper Floridan, but overlain by freshwater rather than saltwater).
Each boring would be drilled through the full thickness of the undifferentiated surficial sediments and the Upper Floridan confining unit into the top of the Oligocene limestone (Upper Floridan aquifer). Should the seismic reflection survey suggest that onshore extensions of paleochannels exist in accessible areas, such locations would be drilled in addition to the locations indicated above.
The exploratory borings would be drilled using Rotosonic drilling or conventional mud-rotary techniques to obtain continuous cores through the full thickness of the undifferentiated sediments and the Upper Floridan confining unit. All cores that are recovered would be lithologically described in the field and selected samples would be collected for analyses of porosity, grain-size distribution, vertical hydraulic conductivity, and pore-water geochemistry and naturally emplaced radioisotope analyses. A suite of down-hole geophysical logs to include velocity, spontaneous potential, electrical resistivity, and natural gamma suites would be completed at each exploratory borehole location.
All down-hole data would be evaluated, stratigraphic units correlated, and a geologic section developed. Pore-water geochemical data, especially chloride concentrations, would be evaluated to develop concentration profiles that will indicate the current status of saltwater transport into the Upper Floridan confining unit. Vertical hydraulic conductivity data derived during this phase would be incorporated into a data set that includes historical data reported by previous investigators.
Completion of this task would provide the hydraulic, ground-water quality, and geologic characteristics necessary to evaluate saltwater transport and leakage rates through the Upper Floridan confining unit. The data obtained from completion of this task could be used to determine the relation between the downward transport of saltwater and 1) head decline in the Upper Floridan aquifer caused by pumping and 2) thinning of the Upper Floridan confining unit caused by previous harbor and channel dredging. The geologic transect boring locations would subsequently be used as locations for monitoring well clusters (Task 3), the lithologic data obtained during completion of Task 2 will provide subsurface control for determining optimum screened intervals and other appropriate well construction details.
3.3 Task 3: Monitoring well clusters.
Monitoring well clusters should be installed at each of the previously identified transect locations. The Hutchinson Island site would have the highest priority for installation because that location is nearest the pumping center and could be used to monitor head changes within the Upper Floridan confining unit in the event that a significant change in pumpage should occur. An appropriate number of monitoring wells (probably not exceeding six per site) should be installed at each cluster location so that the hydraulic head within the entire thickness of the Upper Floridan confining unit could be monitored. In addition, a monitoring well would be installed into the upper part of the Upper Floridan aquifer.
As an alternative to constructing multiple wells within each cluster, a single, fully penetrating string of pressure transducers could be used. The pressure transducers would be attached to the string at intervals to be specified based on the lithology determined at each site during Task 2. However, the pressure transducers would be limited to providing only hydraulic head data; ground-water sampling could not conducted.
The monitoring well clusters would be used to determine the hydraulic head relations both within the Upper Floridan confining unit and between the confining unit and the Upper Floridan aquifer. These data in turn could be used to determine the magnitude of the vertically downward head gradient from the surficial aquifer and surface water bodies to the Upper Floridan aquifer. The vertical hydraulic gradient is a necessary component in determining estimates of ground-water flow rates and salinity transport.
3.4 Task 4: In-channel exploratory drilling.
After reviewing and evaluating the data derived from Task 1 (seismic reflection survey) and Task 2 (geologic transect), continuous core borings should be drilled at locations within the navigation channel to confirm the thickness and characteristics of the Upper Floridan confining unit where it transects the previously identified paleochannel features.
All cores that are recovered would be lithologically described and selected samples collected for analyses of porosity, grain-size distribution, vertical hydraulic conductivity, and pore-water geochemistry and naturally emplaced radioisotope analyses. Down-hole geophysical logs, including velocity, spontaneous potential, electrical resistivity, and natural gamma suites would be run at each exploratory borehole location.
The in-channel drilling program would focus mainly on determining the thickness and vertical hydraulic conductivity of the paleochannel infill sediments and the salinity of pore water at various depths within the infill sediments. This task would provide direct information regarding the hydrogeology of the most critical areas to be dredged with respect to vertical saltwater encroachment into the Upper Floridan aquifer, as well as a determination of the current, pre-dredging saltwater distribution within the paleochannel in-fill sediments.
3.5 Task 5: Historical data research and database development.
Prior to development of a numerical model of the hydrologic system underlying the navigation channel, relevant historical data should be researched and incorporated into a digital database. For example, the range of vertical hydraulic conductivity values obtained from previous studies and historical data regarding ground-water pumpage from the Upper Floridan aquifer and dredging activities within the Savannah Harbor area would be input into the database.
Historical data are necessary to model pre-dredging conditions that will provide the base-line conditions regarding saltwater transport into the Upper Floridan confining unit.
3.6 Task 6: Develop numerical model of the hydrologic system including and underlying the navigation channel.
It is recommended that a numerical ground-water flow and solute transport model be developed for the areas that might be affected by the proposed harbor expansion. Simulations would be made using the data acquired during completion of the previous tasks as input parameters to predict the change in rate and magnitude of brackish water or saltwater transport into and through the Upper Floridan confining unit.
The model would be calibrated and verified by comparing simulations to the observed salinity distribution within the Upper Floridan confining unit. Upon satisfactory calibration of the model, a pre-dredging simulation would be completed to establish the existing saltwater distribution within the Upper Floridan confining unit, as well as within the Upper Floridan aquifer. Post-dredging simulations for each deepening alternative would then be conducted and the predicted salinity transport for each alternative compared to the existing conditions to assess the magnitude of potential saltwater encroachment of the Upper Floridan aquifer resulting from the proposed harbor expansion.
The model would be sufficiently flexible to handle dynamic conditions, such as changes in hydraulic gradient and salinity over time. The simulations would include “no additional dredging” scenarios projected into the future, as well as simulations of the various dredging alternatives. Such simulations would provide comparisons of predicted saltwater encroachment rate and magnitude under no project conditions versus encroachment rates and magnitudes predicted to occur under the various harbor expansion alternatives. These comparisons could be used for future decision- making.
It is recommended that the numerical model code be a widely accepted, public domain one
such as is available through the U. S. Geological Survey. The navigation
channel would be modeled as a line source of brackish water or saltwater chloride;
boundaries and boundary conditions would be specified as part of the development process
of the model.
Bartholomew, M.J., Rich, F.J., Whitaker, A.E., Lewis, S.E., Brodie, B.M., and Hill, A.A., 2000. Preliminary Interpretations of Fracture Sets in Upper Pleistocene and Tertiary Strata of the Lower Coastal Plain in Georgia and South Carolina. A Compendium of Field Trips of South Carolina Geology with Emphasis on the Charleston, South Carolina Area. Geological Society of America-Southeastern Section Meeting, March 23-24, 2000.
Bush, P.W. and Johnston, R.H., 1988. Ground-Water Hydraulics, Regional Flow, and Ground-Water Development of the Floridan Aquifer System in Florida and in Parts of Georgia, South Carolina, and Alabama. United States Geological Survey Professional Paper 1403-C.
Clarke, J. S., Hacke, C. M., and Peck, M.F., 1990. Geology and Ground-Water Resources of the Coastal Area of Georgia. Georgia Geological Survey Bulletin 113. Prepared in cooperation with the Department of the Interior, U.S. Geological Survey, Water Resources Division and Georgia Department of Natural Resources Environmental Protection Division.
Counts, H.B. and Donesky, E., 1963. Salt-Water Encroachment, Geology and Ground-Water Resources of Savannah Area Georgia and South Carolina. United States Geological Survey Paper 1611. Prepared in cooperation with the Georgia Department of Mines, Mining, and Geology, the City of Savannah, and Chatham County.
Davis, G.H., 1987. Land Subsidence and Sea Level Rise on the Atlantic Coastal Plain of the United States. Environmental Geology and Water Science, Vol. 10, Number 2.
Furlow, J.W., 1969. Stratigraphy and Economic Geology of the Eastern Chatham County Phosphate Deposit. Georgia Geologic Survey Bulletin 82. Georgia State Division of Conservation, Department of Mines, Mining, and Geology.
Garza, Reggina and Krause, R.E., 1996. Water-Supply Potential of Major Streams and the Upper Floridan Aquifer in the Vicinity of Savannah, Georgia: U.S. Geological Survey Water-Supply Paper 2411.
HydroVision, Inc., 2000. Review Comments and Recommendations Regarding the Study and Report: “Potential Ground-Water Impacts-Savannah Harbor Expansion Feasibility Study.” Prepared for the City of Savannah.
Jordan, Jones, and Goulding, Inc., 2001. Assessment of the Feasibility of Conducting an Aquifer Test at the Tybee Island Test Site to Determine Vertical Hydraulic Conductivity of the Miocene Confining Beds, Savannah Harbor Deepening Project. Prepared for Georgia Department of Natural Resources, Environmental Protection Division, Georgia Geologic Survey.
Krause, R.E., Matthews, S.E., and Gill, H.E., 1984. Evaluation of the Ground-Water Resources of Coastal Georgia: Preliminary Report on the Data Available as of July 1983. Georgia Department of Natural Resources, Environmental Protection Division, Georgia Geologic Survey, Information Circular 62. Prepared in cooperation with the U.S. Geological Survey.
Krause, R.E. and Randolph, R.B., 1989. Hydrology of the Floridan Aquifer System in Southeast Georgia and Adjacent Parts of Florida and South Carolina. United States Geological Survey Professional Paper 1403-D.
Landmeyer, J.E. and Belval, D.L., 1996. Water-Chemistry and Chloride Fluctuation in the Upper Floridan Aquifer in the Port Royal Sound Area, South Carolina, 1917-93. United States Geological Survey, Water-Resources Investigations Report 96-4102. Prepared in cooperation with the South Carolina Department of Natural Resources, Water Resources Division.
Law Engineering and Environmental Services, Inc., 2000. Report of Evaluation of Potential Hydrologic Significance of Fractures in Coastal Plain Sediments. Prepared for Georgia Department of Natural Resources, Environmental Protection Division, Georgia Geologic Survey.
McCollum, M.J., 1964. Salt-Water Movement in the Principal Artesian aquifer, Georgia and South Carolina. Ground Water, Journal of the technical Division, National Water Well Association, Vol. 2, Number 4.
McCollum, M.J. and Counts, H.B., 1964. Relation of Salt-Water Encroachment to the Major Aquifer Zones Savannah Area, Georgia and South Carolina. United States Geological Survey, Water-Supply Paper 1613-D. Prepared in cooperation with the Georgia Department of Mines, Mining, and Geology, the City of Savannah, and Chatham County.
Peck, M.F., Clarke, J.S., Ransom, C., and Christopher, J., 1999. Potentiometric Surface of the Upper Floridan Aquifer in Georgia and Adjacent Parts of Alabama, Florida, and South Carolina, May 1998 and Water-Level Trends in Georgia, 1990-98. Georgia Department of Natural Resources, Environmental Protection Division, Georgia Geologic Survey. Prepared in cooperation with the United States Department of the Interior, United States Geological Survey.
Randolph, R.B. and Krause, R.E., 1984. Analysis of the Effects of Proposed Pumping from the Principal Artesian Aquifer, Savannah, Georgia. U.S. Geological Survey, Water-Resources Investigations Report 84-4064. Prepared in cooperation with the U.S. Army Corps of Engineers.
Sharp, B., Watson, S., and Hodges, R.A., 1997. Aquifer Performance Test Report, Tybee Island (Upper Brunswick) Aquifer, Chatham County, Georgia (Preliminary: Subject to Review). Department of Geological Sciences Clemson University.
Spigner, B.C. and Ransom, C., 1979. Report on Ground-Water Conditions in the Low Country Area, South Carolina. South Carolina Water Resources Commission Report Number 132.
United States Army Corps of Engineers, Savannah District, 1998. Potential Ground-Water Impacts Savannah Harbor Expansion Feasibility Study.
Warren, M.A., 1944. Artesian Water in Southeastern Georgia With Special Reference to the Coastal Area. Georgia State Division, Department of Mines, Mining and Geology. Geological Bulletin 49. Prepared in cooperation with the United States Geological Survey.