Memorandum

To: Aquifer Committee Working Group

From: John Cox

Date: March 18, 2001

Re: Points of discussion during first meeting

The following paragraphs summarize the substantive points of discussion and consensus as I recall them. Should your recollection be different than those provided herein, please communicate such to me at your earliest convenience.

1) The working group convened at approximately 9:00 AM March 13, 2001. The attendees and participants and their affiliations were as follows:

Larry Keegan (Lockwood Green) attended as an observer. John Cox served as Chairman.

2) The first few minutes of the meeting were devoted to introductions of the Working Group participants and brief summaries of their respective areas of expertise and interest. Upon completion of the introductions and opening remarks. John Cox stated that the objective of the first meeting was to reach a consensus regarding the scientific issues and concerns directly related to the proposed deepening of the Savannah River Navigation Channel and the subsequent impact to the Upper Floridan aquifer. The concerns will then be listed and recommended to the Aquifer Committee at the next Aquifer Committee meeting. The list of technical concerns will be used to identify data gaps and ultimately lead to the development of a Draft Aquifer Study Plan, which in turn, will be recommended to the Aquifer Committee in the future.

3) It was suggested by the Chairman that some time be spent in a discussion of the historic and existing conditions of the hydrogeologic system within the area of the Savannah River and environs.

4) This discussion led to the following major points:

a) The cone of depression centered at Savannah has grown in extent to its present day configuration due to relatively low transmissivity of the Upper Floridan aquifer in the Savannah area. Pumping at Savannah and Hilton Head has caused salt water intrusion to occur. This process was first identified at Dolphin Head on Hilton Head Island.

b) The top of the Upper Floridan aquifer occurs at shallow depth near Beaufort, South Carolina (Port Royal Sound).

c) The existing net vertical hydraulic gradient from the surficial aquifer to the Upper Floridan aquifer is downward.

d) Preliminary work by Camille Ransom near Savannah, indicated that there appears to be a vertical salinity profile in pore water within the Miocene confining unit. Camille indicated that he had found chloride concentrations ranging from about 8,000 mg/l near the top of the confining unit to 6,000 mg/l at 50 feet depth to about 1,000 mg/l at 90 feet depth. Camille raised the question as whether the salinity profile observed was the result of connate water being flushed out of the confining unit or saline shallow water migrating vertically downward. Camille indicated that he plans to radiometrically date pore water samples during the next investigation. Field work will begin on April 9, 2001 by using a Rotosonic drill rig to collect core samples from the Miocene and Upper Oligocene sections at a site on U.S. Hwy 80, near Bull River.

e) John Clarke indicated that off-shore drilling program had drilled core holes at distances of seven, ten, and fifteen miles off-shore of Tybee Island. The 15 mile well produced saline water; the 10 mile well contained fresh water at the top of the Upper Floridan, but chloride concentrations increased with depth, and that the 7 mile well contained saline water near the top of the Upper Floridan aquifer in area where the upper confining unit was thin.

5) As a starting point for the discussion of technical concerns, two points of agreement were reached:

a) The net vertical hydraulic gradient from the surficial aquifer to the Upper Floridan aquifer within the Savannah cone of depression is downward, this includes the area of proposed harbor expansion; and b) Leakage through the Miocene confining unit will increase if additional material is removed; ie. the rate of leakage and quantity of leakage will increase.

6) The following technical concerns were agreed upon by consensus:

a) How much of a rate and quantity of leakage increase resulting from harbor deepening is unknown.

b) Changes in chloride concentrations with time in the Upper Floridan aquifer caused by harbor deepening are unknown.

c) Hydraulic properties, salinity and hydraulic head of the Miocene unit are poorly known, as is the variability of these parameters.

d) Hydraulic properties and geometry of the previously identified paleochannels are poorly understood.

e) The geologic frame work is not as well refined as it should be.

7) Following the lunch break, Lynn Torak (USGS-Georgia District) made a presentation of how geostatistical techniques could be applied to the problem. Although, Lynn's presentation was more focused on potential methodology rather than specific technical concerns or data gaps, his presentation was well received, and it directly addressed how some of the relevant data gaps could be addressed. The specific major points of the presentation were as follows:

a) The complex and heterogenous nature of the system reflects geologic, hydrologic and geochemical processes including lithologic variations, contaminant sources, and changes in flow paths.

b) The overall problem includes spatial analysis, data set comparisons, computation of upper confidence levels, determination of design values, and estimation/prediction of contaminant concentrations.

c) Classical statistical analyses may be inadequate because the date are clustered and biased.

d) The objective of geostatistics are to calculate the most accurate predictions, quantify the accuracy of the predictions, design optimal sampling plans, and to quantify the worth of the data.

8) The first day meeting adjourned at approximately 3:45 PM. Although the objective for the first meeting had been met, it was decided to re-convene the following day to discuss potential methodologies to address the concerns listed.

9) The second day meeting convened March 14, 2001 at approximately 9:00 AM. The following persons attended the second day of the meeting:

John Clarke

Bill McLemore

Harold Gill

Rick Krause

Jim Henry

Cardwell Smith

Camille Ransom

John Cox

Larry Keegan (Observer).

10) Bill McLemore raised the question of gaining a better understanding of the paleochannels through additional seismic surveying. Jim Henry indicated that approximately a week of additional seismic work would be required. He suggested running two parallel seismic lines, one on each side of the existing navigation channel from about Fields Cut to two or three miles off-shore with several tie lines included. Jim indicated that the paleochannel infill seems to be coarser grain size than most of the Miocene sediments. He also raised an issue regarding slope stability around one of the slurry ponds across from the former Union Camp facility. Larry Keegan indicated that GPA is aware of this issue and that USACE has been checking on the stability concerns.

11) The following approach was suggested to resolve concerns about the paleochannels:

a) Complete the additional seismic work as described above and attempt to track one or more of the paleochannel features on-shore so that it can be cored or investigated with a DPT rig. Preferably, cores could be collected and pore water samples would collected to determine salinity profiles.

b) A jack-up rig would be used to drill areas within the existing navigation channel.

c) use geostatistical analyses to plan optimal drilling locations.

d) Three of the identified paleochannels have been drilled, there appear to be five deep paleochannels located over the Tybee High; the additional paleochannels should be investigated.

12) Considerable discussion focused upon the range of reported values of vertical hydraulic conductivity for the Miocene sediments and how to better define this parameter. All seemed to be in agreement that the scale of a pumping test would be preferable to that afforded by laboratory tests; however, the feasibility of conducting a meaningful pumping test was questioned, primarily because of the volume of water that will probably have to be removed to induce drawdown effects within the Miocene unit. It was agreed that Harold Gill will complete an assessment of the available data to gain a better understanding of the feasibility of conducting a successful pumping test. Should a pumping test not be feasible, or if a pumping test is conducted but is unsucessful, then laboratory hydraulic conductivity data will have to suffice.

13) It was suggested that the understanding of the geologic framework could be improved by constructing cross-sections and transects, especially in the up-river reaches. The cross-sections should identify the Miocene A and B units. John Clarke recommended that all core samples should be carefully inspected for joints and fractures. Bill McLemore pointed out that the existing hydrostatic and lithostatic pressures are probably large enough to keep any joints or fractures that may be present closed.

14) John Clarke recommended using the existing monitoring wells as locations for well clusters. The USGS can conduct sampling and analyses at monitoring well clusters at Tybee Island and Ft. Pulaski. He also suggested completing radio-isotope analyses on samples collected from these wells and from the Rotosonic borehole that will be drilled in early April. Additional monitoring well clusters should be considered at the Bull River Rotosonic boring site (Camille Ransom's project), Elba Island, and at two spoil areas on the South Carolina side of the Savannah River (up-stream from Elba Island). It was also suggested that the Rotosonic boring (Bull River site) be converted into a monitoring well, screened within the Upper Floridan aquifer and that additional monitoring wells be completed within selected intervals in the Miocene section.

Bill McLemore indicated that GAEPD could subsidize the construction of two or three Miocene well clusters.

15) John Cox suggested that geotechnical parameters such as grain-size distribution, Atterberg Limits, etc. also be analyzed from core samples collected during the April 9-11 drilling program.

16) Rick Krause stated that in order to predict the leakage increase resulting from harbor deepening some level of numerical modeling would be necessary. In order to complete a model; the history of pumpage would be needed, the history of dredging would be needed, and the history of chloride concentrations or surrogate values would be needed.

17) Bill McLemore indicated that it may be possible to fast-track funding for development of SUTRA-3D model for the Savannah River. Rick responded that a 2-D model should be used first to predict vertical migration of salt water. If the 2-D profile shows that leakage is nominal or otherwise insignificant, then it probably wouldn't be necessary to address the lateral transport component.

18) John Cox indicated that he has developed a simple spread sheet analysis of leakage and chloride concentration scenarios. The spread sheet would be distributed to the Working Group for evaluation (this has been done).

19) It was agreed by the Working Group that an initial phase (Phase I) of limited field work and analysis should be completed. The limited field work would include:

- completion of nested monitoring wells within the Miocene section

- completion of the seismic survey as described by Jim Henry and Card Smith

- completion of additional core holes

- isotope analyses.

These new data would then be used to simulate possible chloride transport through the Miocene section into the Upper Floridan. Long-term monitoring of the Upper Floridan would be recommended, regardless of the level of future study.

The meeting was adjourned at approximately 12:00 PM.