Assessment of Groundwater Flow Near the Savannah River Site, Georgia and South Carolina
The U.S. Department of Energy's (DOE) Savannah River Site (SRS) has manufactured nuclear materials for the National defense since the early 1950s. A variety of hazardous materials including radionuclides, volatile organic compounds, and trace metals, are either disposed of or stored at several locations at the SRS. As a result, contamination of ground water has been detected at several locations within the site and concern has been raised over the possible migration of water-borne contaminants offsite. Two issues have been raised:
- Is ground water flowing from the SRS and beneath the Savannah River into Georgia?
- Under what pumping scenarios could such ground-water movement occur?
Objectives
The U.S. Geological Survey, in cooperation with the DOE, conducted a comprehensive study during 1991-97 that simulated groundwater flow and stream-aquifer relations in the vicinity of the SRS. These groundwater simulations are limited by simplification of the conceptual model, which was based on available data through 1992. Large increases in groundwater pumping in Burke and Screven Counties, Georgia, since 1992, and a pronounced drought during 1998-2002 may have changed hydraulic gradients near the river and affected the potential for trans-river flow. To provide a more accurate and up-to-date evaluation of trans-river flow near the SRS, the earlier model is being updated to incorporate new data and simulate 2002 conditions. The revised model will be used to simulate a variety of water-management scenarios that could impact trans-river flow in the SRS area.
- Update the previously developed groundwater flow model to better define present-day (2002) ground-water flowpaths near SRS.
- Utilize the 2002 calibrated model to identify groundwater flowpaths and quantitatively describe current groundwater flowpaths near SRS under a variety of hypothetical pumping scenarios.
Progress and Significant Results, 2004—2005
- The previous model (Clarke and West, 1998) was updated to simulate ground-water flow under 2002 hydrologic conditions and for four hypothetical pumping scenarios based on ground-water-use trends from 1980 to 2000 (Fanning, 2003).
- Four steady-state pumping scenarios were developed to simulate a range of pumping and climatic conditions affecting potential contaminant migration from the SRS:
- 2002 observed pumping and boundary conditions for an average year.
- 2002 observed pumping and boundary conditions for an average year with SRS pumping discontinued.
- Projected 2020 pumping and boundary conditions for an average year.
- Projected 2020 pumping and boundary conditions for a dry year.
- The USGS particle-tracking code MODPATH (Pollock, 1994) was used to generate advective water-particle path lines and time-of-travel based on MODFLOW simulations of the four scenarios. Results of model simulations and particle tracking were summarized in USGS Scientific Investigations Report 2006-5195 (Cherry, 2006). Major findings include:
- Simulated ground-water flowpaths for each of the four pumping scenarios indicate that time-of-travel from recharge areas originating near central SRS (D and K Areas) westward into Georgia range from 110 years to 800 years (facing page).
- Particle-tracking analysis indicates travel times and flowpaths are similar for the various pumping scenarios; however, the shutdown of the SRS production wells allows fewer particles to penetrate into deeper units (layers A3– A5), and median travel times are decreased by about 90 years.
Progress and Significant Results, 2002—2003
- Collected water-level measurements from 282 wells in Georgia and South Carolina during September 9 –13, 2002, and constructed potentiometric-surface maps for four major aquifers. The potentiometric-surface maps were integrated into a Geographic Information System to determine the horizontal and vertical hydraulic gradients for the aquifers along with any interaction with streams and rivers. The water-level measurements were used to adjust boundary conditions and determine if any additional calibration is required to the model under 2002 hydrologic conditions.
- Updated ground-water use estimates within the eight-county study area to reflect the changes that have occurred since the previous study (Clarke and West, 1998). The major increase in ground-water use between 1995 and 2000 is evident for Burke, Jefferson, and Screven Counties, Ga.; and Allendale and Barnwell Counties, S.C. In these counties, ground-water use for irrigation increased from 16.7 million gallons per day (Mgal/d) during 1995 to 53.1 Mgal/d during 2000 and irrigated acreage increased from 61,690 acres during 1995 to 97,690 acres during 2000 (Fanning, 2003).
- Converted existing regional groundwater model to Graphical User Interface (GUI) environment to generate current model input for MODFLOW-2000 simulations. The new MODFLOW GUI incorporates the hydrogeologic framework (Falls and others, 1997) into the various model layers and is essential when performing three-dimensional particle-tracking analysis.
- Adjusted specified heads in the source-sink layer (A1) of the model to conform with the 2002 potentiometric-surface map of the Upper Three Runs aquifer, and lowered the specified heads along the lateral boundaries of the model in layers A2–A7 based on observation points in each aquifer. The specified heads in the source-sink layer were lowered to reflect the decline in water levels that resulted from the drought that occurred from 1998 to 2002.
- Evaluated ground-water model under steady-state conditions for 2002 to determine if additional calibration is necessary. The model simulations conducted using updated pumping estimates, observed aquifer heads, and recharge rates from the source-sink layer (A1), indicated that no additional calibration was required.
Related Publications
- Development of Water-Use Projections for Groundwater Flow Models in the Coastal Plain of Georgia and South Carolina
- Precipitation, Ground-Water Use, and Ground-Water Levels in the Vicinity of the Savannah River Site, Georgia and South Carolina, 1992–2002
- Ground-Water Conditions and Studies in Georgia, 2004-2005, p 62-63
- Ground-water conditions and studies in Georgia, 2002-03, p 62-63
The U.S. Department of Energy's (DOE) Savannah River Site (SRS) has manufactured nuclear materials for the National defense since the early 1950s. A variety of hazardous materials including radionuclides, volatile organic compounds, and trace metals, are either disposed of or stored at several locations at the SRS. As a result, contamination of ground water has been detected at several locations within the site and concern has been raised over the possible migration of water-borne contaminants offsite. Two issues have been raised:
- Is ground water flowing from the SRS and beneath the Savannah River into Georgia?
- Under what pumping scenarios could such ground-water movement occur?
Objectives
The U.S. Geological Survey, in cooperation with the DOE, conducted a comprehensive study during 1991-97 that simulated groundwater flow and stream-aquifer relations in the vicinity of the SRS. These groundwater simulations are limited by simplification of the conceptual model, which was based on available data through 1992. Large increases in groundwater pumping in Burke and Screven Counties, Georgia, since 1992, and a pronounced drought during 1998-2002 may have changed hydraulic gradients near the river and affected the potential for trans-river flow. To provide a more accurate and up-to-date evaluation of trans-river flow near the SRS, the earlier model is being updated to incorporate new data and simulate 2002 conditions. The revised model will be used to simulate a variety of water-management scenarios that could impact trans-river flow in the SRS area.
- Update the previously developed groundwater flow model to better define present-day (2002) ground-water flowpaths near SRS.
- Utilize the 2002 calibrated model to identify groundwater flowpaths and quantitatively describe current groundwater flowpaths near SRS under a variety of hypothetical pumping scenarios.
Progress and Significant Results, 2004—2005
- The previous model (Clarke and West, 1998) was updated to simulate ground-water flow under 2002 hydrologic conditions and for four hypothetical pumping scenarios based on ground-water-use trends from 1980 to 2000 (Fanning, 2003).
- Four steady-state pumping scenarios were developed to simulate a range of pumping and climatic conditions affecting potential contaminant migration from the SRS:
- 2002 observed pumping and boundary conditions for an average year.
- 2002 observed pumping and boundary conditions for an average year with SRS pumping discontinued.
- Projected 2020 pumping and boundary conditions for an average year.
- Projected 2020 pumping and boundary conditions for a dry year.
- The USGS particle-tracking code MODPATH (Pollock, 1994) was used to generate advective water-particle path lines and time-of-travel based on MODFLOW simulations of the four scenarios. Results of model simulations and particle tracking were summarized in USGS Scientific Investigations Report 2006-5195 (Cherry, 2006). Major findings include:
- Simulated ground-water flowpaths for each of the four pumping scenarios indicate that time-of-travel from recharge areas originating near central SRS (D and K Areas) westward into Georgia range from 110 years to 800 years (facing page).
- Particle-tracking analysis indicates travel times and flowpaths are similar for the various pumping scenarios; however, the shutdown of the SRS production wells allows fewer particles to penetrate into deeper units (layers A3– A5), and median travel times are decreased by about 90 years.
Progress and Significant Results, 2002—2003
- Collected water-level measurements from 282 wells in Georgia and South Carolina during September 9 –13, 2002, and constructed potentiometric-surface maps for four major aquifers. The potentiometric-surface maps were integrated into a Geographic Information System to determine the horizontal and vertical hydraulic gradients for the aquifers along with any interaction with streams and rivers. The water-level measurements were used to adjust boundary conditions and determine if any additional calibration is required to the model under 2002 hydrologic conditions.
- Updated ground-water use estimates within the eight-county study area to reflect the changes that have occurred since the previous study (Clarke and West, 1998). The major increase in ground-water use between 1995 and 2000 is evident for Burke, Jefferson, and Screven Counties, Ga.; and Allendale and Barnwell Counties, S.C. In these counties, ground-water use for irrigation increased from 16.7 million gallons per day (Mgal/d) during 1995 to 53.1 Mgal/d during 2000 and irrigated acreage increased from 61,690 acres during 1995 to 97,690 acres during 2000 (Fanning, 2003).
- Converted existing regional groundwater model to Graphical User Interface (GUI) environment to generate current model input for MODFLOW-2000 simulations. The new MODFLOW GUI incorporates the hydrogeologic framework (Falls and others, 1997) into the various model layers and is essential when performing three-dimensional particle-tracking analysis.
- Adjusted specified heads in the source-sink layer (A1) of the model to conform with the 2002 potentiometric-surface map of the Upper Three Runs aquifer, and lowered the specified heads along the lateral boundaries of the model in layers A2–A7 based on observation points in each aquifer. The specified heads in the source-sink layer were lowered to reflect the decline in water levels that resulted from the drought that occurred from 1998 to 2002.
- Evaluated ground-water model under steady-state conditions for 2002 to determine if additional calibration is necessary. The model simulations conducted using updated pumping estimates, observed aquifer heads, and recharge rates from the source-sink layer (A1), indicated that no additional calibration was required.
Related Publications
- Development of Water-Use Projections for Groundwater Flow Models in the Coastal Plain of Georgia and South Carolina
- Precipitation, Ground-Water Use, and Ground-Water Levels in the Vicinity of the Savannah River Site, Georgia and South Carolina, 1992–2002
- Ground-Water Conditions and Studies in Georgia, 2004-2005, p 62-63
- Ground-water conditions and studies in Georgia, 2002-03, p 62-63