Transport simulations were developed to investigate contaminant mass transport issues, evaluate the future movement of contaminant plumes, and identify and quantify potential groundwater quality problems that could affect on-site and off-site use of groundwater. Monitoring of groundwater across the Hanford Site has characterized a number of radioactive contaminant plumes that emanate from various operational areas at the site, including large plumes of tritium and iodine-129 and smaller plumes of strontium-90, technetium-99, plutonium, uranium, cesium-137, and cobalt-60.
The focus of this analysis was to examine the transport of those radionuclides with the potential for near-term mobility and future risk. Therefore, this analysis included an evaluation of tritium, iodine-129, technetium-99, uranium, and strontium-90. Because of the strong tendency of plutonium, cesium-137, and cobalt-60 to be sorbed on sediments within Hanford groundwater, the long-term transport of those radionuclides was not considered.
The following section describes the results derived from applying the three-dimensional transport model to existing site-wide plumes of tritium, iodine-129, technetium-99, uranium, and strontium-90.
The finite-element grid used to simulate three-dimensional flow conditions in the unconfined aquifer was refined in the 200-Area plateau to add horizontal and vertical discretization of the hydrogeologic units. This was done to 1) provide adequate resolution to represent the areal variations of contaminant concentrations used as initial conditions, 2) more accurately represent flow paths, 3) minimize numerical dispersion in the transport calculations, and 4) allow for appropriate specification of initial vertical contaminant distributions (initial conditions). Because the tritium plume has the greatest areal extent of all plumes considered in the analysis, the grid refinement was primarily based on the examination of issues related to resolving the areal distribution and subsequent transport of the current tritium plume.
The finite-element grid used for transport calculations of all existing plumes (Figure 5.1) was primarily refined in the central area of the Hanford Site in the vicinity of the 200-Area plateau. In this area, each 750 m grid was subdivided into four grid spaces so that the final grid resolution was 375 m on a side.
Within all areas of the grid, additional vertical discretization was added to minimize numerical dispersion in the vertical direction and to facilitate the assignment of initial concentrations of all the existing plumes to the uppermost computational layers of the model. The general approach was to subdivide the principal hydrologic units found at the water table (Unit 1 and Unit 5) into multiple 8-m-thick layers. A maximum of five layers was used to represent each unit. The Upper Ringold unit (Unit 4) was subdivided into two layers to provide for the isolating capabilities of this important mud unit. The original hydrogeologic layering used to represent all other units remained unchanged.
The initial concentrations of all the existing plumes were assigned to the uppermost computational layers of the model. This was done to approximate the current understanding that the bulk of contamination is found in the uppermost part of the aquifer (Eddy et al. 1978). At all locations where a contaminant plume exists, the initial conditions were applied to all nodes found within 25 m of the top of the aquifer (i.e., initial water-table conditions). In some areas, the aquifer is thinner than 25 m, so the initial conditions were applied through the entire thickness of the aquifer. In finite-element solutions, nodes are involved in more than one element, resulting in the effective depth of contaminant being about 28 m.
Results of both the combined horizontal and vertical refinement for the transport simulations yielded a final transport finite-element grid with a surface grid of 3108 nodes and 2991 elements and a total grid of 23,668 nodes and 23,128 elements.
Tritium was present in many Hanford Site waste streams discharged to the soil and is the most mobile and most widely distributed radionuclide on the site. As a result, the contaminant plumes from tritium reflect the maximum extent of contamination in the groundwater at the Hanford Site.
In 1996, tritium concentrations were greater than the interim DWS (20,000 pCi/L) in nearly all the operational areas, including parts of the 100-K, 100-N, 100-D, 100-F, 200-West, 200-East, 400, and 600 Areas (Hartman and Dresel 1997). The highest concentrations of tritium were found in wells near waste discharge cribs from the PUREX plant, exceeding Derived Concentration Guidelines (2,000,000 pCi/L) in one well. The plume from the 200-East Area has migrated eastward into the 600 and 400 Areas and to the Columbia River over a broad area extending from the Old Hanford townsite to the 300 Area. Maximum concentrations of tritium between 150,000 and 180,000 pCi/L can be found in the vicinity of the Old Hanford townsite. The plume has also migrated northwest through the gap between Gable Mountain and Gable Butte. In 1996, maximum concentrations of tritium in the vicinity of the gap ranged from 27,000 to 71,000 pCi/L in a few wells.
In the 200-West Area, tritium plumes originate in the vicinity of T plant, the SX Tank Farm, the REDOX plant, and associated storage and disposal facilities. Near T plant, maximum tritium concen-trations were 120,000 pCi/L in one well. In the vicinity of the REDOX facilities, tritium concentrations reached a maximum of 1,100,000 pCi/L in one well. The larger of the two plumes originating from the REDOX plant area has been slowly migrating eastward toward the 200-East Area and has reached the area of the commercial low-level radioactive waste disposal facility operated by US Ecology.
The overall approach in predicting future transport of tritium was to model the current distribution of tritium forward in time until the entire tritium plume is discharged into the Columbia River or has decayed away. To establish some confidence in the general validity of the calculations, transport calculations were initiated in 1979 and continued through 1996, using historical records of tritium discharges to the land surface at various facilities on site. These results were then compared to conditions measured in 1996. The average annual discharges used in these calculations are provided in Table A.1 in Appendix A. Beyond 1996, the only tritium discharges simulated were currently projected discharges at the SALDS in the 200-West Area, which are expected to continue through the year 2025. Current records of discharges and future projected discharges used in the tritium transport calculations are provided in Table A.2 of Appendix A. A separate detailed analysis of the impact of tritium discharges at the SALDS made with the current three-dimensional transport model has recently been completed and documented in a report by Barnett et al. (1997).
Initial conditions for the tritium plume transport simulation were defined based on the extent of the tritium plume as interpreted in 1979. A surface representing the distribution of tritium in wells monitoring the upper portion of the unconfined aquifer at the Hanford Site during 1979 was prepared with data available in the Hanford Environmental Information System (HEIS). Tritium results for groundwater samples from wells completed in the upper portion of the Hanford Site unconfined aquifer were reviewed, and suspect (off-trend) data were removed. Log (base 10) values of mean tritium concentrations for selected wells were interpolated to a regular grid, using the minimum tension interpolation (gridding) algorithm supplied with EarthVision.® Contours of the logged values from the gridded surface were compared to contours presented by Eddy and Wilbur (1980).
In areas where the contours generated from the interpolated surface differed from the contours presented by Eddy and Wilbur (1980), groundwater data were reviewed to determine which surface better represented tritium concentrations in the unconfined aquifer during 1979. Where the spatial distribution of groundwater data was not sufficient to control interpolation, additional control was added using a graphic editor in EarthVision,® after which the surface was reinterpolated. When a satisfactory fit of the interpolated surface to the observed tritium concentrations was achieved that was also consistent with the conceptual understanding of the tritium distribution in the unconfined aquifer, the surface was converted from logged to actual values. This new surface, representing the distribution of tritium concentrations, was then checked against the observed concentrations of tritium in groundwater and the conceptual understanding of tritium distribution in the Hanford Site unconfined aquifer during 1979. The final distribution of tritium in the unconfined aquifer for 1979 (Figure 5.2) was used to represent the initial tritium concentrations in the three-dimensional transport model simulations.
As indicated earlier, the initial concentrations were assigned only to the uppermost layers of the model to approximate the current understanding that the bulk of contamination is found in the uppermost portion of the aquifer. At all locations where the tritium plume exists, initial conditions were applied to all nodes found within 25 m of the top of the aquifer (i.e., initial water-table conditions). In some areas, the aquifer is thinner than 25 m (e.g., near basalt outcrops), so the initial conditions were applied through the entire thickness of the aquifer.
Results of tritium transport for the period from 1979 through 1996 (Figures 5.2, 5.3, and 5.4) showed the same overall trends of contaminant migration as reported by the HGWP. Model results showed that the tritium plumes originating from the 200-East and 200-West areas slowly migrate laterally in a general easterly direction and discharge to the Columbia River along a broad area between the Old Hanford townsite and north of the 300 Area. Maximum concentrations of tritium in the 600 Area (downgradient of the 200-East Area) decline from over the 2 million pCi/L level in 1979 to above 200,000 pCi/L in 1996. In 1996, tritium levels in wells within the maximum area of concentration ranged from 150,000 to 180,000 pCi/L.
Within the 200-East Area, maximum tritium concentrations, which were initially at several hundred thousand pCi/L in 1979, declined steadily until 1985, when discharges originating from PUREX plant operations startup caused the tritium concentration to rise rapidly above the 5 million pCi/L level. With the cessation of these discharges in 1988, simulated tritium levels declined to the several hundred thousand pCi/L level by 1996. These simulated levels of tritium are consistent with levels measured in wells within the same area, which ranged from about 200,000 to 1 million pCi/L.
In the 200-West Area, initial maximum tritium concentrations exceeding 5 million pCi/L were simulated to decline steadily to several hundred thousand pCi/L by 1996. These simulated levels are, in general, consistent with tritium levels currently measured in wells in the same area.
A comparison of simulated (Figure 5.4) and observed (Figure 5.5) distributions of tritium for 1996 conditions show a few differences worth noting:
Model results from 1996 through 2100 (Figures 5.6, 5.7, 5.8, and 5.9) show that the current tritium plumes will continue to slowly migrate laterally from source locations in the 200 Areas to the Columbia River north and east of the 200-Area plateau. Concentration levels of the simulated plumes in nearly all areas will decline by processes of radioactive decay and plume dispersion as they migrate. The only source of future tritium that was simulated was assumed to originate from projected discharges of tritium at the SALDS from 1996 through 2026.
In the 200-West Area, maximum tritium levels exceeding about 3 million pCi/L simulated at the SALDS decline to 1 million pCi/L in 2000, 80,000 pCi/L in 2020, several thousand pCi/L in 2050, and below detection by 2100. In the 200-East Area, maximum tritium levels exceeding several hundred thousand pCi/L simulated near the PUREX plant decline to tens of thousand pCi/L in 2000, several thousand pCi/L in 2020, below 1000 pCi/L in 2050, and below detection by 2100. In the 600 Area (downgradient of the 200-Area plateau), maximum tritium levels exceeding several hundred thousand pCi/L simulated near the Old Hanford townsite decline to just above 200,000 pCi/L in 2000, below 200,000 pCi/L in 2020, below 80,000 pCi/L in 2050, below 2000 by 2100, and below 500 pCi/L by 2150. Maximum tritium concentrations discharge from the unconfined aquifer to the Columbia River near the Old Hanford townsite east of the 200-East Area. Groundwater containing tritium concentrations exceeding 20,000 pCi/L are predicted to discharge at this location for 50 to 60 years. Groundwater containing tritium concentrations exceeding 2,000 pCi/L are predicted to discharge at this location over a period of 80 to 90 years.
Tritium contamination found in the area north of Gable Mountain and Gable Butte slowly migrates toward the Columbia River. The combination of radioactive decay, dispersion during transport, and discharge of groundwater containing tritium in the 100 Areas causes tritium levels in this area to decline steadily from 1996 levels to below detection (500 pCi/L) by 2070.
According to Hartman and Dresel (1997), the main on-site contributor of iodine-129 to groundwater is liquid discharges to cribs located near fuel-processing facilities in the 200 Areas. Extensive plumes of iodine-129 exceeding the interim DWS (1 mg/L) originate in the 200 Areas and in downgradient portions of the 600 Areas. The presence of iodine-129 is of concern because of its low interim DWS (1 pCi/L), its long-term release from fuel reprocessing facilities, and its long half-life (about 16 million years).
In the 200-East Area, wells exceed the interim DWS of 1 pCi/L along the eastern edge and the northern half of the area. Peak values exceeding 10 pCi/L are found in several wells located near the PUREX facility. The plume originating in the 200-East Area has migrated eastward as much as 10 km toward the Columbia River. In this area, maximum concentrations of iodine-129 between 6 and 9 pCi/L can be found in two wells.
In the 200-West Area, plumes of iodine-129 originate from liquid discharge points near T plant, U Plant, and the REDOX plant. Wells in both of these areas contain iodine-129 levels that exceed the interim DWS (1 pCi/L). The highest concentrations of iodine-129 are found in wells near waste discharge cribs in the vicinity of the REDOX plant, exceeding 80 pCi/L in one well and 20 pCi/L in other wells. The plume from the 200-West Area has migrated slowly eastward toward the 200-East Area.
The approach to simulating future transport of the iodine-129 plume was to start the simulation with initial conditions from 1996 and calculate future transport of iodine-129 based on the three-dimensional transient-flow field discussed in Section 4.3.2. For the purpose of this analysis, no further contributions of iodine-129 were considered. Future releases of iodine-129 are being estimated and analyzed as part of the Hanford Site Composite Analysis.
Initial conditions for the iodine-129 plume were defined based on the extent of the iodine-129 plume in 1996. Because no additional discharges of iodine-129 were assumed to occur, the area defined by the 1996 plume is sufficient to represent the maximum extent of that plume. The initial conditions for iodine-129 were generated in the same manner as initial conditions for tritium (i.e., the 1996 plume was used as an initial condition). The same grid defined for the tritium plume (Figure 5.1) was used for the iodine-129 plume. The simulation was started with 1996 iodine-129 concentrations as initial conditions (Figure 5.10). The 1996 concentrations were used because the iodine-129 data for 1979 were not available.
Transport model results for iodine-129 from 1996 through 2300, shown in Figures 5.11, 5.12, 5.13, and 5.14, demonstrate that current iodine-129 plumes will continue to slowly migrate laterally from source locations in the 200 Areas to the Columbia River north and east of the 200-Area plateau. Concentration levels of the simulated plumes in nearly all areas decline by the processes of plume dispersion and dilution related to infiltration.
In the 200-West Area, maximum iodine-129 levels exceeding 50 pCi/L near the REDOX facilities decline to less than 50 pCi/L in 2000, less than 30 pCi/L in 2031, about 20 pCi/L in 2050, less than 10 pCi/L in 2100, and less than 5 pCi/L by 2400. In the 200-East Area, maximum iodine-129 levels exceeding 10 pCi/L, simulated near the PUREX plant, decline to less than 5 pCi/L by 2031 and less than 1 pCi/L by 2100. In the 600 Area (downgradient of the 200-Area plateau), maximum iodine levels exceeding 5 pCi/L simulated between the 200-East Area and the Old Hanford townsite decline to less than 5 pCi/L in 2200 and below 1 pCi/L in 2400. Maximum iodine-129 concentrations discharge from the unconfined aquifer to the Columbia River near the Old Hanford townsite east of the 200-East Area. Groundwater containing iodine-129 concentrations exceeding 1 pCi/L are predicted to discharge at this location over a period of about 570 years, from about 2030 to 2600.
Iodine-129 contamination found in the area north of Gable Mountain and Gable Butte slowly migrates toward the Columbia River. The combination of radioactive decay and dispersion during transport causes iodine-129 levels in this area to decline steadily from 1996 levels (less than 2 pCi/L) to less than 0.1 pCi/L by 2700.
According to Hartman and Dresel (1997), technetium-99 is a highly mobile radionuclide of concern that has been present in a number of waste streams associated with fuel reprocessing at the Hanford Site. Technetium-99 has been detected at levels exceeding the DWS of 900 pCi/L in a number of areas on the site. In the 200-West Area, technetium-99 exceeded the DWS in several areas. Highest concentrations are found in the plume originating from the U Plant area where technetium-99 levels exceed 9,000 pCi/L in a number of wells and 20,000 pCi/L in one well. Technetium-99 concentration levels exceed 900 pCi/L in wells located within a plume originating from the northern 200-East Area that is migrating through the gap between Gable Mountain and Gable Butte. In the 100-H Area, technetium-99 concentrations exceed 900 pCi/L in wells downgradient of the 183-H solar evaporation basins, where fuel-processing wastes have leaked into the ground.
The approach to simulating future transport of the technetium-99 plume was the same as that used for the iodine-129 plume. The simulation was initiated using the 1996 conditions, and future transport of technetium-99 was based on the three-dimensional transient-flow field discussed in Section 4.3.2. For purposes of this analysis, no further contributions of technetium-99 were considered. As in the case of iodine-129, future releases of technetium-99 are being estimated and analyzed as part of the Hanford Site Composite Analysis.
Initial conditions used for the technetium-99 plume were defined based on the extent of the technetium-99 plume in 1996. Because no additional discharges of technetium-99 were assumed to occur, the area defined by the 1996 plume is sufficient to represent the maximum extent of that plume. The initial conditions for technetium-99 were generated in the same manner as initial conditions for tritium (i.e., the 1996 plume was used as an initial condition). It was found that the same three-dimensional grid defined for the tritium plume originating from the 200 areas (Figure 5.1) could be used for the technetium-99 plume. The 1996 technetium-99 concentrations used as initial conditions are presented in Figure 5.15.
Transport model results for technetium-99 from 1996 through 2300 shown in Figures 5.16, 5.17, 5.18, and 5.19 illustrate how the plumes in the 200-West Area decline to levels below regulatory concern as they migrate eastward toward the 200-East Area. The technetium-99 plume in the northern part of the 200-East Area, which is migrating northward through the gap between Gable Butte and Gable Mountain, will continue to migrate in this direction toward the Columbia River. However, concentration levels of the simulated plumes decline to below regulatory levels over a period of 100 years by the processes of plume dispersion and dilution related to infiltration.
In the 200-West Area, maximum technetium-99 levels exceeding 7000 pCi/L near the REDOX facilities decline to less than 5000 pCi/L in 2000, less than 500 pCi/L in 2031, and less than 90 pCi/L by 2100. In the 200-East Area, maximum technetium-99 levels exceeding 1500 pCi/L, simulated north of the BY crib area, decline to less than 500 pCi/L by 2000 and less than 90 pCi/L by 2011.
Technetium-99 contamination levels found in the area north of Gable Mountain and Gable Butte slowly migrate toward the Columbia River. The combination of dispersion and infiltration from recharge during transport causes technetium-99 levels in this area to decline steadily from 1996 levels (above 1000 pCi/L) to less than 90 pCi/L by 2020.
Groundwater containing technetium-99 levels above 900 pCi/L in the 100 areas are expected to quickly migrate into the Columbia River within the next 10 years and were not specifically considered in this simulation.
According to Hartman and Dresel (1997), uranium releases at the Hanford Site have numerous potential sources, including discharges from fabrication, fuel reprocessing, and uranium recovery. Uranium mobility has been found to be slower than that of tritium and technetium-99 because of its dependency on both Eh and pH conditions. At Eh/pH conditions in the unconfined aquifer, U6+ is found to be the most mobile species of uranium.
Uranium has been detected at concentrations above the maximum concentration limit proposed by EPA of 20 g/L in the 100-H, 100-F, 200, and 300 areas. Contamination in the 100-H, 100-F, and 200-East areas has been very localized. In the 200-West Area, the most extensive contamination occurs in a plume that originates from two cribs located at the U Plant. This plume is currently migrating slowly toward the 200-East Area. Uranium concentrations in this plume exceed 1,000 g/L in a number of wells, and the plume is currently being contained by a pump-and-treat system.
In the 300 Area, elevated concentrations of uranium are found downgradient of the 316-5 process trenches and moving in a southeasterly direction. An expedited response action was performed on the trenches in mid-1991 to reduce the uranium source in this area. Current levels of uranium exceed 20 g/L in many wells downgradient of the trenches; uranium concentrations in a smaller number of wells exceed 100 g/L.
The overall approach for simulating future transport of the uranium plumes was to initiate the simulation from 1996 conditions as was done for the iodine-129 and technetium-99 plumes. For purposes of this analysis, no further contributions of uranium were considered. As in the case of iodine-129 and technetium-99, future releases of uranium are being estimated and analyzed as part of the Hanford Site Composite Analysis.
The initial conditions used for the uranium plume were defined based on the extent of the uranium plume in 1996 and future transport of uranium was calculated based on the three-dimensional transient-flow field discussed in Section 4.3.2. Because no additional discharges of uranium will likely occur at the Hanford Site, the area defined by the 1996 plume is sufficient to represent the maximum extent of that plume. The initial conditions for uranium were generated in the same manner as initial conditions for tritium (i.e., the 1996 plume was used as an initial condition). It was found that the three-dimensional grid defined for the 200 Area tritium plume (Figure 5.1) could be used for the uranium plume. The 1996 uranium concentrations used as initial conditions are presented in Figure 5.20.
Transport model results for uranium from 1996 through 2200, as shown in Figures 5.21, 5.22, 5.23, and 5.24, indicated that because of the process of adsorption, the current uranium plumes in the 200-Area plateau will not migrate significantly from current source locations in the 200 areas outside of the 200 Area exclusion and buffer zones. Concentration levels of the simulated plumes in the 200 areas will decline to below 20 g/L over the next 50 years by the process of plume dispersion and dilution related to areal recharge.
In the 200-West Area, maximum uranium levels exceeding 400 g/L near the REDOX facilities decline to less than 400 g/L in 2000, less than 300 g/L in 2011, less than 100 g/L by 2100, and less than 20 g/L by the year 3000. Maximum uranium levels exceeding 50 g/L simulated in the northern part of the 200-East Area decline to less than 50 g/L by 2000, less than 20 g/L by 2005, and less than 2 g/L by the year 2100.
Groundwater containing uranium levels above 20 g/L in the 100, and 300 areas is expected to discharge into the Columbia River within the next 40 to 50 years. Consequently, these plumes were not specifically considered in this simulation.
According to Hartman and Dresel (1997), strontium-90 has been present and may have been released in waste streams associated with fuel processing. Release of strontium-90 by fuel-element failures during reactor operations is also suspected. Strontium-90 is considered to be the least mobile of all radioactive constituents being considered in this analysis. However, the occurrence of strontium-90 in groundwater is of concern because of its moderately long half-life (28.8 years), its potential for concentrating in bone tissue, and the relatively high energy of the beta decay from its yttrium-90 radioactive decay product.
In 1996, strontium-90 concentrations exceeded the interim DWS of 8 pCi/L in wells within the 100 and 200 Areas. Strontium-90 concentrations exceeded the 1,000 pCi/L derived concentration guidelines in wells within the 100-K, 100-N, and 200-East Areas and in wells near the former Gable Mountain Pond area. The most widespread contamination and highest levels of strontium-90 contamination are found in the 100-N Area. At present, the overall extent of the 100-N Area strontium-90 plume is not increasing perceptibly. A pump-and-treat system is currently in operation in the 100-N Area to reduce the flux of strontium-90 to the Columbia River. Within the 200-Area plateau, the largest concentrations of strontium-90 are found in wells in the north central part of the 200-East Area, in the vicinity of the BY crib area.
The overall approach for modeling of future transport of the strontium-90 plumes was to initiate the simulation from 1996 conditions and calculate future transport of strontium-90 based on the three-dimensional transient-flow field discussed in Section 4.3.2. For purposes of this analysis, no further contributions of strontium-90 were considered. As in the case of iodine-129, technetium-99, and uranium, future releases of strontium-90 are being estimated and analyzed as part of the Hanford Site Composite Analysis.
Initial conditions used for the strontium-90 plume were defined based on the extent of the strontium-90 plume in 1996. Because no additional discharges of strontium-90 were assumed to occur, the area defined by the 1995 plume is sufficient to represent the maximum extent of that plume. The initial conditions for the strontium-90 plume were generated in the same manner as initial conditions for the tritium plume (i.e., the 1995 plume was used as an initial condition). It was found that the same three-dimensional grid defined for the 200-Area tritium plume (Figure 5.1) could be used for the strontium-90 plume. The 1996 strontium-90 concentrations used as initial conditions are presented in Figure 5.25.
Transport model results for strontium-90 from 1996 through 2300, as shown in Figures 5.26, 5.27, 5.28, and 5.29, demonstrate that the current strontium-90 plume will likely migrate very little from its current location within the 200-East Area because strontium-90 is adsorbed to Hanford sediments. Concentration levels of the simulated plume near the BY crib area decline over the next 100 years to well below 8 pCi/L by the process of plume dispersion. The concentrations are further reduced by radioactive decay.
In the 200-East Area, maximum strontium-90 levels exceed 3000 pCi/L and decline to less than 3000 pCi/L in 2000, to less than 500 pCi/L by 2031, to less than 50 pCi/L by the year 2100, and to below the interim DWS of 8 pCi/L by the year 2200. Maximum strontium-90 levels exceeding 8 pCi/L, simulated in the Gable Mountain Pond area north of the 200-East Area, decline to less than 8 pCi/L by 2011 and less than 0.8 pCi/L by 2100.
Groundwater containing elevated strontium-90 levels in the 100-K and 100-N
areas is expected to migrate into the Columbia River within the next 100 years
before decaying to levels below regulatory concern. Strontium-90 transport at
these specific locations was not specifically considered in this analysis.
