DISCLAIMERPortions of this document may be illegible in electronic image products. Images are produced from the best available original document. Executive SummaryA three-dimensional ground-water flow model has been developed for the uppermost unconfined aquifer at the Hanford Site in south-central Washington. This model is being developed to support the Hanford Site Ground-Water Surveillance Project objectives of 1) identifying and quantifying existing, emerging, or potential ground-water quality problems, and 2) assessing the potential for contaminants to migrate from the Hanford Site through the groundwater pathway.The approach to develop the three-dimensional model was to 1) refine the conceptual hydrogeologic model, 2) calibrate a two-dimensional version of the model with an inverse calibration method, and 3) develop the three-dimensional model based on all available information.The model is based on the Coupled Fluid, Energy, and Solute Transport (CFEST) code; however, the database of geologic and hydrologic information is constructed so that the conceptual model can be implemented in codes other than CFEST.The model consists of 10 hydrogeologic layers. Layers 1 through 9 correspond to the hydrogeologic units comprising the unconfined aquifer system. Layer 10 corresponds to basalt. The conceptual model was developed using the ground surface as the top of the model. The numerical model uses only the saturated portion of the sediments below the ground surface (i.e., the unconfined aquifer). Hydraulic conductivities were assigned to layers based on data from aquifer pumping tests, slug tests, and some laboratory tests. Some layers were assigned a constant value of hydraulic conductivity, while others were zoned with different hydraulic conductivities being assigned to each zone. The goal of the three-dimensional model calibration was to preserve the total transmissivity calculated by the two-dimensional inverse calibration, while honoring data available for hydrogeologic layers.The model grid consists of 821 surface nodes and 721 surface elements. The placement of nodes in the grid was selected so that 1) sources would be represented as closely as possible by nodes, 2) potential contours and gradients could be resolved, and 3) geologic layer extents and other important three-dimensional hydrogeologic aspects of the problem could be represented as accurately as possible. Boundary and source conditions used in the model are similar to previous ... lu efforts at the Site. A natural areal recharge distribution was included to reflect infiltration and recharge to the unconfined aquifer.Steady-state predictions from the three-dimensional model for the 1979 calibration period compared well with the observed water table. There is close agreement in most areas of the Site. The three-dimensional model is ready for routine application in steady-state, and we are proceeding with the construction of transport models for tritium, iodine-129, and nitrate based on the flow model.In addition to the three-dimensional model dev...

et al. (1996) in support of the Environmental Restoration Program. However, the current three-dimensional model has resulted in higher estimates of peak concentrations at shallow depths and less vertical migration than was predicted in previous site-wide yodel applications. These discrepancies are attributable to differences in basic assumptions made about the hydrogeologic framework and the horizontal and vertical V discretization used in each model. The differences in assumptions resulting from each modeling approach affect lateral and vertical distributions of predicted hydraulic heads and contaminants in the unconfined aquifer.The near-term impacts of existing plumes to on-site drinking water supplies were examined by evaluating predicted concentration levels in the vicinity of identified on-site water supplies over the next Farm, will be impacted by future transport of these contaminants.vi

iii Executive SummaryThe purpose of this research is to develop and maintain numerical groundwater flow and transport models that can be used to refine the conceptual site model for groundwater beneath the 300 Area, and to assist in evaluating alternative remediation technologies focused on the 300 Area uranium plume. Groundwater flow rates and directions in the 300 Area are very dynamic because of the high hydraulic conductivities, along with the large daily, weekly, and seasonal fluctuations in the Columbia River stage. Quantifying the dynamics of groundwater flow and transport in the 300 Area aquifer will help understand the significant seasonal variability of uranium plume concentrations seen in biannual groundwater monitoring, and will help evaluate remediation options. Groundwater flow rates are very high in the upper portion of the 300 Area unconfined aquifer (within the Hanford formation), with velocities up to 10 to 15 m/d (35 to 50 ft/d) based on a tracer test and limited plume-migration data. Variability in the groundwater-flow directions is apparent from analysis of hourly and subhourly automated water-level measurements from monitoring networks established in the 300 Area. Generalized flow directions in the area between the north and south process ponds are toward the east to south, with the directions changing toward the south and west during periods of increases in the river stage (daily and seasonal).High-resolution water level and river stage data were required to simulate the dynamics of the 300 Area aquifer. Two scales of groundwater flow and transport models were developed based on the availability of high-resolution water-level monitoring data. A larger-domain model was developed that includes the 300 Area and extends north and south using data from the early 1990s water-level monitoring network. A smaller domain model was developed for a portion of the large scale model domain in the north of the 300 Area that used water-level data from another smaller monitoring well network that was established in 2004. These models focus on the highly permeable upper portion of the unconfined aquifer within the Hanford formation that has hydraulic conductivity values 2 to 3 orders of magnitude higher than the underlying Ringold Formation aquifers. These models simulate saturated and unsaturated groundwater flow and transport with the STOMP code, which was developed at Pacific Northwest National Laboratory. 1The hydrostratigraphy, topography, and bathymetry for the three-dimensional models used a consistent framework using EarthVision  software.The model domains include the lower portion of the vadose zone to encompass the range of river stage and water- The hydrostratigraphic units were determined from previously published interpretations of the 300 Area, along with data from additional wells installed since those studies. A reanalysis of some of the older geologic unit picks from well logs in the area, along with using geophysical logs, was conducted based on the detailed knowledge gained from the 300 ...

Table 3.1. Drilling, sampling, geophysical logging, and well construction information for IFRC wells Well Name Well ID Function Samples Drilling Order Start Drilling End Drilling Geophysical Logging Total Depth (ft) Length Screen (ft) Top Screen Depth (ft) Bottom Screen Depth (ft) Bottom End Cap (ft) Instrument Depth (ft) % Recovery Hanford/Ringold Contact Depth (ft)

Nimmons have provided new ideas and concepts regarding the 300 Area uranium plume, and their contributions are appreciated. The authors would like to acknowledge the helpful support of Chris Newbill and JoAnne Rieger for graphics and data compilations, respectively. Text editing and word processing were accomplished by Wayne Cosby, Andy Fish, and Mike Parker under the capable direction of Hope Matthews. Acronyms and Abbreviations μEXAFS microprobe x-ray absorption fine structure μXANES microprobe x-ray absorption near-edge structure μXAS microprobe x-ray spectroscopy μXRF microprobe x-ray fluorescence BWTF Buried Waste Test Facility CERCLA Comprehensive Environmental Response, Compensation, and Liability Act CSM conceptual site model DOE U.S. Department of Energy DOE-RL U.S. Department of Energy, Richland Operations Office DRM distributed rate model EPA U.S. Environmental Protection Agency ERDF Environmental Restoration Disposal Facility GEA gamma energy analysis HEIS Hanford Environmental Information System HRM Hanford River Marker ICP-MS inductively coupled plasma mass spectrometry LFI limited field investigation MCL maximum contaminant level NPL National Priorities List RACS Remediation and Closure Science Project RCRA Resource Conservation and Recovery Act