The 100-BC Groundwater Fate and Transport Model: A Case Study in Savings - 21124

Groundwater flow and transport modeling can play a key role in the decision-making process for selecting the most efficient and cost-effective groundwater contaminant remediation alternative. A well-calibrated groundwater flow and transport model could perform poorly in predicting the future conditions if the conceptual site model (field conditions, underlying features, and processes) are not represented well enough in the numerical models. A site-specific groundwater flow and transport model for the 100-BC-5 Operable Unit (OU) of the U.S. Department of Energy's (DOE) Hanford Site was developed to provide the computational basis for simulation of fate and transport of groundwater contaminants. Contaminated groundwater in the 100-BC-5 OU flows northward and discharges into the Columbia River. The groundwater table descends below the Hanford/Ringold contact a few hundred meters before the hexavalent chromium [Cr(VI)] plume discharges to the river, and it is in this area that a pronounced increase in hydraulic gradient occurs due to loss of transmissivity. The hydraulic gradient is typically on the order of or less than 10-4 in the saturated Hanford and Ringold sediments. Wells in the automated water level network (AWLN) away from the river show time delays from river stage changes as a function of distance. Wells proximal to the river respond immediately to daily stage fluctuations, and on the order of 10's of days several hundred meters away from the river. Current interpretation is that there is flow from the west (upriver) and possibly leakage from the 182-B pond directing plume migration to the north-northeast into the river. The model domain encompasses the sediments above Ringold Upper Mud (RUM) and basalt in the vicinity of the 100-BC-5 OU including the Ringold E and Hanford Formations - the Ringold Upper Mud (RUM) and basalt are considered impermeable relative to the Ringold E and Hanford sediments. The flow model was calibrated to observed heads, hydraulic testing characterization data, hydraulic gradient, and inferred average groundwater velocity from 2012 through 2014. In addition, the correlation between river stage fluctuation and aquifer head changes was used to further constrain the aquifer properties. The calibrated flow model was then assessed using the observed data from 2006 to 2014. As the primary purpose of the groundwater model was to simulate contaminant plume migration and its remedy, the plume sources were estimated and the flow model adjusted to match observed groundwater contaminant plume behavior from 2012 to 2015. The Cr(VI) plume migration in the groundwater was simulated for all the alternatives using the calibrated flow model, the most representative initial concentration plume, site-specific transport properties, and secondary continuing contaminant sources to groundwater. This analysis supported monitored natural attenuation as the preferred alternative, saving millions of dollars. Recently the simulated plume predictions for 2016 through 2020 were evaluated against data, and the model matched reasonably well with the plume inferred from the observed monitoring groundwater concentration data. This further increased confidence in the model-recommended remedy. (authors)