The Science

A simple non-electrostatic model was developed through a step-by-step calibration procedure to describe uranium (U) plume behavior at the Savannah River site. This simple model was found to be more numerically efficient than a complex mechanistic model with electrostatic correction terms in predicting long-term uranium behavior at the site and by extension other uranium-contaminated sites.

The Impact

Uranium geochemistry has been extremely challenging to describe and predict. Although complex mechanistic models have been used to describe uranium sorption in field settings, there is significant uncertainty in model predictions due to scarce field data and modeling assumptions concerning mineral assemblage and subsurface heterogeneity. This study demonstrates that a simpler non-electrostatic model is a powerful alternative for describing uranium plume evolution at the Savannah River Site (SRS) because it can describe U(VI) sorption much more accurately than a constant coefficient (K_d) approach, while being more numerically efficient than a complex model with electrostatic correction terms. This study provides valuable insight into predicting uranium plume persistence from contaminated sites using simple non-electrostatic models.

Summary

The aim of this study was to test whether a simpler, semiempirical, non-electrostatic U(VI) sorption model (NEM) could achieve the same predictive performance as a model with electrostatic correction terms in describing pH and U(VI) behavior at multiple locations within the SRS F-Area. Modeling results indicate that the simpler NEM was able to perform as well as the electrostatic surface complexation model, especially in simulating uranium breakthrough tails and long-term trends. However, the model simulations differed significantly during the early basin discharge period; however, model performance cannot be assessed during this period due to a lack of field observations (e.g., initial pH of the basin water) that would better constrain the models. In this manner, modeling results highlight the importance of the range of environmental data that are typically used for calibrating the model.

Principal Investigator

Haruko Wainwright
Lawrence Berkeley National Laboratory
HMWainwright@lbl.gov

Program Manager

Paul Bayer
U.S. Department of Energy, Biological and Environmental Research (SC-33)
Environmental System Science
paul.bayer@science.doe.gov

Funding

This material is based on work supported as part of the Advanced Simulation Capability for Environmental Management (ASCEM) project, which is funded by the U.S. Department of Energy (DOE) Office of Environmental Management, and as part of the Genomes to Watershed Science Focus Area, which is funded by the Office of Biological and Environmental Research (BER), within the DOE Office of Science. Both are under Award Number DE-AC02-05CH11231 to Lawrence Berkeley National Laboratory.

References