Uncertainty Analyses of Infiltration and Subsurface Flow and Transport for SDMP Sites (NUREG/CR-6565, PNNL-11705)
On this page:
Download complete document
Manuscript Completed: September 1997
Date Published: September 1997
P.D. Meyer, M.L. Rockhold, G.W. Gee
Pacific Northwest National Laboratory
P.O. Box 999
Richland, Washington 99352
T.J. Nicholson, NRC Project Manager
Division of Regulatory Applications
Office of Nuclear Regulatory Research
U.S. Nuclear Regulatory Commission
Washington, DC 20555-0001
NRC Job Code W6503
U.S. Nuclear Regulatory Commission staff have identified a number of sites requiring special attention in the decommissioning process because of elevated levels of radioactive contaminants. Traits common to many of these sites include limited data characterizing the subsurface, the presence of long-lived radio nuclides necessitating a long term analysis (1000 years or more), and potential exposure through multiple pathways. As a consequence of these traits, the uncertainty in predicted exposures can be significant. In addition, simplifications to the physical system and the transport mechanisms are often necessary to reduce the computational requirements of the analysis. Several multiple-pathway transport codes exist for estimating dose, two of which were used in this study. These two codes have built-in Monte Carlo simulation capabilities that were used for the uncertainty analysis.
Several tools for improving uncertainty analyses of exposure estimates through the groundwater pathway have been developed and are discussed in this report.
Generic probability distributions for unsaturated and saturated zone soil hydraulic parameters are presented. These distributions can be used with available dose assessment codes to estimate exposure uncertainty in screening level and preliminary analyses where site-specific data is limited. Tables of the distributions are contained in an appendix, categorized by soil texture. Parameters for the van Genuchten, Brooks-Corey, and Campbell water retention and hydraulic conductivity models are included.
The use of the generic probability distributions for soil hydraulic parameters is illustrated in a method for the estimation of net infiltration uncertainty. This method uses a relatively simple water budget calculation contained in an existing multiple pathway dose assessment code. Onsite meteorological data were used. A distribution for the soil parameter required (plant available water capacity) was selected from the report appendix based solely on a description of the lysimeter soil texture. A comparison between the distribution of predicted annual net infiltration and the observed lysimeter drainage (mean and standard error) showed an agreeable match. For sites without local measurements of precipitation, temperature, etc., meteorological parameters can be estimated from National Climatic Data Center data.
The generic distributions are useful for modeling the uncertainty in soil hydraulic parameters when information about the soils at a site is limited to the soil texture. At many sites, however, there may be some site-specific soil hydraulic property data available. A method is presented to combine the generic distributions with site-specific water retention data using a Bayesian analysis. The resulting updated soil hydraulic parameter distributions can be used to obtain an updated estimate of the probability distribution of dose. The method is illustrated using a hypothetical decommissioning site.