The following is the abstract for a poster presentation at the
Fall 2002 Meeting of the American Geophysical Union (poster NG12B-1033).
Evaluation and Quantification of Uncertainty
in the Modeling of Contaminant Transport and Exposure Assessment
at a Radioactive Waste Disposal Site
J D Tauxe, (Neptune and Company, Los Alamos, NM 87544;
505-662-2121; e-mail: jtauxe(a)neptuneinc.org
(replace (a) with @ to make the address legitimate)
The disposal of low-level radioactive waste (LLW) in the United States
is a highly regulated undertaking. The US Department of Energy (DOE),
itself a large generator of such wastes, requires a substantial amount
of analysis and assessment before permitting continuing operation of
sites for disposal of LLW. One of the requirements that must be met
in assessing the performance of a disposal site and technology is
that a Performance Assessment (PA) demonstrate "reasonable expectation”
that certain performance objectives, such as dose to a hypothetical
future receptor, not be exceeded. The phrase "reasonable assurance"
implies recognition of uncertainty in the assessment process. In
order for this uncertainty to be quantified and communicated to
decision makers, the PA computer model must accept probabilistic
(uncertain) input (parameter values) and produce results which
reflect that uncertainty as it is propagated through the model
calculations. Model parameters range from water content and other
physical properties of alluvium to the activity of radionuclides
disposed to the amount of time a future resident might be expected
to spend tending a garden. The decision maker has the difficult
job of evaluating the uncertainty of modeling results in the context
of granting permission for LLW disposal.
In addition to providing decision makers with realistically uncertain
modeling results, a probabilistic assessment is also useful in guiding
research efforts aimed at reducing the uncertainty in key components
of the model. A sensitivity analysis of the modeling results
identifies which model parameters are most significant in determining
estimated doses, for example, thus providing justification for the
allocation of limited research funding.
reference for this presentation:
Tauxe, J., P. Black, J. Carilli, K. Catlett, B. Crowe,
M. Hooten, S. Rawlinson, A. Schuh, T. Stockton, V. Yucel,
Evaluation and Quantification of Uncertainty in the Modeling
of Contaminant Transport and Exposure Assessment at a
Radioactive Waste Disposal Site, Eos Trans. AGU, 83(47),
Fall Meeting Supplement, Abstract NG12B-1033, 2002