Methodology Used for Total System Performance Assessment of the Potential Nuclear Waste Repository at Yucca Mountain (USA)

Devibec, E. ; Sevougian, S.D. ; Mattie, P.D. ; McNeish, J.A. ; Mishra, S.

United States. Department of Energy. Yucca Mountain Project Office.

关键词: Radioactive Wastes;    Usa;    Environmental Impacts;    Hydrology;    Biosphere;   

DOI  :  10.2172/806076
RP-ID  :  NA
RP-ID  :  NONE
RP-ID  :  806076

美国|英语

来源: UNT Digital Library

【 摘 要 】

The U.S. Department of Energy and its contractors are currently evaluating a site in Nevada (Yucca Mountain) for disposal of high-level radioactive waste from U.S. commercial nuclear plants and U.S. government-owned facilities. The suitability of the potential geologic repository is assessed, based on its performance in isolating the nuclear waste from the environment. Experimental data and models representing the natural and engineered barriers are combined into a Total System Performance Assessment (TSPA) model [1]. Process models included in the TSPA model are unsaturated zone flow and transport, thermal hydrology, in-drift geochemistry, waste package degradation, waste form degradation, engineered barrier system transport, saturated zone flow and transport, and biosphere transport. Because of the uncertainty in the current data and in the future evolution of the total system, simulations follow a probabilistic approach. Multiple realization simulations using Monte Carlo analysis are conducted over time periods of up to one million years, which estimates a range of possible behaviors of the repository. The environmental impact is measured primarily by the annual dose received by an average member of a critical population group residing 20 km down-gradient of the potential repository. In addition to the nominal scenario, other exposure scenarios include the possibility of disruptive events such as volcanic eruption or intrusion, or accidental human intrusion. Sensitivity to key uncertain processes is analyzed. The influence of stochastic variables on the TSPA model output is assessed by ''uncertainty importance analysis'', e.g., regression analysis and classification tree analysis. Further investigation of the impact of parameters and assumptions is conducted through ''one-off analysis'', which consists in fixing a parameter at a particular value, using an alternative conceptual model, or in making a different assumption. Finally, robustness analysis evaluates the performance of the repository when various natural or engineered barriers are assumed to be degraded. The objective of these analyses is to evaluate the performance of the potential repository system under conditions ranging from expected to highly unlikely though physically possible conditions.