¶ Uncertainty and Sensitivity Analysis
The uncertainty and sensitivity analysis accounted for all model parameters affected by uncertainties, such as longitudinal and transverse dispersivity, diffusion coefficients, and porosity. Also, the above described model was extended by the simulation of radionuclide transport. The model’s output is post-processed in order to obtain the maximum radionuclide concentration at the surface, and the impact of model’s parameter over this quantity has been evaluated through variance decomposition methods. This approach ensures both a robust risk assessment that considers all critical parameters and computational efficiency by disregarding the variability of parameters that have no impact on the model’s output. In the following table, Sobol-indices (both first order and total effect of the model’s parameter) are presented:
¶ Life Expectancy Simulations
Velocities from the density-dependent flow and transport simulations are used to simulate subsequent groundwater age and life expectancy, where the latter is used as safety indicator for radionuclide travel times. For clarification, groundwater age and life expectancy are calculated using transport equations including diffusion coefficient and dispersivity:
Findings include:
- Effect of dispersivity: Higher longitudinal dispersivity (αL) and transverse dispersivity (αT) lead to lower life expectancy along the salt dome boundary.
- Diffusion coefficient: Variability in diffusion coefficients significantly influences life expectancy, with higher coefficients resulting in lower maximum life expectancy.
- Underestimation the associated parameters may lead to overestimation of life expectancy and thus to critical overestimation of repository safety.
|
|