Technical Report NTB 14-10

Modelling of Radionuclide Transport along the Underground Access Structures of Deep Geological Repositories

The arrangement and sealing of the access routes to a deep geological repository for radioactive waste should ensure that any radionuclide release from the emplacement rooms during the post-closure phase does not by-pass the geological barriers of the repository system to a significant extent. The base case of the present study, where realistic values for the hydraulic properties of the seals and the associated excavation damage zones were assumed, assesses to what extent this is actually the case for different layout variants (ramp and shaft access and shaft access only). Furthermore, as a test of robustness of system performance against uncertainties related to such seals and the associated excavation damage zones, the present study also considers a broad spectrum of calculation cases including the hypothetical possibility that the seals perform much more poorly than expected and to check whether, consequently, the repository tunnel system and the access structures may provide significant release pathways.

The study considers a generic repository system for high-level waste (HLW repository) and a generic repository system for low- and intermediate-level waste (L/ILW repository), both with Opalinus Clay as the host rock. It also considers the alternative possibilities of a ramp or a shaft as the access route for material transport (waste packages, etc.) to the underground facilities. Additional shafts, e.g. for the transport of persons and for ventilation, are included in both cases.

The overall modelling approach consists of three broad steps:

  1. The network of tunnels and access structures is implemented in a flow model, which serves to calculate water flow rates along the tunnels and through the host rock.
  2. All relevant transport paths are implemented in a radionuclide release and transport model, the water flow rates being obtained from the preceding flow model calculations.
  3. Individual effective dose rates arising from the radionuclides released from the considered repository system are finally evaluated using biosphere dose conversion factors.

The detailed modelling approaches used are to some extent rather simple and stylised, which is deemed acceptable for the purpose of the present study. In addition, all simplifications clearly lead to an overestimate of releases.

The results obtained for the HLW and for the L/ILW repository are generally similar. Quantitative differences arise from the size of the main facility, from the inventory and the properties of the emplaced waste, as well as from the concepts of how radionuclides are released from the different waste forms.
The results show that, for realistic parameter values, radionuclide release along the access structures of a deep geological repository is extremely low. Thus, they confirm the reference assumption that radionuclide release occurs predominantly through the host rock. Globally increasing the hydraulic conductivities that are assumed for the tunnel system and the seals (including the excavation damage zone along these underground structures) increases the calculated flows along all access routes, but the increase is found to level off with increasing hydraulic conductivity of seals and excavation damage zones, as flow becomes ultimately controlled by the limited capacity of the host rock to supply water. The dose rate maxima due to releases via the access structures show the same asymptotic behaviour as the flow, and remain low in all cases.

The calculated release rates along the access structures for cases where realistic hydraulic properties of the seals and the excavation damage zones were assumed are very low for all variants considered and, in particular, far less than the release rates from the host rock. It is thus concluded that the type of main access route to the underground facilities of a deep geological repository with properties as assumed for the present study is not relevant to its post-closure safety.

Finally, even for highly unfavourable parameter values for the hydraulic properties of the seals and the associated excavation damage zones, the calculated dose rates remain well below the regulatory protection criterion of 0.1 mSv per year, often by several orders of magnitude. This finding, and taking into account the fact that for these calculations a number of simplifications were made that clearly lead to an overestimate of releases, demonstrates the robustness of the repository systems considered in the present study with respect to variations of the hydraulic properties of the seals and of the associated excavation damage zones.