Two sets of analyses for the cases of groundwater release to well and sea ecosystems were conducted for the environmental impact assessment of high-level radioactive waste disposal facilities. After obtaining the respective BDCF (Biosphere Dose Conversion Factor) results for the scenarios of well-farming and marine water fishing using different biosphere assessment conceptual models implemented in ECOLEGO, they were compared each other. The purposes of these analyses are to identify reference generic biosphere conceptual models and to get insight on model uncertainty. In this study, the endpoint used for the comparison of the ECOLEGO biosphere models was the socalled Biosphere Dose Conversion Factor (BDCF), which is defined as the maximum value of the total dose to the exposed group, in Sv/yr, resulting from a continuous unit release of 1 Bq/yr during the whole simulation time either to the well compartment (BDCF_Well) or to the marine water compartment (BDCF_Sea). The radionuclides considered in the comparison were Cs-137, I-129, Nb-94, Ni-59, Ni- 63, Sr-90 and Tc-99. The conceptual models used in the biosphere assessment of the releases to a well are based on models that have been used by the DOE (simple-soil model) and SKB (complex-soil model) in safety assessments of radioactive waste repositories, respectively. Difference between two conceptual models used in the assessment of the releases to a sea is the number of compartments representing the sea; i.e., one model represents the sea with one compartment for the water and one for the sediment (singlecompartment model), whereas the alternative model uses two compartments for the water and the sediments: one for the inner coast and one for the outer coast (double-compartment model). The results of the BDCF_Well to a farmer obtained with the DOE and SKB models are shown to be very close to each other. Despite the differences in conceptual models and parameters, the results are within a maximum difference of a factor of 4. The results from the SKB model were higher for all radionuclides. The values of the BDCF_Sea obtained with the single- and double-compartment models are shown to be larger differences with a maximum order of 2. For all studied radionuclides, the double-compartment model produces higher BDCFs than does the single-compartment model. The differences would be due to activity concentrations in both water and sediments. Since the hydrodynamic behavior assumed for flow in the sea could significantly influence the dilution volumes and hence the concentrations, it is found that site-specific investigations are necessary to establish an appropriate marine biosphere conceptual model.

As part of the safety case development for generic disposal sites in Korea, it is necessary to develop generic assessment models using various geosphere–biosphere interfaces (GBIs) and potentially exposed groups (PEGs) that reflect the natural environmental characteristics and the lifestyles of people in Korea. In this study, a unique modeling strategy was developed to systematically construct and select Korean generic biosphere assessment models. The strategy includes three process steps (combination, screening, and experts’ scoring) for the biosphere system conditions. First, various conditions, such as climate, topography, GBIs, and PEGs, were combined in the biosphere system. Second, the combined calculation cases were configured into interrelation matrices to screen out some calculation cases that were highly unlikely or less significant in terms of the exposure dose. Finally, the selected calculation cases were prioritized based on expert judgment by scoring the knowledge, probability, and importance. The results of this study can be implemented in the development of biosphere assessment models for Korean generic sites. It is believed that this systematic methodology for selecting the candidate calculation cases can contribute to increasing the confidence of future site-specific biosphere assessment models.

The “shadow zone” is defined as a region below a flow obstacle, such as a vault, in unsaturated soils. Due to the capillary discontinuity of the cavity, water saturation on the top and side of the cavity is higher than the ambient saturation. On the bottom of the cavity, however, there is a region where water saturation is lower than ambient saturation. Undoubtedly, a shadow zone may also exist below a LILW disposal vault built in subsurface soils above the water table before the vault is fully degraded. During the degradation, flow in the shadow zone is controlled by the rate of water infiltrating the degrading vault. In this study, as one of the efforts to be made for enhancing safety margin by a realistic safety assessment of the engineered vault type LILW disposal facility, the shadow zone effect is investigated by a numerical parametric study using AMBER code. The conceptual model and data were excerpted from IAEA, ISAM Vault Test Case for the liquid release design scenario. It is assumed that the nearfield barriers degrade with time. In order to compare a visible shadow zone effect, the vault degradation period is assumed to be both 500 and 1,000 years, and the shadow zone depth to be varied according to unsaturated zone lithology. It can be seen that with a shorter shadow zone (2.7 m), radionuclides arrive at the water table earlier than with a full shadow zone (55 m) due to increased advection rate in the unsaturated zone. This effect tends to be more visible in the case of a longer degradation period. For radionuclides with short residence time relative to their half-lives in the unsaturated zone, such as Tc-99 and I-129, the radionuclides are shown to come out because they will arrive sooner, thereby allowing less peak release rate, when the shadow zone effect is considered. Once the vault is completely degraded and the infiltration rate of water flowing through the vault is equal to the ambient rate, the shadow zone effect disappears. In this example calculations using IAEA ISAM Vault Test Case input parameters, it might not be shown a significant shadow zone effect. Nevertheless, when the extent of the shadow zone is determined through more sophisticated hydraulic studies in the unsaturated soils surrounding the vault, the shadow zone effect would be checked up on the realistic near-field radionuclide transport modeling in order to contribute to gaining safety margins for post-closure safety assessment of the Wolsong 2nd phase LILW disposal facility.

In biosphere assessment modeling for the safety assessment of the Wolsong LILW disposal facility, the multi-compartment modeling in which all radionuclides transport is described quantitatively in terms of transfer factors between various environmental compartments has been implemented. In order to reflect the actual transfer mechanisms of 14C in the environment the specific activity (SA) modeling approach can be applied as an alternative to the previous transfer factors (TF) approach. The assumption of full SA equilibrium throughout the terrestrial environment is completely satisfactory for 14C release to the atmosphere if the 12C is emitted as 14CO2. This is the only form that is readily taken up by plants, so that active carbon is incorporated into the plant via photosynthesis at the same rate as stable carbon. Accordingly, the 14C concentration in Bq/g stable carbon is the same in the plant as it is in the air. And animals take up carbon almost entirely through ingestion and the SA ratio in the plant is maintained in the animal. In this study, a specific activity model for 14C was implemented in a GoldSim biosphere assessment model. From the literature survey for existing specific activity models developed, the IAEA model was selected. The farming scenario utilizing well water was simulated and the resulting ingestion dose conversion factors (DCFs) from the IAEA SA model were compared with those of the TF approach. The parameter value for the concentration of stable carbon in the air (gC/m3) is used as 0.20 gC/m3 considering the Suess effect. The dose coefficient for food ingestion used for dose calculations was taken from ICRP-72 as 5.8E-10 Sv/Bq. It was found that the ingestion DCFs of the SA model showed about 3 times lower than those of the TF model in the farming scenario through irrigation of well water, so it is expected that the SA approach could be applied for a more realistic assessment. Though the comparisons were made on the results from the terrestrial ecosystem only in this study, it would be necessary to investigate the applicability of the SA modeling approach for 14C through extensive comparisons and analysis including an aquatic ecosystem, and through parameters survey suitable to the domestic condition.