Abstract. In Finland, Olkiluoto Island has been selected for constructing a spent nuclear fuel disposal facility. With the approaching licensing step, submitting the application for nuclear construction licence in 2012, all components of the safety case are becoming more site-specific. In order to assess compliance with regulatory requirements, a prospective deterministic dose assessment methodology based on the ICRP concept of assessing doses to the representative person was developed and implemented in the interim safety case 2009, to be further refined for the safety case 2012. To facilitate the assessment of doses to the most exposed people and other people at the evolving site, full dose distributions are derived, comprising the dose to each potentially exposed person in each generation.

Uncertainty in estimations of the exposure of non-human biota to ionising radiation may arise from a number of sources including values of the model parameters, empirical data, measurement errors and biases in the sampling. The significance of the overall uncertainty of an exposure assessment will depend on how the estimated dose compares with reference doses used for risk characterisation. In this paper, we present the results of a study of the uncertainty in estimation of the exposure of non-human biota using some of the models and parameters recommended in the FASSET methodology. The study was carried out for semi-natural terrestrial, agricultural and marine ecosystems, and for four radionuclides (137Cs, 239Pu, 129I and 237Np). The parameters of the radionuclide transfer models showed the highest sensitivity and contributed the most to the uncertainty in the predictions of doses to biota. The most important ones were related to the bioavailability and mobility of radionuclides in the environment, for example soil-to-plant transfer factors, the bioaccumulation factors for marine biota and the gut uptake fraction for terrestrial mammals. In contrast, the dose conversion coefficients showed low sensitivity and contributed little to the overall uncertainty. Radiobiological effectiveness contributed to the overall uncertainty of the dose estimations for alpha emitters although to a lesser degree than a number of transfer model parameters.

The development of a radionuclide transport model for a tunnel of a repository for short-lived low and intermediate level waste is presented in this work. The model is two-dimensional and uses the finite element method to describe a cross-section perpendicular to the tunnel's main axis. It is shown that the use of a realistic geometry of the system, albeit in 2D, results in a transparent model that can handle the equations in a flexible way. The geometry can easily be changed and also extended to a 3D representation without affecting the numerical code. The disadvantage is the computational requirement. It is expected however that radionuclide transport modelling will evolve to 2D or 3D representations that are more realistic and transparent than today's 1D analytical models which involve many simplifications whose impact on the output is difficult to evaluate.,1752'8&7,21An important fraction of the low and intermediate level waste from the decommissioning of nuclear power plants is short-lived. In Sweden it is envisaged that this short-lived low and intermediate level waste (LILW) will be isolated in underground tunnels at a depth of approximately 300 meters in a crystalline rock formation. Safety studies related to the final disposal of this type of waste are not at the same advanced stage as for high-level nuclear waste or spent nuclear fuel. Nevertheless, the disposal of LILW poses some major technological challenges that require a substantial effort to be put on performance studies of a repository for this kind of waste in the years to come.In this work a model for the estimation of radionuclide migration in the near-field of a repository tunnel has been developed. The investigation has been conducted for regulatory purposes. In the next section we briefly describe the Swedish Nuclear Fuel and Waste Management CO (SKB) concept of the SFL-5 repository. In Section Three we introduce the 2Dmodel and the input data. The discussion of the results is given in Section Four. Section Five presents the conclusions and the summary.$ 5(326,725< )25 6+257/,9(' /,/:A convenient way to isolate the low-level waste is to enclose it in tunnels. The Swedish Nuclear Waste and Management Company, SKB, has performed one preliminary analysis study of a repository 'concept' for this type of waste [1]. According to SKB, the short-lived waste will be placed in a system of tunnels which form the SFL 3-5 repository. A schematic view of this Mat. Res. Soc. Symp. Proc. Vol. 807