Review comments on the SR 97 post-closure safety assessment

These review comments concern an assessment of the long-term safety of a deep repository for spent nuclear fuel, titled Safety Report 97 (SR 97), which was prepared by the Swedish Nuclear Fuel Waste Management Company (SKB). The primary focus of this review is on hydrogeologic issues relating to groundwater flow, hydrologic uncertainty, and the potential for radionuclide transport from leaking canisters. The main hydrological model that was used in SR 97 is based on a continuum conceptual model of groundwater flow in fractured bedrock. Major problems with this model include the following: The validity of the continuum model is arguable for the type of rock that is present at these sites. The suitability of the model for the intended purpose of predicting streamlines and travel times for groundwater flow through the rock mass has not been adequately demonstrated. The comparison with alternative, discrete models yielded more divergent results than has been recognized in the SR 97 reports. The comparison with alternative models did not consider significant, realistic sources of uncertainty in the alternative models, evaluation of which would have likely led to greater divergence. The SR 97 model of radionuclide transport is based on a 1-D streamtube formulation, within which the predicted release of radionuclides to the biosphere is dominated by a parameter called the F ratio. A key factor in this parameter is the flow wetted surface. All of the hydrologic models used in SR 97 relied upon essentially the same set of geometric assumptions to estimate flow wetted surface from conductive fracture frequency in boreholes. Hence the predictions of the alternative models are not independent. Alternative methods of estimating flow wetted surface are needed to obtain a realistic evaluation of the uncertainty regarding radionuclide release. The alternative 3-D hydrologic models were used only to predict streamtube parameters, not for actual transport simulations. Hence the comparison between the main hydrologic conceptual model and alternative models does not include a full assessment of the effects of flow field complexity on radionuclide transport. As this is one of the major distinctions between the continuum model and the alternative models, the comparison must be regarded as incomplete. Besides these major hydrological issues, miscellaneous comments are offered on aspects of the repository system design and site descriptions that relate to hydrology