Brine intrusion by upconing for a high-level nuclear waste repository at Forsmark: Scoping calculations

Abstract

SSM currently reviews a license application for a spent nuclear fuel repository that is proposed to be located at Forsmark, Sweden. The repository is to be situated at 500 m depth in the rock and copper canisters are deposited in holes excavated from the tunnel system. To protect the canisters they are surrounded by a bentonite clay buffer, which is to swell when getting in contact with water. The swelling properties are dependent on the salt content of the water and excessively high salt contents may inhibit the swelling. Thus it is important to ensure that the bentonite is not subjected to water with too high salt contents. The salt content of the groundwater increases with depth and is expected to reach levels that may affect buffer performance at large depths. When excavating the repository very high hydraulic gradients are established and water and salt movement from the depth to the repository, so-called ‘upconing’, could possibly occur. The objective of this study is to evaluate the possibility of salt-water migration to the repository. This objective is motivated by the adverse impacts of water with too high salinity entering the repository and by the uncertainty of the relevant hydraulic and hydrogeochemical conditions at the Forsmark site at great depths. To analyse density dependent flow and salt transport at the Forsmark site the USGS’ SUTRA code is used. This study proceeds by finding critical model cases for which upconing does or does not occur, while assessing whether the parameterizations of these cases are realistic for the Forsmark site. In addition, the fall of the upconed salt mound (i.e. downconing) following closure of the repository is also evaluated. In particular the objectives are (1) to determine the factors that control saltwater upconing in a hydrogeological setting representative of Forsmark; (2) to relate these factors to the plausible conditions prevailing at the repository site; (3) to investigate whether the proposed repository is likely to generate saltwater upconing, given the range of uncertainty in hydrogeologic structure and parameter values; and (4) to evaluate the timing of upconing (salinization) and the timing of downconing (freshening) following repository closure for cases where upconing occurs. The results of this simulation analysis show that upconing behavior is strongly affected by the ratio of permeability to porosity in any zone in which upconing might occur. Within the full range of parameters that are likely to occur at the Forsmark site, the model yields either no significant upconing at all during the operational period of the repository or intrusion of brine-type waters after only one to a few decades.