Modelling groundwater flow changes due to thermal effects of radioactive waste disposal at a hypothetical repository site near Sellafield, UK

High-level radioactive waste (HLW) and spent fuel from nuclear power plants (SF) require very long duration disposal because of its long-lived toxicity. Specific repository design is necessary to accommodate the associated large heat generation from continual decay. Conceptual designs propose geological containment within salt, claystone, or crystalline bedrock. However, while many studies have investigated the safety of disposal in these host rocks, the interaction of near-field heat coupled to far-field groundwater flow for UK-relevant conditions has not been fully investigated. Using an excellent dataset from a repository investigation in saturated fractured rock at Sellafield, UK, with deep groundwater flow, a preliminary investigation of the effect of the addition of heat from HLW and SF on regional groundwater flow is undertaken. A coupled thermo-hydrogeological groundwater flow model dependent on fluid density and viscosity has been calibrated to field data, before heat is added to simulate waste. Substantial heating reduces fluid density and dynamic viscosity. Groundwater flows upwards through permeable damage zones of high-angle faults and increases natural velocity by 1.5x. Elevated groundwater temperatures are predicted at the surface in just hundreds of years. Travel times for conservative chemical species from the hypothetical repository to the St. Bees Sandstone aquifer are reduced by over one-third if engineered barriers fail at the point of containment, and remain reduced by hundreds of years for particles released thousands of years after containment. The results imply that in conditions similar to those modelled, the high-performance guarantee of the engineered barrier is essential to repository safety. A very good understanding of the coupled interaction of near-field heat and far-field groundwater flow will be important for safety cases, especially accuracy concerning heterogeneous fault discontinuities. It is necessary to model heat advection and faults at fractured sites.