A Dynamic Bayesian Network for the Performance Assessment of Nuclear Waste Repositories Undergoing Chemical Degradation due to Climate Change

Concrete is used as a barrier to prevent the release of radioactive contaminants from nuclear waste repositories. Concrete suffers of two main chemical degradation processes that can be expedited by climate change: carbonation-induced corrosion and chloride ingress. In this work, a Dynamic Bayesian Network (DBN) is developed for evaluating the concrete degradation induced by variations in climatic factors that affect corrosion, e.g., temperature, relative humidity and CO2. The DBN is applied with reference to near-surface waste disposal of literature, in support to the assessment of the risk of aquifer contamination due to radioactive contaminants release.