Building Stakeholder Confidence by Reducing the Gulf between Experimental Data and Model Predictions in Assessments of Repository Performance

Abstract

Confidence in the ability of repository systems to isolate high level wastes from the environment can be strengthened by placing greater reliance on robust designs for the repository system, and by using multiple lines of evidence to demonstrate that parameters, models and predictions developed for the repository are relevant. A particularly useful approach is to demonstrate that models and predictions incorporate processes that have been shown to be important in existing natural systems.