The integration and abstraction of EBS models in yucca mountain performance assessment

Design and performance assessment of the EBS requires models that describe how the EBS and near field behave under anticipated repository-relevant conditions. These models must describe coupled hydrologic, thermal, chemical, and mechanical (TFICM) processes that drive radionuclide transport in a highly fractured host rock, consisting of a relatively permeable network of conductive fractures in a setting of highly impermeable tuff rock matrix. An integrated performance assessment of the EBS must include a quantification of the uncertainties that arise from (1) incomplete understanding of processes and (2) from lack of data representative of the large spatial scales and long time scales relevant to radioactive waste disposal (e.g. long-term metal corrosion rates and heterogeneities in rock properties over the large 5 km- emplacement area of the repository). A systematic approach to EBS model development and performance assessment should include as key elements: (1) implementation of a systematic FEPs approach, (2) quantification of uncertainty and variability, (3) sensitivity analyses, and (4) model validation and limitations. The approaches used for these key elements in the Yucca Mountain repository programme are described in Section 2 of this paper. A specific example of Yucca Mountain EBS model development and integration, related to the modeling of localised corrosion of Alloy 22, is discussed in Sections 3 and 4. (authors)