No abstract

The characterisation of potential repository sites will produce huge volumes of information, which must be correlated, quality assured, integrated, analysed, documented and archived in a rigorous and efficient manner. While some of this work involves rather routine data handling that may be easily automated, much of it requires input of tacit knowledge which involves the experience of expert staff. To provide support for the Japanese implementer and also the regulator, a JAEA team is attempting to capture both Japanese and international geosynthesis experience within a Knowledge Management System (KMS) framework, which is termed ISIS. This is a hybrid system that combines “smart” software with human experts, although an aim is to capture tacit knowledge within expert systems to the maximum extent practicable. Initial tests, based mainly on field work carried out by JAEA at the sites of the Mizunami and Horonobe underground research laboratories, have utilised expert systems as modules in a “blackboard system” approach to planning or implementing the processing of field data. Examples are presented of sub-systems where this approach has already been demonstrated and perspectives for more extensive application to integrated geosynthesis management are discussed.

Matrix diffusion is one of the key processes to be considered by safety assessments concerned with varied waste disposal in geological host rocks, but especially disposal of radioactive waste. An appropriate method is proposed for evaluating nuclide retardation by matrix diffusion along flow-paths in crystalline rock, using in-situ geological information. The usefulness of the method is evaluated and the results are summarized as follows: DDetailed in-situ characterization of flow-path structures shows that a flow-path can be classified into three types by its structural features. The structural information is important for supporting reliable qualitative analysis of matrix diffusion by laboratory experiments, e.g. to ensure that porosity measurements and diffusion experiments investigate relevant parts of the rock mass. aInformation from natural analogues, such as studies of the migration of natural uranium-series isotopes along flowpaths, is useful to build confidence in short-term laboratory experiments for evaluating the long-term diffusional processes. ®Mineralogical studies, microscopic observations and dating of the fracture-fillings would be valuable for estimating the long-term stability of fractures and are important for evaluating the usefulness of the methodology for evaluating matrix diffusion in the host rock.

No abstract

It is important for site characterization projects to manage the decision-making process with transparency and traceability and to transfer the technical know-how accumulated during the research and development to the implementing phase and to future generations. The modeling for a geological environment is to be used to synthesize investigation results. Evaluation of the impact of uncertainties in the model is important to identify and prioritize key issues for further investigations. Therefore, a plan for site characterization should be made based on the results of the modeling. The aim of this study is to support for the planning of initial surface-based site characterization based on the technical know-how accumulated from the Mizunami Underground Research Laboratory Project and the Horonobe Underground Research Laboratory Project. These projects are broad scientific studies of the deep geological environment that are a basis for research and development for the geological disposal of high-level radioactive wastes. In this study, the work-flow of the groundwater flow modeling, which is one of the geological environment models, and is to be used for setting the area for the geological environment modeling and for groundwater flow characterization, and the related decision-making process using literature data have been summarized.