Geophysical monitoring of high-level radioactive waste repositories

Non-invasive monitoring of radioactive waste repositories is one of the key objectives addressed in the MODERN2020 project. For this task, geophysical techniques offer excellent means. Previous studies have identified seismic full waveform inversion (FWI) to be the most promising option for delineating subtle changes within a repository using data acquired outside of the repository. Significant anisotropy of the host rock, particularly in clay environments, precluded so far application of FWI technology for repository monitoring. With the development of a novel model parameterization, this problem could be resolved. Moreover, incorporation of structural constraints further improved the quality and reliability of our FWI algorithms. This was demonstrated with a field data set acquired in the Mont Terri rock laboratory. For a better characterization of small differential changes between two consecutive experiments, a novel differential tomography methodology was developed. It was tested with field data sets, with which differential traveltime inversions were performed. It is expected that this new method can be transferred in a straightforward manner to FWI problems. FWI technologies require extensive data analyses and substantial computer resources. Therefore, it was checked, if it is possible to employ quick and inexpensive tools, with which temporal changes in a repository can be detected, but not necessarily imaged. For that purpose, an anomaly detection algorithm was developed, and it will be tested with field data. In addition to seismic methods, geoelectrical techniques can provide valuable information for repository monitoring. For that purpose, tomographic algorithms for geoelectrical and induced polarization data were established and tested with laboratory data. For transferring the electrical parameters, obtained from these tomographic inversions, into relevant physical parameters, such as temperature and moisture content, calibration measurements were performed, and constitutive relationships between these parameters were established.