[en] We implemented a quantitative method based on a 1D finite-differences code for modeling the transport of radionuclides in the subsurface. The code 'TransPyREnd' includes diffusion, advection, equilibrium sorption, decay of radionuclides, and the build-up of the daughter nuclides. An outline of the method was give. While this approach is robust, it crucially relies, for instance in clay rocks, on knowledge of parameters like the porosity, the nuclide- and geology-specific diffusion coefficients, sorption properties, and other transport parameters. While in addition to existing research different research programs are funded to close existing gaps in the data, all of the relevant parameters come at a varying degree of uncertainty that needs to be accounted for. In order to exploit the limited data as efficiently as possible, we use and modify a parameter workflow proposed by Nagra for calculating the diffusion coefficients and the sorption parameters depending on a small number of input parameters, like the temperature and the effective porosity of the relevant layer. We extend this workflow to allow for differing geochemical conditions as expected in the respective German areas. In post-processing, we define several indicators to consolidate the results of these assessments. All indicators are based on the notion of an effective transport length, and defined as the minimum distance from the repository for which the legal limits on radionuclide migration are fulfilled. This transport length is then compared with the total distance from the repository to the boundary of the host rock, resulting in intuitive and easily understandable measures for the performance of containment. Depending on context, different variants of this measure are calculated. To account for parameter uncertainties, we make extensive use of non-deterministic ensemble calculations to propagate the uncertainties of the input parameters to the model results, namely, the transport length. Since the total number of relevant parameters is of the order of 1000, we apply methods of sensitivity analysis to allow a reduction of the number of parameters considered in the ensemble calculations. This also helps in building a deeper understanding of the modelled system behavior and its most relevant properties

[en] In this contribution, we present the general concept, technical layout, and expected scientific impact of the DR-D experiment, as well as first results from field and related laboratory studies