Statistical analyses on the relevance of thermal data for the safety-related assessment of repository systems - first results from the project ThermoBase

The ThermoBase project aims to investigate the thermal field in sedimentary areas in Germany. It was initiated by the Federal Company for Radioactive Waste Disposal (BGE) and motivated by design calculations and safety analyses that are to be carried out as part of the site selection process for a high-level radioactive waste repository. This research project is carried out jointly by the GFZ German Research Centre for Geosciences and the Federal Institute for Geosciences and Natural Resources (BGR). This contribution focuses on stochastic computations from the BGR on the influence of parameters on the transient temperature field in the underground.
Based on research previously completed by the BGR [1, 2], comparative numerical analyses of the temperature distribution for typical geological situations in Germany are conducted. The focus is on the configuration of thermal material parameters, boundary conditions, and the uncertainty in these parameters. Another objective is to investigate the temperature development and penetration depth of permafrost during a possible future glaciation. Statistical methods, such as the Monte Carlo method and stochastic collocation, will be used to make quantitative statements on the impact of changes in these parameters. Methods for the quantification of uncertainties in geological formations related to heat-generating waste repositories [3, 4], which have recently been developed at the BGR and in the joint project MeQur/URS [5], are used here.

Different generic models are intended to be used for this purpose: (1) a 1D-model in which wide parameter variations will be considered, covering a broad range of different rock types, (2) a 2D-model, representing a case study of a repository system in the Lower Cretaceous claystone in northern Germany [1], and (3) a case study of a salt dome in northern Germany, represented by a 2d rotational symmetric model. The salt dome model includes a rock salt sequence within the Zechstein salt layer where the repository is located.

The first results of this work will be presented. Different studies will be shown for the one-dimensional model (1) quantifying the impact of the modeled uncertainties, including global sensitivity analyses. For the two-dimensional model (2), the effect on the temperature distribution of a stochastic state space, defined by uncertain parameters selected based on results from [3], will be demonstrated and first investigations of the 2D salt dome model (3) will be discussed.

References
[1] Maßmann, J. & Ziefle, G. (2017): Integritätsnachweis geologische Barriere. (In: Jobmann, M. et al.: Systemanalyse für die Endlagerstandortmodelle - ANSICHT). Ber.-Nr.: TEC-29-2016-TB; Peine, Hannover, Braunschweig (DBE TECHNOLOGY, BGR, GRS).
[2] Mönig, et al. (2020): RESUS: Synthesebericht. GRS; 567; Köln.
[3] Maßmann, J. et al. (2022): ANSICHT-II – Methode und Berechnungen zur Integritätsanalyse der geologischen Barriere für ein generisches Endlagersystem im Tongestein. BGR, Hannover. DOI:10.25928/n8ac-y452.
[4] Bittens, M., & Gates, R. L. (2023): DistributedSparseGrids. jl: A Julia library implementing an Adaptive Sparse Grid collocation method. Journal of Open Source Software, 8(83), 5003.
[5] Nagel, T, et. al. (2023): MeQur - Uncertainties in THM-coupled integrity calculations, Project web page; URL: https://urs.ifgt.tu-freiberg.de/en/topics/mequr, [Accessed 09-01-2024]