???ViewItemPage???

???ViewItemOverview_lblAbstractSection???

As part of the project ThermoBase we try to understand the underground temperature distribution as it is essential to determine the geothermal energy potential, heat flow, and mechanical behavior of lithological units, as well as the long-term safety of heat-producing waste in repositories. As temperature observations in the subsurface are sparse, numerical modelling is utilized to make predictions and risk-assessments. In contrast to statistical models, numerical physical approaches can account for contrasts in thermal properties, given well informed structural and parametric models of the subsurface. Recent studies suggest that the Last Glacial Period and the Holocene Climate Optimum present the largest contribution to the climate impact on the modern subsurface temperature distribution in Germany. Previous studies, mainly in Northern Europe, have shown, that an additive effect of the Pleistocene Glaciations can be observed, with a total cooling of several Kelvin in up to two kilometer depth.
In the NGB, complex salt dynamics shape the structure of overlying sediment layers. Given the spatial orientation and contrast in thermal parameters, heat refraction may play a significant role on how the paleoclimatic imprint is distributed in the subsurface thermal field. To understand the influence of transient processes we parametrize thermally relevant units in unstructured 3D finite element meshes considering regional constraints.
We apply a heatflow dervied from boreholes in the model regions as lower thermal boundary condition. The upper boundary condition is derived from soil temperatures in transient global circulation models starting at the last glacial maximum and scaled δ18O
as a proxy for the time prior. Furthermore, we define stochastic distributions for petrophysical parameters in the different homogeneous regions from laboratory measurements, core-log interpretation and the KW-DB. Stochastic sampling methods such as Monte Carlo and stochastic collocation are employed to quantify the impact of uncertain parameters on the temperature field.
We present and discuss preliminary results showing a strong interaction between salt-dome geometry and transient boundary conditions, with heat refraction, latent heat, and anisotropy shaping the spatial pattern of anomalies.