SpannEnD
Steffen_Ahlers_SpannEnD_Figure1Figure 1: Screenshot of the geomechanical-numerical model used. The Dimensions are ~1000 x 1250 x 100 km3. The central more colorful part corresponds approximately to the area of Germany.
Steffen AhlersEinleitung
The crustal stress state is an important information for many geological applications, e.g. directional drilling, stimulation and save usage of geothermal reservoirs or the search for and long-term safety of a high-level radioactive waste deposit. However, the state of knowledge for Germany is mainly based on pointwise, sparse and frequently incomplete data sets. Therefore, we developed a geomechanical-numerical model of Germany to provide a continuum mechanics based prediction of the crustal stress state in 3D which is calibrated with available data sets. Our recent model contains about 11.1 million available elements and therefore a main memory of up to 1.5 TB was needed. Since, the main memory of the server of our working group was too small we decided to use the Lichtenberg HPC for our calculations.
Methoden
The 3D geomechanical-numerical is based on available geological geometry models which describe the distribution of geological units in the subsurface of our model region. If necessary, additional geological data have been used to close gaps between or within these models. The final model covers an area of 1000 x 1250 km2 and reaches a depth of 100 km. The geometry model was extended by discretization and parameterization to a geomechnical-numerical model. Since, we assume linear elasticity an individual Poisson’s ratio and Young’s modulus was assigned to each geological unit (overall 22 units). Additionally, densities were assigned. Displacement boundary conditions are defined at the edges of the final model which are used for calibration with magnitudes of the minimum and maximum horizontal stress (Shmin and SHmax). For the validation additional magnitude data sets and orientations of SHmax are used. As FEM solver we used Abaqus™ v2019.
Ergebnisse
An improved prediction of the crustal stress state of Germany could be achieved by our geomechanical-numerical model, which provides the complete 3D stress tensor. Our model predicts an almost homogeneous N-S orientated pattern of the SHmax orientation. A median deviation of 0.3° in comparison to the mean orientation of SHmax derived from the World Stress Map (WSM, www.world-stress-map.org) project indicates an overall good fit. Furthermore, almost all results are within the standard deviation of the mean WSM data. However, some model regions show significant deviations. The results also show a good fit to all three principal stress magnitudes of the vertical stress (SV), Shmin and SHmax indicated by absolute differences of 0.0 MPa for SV, 4.6 MPa for Shmin and 6.4 MPa for SHmax. The differences to the calibration (Shmin and SHmax) data are mainly within in a range of +/- 10 MPa for the Shmin magnitudes and within a range of +/- 20 MPa for the SHmax magnitudes. Despite the overall good fit, some data indicate too low Shmin values in the upper part of our model. The Results are described and discussed in detail in Ahlers et al., 2022a. The model geometry and the results of our model have been published (Ahlers et al., 2022b).
Diskussion
Our model predictions are in good agreement with the mean orientations of the SHmax derived from the WSM and with magnitude data used for calibration and additional stress magnitude data of different geological regions, e.g. Molasse Basin, Upper Rhine Graben or North German Basin. However, some local differences occur and also a minor but possibly systematic deviation of the Shmin occur in the uppermost 1.5 km of our model. Some of these differences occur likely due to geological units which are not well represented in our model, e.g. salt units or weak sedimentary layers since we assume linear elasticity or thin units which can not be represented by our coarse vertical resolution of ~250 m.
