The presence of faults in the subsurface, crossing geothermal targeted aquifers introduces concerns related to induced seismicity. Improving the understanding of geomechanical processes due to geothermal operations is therefore important. Previous research has indicated that ther
...
The presence of faults in the subsurface, crossing geothermal targeted aquifers introduces concerns related to induced seismicity. Improving the understanding of geomechanical processes due to geothermal operations is therefore important. Previous research has indicated that thermally induced stresses can have a significant effect on fault stability, especially in the long term. In depleting gas fields, the effect of fault offset has been studied and found to promote the onset of fault reactivation. However, little research has been conducted on this effect in geothermal reservoirs. This thesis will investigate the effect of juxtaposition of sandstone and claystone across faults on thermo-elastic stress changes and induced fault slip in geothermal projects targeting porous sandstone reservoirs.
This research question will be answered by constructing three different model geometries in a semi- analytical 2D plane-strain model assessing thermo-elastic stress changes and fault slip. The geology and properties of the Pijnacker Geothermie reservoir, located in the West Netherlands Basin and targeting the Delft Sandstone Member as its reservoir rock are used for the input of the model. The three geometries include a model scenario in which sandstone is fully juxtaposed against sandstone, a scenario in which sandstone is fully juxtaposed against claystone and lastly, a scenario with half-a-reservoir-thickness fault offset, juxtaposing sandstone against both sandstone and claystone. These geometries were chosen to represent a fault that experiences cooling on both sides, on a single side and on both sides, but with an offset. A comparison was made between the three model scenarios using base cases and evaluating the differences in stress changes and fault slip. A sensitivity analysis was performed in order to determine the key parameters influencing the results the most. Then, three metrics were extracted from the base cases and plotted against the varying key parameters. These metrics are: the temperature change needed to reactivate the fault, the maximum slip and the slip patch length. Lastly, heterogeneous friction was used in the model in order to assess its effect and make a step towards a slightly more realistic model. The results showed that the double sided cooling case with fault offset was the most destabilizing configuration, needing the least amount of cooling before reaching the onset of fault reactivation and also consistently experiencing the largest maximum slip. The single sided cooling model scenario is the most stable, however its slip patch length is the most sensitive to variations in key parameters, generating the largest slip length for destabilizing values of the key parameters. The inclusion of heterogeneous friction had a destabilizing effect on all scenarios. As the degree to which sandstone is juxtaposed against sandstone affects the amount of thermo-elastic stress changes and slip a fault experiences, it should be considered to include fault offset when assessing fault slip in geothermal projects. Not including fault offset will likely underestimate the thermo-elastic stress changes and subsequent induced fault slip. Panther can capture the effect of thermo-elastic stress changes in a first-order assessment on a simple reservoir geometry containing the essential parameters while avoiding extra model complexity. Further improvements to the model can be made to capture the effect of certain aspects that have been proven to affect thermo-elastic stress changes. These include property contrasts between the reservoir and surrounding rock as well as the addition of clay-rich layers in the sandstone formation and vice versa. Further geological work could be performed to decrease the uncertainty around the location of
the fault and its dip angle.