Reliable induced-seismicity forecasting based on a coupled-processes model

Forecasting induced seismicity is a challenging task. Most forecasting tools are phased on statistical approaches, incorporating little physics at most. As a result, these tools fail to forecast the complex subsurface response to fluid injection/extraction, like maximum-magnitude earthquakes occurring after the stop of injection. To overcome this issue, we have developed a physics-based forecasting tool that also takes into account the statistics of induced seismicity to forecast the frequency and magnitude of future events (Boyet et al., 2014a). The physics-based model solves the hydro-mechanical coupling and could also solve the thermo-hydro-mechanical coupling. This coupled model permits accounting for triggering mechanisms of induced seismicity other than just pore pressure diffusion. In particular, it considers poroelastic stresses, poromechanical stress relaxation after the stop of injection, and shear-slip stress transfer. We have applied this forecasting tool to the case of the enhanced geothermal system at Basel (Switzerland), where the maximum-magnitude earthquake was induced a few hours after the stop of injection. Our tool successfully forecasts the post-injection maximum-magnitude earthquake when reproducing the step-rate stimulation scheme that was used at Basel. Interestingly, a constant-rate or a cyclic stimulation would have not induced large-magnitude post-injection seismicity, according to the forecasting tool. We have also explored the effect of how injection is stopped has on induced seismicity (Boyet et al., 2024b). Simulation results reveal that, for the case of Basel, shutting-in the well would have led to lower magnitude earthquakes than bleeding-off the well, as performed at Basel. Additionally, a progressive decrease in the injection rate would have stabilize faults even further in the post-injection stage.

REFERENCES

Boyet, A. Vilarrasa, V, Rutqvist, J. and De Simone, S., 2024a. Forecasting fluid-injection induced seismicity to choose the best injection strategy for safety and efficiency. Philosophical Transactions Royal Society A, 382: 20230179

Boyet, A., De Simone, S. and Vilarrasa, V., 2024b. To bleed-off or not to bleed-off? Geophysical Research Letters, 51, e2023GL107926