Estimation of ground motion due to induced seismicity at a geothermal power plant near Munich, Germany, using numerical simulations

In the North Alpine Foreland Basin, especially in the greater Munich area, several geothermal plants exploit a deep hydrothermal reservoir. So far at three sites events with ML > 2.0 were detected, in an otherwise seismically inactive region. In this study we investigate a site east of Munich, where moderate seismicity started to appear around five years after the beginning of geothermal circulation. Two larger events (ML2.1/1.8) occurred in December 2016, followed by an additional ML2.1 event about 10 months later in September 2017. All three events occurred at about 3 km depth and were felt by the population. As this is a densely inhabited area, the estimation of the maximum ground motion and its distribution are of great importance for the authorities and the public. We perform 3D seismic simulations of the main events using the spectral element code SALVUS. The results can supplement recorded and macroseismic data in order to estimate the possible seismic impact in the area. In addition, we evaluate the influence of the uncertainties contained in the event parameters and in the subsurface model on the maximum peak ground velocity (PGV) values to calibrate the simulations. The simulated waveforms are mostly in good agreement with the ground motion recordings. Furthermore, the ground motion distribution coincides with the macroseismic data. According to the simulation results, even the largest event in the area did not exceed the critical PGV value of 5 mm/s defined by German norms and therefore had no damage potential. Such a numerical approach can help to improve the seismic monitoring network, identify affected zones and mitigate the seismic risk.

Automated summary

In this study, moderate seismicity appeared around five years after the beginning of geothermal circulation in a site east of Munich. By performing 3D seismic simulations, the possible seismic impact in the area can be estimated and used to improve the seismic monitoring network.