Microseismic Activity in Low-Hazard Geothermal Settings in Southern Germany

In the last few years several geothermal power plants have taken up production in southern Germany and many more are currently in exploration or construction stages. One of the most promising areas in Germany with respect to geothermal exploitation is the Bavarian Molasse Basin: A karstified limestone formation at a few kilometers depth provides sufficiently high temperatures also for power production and its high permeability obviates the need for hydraulic stimulation of the reservoir. Furthermore the local setting is being considered as generally aseismic with very sparse and weak seismicity. The seismic hazard going along with production is therefore being considered negligible and no particular efforts for seismic monitoring were made. In 2008, however, an unexpected Ml 2.3 event which was felt by locals attracted public attention. With no stations near the epicenter, the event still was located in the general vicinity of a geothermal plant that took up production about half a year earlier. In the last two years a temporary network was set up that recorded more than thirty events with magnitudes mainly ranging from Ml 0 to 1.5. The data recorded in the local network are used for absolute locations. A high resolution 3-D P-wave velocity model is constructed from data of a dense 3-D seismic survey conducted to image the geothermal reservoir. An S-wave velocity model is compiled from converted shear waves, an old survey with shear wave excitation and cluster analysis of Vp/Vs ratios using the recorded events. Results show the hypocenters close to the bottom of the injection well. Results of waveform similarity analysis and clustering of events are presented along with preliminary results of further work focusing on hypocenter relocation via master-slave analysis and differential location techniques. Still, the exact extent of the man-made influence on the seismicity remains arguable. Events below magnitude 1.5 could not be detected prior to the production of the geothermal plant in the main network of the local earthquake service. Questions also remain as to the unique features of this geothermal site in comparison with others not far away that do not produce notable induced seismicity. How the fault targeted by the well is oriented in the present-day stress field is suspected to have a major impact on fault reactivation. An upcoming field experiment is about to address these open questions and test some working hypotheses. It is located at another geothermal reservoir in a very similar setting where a geothermal power plant is currently in the construction stage. In addition to monitoring the production stage, we intend to collect one year of pre-production data of the unperturbed reservoir. This will enable us to better assess the influence of fluid injections on local seismicity at geothermal plants and will lead to a better understanding of the reasons for the observed microseismicity.