I started at the SED as an intern—and with a particular interest: communicating the potential impact of earthquake shaking through seismic risk. I quantified the seismic risk in terms of human loss; I process varied data in a system of four models (viz. seismicity, hazard, risk, and loss model). Since our group works on earthquake forecasting too, I approached time-varying (i.e. short-term) loss forecasts. With this concept, I searched for ways to improve risk communication to decision-makers and to the public during a seismic sequence. I dedicated the first part of my PhD to that research topic.
My current research focusses on the long-term behavior of induced seismicity at the terminated Enhanced Geothermal System (EGS) in Basel. Seismic monitoring at this site is running for more than ten years and covers several stages: hydraulic pre-stimulation in November 2006, 6-day-long fluid injection for reservoir creation in December 2006 with subsequent injection stop after a widely felt ML3.4, ultimate shut-in (closure) of the borehole in 2011, and renewed increase of seismicity from mid-2012 onwards.
The existing catalog is incomplete. To analyze seismicity of the geothermal reservoir over its whole lifetime, I need to produce a seismic catalog with homogeneous detection sensitivity and consistent magnitude estimates. I apply the template-matching technique that takes advantage of waveform similarity: scanning with waveforms of known earthquakes for finding unknown earthquakes (especially smaller events) in the 10-year-long seismic recordings. For best sensitivity, I scan the data of the deepest borehole station (OTER2, 2.7km depth) which is very close (1.5–2.5km) to the ~4.5km-deep reservoir.
Compared to the previous catalog, we find ~ten times as much events (>100’000). The high temporal and spatial resolution of the produced catalog allows us to analyze the statistics of the induced Basel earthquakes in great detail. We resolve spatio-temporal variations of the a- and b-value that have not been identified before, enabling us to derive the first high-resolution temporal development of the seismic hazard for the Basel EGS reservoir for the last 10 years.
I have also started to extend such analyses to natural sequences in Switzerland (see the news section and the news archive).