Seismostatistical characterization of microseismicity observed at geothermal fields

Recently, occurrence of felt earthquakes has been recognized as one of the most critical environmental burdens associated with geothermal development. We have taken seismostatistical approach to evaluate characteristics of the microseismicity at geothermal fields to establish realtime and automated monitoring techniques of the reservoir changes and risk assessment of the felt earthquakes. In this study, we have introduced the Epidemic Type Aftershock Sequence (ETAS) model (Ogata, JASA, 1988) to statistically model the time series of occurrences and the magnitude of microseismic events from hydrothermal and EGS fields. Here maximum likelihood estimation has been employed to estimate optimum parameters of the ETAS model. We analyzed microseismic events observed at Yanaizu Nishiyama, one of the largest hydrothermal fields in Japan. In this field, four felt earthquakes with local magnitude larger than 3.0 occurred during production operation since 1996, although no clear correlation between the occurrence of the felt earthquakes and operation to the reservoir has been observed (Asanuma et al., Trans. GRC, 2011). We found that the occurrence rate of primary fluid signals, which are the events triggered by external forcing and have been interpreted to be independent from a series of aftershocks (Hainzl and Ogata, JGR, 2005), correlated to the reinjection rate (Fig. 1). However, no significant change in the other parameters in the ETAS model has been observed. We also analyzed microseismic events observed at Basel EGS site in Switzerland, where some felt earthquakes occurred during and after hydraulic stimulation. The estimated ETAS model demonstrated that there is a correlation between the occurrence rate of primary fluid signals and injection rate. We, however, found that there is limitation to fit the ETAS model to the induced seismic events and new seismostatistical model is required for microseismic reservoir monitoring.ig. 1 A relation among production/injection rate, seismic activity, and occurrence rate of primary fluid signal at Yanaizu Nishiyama. (a) Injection rate of water (blue) and injected volume (green). (b) Production rate of hot water. (c) Production rate of steam. (d) Occurrence rate of microearthquakes. (e) Cumulative number of microearthquakes. (f) Magnitude of the events. (f) μ (occurrence rate of primary fluid signal) (0<μ<3.2).