Monitoring induced seismicity in the North of the Netherlands: status report 2010

This report presents an update on the results from monitoring induced seismicity in the North of the Netherlands. The latest report on this issue (Van Eck et al., 2006) is based on seismicity data from 1986 to 2003. In the mean time, the dataset more than doubled. The analysis is extended to include data up to 2010 and it has been decided to update reports every 5 years. The borehole monitoring network has been expanded, now also covering the northern part of the province of Friesland, and upgraded to a full real-time system. Expansion includes borehole seismograph stations and accelerometer stations. The upgrade enables a faster response at felt earthquakes and the determination of automatic first locations, which is subject of further development. New detection and location threshold maps have been calculated.

Seismicity continued to develop and shows a clear connection with existing faults at reservoir level. Since 2003 the Groningen field dominates in activity, with an increase in mainly smaller magnitude (M <1.5) events. This behaviour is reflected in the calculation of the Gutenberg-Richter parameter b, which shows a significant difference between the Groningen field (b=1.0) and the other fields (b=0.6). Other parameters of importance to hazard calculations did not change significantly. The maximum magnitude Mmax is stable at a value of 3.9. New attenuation relations for small shallow events are being constructed within the EC funded GEISER project. New hazard maps will be constructed when these relations are available. The hazard analysis assumes that the seismicity rate is a stationary process. We do see, however, from the cumulative seismic energy a change in character over time and it its effect will be taken into account in future studies.

The source mechanism of induced earthquakes provides important information on the reactivation of existing faults and constraints for geomechanical modelling of reservoirs. An overview of results for different gas fields in the Netherlands is given, including new results for the largest earthquakes in the Groningen field. The influence of a thick salt layer, acting as a high velocity layer on top of this reservoir, complicates the analysis and a special modelling study was needed to correct for this effect. Waveform modelling of the accelerometer recordings in the region is expected to improve depth estimates of the induced events.

This report concludes with a discussion of the present results and gives recommendations for improved monitoring and further research. The focus in monitoring is on expanding the real-time data exchange to include the accelerometer network and on microseismic monitoring, while for further research a continuation of the hazard research and waveform modelling is proposed.