1365-246X/asset/GJI_centre.gif?v=1&s=16d020b89eb31018f67640eaeffa549771f20e71)
1365-246X/asset/GJI_right.gif?v=1&s=43d3e3120b4ec2a9aca8a69c69c19714269fc4cf)
Probabilistic seismic hazard estimation in low-seismicity regions considering non-Poissonian seismic occurrence
Article first published online: 8 FEB 2006
DOI: 10.1111/j.1365-246X.2006.02863.x
Additional Information
How to Cite
Beauval, C., Hainzl, S. and Scherbaum, F. (2006), Probabilistic seismic hazard estimation in low-seismicity regions considering non-Poissonian seismic occurrence. Geophysical Journal International, 164: 543–550. doi: 10.1111/j.1365-246X.2006.02863.x
Publication History
- Issue published online: 8 FEB 2006
- Article first published online: 8 FEB 2006
- Accepted 2005 September 27. Received 2005 July 4; in original form 2005 May 7
- Abstract
- Article
- References
- Cited By
Keywords:
- aftershocks;
- low-seismicity regions;
- probabilistic hazard;
- seismic modelling;
- statistical methods;
- synthetic-earthquake catalogues
SUMMARY
In low-seismicity regions, such as France or Germany, the estimation of probabilistic seismic hazard must cope with the difficult identification of active faults and with the low amount of seismic data available. Since the probabilistic hazard method was initiated, most studies assume a Poissonian occurrence of earthquakes. Here we propose a method that enables the inclusion of time and space dependences between earthquakes into the probabilistic estimation of hazard. Combining the seismicity model Epidemic Type Aftershocks-Sequence (ETAS) with a Monte Carlo technique, aftershocks are naturally accounted for in the hazard determination. The method is applied to the Pyrenees region in Southern France. The impact on hazard of declustering and of the usual assumption that earthquakes occur according to a Poisson process is quantified, showing that aftershocks contribute on average less than 5 per cent to the probabilistic hazard, with an upper bound around 18 per cent.