Stochastic estimation of French annual mainshock frequencies

Corentin Gouache and Fran\c cois Bonneau and Pierre Tinard. ( 2019 )
in: 2019 Ring Meeting, ASGA

Abstract

Seismic catalogues are composed of both independent events, i.e. main shocks, and spatiotemporal dependent events, i.e. fore- and aftershocks. Independent events present a time distribution that follows a Poisson law. Correlated events present a particular organization that can be described by scaling laws through time, space and magnitude. Seismic hazard models like ETAS (Epidemic-Type AfterShocks) generate time series of independent earthquakes thanks to Poisson law before merging series of dependent earthquakes through laws that characterize aftershocks activity (e.g. Omori-Utsu law). In this spirit, we propose to evaluate the annual main shock frequency calculated from inter event times. The original method was proposed in the beginning of the 21th century and is assumed to calculate main shocks proportion through a seismic catalogue. We made two assumptions: (i) high magnitude earthquakes are exhaustively recorded in the dataset and (ii) a magnitude threshold above which events are considered as main shocks is defined. We validate the method in the light of reference works. We calculate the annual main shock frequencies of mainland France from the historical - instrumental catalogue FCAT17. We also analyse Lesser Antilles seismicity thanks to earthquakes recorded by the USGS. We take into account magnitude uncertainties thanks to a Monte Carlo scheme. Sensitivity tests are proceeded to explore the stability of this method on French data.

Download / Links

BibTeX Reference

@INPROCEEDINGS{GouacheRM2019,
    author = { Gouache, Corentin and Bonneau, Fran\c cois and Tinard, Pierre },
     title = { Stochastic estimation of French annual mainshock frequencies },
 booktitle = { 2019 Ring Meeting },
      year = { 2019 },
 publisher = { ASGA },
  abstract = { Seismic catalogues are composed of both independent events, i.e. main shocks, and spatiotemporal dependent events, i.e. fore- and aftershocks. Independent events present a time distribution that follows a Poisson law. Correlated events present a particular organization that can be described by scaling laws through time, space and magnitude. Seismic hazard models like ETAS (Epidemic-Type AfterShocks) generate time series of independent earthquakes thanks to Poisson law before merging series of dependent earthquakes through laws that characterize aftershocks activity (e.g. Omori-Utsu law). In this spirit, we propose to evaluate the annual main shock frequency calculated from inter event times. The original method was proposed in the beginning of the 21th century and is assumed to calculate main shocks proportion through a seismic catalogue. We made two assumptions: (i) high magnitude earthquakes are exhaustively recorded in the dataset and (ii) a magnitude threshold above which events are considered as main shocks is defined. We validate the method in the light of reference works. We calculate the annual main shock frequencies of mainland France from the historical - instrumental catalogue FCAT17. We also analyse Lesser Antilles seismicity thanks to earthquakes recorded by the USGS. We take into account magnitude uncertainties thanks to a Monte Carlo scheme. Sensitivity tests are proceeded to explore the stability of this method on French data. }
}