eprintid: 1347 rev_number: 19 eprint_status: archive userid: 2 dir: disk0/00/00/13/47 datestamp: 2015-02-10 14:02:24 lastmod: 2017-02-08 12:21:40 status_changed: 2015-04-27 12:10:44 type: article metadata_visibility: show creators_name: Ouillon, Guy creators_name: Ducorbier, C. creators_name: Sornette, Didier creators_id: creators_id: creators_id: dsornette@ethz.ch corp_creators: Lithophyse, Nice, France. corp_creators: Geosciences Azur, Mecatec Unit, Valbonne, France. corp_creators: Department of Management, Technology and Economics, ETH Zurich, Zu¨rich, Switzerland. corp_creators: Department of Earth and Space Sciences and Institute of Geophysics and Planetary Physics, University of California, Los Angeles, California, USA. title: Automatic reconstruction of fault networks from seismicity catalogs: Three-dimensional optimal anisotropic dynamic clustering ispublished: pub subjects: MP2_3 divisions: EPOS-P full_text_status: none abstract: We propose a new pattern recognition method that is able to reconstruct the three- dimensional structure of the active part of a fault network using the spatial location of earthquakes. The method is a generalization of the so-called dynamic clustering (or k means) method, that partitions a set of data points into clusters, using a global minimization criterion of the variance of the hypocenters locations about their center of mass. The new method improves on the original k means method by taking into account the full spatial covariance tensor of each cluster in order to partition the data set into fault- like, anisotropic clusters. Given a catalog of seismic events, the output is the optimal set of plane segments that fits the spatial structure of the data. Each plane segment is fully characterized by its location, size, and orientation. The main tunable parameter is the accuracy of the earthquake locations, which fixes the resolution, i.e., the residual variance of the fit. The resolution determines the number of fault segments needed to describe the earthquake catalog: the better the resolution, the finer the structure of the reconstructed fault segments. The algorithm successfully reconstructs the fault segments of synthetic earthquake catalogs. Applied to the real catalog constituted of a subset of the aftershock sequence of the 28 June 1992 Landers earthquake in southern California, the reconstructed plane segments fully agree with faults already known on geological maps or with blind faults that appear quite obvious in longer-term catalogs. Future improvements of the method are discussed, as well as its potential use in the multiscale study of the inner structure of fault zones. date: 2008-01-17 date_type: published publication: Journal of Geophysical Research volume: 113 number: B1 publisher: American Geophysical Union id_number: doi:10.1029/2007JB005032 refereed: TRUE issn: 0148-0227 official_url: http://dx.doi.org/10.1029/2007JB005032 access_IS-EPOS: limited owner: Publisher citation: Ouillon, Guy and Ducorbier, C. and Sornette, Didier (2008) Automatic reconstruction of fault networks from seismicity catalogs: Three-dimensional optimal anisotropic dynamic clustering. Journal of Geophysical Research, 113 (B1). DOI: https://doi.org/10.1029/2007JB005032