baldassarri baronchelli barrat caglioti castellano cattuto complexity dattoli evolutionary_dynamics granular_media granular_media herrmann information_theory innovation_dynamics kreyon language_dynamics loreto mari mukherjee petri pietronero puglisi quantum_optics self_organization servedio statistical_physics techno_social_systems tria zapperi zippers
2008 |
Toris, Patrizia; Rubeis, Valerio De; Loreto, Vittorio; Pietronero, Luciano Space-time correlations of earthquakes Journal Article GEOPHYSICAL JOURNAL INTERNATIONAL, 173(3) , pp. 932–941, 2008. Abstract | Links | BibTeX | Tags: clustering, correlation, derubeis, loreto, memory, persistence, pietronero, statistical seismology, statistical_physics, tosi @article{b, title = {Space-time correlations of earthquakes}, author = {Patrizia Toris and Valerio De Rubeis and Vittorio Loreto and Luciano Pietronero}, url = {http://gji.oxfordjournals.org/content/173/3/932, http://www.scopus.com/inward/record.url?eid=2-s2.0-43449129493&partnerID=65&md5=781012778cff79bd9d3baf409098ca71 http://gateway.webofknowledge.com/gateway/Gateway.cgi?GWVersion=2&SrcApp=PARTNER_APP&SrcAuth=LinksAMR&KeyUT=000255708100014&DestLinkType=FullRecord&DestApp=ALL_WOS&UsrCustomerID=0c7ff228ccbaaa74236f48834a34396a}, year = {2008}, date = {2008-01-01}, journal = {GEOPHYSICAL JOURNAL INTERNATIONAL}, volume = {173(3)}, pages = {932--941}, publisher = {Blackwell Science Limited:PO Box 88, Oxford OX2 0NE United Kingdom:011 44 1865 776868, 011 44 1865 206038, EMAIL: journals.cs@blacksci.co.uk, INTERNET: http://www.blackwell-science.com, Fax: 011 44 1865 721205}, abstract = {Seismicity is a complex process featuring non-trivial space-time correlations in which several forms of scale invariance have been identified. A frequently used method to detect scale-invariant features is the correlation integral, which leads to the definition of a correlation dimension separately in space and time. In this paper, we generalize this method with the definition of a space-time combined correlation integral. This approach allows us to analyse medium-strong seismicity as a point process, without any distinction among main, after or background shocks. The analyses performed on the catalogue of worldwide seismicity and the corresponding reshuffled version strongly suggest that earthquakes of medium-large magnitude are time clustered inside specific space-time regions. On the basis of this feature, we recognize a space-time domain statistically characterized by sequences' behaviour and a domain of temporal randomness. Then, focusing on the spatial distribution of hypocentres, we find another domain confined to short distances and characterized by a relatively high degree of spatial correlation. This spatial domain slowly increases with time: we interpret this as the 'afterevent' zone representing the set of all subsequent events located very near (about 30 km) to each reference earthquake and embedded on specific seismogenic structures such as faults planes.}, keywords = {clustering, correlation, derubeis, loreto, memory, persistence, pietronero, statistical seismology, statistical_physics, tosi}, pubstate = {published}, tppubtype = {article} } Seismicity is a complex process featuring non-trivial space-time correlations in which several forms of scale invariance have been identified. A frequently used method to detect scale-invariant features is the correlation integral, which leads to the definition of a correlation dimension separately in space and time. In this paper, we generalize this method with the definition of a space-time combined correlation integral. This approach allows us to analyse medium-strong seismicity as a point process, without any distinction among main, after or background shocks. The analyses performed on the catalogue of worldwide seismicity and the corresponding reshuffled version strongly suggest that earthquakes of medium-large magnitude are time clustered inside specific space-time regions. On the basis of this feature, we recognize a space-time domain statistically characterized by sequences' behaviour and a domain of temporal randomness. Then, focusing on the spatial distribution of hypocentres, we find another domain confined to short distances and characterized by a relatively high degree of spatial correlation. This spatial domain slowly increases with time: we interpret this as the 'afterevent' zone representing the set of all subsequent events located very near (about 30 km) to each reference earthquake and embedded on specific seismogenic structures such as faults planes. |
Other subjects
