Estimation of macroseismic scenarios in European countries (Comunicazione a convegno)

  • Estimation of macroseismic scenarios in European countries (Comunicazione a convegno) (literal)
  • 2012-01-01T00:00:00+01:00 (literal)
Alternative label
  • R. Rotondi (a) R. Sigbjornsson (b) E. Varini (a) C. Brambilla (a) (2012)
    Estimation of macroseismic scenarios in European countries
    in European Seismological Commission - 33rd General Assembly, Moscow, August 19-24, 2012
  • R. Rotondi (a) R. Sigbjornsson (b) E. Varini (a) C. Brambilla (a) (literal)
  • (literal)
  • (a) C.N.R. - Istituto di Matematica Applicata e Tecnologie Informatiche, sezione di Milano, Milano (b) Earthquake Engineering Research Centre, Iceland (literal)
  • Estimation of macroseismic scenarios in European countries (literal)
  • This study is being carried out in the framework of the EU research project \"Urban Disaster Prevention Strategies Using MAcroseismic Fields and FAult Sources\" (Grant Agreement n. 230301/2011/613486/SUB/A5), started in January 2012 and involving research teams from Italy, Iceland, Portugal and Spain. We present the analysis of macroseismic fields related to some geographic European areas prone to earthquakes with the aim to characterize the trends of the intensity decay in areas having different geological and geophysical features and to put them in comparison. We applied the following statistical framework. The site-intensity Is is modelled by the probabilistic binomial-beta model: given the epicentral intensity I0 and the parameter p, which depends on the site-epicentre distance and which is assumed to have a beta distribution (Zonno et al. (2009)). According to the Bayesian paradigm, a priori distribution is assigned to the parameter p; this distribution represents our belief on the attenuation behaviour of that area and it is deduced from suitable learning sets. The posterior beta distribution of p is then obtained updating its parameters in the light of the macroseismic data related to the area under investigation. At the end of this procedure, the posterior mean of p is available at each distance bin of a partition of the space around the epicentre. These values are then smoothed through an inverse power function so that, assuming that an earthquake of epicentral intensity I0 occurs, the value of the intensity Is at a given site may be forecast by the mode of the binomial distribution with parameters I0 and the smoothed value of p. The smallest sets of intensity values covering at least a given probability can be computed too. Scenarios were produced for several earthquakes and then compared to their observed macroseismic fields by some validation criteria. We applied the model under the assumption of isotropic decay of the intensity; when the know-how about the earthquakes and the results based on the isotropic assumption highlighted the need to enhance representation of finite sources and to address the problem of directivity effects we considered also the anisotropic model. In particular we considered the elliptical attenuation pattern. According to it, the first bin of the partition of the space around the epicentre is given by the ellipse having the fault length as major axis and the subsequent ones increase with assigned width. By carrying out suitable transformations, the anisotropic model can be traced back to the isotropic one. References Zonno G, Rotondi R, Brambilla C. Mining macroseismic fields to estimate the probability distribution of the intensity at site. BSSA (2009) 98-5, 2876-2892, doi: 10.1785/0120090042. (literal)
Prodotto di
Autore CNR

Incoming links:

Autore CNR di