GNGTS 2017 - 36° Convegno Nazionale

276 GNGTS 2017 S essione 2.1 Italian Macroseismic Intensity Attenuation Model as a function of Mw and distance A.A. Gómez Capera 1 , M. Santulin 3 , M. D’Amico 1 , V. D’Amico 2 , M. Locati 1 , L. Luzi 1 , M. Massa 1 , C. Meletti 2 1 Istituto Nazionale di Geofisica e Vulcanologia, sezione di Milano, Italy 2 Istituto Nazionale di Geofisica e Vulcanologia, sezione di Pisa, Italy 3 Istituto Nazionale di Geofisica e Vulcanologia, sezione di Milano c/o OGS, Trieste, Italy Introduction. Following the goal of CPS ( Centro di Pericolosità Sismica , Center for Seismic Hazard) that is to develop a new Italian seismic hazard map, we propose an updated macroseismic intensity attenuation model for the whole Italian territory and Sicily (without Mount Etna area) that can be applied for computing PSHA in terms of macroseismic intensity. The adopted methodological criterion is focused on developing a new macroseismic intensity attenuation model calibrated as a function of the moment magnitude (Mw) rather than the epicentral intensity (I 0 ). The standard deviation, related to a normal distribution of residuals macroseismic data, is given and can be used to model the uncertainty of ground shaking in hazard studies. Current knowledge status. Regarding PSHA studies in terms of macroseismic intensity, the most recent macroseismic intensity attenuation models for the whole Italian territory are proposed in literature by Pasolini et al . (2008) and Gomez Capera et al . (2010). Both models were developed in the frame of a seismological project (agreement INGV-DPC 2004-2006/ Project S1), and both using the same datasets (DBMI04, Stucchi et al ., 2007 and CPTI04, CPTI Working Group, 2004) adopted for assessing the national reference seismic hazard model (MPS04, MPS Working Group, 2004). The two attenuation models adopt different functional forms to evaluate the intensity attenuation with distance and source parameters. Moreover, they are focalized on different spatial and source size range and consequently on higher and lower intensities. A more recent, but not published, attenuation intensity model was calibrated for Italy in the frame of the module NA4 “Distributed Archive of Historical Earthquake Data” of the EC FP6 NERIES Project (Gomez Capera et al ., 2009) and CPTI11 (Rovida et al ., 2011; Gomez Capera et al. , 2008) for determining earthquake parameters from macroseismic data. The relationship developed by NA4 was disregarded as it could be improved by using a larger dataset. The prediction equation proposed in this work is based on the most recent and publicly available datasets, compiled in the framework of the new Italian hazard map, the version 2015 of the Italian Macroseismic Database (DBMI15, Locati et al ., 2016) and the 2015 version of Parametric Catalogue of Italian Earthquakes (CPTI15, Rovida et al ., 2016), and this allowed to developed new studies. Input data. A dataset of 16,261 Macroseismic Data Points (MDPs) related to a set of 118 earthquakes occurred in the time window 1908-2013 with reliable instrumental location and moment magnitudes (Mw) was used to calibrate the coefficients of the macroseismic intensity attenuation relation. The so-called “input calibration dataset” is the input data carefully selected for addressing a series of general conditions: 1) earthquake in CPTI15 covering the largest possible magnitude range and 2) spatial distribution, and having good quality 3) instrumental Mw, 4) epicentral locations, and 5) set of MDPs from DBMI15. An additional subset of events called “validation dataset”, was then selected with the same criteria, and was used to test the attenuation model developed in this study. Input data selection criteria. In order to derivate a macroseismic intensity attenuation model a careful selection of set of MDPs and earthquakes was carried out considering diverse criteria (Gomez Capera, 2006; Gomez Capera et al ., 2010): - earthquakes relative to the volcanic area of Mount Etna were removed because the attenuation pattern in this zone is different from active crustal regions (Ciccotti et al ., 2000);