GNGTS 2021 - Atti del 39° Convegno Nazionale

GNGTS 2021 S essione 1.1 66 ANALYSIS OF EARTHQUAKES NEAR-FIELD BY INTEGRATING INSAR, SEISMOLOGICAL AND GROUND ACCELERATION DATA P. Petricca 1 , C. Bignami 2 , C. Doglioni 1,2 1 Sapienza Università di Roma - Rome - Italy 2 Istituto Nazionale di Geofisica e Vulcanologia - Rome - Italy 1. Introduction The epicentral area of earthquakes is usually assumed to represent the zone of largest shak- ing and damage, also called near-field, in contrast with the far-field where shaking is less severe due to the energy dissipation governed by the anelasticity parameters and the increase of the in- volved crustal volume. The near-field is conventionally attributed to an undefined radial distance from the epicenter and the source fault. Moreover, due to directivity of the rupture propagation (Calderoni et al ., 2017), sub-events migration (e.g., Yue and Lay, 2020) and site amplification ef- fects (e.g., Milana et al ., 2020), the strongest macroseismic intensity does not often rely on the epicenter or in a restricted area. In the near-field, vertical and horizontal stronger ground motion concur and the ratio of the maximum vertical spectral response to the horizontal can exceed one at very short periods (0.15 s), but falls off rapidly with period reaching a value of about 0.5 for long periods (Ambraseys and Douglas, 2003). Nowadays, the displacement and deformation in the epicentral area can be analyzed by a number of classic and new techniques such as strong- motion data, Global Navigation Satellite System (GNSS) and Interferometric Synthetic Aperture Radar (InSAR). Indeed, coseismic surface deformation can be obtained by dislocation theory of an elastic medium as shown by Okada (1985), where the affected area is the combination of the fault dislocation and its upward propagation through the medium, and it depends on the dip and depth of the assumed activated main fault plane. It is worth to note that not accurate estimation of fault parameters (especially in poorly instrumented areas) can provide not reliable information about the affected areas (Bignami et al ., 2021). Where present, GNSS stations provide excellent measuring of the horizontal and vertical components of the coseismic movements of the single sites (e.g., Wilkinson et al ., 2017). However, they cannot precisely cover the whole deformed area. Here, we discuss the powerful information resulting from InSAR data in describing the area affected by vertical deformation during an earthquake, providing a tool for better understand- ing the near-field and far-field distinction, based on the geological control of the seismological signatures. This analysis provides a tool for better seismic hazard assessment that should focus specifically on the near-field shaking. 2. Data and method Integrating InSAR images, seismological and accelerometric data we extract important infor- mation that characterizes an earthquake: the amount of deformed area, the corresponding in- volved crustal volume, the concentration of vertical and horizontal shaking. These values can be expressed as a function of the event magnitudes. We scrutinize a list of 32 earthquakes (Fig. 1) as a function of their tectonic setting, magnitude, dimension of the deformed area and ground shaking. The earthquakes selected in the present study have the following requirements: i) both the InSAR deformation data and the seismological data of the earthquakes are available; ii) SAR data are provided in the literature (e.g., as auxiliary material or public databases) or the figure de- scribing the deformation pattern in the corresponding paper is readable (i.e., deformation fringes are traceable) and correctly georeferenced. Coseismic fringes (or measured deformation) illumi- nated by SAR images were georeferenced in a geographic information system (QGis software,