GNGTS 2013 - Atti del 32° Convegno Nazionale

The Emilia area. In this case study, we processed COSMO-SkyMed ascending and descending data to monitor the deformation occurred in the areas affected by the May, 2012 earthquakes. We also considered an Envisat-ASAR descending dataset spanning from September, 1992 to June, 1999. The aim was to investigate the possible presence of transient deformation signals occurring before any possible seismic activity in the area, i.e. before the mainshock and any following aftershocks. Thus the analysis included the measurement of the deformation occurring in the post-seismic period of the 2012 sequence. Ground deformation occurred during the May-June, 2012 seismic sequence. The detection of possible transients associated to the largest aftershocks of the sequence (those occurred up to June 4 th 2012) was not possible using time-series InSAR, since there were not enough COSMO images acquired before May 20 th (Salvi et al. , 2012). Using the classical two-pass InSAR technique on COSMO and Radarsat-1 data (Pezzo et al. , 2013), the coseismic ground displacement due to the main shocks of the sequence from May 20 th to June 4 th was mapped. InSAR, GPS, geological and seismological data were then used to constrain a source model for the May 20 th and May 29 th events, which suggested that the activated structures were the Ferrara and Mirandola thrusts, respectively (Pezzo et al. , 2013). The results of the coseismic deformation analysis, source modeling, and interpretation allowed to estimate a 6-bar stress increase caused by the May 20 th mainshock on the fault of the May, 29 th event, which suggests a possible triggering (or clock-advancing) of the dislocation on the Mirandola thrust by the Ferrara thrust dislocation. The InSAR analyses revealed also that during the 9-day period separating the two largest earthquakes, a ~7-8 cm aseismic deformation transient occurred in the area between the two dislocations. Unfortunately, as mentioned above, there were not enough pre-event images to investigate in detail this transient, but its closeness in space and time with the May 29 th aftershock might imply a cause-effect relationship. This pattern is not associated with any significant (Ml ≥ 5) aftershock or foreshock of the May 20 th event, and it can be modeled as slip occurring before the May, 29 th aftershock fault plane (Pezzo et al. , 2013) or it could represent a slowslip event with no seismic signature. Ground deformation occurred in the post-seismic phase. The temporal evolution of the ground deformation in this area could be investigated in detail only for the post-seismic time period following May 30 th . To this aim, we processed 2 ascending and 1 descending 9-month COSMO data sets covering the epicentral zone and the adjoining areas to the East and to the West (Tab. 1). We obtained the mean ground velocity maps and the displacement time series for the frames listed in Tab. 1, using the SBAS (B3 asc and B10h asc) and the Persistent Scatterer (for the B4 asc North, Hooper, 2007) techniques. The SRTM-1 DEM (30m) was used to remove the topographic component during the processing. Tab. 1 – The 3 COSMO-SkyMed frames processed. Orbit type Beam Number of images Number of pairs Temporal span Resolution Incidence angle Ascending B4North 18 17 11/6/2012 26/3/2013 180m 32.2 Ascending B10h 17 24 04/07/2012 27/4/2013 180m 41.3 Descending B3 13 26 23/5/2012 8/4/2013 180m 29.3 205 GNGTS 2013 S essione 1.2