Geophysical characterization of co-seismic fractures due to liquefaction: case study following the Ml 5.9 magnitude earthquake that hit the Emilia on May 20, 2012 Nasser ABU ZEID 1* , Samuel Bignardi 1 , Giovanni Santarato 1 , Riccardo Caputo 1 1 Department of Earth Sciences, University of Ferrara Via Saragat, 1 Blocco B I-44122 FERRARA (FE), ITALY. ABSTRACT Following the main shock (ML 5.9) that hit North Italy on May 20, 2012 at 4:03:53 considerable coseismic site effects, represented by surface fractures, sand boils and lateral spreading, occurred. These were caused by the liquefaction of a saturated sand layer(s) located at shallow depths (< 15 m). The spatial distribution of these coseismic features follow the limits of the paleo-river beds which are widely present in the Padana Plain. In particular, the villages of Sant’Agostino, San Carlo, Mirabello and Vigarano Mainarda which are situated along the paleo-river bed of the Reno River, were severely affected and considerable damage to both cultural and industrial warehouses occurred while residential buildings suffered less damage. These occupy the western and south western portion of the Province of Ferrara, North Italy (Fig. 1). The observed seismic activities are related to the buried active front of the Romagna and Ferrara thrust belt. In this area, this folded belt represents the advanced northern rim of the Apennines mountains. It is overlain by a thick succession of Pliocene and Quaternary sediments forming a wedge-like shape of sediments underlying the Po Plain. The superficial geology of the test sites is composed mainly of alluvial deposits that has been deposited in different environments comprising: channel and proximal and distal levee, inter-fluvial, meander and swamp (Fig. 1). These sediments form also the main hydrogeological units overlying the bedrock which can be found at depths ranging between several hundreds to few kilometres. This work focuses on the application of geoelectromagnetic methods including: Electrical Resistivity Tomography (ERT), Induce Polarisation (IP) and Ground Penetrating Radar (GPR) techniques. for the characterisation of the subsurface continuity of these ruptures and possibly the geometry of the liquefied sand layer. The geophysical survey was carried out in three sites where ground ruptures are well visible on the ground surface and at the same time being located in the vicinity to residential and industrial building as well as in free field (1, 2 and 3 in Fig. 1). High resolution ERT/IP data acquired at the three test sites were accomplished along 2D profiles (1 to 2 meter electrode spacing) employing the ABEM SAS4000 and MPT-DAS1 georesistivity meters, while the GPR data was collected in site 3 using the IDS DAD unit (Italy) attached to 70 MHz sub-ecco monostatic antenna (Sweden). In one site (No. 2) a 50 m long and 5 meters depth paleoseismological trench was excavated in order to map these ruptures and to search for the eventual presence of features belonging to past earthquakes (Caputo et al., 2012). In the first site (1 in Fig. 1), a small-scale resistivity survey was carried out along a 46.5 m length profile that crossed the densely ruptured public park in the centre of San Carlo Village, North Italy. Here, the resistivity image has captured the depth extent of all the surface cracks (see squares, Fig. 2) which can be easily traced to 2-2.5 m depth. Again, a wide resistive anomaly crosses the conductive substratum (arrows) possibly associated with liquefaction dykes. The source liquefied layer is a sandy level encountered at 7 m depth in a borehole drilled in the same area. The observed irregular morphology of the low resistivity layer (ρ=5-10 Ω.m/arrows in Fig. 2), between 14 and 34 m provide strong clues about the occurrence of