Three-dimensional imaging of active structures combining seismological and geological data: the 1980 Irpina earthquake (Ms 6.9)

The connection between fault-zone architecture and seismic behaviour is central to study the seismic potential of an active area. On November 23th, 1980, a Ms 6.9 earthquake struck the Irpinia area in Southern Italy. After the earthquake, several investigations produced divergent seismotectonic models based on different techniques, among others field geology observations, aftershock distribution, geodetic analysis, teleseismic observations, etc.. We re-analyzed all available data resulting from the aforementioned techniques with the aim to understand and reduce such divergence, and to achieve fuller understanding of the 1980 Irpinia sequence and to devise an accurate seismotectonic model. We used different data sets available in literature to create a tree-dimensional database using Midland Valley Software package (Move), and it is aimed in particular at integrating surface and subsurface data. The Database contains: 1) surface geological information extracted by available geological maps; 2) the seismicity from 1980 to 2011 (Italian Seismic Bulletin, INGV) and a detail on the local seismicity from 2007 to 2012 (Irpinia Seismic Network, ISNet bulletin of AMRA scarl); 3) all the seismic sections publicly available from the Videpi database and the crustal section Crop04, as well as deep well data; 4) seismogenic sources proposed by different authors. The combination of these data allowed us to constrain the orientation of the normal fault system responsible of the 1980 Irpinia events. We confirm that three different fault segments are the causative faults of the three main events recorded during the sequence, though we propose a new orientation of these structures. To test our model, we explore how these sub-events interact, cluster, propagate, and inhibit each other, through Coulomb static stress change analysis. We then show how our model capture much of the observed features of the seismicity of the area: both about the 1980 sequence and to recent (2007- 2012) background seismicity well recorded by local seismic network (ISNet). We conclude that the cross-correlation between geological and seismological data allow to model in detail seismogenic structures, and to achieve fuller understanding of seismic sequence in complex geological areas.