The Strait of Messina: Seismotectonics and the source of the 1908 earthquake

Abstract

More than 100 years after the devasting Messina-Reggio Calabria earthquake (M = 7.1), the largest seismic event ever recorded in southern Europe in the instrumental epoch, its causative seismic source is still unknown, and the several rupture models proposed in the last decades are far from any shared solution. Data interpretation on a new dataset of sub-seafloor geophysical soundings with unprecedented resolution, relocated seismicity, and Vp model, together with morphotectonic investigations and inverse modelling of available levelling data, provide additional constraints on the deformation mechanisms and seismotectonics of the Strait of Messina area. Highresolution seismic lines in the offshore, along with displaced Quaternary marine terraces on land, point to active deformation along a previously unmapped ~34.5 km-long extensional fault. Spatial distribution of relocated earthquakes highlighted that a cut-off of the seismicity occurs within the crustal depth. The seismic discontinuity roughly delineates a foreland-dipping and low-angle discontinuity apparently confirming previous studies predicting low-angle seismogenic sources for the 1908 seismic event. However, according to the overburden stress and the attitude of the discontinuity, stress analysis suggests that a seismic slip is unlikely along it. This therefore weakens the hypothesis that a large earthquake may have nucleated along a low-angle discontinuity. Rather, aseismic creeping is instead expected since movement is allowed only by assuming a mechanical weakness of the plane. This mechanical behaviour is currently also supported by the large interseismic strain-rate recorded in the area. Both seismic tomography and crustal-to-subcrustal scale 3D-modelling strongly suggest a cause-and-effect relationship between slab retreat, mantle wedging, uplift in the upper plate block, and active extension in the Strait of Messina area. Lithospheric doming of the upper plate is here interpreted to be the main process controlling uplift in the Peloritani Mts. of Sicily and subsidence in the Strait of Messina region where deformation is mainly accommodated by the weak low-angle discontinuity. In this frame, an almost aseismical slip towards the foreland of the low-angle discontinuity is here accounted to produce stress perturbation in the area. Coulomb stress change modelling revealed that simulated normal slip on the foreland-dipping discontinuity can induce additional stress and promote failure in the overlying brittle faults. An excellent fit between calculated and observed subsidence is achieved by geodetic data inversion that resolved a normal slip on the low-angle discontinuity and a transtensional (slightly left-lateral) motion on the 34.5 km-long and previously unknown extensional fault. The fault-length along with the observed seafloor displacement make this tectonic structure as the most likely to have produced large earthquakes in the Strait of Messina area.