The December 28, 1908, Messina Straits, southern Italy, earthquake: Waveform modeling of regional seismograms

Tho 1908 Messins Straits earthquake is one of the most catastrophic events in history. There were 60,000 to more than 100,000 deaths, and the cities o f Messina and Reggio Calabria, on the opposite sides of the straits, were a lmost completely destroyed. During the last decades only low magnitude even ts occurred in the area. The 1908 earthquake is then crucial for understand ing the mode of stress release in the area. We collected and digitized seve ral regional seismograms of this event recorded in central Europe with the aim of studying the source characteristics. In order to separate the path e ffects, we analyzed recent events with simple and known sources. Owing to t he small azimuth range spanned by the available stations, we could not dete rmine a fault plane solution for the Messina Straits event, nor discriminat e between the published focal mechanisms. However, the normal fault charact er of the rupture is confirmed. By inverting the historical P waveforms we derived source time functions. and obtained seismic moment of 5.38 (+/-2.16 ) x 10(19) N m (M-w=7.1). This is in good agreement with the results obtain ed by several authors from the inversion of the historical leveling data. T his value is confirmed by the SH wave modeling, which also allowed the asse ssment of the unilateral northward character of the rupture propagation alo ng an similar to 43 km fault. Finally, we applied it simple rupture model i n order to derive the slip distribution along the fault. The resulting func tion is in good correspondence with the geodetic inversions performed by Bo schi et al. [1989] and De Natale and Pingue [1991]. In particular, a maximu m slip of similar to4 is located in proximity of the center of the slipped area. The high dislocation patch beneath the Messina harbor, as depicted by De Natale and Pingue, is not confirmed by our analysis and is probably con nected to surface collapse of some of the bench marks.