The 1908 Messina Tsunami: the most likely sources from marine data
Abstract
Dangerous, if not devastating, tsunamis can be sourced from submarine mass failures. Research over the past decade has resulted in an improved recognition of this hazard. Numerous anomalous tsunamis may have a submarine landslide component, and these are identified by runups that, to some degree, are too large in relation to their proposed earthquake source. They include Sanriku, 1896, Messina, 1908, Puerto Rico 1918, Indian Ocean, 1945, Alaska, 1946 and Flores Island, 1982. Re-evaluation of these events, most notably using improved sea bed mapping technologies, has improved our understanding of their source, but not always with a proven result. Thus we are still challenged in our ability to differentiate historical tsunamis that may be sourced from either earthquakes or submarine landslides. Qualification of tsunami sources is critical if we are to validate submarine landslides as an important hazard. The 1908 Messina tsunami is one such enigmatic event, but (for its time) unique in that there is a remarkable catalogue of tsunami inundation and runup data that provides a control on the actual tsunami source(s). The centenary of the 1908 Messina earthquake and tsunami has come and gone, yet over 100 years after this event took place, no complete model of the tsunami source mechanism(s) has yet been produced. Previous analyses of the 1908 tsunami have looked at a wide variety of earthquake rupture mechanisms, and several possible mass failure events. Whereas progress has been made in understanding the event, there still remains some doubt as to the validity of the geological interpretations and, therefore, subsequent tsunami modeling results. Using marine geology data, notably multibeam bathymetry, we present a new unified model of the tsunami and its sources, providing a complete simulation of the event (Watts and Tappin, 2010). The earthquake rupture mechanism is by faulting along stepovers in the Messina Strait. Individual fault segments have NNW-SSE trending strike, and rupture proceeds from south to north trending NNE-SSW along the Messina Strait. These two facts explain the different seismic wave analyses and fault orientations proposed by various researchers. An ab initio mechanical model of fault segments reproduces both seismic inversion solutions and sea floor topography. We also examine several mass failures and discriminate between older events and those that probably contributed to the tsunami. We provide a thorough and rigorous examination of their precise locations, dimensions, displacements, and deformations. We reproduce mass failure motion with a state of the art mechanical model.
- Publication:
-
AGU Fall Meeting Abstracts
- Pub Date:
- December 2010
- Bibcode:
- 2010AGUFMOS13E1299T
- Keywords:
-
- 3022 MARINE GEOLOGY AND GEOPHYSICS / Marine sediments: processes and transport;
- 3045 MARINE GEOLOGY AND GEOPHYSICS / Seafloor morphology;
- geology;
- and geophysics;
- 3070 MARINE GEOLOGY AND GEOPHYSICS / Submarine landslides;
- 7215 SEISMOLOGY / Earthquake source observations