Nature of the backthrust and the frontal thrust using full waveform inversion of ultra-long offset streamer data, offshore central Sumatra

The Sumatra subduction zone has hosted three great in the last decade. Although the 2004 great earthquake produced a devastating tsunami, the other two did not. On the other hand, the Mw 7.8 2010 earthquake produced a large tsunami with run-up heights up to 8 m. Addressing the problems of tsunameginic behavior of the earthquakes is fondamental as a 500 km long segment, the Mentawai segment, is likely to produce a great earthquake in the foreseeable future. In this thesis, I utilize the 15-km long offset streamer data recorded in 2009 by CGG in the 2007 and 2010 earthquake rupture zones, and subject them to advanced analyses techniques such as downward continuation, travel time tomography followed by seismic full waveform inversion, providing information on tens of meter scale. Above the backthrust, the P-wave velocity decreases from 4.9 km/s to 2.4 km/s in the compacted accretionary wedge sediments. Such a large decrease in the P-wave velocity could be due to 17- 40% of fluid or 2% to 13% of gas. The presence of bottom simulating reflections above supports the presence of gas. These results combined with the enhanced reflectivity of the backthrust in 2007 earthquake suggest that the gas might have the mantle wedge origin. At the subduction front, the results show low P-wave velocity associated with normal bend-faults faults, and a 200 m thick low velocity layer above the oceanic crust, where the porosity could be up to 70%. The overlaying 4 km/s sediments could be due to phase change from smectite to illite. This layer would set a suitable décollement surface for the frontal rupture yielding to powerful tsunami generation.