Major Tsunami Risks from Splay Faulting

Abstract

One of the main lessons learned from the great 2004 Sumatra-Andaman earthquake was the fact that tsunami generation process due to large subduction earthquakes is rather complicated. Hence, modelling tsunamis by assuming a simple rupture on a megathrust may not account for actual variations of observed tsunami runups in the near field. From this viewpoint, the 2004 Indian Ocean tsunami was a milestone in tsunami research in that it clearly showed the effect of secondary tsunami sources on intensifying the near-field tsunami heights. The phenomena that are triggered by the main subduction earthquakes and locally contribute to tsunami in addition to the main slip on the subduction zone are known as secondary tsunami sources. The most important secondary sources are submarine landslides, whose effect was mainly evidenced during the 1992 Flores Island tsunami (Synolakis and Okal, 2005; Hidayat et al., 1995), and splay fault branching which was observed during some large subduction earthquakes such as the 1946 Nankai tsunami (Cummins and Kaneda, 2000), 1960 Chilean and 1964 Alaskan tsunamis (Plafker, 1972), and most recently during the 2004 Sumatra-Andaman earthquake and tsunami (Sibuet et al., 2007). Here, in this chapter we focus on splay faults which are known as one of the important secondary tsunami sources and were responsible for a large part of tsunami deaths during past tsunamis. Splay faults, sometimes known as imbricate faults, are steeply-dipping thrust faults which branch upward from the subduction zone to the seafloor. As splay faults often have steep dip angles, they are capable of producing large seafloor deformation which can significantly increase tsunami runup heights in the near-field. Figure 1 schematically shows a splay fault which branches from the plate boundary. As shown, an abrupt increase in seafloor uplift happens at the location of the splay fault. It is clear that this enhanced seafloor uplift will cause larger tsunami wave heights in the near field. With this background, it is clear that tsunami hazard assessment without taking into account the effect of possible splay fault branching due to large subduction zone earthquakes may result in underestimating the actual existing tsunami hazards. Hence, in this chapter we study the consequences of splay faulting on tsunami waves in the near-field. First, we make a review of some actual splay faulting cases. Then, the characteristics of splay fault branching from the main plate boundary will be discussed. In the next section, numerical modeling of tsunami will be performed to investigate the effect of splay faulting on tsunami wave heights in the near-field. Finally, we make some practical