Implications for tsunami earthquakes

Traditional ideas

Generally a tsunami earthquake is defined as a seismic event whose tsunamis are much larger than expected from it surface wave magnitude Ms. One possible cause is a slow slip on the fault plane. If the slip is slow, the high frequency component of the fault slip is decreased compared with the long period component, then its moment magnitude becomes larger than Ms, resulting in larger tsunamis. However, there is a limitation for this model because very long period components can not excite tsunamis either. To conquer this difficulty, Fukao (1979), by analyzing two tsunami events at the southern Kuril trench, proposed a model in which a slow slip in the accretionary wedge and the sedimentary basin causes a large uplift at the lower trench slope. However, this model has also a difficulty to explain a large amplitude of tsunamis, because a dip angle of a ramp from the basal decollement is still small, i.e. such as 30 deg, and the low rigidity of the accretionary and sedimentary wedge cannot sustain a large enough stress to cause a large slip. The model which has been proposed by Dr. Kenji Satake (Geol Survery of Japan) and his colleges is geophysically impossible because it poduces an incredibly large stress drop at the shallowest portion of the decollement. In fact, in the case of the Chichi earthquake, the amount of the displacements at the accretionary wedge near the fault surface, revealed by GPS, is almost the same as the maximum slip of the asperity revealed by Yagi and Kikuchi (1999).

Implications of the anomalous uplift of the NW corner for tsunami generation

I explained the anomalous uplift at the NW corner by the ductile behavior of the sediment (See Why did a large uplift appear at the NW corner? ). The uplift of this corner amounting to 6-9 m is two to three times larger than the uplift revealed by GPS (See GPS data supports the inelastic behavior). If an earthquake, similar to the Chichi earthquake, occurs in the submarine area, and breaks through the accretionary wedge, it will push the sediment in the trench wedge, and cause a large uplift similar to the uplift at the NW corner. Such event would become a tsunami earthquake. It should be noted that the two S. Kuril events analyzed by Fukao (1979), the 1896 Sanriku event by Tanioka and Satake (1996), the 1946 Aleutian event by Johnson and Satake (1997), and the 1992 Nicaragua event by Satake (1994), all occurred in a very similar tectonic situation, i.e. close to the trench axis.

Implications for anomalous tsunamis

The fact that the deformation which does not follow the elastic dislocation theory can occur near the sedimentary wedge has large implications for analyses of anomalous tsunamis or tsunami earthquakes observed in the past and in the future. For example, the tsunami associated with the July 17, 1998 Papua New Guinea earthquake (Ms = 7.0) has been an enigma. Its strong tsunami needs a localized uplift with a large amplitude, but far from the coast if we believe the inhabitants' witness. Tanioka (1999) proposed a fault model, located in the accretionary prism of the New Guinea Trench system, with a high dip angle. But in his model the fault dips to the north!, opposite to the subduction direction. It cannot explain the far distance of the tsunami source area from the coast. If the sediment in the New Guinea Trench had large deformation associated with the slow faulting at the decollement dipping to the south, it apparently can cause a large localized uplift neat the trench wedge, and the tsunami source area becomes much far from the coast than Tanioka's fault.

Where do tsunami earthquakes occur?

References Fukao, Y., Tsunami earthquakes and subduction processes near deep-sea trenches, J. Geophys. Res., 84, 2303-2314, 1979. Johnson, J. M. and K. Satake, Estimation of seismic moment and slip distribution of the April 1, 1946, Aleutian tsunami earthquake, J. Geophys. Res., 102, 11765-11774, 1997. Satake, K., Mechanism of the 1992 Nicaragua tsunami earthquake, Geophys. Res. Lett., 21, 2519-2522, 1994. Tanioka, Y., and K. Satake, Fault parameters of the 1896 Sanriku tsunami earthquake estimated from tsunami numerical modeling, Geophys. Res. Lett., 23,1549-1552, 1996. Tanioka, Y., GRL, in press, 1999.