(FIG-3)
This difference simply tells us that subduction zones don't behave in the same way mechanically. At some subduction zones the plate boundary is coupled-when it slips, it produces a great earthquake. But in other subduction zones, the slip is more or less continuous. We often call it aseismic subduction. To explain this situation, I can show you this. This is the ratio of seismic slip to plate motion. So if the entire plate motion is taken up by seismic activity, this ratio η(eta) should be equal to 1, but if subduction is essentially aseismic, it should be 0. The blue zones are the places where basically the essential part of subduction is seismic, like the Chilean subduction zone, Peru, Bolivia, Colombia and Alaska subduction zones. However, in the Marianas for example and Tonga-Kermadec, subduction is essentially aseismic - plates are just subducting without producing earthquakes.
(FIG-4)
This distinction is extremely important. In particular in Japan, the northern part of Japan, Sanriku, only one quarter of subduction is taken up by seismic slip. On the other hand, along the Nankai Trough here, almost 100% of subduction is seismic. So even in a small area, the mode of subduction is entirely different. This difference must be caused by the difference in the subduction zone boundary, for which drilling projects can certainly produce very useful information.
First, I want to focus on the Sanriku coast. This is the northern part of Japan here, and in 1896 there was a very strange earthquake - we call it the Meiji Sanriku earthquake - which happened somewhere around here. And this of course produced some seismic shaking so people noticed ground shaking here, but it wasn't very strong and they were not particularly alarmed by this activity. But about 30 minutes later, one of the largest tsunamis in the Japanese history attacked the northern coast of Japan, killing more than 20,000 people. As you know, in terms of the seismic-aseismic kind of concept, this kind of earthquake can be viewed as a slow earthquake. There was slip, but the slip was relatively slow, so the earthquake didn't produce much seismic shaking, but in terms of deformation responsible for tsunami generation, it was a great earthquake.
(FIG-5)
This was a very curious event. Of course in 1896 the seismic instruments were not very good, so you might wonder how we could study this one, and of course we couldn't study it very well. But there is a kind of very interesting picture here. Photography hadn't been developed very well, so most of the data were presented in terms of paintings. This is a huge boulder which was probably carried on-shore by huge tsunami, and this suggests that there was large wave motion that brought this boulder on-shore.
OK up to here I have slides, and I'll give you one more evidence that this earthquake was a slow earthquake. [I'll probably go down there so that I can handle the overhead projector myself.]
There were some seismic data too, but I thought maybe this is a simpler evidence that you can understand. [You can turn off the slide projector.]
(FIG-6)
Well, this is the report on the Sanriku earthquake from one of the Japan Meteorological Agency Offices running around 1896. [Is this one running ? Can you see this one ?] The report on the right is for the main shock, and it says the intensity is subtle ? So people didn't feel it. And this is for one of the aftershocks which was a much smaller earthquake. [I always have this problem here. How do you do this one? OK.] It says ground motion is weak; actually weak is stronger than subtle. This really tells us that the main shock was a very subtle event in terms of seismic shaking, but the tsunami was really very large. So this is a very good description of the difference in the type of earthquakes.
