The Earth's mantle is almost two-thirds of the Earth by mass, and we think we know what it is, because we can measure its physical properties. There are some samples we think are brought up by volcanic eruptions. There are some samples even of the sea floor that seem to be like those we expect the mantle to be, but we have never sampled the mantle directly. We have never sampled it in situ, in its place. And that has been a dream of ocean drillers since before the Deep Sea Drilling Program, and it's one that is now a worthy goal and within reach. Achieving this goal will allow us to definitively test this fundamental understanding of the plate tectonic cycle in the oceans, and the whole process of generating oceanic crust, which so dominates the composition of the Earth's crust, the loss of heat from the Earth's interior, and the interaction of water with the upper part of the crust.
I said I wanted to talk about the dynamics of plate boundaries, and by dynamics I mean the physical dynamics and the chemical dynamics, and I want to focus for the purpose of confining the discussion to the convergent plate boundaries or subduction zones. You may know that all of the worlds largest earthquakes occur around the Pacific. Here in fig.35 are shown all the very large earthquakes of this century, and they occur in subduction zones. They occur at the interface between the subducting plate and the over-riding plate.
And, as already mentioned, these subduction zones are important as well because they transport a variety of materials back into the mantle (fig.36). Some of these materials are released and help generate magmas that give rise to volcanos. The chemical exchange and the partitioning of material that accompanies subduction, the release of water, and the generation of magmas is a very complicated process that ultimately gives rise to island chains, to continents, and perhaps to the entire continental crust of the Earth. So it's a very fundamental process in the differentiation of our planet about which we need to understand more.
There's been a lot of recent focus on subduction zones and in particular the shallow seismogenic part of the subduction zone, and I borrowed four slides from a colleague at the University of California at Santa Cruz, Casey Moore who is actively involved with Japanese colleagues in a seismic imaging experiment that is part of a larger project called SEIZE for Seismic Zone Experiment. SEIZE aims to use seismic techniques to image that part of the subduction zone at the interface between the subducting plate and the over-riding plate that generates earthquakes (fig.37).
There's a rather technical slide on the right (fig.38) that shows, as a function of depth, a variety of quantities, but I draw your attention to two in particular. Porosity drops from very high values to very low values over the top three or four kilometers. (For these sediments at the top of the subduction zone, you can think of porosity as water content by volume).
Also very important is temperature. And of course pressure is contributing to this drop in porosity, so as a consequence of pressure and temperature, it is thought that the composition of the sediments and the mechanical properties of the sediments change in important ways. In particular this drawing suggests that a particular chemical reaction involving dehydration - de-watering of a kind of clay to a different composition - is the key event that marks the mechanical transition from stable sliding to stick - slip, which is another term for seismogenic behaviour, at some depth between 3 and 5 km. This is a hypothesis, capable of being tested.
Again I draw your attention to the possible importance of water - in controlling, in this case, the seismic properties at the interface between the two plates, with this chemical reaction involving water perhaps playing a key role in the transition in properties.
There's a very exciting experiment going to go on later this year in the Nankai Trough (fig.39) by a consortium of three US institutions and six Japanese institutions to do three-dimensional seismic imaging.
