Tichelaar, B.W., 1991, Depth of seismic coupling along subduction zones: University of Michigan, Ann Arbor, Ph.D. dissertation, 235 p., illust.
The largest earthquakes of our planet occur in narrow bands around the Earth. Plate boundaries of a specific tectonic regime, the thrust interface in subduction zones, generate most large earthquakes. While underthrusting at subduction zones can cause large earthquakes at shallow depths, it is always accommodated by aseismic creep below a certain depth. With respect to the seismogenic character, this transition from seismically coupled at shallow depths to uncoupled at deeper depths is the most important vertical segmentation of plate boundaries. I have mapped the depth extent of the coupled zone for many subduction zones around the Pacific Ocean. The method I use combines accurate depth estimates of smaller underthrusting earthquakes with aftershock regions of the great earthquakes. For the depth estimation, I use omnilinear inversion of body waves, with bootstrapping for the statistical uncertainty. The results indicate that, for most subduction zones that generate great earthquakes (Chile, Alaska, Aleutians, Kamchatka, Kuriles, and Japan), the seismically coupled zone is a shallow, superficial feature: below about 50 km, the plates are decoupled and great earthquakes do not occur. In one subduction zone where the subducting oceanic plate is young, and the plate convergence rate is small (Mexico), coupling is confined to only the upper 25 km of the interface. There are many mechanisms that may control the depth extent of the coupled zone. I found indications that the maximum depth of coupling coincides with critical isotherms, possibly in conjunction with crustal thickness of the upper plate. This implies that for temperatures higher than the critical temperature frictional sliding changes from unstable to stable (from coupled to uncoupled). Assuming a constant shear stress (10 MPa) on the plate interface, I find a critical isotherm of 260 +/- 80°C. On the other hand, assuming a constant coefficient of friction (0.036), I find two critical isotherms, 370 +/- 40°C and 500 +/- 40°C. An interpretation is that for temperatures higher than 370oC the interface between the continental crust and the oceanic lithosphere becomes uncoupled, while for temperatures above 500oC the interface between the continental mantle and the oceanic lithosphere becomes uncoupled. Critical isotherms have been proposed for other tectonic environments, and can explain my measurements of the depth of seismic coupling in subduction zones. However, considerable uncertainties in the temperature calculations, primarily due to the unknown shear stress and crustal radioactive heat production, make it difficult to 'prove' the importance of critical isotherms, and other mechanisms may be controlling the depth extent of coupling in subduction zones.
Theses and Dissertations