In situ Seismic Anisotropy in the Source Region of Global Deep Earthquakes

University Of Houston, Houston TX

Investigators

Abstract

Earthquakes deeper than ~60km occur at subduction zones in the downgoing lithospheres known as slabs. Depending on the depth range at which they occur, they are known as either intermediate-depth, which can be destructive to human society, or deep-focus earthquakes. It is very challenging to explain the occurrence of these events since normal mantle rocks become ductile and this, in principle, should inhibit the occurrence of earthquakes. Yet more than 20% of the earthquakes worldwide occur at 60 km or larger depths. Moreover, it has been observed that a large proportion of deep earthquakes radiate seismic waves differently than that expected from the double-couple mechanism of shallow earthquakes. These lines of evidence often lead to believe that the physics of deep earthquakes are fundamentally different from that of shallow earthquakes. The research goal of this study is to understand the physics of deep earthquakes and their seismic radiation by taking into account the seismic structure of the subducting oceanic slab where they occur. The focus of this project is to elucidate why many deep earthquakes show a large non-double-couple-components in the results of moment tensor source analysis. The source focal mechanism solution is often used to infer important earthquake source mechanics. Specifically, this study tests the following well-formulated hypothesis for a particular subduction zone: could the presence of a uniform seismic anisotropic layer in a small volume inside the slab explain the non-double-couple component observed in all earthquakes contained in that region without the need of invoking anomalous earthquake source processes? Such anisotropy is expected since significant shearing should occur at the plate interface due to slab sinking. First, by using forward modeling, this study validates the approach of using earthquake radiation patterns to directly invert for in situ anisotropy around the earthquakes. Then 22 deep-earthquake groups in six global subduction zones: Tonga, Mariana, Japan, Kurile, South America, and Java subduction zones are analyzed. Second, for each of the 22 groups, the researchers invert for a single common anisotropy. Preliminary results show that the existence of the in situ laminated anisotropic fabric (i.e., tilted transverse isotropy) around the earthquakes could produce the observed non-double couple component in earthquake radiation pattern using earthquakes with conventional double-couple mechanisms due to faulting. In all subducting slabs, the inverted anisotropic fabric is parallel to the subducting slab interfaces, regardless of depth. The typical inverted shear-wave anisotropy is large, ~25%, in slabs worldwide. The consistency of the inferred anisotropy with respect to the slab geometry and large anisotropy values across many subduction zones fill an important knowledge gap regarding the slab internal structure produced by stress and chemical metamorphism. These new results can provide a critical synergistic link for a wide range of deep slab processes in understanding slab dynamics and plate tectonics.