Eclogitization of the Subducted Oceanic Crust and Its Implications for the Mechanism of Slow Earthquakes

Abstract

The generating mechanism and process of slow earthquakes can help us to better understand the seismogenic process and the petrological evolution of the subduction system, but they are still not very clear. In this work we present robust P and S wave tomography and Poisson's ratio images of the subducting Philippine Sea Plate beneath the Kii peninsula in Southwest Japan. Our results clearly reveal the spatial extent and variation of a low-velocity and high Poisson's ratio layer which is interpreted as the remnant of the subducted oceanic crust. The low-velocity layer disappears at depths >50 km, which is attributed to crustal eclogitization and consumption of fluids. The crustal eclogitization and destruction of the impermeable seal play a key role in the generation of slow earthquakes. The Moho depth of the overlying plate is an important factor affecting the depth range of slow earthquakes in warm subduction zones due to the transition of interface permeability from low to high there. The possible mechanism of the deep slow earthquakes is the dehydrated oceanic crustal rupture and shear slip at the transition zone in response to the crustal eclogitization and the temporal stress/strain field. A potential cause of the slow event gap existing beneath easternmost Shikoku and the Kii channel is the premature rupture of the subducted oceanic crust due to the large tensional force. Plain Language Summary Various phenomena of slow earthquakes have been observed seismologically and geodetically in recent years around the circum-Pacific subduction zones. It has been generally considered that slow earthquakes could be used as a proxy to monitor aseismic slip and precursors to regular earthquakes, and they may play a role in stress transfer from slow to huge earthquakes. However, the generating mechanism and process of slow earthquakes are still not very clear. This study focuses on the process of the subducted oceanic crustal eclogitization and its relationship with the generating mechanism of the slow earthquakes. We suggest that the generating mechanism of the slow earthquakes is the dehydrated oceanic crustal rupture at the transition zone in response to the prograde eclogitization and the stress/strain field. The present results shed new light on the generating mechanism and process of slow earthquakes.

Age of the Subducting Philippine Sea Slab and Mechanism of Low‐Frequency Earthquakes

“…Our results show that the LFE gap is located at the youngest part (~15 Myr) of the PHS slab, suggesting that the slab's lithosphere age may also affect the formation of the LFE gap. According to the results of the present study and previous works (Liu & Zhao, ; Nakajima & Hasegawa, ; Seno & Yamasaki, ; X. Wang et al, ; Z. W. Wang et al, ; Wang et al, ; Yu et al, ), we deem that the LFE gap is caused by joint effects of several factors, including the young slab age, high temperature, low fluid content, and high permeability of the overlying Eurasian plate.…”

“…Previous studies have suggested that the LFE generation depends on the permeability of the overlying plate (Gao & Wang, ; Nakajima & Hasegawa, ) and the metamorphic dehydration of the subducting sediments and oceanic crust (Becken et al, ; Kodaira et al, ; Peng & Gomberg, ; Shelly et al, ; X. Wang et al, ). The location of slab dehydration reactions depends primarily on the temperature and pressure conditions at the top of the subducting slab and hence on the detailed thermal structure of subduction zones (van Keken et al, ).…”

High‐Resolution Ps Receiver Function Imaging of the Crust and Mantle Lithosphere Beneath Southern New England and Tectonic Implications

“…Given its depth, continuity, and gently dipping geometry, we hypothesize that the converter corresponds to the Moho of a relict slab from a past subduction event; the deeper portion of the slab likely broke off and sank into the deeper mantle, while the shallower part remained stuck in the stiff, highly viscous lithospheric mantle. Since oceanic crust generally starts to undergo transformation to eclogite at >50 km depth (e.g., Wang et al., 2017), and eclogite has seismic velocities more similar to mantle peridotite (e.g., Worthington et al., 2013), the Ps converted signal from an oceanic Moho at depth is expected to be much weaker in amplitude than that from the continental Moho, as observed. It is notable that while we seem to observe the oceanic Moho, we do not observe a corresponding NVG interface above it that represents the top of the oceanic crust.…”

Cascadia low frequency earthquakes at the base of an overpressured subduction shear zone

“…Locations of the inter-plate boundary at depths >20 km have also been proposed based on the migration of teleseismic receiver functions 16 , and in several subduction zones the underlying low-velocity zones (LVZs) have been interpreted as the igneous oceanic crust 17 – 20 , including internal stratification 21 , with the downdip extent of the LVZ upper boundary consistent with thermal-petrological modelling of the basalt-eclogite transition 17 . In Cascadia, detection of an approximately coincident 3–5 km thick, landward dipping zone with anomalously high Poisson’s ratio of 0.3–0.4 has led to the proposal that the upper oceanic crust is maintained at near-lithostatic pore pressure by a low-permeability inter-plate boundary immediately above the LVZ 21 , 22 , and a similar interpretation has been made in Japan 23 , 24 . In Cascadia, however, interpretations of the inter-plate boundary using receiver functions derived from teleseismic data are systematically shallower than interpretations based on active source, e.g., normal-incidence reflection and wide-angle, seismic surveys 9 , 10 .…”