Abstract

The lithosphere-asthenosphere boundary (LAB) is the seismic discontinuity with the negative velocity contrast in the upper mantle. The plate tectonic theory describes that the rigid lithosphere translate coherently over the ductile asthenosphere, and the LAB implies the decoupling between the two layers. The seismic velocity, viscosity, electrical resistivity and other physical parameters change with the depths across the boundary. The LAB has become a focus of seismology studies for its significance in understanding the plate motions, the mantle convection and the lithospheric evolution processes. Seismic detections on the LAB in subduction zone areas are helpful to understand the interactions between the lithosphere and asthenosphere layers and the geodynamic processes related with the slab subductions. For the dense permanent networks and transportable arrays of the USArray, the abundant waveforms of the deep earthquake events beneath the Izu-Bonin area can be recorded and retrieved. In this study, the vertical broadband waveforms are collected from three earthquake events occurring from 2002 to 2014 with the focal depths of 400–600 km. The source parameters are taken from the Preliminary Determinations of Epicenters catalog of US Geological Survey, and the focal fault solutions are taken from the global Centroid-Moment-Tensor project. The seismic waveforms are manually selected with two criteria: (1) simple sP arrivals with clear peaks, and (2) high signal-to-noise ratios and without abnormal spikes. In order to enhance the weak unknown precursors, the linear slant stack method, which is the special form of the N -th root slant stack method and phase-weighted stack method, is applied to scan the differential slowness and stack the seismograms in the domain of differential travel-time and differential slowness. The selected waveform data is processed to obtain the vespagrams and the stacked waveforms. The sP precursors reflected on the LAB ( s LAB P ), which have the opposite polarities with the amplitude ratios of 0.17–0.21 relative to the sP phases, are successfully extracted in our results. To reduce the possible effects of the crustal structures on the propagations of sP phases, the one-dimensional modified velocity model (IASP91-IB) is established by replacing the crust part of the IASP91 model with the CRUST1.0 model. On the basis of the modified model, we obtain the distributions for six reflected points of the s LAB P phases near the source area. Our results reveal that the LAB depths range between 58 and 65 km beneath the Izu-Bonin Arc, with the average depth of 62 km and the small topography of 7 km. Based on the half-space cooling model for the oceanic lithosphere, the lithosphere beneath the Izu-Bonin Arc is about located at the isotherm of 950°C, and the lithosphere beneath the West Philippine Basin (WPB) and Parece Vela Basin (PVB) is about located at the isotherm of 1100°C. Compared with the results of the tectonic stable areas (WPB and PVB) in the Philippine Sea, the lithosphere beneath the Izu-Bonin Arc shows the obvious thinning phenomenon. The numerical modeling experiments have revealed the strong erosions of the convecting asthenosphere in the mantle wedge, resulting in the destruction of the craton. We infer that the lithospheric thinning beneath the Izu-Bonin Arc is closely related with the partial melting, which is caused by the volatiles continuously released from the subducted western Pacific slab, and the strong erosions of the small-scale mantle convections in the back-arc mantle wedge.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call