Radial anisotropy beneath northeast T ibet, implications for lithosphere deformation at a restraining bend in the K unlun fault and its vicinity

Abstract

Three‐dimensional shear wave velocity and radial anisotropy models of the crust and upper mantle beneath the NE Tibetan plateau are constructed from new measurements of Love wave dispersions (20–77s) and previously obtained Rayleigh wave dispersions (20–87s) using a two‐plane‐wave method. The mid‐lower crust is characterized with positive anisotropy (VSH > VSV) with large strength beneath the Qinling and Qilian Mountains and small values beneath the Anyemaqen Mountain. The large positive anisotropy can be explained by horizontal alignment of anisotropic minerals in the mid‐lower crust due to crustal flow. The mantle lithosphere above 90 km is largely isotropic while weak positive anisotropy appears beneath 90 km, which probably marks the lithosphere‐asthenosphere boundary (LAB). A low shear wave velocity anomaly and relatively negative radial anisotropy are imaged in the entire lithosphere beneath the restraining bend in the eastern Kunlun fault, consistent with a weak lithosphere experiencing vertical thickening under horizontal compression. The asthenosphere at the restraining bend is characterized by significant low velocity and positive radial anisotropy, reflecting that the asthenosphere here is probably hotter, has more melts, and deforms more easily than the surrounding region. We propose that the lithosphere at the restraining bend was vertically thickened and subsequently delaminated locally, and induced asthenosphere upwelling. This model explains the observations of velocity and anisotropy anomalies in the lithosphere and asthenosphere as well as geological observations of rapid rock uplift at the restraining bend of the Kunlun fault.

Lithospheric SH Wave Velocity Structure Beneath the Northeastern Tibetan Plateau From Love Wave Tomography

“…Figures and show seven horizontal and four vertical sections of the SH wave velocity absolute values and perturbations from the 3‐D inversion at depths from the surface to 130 km, respectively. Our results in the QOB and the SGT largely agree with the previous SH wave velocity model obtained by Li et al () using the same method but different seismic data from an irregular array with a small number of stations. However, our current V SH model could provide more details on the QOB region with high station coverage compared to the previous V SH model (Li et al, ).…”

“…Our results in the QOB and the SGT largely agree with the previous SH wave velocity model obtained by Li et al () using the same method but different seismic data from an irregular array with a small number of stations. However, our current V SH model could provide more details on the QOB region with high station coverage compared to the previous V SH model (Li et al, ). Low velocities characterize the crust and uppermost mantle beneath the QOB and the SGT (Figures and ), which is consistent with the slow P wave velocity (Li et al, ; Lei & Zhao, ), slow SV wave velocity from earthquake surface wave tomography (Li et al, ) and from joint inversion of receiver functions and Rayleigh wave dispersions (Zheng et al, ).…”

“…A continuous low velocity anomaly in the QOB, an area between the Haiyuan fault and the east Kunlun fault, is generally imaged at periods from 20 to 100 s (Figures a–e). The observation of this low velocity anomaly is consistent with previous Rayleigh wave tomography models (Bao, Song, & Li, ; Li et al, ), while it is partially different from a relatively high velocity in the north QOB obtained from previous Rayleigh and Love wave tomography study (Li et al, , ), which has limited station coverage and crossing ray paths there. The Songpan‐Ganzi terrane (SGT) is characterized by low velocity anomaly at periods less than 50 s (Figures a and b) and relatively high velocity compared with the QOB at periods of 59–100 s. Another interesting feature is the sharp velocity contrast between high phase velocities in the Ordos block (OB) and low phase velocities in the Alxa block (AB) at intermediate‐long periods (>45 s) (Figures c–e).…”

“…Clark et al () suggested the presence of crustal flow in the NE Tibetan plateau via fitting the observed northeastward sloping topography by inducing an underlying lower crustal channel with a thickness of 15 km. Recent geophysical observations offer several lines of evidence, such as low velocity zone (Bao et al, ; Schoenbohm et al, ; Yang et al, ) and strong radial anisotropy (Duret et al, ; Li et al, ; Shapiro et al, ; Xie et al, ) in the middle and/or lower crust, in support of the existence of crustal flow in this region. However, the low and moderate Poisson's ratio (Wang et al, ; Xu et al, ) indicated a generally felsic crust in this region, which is contradictory to the mafic crustal composition predicted by the lower crustal flow model (Pan & Niu, ).…”

Complex, multilayered azimuthal anisotropy beneath Tibet: evidence for co-existing channel flow and pure-shear crustal thickening

“…Radial anisotropy (the difference between the vertically and horizontally polarized waves: V SV and V SH , respectively, in the case of S waves) is also well documented (e.g., Shapiro et al 2004;Huang et al 2010;Duret et al 2010;Guo et al 2012;Xie et al 2013;Li et al 2016). …”