Upper-mantle velocity structures beneath the Tibetan Plateau and surrounding areas inferred from triplicated P waveforms

Abstract

P‐wave waveforms in the distance range between 12° and 30° were analyzed to investigate upper‐mantle P velocity structures beneath the Tibetan Plateau and surrounding areas. The waveform data from 504 earthquakes with magnitudes larger than 5.0 between 1990 and 2005 that occurred within 30° from the center of the Plateau were modelled. We divided the study area into 6 regions and modeled upper‐mantle‐distance P waveforms with turning points beneath each region separately. The results show that the upper‐mantle P‐wave velocity structures beneath India, the Himalayas, and the Lhasa Terrane are similar and contain a high‐velocity lid about 250 km thick. The upper‐mantle velocities down to 200 km beneath the Qiangtang Terrane, the Tarim Basin, and especially the Songpan‐Garzê Terrane are lower than those in the south. The 410‐km discontinuity beneath these two terranes is elevated by about 20 km. High‐velocity anomalies are found in the transition zone below 500 km under the Lhasa and Qiangtang Terranes. The results suggest that the Tibetan Plateau was generated by thrusting of the Indian mantle lithosphere under the southern part of Tibet. Portions of the thickened Eurasian mantle lithosphere were delaminated; they are now sitting in the transition zone beneath southern Tibet and atop of the 410‐km discontinuity underneath northern Tibet.

Velocity Anomalies Around the Mantle Transition Zone Beneath the Qiangtang Terrane, Central Tibetan Plateau From Triplicated P Waveforms

“…Relative to the IASP91 model predictions, the observed differential times between branches AB and BC or between AB and CD increase with increasing epicentral distances, and the length of observed branch BC becomes larger for all the four events until CB' (Figure 4). These obser- −1 0 1 2 3 4 5 6 7 8 J S MTZ thickening area from triplicated wave (Chen and Tseng, 2007;Tseng and Chen, 2008;Chu et al, 2019).…”

“…Seismic triplications for the MTZ discontinuities include the seismic waves that propagate above the 410 (branch AB), reflect off the 410 (branch BC), in the MTZ (branch CD), reflect off the 660 (branch DE), and below the 660 (branch EF; Figure 2). As the ray paths for the triplicated waves near the source and receiver are very similar, the differential travel times and amplitude ratios between different branches of the triplicated waves are mainly related to the radial velocity variation near the 410 or the 660 (e.g., Chen & Tseng, 2007; Chu et al., 2019; Li et al., 2017).…”

“…The Indian continent lithosphere imaged as high velocity anomalies using seismic tomography has foundered atop (e.g., Huang & Zhao, 2006; Tilmann et al., 2003) or subducted into the MTZ (e.g., Chen et al., 2017; Li et al., 2008; Xiao et al., 2020). The thickness of the MTZ as defined by the depths of the 410 and 660 km discontinuities (hereafter referred to as the 410 and the 660) estimated from receiver function and seismic triplication analyses is either normal (e.g., Kind et al., 2002; Zhao et al., 2010; Zhao et al., 2011) or thickened (e.g., Chen & Tseng, 2007; Chu et al., 2019; Duan et al., 2017; Tseng & Chen, 2008) because of the plate subduction.…”

“…Seismic station distribution in the Qiangtang terrane is sparse because of the harsh natural environment. Seismic triplicated waveforms produced by the MTZ are sensitive to velocity variation around the MTZ halfway between the earthquakes and seismic stations and thus ideal to study the Qiangtang terrane (e.g., Chen & Tseng, 2007; Chu et al., 2019; Li et al., 2017; Li et al., 2020). We adopted the iterative grid search method (Li et al., 2020) to match the synthetics with observed P ‐wave triplicated waveforms recorded by the China Digital Seismograph Network (CDSN; Figure 1; Zheng et al., 2010).…”

A Partial Molten Low‐Velocity Layer Atop the Mantle Transition Zone Beneath the Western Junggar: Implication for the Formation of Subduction‐Induced Sub‐Slab Mantle Plume

“…The synthetic triplicated waveforms for the four earthquakes are simulated by the reflectivity method (Wang, 1999). We calculate the synthetics of the Event 20111227 for three representative 1‐D reference Earth's models: PREM (Dziewonski & Anderson, 1981), IASP91 (Kennett & Engdahl, 1991) and STW105 (Kustowski et al., 2008; Figure 3a), and three modified models based on the IASP91: TB for the Tarim basin to the south of our study region (Chu et al., 2019), LITH whose lithosphere velocity increases by 1.0%, and ASTHEN whose asthenosphere velocity decreases by −1.0% (Figure 3e). Their crustal structures are replaced by the CRUST1.0 model (Laske et al., 2013) because of deeper Moho in our study region.…”