Surface waves are generally assumed to propagate along the great-circle paths in most of the surface wave tomographies. However, when lateral heterogeneity is strong, the off-great-circle propagation may occur and deteriorate surface wave tomography results based on the great-circle assumption. In this study, we used teleseismic waveforms recorded by the NorthEast China Extended SeiSmic Array in Northeast China to study the off-great-circle propagation of Rayleigh waves using the beamforming method and evaluated the influence of the off-great-circle propagation on event-based surface wave tomography. The results show that arrival angle anomalies generally increase with decreasing period. The arrival angle anomalies at 60 and 50 s periods are smaller than that at 40 and 30 s periods, which indicates that the off-great-circle propagation is relatively weak for longer periods. At 30 s period, the arrival angle anomalies are relatively larger and some of the measurements can exceed 20°, which represents a strong off-great-circle propagation effect. In some areas, the arrival angle anomalies of adjacent events differ significantly, which may be attributed to multipathing propagation of surface waves. To evaluate the influence of the off-great-circle propagation on event-based surface wave tomography, we used measured arrival angle anomalies to correct two-station phase velocity measurements, and performed azimuthal anisotropy tomography using dispersion data sets with and without the arrival angle correction. At longer periods, such as 60 s, the influence of the off-great-circle propagation on surface wave tomography is weak, even though the corrected model has better data fit than the uncorrected model. However, the influence of the off-great-circle propagation is non-negligible at short periods. The tomography results at 30 s period show that the differences in phase velocity, the strength of anisotropy and the fast direction can be as large as 1.5, 1.0 per cent and 30°, respectively. Furthermore, the corrected phase velocity is systematically lower than that without correction. This study illustrates the necessity of studying the off-great-circle propagation of surface waves to improve the accuracy of event-based surface wave tomography, especially for shorter periods.
Abstract Rayleigh waves are used to carry out tomographic inversion and obtain group velocity for a period range from 15 s to 120 s in East Asia (70°E–145°E, 10°N–55°N). The Tarim Basin is obviously recognizable for the period of 15s appearing as a low‐velocity region, however, it can not be seen for 16 s–33 s periods, and then appears as a high velocity zone for 36s–85s periods. This suggests that there exists a deep root under the Tarim Basin. With clear demarcation lines to the Indian plate on the south and to the Tarim Basin and Qaidam Basin on the north, Qinghai‐Xizang Plateau is the most prominent low velocity block for periods of 44 s–120 s, and the velocity in its central and western portion is lower than that in the east. In the vicinity of Chiang Mai, Thailand, we can see a low velocity block with a scale of 1 000 km, which may be created as the mass of Qinghai‐Xizang Plateau migrates southeast. The South‐North seismic belt appears to be a region with high group velocity gradient. The velocities in the east part are noticeably higher than that in the west part. The central of South China Sea, the Japan Sea, and Philippine Sea appear as oceanic crust. The image of Philippine Sea is completely consistent with the geological topography and the seismic zone. Around the Philippine Sea and the Japan Sea, there is a low velocity belt about 400km wide, which may be a zone with magmatic activities.
Abstract We propose an improved block modeling technique to describe complex 3‐D media. A geological medium is represented as an aggregate of geological blocks rather than layers, which have their own attributes such as shape, size, density and seismic wave velocity. The structure of a blocky model is Body → Block → Interface → Point. We construct interfaces in the model with triangles, which fit most complex media. For normal vectors change abruptly across the linked boundaries, normal vectors to the interfaces are everywhere smoothed by an approximate estimation and thus approximately C2 continuous to generate relatively accurate ray tracing. We present some methods for shooting ray tracing in 3‐D: triangle‐subdivision, triangle‐division and subtriangle methods. Our calculation indicates that the subtriangle method is most efficient. We present real cases for some complex models and ray tracing results on these models.
We imaged the azimuthal anisotropy of Rayleigh wave phase velocity (10–60 s) in northeast North China Craton using the teleseismic data recorded by a dense temporary array, and then inverted for the 3-D azimuthal anisotropy of the crust and uppermost mantle (20–110 km). The results reveal that the azimuthal anisotropy varies both horizontally and vertically. Obvious stratified azimuthal anisotropy is shown in the Central Orogenic Belt, where the fast direction is NE–SW to NNE–SSW in the depth range of 20–40 km and changes to NW–SE to NWW–SEE in the depth range of 60–110 km. In the depth range of 30–40 km, a prominent low velocity belt is shown on the southwest of Zhangjiakou-Penglai fault zone (ZPFZ) and the fast direction is subparallel to the strike of the low velocity belt. Distinct lateral variations of azimuthal anisotropy are clearly shown at 110 km. Our results provide new evidence for the existence of upwelling asthenosphere beneath the Datong volcano and support the assumption that ZPFZ may act as the channel of upwelling asthenosphere. Historical strong earthquakes (M ≥ 6.0) mainly occurred in the transition zone between low and high velocity anomalies in the upper and middle crust. The upwelling asthenosphere may prompt the generation of large earthquake.
Abstract It's the first time that interstation method for Rayleigh wave is used to study the 3‐D V S structure in the crust and upper mantle beneath western China and its adjacent area. More than 3000 interstation great circle paths were processed to determine the phase velocity of the fundamental mode of Rayleigh wave, and finally 110 paths of high quality dispersion data were selected which show good spatial coverage in western China and its neighboring regions. Phase velocity maps from 15 s to 120 s were obtained by Tarantola's probability method. The shear wave velocity structure in the crust and upper mantle on 2° × 2° grid points in our studied area was inverted by Tarantola's least square method for nonlinear problem. Based on the physical property that the phase velocity of fundamental Rayleigh wave at period T is most sensitive to the V S structure around λ /3 depth, we provided a layer‐velocity self‐adaptive technique in the structure inversion, which can well accelerate converging speed and improve stability in the inversion. The main conclusions concerning the V S structure of the researched area are: (1) The western Qinghai‐Tibet Plateau shows very high VS value in the lower‐crust and top of the upper mantle, and the asthenosphere is not developed. But the eastern margin of the Qinghai‐Tibet Plateau clearly shows low‐velocity anomaly in the lower crust and top of the upper mantle, which probably created a southward extrusion channel for the low velocity crust material of the Qinghai‐Tibet Plateau along the eastern margin of the plateau through the western Sichuan and Yunnan region and then linking the low velocity zone in northern Burma. In the northeastern margin of the Qinghai‐Tibet Plateau, the VS of the lower crust is apparently lower than that of the mid‐crust. (2) The Lhasa block of the southern Qinghai‐Tibet Plateau has a high velocity lid in the upper mantle and from south to north the depth of the asthenosphere decreases from 130 km to 100 km, the thickness of the underlying asthenosphere increases from about 40 km to 80 km. The Qiangtang block of the northern Tibet shows lower velocity in lower crust, lacks high velocity lid in the upper mantle and displays much thicker asthenosphere. (3) The Tarim basin and Junggar basin both show high velocity structure in the upper mantle, asthenosphere layer does not clearly appear. Thickness and velocity of the lower crust in the Junggar basin is apparently higher than that in the Tarim basin. (4) Velocity in the lower crust and uppermost mantle beneath the western Mongolia plateau is apparently lower than its eastern part. In the middle‐western Mongolia plateau a thick and large scale asthenosphere is clearly shown. This thick asthenosphere gradually becomes thinner from middle to eastern Mongolia plateau and above which the higher velocity lid becomes thicker and thicker. Geological explanations are accomplished based on results of inversion.
Abstract Through conformal mapping, in which the group velocity keeps invariant, a regional spherical area can be extended to a much larger spherical segment, which is then extended to the whole sphere by analytical continuation. Representing velocities by spherical harmonic expansion on the sphere after transformation can greatly reduce the number of spherical harmonic coefficients. The inversion of surface‐wave velocities turns into the linear inversion of spherical harmonic coefficients. The resolution kernel and the root square deviation of the inversion solution can be obtained by analyzing the spherical harmonic coefficients. The algorithm has advantages in the following aspects: its computing speed is very fast; the contours are smooth; and the tectonic boundaries are clear. Theoretically, this method is not only applicable to the surface‐wave group velocity inversion, but also to various analysis of regional fields.
'/%3e%3cuse xlink:href='%23a' transform='rotate(23.036 .093 25.536)'/%3e%3cuse xlink:href='%23a' transform='rotate(45.87 1.273 16.18)'/%3e%3cuse xlink:href='%23a' transform='rotate(69.945 .996 12.078)'/%3e%3cuse xlink:href='%23a' transform='rotate(20.66 -19.689 31.932)'/%3e%3c/svg%3e)