Model AniCD-L: Reveals the zonation of lithospheric anisotropy in the Sichuan–Yunnan region
image: Capital letters A–N and lowercase letters a–e denote subzone codes. Black line segments represent the mean XKS fast polarization direction for each corresponding subzone. The color of each subzone indicates the magnitude of its splitting time. The thick blue-black horizontal line marks the 26.5°N. The yellow areas outlined by red circles indicate regions where results from different researchers show inconsistency.
Credit: ©Science China Press
The Sichuan–Yunnan (i.e. Chuan-Dian, CD in short) region in the eastern and southeastern margins of the Tibetan Plateau characterizes with strong seismicity and complex intracontinental deformation. Seismic anisotropy provides an effective tool to study lithospheric deformation and deep mantle mass migration patterns. The teleseismic XKS (including SKS, PKS, and SKKS phases) splitting method offers particular advantages for investigating the azimuthal anisotropy of the lithosphere and the uppermost asthenosphere beneath seismic stations. It makes it possible to reliably measure S-wave splitting parameters, including the fast polarization direction and splitting time. Numerous studies have investigated lithospheric anisotropy in the CD region by XKS splitting. These studies generally showed consistent regional patterns in fast polarization directions, though localized complexity and discrepancies in anisotropic parameters.
By integrating existing XKS splitting measurements in the CD region, and considering the spatial pattern of fast polarization directions together with the regional tectonic framework, including block boundaries and major deep faults, the study area is divided into 12 subzones (A–N), companying with five small local zones (a–e) characterized by significant discrepancies. In the averaging process, results reported by different studies for the same station are treated as independent effective records. By calculating the mean values and standard deviations (SDs) of the fast polarization directions and splitting times for each subzone, the body-wave anisotropic model AniCD-L is obtained (Figure 1).
The mean fast polarization directions of model AniCD-L exhibit a pronounced north–south zonation, with an approximate boundary at 26.5°N. In the northern part of the CD region, the fast polarization directions are predominantly near NW–SE. Subzone A, located within the Songpan–Ganzi block, shows a dominant NW–SE fast polarization direction with a mean of 138.5°, while subzone E in the north Sichuan–Yunnan block exhibits a dominant NNW–SSE fast polarization direction with a mean of 153.9°. In the southern part of the CD region, the fast polarization directions are mainly near E–W. Subzone M, which covers the majority of Yunnan and the western margin of the Yangtze Craton, has a dominant near E–W fast polarization direction with a mean of 94.2°. In the western part of subzone M, the Tengchong volcanic area (subzone L) shows a near NW–SE fast polarization direction with a mean of 137.4°, whereas subzone N along the southern margin has a mean direction of 125.3°.
According to model AniCD-L, the splitting times are relatively low in the central part of the region, while higher values are observed in both the north and south. The maximum splitting time is observed in subzone M, encompassing the majority of Yunnan and the Yangtze Craton, with a mean value of 1.05 s. Conversely, the minimum splitting time occurs in subzone D (the Sichuan Basin), with a mean of only 0.57 s. In the north CD Region, the Songpan–Ganzi block (subzones A and B) exhibits strong anisotropic characteristics, with splitting times exceeding 1.00 s in both subzones and mean values of 1.01 s and 1.04 s, respectively. In the central part of the CD Region, subzones E–J show relatively consistent splitting times around 0.80 s.
By integrating model AniCD-L with GNSS surface deformation, crustal anisotropy and deep seismic tomography, three main insights were obtained.
(1) North of 26.5°N, the fast polarization directions are predominantly NW–SE and show good consistency with the GNSS velocity field and the strike of major faults. The observations support crust–mantle coupling, reflecting likely coherent eastward extrusion of Tibetan Plateau crust and mantle material under the ongoing northward subduction of the Indian Plate.
(2) South of 26.5°N, the nearly E–W fast polarization directions show significant differences from crustal deformation patterns, supporting crust–mantle decoupling. This pattern is associated with likely mantle flow induced by slab rollback.
(3) At the lithospheric scale, model AniCD-L provides new constraints on the southwestern boundary of the South China block. The Lijiang–Xiaojinhe and the Jinpingshan faults (i.e. general Lijiang–Xiaojinhe fault) indicate the likely position of the west boundary of the South China block, and this boundary extends southward roughly along the vicinity of the Red River fault.