Based on the teleseismic records of the dense broadband seismic array, the crustal anisotropy parameters of the west Ordos block and its adjacent regions were determined with P wave receiver functions. The results indicate that the dominant direction of the fast wave is N-S in the Alxa block, NEE-SWW in north Ordos block, N-S in south Ordos block, and E-W in the Yangtze block. The fast wave direction of crust anisotropy in north Ordos block is distinctly different from the south Ordos block and the Alxa block. The Dabashan thrust fold belt is a visible boundary of crust anisotropy, which implies that the crust anisotropy kept the information of ancient South China block and North China block. Comparing with the direction of the fast wave from the SKS splitting, we tentatively discussed the interaction mode between the crust and mantle. The crust and mantle are decoupled in south Ordos block and coupled in north Ordos block. Combining the results of previous geophysical studies, we presented a crust-mantle interaction model to explain the geophysical observations. The most prominent features of the model are the horizontal eastward expansion of the mantle material in the southern Ordos and the vertical upwelling of the mantle material in the northern Ordos. The different modes of movement of the mantle material led to the deep contrasting structures of north and south Ordos, including the crust anisotropy. The mantle upwelling also implies that north Ordos block might be currently experiencing craton destruction.

Based on the massive teleseismic waveforms collected by the China Array II project, we investigate the 410 km and 660 km discontinuities (referred to as ‘410’ and ‘660’) beneath the northeastern Tibetan Plateau. After all of the 95747 radial receiver functions are stacked into each 0.5x0.5° grid, the P410s and P660s arrival times are Picked. Referring to the local 3-D tomographic model, the ‘410’ and ‘660’ depths are 416±0.6 km and 676±0.8 km, respectively, and the mantle transition zone (MTZ) thickness is ~260 km. Considering the remote effect of the subducted Indian Plate and Pacific Plate, we focused on the depressed two discontinuities. In particular, along the northeastern boundary of the Tibetan Plateau, the ‘410’ is confirmed as a dynamic barrier to hot asthenospheric flow, which escapes from the Tibetan Plateau and is obstructed by the rigid cratonic lithosphere in its surrounding area. The ‘660’ is obviously affected by cold anomalies that are delaminated or removed from the thickened lithosphere. In the western NCC(North China Craton), combining with the depressed ‘660’ and relatively larger P660s-P amplitude ratio, we speculate that a certain amount of water has accumulated in the lower MTZ, while the low Vs velocity around the 660 km depth may be related to melting. According to previous results, we propose a mantle convection model driven by the edge-derived convection of the subducted Pacific Plate. The cratonic lithosphere beneath the Ordos block may also undergo slow modification.