On November 14th 2001 an earthquake of magnitude Ms8.1 occurred at the Qinghai‐Xizang Plateau in China. The deformation observed by GPS showed that there were many differences between the deformations of south and north sides of the Kunlun fault. And after a short span of about three months, the deformations at all GPS stations were eastward. In this paper, the postseismic deformation observed by GPS is used as constraints and we analyze the possible mechanism of postseismic deformation after Kunlun earthquake. Firstly, this paper constructs the virtual work equation and theoretically analyzes the effects of heterogeneous, viscoelastic, poroelastic characters on the postseismic deformation. Then this paper uses grid‐search procedure to determine the viscosity for different models. The best‐fitting viscosity of lower crust is 5.0 × 1017 Pa·s and 9.0 × 1018 Pa·s for the south and north side of Kunlun fault, respectively. The difference of viscosities brought on the special postseismic deformation. And this difference is not only the result of long‐term geological action but also one of the crucial factors of modern geodynamic environment. On the other hand, the simulation shows that the postseismic deformation after 2001 Kunlun Ms8.1 earthquake is caused not only by the viscoelastic relaxation but also by poroelastic relaxation. So, it is necessary that the viscoelastic relaxation and poroelastic rebound should be combinedly considered when analyzing poseismic deformation.

Regarding all pores in rock as a system consisting of spherical pore and capillary pores, this paper presents a model of Sphere‐Capillary. After all pores are divided into different groups by their radii, the relaxation time of each group can be calculated with the Sphere‐Capillary Model. The transversal relaxation time (T2i), spaced on the Sphere‐Capillary Model, can be utilized to conduct the inversion of NMR relaxation signal. Our research suggests that the T2 distributions from the inversion are relative to special pore structures defined by the Sphere‐Capillary model. Using different Sphere‐Capillary Models to derive different relaxation times, then conduct the inversion relaxation signal with the relaxation time. When a T2 distribution fits the relaxation signal in the least squares, the pore structure defined by the Sphere‐Capillary Model characterizes the pore system in rock best. Data from lab NMR measurements are analyzed with the Sphere‐Capillary Model, and the results are compared with the pressure data from mercury injection. This research shows that the Sphere‐Capillary Model describes properly the relaxation characteristic relative to pore structure. Furthermore, the relaxation characteristic may hold a relationship with pore fluids.

[Abstract] A numerical simulation of the geodynamical process of the compression uplift of the Qinghai-Xizang plateau is addressed in this paper. The continental lithosphere of the East Asia is regarded as a continuum in a power law rheology. It lies on a relatively soft upper mantle and is limited in a trapezoid geological frame. The movement of the Indian Plate at the rate of 5cm/a is assumed to be the main driving force for the plateau's uplift. The results show that the predicted topography of the compression uplift can fit the observed pattern of this region. The results also indicate that the process of the compression uplift of the Qinghai-Xizang plateau is constrained by many factors such as the boundary conditions, the dynamical parameters of the continental lithosphere and the denudation rate as well. It is clear that the process of the plateau's uplift is not uniform in space and time.

The data of Bouguer gravity and topography are inverted to obtain the crust thickness of China. In order to reduce the effect of regional non-isostasy we corrected the reference Moho depth in the inversion with regional topography relief, and performed multiple iterations to make the result more reliable. The obtained crust thickness of China is plotted on a map in cells of 1°×1°. Then we analyzed the correlation between the Bouguer gravity anomaly and fluctuation of the Moho depth. A good linear correlation is found, with a correlation coefficient of −0.993. Different correlation coefficients, 0.96 and 0.91, are found for the data in land and ocean region, respectively. The correlation result also shows that the boundary between land and ocean is generally along the bathymetric line of −800 m. In order to examine the influence of the Earth′s curvature on the calculated result, we tried two inversion models: the inversion for the whole region and the inversion for 4 sub-regions. The difference in the crust thickness deduced from the two models is less than 5 km. Possible explanation for the difference is discussed. After comparing our result with that of other studies, we suggest that with our method the Bouguer gravity and the topography data can be independently inverted to obtain the crust thickness of China and its adjacency.