Three-dimensional mechanical modeling of large-scale crustal deformation in China constrained by the GPS velocity field

We present a quantitative model for the crustal movement in China with respect to the Eurasia plate by using the three-dimensional finite element code ADELI The model consists of an elastoplastic upper lithosphere and a viscoelastic lower lithosphere. The lithosphere is supported by the hydrostatic pressure at its base. The India-Eurasia collision is modeled as a velocity boundary condition. Ten large-scale faults are introduced as Coulomb-type frictional zones in the modeling. The values for the root mean square (RMS) of the east and north velocity components differences (RMS(Ue) and RMS(Un)), which are between the observation and the prediction, are regarded as the measurements to evaluate our simulations. We model the long-term crustal deformation in China by adjusting the faults frictions ranged from 0.01 to 0.5 and considering the effects resulted from lithospheric viscosity variation and topographic loading. Our results suggest most of the large-scale faults frictions are not larger than 0.1, which is consistent with other large-scale faults such as the North Anatolian fault (Provost, A.S., Chery, J., Hassani, R., 2003. Three-dimensional mechanical modeling of the GPS velocity field along the North Anatolian fault. Earth Planet. Sci. Lett. 209, 361-377) and the San Andreas fault (Mount, VS., Suppe, J., 1987. State of stress near the San Andreas fault: implications for wrench tectonics. Geology, 15, 1143-1146). Further, we examine the effects on the long-term crustal deformation in China of three causes: the large-scale faults, lithospheric viscosity structure and topographic loading. Results indicate that the lithospheric viscosity structure and the topographic loading have important influences on the crustal deformation in China, while the influences caused by the large-scale faults are small. Although our simulations satisfactorily reproduce the general picture of crustal movement in China, there is a poor agreement between the model and the observed GPS velocity field in Sichuan-Yunnan area. It may result from our simple models such as that the faults are all vertical from model surface to bottom and that the effects caused by the subduction of Burma slab are neglected. (c) 2007 Elsevier B.V. All rights reserved.