The Zipingpu Reservoir Triggering of the 2008 Mw 7.9 Wenchuan Earthquake, China, Due to Poroelastic Coupling

Abstract:

Impoundment of the Zipingpu Reservoir (ZR), China, began in September 2005 and culminated in a water depth averaging 100 meters. This impoundment was followed 2.7 years later by the 2008 Mw 7.9 Wenchuan Earthquake (WE), which ruptured the Longmen Shan Fault (LSF) with its epicenter about 12 kilometers away from the ZR. The close proximity of reservoir impoundment and the WE in both space and time suggests that the events are coupled. Previous studies, however, obtained inconclusive results. Based on the fully-coupled poroelastic theory, we employ three-dimensional Finite Element Models (FEMs) to simulate the evolution of stress and pore pressure due to reservoir impoundment, and its effect on the Coulomb stress on the LSF. The results indicate that the reservoir impoundment broke the regional balance of pore pressure, and formed a pore-pressure front that slowly propagated through the crust with fluid diffusion. Due to the diffusion of the pore pressure, the effective normal stress increased steadily, and so did the Coulomb stress change on LSF. The FEMs predict the positive Coulomb stress changes on the LSF propagating from the upper crust to lower crust. The reservoir loading had increased the Coulomb stress on the shallow part of the LSF up to tens to hundreds of kPa, significantly in favor of the failure of the LSF. It is still quite uncertain about the geometry and extent of the LSF at depth, we therefore calculate the Coulomb stress changes over a range of scenarios of hypocenter locations and fault dip angles. Our result shows that in most of the scenarios the Coulomb stress at the hypocenter either had no increase or increased moderately prior to the WE. However, the Coulomb stress grew significantly in the shallow to mid depth range of the LSF, which “advanced” the leveling of tectonic loading of the fault system for tens to hundreds of years. The initial slip of the WE might or might not be triggered by the impoundment of ZR; but a small event at the hypocenter could trigger large rupture on fault plane above, where the asperities were located. The micro-seismicity around the ZR also showed an expanding pattern from the ZR since its impoundment, which could be associated with diffusion of positive pore pressure. These results suggest a poroelastic triggering effect of the WE due to the impoundment of the ZR.