Probing deep rheology across the eastern margin of the Tibetan plateau: Constraints from the 2008 Mw 7.9 Wenchuan earthquake

Figure 1. The 3D representation of the rheologic model in eastern Tibet and western Sichuan basin. These two geologic structures are separated by the Longmen Shan. The co- and post-seismic GPS displacements are shown in the black and red arrows, respectively. The two possible mechanisms of the postseismic deformation are: (1) deep afterslip (the light blue region), and (2) lower crustal flow (the purple layer). The coseismic slip is inverted from the coseismic GPS measurements. In the lower left, the postseisic displacement during the first year is compared with the two end-member mechanisms. The two afterslip models in the lower right are inverted from the one year postseismic GPS measurements and from the LCF model residual, respectively. The deep afterslip in the multiple mechanism model is largely reduced.

Project Summary The fundamental geological structure and rheology of the Tibetan plateau have been debated for decades. Two major models have been proposed: (1) the deformation in Tibet is distributed, and associated with ductile flow in the mantle or lower crustal flow (LCF); (2) the Tibetan plateau was formed during interactions among rigid blocks with localization of deformation along major faults. On 12 May, 2008, a Mw 7.9 earthquake occurred on the Longmen Shan that separates the eastern Tibetan plateau and the Sichuan basin. The earthquake ruptured ~235 km of the Beichuan fault (BCF) and the entire Pengguan fault (PGF) (Shen et al., 2009). Geodetic inversions show more than 5 slip asperities and ~16 m peak slip on SW BCF (Fig. 1). All of the slip models show oblique thrusting along the SW BCF and a right-slip component gradually increases towards the NE end of the BCF. The postseismic displacement is a response to the redistribution of stresses induced by the earthquake and can be used to probe the deep rheologic properties underneath the surface (Wang et al., 2012). Here we incorporate two-year long geodetic measurements and numerical modeling to examine two end-member hypotheses to provide further evidence to the deep rheology in eastern Tibetan plateau.

The Postseismic Displacement The GPS measurements show an overall NW-SE convergent displacement in SW BCF, and turn into right lateral strike-slip motion in NE BCF. This pattern is similar to the coseismic displacement (Fig. 1), but the peak displacement is about 40 km away from the coseismic surface rupture where the peak coseismic displacement is located (Shen et al., 2009). In the hanging wall, the amplitude of displacement increases from 0-2 cm near the surface rupture to about the location of the Wenchuan-Maowen fault (WMF), and then decays from 5-7 cm at WMF to 3-4 cm in the far field. Comparing with the coseismic displacement (black arrows in Fig. 1), the gradient of the displacement away from the fault is much lower and might imply either a deeper slip on the fault or viscous relaxation from the deeper part of the lithosphere. In the footwall, all of the displacement moves toward NW and the amplitude is much smaller than in the hanging wall.

Model of the Postseismic Displacement The afterslip is the continuous slip of the fault after the main shock and is often considered downdip of the fault rupture zone (Wang et al., 2012). We use a dislocation model with layered structures to investigate the afterslip distribution by inverting the geodetic data. We modify the fault geometry proposed by Shen et al. (2009) and extend the fault width to 65 km depth for afterslip at the downdip extension (afterslip model in Fig. 1). The afterslip distributes on both shallow and deep parts of BCF that represents the fit to the near- and far field displacement. We use a 3D finite element model to construct a regional rheologic model composed of an elastic Tibet upper crust and Sichuan crust, a viscoelastic Tibet lower crust, and a viscoelastic upper mantle. We use the bi-viscous Burger’s rheology to represent the transient and steady state periods of the postseismic deformation. The Burger’s rheology is composed of a Maxwell fluid connected in series with a Kelvin solid to represent the steady state and transient viscosities (η1 and η2, respectively). The best-fitting model is composed of the LCF located between 45 and 60 km in depth and can produce more far field postseismic displacement. The afterslip model can explain the postseismic displacement in the near field but there is larger misfit in the far field. On the other hand, the viscoelastic relaxation model can explain the far field postseismic displacement better than the near field. It appears that a single mechanism cannot solely explain the postseismic displacement. A multiple- mechanism model is needed to fit both near- and far field displacements. We consider the 15 km thick LCF to be the main mechanism of the far field displacement, so the afterslip model may explain the misfit of the LCF model. The inversion (the 2nd afterslip model in the lower right in Fig. 1) of the LCF residual displacement shows a significant reduction of the deep afterslip. As a result, the afterslip alone model requires more than 45 cm slip in the first year below Tibet’s Moho that may already undergo ductile deformation, whereas the viscoelastic relaxation in a 15 km thick LCF can explain the GPS measurements. Consequently, the result of the Wenchuan postseismic displacement supports a weak lower crustal flow underneath eastern Tibet.

Tools InSAR processing and analysis, GPS data analysis, forward and inverse elastic and viscoelastic modeling

Geographic Location Eastern Tibetan Plateau

Group Members Involved Mong-Han Huang, Roland Bürgmann

Project Duration In Progress: start of 2009 through 2014