[J05-2] Crustal dynamics: Multidisciplinary approach to seismogenesis II

[J05-2-01] A unified representation of Earth's quasi-dynamic deformation processes

^○Sylvain Barbot (Earth Observatory of Singapore)

invited

The last three decades have witnessed an explosion of studies on fault processes, from kinematic modeling of geodetic data to dynamic modeling of fault rheology. These studies were made possible by fundamental solutions that describe the stress and displacements caused by slip on a fault. In contrast, direct imaging of the kinematics of off-fault deformation is still impractical, and the dynamic modeling of anelastic deformation still relies on computationally intensive numerical methods (e.g., Barbot & Fialko, 2010). Here, we describe a novel approach that allows us to resolve distributed processes in kinematic inversions of geodetic data and to incorporate off-fault processes in numerical models of earthquake cycles. We quantify analytically the displacement and stress incurred by distributed anelastic strain in finite shear zones (Barbot, Moore & Lambert, BSSA, 2017). We use these elementary solutions to simultaneously invert for slip on faults and distributed strain in the surrounding rocks. We apply this new technique to study postseismic relaxation following the 2016 Mw 7.0 Kumamoto, Japan earthquake (Moore et al., 2017). We directly image the spatial distribution of effective transient viscosity in the Kyushu lower crust, revealing the systematic variations associated with the volcanic arc and plutonic bodies. Our formulation also allows the dynamic simulation of earthquake cycles with distributed deformation using the integral method, an approach more accurate and many orders of magnitude faster than classic finite-element techniques. We simulate earthquakes cycles within the lithosphere-asthenosphere system using rate-and-state friction and the power-law flow of olivine (Lambert & Barbot, 2017), revealing the prevalence of viscoelastic flow in the early stage of postseismic relaxation. Our approach will be instrumental in building comprehensive physical models of stress evolution at plate boundaries.