Fig. 1
From: Rapid mantle flow with power-law creep explains deformation after the 2011 Tohoku mega-quake

Post-earthquake deformation 2.8 years after the 2011 Tohoku-Oki Earthquake and surrounding material properties. a Measured displacement in the land stations (triangles) and the seafloor stations on both the continental plate and the pacific plate (inverse triangles). We removed some land stations for visibility. Coseismic displacement is not available in the station G01. Dashed-dotted and dotted lines are the location of the vertical cross-section (A−A′ profile) and the depth of the plate boundary, respectively. b Assumed temperature structure and frictional properties in the A−A′ profile. In the “unstable” region, where coseismic slip is input in our simulation, friction parameters are set as −0.2 ≤ A−B ≤ −0.1 MPa and 0.2 ≤ L ≤ 0.3 m. In the “stable” region, where afterslip occurs in our simulation, A−B = 0.1 MPa and L = 13 m (also see Fig. 2b). The temperature values in the layers of elastic materials are not used in the simulation. c The assumed viscoelastic structure before the earthquake in the A−A′ profile. The mantle wedge and oceanic mantle are viscoelastic with Gv = 65 GPa. The remaining volume is elastic with Ge = 45 GPa. Poisson’s ratio is ν = 0.25 everywhere. The color indicates the effective viscosity in the Maxwell element before the earthquake. We used the same color scale as in Fig. 5 here to highlight the change due to the earthquake. Contribution from dislocation creep is dominant in the light green area, while viscosity in the linear term is dominant (see Methods) elsewhere