No abstract

S U M M A R YBenchmark comparisons are an essential tool to verify the accuracy and validity of computational approaches to mantle convection. Six 2-D Cartesian compressible convection codes are compared for steady-state constant and temperature-dependent viscosity cases as well as time-dependent constant viscosity cases. In general we find good agreement between all codes when comparing average flow characteristics such as Nusselt number and rms velocity. At Rayleigh numbers near 10 6 and dissipation numbers between 0 and 2, the results differ by approximately 1 per cent. Differences in discretization and use of finite volumes versus finite elements dominate the differences. There is a small systematic difference between the use of the anelastic liquid approximation (ALA) compared to that of the truncated ALA. In determining the onset of time-dependence, there was less agreement between the codes with a spread in the Rayleigh number where the first bifurcation occurs ranging from 7.79 × 10 5 to 1.05 × 10 6 .

Abstract.The formation of localized shear zones is important for understanding many local and global processes in geodynamics. We have developed a selfconsistent thermal-mechanical model together with a rheology which depends on temperature, strain-rate and grain-size distribution. The grain-size distribution has contributions from both dynamic recrystallization and grain-growth processes, and is governed locally We will focus especially on the delicate interplay between the fault-narrowing tendency from viscous heating and the fault-healing influence from grain-growth in controlling the fate of these localized shear zones, with dimensions on the order of hundreds of meters.

No abstract

No abstract

No abstract

Numerical models of bottom-heated thermal convection of highly compressible fluid with strongly temperature-dependent viscosity are presented to understand how the Rayleigh number Ra and the temperature dependence of viscosity exert control over the regimes of thermal convection in massive super-Earths. Thermodynamic properties of mantle materials are pressure dependent, but other material properties including the viscosity are not. A stagnant lid develops along the surface of the planet, when the viscosity contrast across the mantle due to temperature dependence r exceeds 10 6 at high Rayleigh number relevant to super-Earths. The threshold in r, which increases with increasing Ra, is higher than that expected for the Earth from earlier Boussinesq models. The efficiency of convective heat transport measured by the Nusselt number Nu is considerably lower than that expected from Boussinesq models; Nu depends on Ra and r as Nu = 59 Á r . Strong adiabatic compression significantly reduces the activity of hot ascending plumes especially at high r. At r relevant for super-Earths, hot ascending plumes lose their buoyancy on their way and hardly reach the surface boundary: hot spot volcanism due to ascending plumes is probably suppressed on super-Earths. The lithosphere is considerably thicker than that suggested by earlier Boussinesq models and is unlikely to show a plate-like behavior.

We have developed a two-dimensional dynamical model of asymmetric subduction integrated into the mantle convection without imposed plate velocities. In this model we consider that weak oceanic crust behaves as a lubricator on the thrust fault at the plate boundary. We introduce a rheological layer that depends on the history of the past fracture to simulate the effect of the oceanic crust. The thickness of this layer is set to be as thin as the Earth's oceanic crust. To treat 1-kilometer scale structure at the plate boundary in the 1000-kilometer scale mantle convection calculation, we introduce a new numerical method to solve the hydrodynamic equations using a couple of uniform and nonuniform grids of control volumes. Using our developed models, we have systematically investigated effects of basic rheological parameters that determine the deformation strength of the lithosphere and the oceanic crust on the development of the subducted slab, with a focus on the plate motion controlling mechanism. In our model the plate subduction is produced when the friction coefficient (0.004 -0.008) of the modeled oceanic crust and the maximum strength (400 MPa) of the lithosphere are in plausible range inferred from the observations on the plate driving forces and the plate deformation, and the rheology experiments. In this range of the plate strength, yielding induces the plate bending. In this case the speed of plate motion is controlled more by viscosity layering of the underlying mantle than by the plate strength. To examine the setting of the overriding plate, we also consider the two end-member cases in which the overriding plate is fixed or freely-movable. In the case of the freely-movable overriding plate, the trench motion considerably changes the dip angle of the deep slab. Especially in the case with a shallow-angle plate boundary, retrograde slab motion occurs to generate a shallow-angle deep slab.