5:15 PM - 7:15 PM
[MGI30-P11] Plate motion reproduced in a spherical shell mantle convection model with stress-history-dependent viscosity and its characteristics
Keywords:mantle convection, plate motion, stress-history-dependent rheology
On the Earth, the lithosphere is formed by strong temperature-dependent viscosity and is divided into several pieces of rigid plates that move relative to each other. To understand the activity and evolution process of the solid Earth, a mantle convection model that can accurately handle plate motion is indispensable.
Deformation in the lithosphere is concentrated to very narrow regions (compared to the size of the plates) at plate boundaries. These boundaries allow each plate to move as a solid body, and in turn lead to stable and steady plate motions on the present Earth over the time scale of O(102) million years. Formation of plate boundaries is, therefore, a crucial step in plate tectonics. On the Earth, once a plate boundary is newly formed by a rifting of plates under very high stress, it remains even if the stress is reduced below the rupture strength. The mechanical state of a plate is determined not only by the stress at that moment, but also by its "memory" (stress history) on whether or not it has exceeded the rupture strength in the past.
We used a spherical shell mantle convection model (ACuTEMan, Kameyama et al. 2005; 2008) with stress-history-dependent viscosity (Ogawa 2003) to simulate the characteristics of plate motion and the formation of plate boundaries more properly.
The effective viscosity in narrow plate boundary regions is several orders of magnitude lower (approximately 103 or more) than that in the plate interiors. Besides, the lithosphere as a stiff lid over the convecting mantle develops only when there is a viscosity contrast larger than 103 between the surface and the deep mantle due to temperature-dependence of viscosity. These extremely large viscosity variations have been one of the main obstacles for numerical simulations of tectonic plates in mantle convection.
We resolved this obstacle and carried out calculations in a spherical shell mantle on a mesh system with six times larger mesh numbers than that we used in our previous calculation in a 3D box (Miyagoshi et al. 2020). Namely, narrow plate boundaries are successfully formed associated with a sharp viscosity drop, leading to stable motions of rigid plates for more than one billion years. This plate regime is clearly different from the regime where the viscosity ratio between the plate boundary and the plate interior is not so large (called the weak plate regime). On the weak plate regime, there is no region on the surface similar to rigid plates which move coherently or stably.
The calculated ridges are simple narrow bands of low-viscosity region, but the calculated subduction zones have a more complicated structure: "micro-plates" often develop along trenches as observed for Earth. Cold and solid plates sink deep into the mantle while retaining its entity as slabs, and reach the bottom of the mantle. The slabs cause substantial horizontal temperature heterogeneities at the core-mantle boundary (CMB). Our results may be useful in studying the geodynamo driven by horizontal thermal inhomogeneities at the CMB.
We also found that the secondary convection occurs beneath the stably moving plates away from the "ridges" as seen in the two-dimensional model (Ogawa, 2003) and the box model. The secondary convection significantly affects the heat flow distribution on the surface, resulting in its deviation from the prediction from the half-space cooling.
Deformation in the lithosphere is concentrated to very narrow regions (compared to the size of the plates) at plate boundaries. These boundaries allow each plate to move as a solid body, and in turn lead to stable and steady plate motions on the present Earth over the time scale of O(102) million years. Formation of plate boundaries is, therefore, a crucial step in plate tectonics. On the Earth, once a plate boundary is newly formed by a rifting of plates under very high stress, it remains even if the stress is reduced below the rupture strength. The mechanical state of a plate is determined not only by the stress at that moment, but also by its "memory" (stress history) on whether or not it has exceeded the rupture strength in the past.
We used a spherical shell mantle convection model (ACuTEMan, Kameyama et al. 2005; 2008) with stress-history-dependent viscosity (Ogawa 2003) to simulate the characteristics of plate motion and the formation of plate boundaries more properly.
The effective viscosity in narrow plate boundary regions is several orders of magnitude lower (approximately 103 or more) than that in the plate interiors. Besides, the lithosphere as a stiff lid over the convecting mantle develops only when there is a viscosity contrast larger than 103 between the surface and the deep mantle due to temperature-dependence of viscosity. These extremely large viscosity variations have been one of the main obstacles for numerical simulations of tectonic plates in mantle convection.
We resolved this obstacle and carried out calculations in a spherical shell mantle on a mesh system with six times larger mesh numbers than that we used in our previous calculation in a 3D box (Miyagoshi et al. 2020). Namely, narrow plate boundaries are successfully formed associated with a sharp viscosity drop, leading to stable motions of rigid plates for more than one billion years. This plate regime is clearly different from the regime where the viscosity ratio between the plate boundary and the plate interior is not so large (called the weak plate regime). On the weak plate regime, there is no region on the surface similar to rigid plates which move coherently or stably.
The calculated ridges are simple narrow bands of low-viscosity region, but the calculated subduction zones have a more complicated structure: "micro-plates" often develop along trenches as observed for Earth. Cold and solid plates sink deep into the mantle while retaining its entity as slabs, and reach the bottom of the mantle. The slabs cause substantial horizontal temperature heterogeneities at the core-mantle boundary (CMB). Our results may be useful in studying the geodynamo driven by horizontal thermal inhomogeneities at the CMB.
We also found that the secondary convection occurs beneath the stably moving plates away from the "ridges" as seen in the two-dimensional model (Ogawa, 2003) and the box model. The secondary convection significantly affects the heat flow distribution on the surface, resulting in its deviation from the prediction from the half-space cooling.