Formation and dynamics of a chemically stratified layer below the Earth's CMB
Abstract
Seismological and magnetic observations are compatible with the presence of a stratified layer below the Earth's CMB (Lay and Young, 1990; Tanaka, 2007; Gubbins, 2007; Helffrich and Kaneshima, 2010; Lesur et al., 2015) and the existence of such a layer has also been predicted by several theoretical arguments listed below. The proposed thickness varies from 60 km to several hundreds of kilometers across the literature, but is usually close to 100 km. The layer may be thermally stratified if the CMB heat flow is subadiabatic (Gubbins et al., 1982; Labrosse et al., 1997; Lister and Buffett, 1998; Labrosse, 2015) but the possibility of a stratification of chemical origin has also been evoked. Various mechanisms have been proposed for the formation of a chemically stratified layer and include barodiffusion i.e. diffusion of light elements against the pressure gradient (Fearn and Loper, 1981; Braginsky, 2006; Gubbins and Davies, 2013), chemical plumes and blobs that would be able to reach the CMB where they would accumulate (Loper, 1989; Braginsky, 1994; Moffatt and Loper, 1994; Loper, 2007) or ascending droplets in a Fe-S system kept from mixing by surface tension (Franck, 1982). Layering may also be present if immiscible liquids evolve as the composition changes due to inner core growth (Helffrich and Kaneshima, 2004). To finish, Buffett and Seagle (2010) also studied the possibility that light elements be dissolved from the mantle into the core, forming a lighter layer that could grow by diffusion over long time scales. So far, no numerical simulation of core dynamics has been able to validate any of these potential mechanisms and produce a chemically stratified layer in a self-consistent manner. Using a particle-in-cell method newly implemented in the code PARODY (E. Dormy, J. Aubert) allowing to perform simulations of thermochemical convection in the infinite Lewis number limit (neglecting the compositional diffusivity), I will show that a chemically stratified layer systematically forms in simulations via the accumulation of undiffused plumes and blobs that reach the top boundary. I will discuss the dynamics of such a layer and will present scaling laws of its thickness as a function of the control parameters. Extrapolation to the terrestrial case and the potential implications for core dynamics will also be discussed.
- Publication:
-
AGU Fall Meeting Abstracts
- Pub Date:
- December 2017
- Bibcode:
- 2017AGUFMDI31B..03B
- Keywords:
-
- 1507 Core processes;
- GEOMAGNETISM AND PALEOMAGNETISM;
- 1510 Dynamo: theories and simulations;
- GEOMAGNETISM AND PALEOMAGNETISM;
- 3924 High-pressure behavior;
- MINERAL PHYSICS;
- 7207 Core;
- SEISMOLOGY