 |
| Titel |
Modelling the interaction between subducted slabs and thermo-chemical piles |
| VerfasserIn |
R. Gassmöller, B. Steinberger |
| Konferenz |
EGU General Assembly 2012
|
| Medientyp |
Artikel
|
| Sprache |
Englisch
|
| Digitales Dokument |
PDF |
| Erschienen |
In: GRA - Volume 14 (2012) |
| Datensatznummer |
250064191
|
|
|
|
|
|
| Zusammenfassung |
| Based on the reconstructed eruption locations of volcanic products (Large Igneous Provinces
and kimberlites) it has recently been proposed that most plumes get generated at the steep
edges of the large low shear-velocity provinces (LLSVPs) and that these edges have not
discernibly moved over the past few hundred Myr. Following this idea, it is of specific
importance to investigate the influences on the movement of these edges, which would
determine the surface plume positions. Independent of this argumentation, the position and
shape of the LLSVPs are a main factor in mantle convection and should be reproduced
by models which show an earth-like convection pattern. In current models this is
done in a rather qualitative way, with a focus either only on position or only on
shape.
In our 3D geodynamic numerical models of the global mantle we combine
several of the improvements to mantle convection models made during the last few
years: A complex plate reconstruction is used as a kinematic boundary condition. A
self-consistent thermodynamic material model for basalt, harzburgite and peridotite is used
to derive a temperature- and/or pressure-dependent database for parameters like
density, thermal expansivity and specific heat. Furthermore, we use a viscosity
profile derived from surface observations and mineral physics constraints. We use the
model to clarify the influence of surface motions on processes at the CMB and vice
versa.
The results of our models show not only that it is possible to generate LLSVPs at the
actual position and with a similar shape compared to what is observed through seismic
tomography, but even to recreate plumes at positions that match many of today’s hot spots. A
thorough investigation on the statistical significance of this observation will be presented. We
will also show the influence of boundary and initial conditions. Furthermore, we will
discuss the possibility to create a model matching seismic tomography and plume
positions without a chemical boundary layer at the core-mantle boundary, and the
difference in stability of the LLSVPs between a harzburgitic and a peridotitic ambient
mantle. |
|
|
|
|
|
|