 |
| Titel |
Grain size reduction and shear heating: a recipe for intermediate-depth earthquake generation? |
| VerfasserIn |
Marcel Thielmann, Antoine Rozel |
| Konferenz |
EGU General Assembly 2016
|
| Medientyp |
Artikel
|
| Sprache |
en
|
| Digitales Dokument |
PDF |
| Erschienen |
In: GRA - Volume 18 (2016) |
| Datensatznummer |
250134448
|
| Publikation (Nr.) |
 EGU/EGU2016-15176.pdf |
|
|
|
|
|
| Zusammenfassung |
| The mechanisms resulting in intermediate-depth earthquakes remain enigmatic, with two
processes - dehydration embrittlement and thermal runaway – being the most promising
candidates. Using a simple shear one-dimensional model, Thielmann et al. (2015) have
shown that the feedback between grain size evolution and shear heating significantly reduces
the stress needed to initiate thermal runaway.
However, at intermediate depths, Peierls creep as well as dislocation accommodated grain
boundary sliding (disGBS) are also viable deformation mechanisms. Here we investigate the
impact of those additional creep mechanisms (grain boundary sliding and Peierls creep)
on the formation of shear zones. As in Thielmann et al. (2015), we consider both
thermal and microstructural damage mechanisms (shear heating and grain size
reduction).
Depending on material and deformation parameters different creep mechanisms are
dominant during deformation, which affects the occurrence and timing of thermal
runaway (e.g. at low temperatures and/or high strain rates Peierls creep is dominant
and limits the strength of the material which delays thermal runaway). We derive
regime diagrams and from them regime boundaries that allow for easy determination
of the governing mechanisms and of the localization potential for given material
parameters.
In one-dimensional models however, the shear zone - once formed – extends infinitely. In
nature however, this is not the case. This has potentially a large impact on rupture velocities
during shear zone formation. For this reason, we compare the 1D predictions to 2D
simulations where fault length is finite. |
|
|
|
|
|
|