Choosing rheologies for different compositional fileds in Visco-Plastic model

ASPECT
1

Hi all,

I have a question about how to choose rheologies for different compositional fields in Visco-Plastic model. For example, I have three fileds named ‘f_1’,‘f_2’ and ‘f_3’, which correspond to diffusion creep, dislocation creep and peierls creep. How should I express such settings in ‘subsection compositional fields’ and ‘subsection Material model’ ？

Best,
Xie

2

Hi Xie,

Thank you for posting to the forum!

To answer your question, it would help to have a bit more context about what specifically you are trying to achieve.

Is the idea to have three fields, representing different rock types, that each only deform according to a single deformation mechanism? The visco plastic model does not currently support having different mixes of deformation mechanisms for distinct compositional fields.

However, one can adjust the parameters for each deformation mechanism such that their average viscosity between will be equivalent to the value of one specific mechanism.

For example, imagine at a specific temperature, pressure and pressure the diffusion creep parameters give a viscosity of 1e22 Pa s. If you set the dislocation and Peierls creep parameters to have no dependence on temperature, pressure, and strain rate and have a fixed viscosity of say 1e50 Pa s, the average viscosity will be ~ 1e22 Pa s.

Cheers,
John

3

Hi John,

Thanks for replying. Your understanding is correct. Sorry to hear that the VP model does not currently support having different mixes of deformation mechanisms for distinct compositional fields. I think your suggestion will be a feasible solution.

Best,
Xie

4

Hi Xie,

Glad to hear that the proposed solution will be a feasible one in the short term.

I should add that @bobmyhill has been working on new composite rheology formulations, which I believe will allow specifying what mechanisms to use for each compositional field. We are planning to test these formulations with the visco plastic model as a next step.

Cheers,
John