Implementing different rheological parameters for upper and lower mantle in the viscoplastic model

ASPECT
1

Hello all,

I have a question regarding the viscoplastic material model. Is there an option where we can impose different rheological parameters for the upper and lower mantles without having them as separate compositional fields? For instance, arbitrary density crossovers can be implemented here for example at 660 km if we set ‘Phase transition depths=background:660e3’ and then later on having “set Densities = background: 3300|4100”. Can we do the same, say for example the activation energies for diffusion creep with “set Activation energies for diffusion creep = background:300e3|200e3”. This has been done in the paper: Non-collisional orogeny in northeast Japan driven by nearby same-dip double subduction | Nature Geoscience, however, they did not impose a modified viscoplastic model plugin in their input file, so I assumed they simply used the built-in viscoplastic model functionality of ASPECT 2.5.0.

Thank you in advance!

Kind regards,
John

2

Hi @jmagali,

Can we do the same, say for example the activation energies for diffusion creep with “set Activation energies for diffusion creep = background:300e3|200e3”.

The PRM files in the zenodo repository for the linked paper illustrate their approach for the phase transitions, and there are also additional examples in the aspects tests suite (one example).

Please let us know if you have any further questions about this!

Cheers,
John

3

Hi John,

Thanks for the quick reply! Yes, I understand the formatting when including phases in the viscoplastic material model. However, in the case of diffusion and/or dislocation, the format “background: value|value” only seems to work in the “Prefactors for diffusion/dislocation creep”. Doing so in the activation energies/volumes give the following error:

Exception 'ExcMessage (“The string for property <” + options.property_name + "> does not have the expected format. " + "Check that the string is either a " + "comma separated list of <double>' or " + " : ||…, " + " : |… , … ‘. " + "If the string looks correct, " + "it is likely that the length of the " + "list of keys passed to " + "parse_map_to_double_array does not " + "match the length of the " + “comma separated property list.”)’ on rank 30 on processing:

Or i think i am missing something.. Here is a snippet of the viscoplastic model in the input file:

        set Activation volumes for dislocation creep = background:12e-6|0,WZ:0|0,LLVP:12e-6|0
        set Activation energies for dislocation creep = background:740e3|0,WZ:0|0,LLVP:740e3|0
        set Prefactors for dislocation creep = background:5e-16|1e-24,WZ:1e-21|1e-21,LLVP:5e-16|1e-24
        set Stress exponents for dislocation creep = 3.5 #background:3.5|1,WZ:1|1,LLVP:3.5|1

Thank you again!

John

4

@jmagali,

Thanks for the clarificaiton.

Can you post the PRM file or just the material section of it? It seems as if there is an issue with the formatting.

I tried modifying the test visco_plastic_phases_plasticity.prm to use the configuration below, which ran without issue.

    set Activation energies for dislocation creep = background: 0. | 220e3
    set Activation volumes for dislocation creep = background: 0. | 0.

Cheers,
John

5

Hi John,

Here is a snippet of the material model part:

subsection Material model
  #set Material averaging = harmonic average 
  set Model name = visco plastic
  subsection Visco Plastic

        set Reference temperature = 1573

        set Phase transition depths = background: 660e3, WZ: 660e3, LLVP: 660e3
        set Phase transition widths = background: 1e3, WZ: 1e3, LLVP: 1e3
        set Phase transition temperatures = background: 1850, WZ: 1850, LLVP: 1850
        set Phase transition Clapeyron slopes = background: 0, WZ: 0, LLVP: 0

        set Minimum viscosity = 1e20
        set Maximum viscosity = 1e25

        set Thermal diffusivities = 1e-6
        set Heat capacities = 1250
        set Densities = background: 3300|4150,WZ:3300|4150,LLVP:3550|4400
        set Thermal expansivities = 3e-5

        set Viscous flow law = composite
        set Activation volumes for diffusion creep = background: 4e-6|1.5e-6, WZ: 0|0, 4e-6,LLVP:4e-6|1.5e-6
        set Activation energies for diffusion creep = background:300e3|200e3,WZ:0|0,LLVP:300e3|200e3
        set Grain size =1
        set Prefactors for diffusion creep = background: 3e-11|2e-17,WZ:1e-20|1e-20, LLVP:3e-11|2e-17 #background:1e-9|1e-10,WZ:1e-20|1e-20,LLVP:1e-9|1e-10

        set Activation volumes for dislocation creep = background:12e-6|0,WZ:0|0,LLVP:12e-6|0
        set Activation energies for dislocation creep = background:540e3|0,WZ:0|0,LLVP:540e3|0
        set Prefactors for dislocation creep = background:5e-16|1e-24,WZ:1e-20|1e-20,LLVP:5e-16|1e-24
        set Stress exponents for dislocation creep = 3.5 #background:3.5|1,WZ:1|1,LLVP:3.5|1


        #set Yield mechanism = drucker
        set Cohesions = 1e15
        #set Angles of internal friction = 5
        #set Maximum yield stress = 1e15
        set Constant viscosity prefactors = 1,1,10

   end


end

Kindly,
John

6

Hi again, John
I found the error!
Thank you again for the assistance
With gratitude,
John

7

Glad you found the error! At first glance, it looks the issue was at the following snippet under set Activation volumes for diffusion creep - WZ: 0|0, 4e-6,. Can you confirm?

set Activation volumes for diffusion creep = background: 4e-6|1.5e-6, WZ: 0|0, 4e-6,LLVP:4e-6|1.5e-6

As an aside, I find using the following formatting approach that breaks parameter assignments into multiple lines quite helpful in mitigating these types of issues:

set Activation volumes for diffusion creep = background: 4e-6 | 1.5e-6, \
                                             WZ: 0 | 0, \
                                             ....

Cheers,
John