Create new viscosity prefactor feature

[danieldouglas92]

This pull request implements the framework that will allow the user to apply more complicated viscosity prefactors to the visco-plastic material model, like accounting for the water fugacity. This is what I implement here. The compositional field 'bound_water' is queried and used to determine the water fugacity for olivine using Hirth and Kohlstaedt 2004 (10.1029/138GM06). This should be set up in such a way that it can be applied to the other rheologies, like diffusion/dislocation creep. This PR takes the place of #4965

[bobmyhill]

Hi @danieldouglas92, thanks for doing this!

The code structure is what I had envisaged, but the content of ViscosityPrefactors<dim>::compute_viscosity(...) is unexpected to me. I think what you want is to vary your viscosity as a function of the values of the compositional fields, right? If so, the viscosity should not be an explicit function of time.

My thought was that compute_viscosity would have different / additional arguments compared to the function in constant_viscosity_prefactors.cc. For example, you could have base_viscosity and a double or vector-double of state_parameters, and then have the function calculate the new viscosity based on the values of those state parameters. I think you just need a power law, right?

To start with, maybe just code up the functional form that you will need for your scientific case; we can extend in later PRs as needed.

[naliboff]

@bobmyhill - here is a summary of my conversation with @danieldouglas92 about this PR.

1. The long-term goal is to have a single rheology module that supports multiple cases for handling prefactor terms that are multiplied with the viscous flow laws.

2. The way for selecting this could be through enum variable, similar to ViscosityScheme in rheology/visco_plastic

      enum ViscosityPrefactorSheme
      {
        constant,
        mantle_hydration
      } viscosity_prefactor_scheme;
   

with more schemes added on through time.

3. As a starting point, this PR could be modified to just re-implement the constant scheme and re-use the same test? Or perhaps it would be better to add in a slightly different (but simple to implement) test case so that we are not re-implementing a constant viscosity prefactor multiplication case in visco_plastic?

[naliboff]

This is looking good, thanks! Mainly comments on documentation and naming schemes.

[naliboff]

@danieldouglas92 - Here are a few additional comments on the test case.

[danieldouglas92]

This should be ready for another round of reviews @naliboff and @bobmyhill, thanks in advance!

[bobmyhill]

Nice! I think this is almost there, just a few comments.

[naliboff]

A few more minor renaming suggestions

[danieldouglas92]

Should be ready for another run through @naliboff

[naliboff]

@danieldouglas92 - Thanks for the updates and the PR is in great shape! Only a few minor cleanup requests below.

[danieldouglas92]

As long as the testers pass, I think I just need one last review from you @bobmyhill!

[naliboff]

/rebuild

[bobmyhill]

Nice! Just a few doc comments and then this is good to go.

[danieldouglas92]

Ready for another (hopefully last) run through @naliboff and @bobmyhill! Thanks again for all the helpful feedback

[naliboff]

Good to go from my side, thanks!