[aspect-devel] Aspect melt_fraction

[Wolfgang Bangerth]

> I think the melt postprocessor that calculates the melt fraction is already in
> the main version of aspect for a few months.

Yes.


> However, in the last weeks I
> tested it and it should now calculate the melt fraction exactly as in the
> sources of the melting models (to have a look at the p-T-diagrams, you can run
> the tests melt_postprocessor_peridotite and melt_postprocessor_pyroxenite).
>
> If there are no compositional fields, this postprocessor will visualize the
> melt fraction of peridotite (calculated using the anhydrous model of the Katz,
> 2003 paper). If there is at least one compositional field, the postprocessor
> assumes that the  first compositional field is the content of pyroxenite, and
> will visualize the melt fraction for a mixture of peridotite and pyroxenite
> (using the melting model of Sobolev, 2011 for pyroxenite).
> All the parameters that were used in these calculations can be changed in the
> input file, too, the most relevant maybe being the mass fraction of Cpx in
> peridotite in the Katz melting model ("Mass fraction cpx"), which right now
> has a default of 15%. (The code is implemented in
> source/postprocess/visualization/melt_fraction.cc.)

Would you mind adding essentially these two paragraphs to the documentation of 
the MeltFraction plugin at the bottom of melt_fraction.cc? I think this is 
actually a very useful description of what it does.


> In addition, to make the rheology dependent on melt fraction, we now have the
> possibility to access the calculation of a postprocessor from the material
> model. So to access the melt postprocessor, you could use the function
> "find_postprocessor ()" (in include/aspect/simulator_access.h) to get the
> postprocessor calculation. (Wolfgang, is this how it works? I never used it
> myself...).

Not quite. find_postprocessor() will find "normal" postprocessors, of which 
the visualization module is one. The visualization model has plugins itself, 
such as the melt fraction one. We would need another function that queries the 
visualization postprocessors but that should be a 5-minute task once someone 
needs it.

Best
  Wolfgang

-- 
------------------------------------------------------------------------
Wolfgang Bangerth               email:            bangerth at math.tamu.edu
                                 www: http://www.math.tamu.edu/~bangerth/