Change Core/Mantle boundary in 2D reference profile

ASPECT
1

Dear community,

I’m trying to produce a 2D model convection following the cookbook of BurnMan with a reference profile. But, for my case of study I need to study the parameter space of Core/Mantle ratios, which means that I need to change the Core/Mantle boundary in every case. How can I change this parameter in the Cookbook: 2D compressible convection with a reference profile and material properties from BurnMan ? Is there a direct/indirect way of introducing this? As the user’s input parameters just include the outer radii. Thanks a lot for your help.

Best,

Mariana V

2

Dear Mariana,

I think you misunderstood my earlier suggestion. I wanted to signpost you to the entropy method as an alternative to using the Burnman 1D profile, as the entropy method is a more advanced method for studying isochemical convection. I also intended to suggest that if you want to use that method, you give it a try, and then ask for help if you needed advice with a specific problem.

Hopefully my answer to your original question will help you get the Burnman coupling working if that is what you decide is the best method for your use case.

Best wishes,
Bob

1 Like
3

Dear Bob,

Thanks a lot for your answers. I found an ipython version in your github ( Create relaxed geodynamic 1D profile) that let me change the thickness of the layer easily and it worked, there are no convergence problems anymore! But, as I want to study the parameter space of core/mantle boundaries at different Rayleigh numbers for Earth-like exoplanets, I tried to create the mantle layer with the ‘BoundaryLayerPerturbation’ tool included on BurnMan that let me change these Ra values. Still, I need to create a new PerpleX table for this configuration. I will try this solution as a plan B. The entropy method you suggest now looks more suitable for what I need. I’m going to try it as well to compare both methods! Thanks a lot and sorry for the confusion.

Best,

Mariana V

4

Hi Mariana,

Glad to hear you found a way forward using my script, and no need to apologise!

I want to study the parameter space of core/mantle boundaries at different Rayleigh numbers for Earth-like exoplanets, [so] I tried to create the mantle layer with the ‘BoundaryLayerPerturbation’ tool included on BurnMan that let me change these Ra values

The ‘BoundaryLayerPerturbation’ tool in BurnMan uses an analytical expression to estimate average layer temperatures given a simple rheological model and layer thicknesses. It was parameterised using analogue experiments conducted in the 1970s and 1980s. It is a very useful “back of the envelope” calculation, and approximates what one might see from the results of a geodynamic simulation without having to run that simulation.

If you’re running geodynamic simulations, thermal boundary layers develop naturally as a result of your input conditions during the run - they need not and can not be imposed by user input.

Finally, ASPECT mostly takes material properties as input. Rayleigh number isn’t a material property (see the equations in various textbooks and on wiki). In my opinion, (approximate) Rayleigh numbers are best used (a) to avoid running simulations at all if the setup is simple enough, or the desired outputs are insensitive to model complexity, (b) figure out which parameter ranges might be interesting in your simulations, or (c) as a sanity or interestingness check after you have run your simulations.

I hope that’s useful!

Best wishes,
Bob

1 Like