[aspect-devel] Initial mesh with heterogeneous mesh-spacing

[Max Rudolph]

Try using the 'Minimum refinement function' criterion for the initial
adaptive refinement.

On Mon, Jun 25, 2018 at 1:26 AM Ravi Kanda <kanda.vs.ravi at gmail.com> wrote:

> Hi all,
>
> I am trying to figure out how to define an initial mesh configuration
> such that regions near the boundary have coarser resolution than the
> interior (2D cartesian only, at the moment). Specifically, I want an
> interior sub-domain (whose top surface is at the free-surface) having a
> much higher resolution - crudely illustrated by the attached JPG.  Note
> that I intend this finer resolution domain to be a purely
> geometric/geographic construct (e.g., corresponding to higher resolution
> geophysical data that I may want to compare to later).  The subdomain
> does not necessarily correspond to any specific state-variable such as
> composition, density temperature, or velocity (/strain) gradients.  This
> initial mesh may be further refined by other strategies (temperature,
> strain-rate, composition, etc.) at subsequent time-steps, but I want the
> model to start with, this initial heterogeneous mesh configuration as a
> minimum starting resolution of the entire domain.
>
> I went through Appendix A of the Aspect 2.0 manual describing the PRM
> runtime parameters, specifically, "Geometry model", and
> "Discretization", and ran some text searches within the document for
> relevant keywords.  I see you can set a uniform discretization (number
> of divisions) along each model dimension (x/y-repetitions), but not sure
> how to vary the nodal frequency as a function(x,y). Given I haven't yet
> done a thorough read of the manual, it is likely I may have missed it .
>
> I will really appreciate any suggestions or tips (or even where to look).
>
> Thanks,
> Ravi.
>
> --
> ---------------------------------------------------------------------
> Ravi Kanda
> Research Fellow
> Department of Geology
> Utah State University
> 4505 Old Main Hill
> Logan UT 84322-4505
> --------------------------
> Web Page: https://ravi-vs-kanda.github.io
>
> ----------------------------------------------------------------------
> For a human being, the unexamined life is not worth living - SOCRATES
> ----------------------------------------------------------------------
>
> _______________________________________________
> Aspect-devel mailing list
> Aspect-devel at geodynamics.org
> http://lists.geodynamics.org/cgi-bin/mailman/listinfo/aspect-devel
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.geodynamics.org/pipermail/aspect-devel/attachments/20180625/1c47b149/attachment.html>