[CIG-MC] velocity BCs

[Craig O'Neill]

Dear all,

Thanks very much for all the swift replies. Apologies for any unwanted
cross-postings, but I thought the resolution of this discussion might be of
broader interest.

So Tan, it seems the GPlates format is naturally v_colat, v_long so no
switch of the velocities required.

Thanks very much for the link Mark.

I encountered a similar wall to Rene, that only global mesh caps could be
exported. I actually wrote my own 'regional' gpml mesh cap file, but it was
not recognized as a mesh as such by GPlates (imported the gpml as velocities
or something).

The workaround I used was to construct a perl script which uses some GMT
calls, to grid the "global" exported velocity files, then re-sample them at
the desired regional grid points.

I've posted it here for those interested:
http://www.gemoc.mq.edu.au/~coneill/software/gpml2citcomR.pl

This seems to be working for me.
Any comments or corrections welcome!

cheers

-Craig.



On 29 June 2011 19:24, Mark Turner <mturner at gps.caltech.edu> wrote:

>
> hi,
>
> The format is the same for both the regional and global
> citcoms mesh input files if you are using .gpml.
>
> You can use GPlates to generate a set of global cap files
> with a small number of nodes per side of the cap,
> something like 3.
>
> Examine those files to see the .gpml format,
> and the ordering of the xml tags and such.
>
> Then create a .gpml citcoms input mesh file
> with your own regional mesh coordinates
> using the values in latitude and longitude.
>
> GPlates will read this data in as a single regular feature,
> with a multi-point geometry with your coordinates.
>
>
> Then the velocity generation process with in GPlates
> is the same for the global case.
>
>
> Yet, I'm not sure if that python script will work with
> a regional case.  I'll check on that tomorrow.
>
>
>
>
>  is there any gplates version, that can create a regional mesh? In my
>> version this checkbox is unselectable and gplates is referring to possible
>> future versions.
>>
>
> yes, possibly ...
> but, I'm not sure if we fixed that in the next public release 1.1.1 ... ?
>
> For the last few years our focus has been with global model runs,
> so there are still gaps in the regional processing we need to fix up.
> Thanks for helping me figure out what needs work!
>
> Mark
>
>
>


-- 
Dr Craig O'Neill
ARC Future Fellow/Senior Lecturer
CCFS - Core to Crust Fluid Systems ARC Centre of Excellence
GEMOC ARC National Key Centre
Department of Earth & Planetary Science
Macquarie University, E7A, Rm 515
Sydney, NSW, 2109, Australia

Phone: +61 2 9850 9673
http://www.gemoc.mq.edu.au/~coneill/
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://geodynamics.org/pipermail/cig-mc/attachments/20110630/9bcf645e/attachment.htm