[CIG-MC] CIG-MC Digest, Vol 79, Issue 1

Magali Billen mibillen at ucdavis.edu
Mon Nov 25 11:36:33 PST 2013 

Hi Mike,
Thanks for correcting me on the published work - I knew about Eh Tan's early paper, but not the others.

Now that you also are using Rhea, which has a more flexible grid structure where you can put in
higher resolution where you need it, do you anticipate using the coupling feature again? I know the 
original intent of creating the coupling framework was to couple codes with different physics (e.g.,
a visco-elastic lithospheric code with a viscous mantle code). Do you know of any groups using it in 
that way?  Its important for the CIG MC group to consider all the implications of removing the pyre
layer and the ability to couple codes in the future. I don't know enough about this to weigh in beyond
asking questions.

Magali



On Nov 24, 2013, at 1:24 PM, Mike Gurnis wrote:

> Hi Magali,
> 
> These are all good points but I did want to correct the point that there are no known published work using coupled codes with CitcomS. We had three papers, a plume study and two subduction studies:
> 
> http://www.gps.caltech.edu/~gurnis/Papers/2006_Tan_etal_G3.pdf
> 
> http://www.gps.caltech.edu/~gurnis/Papers/2011_DiCaprio_etal_Lithosphere.pdf
> 
> http://www.gps.caltech.edu/~gurnis/Papers/2011_Matthews_etal_GondRes.pdf
> 
> We are currently not using the coupler in my group.
> 
> Mike
> 
> 
> 
> On 11/22/13 9:24 AM, Magali Billen wrote:
>> Thorsten and Katie bring up good points. I am more concerned about their being good cookbooks
>> available, then about coupling codes. 
>> 
>> First, there are known issue with cookbooks (some only work  as shown with older version of CitcomS), 
>> so this would be a good opportunity to update these. I suggest soliciting very well documented contributions 
>> from members of this mailing list, which they use when starting their students on problems.
>> 
>> Second, I don't know of any published work using coupled codes (please correct me if I'm wrong). While this was demonstrated to work for some simple models and was a heroic effort for those who made it work, its hard to get to work in practice for more complicated problems. For example, if you have higher resolution for a CitcomS model embedded inside a larger CitcomS models with lower resolution - any thermal boundary layer ends up creating thermal instabilities at the sides of the box due to the mismatch in the thickness). If someone is using the coupler, then we need to keep this capability, but if after this much time no one has used it, that is telling us something important.
>> 
>> -Magali
>> 
>> 
>> On Nov 21, 2013, at 9:15 PM, Catherine Cooper wrote:
>> 
>>> 
>>> I am also concerned about this development.  The cookbooks provided with the manual will only work if the pyre version of CitcomS is installed.  Will there be new cookbooks developed?  Or will they too no longer be supported?  These are valuable learning tools for new users including graduate students.  It would be a shame if there were no longer educational examples for new users to safely learn on.
>>> 
>>> Thanks,
>>> 
>>> Katie
>>> 
>>> 
>>> Catherine Cooper, Ph.D.
>>> Assistant Professor in Geophysics
>>> School of the Environment
>>> Washington State University
>>> (509)335-1501
>>> cmcooper at wsu.edu
>>> 
>>> 
>>> On Nov 21, 2013, at 9:03 PM, Thorsten Becker wrote:
>>> 
>>>> Hi
>>>> 
>>>> While I am not a python/pyre user, I wonder how this affects the abilities to couple regional and global codes, for example? This is another extension I am not using, but this always seemed very helpful potentially? Wouldn't this be useful to maintain?
>>>> 
>>>> When I compile CitcomS, I simply use the --nopyre option or whatever it's called, and this seems to get rid off many of the issues mentioned above?
>>>> 
>>>>  I can see how this must have all been discussed but I am not aware of the reasoning, sorry, so please excuse if these are naive questions. 
>>>> 
>>>> Thanks!
>>>> 
>>>> Thorsten
>>>> 
>>>> 
>>>> 
>>>> Thorsten W Becker - USC     
>>>> 
>>>> 
>>>> On Thu, Nov 21, 2013 at 12:00 PM, <cig-mc-request at geodynamics.org> wrote:
>>>> Send CIG-MC mailing list submissions to
>>>>         cig-mc at geodynamics.org
>>>> 
>>>> To subscribe or unsubscribe via the World Wide Web, visit
>>>>         http://geodynamics.org/cgi-bin/mailman/listinfo/cig-mc
>>>> or, via email, send a message with subject or body 'help' to
>>>>         cig-mc-request at geodynamics.org
>>>> 
>>>> You can reach the person managing the list at
>>>>         cig-mc-owner at geodynamics.org
>>>> 
>>>> When replying, please edit your Subject line so it is more specific
>>>> than "Re: Contents of CIG-MC digest..."
>>>> 
>>>> 
>>>> Today's Topics:
>>>> 
>>>>    1. CitcomS Python Removal (Eric Heien)
>>>> 
>>>> 
>>>> ----------------------------------------------------------------------
>>>> 
>>>> Message: 1
>>>> Date: Thu, 21 Nov 2013 10:47:06 -0800
>>>> From: Eric Heien <emheien at ucdavis.edu>
>>>> To: CIG-MC at geodynamics.org
>>>> Subject: [CIG-MC] CitcomS Python Removal
>>>> Message-ID: <00B7E076-EC0E-4E5B-AA20-4DADB9498AFA at ucdavis.edu>
>>>> Content-Type: text/plain; charset=windows-1252
>>>> 
>>>> Hello,
>>>> 
>>>> The CitcomS developers have recently decided to remove Python and Pyre support from CitcomS.  There are several reasons for this:
>>>> 
>>>> - The Python interface in CitcomS is not compatible with Python 2.7 or above.  Python 2.7 was released in 2010, meaning most systems set up in the last 3 years won?t support Python in CitcomS.
>>>> - The interface requires downloading extra packages - this happens automatically but can cause compilation failure on some platforms.
>>>> - The interface opaquely assigns some parameters with values not controlled by the user.
>>>> - The Python interface adds a level of complexity making it harder to debug the program and understand how data structures are used/modified during a run.
>>>> - Many of the major CitcomS users do not use the Python interface.
>>>> 
>>>> We plan to remove this functionality from CitcomS and provide tools to convert Python based input files to the normal input file format.  Relevant logic that was in the Python framework will be moved into the main CitcomS code.  This change will be implemented over the coming year and incorporated into a future CitcomS release.
>>>> 
>>>> If you have any concerns about this decision or would like to weigh in, feel free to contact me or the cig-mc mailing list.
>>>> 
>>>> Thank you,
>>>> 
>>>> -Eric
>>>> 
>>>> 
>>>> 
>>>> ------------------------------
>>>> 
>>>> _______________________________________________
>>>> CIG-MC mailing list
>>>> CIG-MC at geodynamics.org
>>>> http://geodynamics.org/cgi-bin/mailman/listinfo/cig-mc
>>>> 
>>>> End of CIG-MC Digest, Vol 79, Issue 1
>>>> *************************************
>>>> 
>>>> 
>>>> 
>>>> _______________________________________________
>>>> CIG-MC mailing list
>>>> CIG-MC at geodynamics.org
>>>> http://geodynamics.org/cgi-bin/mailman/listinfo/cig-mc
>>> 
>>> _______________________________________________
>>> CIG-MC mailing list
>>> CIG-MC at geodynamics.org
>>> http://geodynamics.org/cgi-bin/mailman/listinfo/cig-mc
>> 
>> 
>> 
>> _______________________________________________
>> CIG-MC mailing list
>> CIG-MC at geodynamics.org
>> http://geodynamics.org/cgi-bin/mailman/listinfo/cig-mc
> 
> _______________________________________________
> CIG-MC mailing list
> CIG-MC at geodynamics.org
> http://geodynamics.org/cgi-bin/mailman/listinfo/cig-mc

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://geodynamics.org/pipermail/cig-mc/attachments/20131125/cc56dbfb/attachment-0001.html>

More information about the CIG-MC mailing list