[mpas-developers] iro for tracers ….. bug or feature?

[Michael Duda]

This is behavior that we could look at changing, especially if it
would be of benefit to MPAS-O. In the atmosphere, for example, we
currently write out 8 moisture species and number concentrations,
regardless of the microphysics scheme that's being used in the
run; so, we could save ourselves from writing out (and reading in)
3d arrays with nothing in them.

Michael


On Fri, Oct 26, 2012 at 10:19:49AM -0600, Mark Petersen wrote:
> That is correct.  I discovered this about a year ago, and it is just 
> that arrays within a super array need to all behave the same way for the 
> 'iro' settings.
> 
> Mark
> 
> On 10/26/12 09:47, Doug Jacobsen wrote:
> >Hey Todd,
> >
> >I think this is because super arrays can't have mixed input/output 
> >states. It's all or nothing, because they are all packaged into a 
> >single field type.
> >
> >It most likely uses the last tracer's iro values to determine if it's 
> >input/output/restart. So, I bet if you took the o off salinity and 
> >then put temperature below salinity you would get them both written out.
> >
> >It seems like this is more like a feature than a bug, because of the 
> >way super arrays currently work. Super arrays are a bit more 
> >restrictive in terms of what you can and can't do with them, but they 
> >are more useful than having all of the tracers be their own fields.
> >
> >Doug
> >
> >On Fri, Oct 26, 2012 at 9:25 AM, Laura Fowler <laura at ucar.edu 
> ><mailto:laura at ucar.edu>> wrote:
> >
> >    Hi Todd:
> >    I removed a bunch of outputs from the core_nhyd_atmos Registry to
> >    start
> >    long Regional Climate Simulations with mesh refinements (as you did).
> >    I'll double check to see if I am getting the same bug or feature !
> >    Cheers,
> >    Laura
> >
> >    On 10/26/12 8:31 AM, Todd Ringler wrote:
> >    > possible bug report ....
> >    >
> >    > I am using the ocean core on the trunk. I changed Registry to
> >    remove the output of salinity
> >    >
> >    > var persistent real    temperature ( nVertLevels nCells Time ) 2
> >    iro temperature state tracers dynamics
> >    > var persistent real    salinity ( nVertLevels nCells Time ) 2 ir
> >     salinity state tracers dynamics
> >    >
> >    > and to my surprise neither salinity nor temperature was written
> >    to output! These are the only two fields packed in the tracers
> >    super structure.
> >    >
> >    > If I turn back on salinity output with iro, I then get both
> >    fields in the output.
> >    > _______________________________________________
> >    > mpas-developers mailing list
> >    > mpas-developers at mailman.ucar.edu
> >    <mailto:mpas-developers at mailman.ucar.edu>
> >    > http://mailman.ucar.edu/mailman/listinfo/mpas-developers
> >
> >
> >    --
> >    !----------------------------------------------------
> >    Laura D. Fowler
> >    Mesoscale and Microscale Meteorology Division (MMM)
> >    National Center for Atmospheric Research
> >    P.O. Box 3000, Boulder CO 80307-3000
> >
> >    e-mail: laura at ucar.edu <mailto:laura at ucar.edu>
> >    phone : 303-497-1628 <tel:303-497-1628>
> >
> >    !----------------------------------------------------
> >
> >
> >    _______________________________________________
> >    mpas-developers mailing list
> >    mpas-developers at mailman.ucar.edu
> >    <mailto:mpas-developers at mailman.ucar.edu>
> >    http://mailman.ucar.edu/mailman/listinfo/mpas-developers
> >
> >
> >
> >
> >_______________________________________________
> >mpas-developers mailing list
> >mpas-developers at mailman.ucar.edu
> >http://mailman.ucar.edu/mailman/listinfo/mpas-developers

> _______________________________________________
> mpas-developers mailing list
> mpas-developers at mailman.ucar.edu
> http://mailman.ucar.edu/mailman/listinfo/mpas-developers