[MITgcm-devel] PLOT_FIELD_XY in seaice_do_diags.F

[Jean-Michel Campin]

Hi Martin,

Regarding the seaice time-ave code, I don't know what users are doing
so i would tend to be cautious. I was just curious about the 5 index 
arrays for all-2D fields.

But regarding PLOT_FIELD_XY, I would propose:
-to keep the few PLOT_FIELD_XY that we have early-on in initialisation stage:
 ini_curvilinear_grid.F, ini_depths.F, ini_masks_etc.F
 (here it could be usefull to see if something is wrong early in the
  initialisation), 
-call them only if debugMode=T in ini_salt.F & ini_theta.F (as it is
 in do_the_model_io.F)
-remove the commented one in ini_forcing.F, ini_cg2d.F (update_masks_etc.F ?) 
-and in the long term, remove them from all pkgs, starting with 
 seaice_do_diags.F (we could keep some of them in pkg initialisation,
 specially to print various masks ?).
Comments ?

And for the renaming, might just do all of them, i.e., all
the pkg/*_do_diags.F which are called from do_the_model_io.F 
and do not contain any pkg/diagnostics stuff (normally, they 
should not).

Cheers,
Jean-Michel

On Mon, Jun 15, 2009 at 09:10:47AM +0200, Martin Losch wrote:
> Hi Jean-Michel, Dimitris,
>
> most of these questions go directoy to Dimitris. I hardly ever look at  
> the plot_field_xy output, and certainly never (so far) in the seaice  
> output. As far as I am concerned you can remove these calls and rename  
> the subroutine.
>
> I don't know about the tave-fields. I have never touched them (not even 
> looked at them). Maybe it's time to retire the tave-pkg altogether? E.g. 
> put a stop statement somewhere into to code saying:
> "you are using outdated code. all of the functionality has been moved to 
> pkg/diagnostics. pkg/tave will be removed in the near future", keep this 
> for a while (maybe a few months) and then get rid off pkg/tave ?
>
> M.
>
>
>
> On Jun 12, 2009, at 4:50 PM, Jean-Michel Campin wrote:
>
>> Hi Martin,
>>
>> I was going to simplify the MDS part of seaice_do_diags.F
>> (by the way, can I rename it "seaice_output.F, to clarify that
>> it has nothing to do with the diagnostics pkg ?) removing
>> the temp arr copy (since we have pkg/rw routine that
>> allow to write 1 level among several).
>>
>> Do you really use the PLOT_FIELD_XY output ? Personally, most of the
>> time, when I have a problem, it's always to track a relative diff <  
>> 10^-3
>> and the alphabet is not helpful. The only reason why I could look
>> at the PLOT_FIELD_XY output would be for the overlap, but in
>> seaice_do_diags.F they are reset by EXCH call just before the plot.
>> I can either remove all the PLOT_FIELD_XY from seaice_do_diags.F
>> (so that I don't need the temp array anymore, except for MNC), or
>> keep them (much more coding), but then I will not check that the
>> changes are right for those PLOT_FIELD output.
>>
>> other unrelated question:
>> Why all the 2D seaice time-ave fields are stored in 5 indices arrays ?
>> (FUtave, EmPmRtave, QNETtave ...)
>>
>> Cheers,
>> Jean-Michel
>>
>> _______________________________________________
>> MITgcm-devel mailing list
>> MITgcm-devel at mitgcm.org
>> http://mitgcm.org/mailman/listinfo/mitgcm-devel
>
> _______________________________________________
> MITgcm-devel mailing list
> MITgcm-devel at mitgcm.org
> http://mitgcm.org/mailman/listinfo/mitgcm-devel