[MITgcm-support] "gluemnc grid.t*.nc" omits many variables

[Andy Hogg]

Hi Maarten

We also had problems with gluemnc, and have been developing a python script to do a similar job. It's still under development, but if you wanted to give it a try just let me know and I'll put a clean version (and brief instructions) together for you. 

(Anyone else interested in roadtesting this script, please drop me a line.)

Cheers
Andy

On 25/09/2010, at 5:47 AM, Ryan Abernathey wrote:

> Hi Maarten,
> 
> Originally the gluemnc script did not work with the grid files at all. I modified it last year in order to allow glueing of grid files, and it does work for me without producing the errors you described. The problem could depend on your particular versions of nco, perl, shell, etc.
> 
> Maybe if you could find the very first error produced, I might be able to give you some hints about why it is failing. (Many of the subsequent errors could be caused by one initial failure.) But since glueing netcdf grid files is not a very common practice (I'm the only person I know of who does it), you might have to just try debugging the script yourself. A good place to start for this is the nco user's guide:
> http://nco.sourceforge.net/nco.html
> 
> Good luck!
> 
> -Ryan
> 
> 
> 
> On Sep 23, 2010, at 3:13 PM, Maarten Buijsman wrote:
> 
>> Dear modelers
>> 
>> I use the pearl scripts gluemnc/xplodemnc to combine my netcdf tile files.
>> 
>> For example:  "gluemnc state.0000000000.t*.nc" produces a nice state.0000000000.glob.nc file that has the same variables as in the state.0000000000.t*.nc files.
>> 
>> However, when I do "gluemnc grid.t*.nc" I produce a grid.glob.nc file that only has variables RC RF RU RL drC and drF, while the grid.t*.nc files have variables Depth dxC dxF dxG dxV dyC dyF dyG dyU fCoriG fCori HFacC HFacS HFacW rA rAs rAw rAz R_low Ro_surf XC XG YC YG. Clearly many variables are omitted. Why?
>> 
>> By the way, "gluemnc *.nc" does not work either.
>> 
>> These are some error message examples:
>> ncpdq: ERROR unable to determine method for remote retrieval of Depth.grid.gloy.1018.nc
>> ncrcat: ERROR unable to determine method for remote retrieval of dyU.grid.gloy.*.nc
>> ncrcat: ERROR unable to determine method for remote retrieval of dyU.grid.glob.nc
>> 
>> Did other folks had the same issue and has it been addressed?
>> 
>> thanks!
>> 
>> Maarten Buijsman
>> 
>> --
>> ~~ ~~ ~~ ~~ ~~ ~~ ~~ ~~ ~~ ~~ ~~ ~~ ~~ ~~ ~~
>> Maarten Buisman (maarten.buijsman at noaa.gov)
>> Geophysical Fluid Dynamics Laboratory
>> 201 Forrestal Road
>> Princeton, NJ 08540-6649
>> Phone: (609) 452-5305
>> Fax: (609) 987-5063
>> 
>> 
>> 
>> 
>> 
>> 
>> _______________________________________________
>> MITgcm-support mailing list
>> MITgcm-support at mitgcm.org
>> http://mitgcm.org/mailman/listinfo/mitgcm-support
> 
> 
> _______________________________________________
> MITgcm-support mailing list
> MITgcm-support at mitgcm.org
> http://mitgcm.org/mailman/listinfo/mitgcm-support