[gradsusr] How to combine netcdf files.

Mon Dec 10 01:32:04 EST 2012

I find that in using the template options, in the dset, %x3 does not work
for me.
file name ua_Amon_CSIRO-Mk3-6-0_rcp45_r1i1p1_200601-202512.nc
              ua_Amon_CSIRO-Mk3-6-0_rcp45_r1i1p1_202601-204512.nc
dset ua_Amon_CSIRO-Mk3-6-0_rcp45_r1i1p1_%x3601-%x3512.nc this does not work
in diplaying data, it will say "Cannot contour grid - all undefined values"
dset ua_Amon_CSIRO-Mk3-6-0_rcp45_r1i1p1_%ch.nc however works.
chsub 1 240 200601-202512
chsub 241 480 202601-204512
TDEF 480 Linear 12:00z01jan2006 1mo

I do think that%x3601-%x3512 should be replaced with %y401-%y412. This does
not work as well. Does this mean in using the substitution template like
%x, %y and others, the increment must be 1? Since using %ch is considered
as string, it works well with 20 years increment in my files name, I just
considered "200601-202512" and "200601-202512" as strings.

To Jeff, you mean the dtype does not need change. dset need to be changet
to reflect the filename increment.

Yunus

On Fri, Dec 7, 2012 at 1:27 PM, Jeff Duda <jeffduda319 at gmail.com> wrote:

> If you want the data in one file you'll have to use CDO or other netCDF
> tools using Fortran or some other programming language.  You can get the
> same effect by using templating.  You don't need to change the DSET entry.
>  It should be just netcdf.  I'm not 100% the following will work, but it's
> worth a try because it won't take much time.  You'll have to determine
> exactly how many time indices are contained in each individual file and
> keep track of them, because you'll need them for the CHSUB entries.  It
> looks like you'll have 4 or 5 entries to write.  You'll want the text of
> the substitution string to match what is in the files.  The middle two
> numbers should reprsent the time indices in your control file that
> correspond to the times in the data.  These indices will depend on your
> TDEF entry.  It looks like you are dealing with monthly data, so your TDEF
> entry should be something like
> TDEF XXX LINEAR 00Z01JAN2010 1MO,
> where XXX is the number of months in the combined files.  If you're
> looking at 90 years, then this number will probably need to be 90*12+1 =
> 1081? (mental math...check it).  Once your TDEF line is specified, the
> CHSUB entries should look something like
> chsub t1 t2 200601-202512
> chsub t2 t3 202601-204512
> etc., where t1, t2, and t3 mark specific time indices that mark the start
> and end time indices within each file.  t1 is the first time index, so
> t1=1.  t2 should be the last time index found in the
> file a_Amon_CSIRO-Mk3-6-0_rcp45_r1i1p1_200601-202512.nc.  If this spans 18
> years, then the last time index is probably 12*18+1 = 217.  The next data
> file (in temporal order) will start with the next time index, 218, and end
> with the additional number of time indices in that specific file.  If that
> file is ua_Amon_CSIRO-Mk3-6-0_rcp45_r1i1p1_202601-204512.nc, then you
> probably have 218+20*12+1 = 459.  Repeat this until you reach the end of
> the files.  As a help, the last time index you reach by doing this process
> will be the value of XXX you should enter in your TDEF entry.  Don't forget
> to add "options template" to your control file.  Try this and see if it
> works.
>
> Jeff Duda
>
> On Thu, Dec 6, 2012 at 7:57 PM, Muhammad Yunus Ahmad Mazuki <
> ukm.yunus at gmail.com> wrote:
>
>> If I want to use the template option, for example file name
>> "ua_Amon_CSIRO-Mk3-6-0_rcp45_r1i1p1_200601-202512.nc",
>> "ua_Amon_CSIRO-Mk3-6-0_rcp45_r1i1p1_202601-204512.nc",
>> "ua_Amon_CSIRO-Mk3-6-0_rcp45_r1i1p1_204601-206512.nc" and
>> "ua_Amon_CSIRO-Mk3-6-0_rcp45_r1i1p1_206601-208512.nc",
>> "ua_Amon_CSIRO-Mk3-6-0_rcp45_r1i1p1_208601-210012.nc", how do I configure
>> the DSET, and CHSUB? If possible, I actually just wanted 2010-2100 in one
>> file.
>>
>> On Wed, Dec 5, 2012 at 1:12 PM, Furrukh Bashir <furrukhbashir at gmail.com>wrote:
>>
>>> Dear Fahim Ahmed
>>> You may use NCO as well.
>>> http://linux.die.net/man/1/ncrcat
>>>
>>> Regards
>>>
>>>
>>> *FURRUKH BASHIR*
>>> *Pakistan Meteorological Department
>>> Research and Development Division
>>> Islamabad, Pakistan
>>> Mob: 0321-5876770
>>> Office Phone: +92 51 9250334*
>>>
>>>
>>>
>>> On Wed, Dec 5, 2012 at 1:55 PM, NAGARAJU CHILUKOTI <
>>> chilukotinagaraju at gmail.com> wrote:
>>>
>>>> Hi !!!
>>>>
>>>>  you can do with CDO climate data operators (CDO)
>>>>
>>>>      with cdo mergetimes command
>>>>
>>>>    full details you can find here
>>>> https://code.zmaw.de/projects/cdo/wiki/Cdo#Documentation
>>>>
>>>> I hope this will solve your problem
>>>>
>>>>
>>>> With Cheers
>>>>
>>>> On Wed, Dec 5, 2012 at 8:10 AM, fahim ahmad <fahimpmd at yahoo.com> wrote:
>>>>
>>>>> Hi I have downloaded Meteorological Forcing dataset (Daily) from NCEP
>>>>> in order to run CLM (Community Land Model). Dataset include 7 different
>>>>> variables, for each year one file. I want to combine data in monthly files.
>>>>> for example in the file 1985-12.nc i want to combine daily data of
>>>>> all the 7 variables. Using grads how can i do this. or if any one has
>>>>> Fortran script??
>>>>> Thank you very much.
>>>>>
>>>>> **************************************
>>>>> Muhammad FAHIM Ahmad,
>>>>> PhD Scholar,
>>>>> Nanjing Univ. of Info. Science & Tech.
>>>>> Nanjing, China P.R.
>>>>> Cell# +8615715191564.
>>>>> **************************************
>>>>> _______________________________________________
>>>>> gradsusr mailing list
>>>>> gradsusr at gradsusr.org
>>>>> http://gradsusr.org/mailman/listinfo/gradsusr
>>>>>
>>>>>
>>>>
>>>>
>>>> --
>>>>  Nagaraju Chilukoti
>>>>  Computational Earth Science Group,
>>>>  Centre for Development of Advanced Computing (CDAC)
>>>>  Pune University Campus, Ganeshkhind
>>>>  Pune 411007
>>>>  *020- 25704226*
>>>>  Mob.* 08087384939*
>>>>  nagarajuc at cdac.in <http://goog_932359822>
>>>>  chilukotinagaraju at gmail.com
>>>>
>>>> _______________________________________________
>>>> gradsusr mailing list
>>>> gradsusr at gradsusr.org
>>>> http://gradsusr.org/mailman/listinfo/gradsusr
>>>>
>>>>
>>>
>>> _______________________________________________
>>> gradsusr mailing list
>>> gradsusr at gradsusr.org
>>> http://gradsusr.org/mailman/listinfo/gradsusr
>>>
>>>
>>
>> _______________________________________________
>> gradsusr mailing list
>> gradsusr at gradsusr.org
>> http://gradsusr.org/mailman/listinfo/gradsusr
>>
>>
>
>
> --
> Jeff Duda
> Graduate research assistant
> University of Oklahoma School of Meteorology
> Center for Analysis and Prediction of Storms
>
>
> _______________________________________________
> gradsusr mailing list
> gradsusr at gradsusr.org
> http://gradsusr.org/mailman/listinfo/gradsusr
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://gradsusr.org/pipermail/gradsusr/attachments/20121210/7ea6c2b8/attachment-0003.html 