Re: Proposed ncl_convert2nc modification

Dave Brown will comment further on NCL's names and why they changed.
Basically, when NCEP did not prescribe names, NCL made them up.
Subsequently, NCEP specified the name. NCL always goes with the
operational center specifies if a name is available.

========== Some general info =======================

When NCL performs the conversion via "ncl_convert2nc" it *always*
adds global attributes. One is the "NCL_Version"

One of the 'dumps' was created with v4.3.1 which was
released August 2007. This is roughly 2.5 years old.

// global attributes:
                :creation_date = "Wed Feb 10 13:44:16 MST 2010" ;
                :NCL_Version = "4.3.1" ;
                :system = "Linux gtgn 2.6.18-6-amd64 #1 SMP Fri Aug 21 14:53:35 UTC
2009 x86_64 GNU/Linux" ;
                :conventions = "None" ;
                :grib_source = "20100115_i15_f000_WRF-RR.grb2" ;
                :title = "NCL: convert-GRIB-to-netCDF" ;

=====================
The other was create with Version 5.1.0 released in March 2009

// global attributes:
                :creation_date = "Fri Apr 2 11:02:18 MDT 2010" ;
                :NCL_Version = "5.1.0" ;
                :system = "Linux finch 2.6.26-2-amd64 #1 SMP Tue Jan 12 22:12:20 UTC
2010 x86_64 GNU/Linux" ;
                :conventions = "None" ;
                :grib_source = "20100115_i15_f000_WRF-RR.grb2" ;
                :title = "NCL: convert-GRIB-to-netCDF" ;
}
======================

FYI: The current release is: Version 5.1.1 - June 16, 2009
        http://www.ncl.ucar.edu/current_release.shtml

======
There is a beta 5.2.0 already available.

http://www.ncl.ucar.edu/future_release.shtml

On 04/06/2010 11:29 AM, Julie Prestopnik wrote:
> Hi Dave and Dennis. Thank you for the information. One of my
> co-workers indicated that she previously saw TURBKE_P0_L104_GLC0 for
> Turbulent Kinetic Energy, but later saw TKE_P0_L104_GLC0 for Turbulent
> Kinetic Energy for the same grib file (20100115_i15_f000_WRF-RR.grb2).
>
> I have attached the header information for both files.
>
> Both files were converted using ncl_convert2nc. Unfortunately, the file
> with TURBKE_P0_L104_GLC0 was generated two days before that machine was
> upgraded so I cannot determine which version of NCL was used or what
> uname -a would have output at the time. It may be possible that this is
> the result of a version upgrade.
>
> The version used to generate the file with TKE_P0_L104_GLC0 is 5.1.0 and
> the output from uname -a is: Linux finch 2.6.26-2-amd64 #1 SMP Tue Jan
> 12 22:12:20 UTC 2010 x86_64 GNU/Linux
>
> I have been told that there were other instances where this has occurred
> and need to gather more information from the people who have experienced
> this issue.
>
> Thank you for your help.
>
> Julie
>
> David Brown wrote:
>>
>> Hi Julie,
>> We need more information about what you are asking. And as Dennis
>> suggested, it would be helpful to have some sample files. NCL adds
>> information as suffixes to the raw parameter names that come from
>> various versions of the GRIB parameter tables based on the information
>> found in the individual GRIB records. The suffixes are needed to ensure
>> the uniqueness of each variable name as presented in the NCL context.
>> Note that these variables are actually aggregations along level and time
>> of individual GRIB records. It is certainly possible to provide your own
>> parameter table to give any name you would like as parameter names, but
>> the rest of the name (from the first underscore character to the end of
>> the name) is needed to keep the data properly separated into unique
>> individually accessible variables. So we need to know what you actually
>> mean when you say the "field names change". Possibly you are getting an
>> incorrect version of the parameter table. If so it could be a problem
>> with the NCL code.
>> -dave
>>
>> On Apr 5, 2010, at 7:28 PM, jpresto@ucar.edu wrote:
>>
>>> Hello. I have noticed when using ncl_convert2nc to convert WRFRR grib
>>> files to netCDF format, sometimes the field names change. I would
>>> like to
>>> have the field names stay the same. It would be nice if ncl_convert2nc
>>> had the following capability: if the input grib names were known, I could
>>> have a configuration file with the field names I would like to have
>>> ncl_convert2nc output. Is this something that the maintainers of
>>> ncl_convert2nc would be interested in modifying the code to do?
>>>
>>> Thanks,
>>> Julie Prestopnik
>>>
>>> Software Engineer
>>> Research Applications Laboratory
>>>
>>>
>>> _______________________________________________
>>> ncl-talk mailing list
>>> List instructions, subscriber options, unsubscribe:
>>> http://mailman.ucar.edu/mailman/listinfo/ncl-talk
>
>
>
> _______________________________________________
> ncl-talk mailing list
> List instructions, subscriber options, unsubscribe:
> http://mailman.ucar.edu/mailman/listinfo/ncl-talk

-- 
======================================================
Dennis J. Shea                  tel: 303-497-1361    |
P.O. Box 3000                   fax: 303-497-1333    |
Climate Analysis Section                             |
Climate & Global Dynamics Div.                       |
National Center for Atmospheric Research             |
Boulder, CO  80307                                   |
USA                        email: shea 'at' ucar.edu |
======================================================
_______________________________________________
ncl-talk mailing list
List instructions, subscriber options, unsubscribe:
http://mailman.ucar.edu/mailman/listinfo/ncl-talk