Re: NCL does not handle GRIB2 boustrophedonic scanning

Hi Richard,

I have to admit that I have never heard the word ' boustrophedonic'
before. Thanks for the addition to my vocabulary.
However, the issue of GRIB2 data that alternates direction along one
of the dimensions was addressed shortly after the release of version
5.0.0. I believe that your example data is handled properly using the
current development version of NCL. If you would like us to create a
binary that contains the fix, please contact Mary Haley with your
system particulars.
  -dave

On Jun 11, 2008, at 1:02 PM, Mary Haley wrote:

> Hi Richard,
>
> Just FYI, your original message came through just fine. Dave Brown
> has been looking at your problem and will probably report back to
> ncl-talk when he knows more.
>
> Thanks for including your GRIB2 file.
>
> --Mary
>
> On Wed, 11 Jun 2008, Richard Carpenter wrote:
>
>> [It looks like the text of my message earlier today was scrubbed,
>> so I am resending without the attachment.]
>>
>> Hello,
>>
>> It appear that NCL does not handle a complex packing method called
>> boustrophedonic scanning in GRIB2 files. This leads to a "venetian
>> blinds" effect when the attached file is plotted using NCL 5.0.0.
>> IDV and GEMPAK both display the data properly. A workaround for
>> NCL is given below.
>>
>> dimms = dimsizes(array)
>> do j=1,dimms(0)-1,2
>> array(j,:) = array(j,::-1)
>> end do
>>
>> Thank you.
>>
>> Richard Carpenter
>> Weather Decision Technologies, Inc.<http://www.wdtinc.com/>
>>
>>
>>

_______________________________________________
ncl-talk mailing list
ncl-talk_at_ucar.edu
http://mailman.ucar.edu/mailman/listinfo/ncl-talk
Received on Wed Jun 11 2008 - 13:18:35 MDT