lats4d: Trouble in generating correct center name and code

[Arlindo da Silva]

On 10/10/07, Ioannis A. Vamvakas <giavvns at ath.hcmr.gr> wrote:
>
> Hello,
>
> I've been using lats4d to convert a netCDF file into grib1.
> Everything works just fine but the Center name/id.


Do you mean the originating or the processing center?

It won't
> see the name I provide in the lookup table and will use the
> default name (=pcmdi) and code (=100).
>
> The command I use to do this is:
>
> lats4d -i <file>.nc -o <file> -format grib -table mytable
>
> where, 'mytable', is a copy of '.grads.lats.table' with a
> line at the center section added as follows:
>
> hcmr | 96 | 0 | 0


The description of the center record in the lats table reads:

# The format of each record is:
#   center | GRIB_id | GRIB_center | GRIB_subcenter
#
# center = mnemonic for the center
# GRIB_id = GRIB generating process id (PDS octet 6)
# GRIB_center = the id of center managing the data (for AMIP II this is
PCMDI) -
 see GRIB Table 0
# GRIB_subcenter = the id of the subcenter


"0" is not a valid grib center; I believe this is being changed into 100.
Try something like this:

hcmr | 96 | 96 | 2

Here is what "wgrib -V" reports:

rec 190:1133616:date 1987010500 DIST kpds5=8 kpds6=100 kpds7=100
levels=(0,100) grid=255 100 mb anl: bitmap: 3312 undef
  DIST=Geometric height [m]
  timerange 0 P1 0 P2 0 TimeU 2  nx 72 ny 46 GDS grid 0 num_in_ave 0 missing
0
  center 96 subcenter 2 process 96 Table 128 scan: WE:NS winds(N/S)
  latlon: lat  -90.000000 to 90.000000 by 4.000000  nxny 3312
          long 0.000000 to 355.000000 by 5.000000, (72 x 46) scan 2 mode 128
bdsgrid 1
  min/max data 3.40282e+38 -3.40282e+38  num bits 16  BDS_Ref 3.40282e+38
DecScale 0 BinScale -32766

  Good Luck,

--
Arlindo da Silva
dasilva at alum.mit.edu
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://gradsusr.org/pipermail/gradsusr/attachments/20071011/dc6b7397/attachment.html