[gradsusr] GFS grb2 to grb1 conversion issue

[Sam Wilson]

Hi,

I’m running into an issue with the resulting grib1 file produced from converting a long range (>=264 hr forecast) GFS grib2 file with the cnvgrib utility.

I’ve attached the initial grib2 file I’m working with as an example - this file contains only my variables of interest.

I want to convert this file to grib1, which I do using the following:

1.) wgrib2 gfslr2015111912hr384.grb2 -grib temp.grb2

2.) cnvgrib -g21 temp.grb2 gfslr2015111912hr384.grb

When I run wgrib on the file grib1 file, it gives the following output:

1:0:d=15111912:TMP:kpds5=11:kpds6=105:kpds7=2:TR=10:P1=1:P2=128:TimeU=1:2 m above gnd:384hr fcst:NAve=0
2:114114:d=15111912:RH:kpds5=52:kpds6=105:kpds7=2:TR=10:P1=1:P2=128:TimeU=1:2 m above gnd:384hr fcst:NAve=0
3:195648:d=15111912:UGRD:kpds5=33:kpds6=105:kpds7=10:TR=10:P1=1:P2=128:TimeU=1:10 m above gnd:384hr fcst:NAve=0
4:301616:d=15111912:VGRD:kpds5=34:kpds6=105:kpds7=10:TR=10:P1=1:P2=128:TimeU=1:10 m above gnd:384hr fcst:NAve=0
5:407584:d=15111912:APCP:kpds5=61:kpds6=1:kpds7=0:TR=4:P1=116:P2=128:TimeU=1:sfc:116-128hr acc:NAve=0
6:513552:d=15111912:CSNOW:kpds5=143:kpds6=1:kpds7=0:TR=3:P1=116:P2=128:TimeU=1:sfc:116-128hr ave:NAve=0
7:521780:d=15111912:CICEP:kpds5=142:kpds6=1:kpds7=0:TR=3:P1=116:P2=128:TimeU=1:sfc:116-128hr ave:NAve=0
8:530008:d=15111912:CFRZR:kpds5=141:kpds6=1:kpds7=0:TR=3:P1=116:P2=128:TimeU=1:sfc:116-128hr ave:NAve=0
9:538236:d=15111912:LFTX:kpds5=131:kpds6=1:kpds7=0:TR=10:P1=1:P2=128:TimeU=1:sfc:384hr fcst:NAve=0
10:611624:d=15111912:CAPE:kpds5=157:kpds6=1:kpds7=0:TR=10:P1=1:P2=128:TimeU=1:sfc:384hr fcst:NAve=0
11:717592:d=15111912:PWAT:kpds5=54:kpds6=200:kpds7=0:TR=10:P1=1:P2=128:TimeU=1:atmos col:384hr fcst:NAve=0
12:799126:d=15111912:TCDC:kpds5=71:kpds6=10:kpds7=0:TR=3:P1=116:P2=128:TimeU=1:atmos col:116-128hr ave:NAve=0
13:856224:d=15111912:PRMSL:kpds5=2:kpds6=102:kpds7=0:TR=10:P1=1:P2=128:TimeU=1:MSL:384hr fcst:NAve=0

I’ve put in bold the the issues I’m seeing - those variables are giving me undefined values when I try to plot them, which I assume is due to the above issue.

I think #43 at the following link describes this problem, so it seems to be a known issue:

http://ftp.cpc.ncep.noaa.gov/wd51we/wgrib2/tricks.wgrib2

I tried running the suggestion in #43 on my initial grib2 file, then ran cnvgrib but I get the same bad results with the grib1 file.

Is their another workaround to this problem I could try?

The versions of the utilities I’m running are below:

wgrib2 v0.1.9.4
wgrib v1.8.0.13f
cnvgrib-1.1.4

Thanks for any help you can provide.

Cheers,
Sam

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://gradsusr.org/pipermail/gradsusr/attachments/20151119/5c4cb45d/attachment-0001.html 
-------------- next part --------------
A non-text attachment was scrubbed...
Name: gfslr2015111912hr384.grb2
Type: application/octet-stream
Size: 593646 bytes
Desc: gfslr2015111912hr384.grb2
Url : http://gradsusr.org/pipermail/gradsusr/attachments/20151119/5c4cb45d/attachment-0001.obj