[netCDFJava #RXE-920332]: Grib2NetCDF incorrectly converting time durations

Hi Edward:

1) This bug is fixed in version 4.3, which will have an alpha release soon.

2) Grib2Netcdf is deprecated and uses the 4.2 classes, see this page on writing 
files:

  
http://www.unidata.ucar.edu/software/netcdf-java/reference/Cookbook.html#writeClassic

3. We dont have tables for this parameter 0-4-192 from this local table:

 Center        = (54) Montreal (RSMC)
 SubCenter     = (0) null
 Master Table  = 4
 Local Table   = 0

You may note that version 4.2 assigned an incorrect parameter from NCEP (center 
7) tables. Do you have any information on what it should be, or who to contact 
to get that table? 

thanks!

John

> Full Name: Edward Glen
> Email Address: address@hidden
> Organization: MetService New Zealand
> Package Version: toolsUI-4.2
> Operating System: Ubuntu 10.04
> Hardware:
> Description of problem:
> 
> I used the ucar.nc2.iosp.grib.Grib2Netcdf converter to convert the
> attached .grib2 file to NetCDF (result also attached).
> 
> The wgrib2 summary for the grib2 file is:
> 
> 1:0:d=2010110600:var discipline=0 center=54 local_table=0 parmcat=4 
> parm=192:surface:0-6 hour acc fcst:MM-ENS=10
> 
> indicating that the duration for the grib record is 0-6 hour.
> 
> Converted to NetCDF, the time dimension becomes:
> 
> int time(time) ;
> time:long_name = "forecast time for (1 hour intervals)" ;
> time:units = "hour since 2010-11-06T00:00:00Z" ;
> time:bounds = "time_bounds" ;
> int time_bounds(time, bounds_dim) ;
> time_bounds:long_name = "bounds for time" ;
> time_bounds:units = "hour since 2010-11-06T00:00:00Z" ;
> 
> time_bounds =
> 0, 1 ;
> 
> time = 1 ;
> 
> Which I read as meaning 0 to 1 hours, not 0 to 6 hours.
> 
> Am I misunderstanding this or is the converter not correctly taking the
> size of the duration into account?
> 
> Thanks,
> -Edward
> 
> 


Ticket Details
===================
Ticket ID: RXE-920332
Department: Support netCDF Java
Priority: Normal
Status: Open