Hi Sachin-
We recently found this bug with another dataset and it has been
fixed. It was a bug introduced in the grib reading library.
However, a new release of netCDF-Java with the
fix is pending the repair of John's machine. ;-)
Don
Sachin Kumar Bhate wrote:
John,
We are trying to read the coordinate variables from a grib file, but
somehow the variable array is being populated with the same 1st value.
1. Check screen grab lon values in latest toolsUI (Version
4.0.41.20090505.2215) at
http://www.gri.msstate.edu/rsearch_data/nopp/lon_sachin.jpg
2. The other strange thing is that Rich is using the latest toolsUI as
well on a machine
that does not have latest JVM (i.e. 1.6) and on that machine the lon
variables
show up correctly... check his screen grab,
http://www.gri.msstate.edu/rsearch_data/nopp/lon_rich.png
Looks like the attribute 'grid_spacing' in step 1 above is not set
properly.
It says '0.0 degrees_east', while it should be '0.25 degrees_east'.
The file we are trying to read is at
ftp://polar.ncep.noaa.gov/pub/history/waves/wna.dp.200409.grb
The netcdf file for above grib file (converted using NCL) is at
http://omglnx4.meas.ncsu.edu/joseph/skb_help/wna.dp.200409.nc
Any comments,
thanks,
Sachin
--
*************************************************************
Don Murray UCAR Unidata Program
dmurray@xxxxxxxxxxxxxxxx P.O. Box 3000
(303) 497-8628 Boulder, CO 80307
http://www.unidata.ucar.edu/staff/donm
*************************************************************
