Hi Dennis,
Thanks for your response. I appreciate that this is being worked on in
NCL, even though I realize that it is probably a result of a non-standard
ECMWF Grib file.
In the meantime, as I mentioned in my original post, I have managed to
work around this problem. I used wgrib to split the Grib file into two,
with each new Grib file containing data on only a single "time range."
Then, NCL can read each of these files with no problem.
Thanks for your help!
--Larry
On Thu, 7 Feb 2002, Dennis Shea wrote:
> >I have an ECMWF Grib file containing the same set of variables on two
> >different time ranges. Is there any way to access a given variable on both
> >of the time ranges in NCL? It seems that NCL gives both of them the same
> >file variable name, and so is only able to access one.
> >
> >An abbreviated version of the Grib inventory for this file, as output from
> >wgrib, is:
>
> [Big Snip]
>
> The main developer of NCL is aware of this issue.
>
> The problem is with the ECMWF file not with NCL. Unfortunately,
> it is unlikely that ECMWF will change. It is possible that
> the source NCL can be changed to accomodate the file. The
> real issue is to make sure that there will not be an
> undesirable side-effects from this change. This requires
> some thought.
>
> Here is a quote: the problem is that I need to make sure thata data
> which actually uses the time range correctly will not be affected.
> To date I have not found any data that is inconsitent in it's use
> like this set.
>
> Regards
> Dennis Shea
>
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
Larry W. Horowitz phone: (609) 452-6520
GFDL/NOAA fax: (609) 987-5063
P.O. Box 308, Princeton Univ. home page: http://www.asp.ucar.edu/~larryh/
Princeton, NJ 08542-0308 email: larryh AT ucar.edu or lwh AT gfdl.noaa.gov
This archive was generated by hypermail 2b29 : Tue Feb 19 2002 - 08:48:13 MST