Re: Nio.open_file hangs NOT necessarily too few grid points in GRIB1

From: Wenchieh YEN <wenchieh.yen_at_nyahnyahspammersnyahnyah>
Date: Fri, 16 Feb 2007 22:33:09 +0100

Hello David,

David Ian Brown wrote:
> Wenchieh,
> OK, thanks for this information -- it's quite useful. My earlier reply
> was a bit of a shortcut.

no problem, I got what you actually meant in the last reply.

> Since codes 13 and 14 for the forecast time unit are not otherwise
> defined I can (in the case of records from DWD)
> add code to treat these as .25 and .5 hours respectively.

great!

>
> Concerning your problem, it definitely seems odd. Can you say for sure
> that the file lfff00150000.crop_4x4.grb
> that you sent us is one of the files that does not work for you? Because
> on my mac (still at 10.3.9) I have no problem reading it.
> I will try it out on other systems to see if I can get it to fail.

yes. I have problem with that file on both of my systems. If possible I
will test it on other systems too.

I used two methods to crop the original GRIB file from DWD: one
using the official DWD GRIB1 library in fortran; one with my own c
codes. They differ in how BDS and BMS are handled. Both don't work with
Nio. The 4x4 file you have is produced using the DWD GRIB1 lib.

> Also I notice that these files have parameters from DWD table 204 which
> is not yet incorporated into NCL.
> A quick google search gives me the values below for table 204. Is this
> correct? I can see that it is not complete.
> Can you point me to an authoritative version of the table so that I can
> add it into NCL's set of parameter tables.
> Thanks.
> -dave

DWD now has status quo GRIB tables up to 207. Some GRIB parameter have a
new name with the same GRIB table and parameter id. Some even have
auxiliary parameter id defined in the PDS section, so as to subclass
certain parameters. One example is cloudy/clear-sky brightness
temperature/radiances from synthetic satellite channels.

I believe there are no "authoritative" DWD tables out there in internet.
I have the operational DWD GRIB tables here, however, many parameter
names seem differ from what are actually used in the DWD models / NCL
tables / wgrib tables. I am rebuilding the tables to be NCL compliant
(see the link below) and cross-checking them. Besides I am also
contacting possbile internal DWD people who can help on this, and most
importantly get their endorsement...

If you gonna incorporate these tables into NCL, thus Nio, what format do
you prefer? like NCL tables?

-Wenchieh-

http://www.ncl.ucar.edu/Document/Manuals/Ref_Manual/NclFormatSupport.shtml#GRIB
Received on Fri Feb 16 2007 - 14:33:09 MST

This archive was generated by hypermail 2.2.0 : Sun Apr 01 2007 - 11:53:11 MDT