[Met_help] ERROR: read_pds() -> unexpected time range indicator of 10

[John Halley Gotway]

Luke,

In running your data through MET, I found that I was requiring the forecasts to be on the hour.  To make MET more broadly applicable, I removed that check.  However, there are some implications to this.

For example, the output files from point_stat are named based on their initialization time and their forecast hour.  The name of the VSDB output file of point_stat for your data is:
/d1/johnhg/test_MET/METv0.9/out/point_stat/point_stat_2007092818I_F06.vsdb

You'll see that the initialization time is listed as YYYYMMDDHH and the forecast hour is listed as HH.  Inside this file, you'll find the same time info format YYYYMMDDHH initialization and HH
forecast time.  However in your case, the forecast time is really 6 hours and 1 minute, not just 6 hours!

The reason the time information is formatted in this way is to be consistent with the output of the current set of NCEP verification tools.  The MET development team will need to discuss how important
this is, and whether in future releases of MET, we should reformat the initialization and forecast times as YYYYMMDDHHMMSS and HHMMSS to be more explicit.

As for the change you need to support the time range indicator of 10, I've attached a patched file named "read_grib.cc" which replaces the file by the same name in the "vx_met_util" library.  Here are
some instructions for using it:
(1) mv /fuelberg/r0/lpeffers/MET_9/METv0.9/lib/vx_met_util/read_grib.cc /fuelberg/r0/lpeffers/MET_9/METv0.9/lib/vx_met_util/read_grib.cc_orig
(2) cp read_grib.cc /fuelberg/r0/lpeffers/MET_9/METv0.9/lib/vx_met_util/.
(3) cd /fuelberg/r0/lpeffers/MET_9/METv0.9
(4) make clean
(5) make >& make_met.log&
(6) tail -f make_met.log

Please let me know if you run into any more problems or have more questions.

John

Luke Peffers wrote:
> Thank you John:
> 
> I will post these files via ftp after I send this message.  I just wanted to
> let you know my steps up to this problem just in case I may have something
> to do with the error in addition to the bug that you mentioned;
> 
> Because I am fairly new to the world of NWP, I may have made an error
> somewhere in the process of creating these two files.  I post-processed my
> WRF data using wrfpostprocV2.  I had some problems using the run script in
> WPP that may have something to do with my current problem with MET.  The
> script seems to want to cycle through forecast hours in a single wrfout
> file.  However, the user guide for WPP states that there must be only one
> frame per wrfout file so I adjusted the script to only process one wrfout
> file for one time. From there, I had a problem with the output time from
> wrf.  WRF's output time does not always stop at 00:00. Again, to solve this
> in WPP, I adjusted the script to look for a wrfout file with non-zero
> minutes and seconds.
> 
> Now, my question is:  does MET care if the minutes and seconds of the
> forecast file doesn't match the observation file?  My forecast file is valid
> for 2007-09-29_00:01:40 while my observations are valid for
> 2007-09-29_00:00:00.
> 
> Thanks again for all the help.
> 
> Luke Peffers
> FSU
> 
> On 10/1/07, John Halley Gotway <johnhg at rap.ucar.edu> wrote:
>> Luke,
>>
>> Your data uses a setting in Grib that we hadn't anticipated in the beta
>> version.  The fix should involve just adding a few lines of code to one of
>> the library files.
>>
>> Would you be able to send me that Grib file you're using
>> (/fuelberg/r0/lpeffers/WRFV2/run/postprd/WRFPRS_d01.06.01.grb) as well as
>> the NetCDF obs file
>> (/fuelberg/r0/lpeffers/MET_9/METv0.9/data/PREPBEFR_DATA/OBS_00.nc)?  I'd
>> like to make sure that the change I make really does fix the problem.  Once
>> I'm sure of that, I'll send you an updated patch
>> file with instructions for how to update your version of MET.  And I'll
>> make sure the fix makes its way into subsequent versions of MET.
>>
>> If those 2 files are too large to send via email, you could post it on our
>> anonymous FTP site:
>> cp /fuelberg/r0/lpeffers/WRFV2/run/postprd/WRFPRS_d01.06.01.grb
>> WRFPRS_d01.06.01.grb
>> cp /fuelberg/r0/lpeffers/MET_9/METv0.9/data/PREPBEFR_DATA/OBS_00.nc
>> OBS_00.nc
>> ftp ftp.rap.ucar.edu
>> Username = anonymous
>> Password = "you email address"
>> cd incoming/irap/johnhg
>> bin
>> put WRFPRS_d01.06.01.grb
>> put OBS_00.nc
>> bye
>>
>> Thanks,
>> John
>>
>> Luke Peffers wrote:
>>> Hello, I have made it to the point where I can now try using the
>> point_stat
>>> tool.  I didn't make it very far however.  I received this error (ERROR:
>>> read_pds() -> unexpected time range indicator of 10) when I entered the
>>> command;
>>>
>>>  ./point_stat
>>> /fuelberg/r0/lpeffers/WRFV2/run/postprd/WRFPRS_d01.06.01.grb
>>> /fuelberg/r0/lpeffers/MET_9/METv0.9/data/PREPBEFR_DATA/OBS_00.nc     WRF
>>>
>>> Any advice as to where I should start looking for the solution to this
>>> problem?
>>>
>>> Thank you,
>>>
>>> Luke Peffers
>>> FSU
>>>
>>>
>>>
>>> ------------------------------------------------------------------------
>>>
>>> _______________________________________________
>>> Met_help mailing list
>>> Met_help at mailman.ucar.edu
>>> http://mailman.ucar.edu/mailman/listinfo/met_help
> 
-------------- next part --------------
A non-text attachment was scrubbed...
Name: read_grib.cc
Type: text/x-c++src
Size: 27221 bytes
Desc: not available
Url : http://mailman.ucar.edu/pipermail/met_help/attachments/20071001/b7333716/read_grib-0001.bin