[Met_help] WRF Gridstat Question
John Halley Gotway
johnhg at ucar.edu
Wed Mar 24 14:01:01 MDT 2010
Bob,
Thanks for sending the file.
I want to point out one odd thing I noticed in the config file you sent me:
fcst_field[] = [ "240/A24","240/A24","240/A24" ];
obs_field[] = [ "61/A24","61/A24","61/A24"];
fcst_thresh[] = [ "gt0.0","gt1.0","gt12.5" ];
obs_thresh[] = [];
You've listed the same field comparison 3 times with one threshold for each. Instead, you should list it once, using 3 thresholds, like this:
fcst_field[] = [ "240/A24" ];
obs_field[] = [ "61/A24" ];
fcst_thresh[] = [ "gt0.0 gt1.0 gt12.5" ];
obs_thresh[] = [];
When I ran the data you sent, I got the following WARNING:
***WARNING***: process_scores() -> Forecast and observation valid times do not match 1267596000 != 1267509600 for field 1.
Sorry for this, but the times are listed in unixtime. We really should change them to be more legible! Here's how you can convert them to a typical time:
date -ud '1970-01-01 UTC '1267596000' seconds' +%Y%m%d_%H
date -ud '1970-01-01 UTC '1267509600' seconds' +%Y%m%d_%H
These unixtimes correspond to date/time of 20100303_06 and 20100302_06.
It looks like we just have things shifted by 1 day.
Doing an "ncdump" of the observation file, you'll see:
ncdump STR_20100302_06
APCP_24:valid_time = "20100302_060000" ;
And doing a wgrib of the model file, you'll see:
wgrib 20100302_06
date 2010030206
The date listed by wgrib is the model initialization time. Since it's a 24-hour forecast, this forecast is actually valid at 2010030306.
The timestamp of your observation file is listing the valid time while the timestamp of your forecast file is listing the initialization time. So really, you shouldn't be comparing these files. They
don't have the same valid time. Instead, you should compare model file 20100302_06 versus observation file STR_20100303_06.
Hope that helps.
John
Craig, Robert J Civ USAF AFWA 16 WS/WXN wrote:
> John, you have the files.
>
> Bob
>
>
>
> -----Original Message-----
> From: John Halley Gotway [mailto:johnhg at ucar.edu]
> Sent: Wednesday, March 24, 2010 2:27 PM
> To: Craig, Robert J Civ USAF AFWA 16 WS/WXN
> Cc: met_help at ucar.edu
> Subject: Re: [Met_help] WRF Gridstat Question
>
> Bob,
>
> This warning message is stating that the forecast file has a valid time
> of 20100303_030000 and the observation file has a valid time of
> 20100303_060000.
>
> It looks like you sent me information about the previous day, 20100302.
> Regardless, it looks like the observation file times are as expected -
> to be valid at hour 06. But for some reason, MET thinks
> that the model file is valid at hour 03. Based on the wgrib output you
> sent, that doesn't look right. Seems like it should be valid at hour 06
> as well.
>
> If you'd like, you could send us a sample fcst and obs file you're
> using. I could run it through Grid-Stat and try to figure it out. If
> you'd like to do that, please post them to our anonymous ftp site:
> ftp ftp.rap.ucar.edu
> username=anonymous
> password="your email address"
> cd incoming/irap/met_help/craig_data
> put "your files"
> bye
>
> I'd just need the NetCDF output of PCP-Combine and the model file you're
> verifying. You already sent me the Grid-Stat config file.
>
> Thanks,
> John Halley Gotway
> met_help at ucar.edu
>
>
> Craig, Robert J Civ USAF AFWA 16 WS/WXN wrote:
>> I am trying to verify WRF 24hr accumulation forecast and comparing to
> stage
>> iV data and I keep getting the following error:
>>
>>
>>
>> ***WARNING***: process_scores() -> Forecast and observation valid
> times do
>> not match 1267585200 != 1267596000 for field 1.
>>
>>
>>
>> What is causing this time mismatch? The data that lead to this are
> listed
>> below. The config file I am using is attached.
>>
>>
>>
>> The PCP combine file to verify 2 Mar 2010 is:
>>
>>
>>
>> /sphome/qcteam/METv2.0/bin/pcp_combine -sum 00000000_000000 1
>> 20100302_060000 24
> /sphome/qcteam/METv2.0/out/pcp_combine/STR_20100302_06
>> -pcpdir /h/data/global/WXQC/data/temp/met/precip -pcprx ST4.*.01h
>>
>> Performing sum command: init_time/in_accum/valid_time/out_accum Times
> =
>> 0/3600/1267509600/86400
>>
>> Searching for 24 files with accumulation times of 1 hours to sum to a
> total
>> accumulation time of 24 hours in /h/data/global/WXQC/data/t
>>
>> emp/met/precip
>>
>> [1] Reading input file:
>> /h/data/global/WXQC/data/temp/met/precip/ST4.2010030206.01h
>>
>> [2] Reading input file:
>> /h/data/global/WXQC/data/temp/met/precip/ST4.2010030205.01h
>>
>> [3] Reading input file:
>> /h/data/global/WXQC/data/temp/met/precip/ST4.2010030204.01h
>>
>> [4] Reading input file:
>> /h/data/global/WXQC/data/temp/met/precip/ST4.2010030203.01h
>>
>> [5] Reading input file:
>> /h/data/global/WXQC/data/temp/met/precip/ST4.2010030202.01h
>>
>> [6] Reading input file:
>> /h/data/global/WXQC/data/temp/met/precip/ST4.2010030201.01h
>>
>> [7] Reading input file:
>> /h/data/global/WXQC/data/temp/met/precip/ST4.2010030200.01h
>>
>> [8] Reading input file:
>> /h/data/global/WXQC/data/temp/met/precip/ST4.2010030123.01h
>>
>> [9] Reading input file:
>> /h/data/global/WXQC/data/temp/met/precip/ST4.2010030122.01h
>>
>> [10] Reading input file:
>> /h/data/global/WXQC/data/temp/met/precip/ST4.2010030121.01h
>>
>> [11] Reading input file:
>> /h/data/global/WXQC/data/temp/met/precip/ST4.2010030120.01h
>>
>> [12] Reading input file:
>> /h/data/global/WXQC/data/temp/met/precip/ST4.2010030119.01h
>>
>> [13] Reading input file:
>> /h/data/global/WXQC/data/temp/met/precip/ST4.2010030118.01h
>>
>> [14] Reading input file:
>> /h/data/global/WXQC/data/temp/met/precip/ST4.2010030117.01h
>>
>> [15] Reading input file:
>> /h/data/global/WXQC/data/temp/met/precip/ST4.2010030116.01h
>>
>> [16] Reading input file:
>> /h/data/global/WXQC/data/temp/met/precip/ST4.2010030115.01h
>>
>> [17] Reading input file:
>> /h/data/global/WXQC/data/temp/met/precip/ST4.2010030114.01h
>>
>> [18] Reading input file:
>> /h/data/global/WXQC/data/temp/met/precip/ST4.2010030113.01h
>>
>> [19] Reading input file:
>> /h/data/global/WXQC/data/temp/met/precip/ST4.2010030112.01h
>>
>> [20] Reading input file:
>> /h/data/global/WXQC/data/temp/met/precip/ST4.2010030111.01h
>>
>> [21] Reading input file:
>> /h/data/global/WXQC/data/temp/met/precip/ST4.2010030110.01h
>>
>> [22] Reading input file:
>> /h/data/global/WXQC/data/temp/met/precip/ST4.2010030109.01h
>>
>> [23] Reading input file:
>> /h/data/global/WXQC/data/temp/met/precip/ST4.2010030108.01h
>>
>> [24] Reading input file:
>> /h/data/global/WXQC/data/temp/met/precip/ST4.2010030107.01h
>>
>> Writing output file:
> /sphome/qcteam/METv2.0/out/pcp_combine/STR_20100302_06
>>
>>
>> The grib header for the model file is:
>>
>>
>>
>> rec 1:0:date 2010030206 TOTAPCP kpds5=240 kpds6=1 kpds7=0 levels=(0,0)
>> grid=255 sfc 0-24hr acc:
>>
>> TOTAPCP= Total accumulated precipitation [kg/m2]
>>
>> timerange 4 P1 0 P2 24 TimeU 1 nx 342 ny 210 GDS grid 3 num_in_ave
> 0
>> missing 0
>>
>> center 57 subcenter 1 process 11 Table 131
>>
>> Lambert Conf: Lat1 20.586000 Lon1 -122.243000 Lov -96.000000
>>
>> Latin1 60.000000 Latin2 30.000000 LatSP -90.000000 LonSP
> 0.000000
>> North Pole (342 x 210) Dx 15.000000 Dy 15.000000 scan 64
>>
>> min/max data 0 93.31 num bits 14 BDS_Ref 0 DecScale 2 BinScale 0
>>
>>
>>
>> The MET grid stat command is:
>>
>>
>>
>> /sphome/qcteam/METv2.0/bin/grid_stat
>> /h/data/global/WXQC/data/temp/met/20100302_06
>> /sphome/qcteam/METv2.0/out/pcp_combine/STR_20100302_
>>
>> 06 /sphome/qcteam/METv2.0/data/config/WRFGridStatconfig -outdir
>> /sphome/qcteam/METv2.0/out/grid_stat/WRF/T02B
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
> ------------------------------------------------------------------------
>> _______________________________________________
>> Met_help mailing list
>> Met_help at mailman.ucar.edu
>> http://mailman.ucar.edu/mailman/listinfo/met_help
More information about the Met_help
mailing list