[Met_help] [rt.rap.ucar.edu #75640] History for Problems with grid_stat used with ECMWF observations
Julie Prestopnik via RT
met_help at ucar.edu
Wed Mar 30 10:25:39 MDT 2016
----------------------------------------------------------------
Initial Request
----------------------------------------------------------------
Hello everyone,
I have a problem when I use grid_stat comparing WRF forecasted data with
ECMWF analysis.
This problem does not happen when I compare with NCEP analysis.
I modified GridStatConfig_default only in this part:
fcst = {
field = [
{
name = "TMP";
level = [ "P500" ];
// cat_thresh = [ >0.0, >=5.0 ];
}
];
}
obs = fcst;
When I use
./grid_stat WRF.fcst.grib NCEP.obs.grib GridStatConfig_modified
all is fine.
When I use
./grid_stat WRF.fcst.grib ECMWF.obs.grib GridStatConfig_modified -v 4
I get the this output
DEBUG 1: Default Config File:
/home/ssalis/Documents/WRF/Build_WRF/MET/met-5.1_bugfix/share/met/config/GridStatConfig_default
DEBUG 1: User Config File: GridStatConfig_modified
DEBUG 4: Met2dDataFileFactory::new_met_2d_data_file() -> created new
Met2dDataFile object of type "FileType_Gb1".
DEBUG 4: Met2dDataFileFactory::new_met_2d_data_file() -> created new
Met2dDataFile object of type "FileType_Gb1".
DEBUG 4: VarInfoFactory::new_var_info() -> created new VarInfo object of
type "FileType_Gb1".
DEBUG 4: VarInfoFactory::new_var_info() -> created new VarInfo object of
type "FileType_Gb1".
DEBUG 3: Use the matching forecast and observation grids.
DEBUG 3: Grid Definition: Projection: Lat/Lon Nx: 41 Ny: 31 lat_ll: 38.000
lon_ll: -6.500 delta_lat: 0.125 delta_lon: 0.125
GSL_RNG_TYPE=mt19937
GSL_RNG_SEED=1209005142
DEBUG 1: Forecast File: WRF.fcst.grib
DEBUG 1: Observation File: ECMWF.obs.grib
DEBUG 3: Reading forecast data for TMP/P500.
DEBUG 3: MetGrib1DataFile::data_plane_array() -> Found range match for
VarInfo "TMP/P500" in GRIB record 122 of GRIB file "WRF.fcst.grib".
DEBUG 3: MetGrib1DataFile::data_plane_array() -> Found 1 GRIB records
matching VarInfo "TMP/P500" in GRIB file "WRF.fcst.grib".
DEBUG 3: Reading observation data for TMP/P500.
DEBUG 3: MetGrib1DataFile::data_plane_array() -> Found 0 GRIB records
matching VarInfo "TMP/P500" in GRIB file "ECMWF.obs.grib".
WARNING:
WARNING: MetGrib1DataFile::data_plane() -> No exact match found for VarInfo
"TMP/P500" in GRIB file "ECMWF.obs.grib".
WARNING:
WARNING:
WARNING: process_scores() -> TMP/P500 not found in file: ECMWF.obs.grib
WARNING:
DEBUG 2:
DEBUG 2:
--------------------------------------------------------------------------------
DEBUG 2:
Checking my ECMWF.obs.grib with wgrib I obtain
rec 3:5300:date 2016032212 T kpds5=130 kpds6=100 kpds7=1 levels=(0,1)
grid=255 1 mb ECext=1 class=operations type=analysis stream=1025
Version=0001 anl:
T=Temperature [K]
timerange 0 P1 0 P2 0 TimeU 1 nx 41 ny 31 GDS grid 0 num_in_ave 0
missing 0
center 98 subcenter 0 process 146 Table 128 scan: WE:NS winds(N/S)
latlon: lat 41.750000 to 38.000000 by 0.125000 nxny 1271
long 6.500000 to 11.500000 by 0.125000, (41 x 31) scan 0 mode 128
bdsgrid 1
min/max data 252.24 257.431 num bits 16 BDS_Ref 252.24 DecScale 0
BinScale -13
Any suggestion about how to solve this problem?
Thanks,
Samuele
----------------------------------------------------------------
Complete Ticket History
----------------------------------------------------------------
Subject: Problems with grid_stat used with ECMWF observations
From: Julie Prestopnik
Time: Thu Mar 24 13:43:23 2016
Hi Samuele. I see that you are having problems with GridStat when
using
ECMWF data. Unfortunately, MET does not currently include support for
the
ECMWF usage of GRIB. We are aware that this is support that would be
useful and it is a priority for us to add support for it. We're
hoping to
include it in our next release.
For now, one way around this is to modify your obs field. So, instead
of
having obs = fcst, create an obs field that looks like the forecast
field,
but instead of using "name" use "GRIB1_ptv" and "GRIB1_rec" as I show
below, using the table number (128 in this case) for GRIB1_ptv and
using
the grib code (130 in this case) for GRIB1_rec.
obs = {
message_type = [ "<Fill in your message_type here>" ];
sid_exc = [];
field = [
{
GRIB1_ptv = 128;
GRIB1_rec = 130;
level = [ "<Fill in the level here>" ];
cat_thresh = [ ];
}
];
}
Please give that a shot and let us know how it goes.
Thanks,
Julie
On Thu, Mar 24, 2016 at 10:19 AM, Samuele Salis via RT
<met_help at ucar.edu>
wrote:
>
> Thu Mar 24 10:19:09 2016: Request 75640 was acted upon.
> Transaction: Ticket created by samuelesalis at gmail.com
> Queue: met_help
> Subject: Problems with grid_stat used with ECMWF observations
> Owner: Nobody
> Requestors: samuelesalis at gmail.com
> Status: new
> Ticket <URL:
https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=75640 >
>
>
> Hello everyone,
>
> I have a problem when I use grid_stat comparing WRF forecasted data
with
> ECMWF analysis.
> This problem does not happen when I compare with NCEP analysis.
> I modified GridStatConfig_default only in this part:
> fcst = {
>
> field = [
> {
> name = "TMP";
> level = [ "P500" ];
> // cat_thresh = [ >0.0, >=5.0 ];
> }
> ];
>
> }
>
> obs = fcst;
>
>
>
>
> When I use
> ./grid_stat WRF.fcst.grib NCEP.obs.grib GridStatConfig_modified
> all is fine.
>
>
>
> When I use
> ./grid_stat WRF.fcst.grib ECMWF.obs.grib GridStatConfig_modified -v
4
> I get the this output
>
> DEBUG 1: Default Config File:
>
> /home/ssalis/Documents/WRF/Build_WRF/MET/met-
5.1_bugfix/share/met/config/GridStatConfig_default
> DEBUG 1: User Config File: GridStatConfig_modified
> DEBUG 4: Met2dDataFileFactory::new_met_2d_data_file() -> created new
> Met2dDataFile object of type "FileType_Gb1".
> DEBUG 4: Met2dDataFileFactory::new_met_2d_data_file() -> created new
> Met2dDataFile object of type "FileType_Gb1".
> DEBUG 4: VarInfoFactory::new_var_info() -> created new VarInfo
object of
> type "FileType_Gb1".
> DEBUG 4: VarInfoFactory::new_var_info() -> created new VarInfo
object of
> type "FileType_Gb1".
> DEBUG 3: Use the matching forecast and observation grids.
> DEBUG 3: Grid Definition: Projection: Lat/Lon Nx: 41 Ny: 31 lat_ll:
38.000
> lon_ll: -6.500 delta_lat: 0.125 delta_lon: 0.125
> GSL_RNG_TYPE=mt19937
> GSL_RNG_SEED=1209005142
> DEBUG 1: Forecast File: WRF.fcst.grib
> DEBUG 1: Observation File: ECMWF.obs.grib
> DEBUG 3: Reading forecast data for TMP/P500.
> DEBUG 3: MetGrib1DataFile::data_plane_array() -> Found range match
for
> VarInfo "TMP/P500" in GRIB record 122 of GRIB file "WRF.fcst.grib".
> DEBUG 3: MetGrib1DataFile::data_plane_array() -> Found 1 GRIB
records
> matching VarInfo "TMP/P500" in GRIB file "WRF.fcst.grib".
> DEBUG 3: Reading observation data for TMP/P500.
> DEBUG 3: MetGrib1DataFile::data_plane_array() -> Found 0 GRIB
records
> matching VarInfo "TMP/P500" in GRIB file "ECMWF.obs.grib".
> WARNING:
> WARNING: MetGrib1DataFile::data_plane() -> No exact match found for
VarInfo
> "TMP/P500" in GRIB file "ECMWF.obs.grib".
> WARNING:
> WARNING:
> WARNING: process_scores() -> TMP/P500 not found in file:
ECMWF.obs.grib
> WARNING:
> DEBUG 2:
> DEBUG 2:
>
>
--------------------------------------------------------------------------------
> DEBUG 2:
>
>
>
>
>
> Checking my ECMWF.obs.grib with wgrib I obtain
>
> rec 3:5300:date 2016032212 T kpds5=130 kpds6=100 kpds7=1
levels=(0,1)
> grid=255 1 mb ECext=1 class=operations type=analysis stream=1025
> Version=0001 anl:
> T=Temperature [K]
> timerange 0 P1 0 P2 0 TimeU 1 nx 41 ny 31 GDS grid 0 num_in_ave 0
> missing 0
> center 98 subcenter 0 process 146 Table 128 scan: WE:NS winds(N/S)
> latlon: lat 41.750000 to 38.000000 by 0.125000 nxny 1271
> long 6.500000 to 11.500000 by 0.125000, (41 x 31) scan 0
mode 128
> bdsgrid 1
> min/max data 252.24 257.431 num bits 16 BDS_Ref 252.24 DecScale
0
> BinScale -13
>
>
>
>
> Any suggestion about how to solve this problem?
>
> Thanks,
> Samuele
>
>
--
Julie Prestopnik
National Center for Atmospheric Research
Research Applications Laboratory
Phone: 303.497.8399
Email: jpresto at ucar.edu
------------------------------------------------
Subject: Problems with grid_stat used with ECMWF observations
From: Samuele Salis
Time: Fri Mar 25 01:44:09 2016
Hi Julie,
I modified the GridStatConfig_default file as you suggested and
I have not message errors any more and MET produces all the
files requested.
If I check the statistical values they look reasonable, but still I
have a
little doubt because I saw that the ECMWF variable (temperature
at 500 hPa) is considered in the grid_stat standard output as
130 128 "ELEV" "Ocean surface elevation relative to geoid" "m"
in the nceptab_flat.txt table
DEBUG 3: Reading observation data for ELEV/P500.
I think all is fine, but I would like your opinion about this
topic.
Thanks,
Samuele
On 24 March 2016 at 20:43, Julie Prestopnik via RT <met_help at ucar.edu>
wrote:
> Hi Samuele. I see that you are having problems with GridStat when
using
> ECMWF data. Unfortunately, MET does not currently include support
for the
> ECMWF usage of GRIB. We are aware that this is support that would
be
> useful and it is a priority for us to add support for it. We're
hoping to
> include it in our next release.
>
> For now, one way around this is to modify your obs field. So,
instead of
> having obs = fcst, create an obs field that looks like the forecast
field,
> but instead of using "name" use "GRIB1_ptv" and "GRIB1_rec" as I
show
> below, using the table number (128 in this case) for GRIB1_ptv and
using
> the grib code (130 in this case) for GRIB1_rec.
>
> obs = {
> message_type = [ "<Fill in your message_type here>" ];
> sid_exc = [];
>
> field = [
> {
> GRIB1_ptv = 128;
> GRIB1_rec = 130;
> level = [ "<Fill in the level here>" ];
> cat_thresh = [ ];
> }
> ];
>
> }
>
> Please give that a shot and let us know how it goes.
>
> Thanks,
> Julie
>
> On Thu, Mar 24, 2016 at 10:19 AM, Samuele Salis via RT
<met_help at ucar.edu>
> wrote:
>
> >
> > Thu Mar 24 10:19:09 2016: Request 75640 was acted upon.
> > Transaction: Ticket created by samuelesalis at gmail.com
> > Queue: met_help
> > Subject: Problems with grid_stat used with ECMWF observations
> > Owner: Nobody
> > Requestors: samuelesalis at gmail.com
> > Status: new
> > Ticket <URL:
https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=75640 >
> >
> >
> > Hello everyone,
> >
> > I have a problem when I use grid_stat comparing WRF forecasted
data with
> > ECMWF analysis.
> > This problem does not happen when I compare with NCEP analysis.
> > I modified GridStatConfig_default only in this part:
> > fcst = {
> >
> > field = [
> > {
> > name = "TMP";
> > level = [ "P500" ];
> > // cat_thresh = [ >0.0, >=5.0 ];
> > }
> > ];
> >
> > }
> >
> > obs = fcst;
> >
> >
> >
> >
> > When I use
> > ./grid_stat WRF.fcst.grib NCEP.obs.grib GridStatConfig_modified
> > all is fine.
> >
> >
> >
> > When I use
> > ./grid_stat WRF.fcst.grib ECMWF.obs.grib GridStatConfig_modified
-v 4
> > I get the this output
> >
> > DEBUG 1: Default Config File:
> >
> >
> /home/ssalis/Documents/WRF/Build_WRF/MET/met-
5.1_bugfix/share/met/config/GridStatConfig_default
> > DEBUG 1: User Config File: GridStatConfig_modified
> > DEBUG 4: Met2dDataFileFactory::new_met_2d_data_file() -> created
new
> > Met2dDataFile object of type "FileType_Gb1".
> > DEBUG 4: Met2dDataFileFactory::new_met_2d_data_file() -> created
new
> > Met2dDataFile object of type "FileType_Gb1".
> > DEBUG 4: VarInfoFactory::new_var_info() -> created new VarInfo
object of
> > type "FileType_Gb1".
> > DEBUG 4: VarInfoFactory::new_var_info() -> created new VarInfo
object of
> > type "FileType_Gb1".
> > DEBUG 3: Use the matching forecast and observation grids.
> > DEBUG 3: Grid Definition: Projection: Lat/Lon Nx: 41 Ny: 31
lat_ll:
> 38.000
> > lon_ll: -6.500 delta_lat: 0.125 delta_lon: 0.125
> > GSL_RNG_TYPE=mt19937
> > GSL_RNG_SEED=1209005142
> > DEBUG 1: Forecast File: WRF.fcst.grib
> > DEBUG 1: Observation File: ECMWF.obs.grib
> > DEBUG 3: Reading forecast data for TMP/P500.
> > DEBUG 3: MetGrib1DataFile::data_plane_array() -> Found range match
for
> > VarInfo "TMP/P500" in GRIB record 122 of GRIB file
"WRF.fcst.grib".
> > DEBUG 3: MetGrib1DataFile::data_plane_array() -> Found 1 GRIB
records
> > matching VarInfo "TMP/P500" in GRIB file "WRF.fcst.grib".
> > DEBUG 3: Reading observation data for TMP/P500.
> > DEBUG 3: MetGrib1DataFile::data_plane_array() -> Found 0 GRIB
records
> > matching VarInfo "TMP/P500" in GRIB file "ECMWF.obs.grib".
> > WARNING:
> > WARNING: MetGrib1DataFile::data_plane() -> No exact match found
for
> VarInfo
> > "TMP/P500" in GRIB file "ECMWF.obs.grib".
> > WARNING:
> > WARNING:
> > WARNING: process_scores() -> TMP/P500 not found in file:
ECMWF.obs.grib
> > WARNING:
> > DEBUG 2:
> > DEBUG 2:
> >
> >
>
--------------------------------------------------------------------------------
> > DEBUG 2:
> >
> >
> >
> >
> >
> > Checking my ECMWF.obs.grib with wgrib I obtain
> >
> > rec 3:5300:date 2016032212 T kpds5=130 kpds6=100 kpds7=1
levels=(0,1)
> > grid=255 1 mb ECext=1 class=operations type=analysis stream=1025
> > Version=0001 anl:
> > T=Temperature [K]
> > timerange 0 P1 0 P2 0 TimeU 1 nx 41 ny 31 GDS grid 0 num_in_ave
0
> > missing 0
> > center 98 subcenter 0 process 146 Table 128 scan: WE:NS
winds(N/S)
> > latlon: lat 41.750000 to 38.000000 by 0.125000 nxny 1271
> > long 6.500000 to 11.500000 by 0.125000, (41 x 31) scan 0
mode
> 128
> > bdsgrid 1
> > min/max data 252.24 257.431 num bits 16 BDS_Ref 252.24
DecScale 0
> > BinScale -13
> >
> >
> >
> >
> > Any suggestion about how to solve this problem?
> >
> > Thanks,
> > Samuele
> >
> >
>
>
> --
> Julie Prestopnik
> National Center for Atmospheric Research
> Research Applications Laboratory
> Phone: 303.497.8399
> Email: jpresto at ucar.edu
>
>
------------------------------------------------
Subject: Problems with grid_stat used with ECMWF observations
From: Julie Prestopnik
Time: Fri Mar 25 10:23:24 2016
Hi Samuele. I'm glad that you have not had any more errors and that
MET is
producing output.
All the nceptab_flat.txt (and all GRIB1/GRIB2 tables) file does is
assign
abbreviations and units to a sequence of identifying numbers. It is
currently set for NCEP usage. You are welcome to change the contents
of
that table to change the abbreviations used.
At runtime, MET is reading the installed version of that file from
INSTALL/share/met/table_files/nceptab_flat.txt, where INSTALL
indicates
MET's installation path. You can either modify that files manually,
or you
can modify the met-5.1_bugfix/data/table_files/nceptab_flat.txt file
and
then run "make install" to install the file in the share area.
As I mentioned before, we will be adding better support for this in
the
next release.
Please let me know if you have any other questions.
Thanks,
Julie
On Fri, Mar 25, 2016 at 1:44 AM, Samuele Salis via RT
<met_help at ucar.edu>
wrote:
>
> <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=75640 >
>
> Hi Julie,
>
> I modified the GridStatConfig_default file as you suggested and
> I have not message errors any more and MET produces all the
> files requested.
>
> If I check the statistical values they look reasonable, but still I
have a
> little doubt because I saw that the ECMWF variable (temperature
> at 500 hPa) is considered in the grid_stat standard output as
> 130 128 "ELEV" "Ocean surface elevation relative to geoid" "m"
> in the nceptab_flat.txt table
>
> DEBUG 3: Reading observation data for ELEV/P500.
>
> I think all is fine, but I would like your opinion about this
> topic.
>
> Thanks,
> Samuele
>
> On 24 March 2016 at 20:43, Julie Prestopnik via RT
<met_help at ucar.edu>
> wrote:
>
> > Hi Samuele. I see that you are having problems with GridStat when
using
> > ECMWF data. Unfortunately, MET does not currently include support
for
> the
> > ECMWF usage of GRIB. We are aware that this is support that would
be
> > useful and it is a priority for us to add support for it. We're
hoping
> to
> > include it in our next release.
> >
> > For now, one way around this is to modify your obs field. So,
instead of
> > having obs = fcst, create an obs field that looks like the
forecast
> field,
> > but instead of using "name" use "GRIB1_ptv" and "GRIB1_rec" as I
show
> > below, using the table number (128 in this case) for GRIB1_ptv and
using
> > the grib code (130 in this case) for GRIB1_rec.
> >
> > obs = {
> > message_type = [ "<Fill in your message_type here>" ];
> > sid_exc = [];
> >
> > field = [
> > {
> > GRIB1_ptv = 128;
> > GRIB1_rec = 130;
> > level = [ "<Fill in the level here>" ];
> > cat_thresh = [ ];
> > }
> > ];
> >
> > }
> >
> > Please give that a shot and let us know how it goes.
> >
> > Thanks,
> > Julie
> >
> > On Thu, Mar 24, 2016 at 10:19 AM, Samuele Salis via RT <
> met_help at ucar.edu>
> > wrote:
> >
> > >
> > > Thu Mar 24 10:19:09 2016: Request 75640 was acted upon.
> > > Transaction: Ticket created by samuelesalis at gmail.com
> > > Queue: met_help
> > > Subject: Problems with grid_stat used with ECMWF
observations
> > > Owner: Nobody
> > > Requestors: samuelesalis at gmail.com
> > > Status: new
> > > Ticket <URL:
https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=75640
> >
> > >
> > >
> > > Hello everyone,
> > >
> > > I have a problem when I use grid_stat comparing WRF forecasted
data
> with
> > > ECMWF analysis.
> > > This problem does not happen when I compare with NCEP analysis.
> > > I modified GridStatConfig_default only in this part:
> > > fcst = {
> > >
> > > field = [
> > > {
> > > name = "TMP";
> > > level = [ "P500" ];
> > > // cat_thresh = [ >0.0, >=5.0 ];
> > > }
> > > ];
> > >
> > > }
> > >
> > > obs = fcst;
> > >
> > >
> > >
> > >
> > > When I use
> > > ./grid_stat WRF.fcst.grib NCEP.obs.grib GridStatConfig_modified
> > > all is fine.
> > >
> > >
> > >
> > > When I use
> > > ./grid_stat WRF.fcst.grib ECMWF.obs.grib GridStatConfig_modified
-v 4
> > > I get the this output
> > >
> > > DEBUG 1: Default Config File:
> > >
> > >
> >
> /home/ssalis/Documents/WRF/Build_WRF/MET/met-
5.1_bugfix/share/met/config/GridStatConfig_default
> > > DEBUG 1: User Config File: GridStatConfig_modified
> > > DEBUG 4: Met2dDataFileFactory::new_met_2d_data_file() -> created
new
> > > Met2dDataFile object of type "FileType_Gb1".
> > > DEBUG 4: Met2dDataFileFactory::new_met_2d_data_file() -> created
new
> > > Met2dDataFile object of type "FileType_Gb1".
> > > DEBUG 4: VarInfoFactory::new_var_info() -> created new VarInfo
object
> of
> > > type "FileType_Gb1".
> > > DEBUG 4: VarInfoFactory::new_var_info() -> created new VarInfo
object
> of
> > > type "FileType_Gb1".
> > > DEBUG 3: Use the matching forecast and observation grids.
> > > DEBUG 3: Grid Definition: Projection: Lat/Lon Nx: 41 Ny: 31
lat_ll:
> > 38.000
> > > lon_ll: -6.500 delta_lat: 0.125 delta_lon: 0.125
> > > GSL_RNG_TYPE=mt19937
> > > GSL_RNG_SEED=1209005142
> > > DEBUG 1: Forecast File: WRF.fcst.grib
> > > DEBUG 1: Observation File: ECMWF.obs.grib
> > > DEBUG 3: Reading forecast data for TMP/P500.
> > > DEBUG 3: MetGrib1DataFile::data_plane_array() -> Found range
match for
> > > VarInfo "TMP/P500" in GRIB record 122 of GRIB file
"WRF.fcst.grib".
> > > DEBUG 3: MetGrib1DataFile::data_plane_array() -> Found 1 GRIB
records
> > > matching VarInfo "TMP/P500" in GRIB file "WRF.fcst.grib".
> > > DEBUG 3: Reading observation data for TMP/P500.
> > > DEBUG 3: MetGrib1DataFile::data_plane_array() -> Found 0 GRIB
records
> > > matching VarInfo "TMP/P500" in GRIB file "ECMWF.obs.grib".
> > > WARNING:
> > > WARNING: MetGrib1DataFile::data_plane() -> No exact match found
for
> > VarInfo
> > > "TMP/P500" in GRIB file "ECMWF.obs.grib".
> > > WARNING:
> > > WARNING:
> > > WARNING: process_scores() -> TMP/P500 not found in file:
ECMWF.obs.grib
> > > WARNING:
> > > DEBUG 2:
> > > DEBUG 2:
> > >
> > >
> >
>
--------------------------------------------------------------------------------
> > > DEBUG 2:
> > >
> > >
> > >
> > >
> > >
> > > Checking my ECMWF.obs.grib with wgrib I obtain
> > >
> > > rec 3:5300:date 2016032212 T kpds5=130 kpds6=100 kpds7=1
levels=(0,1)
> > > grid=255 1 mb ECext=1 class=operations type=analysis stream=1025
> > > Version=0001 anl:
> > > T=Temperature [K]
> > > timerange 0 P1 0 P2 0 TimeU 1 nx 41 ny 31 GDS grid 0
num_in_ave 0
> > > missing 0
> > > center 98 subcenter 0 process 146 Table 128 scan: WE:NS
winds(N/S)
> > > latlon: lat 41.750000 to 38.000000 by 0.125000 nxny 1271
> > > long 6.500000 to 11.500000 by 0.125000, (41 x 31) scan
0 mode
> > 128
> > > bdsgrid 1
> > > min/max data 252.24 257.431 num bits 16 BDS_Ref 252.24
DecScale 0
> > > BinScale -13
> > >
> > >
> > >
> > >
> > > Any suggestion about how to solve this problem?
> > >
> > > Thanks,
> > > Samuele
> > >
> > >
> >
> >
> > --
> > Julie Prestopnik
> > National Center for Atmospheric Research
> > Research Applications Laboratory
> > Phone: 303.497.8399
> > Email: jpresto at ucar.edu
> >
> >
>
>
--
Julie Prestopnik
National Center for Atmospheric Research
Research Applications Laboratory
Phone: 303.497.8399
Email: jpresto at ucar.edu
------------------------------------------------
Subject: Problems with grid_stat used with ECMWF observations
From: Samuele Salis
Time: Wed Mar 30 10:22:39 2016
Ok Julie,
so it looks like there are not remarkable problems, I mean I can trust
the
result I am going to obtain.
Maybe I will change the table, because the field I verify are quite
recurring.
Thanks again,
Samuele
On 25 March 2016 at 17:23, Julie Prestopnik via RT <met_help at ucar.edu>
wrote:
> Hi Samuele. I'm glad that you have not had any more errors and that
MET is
> producing output.
>
> All the nceptab_flat.txt (and all GRIB1/GRIB2 tables) file does is
assign
> abbreviations and units to a sequence of identifying numbers. It is
> currently set for NCEP usage. You are welcome to change the
contents of
> that table to change the abbreviations used.
>
> At runtime, MET is reading the installed version of that file from
> INSTALL/share/met/table_files/nceptab_flat.txt, where INSTALL
indicates
> MET's installation path. You can either modify that files manually,
or you
> can modify the met-5.1_bugfix/data/table_files/nceptab_flat.txt file
and
> then run "make install" to install the file in the share area.
>
> As I mentioned before, we will be adding better support for this in
the
> next release.
>
> Please let me know if you have any other questions.
>
> Thanks,
> Julie
>
> On Fri, Mar 25, 2016 at 1:44 AM, Samuele Salis via RT
<met_help at ucar.edu>
> wrote:
>
> >
> > <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=75640 >
> >
> > Hi Julie,
> >
> > I modified the GridStatConfig_default file as you suggested and
> > I have not message errors any more and MET produces all the
> > files requested.
> >
> > If I check the statistical values they look reasonable, but still
I have
> a
> > little doubt because I saw that the ECMWF variable (temperature
> > at 500 hPa) is considered in the grid_stat standard output as
> > 130 128 "ELEV" "Ocean surface elevation relative to geoid" "m"
> > in the nceptab_flat.txt table
> >
> > DEBUG 3: Reading observation data for ELEV/P500.
> >
> > I think all is fine, but I would like your opinion about this
> > topic.
> >
> > Thanks,
> > Samuele
> >
> > On 24 March 2016 at 20:43, Julie Prestopnik via RT
<met_help at ucar.edu>
> > wrote:
> >
> > > Hi Samuele. I see that you are having problems with GridStat
when
> using
> > > ECMWF data. Unfortunately, MET does not currently include
support for
> > the
> > > ECMWF usage of GRIB. We are aware that this is support that
would be
> > > useful and it is a priority for us to add support for it. We're
hoping
> > to
> > > include it in our next release.
> > >
> > > For now, one way around this is to modify your obs field. So,
instead
> of
> > > having obs = fcst, create an obs field that looks like the
forecast
> > field,
> > > but instead of using "name" use "GRIB1_ptv" and "GRIB1_rec" as I
show
> > > below, using the table number (128 in this case) for GRIB1_ptv
and
> using
> > > the grib code (130 in this case) for GRIB1_rec.
> > >
> > > obs = {
> > > message_type = [ "<Fill in your message_type here>" ];
> > > sid_exc = [];
> > >
> > > field = [
> > > {
> > > GRIB1_ptv = 128;
> > > GRIB1_rec = 130;
> > > level = [ "<Fill in the level here>" ];
> > > cat_thresh = [ ];
> > > }
> > > ];
> > >
> > > }
> > >
> > > Please give that a shot and let us know how it goes.
> > >
> > > Thanks,
> > > Julie
> > >
> > > On Thu, Mar 24, 2016 at 10:19 AM, Samuele Salis via RT <
> > met_help at ucar.edu>
> > > wrote:
> > >
> > > >
> > > > Thu Mar 24 10:19:09 2016: Request 75640 was acted upon.
> > > > Transaction: Ticket created by samuelesalis at gmail.com
> > > > Queue: met_help
> > > > Subject: Problems with grid_stat used with ECMWF
observations
> > > > Owner: Nobody
> > > > Requestors: samuelesalis at gmail.com
> > > > Status: new
> > > > Ticket <URL:
> https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=75640
> > >
> > > >
> > > >
> > > > Hello everyone,
> > > >
> > > > I have a problem when I use grid_stat comparing WRF forecasted
data
> > with
> > > > ECMWF analysis.
> > > > This problem does not happen when I compare with NCEP
analysis.
> > > > I modified GridStatConfig_default only in this part:
> > > > fcst = {
> > > >
> > > > field = [
> > > > {
> > > > name = "TMP";
> > > > level = [ "P500" ];
> > > > // cat_thresh = [ >0.0, >=5.0 ];
> > > > }
> > > > ];
> > > >
> > > > }
> > > >
> > > > obs = fcst;
> > > >
> > > >
> > > >
> > > >
> > > > When I use
> > > > ./grid_stat WRF.fcst.grib NCEP.obs.grib
GridStatConfig_modified
> > > > all is fine.
> > > >
> > > >
> > > >
> > > > When I use
> > > > ./grid_stat WRF.fcst.grib ECMWF.obs.grib
GridStatConfig_modified -v 4
> > > > I get the this output
> > > >
> > > > DEBUG 1: Default Config File:
> > > >
> > > >
> > >
> >
> /home/ssalis/Documents/WRF/Build_WRF/MET/met-
5.1_bugfix/share/met/config/GridStatConfig_default
> > > > DEBUG 1: User Config File: GridStatConfig_modified
> > > > DEBUG 4: Met2dDataFileFactory::new_met_2d_data_file() ->
created new
> > > > Met2dDataFile object of type "FileType_Gb1".
> > > > DEBUG 4: Met2dDataFileFactory::new_met_2d_data_file() ->
created new
> > > > Met2dDataFile object of type "FileType_Gb1".
> > > > DEBUG 4: VarInfoFactory::new_var_info() -> created new VarInfo
object
> > of
> > > > type "FileType_Gb1".
> > > > DEBUG 4: VarInfoFactory::new_var_info() -> created new VarInfo
object
> > of
> > > > type "FileType_Gb1".
> > > > DEBUG 3: Use the matching forecast and observation grids.
> > > > DEBUG 3: Grid Definition: Projection: Lat/Lon Nx: 41 Ny: 31
lat_ll:
> > > 38.000
> > > > lon_ll: -6.500 delta_lat: 0.125 delta_lon: 0.125
> > > > GSL_RNG_TYPE=mt19937
> > > > GSL_RNG_SEED=1209005142
> > > > DEBUG 1: Forecast File: WRF.fcst.grib
> > > > DEBUG 1: Observation File: ECMWF.obs.grib
> > > > DEBUG 3: Reading forecast data for TMP/P500.
> > > > DEBUG 3: MetGrib1DataFile::data_plane_array() -> Found range
match
> for
> > > > VarInfo "TMP/P500" in GRIB record 122 of GRIB file
"WRF.fcst.grib".
> > > > DEBUG 3: MetGrib1DataFile::data_plane_array() -> Found 1 GRIB
records
> > > > matching VarInfo "TMP/P500" in GRIB file "WRF.fcst.grib".
> > > > DEBUG 3: Reading observation data for TMP/P500.
> > > > DEBUG 3: MetGrib1DataFile::data_plane_array() -> Found 0 GRIB
records
> > > > matching VarInfo "TMP/P500" in GRIB file "ECMWF.obs.grib".
> > > > WARNING:
> > > > WARNING: MetGrib1DataFile::data_plane() -> No exact match
found for
> > > VarInfo
> > > > "TMP/P500" in GRIB file "ECMWF.obs.grib".
> > > > WARNING:
> > > > WARNING:
> > > > WARNING: process_scores() -> TMP/P500 not found in file:
> ECMWF.obs.grib
> > > > WARNING:
> > > > DEBUG 2:
> > > > DEBUG 2:
> > > >
> > > >
> > >
> >
>
--------------------------------------------------------------------------------
> > > > DEBUG 2:
> > > >
> > > >
> > > >
> > > >
> > > >
> > > > Checking my ECMWF.obs.grib with wgrib I obtain
> > > >
> > > > rec 3:5300:date 2016032212 T kpds5=130 kpds6=100 kpds7=1
levels=(0,1)
> > > > grid=255 1 mb ECext=1 class=operations type=analysis
stream=1025
> > > > Version=0001 anl:
> > > > T=Temperature [K]
> > > > timerange 0 P1 0 P2 0 TimeU 1 nx 41 ny 31 GDS grid 0
num_in_ave 0
> > > > missing 0
> > > > center 98 subcenter 0 process 146 Table 128 scan: WE:NS
winds(N/S)
> > > > latlon: lat 41.750000 to 38.000000 by 0.125000 nxny 1271
> > > > long 6.500000 to 11.500000 by 0.125000, (41 x 31)
scan 0
> mode
> > > 128
> > > > bdsgrid 1
> > > > min/max data 252.24 257.431 num bits 16 BDS_Ref 252.24
DecScale
> 0
> > > > BinScale -13
> > > >
> > > >
> > > >
> > > >
> > > > Any suggestion about how to solve this problem?
> > > >
> > > > Thanks,
> > > > Samuele
> > > >
> > > >
> > >
> > >
> > > --
> > > Julie Prestopnik
> > > National Center for Atmospheric Research
> > > Research Applications Laboratory
> > > Phone: 303.497.8399
> > > Email: jpresto at ucar.edu
> > >
> > >
> >
> >
>
>
> --
> Julie Prestopnik
> National Center for Atmospheric Research
> Research Applications Laboratory
> Phone: 303.497.8399
> Email: jpresto at ucar.edu
>
>
------------------------------------------------
Subject: Problems with grid_stat used with ECMWF observations
From: Julie Prestopnik
Time: Wed Mar 30 10:25:35 2016
Thanks, Samuele. I'll go ahead and close this ticket. Please let us
know
if you have any other questions.
Thanks,
Julie
On Wed, Mar 30, 2016 at 10:22 AM, Samuele Salis via RT
<met_help at ucar.edu>
wrote:
>
> <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=75640 >
>
> Ok Julie,
>
> so it looks like there are not remarkable problems, I mean I can
trust the
> result I am going to obtain.
>
> Maybe I will change the table, because the field I verify are quite
> recurring.
>
> Thanks again,
> Samuele
>
> On 25 March 2016 at 17:23, Julie Prestopnik via RT
<met_help at ucar.edu>
> wrote:
>
> > Hi Samuele. I'm glad that you have not had any more errors and
that MET
> is
> > producing output.
> >
> > All the nceptab_flat.txt (and all GRIB1/GRIB2 tables) file does is
assign
> > abbreviations and units to a sequence of identifying numbers. It
is
> > currently set for NCEP usage. You are welcome to change the
contents of
> > that table to change the abbreviations used.
> >
> > At runtime, MET is reading the installed version of that file from
> > INSTALL/share/met/table_files/nceptab_flat.txt, where INSTALL
indicates
> > MET's installation path. You can either modify that files
manually, or
> you
> > can modify the met-5.1_bugfix/data/table_files/nceptab_flat.txt
file and
> > then run "make install" to install the file in the share area.
> >
> > As I mentioned before, we will be adding better support for this
in the
> > next release.
> >
> > Please let me know if you have any other questions.
> >
> > Thanks,
> > Julie
> >
> > On Fri, Mar 25, 2016 at 1:44 AM, Samuele Salis via RT
<met_help at ucar.edu
> >
> > wrote:
> >
> > >
> > > <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=75640 >
> > >
> > > Hi Julie,
> > >
> > > I modified the GridStatConfig_default file as you suggested and
> > > I have not message errors any more and MET produces all the
> > > files requested.
> > >
> > > If I check the statistical values they look reasonable, but
still I
> have
> > a
> > > little doubt because I saw that the ECMWF variable (temperature
> > > at 500 hPa) is considered in the grid_stat standard output as
> > > 130 128 "ELEV" "Ocean surface elevation relative to geoid" "m"
> > > in the nceptab_flat.txt table
> > >
> > > DEBUG 3: Reading observation data for ELEV/P500.
> > >
> > > I think all is fine, but I would like your opinion about this
> > > topic.
> > >
> > > Thanks,
> > > Samuele
> > >
> > > On 24 March 2016 at 20:43, Julie Prestopnik via RT
<met_help at ucar.edu>
> > > wrote:
> > >
> > > > Hi Samuele. I see that you are having problems with GridStat
when
> > using
> > > > ECMWF data. Unfortunately, MET does not currently include
support
> for
> > > the
> > > > ECMWF usage of GRIB. We are aware that this is support that
would be
> > > > useful and it is a priority for us to add support for it.
We're
> hoping
> > > to
> > > > include it in our next release.
> > > >
> > > > For now, one way around this is to modify your obs field. So,
> instead
> > of
> > > > having obs = fcst, create an obs field that looks like the
forecast
> > > field,
> > > > but instead of using "name" use "GRIB1_ptv" and "GRIB1_rec" as
I show
> > > > below, using the table number (128 in this case) for GRIB1_ptv
and
> > using
> > > > the grib code (130 in this case) for GRIB1_rec.
> > > >
> > > > obs = {
> > > > message_type = [ "<Fill in your message_type here>" ];
> > > > sid_exc = [];
> > > >
> > > > field = [
> > > > {
> > > > GRIB1_ptv = 128;
> > > > GRIB1_rec = 130;
> > > > level = [ "<Fill in the level here>" ];
> > > > cat_thresh = [ ];
> > > > }
> > > > ];
> > > >
> > > > }
> > > >
> > > > Please give that a shot and let us know how it goes.
> > > >
> > > > Thanks,
> > > > Julie
> > > >
> > > > On Thu, Mar 24, 2016 at 10:19 AM, Samuele Salis via RT <
> > > met_help at ucar.edu>
> > > > wrote:
> > > >
> > > > >
> > > > > Thu Mar 24 10:19:09 2016: Request 75640 was acted upon.
> > > > > Transaction: Ticket created by samuelesalis at gmail.com
> > > > > Queue: met_help
> > > > > Subject: Problems with grid_stat used with ECMWF
observations
> > > > > Owner: Nobody
> > > > > Requestors: samuelesalis at gmail.com
> > > > > Status: new
> > > > > Ticket <URL:
> > https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=75640
> > > >
> > > > >
> > > > >
> > > > > Hello everyone,
> > > > >
> > > > > I have a problem when I use grid_stat comparing WRF
forecasted data
> > > with
> > > > > ECMWF analysis.
> > > > > This problem does not happen when I compare with NCEP
analysis.
> > > > > I modified GridStatConfig_default only in this part:
> > > > > fcst = {
> > > > >
> > > > > field = [
> > > > > {
> > > > > name = "TMP";
> > > > > level = [ "P500" ];
> > > > > // cat_thresh = [ >0.0, >=5.0 ];
> > > > > }
> > > > > ];
> > > > >
> > > > > }
> > > > >
> > > > > obs = fcst;
> > > > >
> > > > >
> > > > >
> > > > >
> > > > > When I use
> > > > > ./grid_stat WRF.fcst.grib NCEP.obs.grib
GridStatConfig_modified
> > > > > all is fine.
> > > > >
> > > > >
> > > > >
> > > > > When I use
> > > > > ./grid_stat WRF.fcst.grib ECMWF.obs.grib
GridStatConfig_modified
> -v 4
> > > > > I get the this output
> > > > >
> > > > > DEBUG 1: Default Config File:
> > > > >
> > > > >
> > > >
> > >
> >
> /home/ssalis/Documents/WRF/Build_WRF/MET/met-
5.1_bugfix/share/met/config/GridStatConfig_default
> > > > > DEBUG 1: User Config File: GridStatConfig_modified
> > > > > DEBUG 4: Met2dDataFileFactory::new_met_2d_data_file() ->
created
> new
> > > > > Met2dDataFile object of type "FileType_Gb1".
> > > > > DEBUG 4: Met2dDataFileFactory::new_met_2d_data_file() ->
created
> new
> > > > > Met2dDataFile object of type "FileType_Gb1".
> > > > > DEBUG 4: VarInfoFactory::new_var_info() -> created new
VarInfo
> object
> > > of
> > > > > type "FileType_Gb1".
> > > > > DEBUG 4: VarInfoFactory::new_var_info() -> created new
VarInfo
> object
> > > of
> > > > > type "FileType_Gb1".
> > > > > DEBUG 3: Use the matching forecast and observation grids.
> > > > > DEBUG 3: Grid Definition: Projection: Lat/Lon Nx: 41 Ny: 31
lat_ll:
> > > > 38.000
> > > > > lon_ll: -6.500 delta_lat: 0.125 delta_lon: 0.125
> > > > > GSL_RNG_TYPE=mt19937
> > > > > GSL_RNG_SEED=1209005142
> > > > > DEBUG 1: Forecast File: WRF.fcst.grib
> > > > > DEBUG 1: Observation File: ECMWF.obs.grib
> > > > > DEBUG 3: Reading forecast data for TMP/P500.
> > > > > DEBUG 3: MetGrib1DataFile::data_plane_array() -> Found range
match
> > for
> > > > > VarInfo "TMP/P500" in GRIB record 122 of GRIB file
"WRF.fcst.grib".
> > > > > DEBUG 3: MetGrib1DataFile::data_plane_array() -> Found 1
GRIB
> records
> > > > > matching VarInfo "TMP/P500" in GRIB file "WRF.fcst.grib".
> > > > > DEBUG 3: Reading observation data for TMP/P500.
> > > > > DEBUG 3: MetGrib1DataFile::data_plane_array() -> Found 0
GRIB
> records
> > > > > matching VarInfo "TMP/P500" in GRIB file "ECMWF.obs.grib".
> > > > > WARNING:
> > > > > WARNING: MetGrib1DataFile::data_plane() -> No exact match
found for
> > > > VarInfo
> > > > > "TMP/P500" in GRIB file "ECMWF.obs.grib".
> > > > > WARNING:
> > > > > WARNING:
> > > > > WARNING: process_scores() -> TMP/P500 not found in file:
> > ECMWF.obs.grib
> > > > > WARNING:
> > > > > DEBUG 2:
> > > > > DEBUG 2:
> > > > >
> > > > >
> > > >
> > >
> >
>
--------------------------------------------------------------------------------
> > > > > DEBUG 2:
> > > > >
> > > > >
> > > > >
> > > > >
> > > > >
> > > > > Checking my ECMWF.obs.grib with wgrib I obtain
> > > > >
> > > > > rec 3:5300:date 2016032212 T kpds5=130 kpds6=100 kpds7=1
> levels=(0,1)
> > > > > grid=255 1 mb ECext=1 class=operations type=analysis
stream=1025
> > > > > Version=0001 anl:
> > > > > T=Temperature [K]
> > > > > timerange 0 P1 0 P2 0 TimeU 1 nx 41 ny 31 GDS grid 0
num_in_ave
> 0
> > > > > missing 0
> > > > > center 98 subcenter 0 process 146 Table 128 scan: WE:NS
> winds(N/S)
> > > > > latlon: lat 41.750000 to 38.000000 by 0.125000 nxny 1271
> > > > > long 6.500000 to 11.500000 by 0.125000, (41 x 31)
scan 0
> > mode
> > > > 128
> > > > > bdsgrid 1
> > > > > min/max data 252.24 257.431 num bits 16 BDS_Ref 252.24
> DecScale
> > 0
> > > > > BinScale -13
> > > > >
> > > > >
> > > > >
> > > > >
> > > > > Any suggestion about how to solve this problem?
> > > > >
> > > > > Thanks,
> > > > > Samuele
> > > > >
> > > > >
> > > >
> > > >
> > > > --
> > > > Julie Prestopnik
> > > > National Center for Atmospheric Research
> > > > Research Applications Laboratory
> > > > Phone: 303.497.8399
> > > > Email: jpresto at ucar.edu
> > > >
> > > >
> > >
> > >
> >
> >
> > --
> > Julie Prestopnik
> > National Center for Atmospheric Research
> > Research Applications Laboratory
> > Phone: 303.497.8399
> > Email: jpresto at ucar.edu
> >
> >
>
>
--
Julie Prestopnik
National Center for Atmospheric Research
Research Applications Laboratory
Phone: 303.497.8399
Email: jpresto at ucar.edu
------------------------------------------------
More information about the Met_help
mailing list