[Met_help] [rt.rap.ucar.edu #50422] History for Validation of wind speeds from mast
Paul Oldenburg via RT
met_help at ucar.edu
Thu Oct 13 08:53:41 MDT 2011
----------------------------------------------------------------
Initial Request
----------------------------------------------------------------
Dear met users,
I'm interested in using MET to validate WRF wind speed forecasts against observations from masts at (multiple) heights, up to 80m above the surface. Does anyone have experience of this? Looking at the documentation, it seems I might need to interpolate WRF output to the correct heights, and then pretend both the forecast and observations are the 10m wind? Or is there an easier way?
Thanks,
Sam.
----------------------------------------------------------------
Complete Ticket History
----------------------------------------------------------------
Subject: Validation of wind speeds from mast
From: Paul Oldenburg
Time: Mon Oct 03 09:58:24 2011
Sam,
What you have suggested would work. However, there is another way to
perform the verification that should yield closer
height matches between the forecast and obs. If you use the
relatively new Unified Post Processor (see
http://www.dtcenter.org/wrf-nmm/users/overview/upp_overview.php if you
don't have an instance), there is a way to
interpolate the forecast wind speeds at various heights above ground.
In the UPP config file,
[UPP_path]/parm/wrf_cntrl.parm, look for the lines containing the
following four settings, then change the first group
of 00000 in each list to 22220 as I have done here:
(SPE HUM AT FD HEIGHT) SCAL=( 5.0)
L=(22220 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(TEMP AT FD HEIGHTS ) SCAL=( 5.0)
L=(22220 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(U WIND AT FD HEIGHTS) SCAL=( 5.0)
L=(22220 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(V WIND AT FD HEIGHTS) SCAL=( 5.0)
L=(22220 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
I attached a version of this file with this change included. Please
move your original copy and replace it with the
attached. When you run UPP with this configuration, the output should
include interpolated WIND at 30m, 50m, 80m and
100m. Once you have performed this interpolation on the forecast
data, you can then specify the verification fields in
the point_stat config file like so:
fcst_field[] = [ "WIND/Z80" ];
obs_field[] = [ "WIND/Z70-90" ];
This would compare the interpolated 80m forecast wind speed to any
observation whose height is between 70m and 90m.
I'm curious about your verification task, and I was wondering if it
would be possible for you to send me your
observation and forecast data? If you are using ascii2nc to pre-
process your observations, can you please send me the
ascii obs file, too? I will try to perform the type of verification
that I am suggesting you try. If the data is too
large to email, you can upload it to our FTP site by following these
instructions:
http://www.dtcenter.org/met/users/support/met_help.php#ftp
If you have any questions, please let me know.
Thanks,
Paul
On 10/03/2011 09:17 AM, sam.hawkins at vattenfall.com via RT wrote:
>
> Mon Oct 03 09:17:06 2011: Request 50422 was acted upon.
> Transaction: Ticket created by sam.hawkins at vattenfall.com
> Queue: met_help
> Subject: Validation of wind speeds from mast
> Owner: Nobody
> Requestors: sam.hawkins at vattenfall.com
> Status: new
> Ticket <URL:
https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=50422 >
>
>
> Dear met users,
>
> I'm interested in using MET to validate WRF wind speed forecasts
against observations from masts at (multiple) heights, up to 80m
above the surface. Does anyone have experience of this? Looking at
the documentation, it seems I might need to interpolate WRF output to
the correct heights, and then pretend both the forecast and
observations are the 10m wind? Or is there an easier way?
>
> Thanks,
>
> Sam.
>
>
>
------------------------------------------------
Subject: Validation of wind speeds from mast
From: Paul Oldenburg
Time: Mon Oct 03 09:58:24 2011
KGTYPE******I5*******:(00255)********START OF THIS OUTPUT
GRID******
IMDLTY *I5* :(00125)
DATSET *A6* :(WRFPRS)
(PRESS ON MDL SFCS ) SCAL=( 3.0)
L=(11000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(HEIGHT ON MDL SFCS ) SCAL=( 6.0)
L=(11000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(TEMP ON MDL SFCS ) SCAL=( 4.0)
L=(11000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(POT TEMP ON MDL SFCS) SCAL=( 4.0)
L=(00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(DWPT TEMP ON MDL SFC) SCAL=( 5.0)
L=(11000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(SPEC HUM ON MDL SFCS) SCAL=( 3.0)
L=(11000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(REL HUM ON MDL SFCS ) SCAL=( 4.0)
L=(10000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(MST CNVG ON MDL SFCS) SCAL=( 3.0)
L=(00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(U WIND ON MDL SFCS ) SCAL=( 4.0)
L=(11000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(V WIND ON MDL SFCS ) SCAL=( 4.0)
L=(11000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(OMEGA ON MDL SFCS ) SCAL=( 5.0)
L=(00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(ABS VORT ON MDL SFCS) SCAL=( 3.0)
L=(00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(STRMFUNC ON MDL SFCS) SCAL=( 3.0)
L=(00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(TRBLNT KE ON MDL SFC) SCAL=( 3.0)
L=(00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(RCHDSN NO ON MDL SFC) SCAL=( 3.0)
L=(00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(RAIN ON MDL SFCS ) SCAL=( 6.0)
L=(00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(SNOW ON MDL SFCS ) SCAL=( 6.0)
L=(00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(F_RimeF ON MDL SFCS ) SCAL=( 3.0)
L=(00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(CONDENSATE MDL SFCS ) SCAL=( 3.0)
L=(00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(RADAR REFL MDL SFCS ) SCAL=( 4.0)
L=(00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(CLD WTR ON MDL SFCS ) SCAL=( 6.0)
L=(00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(CLD ICE ON MDL SFCS ) SCAL=( 6.0)
L=(00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(GRAUPEL ON MDL SFCS ) SCAL=( 6.0)
L=(00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(CLD FRAC ON MDL SFCS) SCAL=( 3.0)
L=(00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(MASTER LENGTH SCALE ) SCAL=( 3.0)
L=(00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(ASYMPT MSTR LEN SCL ) SCAL=( 3.0)
L=(00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(HEIGHT OF PRESS SFCS) SCAL=( 6.0)
L=(00000 01001 01010 10101 01010 10101 01010 11111 11111 10000 00000
00000 00000 00000)
(TEMP ON PRESS SFCS ) SCAL=( 4.0)
L=(00000 01001 01010 10101 01010 10101 01010 11111 11111 10000 00000
00000 00000 00000)
(POT TEMP ON P SFCS ) SCAL=( 4.0)
L=(00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(DWPT TEMP ON P SFCS ) SCAL=( 4.0)
L=(00000 01001 01010 10101 01010 10101 01010 11111 11111 10000 00000
00000 00000 00000)
(SPEC HUM ON P SFCS ) SCAL=( 4.0)
L=(00000 01001 01010 10101 01010 10101 01010 11111 11111 10000 00000
00000 00000 00000)
(REL HUMID ON P SFCS ) SCAL=( 4.0)
L=(00000 01001 01010 10101 01010 10101 01010 11111 11111 10000 00000
00000 00000 00000)
(MST CNVG ON P SFCS ) SCAL=( 3.0)
L=(00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(U WIND ON PRESS SFCS) SCAL=( 4.0)
L=(00000 01001 01010 10101 01010 10101 01010 11111 11111 10000 00000
00000 00000 00000)
(V WIND ON PRESS SFCS) SCAL=( 4.0)
L=(00000 01001 01010 10101 01010 10101 01010 11111 11111 10000 00000
00000 00000 00000)
(OMEGA ON PRESS SFCS ) SCAL=( 5.0)
L=(00000 01001 01010 10101 01010 10101 01010 11111 11111 10000 00000
00000 00000 00000)
(ABS VORT ON P SFCS ) SCAL=( 3.0)
L=(00000 01001 01010 10101 01010 10101 01010 11111 11111 10000 00000
00000 00000 00000)
(STRMFUNC ON P SFCS ) SCAL=( 3.0)
L=(00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(TRBLNT KE ON P SFCS ) SCAL=( 3.0)
L=(00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(CLOUD ICE ON P SFCS ) SCAL=( 6.0)
L=(00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(CLOUD WATR ON P SFCS) SCAL=( 6.0)
L=(00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(RAIN ON P SFCS ) SCAL=( 6.0)
L=(00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(SNOW ON P SFCS ) SCAL=( 6.0)
L=(00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(F_RimeF ON P SFCS ) SCAL=( 3.0)
L=(00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(RADAR REFL ON P SFCS) SCAL=( 4.0)
L=(00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(GRAUPEL ON P SFCS ) SCAL=( 6.0)
L=(10000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(MAPS SLP ) SCAL=( 6.0)
L=(10000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(MESINGER MEAN SLP ) SCAL=( 6.0)
L=(10000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(SHUELL MEAN SLP ) SCAL=( 6.0)
L=(10000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(SHELTER PRESSURE ) SCAL=( 6.0)
L=(00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(SHELTER TEMPERATURE ) SCAL=( 5.0)
L=(10000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(SHELTER SPEC HUMID ) SCAL=( 3.0)
L=(10000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(SHELTER DEWPOINT ) SCAL=( 5.0)
L=(10000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(SHELTER REL HUMID ) SCAL=( 4.0)
L=(10000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(SHELTER MIX RATIO ) SCAL=( 4.0)
L=(00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(U WIND AT ANEMOM HT ) SCAL=( 4.0)
L=(10000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(V WIND AT ANEMOM HT ) SCAL=( 4.0)
L=(10000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(POT TEMP AT 10 M ) SCAL=( 5.0)
L=(00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(SPEC HUM AT 10 M ) SCAL=( 3.0)
L=(00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(SURFACE PRESSURE ) SCAL=( 6.0)
L=(10000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(SURFACE HEIGHT ) SCAL=( 6.0)
L=(10000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(SURFACE POT TEMP ) SCAL=( 5.0)
L=(00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(SURFACE SPEC HUMID ) SCAL=( 3.0)
L=(00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(SURFACE DEWPOINT ) SCAL=( 4.0)
L=(00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(SURFACE REL HUMID ) SCAL=( 4.0)
L=(00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(SFC (SKIN) TEMPRATUR) SCAL=( 5.0)
L=(10000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(BOTTOM SOIL TEMP ) SCAL=( 5.0)
L=(10000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(SOIL TEMPERATURE ) SCAL=( 3.0)
L=(11111 10000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(SOIL MOISTURE ) SCAL=( 3.0)
L=(11111 10000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(LIQUID SOIL MOISTURE) SCAL=( 3.0)
L=(11110 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(TOTAL SOIL MOISTURE ) SCAL=( 3.0)
L=(10000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(PLANT CANOPY SFC WTR) SCAL=( 3.0)
L=(00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(SNOW WATER EQUIVALNT) SCAL=( 3.0)
L=(10000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(PERCENT SNOW COVER ) SCAL=( 3.0)
L=(10000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(SFC EXCHANGE COEF ) SCAL=( 3.0)
L=(00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(GREEN VEG COVER ) SCAL=( 3.0)
L=(10000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(VEGETATION TYPE ) SCAL=( 3.0)
L=(00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(SOIL TYPE ) SCAL=( 3.0)
L=(00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(SLOPE TYPE ) SCAL=( 3.0)
L=(00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(SNOW FREE ALBEDO ) SCAL=( 3.0)
L=(00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(MAXIMUM SNOW ALBEDO ) SCAL=( 3.0)
L=(00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(CANOPY CONDUCTANCE ) SCAL=( 3.0)
L=(00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(CANOPY COND SOLAR ) SCAL=( 3.0)
L=(00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(CANOPY COND TEMP ) SCAL=( 5.0)
L=(00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(CANOPY COND HUMID ) SCAL=( 3.0)
L=(00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(CANOPY COND SOILM ) SCAL=( 3.0)
L=(00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(SOIL MOIST REFERENCE) SCAL=( 3.0)
L=(00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(SOIL MOIST POROSITY ) SCAL=( 3.0)
L=(00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(NO OF ROOT LAYERS ) SCAL=( 1.0)
L=(00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(MIN STOMATAL RESIST ) SCAL=( 3.0)
L=(00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(SNOW DEPTH ) SCAL=( 6.0)
L=(10000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(AIR DRY SOIL MOIST ) SCAL=( 3.0)
L=(00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(SOIL MOIST WILT PT ) SCAL=( 3.0)
L=(00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(SOIL MOISTURE AVAIL ) SCAL=( 3.0)
L=(10000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(INST GROUND HEAT FLX) SCAL=( 4.0)
L=(10000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(LIFTED INDEX--SURFCE) SCAL=( 3.0)
L=(10000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(LIFTED INDEX--BEST ) SCAL=( 3.0)
L=(10000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(LIFTED INDEX--BNDLYR) SCAL=( 3.0)
L=(10000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(CNVCT AVBL POT ENRGY) SCAL=( 3.0)
L=(11110 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(CNVCT INHIBITION ) SCAL=( 3.0)
L=(11110 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(PRECIPITABLE WATER ) SCAL=( 3.0)
L=(10000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(STORM REL HELICITY ) SCAL=( 3.0)
L=(10000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(ACM TOTAL PRECIP ) SCAL=( 5.0)
L=(10000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(ACM CONVCTIVE PRECIP) SCAL=( 5.0)
L=(10000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(ACM GRD SCALE PRECIP) SCAL=( 5.0)
L=(10000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(ACM GRD SCALE SW ICE) SCAL=( 4.0)
L=(10000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(ACM SNOWFALL ) SCAL=( 5.0)
L=(10000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(BUCKET SNOW PRECIP ) SCAL=( 5.0)
L=(00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(BUCKET TOTAL PRECIP ) SCAL=( 5.0)
L=(00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(BUCKET CONV PRECIP ) SCAL=( 5.0)
L=(00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(BUCKET GRDSCALE PRCP) SCAL=( 5.0)
L=(00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(ACM SNOW TOTAL/MELT ) SCAL=( 4.0)
L=(10000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(ACM STORM SFC RNOFF ) SCAL=( 4.0)
L=(10000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(ACM BSFL-GDWR RNOFF ) SCAL=( 4.0)
L=(00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(INSTANT PRECIP TYPE ) SCAL=( 1.0)
L=(10000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(GSD PRECIP TYPE ) SCAL=( 1.0)
L=(10000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(INSTANT PRECIP RATE ) SCAL=( 3.0)
L=(10000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(CONV PRECIP RATE ) SCAL=( 3.0)
L=(10000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(FROZEN FRAC CLD SCHM) SCAL=( 4.0)
L=(00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(SEA SFC TEMPERATURE ) SCAL=( 5.0)
L=(10000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(LOW CLOUD FRACTION ) SCAL=( 3.0)
L=(10000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(MID CLOUD FRACTION ) SCAL=( 3.0)
L=(10000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(HIGH CLOUD FRACTION ) SCAL=( 3.0)
L=(10000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(AVG TOTAL CLD FRAC ) SCAL=( 3.0)
L=(10000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(AVG STRAT CLD FRAC ) SCAL=( 3.0)
L=(10000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(AVG CNVCT CLD FRAC ) SCAL=( 3.0)
L=(10000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(GSD CLD BOT HEIGHT ) SCAL=( 6.0)
L=(10000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(GSD CLD TOP HEIGHT ) SCAL=( 6.0)
L=(10000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(CLOUD BOT PRESSURE ) SCAL=( 6.0)
L=(10000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(CLOUD TOP PRESSURE ) SCAL=( 6.0)
L=(10000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(CLOUD TOP TEMPS ) SCAL=( 5.0)
L=(10000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(CLOUD BOTTOM HEIGHT ) SCAL=( 6.0)
L=(10000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(CLOUD TOP HEIGHT ) SCAL=( 6.0)
L=(10000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(CONV CLOUD BOT PRESS) SCAL=( 6.0)
L=(00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(CONV CLOUD TOP PRESS) SCAL=( 6.0)
L=(00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(SHAL CU CLD BOT PRES) SCAL=( 6.0)
L=(00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(SHAL CU CLD TOP PRES) SCAL=( 6.0)
L=(00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(DEEP CU CLD BOT PRES) SCAL=( 6.0)
L=(00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(DEEP CU CLD TOP PRES) SCAL=( 6.0)
L=(00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(GRID CLOUD BOT PRESS) SCAL=( 6.0)
L=(00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(GRID CLOUD TOP PRESS) SCAL=( 6.0)
L=(00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(CONV CLOUD FRACTION ) SCAL=( 3.0)
L=(10000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(MAX WIND PRESS LEVEL) SCAL=( 3.0)
L=(00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(MAX WIND HGHT LEVEL ) SCAL=( 3.0)
L=(00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(U COMP MAX WIND ) SCAL=( 5.0)
L=(00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(V COMP MAX WIND ) SCAL=( 5.0)
L=(00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(PBL HEIGHT ) SCAL=( 5.0)
L=(10000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(COMPOSITE RADAR REFL) SCAL=( 4.0)
L=(10000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(RADAR REFL AGL ) SCAL=( 4.0)
L=(00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(VISIBILITY ) SCAL=( 2.7)
L=(10000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(GSD VISIBILITY ) SCAL=( 2.7)
L=(10000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(LCL AGL HEIGHT ) SCAL=( 5.0)
L=(00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(LCL PRESSURE ) SCAL=( 6.0)
L=(00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(AVE GRDSCL RN TMPTDY) SCAL=( 3.0)
L=(00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(AVE CNVCT RN TMPTDY ) SCAL=( 2.7)
L=(00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(RADFLX CNVG TMP TNDY) SCAL=( 3.0)
L=(00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(SW RAD TEMP TNDY ) SCAL=( 3.0)
L=(00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(LW RAD TEMP TNDY ) SCAL=( 3.0)
L=(00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(INSTN OUT SFC SW RAD) SCAL=( 4.0)
L=(10000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(INSTN OUT SFC LW RAD) SCAL=( 4.0)
L=(10000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(AVE INCMG SFC SW RAD) SCAL=( 4.0)
L=(10000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(AVE INCMG SFC LW RAD) SCAL=( 4.0)
L=(10000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(AVE OUTGO SFC SW RAD) SCAL=( 4.0)
L=(10000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(AVE OUTGO SFC LW RAD) SCAL=( 4.0)
L=(10000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(BRIGHTNESS TEMP ) SCAL=( 5.0)
L=(00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(BRIGHTNESS TEMP NCAR) SCAL=( 5.0)
L=(10000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(INSTN INC SFC SW RAD) SCAL=( 4.0)
L=(10000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(INSTN INC SFC LW RAD) SCAL=( 4.0)
L=(10000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(INSTN CLR INC SFC SW) SCAL=( 4.0)
L=(10000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(ROUGHNESS LENGTH ) SCAL=( 2.7)
L=(10000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(FRICTION VELOCITY ) SCAL=( 3.0)
L=(10000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(SFC DRAG COEFFICIENT) SCAL=( 3.0)
L=(00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(SFC U WIND STRESS ) SCAL=( 3.0)
L=(00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(SFC V WIND STRESS ) SCAL=( 3.0)
L=(00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(AVE SFC SENHEAT FX ) SCAL=( 4.0)
L=(10000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(AVE GROUND HEAT FX ) SCAL=( 4.0)
L=(10000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(AVE SNO PHSCNG HT FX) SCAL=( 4.0)
L=(00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(AVE SFC LATHEAT FX ) SCAL=( 4.0)
L=(10000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(AVE SFC MOMENTUM FX ) SCAL=( 4.0)
L=(10000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(ACC SFC EVAPORATION ) SCAL=( 4.0)
L=(10000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(ACC POT EVAPORATION ) SCAL=( 4.0)
L=(10000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(INST SFC SENHEAT FX ) SCAL=( 3.0)
L=(10000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(INST GROUND HEAT FLX) SCAL=( 3.0)
L=(00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(INST SFC LATHEAT FX ) SCAL=( 3.0)
L=(10000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(LATITUDE ) SCAL=( 4.0)
L=(10000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(LONGITUDE ) SCAL=( 4.0)
L=(10000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(LAND/SEA MASK ) SCAL=( 1.0)
L=(10000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(SEA ICE MASK ) SCAL=( 1.0)
L=(10000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(SFC MIDDAY ALBEDO ) SCAL=( 3.0)
L=(00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(LIFT PCL LVL PRESS ) SCAL=( 3.0)
L=(00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(PRESS AT TROPOPAUSE ) SCAL=( 6.0)
L=(00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(HEIGHT AT TROPOPAUSE) SCAL=( 3.0)
L=(10000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(TEMP AT TROPOPAUSE ) SCAL=( 5.0)
L=(00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(POTENTL TEMP AT TROP) SCAL=( 5.0)
L=(00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(U WIND AT TROPOPAUSE) SCAL=( 5.0)
L=(00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(V WIND AT TROPOPAUSE) SCAL=( 5.0)
L=(00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(SHEAR AT TROPOPAUSE ) SCAL=( 3.0)
L=(00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(SPE HUM AT FD HEIGHT) SCAL=( 5.0)
L=(22220 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(TEMP AT FD HEIGHTS ) SCAL=( 5.0)
L=(22220 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(U WIND AT FD HEIGHTS) SCAL=( 5.0)
L=(22220 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(V WIND AT FD HEIGHTS) SCAL=( 5.0)
L=(22220 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(HEIGHT OF FRZ LVL ) SCAL=( 6.0)
L=(10000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(REL HUMID AT FRZ LVL) SCAL=( 4.0)
L=(00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(HIGHEST FREEZE LVL ) SCAL=( 5.0)
L=(10000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(PRESS IN BNDRY LYR ) SCAL=( 6.0)
L=(00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(TEMP IN BNDRY LYR ) SCAL=( 3.0)
L=(00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(POT TMP IN BNDRY LYR) SCAL=( 3.0)
L=(00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(DWPT IN BNDRY LYR ) SCAL=( 3.0)
L=(00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(SPC HUM IN BNDRY LYR) SCAL=( 3.0)
L=(10000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(REL HUM IN BNDRY LYR) SCAL=( 3.0)
L=(00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(MST CNV IN BNDRY LYR) SCAL=( 3.0)
L=(00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(P WATER IN BNDRY LYR) SCAL=( 3.0)
L=(10000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(U WIND IN BNDRY LYR ) SCAL=( 5.0)
L=(10000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(V WIND IN BNDRY LYR ) SCAL=( 5.0)
L=(10000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(OMEGA IN BNDRY LYR ) SCAL=( 5.0)
L=(00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(MAX 10m WIND SPEED ) SCAL=( 4.0)
L=(00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(MAX UPDRAFT VERT VEL) SCAL=( 3.0)
L=(00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(MAX DNDRAFT VERT VEL) SCAL=( 3.0)
L=(00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(MAX 1km REFLECTIVITY) SCAL=( 4.0)
L=(00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(MAX UPDRAFT HELICITY) SCAL=( 3.0)
L=(00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(MEAN VERT VEL ) SCAL=( 3.0)
L=(00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(MAX VERT INTEG GRAUP) SCAL=( 3.0)
L=(00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(LCL AGL HEIGHT ) SCAL=( 5.0)
L=(00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(U COMP STORM MOTION ) SCAL=( 4.0)
L=(00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(V COMP STORM MOTION ) SCAL=( 4.0)
L=(00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(U COMP 0-1 KM SHEAR ) SCAL=( 4.0)
L=(00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(V COMP 0-1 KM SHEAR ) SCAL=( 4.0)
L=(00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(U COMP 0-6 KM SHEAR ) SCAL=( 4.0)
L=(00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(V COMP 0-6 KM SHEAR ) SCAL=( 4.0)
L=(00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(SFC WIND GUST ) SCAL=( 3.0)
L=(10000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(AVE OUTGO TOA SW RAD) SCAL=( 4.0)
L=(00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(AVE OUTGO TOA LW RAD) SCAL=( 4.0)
L=(00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(INSTN OUT TOA LW RAD) SCAL=( 3.0)
L=(10000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(TOTAL CLD FRACTION ) SCAL=( 3.0)
L=(10000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(TOTAL COLUMN CLD WTR) SCAL=( 5.0)
L=(10000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(TOTAL COLUMN CLD ICE) SCAL=( 5.0)
L=(10000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(TOTAL COLUMN RAIN ) SCAL=( 5.0)
L=(10000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(TOTAL COLUMN SNOW ) SCAL=( 5.0)
L=(10000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(TOTAL COL CONDENSATE) SCAL=( 5.0)
L=(10000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(TOTAL COLD LIQUID ) SCAL=( 5.0)
L=(00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(TOTAL MELTING ICE ) SCAL=( 5.0)
L=(00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(CEILING ) SCAL=( 5.0)
L=(10000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(EMISSIVITY ) SCAL=( 4.0)
L=(00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(GOES TB - CH 2 ) SCAL=( 4.0)
L=(00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(GOES TB - CH 3 ) SCAL=( 4.0)
L=(00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(GOES TB - CH 4 ) SCAL=( 4.0)
L=(00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
(GOES TB - CH 5 ) SCAL=( 4.0)
L=(00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
***DONE WITH THIS GRID***
------------------------------------------------
Subject: Validation of wind speeds from mast
From: sam.hawkins at vattenfall.com
Time: Tue Oct 04 02:51:36 2011
Paul,
Thanks for your quick reply. I have the UPP compiled, so I will give
it a try. I don't actually have access to the mast data yet, that
will come in the next week or so. I'm just trying to set up all of the
tools I need first. However, I have some old mast data from another
site which I may use as a test. It's only up to 40m height. Would it
be possible to validate against that? Obviously I can change the
format.
Sam.
-----Original Message-----
From: Paul Oldenburg via RT [mailto:met_help at ucar.edu]
Sent: 03 October 2011 16:58
To: Hawkins Samuel Lennon (AE-DE)
Cc: met_help at mailman.ucar.edu
Subject: Re: [rt.rap.ucar.edu #50422] Validation of wind speeds from
mast
Sam,
What you have suggested would work. However, there is another way to
perform the verification that should yield closer height matches
between the forecast and obs. If you use the relatively new Unified
Post Processor (see http://www.dtcenter.org/wrf-
nmm/users/overview/upp_overview.php if you don't have an instance),
there is a way to interpolate the forecast wind speeds at various
heights above ground. In the UPP config file,
[UPP_path]/parm/wrf_cntrl.parm, look for the lines containing the
following four settings, then change the first group of 00000 in each
list to 22220 as I have done here:
(SPE HUM AT FD HEIGHT) SCAL=( 5.0)
L=(22220 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000) (TEMP AT FD HEIGHTS ) SCAL=( 5.0) L=(22220 00000
00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000) (U WIND AT FD HEIGHTS) SCAL=( 5.0) L=(22220 00000 00000 00000
00000 00000 00000 00000 00000 00000 00000 00000 00000 00000) (V WIND
AT FD HEIGHTS) SCAL=( 5.0) L=(22220 00000 00000 00000 00000 00000
00000 00000 00000 00000 00000 00000 00000 00000)
I attached a version of this file with this change included. Please
move your original copy and replace it with the attached. When you
run UPP with this configuration, the output should include
interpolated WIND at 30m, 50m, 80m and 100m. Once you have performed
this interpolation on the forecast data, you can then specify the
verification fields in the point_stat config file like so:
fcst_field[] = [ "WIND/Z80" ];
obs_field[] = [ "WIND/Z70-90" ];
This would compare the interpolated 80m forecast wind speed to any
observation whose height is between 70m and 90m.
I'm curious about your verification task, and I was wondering if it
would be possible for you to send me your observation and forecast
data? If you are using ascii2nc to pre-process your observations, can
you please send me the ascii obs file, too? I will try to perform the
type of verification that I am suggesting you try. If the data is too
large to email, you can upload it to our FTP site by following these
instructions:
http://www.dtcenter.org/met/users/support/met_help.php#ftp
If you have any questions, please let me know.
Thanks,
Paul
On 10/03/2011 09:17 AM, sam.hawkins at vattenfall.com via RT wrote:
>
> Mon Oct 03 09:17:06 2011: Request 50422 was acted upon.
> Transaction: Ticket created by sam.hawkins at vattenfall.com
> Queue: met_help
> Subject: Validation of wind speeds from mast
> Owner: Nobody
> Requestors: sam.hawkins at vattenfall.com
> Status: new
> Ticket <URL:
https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=50422
> >
>
>
> Dear met users,
>
> I'm interested in using MET to validate WRF wind speed forecasts
against observations from masts at (multiple) heights, up to 80m
above the surface. Does anyone have experience of this? Looking at
the documentation, it seems I might need to interpolate WRF output to
the correct heights, and then pretend both the forecast and
observations are the 10m wind? Or is there an easier way?
>
> Thanks,
>
> Sam.
>
>
>
------------------------------------------------
Subject: Re: [rt.rap.ucar.edu #50422] Validation of wind speeds from mast
From: Paul Oldenburg
Time: Tue Oct 04 08:11:06 2011
Sam,
You should be able to use any data you want as observations, as long
as you can get MET to read it. Once that is done,
the trick will be to set the fcst_field and obs_field settings in the
point_stat config file so that MET will match the
data. If you send me some of your test data, I can take a look.
Paul
On 10/04/2011 02:51 AM, sam.hawkins at vattenfall.com via RT wrote:
>
> <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=50422 >
>
> Paul,
>
> Thanks for your quick reply. I have the UPP compiled, so I will give
it a try. I don't actually have access to the mast data yet, that
will come in the next week or so. I'm just trying to set up all of the
tools I need first. However, I have some old mast data from another
site which I may use as a test. It's only up to 40m height. Would it
be possible to validate against that? Obviously I can change the
format.
>
> Sam.
>
>
>
> -----Original Message-----
> From: Paul Oldenburg via RT [mailto:met_help at ucar.edu]
> Sent: 03 October 2011 16:58
> To: Hawkins Samuel Lennon (AE-DE)
> Cc: met_help at mailman.ucar.edu
> Subject: Re: [rt.rap.ucar.edu #50422] Validation of wind speeds from
mast
>
> Sam,
>
> What you have suggested would work. However, there is another way
to perform the verification that should yield closer height matches
between the forecast and obs. If you use the relatively new Unified
Post Processor (see http://www.dtcenter.org/wrf-
nmm/users/overview/upp_overview.php if you don't have an instance),
there is a way to interpolate the forecast wind speeds at various
heights above ground. In the UPP config file,
[UPP_path]/parm/wrf_cntrl.parm, look for the lines containing the
following four settings, then change the first group of 00000 in each
list to 22220 as I have done here:
>
> (SPE HUM AT FD HEIGHT) SCAL=( 5.0)
> L=(22220 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000 00000) (TEMP AT FD HEIGHTS ) SCAL=( 5.0) L=(22220
00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000) (U WIND AT FD HEIGHTS) SCAL=( 5.0) L=(22220 00000 00000
00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000) (V
WIND AT FD HEIGHTS) SCAL=( 5.0) L=(22220 00000 00000 00000 00000
00000 00000 00000 00000 00000 00000 00000 00000 00000)
>
> I attached a version of this file with this change included. Please
move your original copy and replace it with the attached. When you
run UPP with this configuration, the output should include
interpolated WIND at 30m, 50m, 80m and 100m. Once you have performed
this interpolation on the forecast data, you can then specify the
verification fields in the point_stat config file like so:
>
> fcst_field[] = [ "WIND/Z80" ];
> obs_field[] = [ "WIND/Z70-90" ];
>
> This would compare the interpolated 80m forecast wind speed to any
observation whose height is between 70m and 90m.
>
> I'm curious about your verification task, and I was wondering if it
would be possible for you to send me your observation and forecast
data? If you are using ascii2nc to pre-process your observations, can
you please send me the ascii obs file, too? I will try to perform the
type of verification that I am suggesting you try. If the data is too
large to email, you can upload it to our FTP site by following these
instructions:
>
> http://www.dtcenter.org/met/users/support/met_help.php#ftp
>
> If you have any questions, please let me know.
>
> Thanks,
>
> Paul
>
>
> On 10/03/2011 09:17 AM, sam.hawkins at vattenfall.com via RT wrote:
>>
>> Mon Oct 03 09:17:06 2011: Request 50422 was acted upon.
>> Transaction: Ticket created by sam.hawkins at vattenfall.com
>> Queue: met_help
>> Subject: Validation of wind speeds from mast
>> Owner: Nobody
>> Requestors: sam.hawkins at vattenfall.com
>> Status: new
>> Ticket <URL:
https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=50422
>>>
>>
>>
>> Dear met users,
>>
>> I'm interested in using MET to validate WRF wind speed forecasts
against observations from masts at (multiple) heights, up to 80m
above the surface. Does anyone have experience of this? Looking at
the documentation, it seems I might need to interpolate WRF output to
the correct heights, and then pretend both the forecast and
observations are the 10m wind? Or is there an easier way?
>>
>> Thanks,
>>
>> Sam.
>>
>>
>>
>
>
------------------------------------------------
Subject: Re: [rt.rap.ucar.edu #50422] Validation of wind speeds from mast
From: Paul Oldenburg
Time: Tue Oct 04 08:12:42 2011
Sam,
I just noticed your attached mast data. I'll try to set up an
experiment and get some matched pairs with some WRF data
that we have.
Thanks,
Paul
On 10/04/2011 08:11 AM, Paul Oldenburg wrote:
> Sam,
>
> You should be able to use any data you want as observations, as long
as you can get MET to read it. Once that is done,
> the trick will be to set the fcst_field and obs_field settings in
the point_stat config file so that MET will match the
> data. If you send me some of your test data, I can take a look.
>
> Paul
>
>
> On 10/04/2011 02:51 AM, sam.hawkins at vattenfall.com via RT wrote:
>>
>> <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=50422 >
>>
>> Paul,
>>
>> Thanks for your quick reply. I have the UPP compiled, so I will
give it a try. I don't actually have access to the mast data yet,
that will come in the next week or so. I'm just trying to set up all
of the tools I need first. However, I have some old mast data from
another site which I may use as a test. It's only up to 40m height.
Would it be possible to validate against that? Obviously I can change
the format.
>>
>> Sam.
>>
>>
>>
>> -----Original Message-----
>> From: Paul Oldenburg via RT [mailto:met_help at ucar.edu]
>> Sent: 03 October 2011 16:58
>> To: Hawkins Samuel Lennon (AE-DE)
>> Cc: met_help at mailman.ucar.edu
>> Subject: Re: [rt.rap.ucar.edu #50422] Validation of wind speeds
from mast
>>
>> Sam,
>>
>> What you have suggested would work. However, there is another way
to perform the verification that should yield closer height matches
between the forecast and obs. If you use the relatively new Unified
Post Processor (see http://www.dtcenter.org/wrf-
nmm/users/overview/upp_overview.php if you don't have an instance),
there is a way to interpolate the forecast wind speeds at various
heights above ground. In the UPP config file,
[UPP_path]/parm/wrf_cntrl.parm, look for the lines containing the
following four settings, then change the first group of 00000 in each
list to 22220 as I have done here:
>>
>> (SPE HUM AT FD HEIGHT) SCAL=( 5.0)
>> L=(22220 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000 00000) (TEMP AT FD HEIGHTS ) SCAL=( 5.0) L=(22220
00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000) (U WIND AT FD HEIGHTS) SCAL=( 5.0) L=(22220 00000 00000
00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000) (V
WIND AT FD HEIGHTS) SCAL=( 5.0) L=(22220 00000 00000 00000 00000
00000 00000 00000 00000 00000 00000 00000 00000 00000)
>>
>> I attached a version of this file with this change included.
Please move your original copy and replace it with the attached. When
you run UPP with this configuration, the output should include
interpolated WIND at 30m, 50m, 80m and 100m. Once you have performed
this interpolation on the forecast data, you can then specify the
verification fields in the point_stat config file like so:
>>
>> fcst_field[] = [ "WIND/Z80" ];
>> obs_field[] = [ "WIND/Z70-90" ];
>>
>> This would compare the interpolated 80m forecast wind speed to any
observation whose height is between 70m and 90m.
>>
>> I'm curious about your verification task, and I was wondering if it
would be possible for you to send me your observation and forecast
data? If you are using ascii2nc to pre-process your observations, can
you please send me the ascii obs file, too? I will try to perform the
type of verification that I am suggesting you try. If the data is too
large to email, you can upload it to our FTP site by following these
instructions:
>>
>> http://www.dtcenter.org/met/users/support/met_help.php#ftp
>>
>> If you have any questions, please let me know.
>>
>> Thanks,
>>
>> Paul
>>
>>
>> On 10/03/2011 09:17 AM, sam.hawkins at vattenfall.com via RT wrote:
>>>
>>> Mon Oct 03 09:17:06 2011: Request 50422 was acted upon.
>>> Transaction: Ticket created by sam.hawkins at vattenfall.com
>>> Queue: met_help
>>> Subject: Validation of wind speeds from mast
>>> Owner: Nobody
>>> Requestors: sam.hawkins at vattenfall.com
>>> Status: new
>>> Ticket <URL:
https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=50422
>>>>
>>>
>>>
>>> Dear met users,
>>>
>>> I'm interested in using MET to validate WRF wind speed forecasts
against observations from masts at (multiple) heights, up to 80m
above the surface. Does anyone have experience of this? Looking at
the documentation, it seems I might need to interpolate WRF output to
the correct heights, and then pretend both the forecast and
observations are the 10m wind? Or is there an easier way?
>>>
>>> Thanks,
>>>
>>> Sam.
>>>
>>>
>>>
>>
>>
>
------------------------------------------------
Subject: Validation of wind speeds from mast
From: sam.hawkins at vattenfall.com
Time: Tue Oct 04 09:37:27 2011
Paul,
Thanks for looking into it. I've never had much success with any of
the WRF post-processors. I've just tried running UPP but it produces
output files of zero size.
I've tried turning eveything off in the wrf_cntrl.parm file, apart
from U wind at anemometer height and V wind at anemoter height, but I
still get output files of zero size. I've attached a log from running
unipost. Is it best to direct UPP queries to wrfhelp?
Sam.
-----Original Message-----
From: Paul Oldenburg via RT [mailto:met_help at ucar.edu]
Sent: 04 October 2011 15:13
To: Hawkins Samuel Lennon (AE-DE)
Cc: met_help at mailman.ucar.edu
Subject: Re: [rt.rap.ucar.edu #50422] Validation of wind speeds from
mast
Sam,
I just noticed your attached mast data. I'll try to set up an
experiment and get some matched pairs with some WRF data that we have.
Thanks,
Paul
On 10/04/2011 08:11 AM, Paul Oldenburg wrote:
> Sam,
>
> You should be able to use any data you want as observations, as long
> as you can get MET to read it. Once that is done, the trick will be
> to set the fcst_field and obs_field settings in the point_stat
config file so that MET will match the data. If you send me some of
your test data, I can take a look.
>
> Paul
>
>
> On 10/04/2011 02:51 AM, sam.hawkins at vattenfall.com via RT wrote:
>>
>> <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=50422 >
>>
>> Paul,
>>
>> Thanks for your quick reply. I have the UPP compiled, so I will
give it a try. I don't actually have access to the mast data yet,
that will come in the next week or so. I'm just trying to set up all
of the tools I need first. However, I have some old mast data from
another site which I may use as a test. It's only up to 40m height.
Would it be possible to validate against that? Obviously I can change
the format.
>>
>> Sam.
>>
>>
>>
>> -----Original Message-----
>> From: Paul Oldenburg via RT [mailto:met_help at ucar.edu]
>> Sent: 03 October 2011 16:58
>> To: Hawkins Samuel Lennon (AE-DE)
>> Cc: met_help at mailman.ucar.edu
>> Subject: Re: [rt.rap.ucar.edu #50422] Validation of wind speeds
from
>> mast
>>
>> Sam,
>>
>> What you have suggested would work. However, there is another way
to perform the verification that should yield closer height matches
between the forecast and obs. If you use the relatively new Unified
Post Processor (see http://www.dtcenter.org/wrf-
nmm/users/overview/upp_overview.php if you don't have an instance),
there is a way to interpolate the forecast wind speeds at various
heights above ground. In the UPP config file,
[UPP_path]/parm/wrf_cntrl.parm, look for the lines containing the
following four settings, then change the first group of 00000 in each
list to 22220 as I have done here:
>>
>> (SPE HUM AT FD HEIGHT) SCAL=( 5.0)
>> L=(22220 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000
>> 00000 00000 00000) (TEMP AT FD HEIGHTS ) SCAL=( 5.0) L=(22220
>> 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
>> 00000 00000) (U WIND AT FD HEIGHTS) SCAL=( 5.0) L=(22220 00000
>> 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
>> 00000) (V WIND AT FD HEIGHTS) SCAL=( 5.0) L=(22220 00000 00000
>> 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000)
>>
>> I attached a version of this file with this change included.
Please move your original copy and replace it with the attached. When
you run UPP with this configuration, the output should include
interpolated WIND at 30m, 50m, 80m and 100m. Once you have performed
this interpolation on the forecast data, you can then specify the
verification fields in the point_stat config file like so:
>>
>> fcst_field[] = [ "WIND/Z80" ];
>> obs_field[] = [ "WIND/Z70-90" ];
>>
>> This would compare the interpolated 80m forecast wind speed to any
observation whose height is between 70m and 90m.
>>
>> I'm curious about your verification task, and I was wondering if it
would be possible for you to send me your observation and forecast
data? If you are using ascii2nc to pre-process your observations, can
you please send me the ascii obs file, too? I will try to perform the
type of verification that I am suggesting you try. If the data is too
large to email, you can upload it to our FTP site by following these
instructions:
>>
>> http://www.dtcenter.org/met/users/support/met_help.php#ftp
>>
>> If you have any questions, please let me know.
>>
>> Thanks,
>>
>> Paul
>>
>>
>> On 10/03/2011 09:17 AM, sam.hawkins at vattenfall.com via RT wrote:
>>>
>>> Mon Oct 03 09:17:06 2011: Request 50422 was acted upon.
>>> Transaction: Ticket created by sam.hawkins at vattenfall.com
>>> Queue: met_help
>>> Subject: Validation of wind speeds from mast
>>> Owner: Nobody
>>> Requestors: sam.hawkins at vattenfall.com
>>> Status: new
>>> Ticket <URL:
>>> https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=50422
>>>>
>>>
>>>
>>> Dear met users,
>>>
>>> I'm interested in using MET to validate WRF wind speed forecasts
against observations from masts at (multiple) heights, up to 80m
above the surface. Does anyone have experience of this? Looking at
the documentation, it seems I might need to interpolate WRF output to
the correct heights, and then pretend both the forecast and
observations are the 10m wind? Or is there an easier way?
>>>
>>> Thanks,
>>>
>>> Sam.
>>>
>>>
>>>
>>
>>
>
------------------------------------------------
Subject: Re: [rt.rap.ucar.edu #50422] Validation of wind speeds from mast
From: Paul Oldenburg
Time: Tue Oct 04 09:39:54 2011
Sam,
I'm working on the exact same thing myself. In general, it is best to
send UPP questions to WRF help, but in this case,
I'll try to figure out it. Of course, if I fail to figure it out, the
next step will be to engage WRF help (which
happens to be down the hall from me).
Can you please send me one of your WRF output files? If it's too big
for email, please upload to our FTP site (URL below).
Thanks,
Paul
On 10/04/2011 09:37 AM, sam.hawkins at vattenfall.com via RT wrote:
>
> <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=50422 >
>
> Paul,
>
> Thanks for looking into it. I've never had much success with any of
the WRF post-processors. I've just tried running UPP but it produces
output files of zero size.
>
> I've tried turning eveything off in the wrf_cntrl.parm file, apart
from U wind at anemometer height and V wind at anemoter height, but I
still get output files of zero size. I've attached a log from running
unipost. Is it best to direct UPP queries to wrfhelp?
>
> Sam.
>
>
>
> -----Original Message-----
> From: Paul Oldenburg via RT [mailto:met_help at ucar.edu]
> Sent: 04 October 2011 15:13
> To: Hawkins Samuel Lennon (AE-DE)
> Cc: met_help at mailman.ucar.edu
> Subject: Re: [rt.rap.ucar.edu #50422] Validation of wind speeds from
mast
>
> Sam,
>
> I just noticed your attached mast data. I'll try to set up an
experiment and get some matched pairs with some WRF data that we have.
>
> Thanks,
>
> Paul
>
>
> On 10/04/2011 08:11 AM, Paul Oldenburg wrote:
>> Sam,
>>
>> You should be able to use any data you want as observations, as
long
>> as you can get MET to read it. Once that is done, the trick will
be
>> to set the fcst_field and obs_field settings in the point_stat
config file so that MET will match the data. If you send me some of
your test data, I can take a look.
>>
>> Paul
>>
>>
>> On 10/04/2011 02:51 AM, sam.hawkins at vattenfall.com via RT wrote:
>>>
>>> <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=50422 >
>>>
>>> Paul,
>>>
>>> Thanks for your quick reply. I have the UPP compiled, so I will
give it a try. I don't actually have access to the mast data yet,
that will come in the next week or so. I'm just trying to set up all
of the tools I need first. However, I have some old mast data from
another site which I may use as a test. It's only up to 40m height.
Would it be possible to validate against that? Obviously I can change
the format.
>>>
>>> Sam.
>>>
>>>
>>>
>>> -----Original Message-----
>>> From: Paul Oldenburg via RT [mailto:met_help at ucar.edu]
>>> Sent: 03 October 2011 16:58
>>> To: Hawkins Samuel Lennon (AE-DE)
>>> Cc: met_help at mailman.ucar.edu
>>> Subject: Re: [rt.rap.ucar.edu #50422] Validation of wind speeds
from
>>> mast
>>>
>>> Sam,
>>>
>>> What you have suggested would work. However, there is another way
to perform the verification that should yield closer height matches
between the forecast and obs. If you use the relatively new Unified
Post Processor (see http://www.dtcenter.org/wrf-
nmm/users/overview/upp_overview.php if you don't have an instance),
there is a way to interpolate the forecast wind speeds at various
heights above ground. In the UPP config file,
[UPP_path]/parm/wrf_cntrl.parm, look for the lines containing the
following four settings, then change the first group of 00000 in each
list to 22220 as I have done here:
>>>
>>> (SPE HUM AT FD HEIGHT) SCAL=( 5.0)
>>> L=(22220 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000
>>> 00000 00000 00000) (TEMP AT FD HEIGHTS ) SCAL=( 5.0) L=(22220
>>> 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
>>> 00000 00000) (U WIND AT FD HEIGHTS) SCAL=( 5.0) L=(22220 00000
>>> 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
>>> 00000) (V WIND AT FD HEIGHTS) SCAL=( 5.0) L=(22220 00000 00000
>>> 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000)
>>>
>>> I attached a version of this file with this change included.
Please move your original copy and replace it with the attached. When
you run UPP with this configuration, the output should include
interpolated WIND at 30m, 50m, 80m and 100m. Once you have performed
this interpolation on the forecast data, you can then specify the
verification fields in the point_stat config file like so:
>>>
>>> fcst_field[] = [ "WIND/Z80" ];
>>> obs_field[] = [ "WIND/Z70-90" ];
>>>
>>> This would compare the interpolated 80m forecast wind speed to any
observation whose height is between 70m and 90m.
>>>
>>> I'm curious about your verification task, and I was wondering if
it would be possible for you to send me your observation and forecast
data? If you are using ascii2nc to pre-process your observations, can
you please send me the ascii obs file, too? I will try to perform the
type of verification that I am suggesting you try. If the data is too
large to email, you can upload it to our FTP site by following these
instructions:
>>>
>>> http://www.dtcenter.org/met/users/support/met_help.php#ftp
>>>
>>> If you have any questions, please let me know.
>>>
>>> Thanks,
>>>
>>> Paul
>>>
>>>
>>> On 10/03/2011 09:17 AM, sam.hawkins at vattenfall.com via RT wrote:
>>>>
>>>> Mon Oct 03 09:17:06 2011: Request 50422 was acted upon.
>>>> Transaction: Ticket created by sam.hawkins at vattenfall.com
>>>> Queue: met_help
>>>> Subject: Validation of wind speeds from mast
>>>> Owner: Nobody
>>>> Requestors: sam.hawkins at vattenfall.com
>>>> Status: new
>>>> Ticket <URL:
>>>> https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=50422
>>>>>
>>>>
>>>>
>>>> Dear met users,
>>>>
>>>> I'm interested in using MET to validate WRF wind speed forecasts
against observations from masts at (multiple) heights, up to 80m
above the surface. Does anyone have experience of this? Looking at
the documentation, it seems I might need to interpolate WRF output to
the correct heights, and then pretend both the forecast and
observations are the 10m wind? Or is there an easier way?
>>>>
>>>> Thanks,
>>>>
>>>> Sam.
>>>>
>>>>
>>>>
>>>
>>>
>>
>
>
------------------------------------------------
Subject: RE: [rt.rap.ucar.edu #50422] Validation of wind speeds from mast
From: sam.hawkins at vattenfall.com
Time: Tue Oct 04 10:18:30 2011
Paul,
I'll send you a wrfout file a little later when I get home - I'm
sitting behind a corporate firewall which seems to prevent me sending
things via ftp.
The problem may be related to my domain. I set up a non-rotated 'lat-
lon' projection as a quick test, which is probably a bad idea at this
latitude. I'm going to try something similar with a lambert projection
and see if I have the same problem.
Sam.
-----Original Message-----
From: Paul Oldenburg via RT [mailto:met_help at ucar.edu]
Sent: 04 October 2011 16:40
To: Hawkins Samuel Lennon (AE-DE)
Cc: met_help at mailman.ucar.edu
Subject: Re: [rt.rap.ucar.edu #50422] Validation of wind speeds from
mast
Sam,
I'm working on the exact same thing myself. In general, it is best to
send UPP questions to WRF help, but in this case, I'll try to figure
out it. Of course, if I fail to figure it out, the next step will be
to engage WRF help (which happens to be down the hall from me).
Can you please send me one of your WRF output files? If it's too big
for email, please upload to our FTP site (URL below).
Thanks,
Paul
On 10/04/2011 09:37 AM, sam.hawkins at vattenfall.com via RT wrote:
>
> <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=50422 >
>
> Paul,
>
> Thanks for looking into it. I've never had much success with any of
the WRF post-processors. I've just tried running UPP but it produces
output files of zero size.
>
> I've tried turning eveything off in the wrf_cntrl.parm file, apart
from U wind at anemometer height and V wind at anemoter height, but I
still get output files of zero size. I've attached a log from running
unipost. Is it best to direct UPP queries to wrfhelp?
>
> Sam.
>
>
>
> -----Original Message-----
> From: Paul Oldenburg via RT [mailto:met_help at ucar.edu]
> Sent: 04 October 2011 15:13
> To: Hawkins Samuel Lennon (AE-DE)
> Cc: met_help at mailman.ucar.edu
> Subject: Re: [rt.rap.ucar.edu #50422] Validation of wind speeds from
> mast
>
> Sam,
>
> I just noticed your attached mast data. I'll try to set up an
experiment and get some matched pairs with some WRF data that we have.
>
> Thanks,
>
> Paul
>
>
> On 10/04/2011 08:11 AM, Paul Oldenburg wrote:
>> Sam,
>>
>> You should be able to use any data you want as observations, as
long
>> as you can get MET to read it. Once that is done, the trick will
be
>> to set the fcst_field and obs_field settings in the point_stat
config file so that MET will match the data. If you send me some of
your test data, I can take a look.
>>
>> Paul
>>
>>
>> On 10/04/2011 02:51 AM, sam.hawkins at vattenfall.com via RT wrote:
>>>
>>> <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=50422 >
>>>
>>> Paul,
>>>
>>> Thanks for your quick reply. I have the UPP compiled, so I will
give it a try. I don't actually have access to the mast data yet,
that will come in the next week or so. I'm just trying to set up all
of the tools I need first. However, I have some old mast data from
another site which I may use as a test. It's only up to 40m height.
Would it be possible to validate against that? Obviously I can change
the format.
>>>
>>> Sam.
>>>
>>>
>>>
>>> -----Original Message-----
>>> From: Paul Oldenburg via RT [mailto:met_help at ucar.edu]
>>> Sent: 03 October 2011 16:58
>>> To: Hawkins Samuel Lennon (AE-DE)
>>> Cc: met_help at mailman.ucar.edu
>>> Subject: Re: [rt.rap.ucar.edu #50422] Validation of wind speeds
from
>>> mast
>>>
>>> Sam,
>>>
>>> What you have suggested would work. However, there is another way
to perform the verification that should yield closer height matches
between the forecast and obs. If you use the relatively new Unified
Post Processor (see http://www.dtcenter.org/wrf-
nmm/users/overview/upp_overview.php if you don't have an instance),
there is a way to interpolate the forecast wind speeds at various
heights above ground. In the UPP config file,
[UPP_path]/parm/wrf_cntrl.parm, look for the lines containing the
following four settings, then change the first group of 00000 in each
list to 22220 as I have done here:
>>>
>>> (SPE HUM AT FD HEIGHT) SCAL=( 5.0)
>>> L=(22220 00000 00000 00000 00000 00000 00000 00000 00000 00000
>>> 00000 00000 00000 00000) (TEMP AT FD HEIGHTS ) SCAL=( 5.0)
>>> L=(22220 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000
>>> 00000 00000 00000) (U WIND AT FD HEIGHTS) SCAL=( 5.0) L=(22220
>>> 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
>>> 00000
>>> 00000) (V WIND AT FD HEIGHTS) SCAL=( 5.0) L=(22220 00000 00000
>>> 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000)
>>>
>>> I attached a version of this file with this change included.
Please move your original copy and replace it with the attached. When
you run UPP with this configuration, the output should include
interpolated WIND at 30m, 50m, 80m and 100m. Once you have performed
this interpolation on the forecast data, you can then specify the
verification fields in the point_stat config file like so:
>>>
>>> fcst_field[] = [ "WIND/Z80" ];
>>> obs_field[] = [ "WIND/Z70-90" ];
>>>
>>> This would compare the interpolated 80m forecast wind speed to any
observation whose height is between 70m and 90m.
>>>
>>> I'm curious about your verification task, and I was wondering if
it would be possible for you to send me your observation and forecast
data? If you are using ascii2nc to pre-process your observations, can
you please send me the ascii obs file, too? I will try to perform the
type of verification that I am suggesting you try. If the data is too
large to email, you can upload it to our FTP site by following these
instructions:
>>>
>>> http://www.dtcenter.org/met/users/support/met_help.php#ftp
>>>
>>> If you have any questions, please let me know.
>>>
>>> Thanks,
>>>
>>> Paul
>>>
>>>
>>> On 10/03/2011 09:17 AM, sam.hawkins at vattenfall.com via RT wrote:
>>>>
>>>> Mon Oct 03 09:17:06 2011: Request 50422 was acted upon.
>>>> Transaction: Ticket created by sam.hawkins at vattenfall.com
>>>> Queue: met_help
>>>> Subject: Validation of wind speeds from mast
>>>> Owner: Nobody
>>>> Requestors: sam.hawkins at vattenfall.com
>>>> Status: new
>>>> Ticket <URL:
>>>> https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=50422
>>>>>
>>>>
>>>>
>>>> Dear met users,
>>>>
>>>> I'm interested in using MET to validate WRF wind speed forecasts
against observations from masts at (multiple) heights, up to 80m
above the surface. Does anyone have experience of this? Looking at
the documentation, it seems I might need to interpolate WRF output to
the correct heights, and then pretend both the forecast and
observations are the 10m wind? Or is there an easier way?
>>>>
>>>> Thanks,
>>>>
>>>> Sam.
>>>>
>>>>
>>>>
>>>
>>>
>>
>
>
------------------------------------------------
Subject: RE: [rt.rap.ucar.edu #50422] Validation of wind speeds from mast
From: sam.hawkins at vattenfall.com
Time: Wed Oct 05 02:10:08 2011
Paul,
I've uploaded a wrfout file into 'hawkins-data'.Meanwhile I'll try UPP
with a different domain today and see if I have a similar problem.
Thanks for looking into this.
Sam.
-----Original Message-----
From: Paul Oldenburg via RT [mailto:met_help at ucar.edu]
Sent: 04 October 2011 16:40
To: Hawkins Samuel Lennon (AE-DE)
Cc: met_help at mailman.ucar.edu
Subject: Re: [rt.rap.ucar.edu #50422] Validation of wind speeds from
mast
Sam,
I'm working on the exact same thing myself. In general, it is best to
send UPP questions to WRF help, but in this case, I'll try to figure
out it. Of course, if I fail to figure it out, the next step will be
to engage WRF help (which happens to be down the hall from me).
Can you please send me one of your WRF output files? If it's too big
for email, please upload to our FTP site (URL below).
Thanks,
Paul
On 10/04/2011 09:37 AM, sam.hawkins at vattenfall.com via RT wrote:
>
> <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=50422 >
>
> Paul,
>
> Thanks for looking into it. I've never had much success with any of
the WRF post-processors. I've just tried running UPP but it produces
output files of zero size.
>
> I've tried turning eveything off in the wrf_cntrl.parm file, apart
from U wind at anemometer height and V wind at anemoter height, but I
still get output files of zero size. I've attached a log from running
unipost. Is it best to direct UPP queries to wrfhelp?
>
> Sam.
>
>
>
> -----Original Message-----
> From: Paul Oldenburg via RT [mailto:met_help at ucar.edu]
> Sent: 04 October 2011 15:13
> To: Hawkins Samuel Lennon (AE-DE)
> Cc: met_help at mailman.ucar.edu
> Subject: Re: [rt.rap.ucar.edu #50422] Validation of wind speeds from
> mast
>
> Sam,
>
> I just noticed your attached mast data. I'll try to set up an
experiment and get some matched pairs with some WRF data that we have.
>
> Thanks,
>
> Paul
>
>
> On 10/04/2011 08:11 AM, Paul Oldenburg wrote:
>> Sam,
>>
>> You should be able to use any data you want as observations, as
long
>> as you can get MET to read it. Once that is done, the trick will
be
>> to set the fcst_field and obs_field settings in the point_stat
config file so that MET will match the data. If you send me some of
your test data, I can take a look.
>>
>> Paul
>>
>>
>> On 10/04/2011 02:51 AM, sam.hawkins at vattenfall.com via RT wrote:
>>>
>>> <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=50422 >
>>>
>>> Paul,
>>>
>>> Thanks for your quick reply. I have the UPP compiled, so I will
give it a try. I don't actually have access to the mast data yet,
that will come in the next week or so. I'm just trying to set up all
of the tools I need first. However, I have some old mast data from
another site which I may use as a test. It's only up to 40m height.
Would it be possible to validate against that? Obviously I can change
the format.
>>>
>>> Sam.
>>>
>>>
>>>
>>> -----Original Message-----
>>> From: Paul Oldenburg via RT [mailto:met_help at ucar.edu]
>>> Sent: 03 October 2011 16:58
>>> To: Hawkins Samuel Lennon (AE-DE)
>>> Cc: met_help at mailman.ucar.edu
>>> Subject: Re: [rt.rap.ucar.edu #50422] Validation of wind speeds
from
>>> mast
>>>
>>> Sam,
>>>
>>> What you have suggested would work. However, there is another way
to perform the verification that should yield closer height matches
between the forecast and obs. If you use the relatively new Unified
Post Processor (see http://www.dtcenter.org/wrf-
nmm/users/overview/upp_overview.php if you don't have an instance),
there is a way to interpolate the forecast wind speeds at various
heights above ground. In the UPP config file,
[UPP_path]/parm/wrf_cntrl.parm, look for the lines containing the
following four settings, then change the first group of 00000 in each
list to 22220 as I have done here:
>>>
>>> (SPE HUM AT FD HEIGHT) SCAL=( 5.0)
>>> L=(22220 00000 00000 00000 00000 00000 00000 00000 00000 00000
>>> 00000 00000 00000 00000) (TEMP AT FD HEIGHTS ) SCAL=( 5.0)
>>> L=(22220 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000
>>> 00000 00000 00000) (U WIND AT FD HEIGHTS) SCAL=( 5.0) L=(22220
>>> 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
>>> 00000
>>> 00000) (V WIND AT FD HEIGHTS) SCAL=( 5.0) L=(22220 00000 00000
>>> 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000)
>>>
>>> I attached a version of this file with this change included.
Please move your original copy and replace it with the attached. When
you run UPP with this configuration, the output should include
interpolated WIND at 30m, 50m, 80m and 100m. Once you have performed
this interpolation on the forecast data, you can then specify the
verification fields in the point_stat config file like so:
>>>
>>> fcst_field[] = [ "WIND/Z80" ];
>>> obs_field[] = [ "WIND/Z70-90" ];
>>>
>>> This would compare the interpolated 80m forecast wind speed to any
observation whose height is between 70m and 90m.
>>>
>>> I'm curious about your verification task, and I was wondering if
it would be possible for you to send me your observation and forecast
data? If you are using ascii2nc to pre-process your observations, can
you please send me the ascii obs file, too? I will try to perform the
type of verification that I am suggesting you try. If the data is too
large to email, you can upload it to our FTP site by following these
instructions:
>>>
>>> http://www.dtcenter.org/met/users/support/met_help.php#ftp
>>>
>>> If you have any questions, please let me know.
>>>
>>> Thanks,
>>>
>>> Paul
>>>
>>>
>>> On 10/03/2011 09:17 AM, sam.hawkins at vattenfall.com via RT wrote:
>>>>
>>>> Mon Oct 03 09:17:06 2011: Request 50422 was acted upon.
>>>> Transaction: Ticket created by sam.hawkins at vattenfall.com
>>>> Queue: met_help
>>>> Subject: Validation of wind speeds from mast
>>>> Owner: Nobody
>>>> Requestors: sam.hawkins at vattenfall.com
>>>> Status: new
>>>> Ticket <URL:
>>>> https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=50422
>>>>>
>>>>
>>>>
>>>> Dear met users,
>>>>
>>>> I'm interested in using MET to validate WRF wind speed forecasts
against observations from masts at (multiple) heights, up to 80m
above the surface. Does anyone have experience of this? Looking at
the documentation, it seems I might need to interpolate WRF output to
the correct heights, and then pretend both the forecast and
observations are the 10m wind? Or is there an easier way?
>>>>
>>>> Thanks,
>>>>
>>>> Sam.
>>>>
>>>>
>>>>
>>>
>>>
>>
>
>
------------------------------------------------
Subject: Re: [rt.rap.ucar.edu #50422] Validation of wind speeds from mast
From: Paul Oldenburg
Time: Wed Oct 05 14:09:42 2011
Sam,
I ran your WRF NetCDF output through UPP and it appeared to work
correctly. However, MET does not support Data
Representation Type 205 (Arakawa Non-E Staggered rotated
Latitude/Longitude Grid - see
http://www.nco.ncep.noaa.gov/pmb/docs/on388/table6.html). The reason
is because it is staggered. I tried to use the
UPP copygb utility to regrid the data to a supported grid type (212 -
Lambert Conformal), but it was not successful. I
will take this issue up with UPP support to see if there is a bug in
copygb.
In the meanwhile, if it is easy to use a different grid when you run
WPS and WRF, MET has support for most standard,
non-staggered grids. If you have any questions about this, please let
me know.
Thanks,
Paul
On 10/05/2011 02:10 AM, sam.hawkins at vattenfall.com via RT wrote:
>
> <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=50422 >
>
> Paul,
>
> I've uploaded a wrfout file into 'hawkins-data'.Meanwhile I'll try
UPP with a different domain today and see if I have a similar problem.
Thanks for looking into this.
>
> Sam.
>
> -----Original Message-----
> From: Paul Oldenburg via RT [mailto:met_help at ucar.edu]
> Sent: 04 October 2011 16:40
> To: Hawkins Samuel Lennon (AE-DE)
> Cc: met_help at mailman.ucar.edu
> Subject: Re: [rt.rap.ucar.edu #50422] Validation of wind speeds from
mast
>
> Sam,
>
> I'm working on the exact same thing myself. In general, it is best
to send UPP questions to WRF help, but in this case, I'll try to
figure out it. Of course, if I fail to figure it out, the next step
will be to engage WRF help (which happens to be down the hall from
me).
>
> Can you please send me one of your WRF output files? If it's too
big for email, please upload to our FTP site (URL below).
>
> Thanks,
>
> Paul
>
>
> On 10/04/2011 09:37 AM, sam.hawkins at vattenfall.com via RT wrote:
>>
>> <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=50422 >
>>
>> Paul,
>>
>> Thanks for looking into it. I've never had much success with any of
the WRF post-processors. I've just tried running UPP but it produces
output files of zero size.
>>
>> I've tried turning eveything off in the wrf_cntrl.parm file, apart
from U wind at anemometer height and V wind at anemoter height, but I
still get output files of zero size. I've attached a log from running
unipost. Is it best to direct UPP queries to wrfhelp?
>>
>> Sam.
>>
>>
>>
>> -----Original Message-----
>> From: Paul Oldenburg via RT [mailto:met_help at ucar.edu]
>> Sent: 04 October 2011 15:13
>> To: Hawkins Samuel Lennon (AE-DE)
>> Cc: met_help at mailman.ucar.edu
>> Subject: Re: [rt.rap.ucar.edu #50422] Validation of wind speeds
from
>> mast
>>
>> Sam,
>>
>> I just noticed your attached mast data. I'll try to set up an
experiment and get some matched pairs with some WRF data that we have.
>>
>> Thanks,
>>
>> Paul
>>
>>
>> On 10/04/2011 08:11 AM, Paul Oldenburg wrote:
>>> Sam,
>>>
>>> You should be able to use any data you want as observations, as
long
>>> as you can get MET to read it. Once that is done, the trick will
be
>>> to set the fcst_field and obs_field settings in the point_stat
config file so that MET will match the data. If you send me some of
your test data, I can take a look.
>>>
>>> Paul
>>>
>>>
>>> On 10/04/2011 02:51 AM, sam.hawkins at vattenfall.com via RT wrote:
>>>>
>>>> <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=50422 >
>>>>
>>>> Paul,
>>>>
>>>> Thanks for your quick reply. I have the UPP compiled, so I will
give it a try. I don't actually have access to the mast data yet,
that will come in the next week or so. I'm just trying to set up all
of the tools I need first. However, I have some old mast data from
another site which I may use as a test. It's only up to 40m height.
Would it be possible to validate against that? Obviously I can change
the format.
>>>>
>>>> Sam.
>>>>
>>>>
>>>>
>>>> -----Original Message-----
>>>> From: Paul Oldenburg via RT [mailto:met_help at ucar.edu]
>>>> Sent: 03 October 2011 16:58
>>>> To: Hawkins Samuel Lennon (AE-DE)
>>>> Cc: met_help at mailman.ucar.edu
>>>> Subject: Re: [rt.rap.ucar.edu #50422] Validation of wind speeds
from
>>>> mast
>>>>
>>>> Sam,
>>>>
>>>> What you have suggested would work. However, there is another
way to perform the verification that should yield closer height
matches between the forecast and obs. If you use the relatively new
Unified Post Processor (see http://www.dtcenter.org/wrf-
nmm/users/overview/upp_overview.php if you don't have an instance),
there is a way to interpolate the forecast wind speeds at various
heights above ground. In the UPP config file,
[UPP_path]/parm/wrf_cntrl.parm, look for the lines containing the
following four settings, then change the first group of 00000 in each
list to 22220 as I have done here:
>>>>
>>>> (SPE HUM AT FD HEIGHT) SCAL=( 5.0)
>>>> L=(22220 00000 00000 00000 00000 00000 00000 00000 00000 00000
>>>> 00000 00000 00000 00000) (TEMP AT FD HEIGHTS ) SCAL=( 5.0)
>>>> L=(22220 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000
>>>> 00000 00000 00000) (U WIND AT FD HEIGHTS) SCAL=( 5.0) L=(22220
>>>> 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
>>>> 00000
>>>> 00000) (V WIND AT FD HEIGHTS) SCAL=( 5.0) L=(22220 00000 00000
>>>> 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000)
>>>>
>>>> I attached a version of this file with this change included.
Please move your original copy and replace it with the attached. When
you run UPP with this configuration, the output should include
interpolated WIND at 30m, 50m, 80m and 100m. Once you have performed
this interpolation on the forecast data, you can then specify the
verification fields in the point_stat config file like so:
>>>>
>>>> fcst_field[] = [ "WIND/Z80" ];
>>>> obs_field[] = [ "WIND/Z70-90" ];
>>>>
>>>> This would compare the interpolated 80m forecast wind speed to
any observation whose height is between 70m and 90m.
>>>>
>>>> I'm curious about your verification task, and I was wondering if
it would be possible for you to send me your observation and forecast
data? If you are using ascii2nc to pre-process your observations, can
you please send me the ascii obs file, too? I will try to perform the
type of verification that I am suggesting you try. If the data is too
large to email, you can upload it to our FTP site by following these
instructions:
>>>>
>>>> http://www.dtcenter.org/met/users/support/met_help.php#ftp
>>>>
>>>> If you have any questions, please let me know.
>>>>
>>>> Thanks,
>>>>
>>>> Paul
>>>>
>>>>
>>>> On 10/03/2011 09:17 AM, sam.hawkins at vattenfall.com via RT wrote:
>>>>>
>>>>> Mon Oct 03 09:17:06 2011: Request 50422 was acted upon.
>>>>> Transaction: Ticket created by sam.hawkins at vattenfall.com
>>>>> Queue: met_help
>>>>> Subject: Validation of wind speeds from mast
>>>>> Owner: Nobody
>>>>> Requestors: sam.hawkins at vattenfall.com
>>>>> Status: new
>>>>> Ticket <URL:
>>>>> https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=50422
>>>>>>
>>>>>
>>>>>
>>>>> Dear met users,
>>>>>
>>>>> I'm interested in using MET to validate WRF wind speed forecasts
against observations from masts at (multiple) heights, up to 80m
above the surface. Does anyone have experience of this? Looking at
the documentation, it seems I might need to interpolate WRF output to
the correct heights, and then pretend both the forecast and
observations are the 10m wind? Or is there an easier way?
>>>>>
>>>>> Thanks,
>>>>>
>>>>> Sam.
>>>>>
>>>>>
>>>>>
>>>>
>>>>
>>>
>>
>>
>
>
------------------------------------------------
Subject: RE: [rt.rap.ucar.edu #50422] Validation of wind speeds from mast
From: sam.hawkins at vattenfall.com
Time: Thu Oct 06 04:10:49 2011
Paul,
I've tried with a Lambert projection, but I get the same issue - UPP
appears to run, but the output files are of zero size. I suspect it
is a problem with the way I have compiled UPP. In particular in the
output log the line:
The Posting is using 1 MPI task
There are 0 I/O servers
makes me suspicoius. I get this even if I try and run unipost.exe from
the command line, without using mpi_run or similar. Should I direct a
query to wrfhelp?
Thanks,
Sam.
-----Original Message-----
From: Paul Oldenburg via RT [mailto:met_help at ucar.edu]
Sent: 05 October 2011 21:10
To: Hawkins Samuel Lennon (AE-DE)
Cc: met_help at mailman.ucar.edu
Subject: Re: [rt.rap.ucar.edu #50422] Validation of wind speeds from
mast
Sam,
I ran your WRF NetCDF output through UPP and it appeared to work
correctly. However, MET does not support Data Representation Type 205
(Arakawa Non-E Staggered rotated Latitude/Longitude Grid - see
http://www.nco.ncep.noaa.gov/pmb/docs/on388/table6.html). The reason
is because it is staggered. I tried to use the UPP copygb utility to
regrid the data to a supported grid type (212 - Lambert Conformal),
but it was not successful. I will take this issue up with UPP support
to see if there is a bug in copygb.
In the meanwhile, if it is easy to use a different grid when you run
WPS and WRF, MET has support for most standard, non-staggered grids.
If you have any questions about this, please let me know.
Thanks,
Paul
On 10/05/2011 02:10 AM, sam.hawkins at vattenfall.com via RT wrote:
>
> <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=50422 >
>
> Paul,
>
> I've uploaded a wrfout file into 'hawkins-data'.Meanwhile I'll try
UPP with a different domain today and see if I have a similar problem.
Thanks for looking into this.
>
> Sam.
>
> -----Original Message-----
> From: Paul Oldenburg via RT [mailto:met_help at ucar.edu]
> Sent: 04 October 2011 16:40
> To: Hawkins Samuel Lennon (AE-DE)
> Cc: met_help at mailman.ucar.edu
> Subject: Re: [rt.rap.ucar.edu #50422] Validation of wind speeds from
> mast
>
> Sam,
>
> I'm working on the exact same thing myself. In general, it is best
to send UPP questions to WRF help, but in this case, I'll try to
figure out it. Of course, if I fail to figure it out, the next step
will be to engage WRF help (which happens to be down the hall from
me).
>
> Can you please send me one of your WRF output files? If it's too
big for email, please upload to our FTP site (URL below).
>
> Thanks,
>
> Paul
>
>
> On 10/04/2011 09:37 AM, sam.hawkins at vattenfall.com via RT wrote:
>>
>> <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=50422 >
>>
>> Paul,
>>
>> Thanks for looking into it. I've never had much success with any of
the WRF post-processors. I've just tried running UPP but it produces
output files of zero size.
>>
>> I've tried turning eveything off in the wrf_cntrl.parm file, apart
from U wind at anemometer height and V wind at anemoter height, but I
still get output files of zero size. I've attached a log from running
unipost. Is it best to direct UPP queries to wrfhelp?
>>
>> Sam.
>>
>>
>>
>> -----Original Message-----
>> From: Paul Oldenburg via RT [mailto:met_help at ucar.edu]
>> Sent: 04 October 2011 15:13
>> To: Hawkins Samuel Lennon (AE-DE)
>> Cc: met_help at mailman.ucar.edu
>> Subject: Re: [rt.rap.ucar.edu #50422] Validation of wind speeds
from
>> mast
>>
>> Sam,
>>
>> I just noticed your attached mast data. I'll try to set up an
experiment and get some matched pairs with some WRF data that we have.
>>
>> Thanks,
>>
>> Paul
>>
>>
>> On 10/04/2011 08:11 AM, Paul Oldenburg wrote:
>>> Sam,
>>>
>>> You should be able to use any data you want as observations, as
long
>>> as you can get MET to read it. Once that is done, the trick will
be
>>> to set the fcst_field and obs_field settings in the point_stat
config file so that MET will match the data. If you send me some of
your test data, I can take a look.
>>>
>>> Paul
>>>
>>>
>>> On 10/04/2011 02:51 AM, sam.hawkins at vattenfall.com via RT wrote:
>>>>
>>>> <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=50422 >
>>>>
>>>> Paul,
>>>>
>>>> Thanks for your quick reply. I have the UPP compiled, so I will
give it a try. I don't actually have access to the mast data yet,
that will come in the next week or so. I'm just trying to set up all
of the tools I need first. However, I have some old mast data from
another site which I may use as a test. It's only up to 40m height.
Would it be possible to validate against that? Obviously I can change
the format.
>>>>
>>>> Sam.
>>>>
>>>>
>>>>
>>>> -----Original Message-----
>>>> From: Paul Oldenburg via RT [mailto:met_help at ucar.edu]
>>>> Sent: 03 October 2011 16:58
>>>> To: Hawkins Samuel Lennon (AE-DE)
>>>> Cc: met_help at mailman.ucar.edu
>>>> Subject: Re: [rt.rap.ucar.edu #50422] Validation of wind speeds
>>>> from mast
>>>>
>>>> Sam,
>>>>
>>>> What you have suggested would work. However, there is another
way to perform the verification that should yield closer height
matches between the forecast and obs. If you use the relatively new
Unified Post Processor (see http://www.dtcenter.org/wrf-
nmm/users/overview/upp_overview.php if you don't have an instance),
there is a way to interpolate the forecast wind speeds at various
heights above ground. In the UPP config file,
[UPP_path]/parm/wrf_cntrl.parm, look for the lines containing the
following four settings, then change the first group of 00000 in each
list to 22220 as I have done here:
>>>>
>>>> (SPE HUM AT FD HEIGHT) SCAL=( 5.0) L=(22220 00000 00000 00000
>>>> 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000)
(TEMP
>>>> AT FD HEIGHTS ) SCAL=( 5.0) L=(22220 00000 00000 00000 00000
00000
>>>> 00000 00000 00000 00000 00000 00000 00000 00000) (U WIND AT FD
>>>> HEIGHTS) SCAL=( 5.0) L=(22220 00000 00000 00000 00000 00000
00000
>>>> 00000 00000 00000 00000 00000 00000
>>>> 00000) (V WIND AT FD HEIGHTS) SCAL=( 5.0) L=(22220 00000 00000
>>>> 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000)
>>>>
>>>> I attached a version of this file with this change included.
Please move your original copy and replace it with the attached. When
you run UPP with this configuration, the output should include
interpolated WIND at 30m, 50m, 80m and 100m. Once you have performed
this interpolation on the forecast data, you can then specify the
verification fields in the point_stat config file like so:
>>>>
>>>> fcst_field[] = [ "WIND/Z80" ];
>>>> obs_field[] = [ "WIND/Z70-90" ];
>>>>
>>>> This would compare the interpolated 80m forecast wind speed to
any observation whose height is between 70m and 90m.
>>>>
>>>> I'm curious about your verification task, and I was wondering if
it would be possible for you to send me your observation and forecast
data? If you are using ascii2nc to pre-process your observations, can
you please send me the ascii obs file, too? I will try to perform the
type of verification that I am suggesting you try. If the data is too
large to email, you can upload it to our FTP site by following these
instructions:
>>>>
>>>> http://www.dtcenter.org/met/users/support/met_help.php#ftp
>>>>
>>>> If you have any questions, please let me know.
>>>>
>>>> Thanks,
>>>>
>>>> Paul
>>>>
>>>>
>>>> On 10/03/2011 09:17 AM, sam.hawkins at vattenfall.com via RT wrote:
>>>>>
>>>>> Mon Oct 03 09:17:06 2011: Request 50422 was acted upon.
>>>>> Transaction: Ticket created by sam.hawkins at vattenfall.com
>>>>> Queue: met_help
>>>>> Subject: Validation of wind speeds from mast
>>>>> Owner: Nobody
>>>>> Requestors: sam.hawkins at vattenfall.com
>>>>> Status: new
>>>>> Ticket <URL:
>>>>> https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=50422
>>>>>>
>>>>>
>>>>>
>>>>> Dear met users,
>>>>>
>>>>> I'm interested in using MET to validate WRF wind speed forecasts
against observations from masts at (multiple) heights, up to 80m
above the surface. Does anyone have experience of this? Looking at
the documentation, it seems I might need to interpolate WRF output to
the correct heights, and then pretend both the forecast and
observations are the 10m wind? Or is there an easier way?
>>>>>
>>>>> Thanks,
>>>>>
>>>>> Sam.
>>>>>
>>>>>
>>>>>
>>>>
>>>>
>>>
>>
>>
>
>
------------------------------------------------
Subject: RE: [rt.rap.ucar.edu #50422] Validation of wind speeds from mast
From: sam.hawkins at vattenfall.com
Time: Thu Oct 06 08:05:33 2011
Paul,
I managed to get UPP working by recompiling for serial execution.
I've also defined a lambert domain over the UK and run a forecast for
that, and I've got hold of the real offshore mast data I will be using
to validate forecasts.
I'll see if I can validations working with MET. I'll also put a
forecast and some sample data on the ftp site later, in case you have
a chance to try validation as well.
Thanks,
Sam.
-----Original Message-----
From: Paul Oldenburg via RT [mailto:met_help at ucar.edu]
Sent: 05 October 2011 21:10
To: Hawkins Samuel Lennon (AE-DE)
Cc: met_help at mailman.ucar.edu
Subject: Re: [rt.rap.ucar.edu #50422] Validation of wind speeds from
mast
Sam,
I ran your WRF NetCDF output through UPP and it appeared to work
correctly. However, MET does not support Data Representation Type 205
(Arakawa Non-E Staggered rotated Latitude/Longitude Grid - see
http://www.nco.ncep.noaa.gov/pmb/docs/on388/table6.html). The reason
is because it is staggered. I tried to use the UPP copygb utility to
regrid the data to a supported grid type (212 - Lambert Conformal),
but it was not successful. I will take this issue up with UPP support
to see if there is a bug in copygb.
In the meanwhile, if it is easy to use a different grid when you run
WPS and WRF, MET has support for most standard, non-staggered grids.
If you have any questions about this, please let me know.
Thanks,
Paul
On 10/05/2011 02:10 AM, sam.hawkins at vattenfall.com via RT wrote:
>
> <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=50422 >
>
> Paul,
>
> I've uploaded a wrfout file into 'hawkins-data'.Meanwhile I'll try
UPP with a different domain today and see if I have a similar problem.
Thanks for looking into this.
>
> Sam.
>
> -----Original Message-----
> From: Paul Oldenburg via RT [mailto:met_help at ucar.edu]
> Sent: 04 October 2011 16:40
> To: Hawkins Samuel Lennon (AE-DE)
> Cc: met_help at mailman.ucar.edu
> Subject: Re: [rt.rap.ucar.edu #50422] Validation of wind speeds from
> mast
>
> Sam,
>
> I'm working on the exact same thing myself. In general, it is best
to send UPP questions to WRF help, but in this case, I'll try to
figure out it. Of course, if I fail to figure it out, the next step
will be to engage WRF help (which happens to be down the hall from
me).
>
> Can you please send me one of your WRF output files? If it's too
big for email, please upload to our FTP site (URL below).
>
> Thanks,
>
> Paul
>
>
> On 10/04/2011 09:37 AM, sam.hawkins at vattenfall.com via RT wrote:
>>
>> <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=50422 >
>>
>> Paul,
>>
>> Thanks for looking into it. I've never had much success with any of
the WRF post-processors. I've just tried running UPP but it produces
output files of zero size.
>>
>> I've tried turning eveything off in the wrf_cntrl.parm file, apart
from U wind at anemometer height and V wind at anemoter height, but I
still get output files of zero size. I've attached a log from running
unipost. Is it best to direct UPP queries to wrfhelp?
>>
>> Sam.
>>
>>
>>
>> -----Original Message-----
>> From: Paul Oldenburg via RT [mailto:met_help at ucar.edu]
>> Sent: 04 October 2011 15:13
>> To: Hawkins Samuel Lennon (AE-DE)
>> Cc: met_help at mailman.ucar.edu
>> Subject: Re: [rt.rap.ucar.edu #50422] Validation of wind speeds
from
>> mast
>>
>> Sam,
>>
>> I just noticed your attached mast data. I'll try to set up an
experiment and get some matched pairs with some WRF data that we have.
>>
>> Thanks,
>>
>> Paul
>>
>>
>> On 10/04/2011 08:11 AM, Paul Oldenburg wrote:
>>> Sam,
>>>
>>> You should be able to use any data you want as observations, as
long
>>> as you can get MET to read it. Once that is done, the trick will
be
>>> to set the fcst_field and obs_field settings in the point_stat
config file so that MET will match the data. If you send me some of
your test data, I can take a look.
>>>
>>> Paul
>>>
>>>
>>> On 10/04/2011 02:51 AM, sam.hawkins at vattenfall.com via RT wrote:
>>>>
>>>> <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=50422 >
>>>>
>>>> Paul,
>>>>
>>>> Thanks for your quick reply. I have the UPP compiled, so I will
give it a try. I don't actually have access to the mast data yet,
that will come in the next week or so. I'm just trying to set up all
of the tools I need first. However, I have some old mast data from
another site which I may use as a test. It's only up to 40m height.
Would it be possible to validate against that? Obviously I can change
the format.
>>>>
>>>> Sam.
>>>>
>>>>
>>>>
>>>> -----Original Message-----
>>>> From: Paul Oldenburg via RT [mailto:met_help at ucar.edu]
>>>> Sent: 03 October 2011 16:58
>>>> To: Hawkins Samuel Lennon (AE-DE)
>>>> Cc: met_help at mailman.ucar.edu
>>>> Subject: Re: [rt.rap.ucar.edu #50422] Validation of wind speeds
>>>> from mast
>>>>
>>>> Sam,
>>>>
>>>> What you have suggested would work. However, there is another
way to perform the verification that should yield closer height
matches between the forecast and obs. If you use the relatively new
Unified Post Processor (see http://www.dtcenter.org/wrf-
nmm/users/overview/upp_overview.php if you don't have an instance),
there is a way to interpolate the forecast wind speeds at various
heights above ground. In the UPP config file,
[UPP_path]/parm/wrf_cntrl.parm, look for the lines containing the
following four settings, then change the first group of 00000 in each
list to 22220 as I have done here:
>>>>
>>>> (SPE HUM AT FD HEIGHT) SCAL=( 5.0) L=(22220 00000 00000 00000
>>>> 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000)
(TEMP
>>>> AT FD HEIGHTS ) SCAL=( 5.0) L=(22220 00000 00000 00000 00000
00000
>>>> 00000 00000 00000 00000 00000 00000 00000 00000) (U WIND AT FD
>>>> HEIGHTS) SCAL=( 5.0) L=(22220 00000 00000 00000 00000 00000
00000
>>>> 00000 00000 00000 00000 00000 00000
>>>> 00000) (V WIND AT FD HEIGHTS) SCAL=( 5.0) L=(22220 00000 00000
>>>> 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000)
>>>>
>>>> I attached a version of this file with this change included.
Please move your original copy and replace it with the attached. When
you run UPP with this configuration, the output should include
interpolated WIND at 30m, 50m, 80m and 100m. Once you have performed
this interpolation on the forecast data, you can then specify the
verification fields in the point_stat config file like so:
>>>>
>>>> fcst_field[] = [ "WIND/Z80" ];
>>>> obs_field[] = [ "WIND/Z70-90" ];
>>>>
>>>> This would compare the interpolated 80m forecast wind speed to
any observation whose height is between 70m and 90m.
>>>>
>>>> I'm curious about your verification task, and I was wondering if
it would be possible for you to send me your observation and forecast
data? If you are using ascii2nc to pre-process your observations, can
you please send me the ascii obs file, too? I will try to perform the
type of verification that I am suggesting you try. If the data is too
large to email, you can upload it to our FTP site by following these
instructions:
>>>>
>>>> http://www.dtcenter.org/met/users/support/met_help.php#ftp
>>>>
>>>> If you have any questions, please let me know.
>>>>
>>>> Thanks,
>>>>
>>>> Paul
>>>>
>>>>
>>>> On 10/03/2011 09:17 AM, sam.hawkins at vattenfall.com via RT wrote:
>>>>>
>>>>> Mon Oct 03 09:17:06 2011: Request 50422 was acted upon.
>>>>> Transaction: Ticket created by sam.hawkins at vattenfall.com
>>>>> Queue: met_help
>>>>> Subject: Validation of wind speeds from mast
>>>>> Owner: Nobody
>>>>> Requestors: sam.hawkins at vattenfall.com
>>>>> Status: new
>>>>> Ticket <URL:
>>>>> https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=50422
>>>>>>
>>>>>
>>>>>
>>>>> Dear met users,
>>>>>
>>>>> I'm interested in using MET to validate WRF wind speed forecasts
against observations from masts at (multiple) heights, up to 80m
above the surface. Does anyone have experience of this? Looking at
the documentation, it seems I might need to interpolate WRF output to
the correct heights, and then pretend both the forecast and
observations are the 10m wind? Or is there an easier way?
>>>>>
>>>>> Thanks,
>>>>>
>>>>> Sam.
>>>>>
>>>>>
>>>>>
>>>>
>>>>
>>>
>>
>>
>
>
------------------------------------------------
Subject: Re: [rt.rap.ucar.edu #50422] Validation of wind speeds from mast
From: Paul Oldenburg
Time: Thu Oct 06 08:59:37 2011
Sam,
That is good news. I already submitted the grid 205 issue to
wrf_help, but I think your execution difficulty prior to
recompiling for serial execution may be a different issue. If you are
inclined, you may consider submitting a separate
wrf_help request on that topic.
I put a bundle called hawkins_ps_test.tar.gz on our FTP site that has
a few files in it that I used for testing. If you
pull it down you will find:
format_mast.pl: I wrote this to format your CSV into the ASCII format
that is read by the MET tool ascii2nc. It is
important that you change the latitude and longitude hard coded in the
script to the location of your obs if you use it.
I use it like this:
$ cat mast_sample.csv | perl format_mast.pl > mast_sample.txt
$ $MET_BASE/bin/ascii2nc mast_sample.txt mast_sample.nc -v 3
Creating NetCDF Observation file: mast_sample.nc
Reading ASCII Observation file: mast_sample.txt
Processing 32 observations.
Finished processing 32 observations.
This generates mast_sample.nc, which can be read as observations
directly by MET.
PointStatConfig: This is a point_stat config file that I set up to
verify the sample data. Note especially the two
lines fcst_field and obs_field:
fcst_field[] = [ "WIND/Z10-30", "WIND/Z30", "WIND/Z30-50" ];
obs_field[] = [ "WIND/Z20", "WIND/Z30", "WIND/Z40" ];
These settings will interpolate your model data, assuming UPP has
outputted wind at the levels that we discussed
earlier, and verify them against the mast wind obs at 20, 30 and 40m.
Unfortunately, point_stat does not verify wind
direction at this time.
rc_2008112200_F000.grb: Some bogus GRIB data that I pulled from
another project for testing. Note that if you test
using this data, you will have to change the ascii2nc input file,
mast_sample.txt, to set the date to 20081122_000000.
Otherwise, you could change the beg_ds and end_ds settings in
PointStatConfig to be very large.
Please let me know if you have any questions.
Thanks,
Paul
On 10/06/2011 08:05 AM, sam.hawkins at vattenfall.com via RT wrote:
>
> <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=50422 >
>
> Paul,
>
> I managed to get UPP working by recompiling for serial execution.
I've also defined a lambert domain over the UK and run a forecast for
that, and I've got hold of the real offshore mast data I will be using
to validate forecasts.
>
> I'll see if I can validations working with MET. I'll also put a
forecast and some sample data on the ftp site later, in case you have
a chance to try validation as well.
>
> Thanks,
>
> Sam.
>
> -----Original Message-----
> From: Paul Oldenburg via RT [mailto:met_help at ucar.edu]
> Sent: 05 October 2011 21:10
> To: Hawkins Samuel Lennon (AE-DE)
> Cc: met_help at mailman.ucar.edu
> Subject: Re: [rt.rap.ucar.edu #50422] Validation of wind speeds from
mast
>
> Sam,
>
> I ran your WRF NetCDF output through UPP and it appeared to work
correctly. However, MET does not support Data Representation Type 205
(Arakawa Non-E Staggered rotated Latitude/Longitude Grid - see
http://www.nco.ncep.noaa.gov/pmb/docs/on388/table6.html). The reason
is because it is staggered. I tried to use the UPP copygb utility to
regrid the data to a supported grid type (212 - Lambert Conformal),
but it was not successful. I will take this issue up with UPP support
to see if there is a bug in copygb.
>
> In the meanwhile, if it is easy to use a different grid when you run
WPS and WRF, MET has support for most standard, non-staggered grids.
If you have any questions about this, please let me know.
>
> Thanks,
>
> Paul
>
>
> On 10/05/2011 02:10 AM, sam.hawkins at vattenfall.com via RT wrote:
>>
>> <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=50422 >
>>
>> Paul,
>>
>> I've uploaded a wrfout file into 'hawkins-data'.Meanwhile I'll try
UPP with a different domain today and see if I have a similar problem.
Thanks for looking into this.
>>
>> Sam.
>>
>> -----Original Message-----
>> From: Paul Oldenburg via RT [mailto:met_help at ucar.edu]
>> Sent: 04 October 2011 16:40
>> To: Hawkins Samuel Lennon (AE-DE)
>> Cc: met_help at mailman.ucar.edu
>> Subject: Re: [rt.rap.ucar.edu #50422] Validation of wind speeds
from
>> mast
>>
>> Sam,
>>
>> I'm working on the exact same thing myself. In general, it is best
to send UPP questions to WRF help, but in this case, I'll try to
figure out it. Of course, if I fail to figure it out, the next step
will be to engage WRF help (which happens to be down the hall from
me).
>>
>> Can you please send me one of your WRF output files? If it's too
big for email, please upload to our FTP site (URL below).
>>
>> Thanks,
>>
>> Paul
>>
>>
>> On 10/04/2011 09:37 AM, sam.hawkins at vattenfall.com via RT wrote:
>>>
>>> <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=50422 >
>>>
>>> Paul,
>>>
>>> Thanks for looking into it. I've never had much success with any
of the WRF post-processors. I've just tried running UPP but it
produces output files of zero size.
>>>
>>> I've tried turning eveything off in the wrf_cntrl.parm file, apart
from U wind at anemometer height and V wind at anemoter height, but I
still get output files of zero size. I've attached a log from running
unipost. Is it best to direct UPP queries to wrfhelp?
>>>
>>> Sam.
>>>
>>>
>>>
>>> -----Original Message-----
>>> From: Paul Oldenburg via RT [mailto:met_help at ucar.edu]
>>> Sent: 04 October 2011 15:13
>>> To: Hawkins Samuel Lennon (AE-DE)
>>> Cc: met_help at mailman.ucar.edu
>>> Subject: Re: [rt.rap.ucar.edu #50422] Validation of wind speeds
from
>>> mast
>>>
>>> Sam,
>>>
>>> I just noticed your attached mast data. I'll try to set up an
experiment and get some matched pairs with some WRF data that we have.
>>>
>>> Thanks,
>>>
>>> Paul
>>>
>>>
>>> On 10/04/2011 08:11 AM, Paul Oldenburg wrote:
>>>> Sam,
>>>>
>>>> You should be able to use any data you want as observations, as
long
>>>> as you can get MET to read it. Once that is done, the trick will
be
>>>> to set the fcst_field and obs_field settings in the point_stat
config file so that MET will match the data. If you send me some of
your test data, I can take a look.
>>>>
>>>> Paul
>>>>
>>>>
>>>> On 10/04/2011 02:51 AM, sam.hawkins at vattenfall.com via RT wrote:
>>>>>
>>>>> <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=50422 >
>>>>>
>>>>> Paul,
>>>>>
>>>>> Thanks for your quick reply. I have the UPP compiled, so I will
give it a try. I don't actually have access to the mast data yet,
that will come in the next week or so. I'm just trying to set up all
of the tools I need first. However, I have some old mast data from
another site which I may use as a test. It's only up to 40m height.
Would it be possible to validate against that? Obviously I can change
the format.
>>>>>
>>>>> Sam.
>>>>>
>>>>>
>>>>>
>>>>> -----Original Message-----
>>>>> From: Paul Oldenburg via RT [mailto:met_help at ucar.edu]
>>>>> Sent: 03 October 2011 16:58
>>>>> To: Hawkins Samuel Lennon (AE-DE)
>>>>> Cc: met_help at mailman.ucar.edu
>>>>> Subject: Re: [rt.rap.ucar.edu #50422] Validation of wind speeds
>>>>> from mast
>>>>>
>>>>> Sam,
>>>>>
>>>>> What you have suggested would work. However, there is another
way to perform the verification that should yield closer height
matches between the forecast and obs. If you use the relatively new
Unified Post Processor (see http://www.dtcenter.org/wrf-
nmm/users/overview/upp_overview.php if you don't have an instance),
there is a way to interpolate the forecast wind speeds at various
heights above ground. In the UPP config file,
[UPP_path]/parm/wrf_cntrl.parm, look for the lines containing the
following four settings, then change the first group of 00000 in each
list to 22220 as I have done here:
>>>>>
>>>>> (SPE HUM AT FD HEIGHT) SCAL=( 5.0) L=(22220 00000 00000 00000
>>>>> 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000)
(TEMP
>>>>> AT FD HEIGHTS ) SCAL=( 5.0) L=(22220 00000 00000 00000 00000
00000
>>>>> 00000 00000 00000 00000 00000 00000 00000 00000) (U WIND AT FD
>>>>> HEIGHTS) SCAL=( 5.0) L=(22220 00000 00000 00000 00000 00000
00000
>>>>> 00000 00000 00000 00000 00000 00000
>>>>> 00000) (V WIND AT FD HEIGHTS) SCAL=( 5.0) L=(22220 00000 00000
>>>>> 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000)
>>>>>
>>>>> I attached a version of this file with this change included.
Please move your original copy and replace it with the attached. When
you run UPP with this configuration, the output should include
interpolated WIND at 30m, 50m, 80m and 100m. Once you have performed
this interpolation on the forecast data, you can then specify the
verification fields in the point_stat config file like so:
>>>>>
>>>>> fcst_field[] = [ "WIND/Z80" ];
>>>>> obs_field[] = [ "WIND/Z70-90" ];
>>>>>
>>>>> This would compare the interpolated 80m forecast wind speed to
any observation whose height is between 70m and 90m.
>>>>>
>>>>> I'm curious about your verification task, and I was wondering if
it would be possible for you to send me your observation and forecast
data? If you are using ascii2nc to pre-process your observations, can
you please send me the ascii obs file, too? I will try to perform the
type of verification that I am suggesting you try. If the data is too
large to email, you can upload it to our FTP site by following these
instructions:
>>>>>
>>>>> http://www.dtcenter.org/met/users/support/met_help.php#ftp
>>>>>
>>>>> If you have any questions, please let me know.
>>>>>
>>>>> Thanks,
>>>>>
>>>>> Paul
>>>>>
>>>>>
>>>>> On 10/03/2011 09:17 AM, sam.hawkins at vattenfall.com via RT wrote:
>>>>>>
>>>>>> Mon Oct 03 09:17:06 2011: Request 50422 was acted upon.
>>>>>> Transaction: Ticket created by sam.hawkins at vattenfall.com
>>>>>> Queue: met_help
>>>>>> Subject: Validation of wind speeds from mast
>>>>>> Owner: Nobody
>>>>>> Requestors: sam.hawkins at vattenfall.com
>>>>>> Status: new
>>>>>> Ticket <URL:
>>>>>> https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=50422
>>>>>>>
>>>>>>
>>>>>>
>>>>>> Dear met users,
>>>>>>
>>>>>> I'm interested in using MET to validate WRF wind speed
forecasts against observations from masts at (multiple) heights, up
to 80m above the surface. Does anyone have experience of this?
Looking at the documentation, it seems I might need to interpolate
WRF output to the correct heights, and then pretend both the forecast
and observations are the 10m wind? Or is there an easier way?
>>>>>>
>>>>>> Thanks,
>>>>>>
>>>>>> Sam.
>>>>>>
>>>>>>
>>>>>>
>>>>>
>>>>>
>>>>
>>>
>>>
>>
>>
>
>
------------------------------------------------
Subject: RE: [rt.rap.ucar.edu #50422] Validation of wind speeds from mast
From: sam.hawkins at vattenfall.com
Time: Thu Oct 06 09:15:58 2011
Paul,
Thanks so much for your help, I'll give that a go. Just a quick
question, I seem to remember the old flight heights in ARWPost were
above sea level, not terrain level. Is this still the case? Most of
my mast data is offshore, so it doesn't really matter, but I may use
some onshore masts as well.
Sam.
-----Original Message-----
From: Paul Oldenburg via RT [mailto:met_help at ucar.edu]
Sent: 06 October 2011 16:00
To: Hawkins Samuel Lennon (AE-DE)
Cc: met_help at mailman.ucar.edu
Subject: Re: [rt.rap.ucar.edu #50422] Validation of wind speeds from
mast
Sam,
That is good news. I already submitted the grid 205 issue to
wrf_help, but I think your execution difficulty prior to recompiling
for serial execution may be a different issue. If you are inclined,
you may consider submitting a separate wrf_help request on that topic.
I put a bundle called hawkins_ps_test.tar.gz on our FTP site that has
a few files in it that I used for testing. If you pull it down you
will find:
format_mast.pl: I wrote this to format your CSV into the ASCII format
that is read by the MET tool ascii2nc. It is important that you
change the latitude and longitude hard coded in the script to the
location of your obs if you use it.
I use it like this:
$ cat mast_sample.csv | perl format_mast.pl > mast_sample.txt $
$MET_BASE/bin/ascii2nc mast_sample.txt mast_sample.nc -v 3 Creating
NetCDF Observation file: mast_sample.nc Reading ASCII Observation
file: mast_sample.txt Processing 32 observations.
Finished processing 32 observations.
This generates mast_sample.nc, which can be read as observations
directly by MET.
PointStatConfig: This is a point_stat config file that I set up to
verify the sample data. Note especially the two lines fcst_field and
obs_field:
fcst_field[] = [ "WIND/Z10-30", "WIND/Z30", "WIND/Z30-50" ];
obs_field[] = [ "WIND/Z20", "WIND/Z30", "WIND/Z40" ];
These settings will interpolate your model data, assuming UPP has
outputted wind at the levels that we discussed earlier, and verify
them against the mast wind obs at 20, 30 and 40m. Unfortunately,
point_stat does not verify wind direction at this time.
rc_2008112200_F000.grb: Some bogus GRIB data that I pulled from
another project for testing. Note that if you test using this data,
you will have to change the ascii2nc input file, mast_sample.txt, to
set the date to 20081122_000000.
Otherwise, you could change the beg_ds and end_ds settings in
PointStatConfig to be very large.
Please let me know if you have any questions.
Thanks,
Paul
On 10/06/2011 08:05 AM, sam.hawkins at vattenfall.com via RT wrote:
>
> <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=50422 >
>
> Paul,
>
> I managed to get UPP working by recompiling for serial execution.
I've also defined a lambert domain over the UK and run a forecast for
that, and I've got hold of the real offshore mast data I will be using
to validate forecasts.
>
> I'll see if I can validations working with MET. I'll also put a
forecast and some sample data on the ftp site later, in case you have
a chance to try validation as well.
>
> Thanks,
>
> Sam.
>
> -----Original Message-----
> From: Paul Oldenburg via RT [mailto:met_help at ucar.edu]
> Sent: 05 October 2011 21:10
> To: Hawkins Samuel Lennon (AE-DE)
> Cc: met_help at mailman.ucar.edu
> Subject: Re: [rt.rap.ucar.edu #50422] Validation of wind speeds from
> mast
>
> Sam,
>
> I ran your WRF NetCDF output through UPP and it appeared to work
correctly. However, MET does not support Data Representation Type 205
(Arakawa Non-E Staggered rotated Latitude/Longitude Grid - see
http://www.nco.ncep.noaa.gov/pmb/docs/on388/table6.html). The reason
is because it is staggered. I tried to use the UPP copygb utility to
regrid the data to a supported grid type (212 - Lambert Conformal),
but it was not successful. I will take this issue up with UPP support
to see if there is a bug in copygb.
>
> In the meanwhile, if it is easy to use a different grid when you run
WPS and WRF, MET has support for most standard, non-staggered grids.
If you have any questions about this, please let me know.
>
> Thanks,
>
> Paul
>
>
> On 10/05/2011 02:10 AM, sam.hawkins at vattenfall.com via RT wrote:
>>
>> <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=50422 >
>>
>> Paul,
>>
>> I've uploaded a wrfout file into 'hawkins-data'.Meanwhile I'll try
UPP with a different domain today and see if I have a similar problem.
Thanks for looking into this.
>>
>> Sam.
>>
>> -----Original Message-----
>> From: Paul Oldenburg via RT [mailto:met_help at ucar.edu]
>> Sent: 04 October 2011 16:40
>> To: Hawkins Samuel Lennon (AE-DE)
>> Cc: met_help at mailman.ucar.edu
>> Subject: Re: [rt.rap.ucar.edu #50422] Validation of wind speeds
from
>> mast
>>
>> Sam,
>>
>> I'm working on the exact same thing myself. In general, it is best
to send UPP questions to WRF help, but in this case, I'll try to
figure out it. Of course, if I fail to figure it out, the next step
will be to engage WRF help (which happens to be down the hall from
me).
>>
>> Can you please send me one of your WRF output files? If it's too
big for email, please upload to our FTP site (URL below).
>>
>> Thanks,
>>
>> Paul
>>
>>
>> On 10/04/2011 09:37 AM, sam.hawkins at vattenfall.com via RT wrote:
>>>
>>> <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=50422 >
>>>
>>> Paul,
>>>
>>> Thanks for looking into it. I've never had much success with any
of the WRF post-processors. I've just tried running UPP but it
produces output files of zero size.
>>>
>>> I've tried turning eveything off in the wrf_cntrl.parm file, apart
from U wind at anemometer height and V wind at anemoter height, but I
still get output files of zero size. I've attached a log from running
unipost. Is it best to direct UPP queries to wrfhelp?
>>>
>>> Sam.
>>>
>>>
>>>
>>> -----Original Message-----
>>> From: Paul Oldenburg via RT [mailto:met_help at ucar.edu]
>>> Sent: 04 October 2011 15:13
>>> To: Hawkins Samuel Lennon (AE-DE)
>>> Cc: met_help at mailman.ucar.edu
>>> Subject: Re: [rt.rap.ucar.edu #50422] Validation of wind speeds
from
>>> mast
>>>
>>> Sam,
>>>
>>> I just noticed your attached mast data. I'll try to set up an
experiment and get some matched pairs with some WRF data that we have.
>>>
>>> Thanks,
>>>
>>> Paul
>>>
>>>
>>> On 10/04/2011 08:11 AM, Paul Oldenburg wrote:
>>>> Sam,
>>>>
>>>> You should be able to use any data you want as observations, as
>>>> long as you can get MET to read it. Once that is done, the trick
>>>> will be to set the fcst_field and obs_field settings in the
point_stat config file so that MET will match the data. If you send
me some of your test data, I can take a look.
>>>>
>>>> Paul
>>>>
>>>>
>>>> On 10/04/2011 02:51 AM, sam.hawkins at vattenfall.com via RT wrote:
>>>>>
>>>>> <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=50422 >
>>>>>
>>>>> Paul,
>>>>>
>>>>> Thanks for your quick reply. I have the UPP compiled, so I will
give it a try. I don't actually have access to the mast data yet,
that will come in the next week or so. I'm just trying to set up all
of the tools I need first. However, I have some old mast data from
another site which I may use as a test. It's only up to 40m height.
Would it be possible to validate against that? Obviously I can change
the format.
>>>>>
>>>>> Sam.
>>>>>
>>>>>
>>>>>
>>>>> -----Original Message-----
>>>>> From: Paul Oldenburg via RT [mailto:met_help at ucar.edu]
>>>>> Sent: 03 October 2011 16:58
>>>>> To: Hawkins Samuel Lennon (AE-DE)
>>>>> Cc: met_help at mailman.ucar.edu
>>>>> Subject: Re: [rt.rap.ucar.edu #50422] Validation of wind speeds
>>>>> from mast
>>>>>
>>>>> Sam,
>>>>>
>>>>> What you have suggested would work. However, there is another
way to perform the verification that should yield closer height
matches between the forecast and obs. If you use the relatively new
Unified Post Processor (see http://www.dtcenter.org/wrf-
nmm/users/overview/upp_overview.php if you don't have an instance),
there is a way to interpolate the forecast wind speeds at various
heights above ground. In the UPP config file,
[UPP_path]/parm/wrf_cntrl.parm, look for the lines containing the
following four settings, then change the first group of 00000 in each
list to 22220 as I have done here:
>>>>>
>>>>> (SPE HUM AT FD HEIGHT) SCAL=( 5.0) L=(22220 00000 00000 00000
>>>>> 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000)
>>>>> (TEMP AT FD HEIGHTS ) SCAL=( 5.0) L=(22220 00000 00000 00000
>>>>> 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000) (U
>>>>> WIND AT FD
>>>>> HEIGHTS) SCAL=( 5.0) L=(22220 00000 00000 00000 00000 00000
00000
>>>>> 00000 00000 00000 00000 00000 00000
>>>>> 00000) (V WIND AT FD HEIGHTS) SCAL=( 5.0) L=(22220 00000 00000
>>>>> 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000)
>>>>>
>>>>> I attached a version of this file with this change included.
Please move your original copy and replace it with the attached. When
you run UPP with this configuration, the output should include
interpolated WIND at 30m, 50m, 80m and 100m. Once you have performed
this interpolation on the forecast data, you can then specify the
verification fields in the point_stat config file like so:
>>>>>
>>>>> fcst_field[] = [ "WIND/Z80" ];
>>>>> obs_field[] = [ "WIND/Z70-90" ];
>>>>>
>>>>> This would compare the interpolated 80m forecast wind speed to
any observation whose height is between 70m and 90m.
>>>>>
>>>>> I'm curious about your verification task, and I was wondering if
it would be possible for you to send me your observation and forecast
data? If you are using ascii2nc to pre-process your observations, can
you please send me the ascii obs file, too? I will try to perform the
type of verification that I am suggesting you try. If the data is too
large to email, you can upload it to our FTP site by following these
instructions:
>>>>>
>>>>> http://www.dtcenter.org/met/users/support/met_help.php#ftp
>>>>>
>>>>> If you have any questions, please let me know.
>>>>>
>>>>> Thanks,
>>>>>
>>>>> Paul
>>>>>
>>>>>
>>>>> On 10/03/2011 09:17 AM, sam.hawkins at vattenfall.com via RT wrote:
>>>>>>
>>>>>> Mon Oct 03 09:17:06 2011: Request 50422 was acted upon.
>>>>>> Transaction: Ticket created by sam.hawkins at vattenfall.com
>>>>>> Queue: met_help
>>>>>> Subject: Validation of wind speeds from mast
>>>>>> Owner: Nobody
>>>>>> Requestors: sam.hawkins at vattenfall.com
>>>>>> Status: new
>>>>>> Ticket <URL:
>>>>>> https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=50422
>>>>>>>
>>>>>>
>>>>>>
>>>>>> Dear met users,
>>>>>>
>>>>>> I'm interested in using MET to validate WRF wind speed
forecasts against observations from masts at (multiple) heights, up
to 80m above the surface. Does anyone have experience of this?
Looking at the documentation, it seems I might need to interpolate
WRF output to the correct heights, and then pretend both the forecast
and observations are the 10m wind? Or is there an easier way?
>>>>>>
>>>>>> Thanks,
>>>>>>
>>>>>> Sam.
>>>>>>
>>>>>>
>>>>>>
>>>>>
>>>>>
>>>>
>>>
>>>
>>
>>
>
>
------------------------------------------------
Subject: Re: [rt.rap.ucar.edu #50422] Validation of wind speeds from mast
From: Paul Oldenburg
Time: Thu Oct 06 09:19:19 2011
Sam,
In the UPP config file, wrf_cntrl.parm, remember that I asked you set
the first four zeroes to 2's:
(U WIND AT FD HEIGHTS) SCAL=( 5.0)
L=(22220 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000 00000 00000)
This setting is not well documented (or maybe not at all), but if you
specify a 1, you get the height in msl. If you
specify a 2, you get height above terrain.
Paul
On 10/06/2011 09:15 AM, sam.hawkins at vattenfall.com via RT wrote:
>
> <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=50422 >
>
> Paul,
>
> Thanks so much for your help, I'll give that a go. Just a quick
question, I seem to remember the old flight heights in ARWPost were
above sea level, not terrain level. Is this still the case? Most of
my mast data is offshore, so it doesn't really matter, but I may use
some onshore masts as well.
>
> Sam.
>
> -----Original Message-----
> From: Paul Oldenburg via RT [mailto:met_help at ucar.edu]
> Sent: 06 October 2011 16:00
> To: Hawkins Samuel Lennon (AE-DE)
> Cc: met_help at mailman.ucar.edu
> Subject: Re: [rt.rap.ucar.edu #50422] Validation of wind speeds from
mast
>
> Sam,
>
> That is good news. I already submitted the grid 205 issue to
wrf_help, but I think your execution difficulty prior to recompiling
for serial execution may be a different issue. If you are inclined,
you may consider submitting a separate wrf_help request on that topic.
>
> I put a bundle called hawkins_ps_test.tar.gz on our FTP site that
has a few files in it that I used for testing. If you pull it down
you will find:
>
> format_mast.pl: I wrote this to format your CSV into the ASCII
format that is read by the MET tool ascii2nc. It is important that
you change the latitude and longitude hard coded in the script to the
location of your obs if you use it.
> I use it like this:
>
> $ cat mast_sample.csv | perl format_mast.pl > mast_sample.txt $
$MET_BASE/bin/ascii2nc mast_sample.txt mast_sample.nc -v 3 Creating
NetCDF Observation file: mast_sample.nc Reading ASCII Observation
file: mast_sample.txt Processing 32 observations.
> Finished processing 32 observations.
>
> This generates mast_sample.nc, which can be read as observations
directly by MET.
>
> PointStatConfig: This is a point_stat config file that I set up to
verify the sample data. Note especially the two lines fcst_field and
obs_field:
>
> fcst_field[] = [ "WIND/Z10-30", "WIND/Z30", "WIND/Z30-50" ];
> obs_field[] = [ "WIND/Z20", "WIND/Z30", "WIND/Z40" ];
>
> These settings will interpolate your model data, assuming UPP has
outputted wind at the levels that we discussed earlier, and verify
them against the mast wind obs at 20, 30 and 40m. Unfortunately,
point_stat does not verify wind direction at this time.
>
> rc_2008112200_F000.grb: Some bogus GRIB data that I pulled from
another project for testing. Note that if you test using this data,
you will have to change the ascii2nc input file, mast_sample.txt, to
set the date to 20081122_000000.
> Otherwise, you could change the beg_ds and end_ds settings in
PointStatConfig to be very large.
>
> Please let me know if you have any questions.
>
> Thanks,
>
> Paul
>
>
>
> On 10/06/2011 08:05 AM, sam.hawkins at vattenfall.com via RT wrote:
>>
>> <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=50422 >
>>
>> Paul,
>>
>> I managed to get UPP working by recompiling for serial execution.
I've also defined a lambert domain over the UK and run a forecast for
that, and I've got hold of the real offshore mast data I will be using
to validate forecasts.
>>
>> I'll see if I can validations working with MET. I'll also put a
forecast and some sample data on the ftp site later, in case you have
a chance to try validation as well.
>>
>> Thanks,
>>
>> Sam.
>>
>> -----Original Message-----
>> From: Paul Oldenburg via RT [mailto:met_help at ucar.edu]
>> Sent: 05 October 2011 21:10
>> To: Hawkins Samuel Lennon (AE-DE)
>> Cc: met_help at mailman.ucar.edu
>> Subject: Re: [rt.rap.ucar.edu #50422] Validation of wind speeds
from
>> mast
>>
>> Sam,
>>
>> I ran your WRF NetCDF output through UPP and it appeared to work
correctly. However, MET does not support Data Representation Type 205
(Arakawa Non-E Staggered rotated Latitude/Longitude Grid - see
http://www.nco.ncep.noaa.gov/pmb/docs/on388/table6.html). The reason
is because it is staggered. I tried to use the UPP copygb utility to
regrid the data to a supported grid type (212 - Lambert Conformal),
but it was not successful. I will take this issue up with UPP support
to see if there is a bug in copygb.
>>
>> In the meanwhile, if it is easy to use a different grid when you
run WPS and WRF, MET has support for most standard, non-staggered
grids. If you have any questions about this, please let me know.
>>
>> Thanks,
>>
>> Paul
>>
>>
>> On 10/05/2011 02:10 AM, sam.hawkins at vattenfall.com via RT wrote:
>>>
>>> <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=50422 >
>>>
>>> Paul,
>>>
>>> I've uploaded a wrfout file into 'hawkins-data'.Meanwhile I'll try
UPP with a different domain today and see if I have a similar problem.
Thanks for looking into this.
>>>
>>> Sam.
>>>
>>> -----Original Message-----
>>> From: Paul Oldenburg via RT [mailto:met_help at ucar.edu]
>>> Sent: 04 October 2011 16:40
>>> To: Hawkins Samuel Lennon (AE-DE)
>>> Cc: met_help at mailman.ucar.edu
>>> Subject: Re: [rt.rap.ucar.edu #50422] Validation of wind speeds
from
>>> mast
>>>
>>> Sam,
>>>
>>> I'm working on the exact same thing myself. In general, it is
best to send UPP questions to WRF help, but in this case, I'll try to
figure out it. Of course, if I fail to figure it out, the next step
will be to engage WRF help (which happens to be down the hall from
me).
>>>
>>> Can you please send me one of your WRF output files? If it's too
big for email, please upload to our FTP site (URL below).
>>>
>>> Thanks,
>>>
>>> Paul
>>>
>>>
>>> On 10/04/2011 09:37 AM, sam.hawkins at vattenfall.com via RT wrote:
>>>>
>>>> <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=50422 >
>>>>
>>>> Paul,
>>>>
>>>> Thanks for looking into it. I've never had much success with any
of the WRF post-processors. I've just tried running UPP but it
produces output files of zero size.
>>>>
>>>> I've tried turning eveything off in the wrf_cntrl.parm file,
apart from U wind at anemometer height and V wind at anemoter height,
but I still get output files of zero size. I've attached a log from
running unipost. Is it best to direct UPP queries to wrfhelp?
>>>>
>>>> Sam.
>>>>
>>>>
>>>>
>>>> -----Original Message-----
>>>> From: Paul Oldenburg via RT [mailto:met_help at ucar.edu]
>>>> Sent: 04 October 2011 15:13
>>>> To: Hawkins Samuel Lennon (AE-DE)
>>>> Cc: met_help at mailman.ucar.edu
>>>> Subject: Re: [rt.rap.ucar.edu #50422] Validation of wind speeds
from
>>>> mast
>>>>
>>>> Sam,
>>>>
>>>> I just noticed your attached mast data. I'll try to set up an
experiment and get some matched pairs with some WRF data that we have.
>>>>
>>>> Thanks,
>>>>
>>>> Paul
>>>>
>>>>
>>>> On 10/04/2011 08:11 AM, Paul Oldenburg wrote:
>>>>> Sam,
>>>>>
>>>>> You should be able to use any data you want as observations, as
>>>>> long as you can get MET to read it. Once that is done, the
trick
>>>>> will be to set the fcst_field and obs_field settings in the
point_stat config file so that MET will match the data. If you send
me some of your test data, I can take a look.
>>>>>
>>>>> Paul
>>>>>
>>>>>
>>>>> On 10/04/2011 02:51 AM, sam.hawkins at vattenfall.com via RT wrote:
>>>>>>
>>>>>> <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=50422 >
>>>>>>
>>>>>> Paul,
>>>>>>
>>>>>> Thanks for your quick reply. I have the UPP compiled, so I will
give it a try. I don't actually have access to the mast data yet,
that will come in the next week or so. I'm just trying to set up all
of the tools I need first. However, I have some old mast data from
another site which I may use as a test. It's only up to 40m height.
Would it be possible to validate against that? Obviously I can change
the format.
>>>>>>
>>>>>> Sam.
>>>>>>
>>>>>>
>>>>>>
>>>>>> -----Original Message-----
>>>>>> From: Paul Oldenburg via RT [mailto:met_help at ucar.edu]
>>>>>> Sent: 03 October 2011 16:58
>>>>>> To: Hawkins Samuel Lennon (AE-DE)
>>>>>> Cc: met_help at mailman.ucar.edu
>>>>>> Subject: Re: [rt.rap.ucar.edu #50422] Validation of wind speeds
>>>>>> from mast
>>>>>>
>>>>>> Sam,
>>>>>>
>>>>>> What you have suggested would work. However, there is another
way to perform the verification that should yield closer height
matches between the forecast and obs. If you use the relatively new
Unified Post Processor (see http://www.dtcenter.org/wrf-
nmm/users/overview/upp_overview.php if you don't have an instance),
there is a way to interpolate the forecast wind speeds at various
heights above ground. In the UPP config file,
[UPP_path]/parm/wrf_cntrl.parm, look for the lines containing the
following four settings, then change the first group of 00000 in each
list to 22220 as I have done here:
>>>>>>
>>>>>> (SPE HUM AT FD HEIGHT) SCAL=( 5.0) L=(22220 00000 00000 00000
>>>>>> 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000)
>>>>>> (TEMP AT FD HEIGHTS ) SCAL=( 5.0) L=(22220 00000 00000 00000
>>>>>> 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000)
(U
>>>>>> WIND AT FD
>>>>>> HEIGHTS) SCAL=( 5.0) L=(22220 00000 00000 00000 00000 00000
00000
>>>>>> 00000 00000 00000 00000 00000 00000
>>>>>> 00000) (V WIND AT FD HEIGHTS) SCAL=( 5.0) L=(22220 00000
00000
>>>>>> 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
00000)
>>>>>>
>>>>>> I attached a version of this file with this change included.
Please move your original copy and replace it with the attached. When
you run UPP with this configuration, the output should include
interpolated WIND at 30m, 50m, 80m and 100m. Once you have performed
this interpolation on the forecast data, you can then specify the
verification fields in the point_stat config file like so:
>>>>>>
>>>>>> fcst_field[] = [ "WIND/Z80" ];
>>>>>> obs_field[] = [ "WIND/Z70-90" ];
>>>>>>
>>>>>> This would compare the interpolated 80m forecast wind speed to
any observation whose height is between 70m and 90m.
>>>>>>
>>>>>> I'm curious about your verification task, and I was wondering
if it would be possible for you to send me your observation and
forecast data? If you are using ascii2nc to pre-process your
observations, can you please send me the ascii obs file, too? I will
try to perform the type of verification that I am suggesting you try.
If the data is too large to email, you can upload it to our FTP site
by following these instructions:
>>>>>>
>>>>>> http://www.dtcenter.org/met/users/support/met_help.php#ftp
>>>>>>
>>>>>> If you have any questions, please let me know.
>>>>>>
>>>>>> Thanks,
>>>>>>
>>>>>> Paul
>>>>>>
>>>>>>
>>>>>> On 10/03/2011 09:17 AM, sam.hawkins at vattenfall.com via RT
wrote:
>>>>>>>
>>>>>>> Mon Oct 03 09:17:06 2011: Request 50422 was acted upon.
>>>>>>> Transaction: Ticket created by sam.hawkins at vattenfall.com
>>>>>>> Queue: met_help
>>>>>>> Subject: Validation of wind speeds from mast
>>>>>>> Owner: Nobody
>>>>>>> Requestors: sam.hawkins at vattenfall.com
>>>>>>> Status: new
>>>>>>> Ticket <URL:
>>>>>>> https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=50422
>>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>> Dear met users,
>>>>>>>
>>>>>>> I'm interested in using MET to validate WRF wind speed
forecasts against observations from masts at (multiple) heights, up
to 80m above the surface. Does anyone have experience of this?
Looking at the documentation, it seems I might need to interpolate
WRF output to the correct heights, and then pretend both the forecast
and observations are the 10m wind? Or is there an easier way?
>>>>>>>
>>>>>>> Thanks,
>>>>>>>
>>>>>>> Sam.
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>
>>>>>>
>>>>>
>>>>
>>>>
>>>
>>>
>>
>>
>
>
------------------------------------------------
Subject: RE: [rt.rap.ucar.edu #50422] Resolved: Validation of wind speeds from mast
From: sam.hawkins at vattenfall.com
Time: Thu Oct 13 06:16:03 2011
Paul,
Thanks for all the help, I have UPP and MET working together now. Just
one final question: are the height levels hard-coded into UPP, or is
it possible to specify your own?
Thanks,
Sam.
-----Original Message-----
From: Paul Oldenburg via RT [mailto:met_help at ucar.edu]
Sent: 12 October 2011 21:55
To: Hawkins Samuel Lennon (AE-DE)
Subject: [rt.rap.ucar.edu #50422] Resolved: Validation of wind speeds
from mast
According to our records, your request has been resolved. If you have
any further questions or concerns, please respond to this message.
------------------------------------------------
Subject: RE: [rt.rap.ucar.edu #50422] Resolved: Validation of wind speeds from mast
From: Paul Oldenburg
Time: Thu Oct 13 06:47:58 2011
Sam,
I'm afraid that those height levels cannot be adjusted. You probably
think that is ridiculous.
So do I. However, those particular heights are used for aircraft
forecasts or some other
application, so that is why they are implemented in the UPP. There
are higher heights available
if you keep setting 0's to 2's on the config file line that we
discussed. Sorry I can't be more
help.
Paul
>
> <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=50422 >
>
> Paul,
>
> Thanks for all the help, I have UPP and MET working together now.
Just one final question: are the
> height levels hard-coded into UPP, or is it possible to specify your
own?
>
> Thanks,
>
> Sam.
>
> -----Original Message-----
> From: Paul Oldenburg via RT [mailto:met_help at ucar.edu]
> Sent: 12 October 2011 21:55
> To: Hawkins Samuel Lennon (AE-DE)
> Subject: [rt.rap.ucar.edu #50422] Resolved: Validation of wind
speeds from mast
>
> According to our records, your request has been resolved. If you
have any further questions or
> concerns, please respond to this message.
>
------------------------------------------------
More information about the Met_help
mailing list