Mark,
GDPOINT can't automatically interpolate between vertical levels. If you're
interested in RELH at 650mb and only RELH (600, 700) are available, you
would need to do the following:
GFUNC=AVG(RELH@600%PRES,RELH@700%PRES)
All GFUNC functions can be found at:
http://www.unidata.ucar.edu/cgi-bin/gempak/manual/apxB_index
NSHARP, is a GEMPAK GUI which displays model soundings at LAT/LON points.
If you don't need to build a script which automatically calculates values at
different heights, you could manually get model soundings using NSHARP which
should give you temp / wind at different levels.
Another option would be to build a script which identifies the geopotential
height at model pressure levels.
IN GFS GEMPAK Grid files:
GFUNC=HGHT
If a pilot needs temp/wind at 5000m, you could identify the two model
pressure levels nearest to 5000m and then interpolate a value using nearest
neighbor or IDW.
You could always look into the IDV which probably does what you need and can
be run using ISL scripts.
Evan
Business Applications Meteorologist
Weather Trends International, Inc.
Phone 610-807-3197
http://www.wxtrends.com
-----Original Message-----
From: Mark Hughes [mailto:mhughes@xxxxxxxx]
Sent: Tuesday, November 17, 2009 7:16 PM
To: Evan Lowery
Subject: Re: [gembud] Interpolating levels in GRIB data
Hi Evan,
Thanks for your reply. I may not understand your example, but my
problem is I need point data at a level inbetween the levels in the model.
My application is to provide wind and temp data for pilots. They think
of altitude in flight levels which don't convert to
the exact millibar levels in the model.
So, in your example below, what if I want RELH at 650m mb for NYC? I
suspect there is some way to use GFUNC.
Thanks,
Mark
BTW: gdpoint will take the input GPOINT=NYC. How cool is that?
Evan Lowery wrote:
>Hi Mark,
>
>You could do the following:
>
>
>Using GDINFO output, identify the GFUNC, GLEVEL, GVCORD of interest. For
>example, using today's 0Z GFS grid I do the following:
>
>GDFILE = gfs_3_20091116_0000_006.grd
> LSTALL = YES
> OUTPUT = T
> GDATTIM = all
> GLEVEL = all
> GVCORD = all
> GFUNC = all
> GEMPAK-GDINFO>r
>
>I'm interested in RELH at 800, 700, 600, and 500 mb in New York City, NYC
>(40.783, -73.883). Within the GDINFO output I see:
>
> 100 091116/0000F006 800 PRES RELH
> 116 091116/0000F006 700 PRES RELH
> 132 091116/0000F006 600 PRES RELH
> 148 091116/0000F006 500 PRES RELH
>
>Using GDPOINT, pull the RELH values of interest:
>
>gdpoint
>GDATTIM = 091116/0000F006
> GDFILE = gfs_3_20091116_0000_006.grd
> GLEVEL = 800
> GPOINT = 40.783;-73.883
> GVCORD = PRES
> GFUNC = RELH
> SCALE = 0
>GEMPAK-GDPOINT>r
>
>GDFILE: gfs_3_20091116_0000_006.grd
>GPOINT: 40.783;-73.883
>GVCORD: PRES
>GLEVEL: 800
>GFUNC : RELH
>SCALE : 0
> 091116/0000F006 : 58.81023
> Parameters requested: GDATTIM,GDFILE,GLEVEL,GPOINT,GVCORD,GFUNC,SCALE.
> GEMPAK-GDPOINT>GLEVEL=700
> GEMPAK-GDPOINT>r
>
>GDFILE: gfs_3_20091116_0000_006.grd
>GPOINT: 40.783;-73.883
>GVCORD: PRES
>GLEVEL: 700
>GFUNC : RELH
>SCALE : 0
> 091116/0000F006 : 27.16799
> Parameters requested: GDATTIM,GDFILE,GLEVEL,GPOINT,GVCORD,GFUNC,SCALE.
> GEMPAK-GDPOINT>GLEVEL=600
> GEMPAK-GDPOINT>r
>
>GDFILE: gfs_3_20091116_0000_006.grd
>GPOINT: 40.783;-73.883
>GVCORD: PRES
>GLEVEL: 600
>GFUNC : RELH
>SCALE : 0
> 091116/0000F006 : 8.15741
> Parameters requested: GDATTIM,GDFILE,GLEVEL,GPOINT,GVCORD,GFUNC,SCALE.
> GEMPAK-GDPOINT>GLEVEL=500
> GEMPAK-GDPOINT>r
>
>GDFILE: gfs_3_20091116_0000_006.grd
>GPOINT: 40.783;-73.883
>GVCORD: PRES
>GLEVEL: 500
>GFUNC : RELH
>SCALE : 0
> 091116/0000F006 : 3.61722
>
>
>GDPOINT uses the 8 closest GRID points to calculate an interpolated value
at
>the LAT/LON you request. Using 8 points could be an issue with coarser
>models as the interpolation would include grid points far removed from your
>LAT/LON point.
>
>Regards,
>Evan
>
>Business Applications Meteorologist
>Weather Trends International, Inc.
>Phone 610-807-3197
>http://www.wxtrends.com
>
>-----Original Message-----
>From: gembud-bounces@xxxxxxxxxxxxxxxx
>[mailto:gembud-bounces@xxxxxxxxxxxxxxxx] On Behalf Of Mark Hughes
>Sent: Tuesday, November 17, 2009 12:48 PM
>To: gembud@xxxxxxxxxxxxxxxx
>Subject: [gembud] Interpolating levels in GRIB data
>
>Hi All,
>
>I am working with models produced from wind and temperature GRIB data.
>(DCGRIB2 produces a .gem model from HT*, HU* and HV* messages).
>I am trying to extract point data using GDPOINT at an arbitrary level.
>Can anyone tell me how to interpolate? Or, as an alternative, how can I
>create a new grid with finer resolution in the level dimension?
>
>Many thanks,
>
>Mark Hughes
>Global Weather Dynamics
>
>_______________________________________________
>gembud mailing list
>gembud@xxxxxxxxxxxxxxxx
>For list information or to unsubscribe, visit:
>http://www.unidata.ucar.edu/mailing_lists/
>
>
>
>
