[Met_help] [rt.rap.ucar.edu #41690] History for Vertical levels

RAL HelpDesk {for John Halley Gotway} met_help at ucar.edu
Fri Oct 29 13:25:34 MDT 2010


----------------------------------------------------------------
  Initial Request
----------------------------------------------------------------

  Hi,

I was just looking through the v3.0 user manual (I've still been using 
v0.9 up until very recently), and I noticed that there is functionality 
in point_stat to verify a single model level (referred to as ZNNN in the 
documentation).  Does this mean if I had a WRF Grib file on the native 
vertical grid, which we do output routinely, that I can verify directly 
on each of the model levels, or will I still need to verify on pressure 
levels?  Ideally, I would like something that verifies on height AGL not 
pressure, which is why I've still been using our in-house verification 
as a supplement to MET's pressure-level verification capabilities.  If 
you could help clarify these things for me, I would really appreciate 
it.  Thanks!


Jeff Zielonka

The Pennsylvania State University
Department of Meteorology
614 Walker Building
University Park, PA 16802


----------------------------------------------------------------
  Complete Ticket History
----------------------------------------------------------------

Subject: Re: [rt.rap.ucar.edu #41690] Vertical levels
From: John Halley Gotway
Time: Fri Oct 22 12:54:36 2010

Jeff,

I'd say the answer to your question is - maybe.

It is true that we've enhanced Point-Stat to better handle verifying
on vertical levels.  For example, if you select "UGRD/Z50", Point-Stat
will look for a U-component of wind GRIB record that's
defined on a vertical level (either as 50m AGL or 50m MSL).  It will
then look at the observation height value and verify that 50m using
any observations that have a height value of 50m.  You could
also verify a range of forecast levels.  Suppose you have UGRD output
every 10 meters up to 100 meters.  You could verify "UGRD/Z40-60" and
Point-Stat would use all observations falling between 40 and
60 meters and interpolate the forecast values to the observation
heights.

Ultimately, it's up to the user to make sure that the vertical level
values match up and make sense.  You'd need to make sure that your
model heights and observation heights are defined consistently -
either both MSL or both AGL.  Point-Stat will not convert between AGL
and MSL for you.

What I'm not sure about it is exactly how your GRIB model output is
defined.  If you have output for vertical levels (either AGL or MSL),
you should be able to use Point-Stat to verify it.  But if
your model output is on the native model levels (sigma levels),
probably not.

Hope that helps.

Thanks,
John Halley Gotway
met_help at ucar.edu

On 10/22/2010 07:43 AM, RAL HelpDesk {for Jeff Zielonka} wrote:
>
> Fri Oct 22 07:43:56 2010: Request 41690 was acted upon.
> Transaction: Ticket created by zielonka at meteo.psu.edu
>        Queue: met_help
>      Subject: Vertical levels
>        Owner: Nobody
>   Requestors: zielonka at meteo.psu.edu
>       Status: new
>  Ticket <URL:
https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=41690 >
>
>
>   Hi,
>
> I was just looking through the v3.0 user manual (I've still been
using
> v0.9 up until very recently), and I noticed that there is
functionality
> in point_stat to verify a single model level (referred to as ZNNN in
the
> documentation).  Does this mean if I had a WRF Grib file on the
native
> vertical grid, which we do output routinely, that I can verify
directly
> on each of the model levels, or will I still need to verify on
pressure
> levels?  Ideally, I would like something that verifies on height AGL
not
> pressure, which is why I've still been using our in-house
verification
> as a supplement to MET's pressure-level verification capabilities.
If
> you could help clarify these things for me, I would really
appreciate
> it.  Thanks!
>
>
> Jeff Zielonka
>
> The Pennsylvania State University
> Department of Meteorology
> 614 Walker Building
> University Park, PA 16802

------------------------------------------------


More information about the Met_help mailing list