Re: vinth2p with CMIP data

From: Dennis Shea <shea_at_nyahnyahspammersnyahnyah>
Date: Thu Mar 29 2012 - 12:21:34 MDT

Hi Kelly,

The vinth2p function was implemented within NCL directly from
an old processor. No changes were made. The vinth2p* suite
of function require that the usint of 'ps' be Pa, the units of
po be 'hpa' and, further that the data be ordered from
top-of-atmosphere to the bottom. The GFDL model has data
ordered from bottom-to-top. The user must reorder. In NCL,
the reordering is simple
;************************************************
; As per documentation, the function requires data to be
; ordered from top to bottom. Reverse the order using ::-1 syntax.
;************************************************
    ua = ua(:,::-1,:,:)
    a = a(::-1)
    b = b(::-1)

Two scripts are attached:
[1] Has a number of print statements to illustrate each step.
[2] The other could be used for production

The test file used is at:

ftp ftp.cgd.ucar.edu
anonymous
email
cd /ftp/pub/shea/GFDL
get ua_6hrLev_GFDL-ESM2M_rcp85_r1i1p1.nc
quit

Good Luck

On 03/28/2012 11:19 AM, kelly lombardo wrote:
> Hi Dennis
>
> The original file was too large, so I transfered over a file containing
> a small part of the data named ua_GFDL-ESM2M.part.nc
>
> Thanks so much
> kelly
>
>
> ------------------------------------------------------------------------
> *From:* Dennis Shea <shea@ucar.edu>
> *To:* kelly lombardo <kellyann7319@yahoo.com>
> *Sent:* Tuesday, March 27, 2012 7:05 PM
> *Subject:* Re: [ncl-talk] vinth2p with CMIP data
>
> Hi Kelly,
>
> What I would like is one of the original GFDL hybrid model
> files. ie, one of the source files. I want to look at the
>
>
> THX
> D
>
> On 03/27/2012 04:14 PM, kelly lombardo wrote:
> > Great, thanks very much.
> > The files I sent over are: sigma.2.pres.emailed.ncl and ua.test.p0.nc
> > NCL version: 6.0.0
> > System: Linux cloud.somas.stonybrook.edu
> > <http://cloud.somas.stonybrook.edu/> 2.6.18-238.el5 #1 SMP Thu Jan 13
> > 15:51:15 EST 2011 x86_64 x86_64 x86_64 GNU/Linux
> >
> > Thanks again
> > kelly
> > ------------------------------------------------------------------------
> > *From:* Dennis Shea <shea@ucar.edu <mailto:shea@ucar.edu>>
> > *To:* kelly lombardo <kellyann7319@yahoo.com
> <mailto:kellyann7319@yahoo.com>>
> > *Sent:* Tuesday, March 27, 2012 3:21 PM
> > *Subject:* Re: [ncl-talk] vinth2p with CMIP data
> >
> > offline
> >
> > You can send a sample file and clean script to
> >
> > ftp ftp.cgd.ucar.edu <http://ftp.cgd.ucar.edu>
> > anonymous
> > <use your email address for the password>
> > cd incoming
> > put ...
> > put ...
> > quit
> >
> > Send email to ncl-talk with the following information:
> >
> > The names of the files after you have successfully completed the
> > transfer.
> > We cannot look at the ftp/incoming directory for security reasons.
> > The version of NCL you are running (ncl -V).
> > The type of system you are on (uname -a).
> >
> > On 3/27/12 12:10 PM, kelly lombardo wrote:
> > > I was just interpolating to 850; 700; 500; 300; 200 hPa, nothing below
> > > ground level. Could it be that this routine isn't calibrated for GFDL
> > > model output, and can only be used with the CESM? I assumed that the A
> > > and B coefficients would be alright since it looks like these
> > > coefficients in both the GFDL and CESM are used for the middle point in
> > > between the model grid levels. Maybe this was an incorrect assumption?
> > >
> > >
> > >
> > >
> ------------------------------------------------------------------------
> > > *From:* Dennis Shea <shea@ucar.edu <mailto:shea@ucar.edu>
> <mailto:shea@ucar.edu <mailto:shea@ucar.edu>>>
> > > *To:* kelly lombardo <kellyann7319@yahoo.com
> <mailto:kellyann7319@yahoo.com>
> > <mailto:kellyann7319@yahoo.com <mailto:kellyann7319@yahoo.com>>>
> > > *Cc:* "ncl-talk@ucar.edu <mailto:ncl-talk@ucar.edu>
> <mailto:ncl-talk@ucar.edu <mailto:ncl-talk@ucar.edu>>"
> > <ncl-talk@ucar.edu <mailto:ncl-talk@ucar.edu>
> <mailto:ncl-talk@ucar.edu <mailto:ncl-talk@ucar.edu>>>
> > > *Sent:* Tuesday, March 27, 2012 1:50 PM
> > > *Subject:* Re: [ncl-talk] vinth2p with CMIP data
> > >
> > > If you are interpolating, there should be no winds that
> > > look "too strong". If you are extrapolating (below ground level)
> > > then this could be the case. vinth2p_ecmwf will extrapolate
> > > temperature (using the lapse rate) and/or geopotential (lapse
> > > rate and hydrostatic formula). *All* other variables are
> > > set to the the nearest non-missing value. This could result
> > > in "too strong" winds.
> > >
> > > As noted in the documentation, the vinth2p was taken directly
> > > from the old model processor. NCL changed nothing.
> > >
> > > My suggestions:
> > > (1) do not extrapolate [*all* extrapolations not guided
> > > by physics are dangerous, even then use with caution]
> > >
> > > (2) Look at "Grid Fill" examples at
> > > http://www.ncl.ucar.edu/Applications/grid_fill.shtml
> > > In particular the ones using 'poisson_grid_fill'
> > > Good Luck
> > >
> > > On 03/27/2012 11:27 AM, kelly lombardo wrote:
> > > > Hey everyone
> > > >
> > > > I am trying to interpolate CMIP GFDL-ESM2M data (such as U, V, T)
> which
> > > > is stored on hybrid sigma-pressure coordinates to pressure
> coordinates
> > > > using vinth2p. I was able to interpolate the data, but the winds look
> > > > too strong. Has anyone encountered a similar problem? Or does anyone
> > > > know where I can find the vinth2p code so I can see how the
> > > > interpolation is calculated?
> > > >
> > > > Thanks
> > > > kelly
> > > >
> > > >
> > > > _______________________________________________
> > > > ncl-talk mailing list
> > > > List instructions, subscriber options, unsubscribe:
> > > > http://mailman.ucar.edu/mailman/listinfo/ncl-talk
> > >
> > >
> > >
> >
> >
>
>
>

_______________________________________________
ncl-talk mailing list
List instructions, subscriber options, unsubscribe:
http://mailman.ucar.edu/mailman/listinfo/ncl-talk

Received on Thu Mar 29 12:21:44 2012

This archive was generated by hypermail 2.1.8 : Mon Apr 09 2012 - 13:43:03 MDT