Re: wrf_user_interp3d documentation

From: Mary Haley <haley_at_nyahnyahspammersnyahnyah>
Date: Wed Oct 16 2013 - 09:24:44 MDT

Hi folks,

I started looking into this yesterday, but didn't get back to it. I think Sam may be correct too.

I've CC'ed wrfhelp. wrfhelp: can you suggest some changes to the documentation that would clear up some of the confusion mentioned here? I'll be happy to make the update.

Thanks,

--Mary

On Oct 15, 2013, at 4:18 PM, Jatin Kala <jatin.kala.jk@gmail.com> wrote:

> Hi Sam,
> I think you are right, should be interpolated *from*.
> I think it should also be added to the documentation, that when "vert" is height, plot type is "h" and and loc to a height like e.g. 50m or whatever, what will be returned is values above sea-level, not ground-level, which most people want. The model terrain height needs to be subtracted from the geopotential height before interpolation to get height above ground level. There are several posts on the WRF-NCL forum on this and i think an example on the NCL web-page would help.
> My 2 cents.
> Cheers,
> Jatin.
>
>
>
> On 16/10/2013, at 2:28 AM, <sam.hawkins@vattenfall.com> <sam.hawkins@vattenfall.com> wrote:
>
>> Maybe I'm missing something here, but I'm finding the documentation of wrf_user_intrp3d confusing. The docs say:
>>
>> function wrf_user_intrp3d (
>> var3d : numeric,
>> vert : numeric,
>> plot_type : string,
>> loc [*] : numeric, ; up to four values
>> angle : numeric,
>> res [1] : logical
>> )
>>
>> var3d: Data on model levels that will be interpolated. The rightmost dimensions of this array is nz x ny x nx.
>>
>> vert: Array of vertical coordinate to interpolate **to**. This must either be pressure/height. Dimensions must be the same as those of var3d. If vert is pressure, the units may be either hPa or Pa.
>>
>> plot_type: Plot orientation. Use "h" for horizontal plots, and "v" for vertical cross sections.
>>
>> Loc: Interpolation information.
>>
>> Angle: Only valid for cross sections where a plane will be plotted through a given point on the model domain. 0.0 represents a S-N cross section and 90.0 a W-E cross section.
>>
>> Res: set to True if plotting a cross section from point A to point B; otherwise set to False.
>>
>>
>>
>> But none of the call signatures in the examples match this. It seems vert should be an array of coordinates to interpolate **from**: i.e.
>> http://www.mmm.ucar.edu/wrf/OnLineTutorial/Graphics/NCL/Examples/LEVELS_INTERP/wrf_Height2.ncl
>>
>> ;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;
>> ; First get the variables we will need
>>
>> tc = wrf_user_getvar(a,"tc",-1) ; T in C
>> u = wrf_user_getvar(a,"ua",-1) ; u averaged to mass points
>> v = wrf_user_getvar(a,"va",-1) ; v averaged to mass points
>> p = wrf_user_getvar(a, "pressure",-1) ; pressure is our vertical coordinate
>> z = wrf_user_getvar(a, "z",-1) ; grid point height
>> rh = wrf_user_getvar(a,"rh",-1) ; relative humidity
>>
>> ; The specific height levels that we want the data interpolated to.
>> ; And interpolate to these levels
>> height_levels = (/ 250., 2000./) ; height levels to plot - in meter
>> nlevels = dimsizes(height_levels) ; number of height levels
>>
>> p_plane = wrf_user_intrp3d( p,z,"h",height_levels,0.,False)
>> tc_plane = wrf_user_intrp3d(tc,z,"h",height_levels,0.,False)
>> rh_plane = wrf_user_intrp3d(rh,z,"h",height_levels,0.,False)
>> u_plane = wrf_user_intrp3d( u,z,"h",height_levels,0.,False)
>> v_plane = wrf_user_intrp3d( v,z,"h",height_levels,0.,False)
>>
>> u_plane = u_plane*1.94386 ; kts
>> v_plane = v_plane*1.94386 ; kts
>> u_plane@units = "kts"
>> v_plane@units = "kts"
>>
>> ;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;
>> _______________________________________________
>> 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

_______________________________________________
ncl-talk mailing list
List instructions, subscriber options, unsubscribe:
http://mailman.ucar.edu/mailman/listinfo/ncl-talk
Received on Wed Oct 16 09:24:59 2013

This archive was generated by hypermail 2.1.8 : Tue Oct 22 2013 - 10:35:27 MDT