[ncl-talk] diagonal cross section with wrf_user_intrp3d

Matthew Fearon Matthew.Fearon at dri.edu
Mon Mar 23 13:14:03 MDT 2015 

Alex,
I'll take a look. Thanks for your time and for sharing this.
best regards,
Matt
________________________________
From: Alexander Schaefer [alexander.schaefer at mines.sdsmt.edu]
Sent: Monday, March 23, 2015 12:01 PM
To: Matthew Fearon
Cc: ncl-talk at ucar.edu
Subject: Re: [ncl-talk] diagonal cross section with wrf_user_intrp3d

Matt,

Attached is a cross-section script I use. It goes rather in-depth.  This one simply loops over the index values from point A to point B, which would be at a 45 deg angle.  To get different angled cross-sections you would have to create an array of all the points between the points on the line you want. It may be easier to use the point/angle method and just control the range of the data plotted.

Best of luck,
-Alex

On Mon, Mar 23, 2015 at 12:01 PM, Matthew Fearon <Matthew.Fearon at dri.edu<mailto:Matthew.Fearon at dri.edu>> wrote:
Alex,
Yes. As you have time, please send your script. Would be good see an alternative.
thank you,
Matt

________________________________
From: Alexander Schaefer [alexander.schaefer at mines.sdsmt.edu<mailto:alexander.schaefer at mines.sdsmt.edu>]
Sent: Monday, March 23, 2015 10:52 AM
To: Matthew Fearon
Cc: Evelyn Grell - NOAA Affiliate; ncl-talk at ucar.edu<mailto:ncl-talk at ucar.edu>; wrf-help at ucar.edu<mailto:wrf-help at ucar.edu>

Subject: Re: [ncl-talk] diagonal cross section with wrf_user_intrp3d

Hi Matt,

I haven't been able to get the diagonal slicing with wrf_user_intrp3d to work right but I have been able to get diagonal cross-sections, though it was not pretty.  Essentially created some arrays of the points that were in the cross-section and then used wrf_interp_1d to interpolate the data at each point in the list to higher resolution in the vertical. Works best with points on the 45 deg angle to reduce horizontal space between points but I suppose that any array of points should work.  I played with the "2-point" example for multiple days before going other routes.  This way does however let you control the horizontal extent like you were looking to do.


I'll search for the script and pass it along if you would like to give it a whirl.

-Alex

On Mon, Mar 23, 2015 at 11:40 AM, Matthew Fearon <Matthew.Fearon at dri.edu<mailto:Matthew.Fearon at dri.edu>> wrote:
Hi Evelyn,
Thank you for your help. I was able to get a diagonal slice using the single pivot point (and angle) example. Still no luck with the two point example. The latter would be better as I could limit the horizontal extent more easily. I'll try again later.
thanks again,
Matt

________________________________
From: Evelyn Grell - NOAA Affiliate [evelyn.grell at noaa.gov<mailto:evelyn.grell at noaa.gov>]
Sent: Monday, March 23, 2015 8:12 AM
To: Matthew Fearon
Cc: wrf-help at ucar.edu<mailto:wrf-help at ucar.edu>; ncl-talk at ucar.edu<mailto:ncl-talk at ucar.edu>
Subject: Re: [ncl-talk] diagonal cross section with wrf_user_intrp3d

Matthew,

Regarding your second question (the easier one), I got confused by this too.  I believe the order of the variables got changed between versions:
In v6.1.2, the order to specify 2 points for a cross section has to be x1,x2,y1,y2.
In v6.2.1, the order to specify 2 points has to be x1,y1,x2,y2.
(maybe wrfhelp could add a note in the example.)

You can check by looking at the code in your
$NCARG_ROOT/lib/ncarg/nclscripts/wrf/WRFUserARW.ncl

Another thing that I ran into is that it assumes you are defining your points in "WRF" space (ie arrays start with 1) rather than in "NCL" space (ie arrays start with 0) so it subtracts 1 for you...

In wrf_user_interp3d:

; set vertical cross section
     if (opts) then
       xy = wrf_user_set_xy( z, loc_param(0)-1, loc_param(2)-1, \    ; the -1 is for NCL dimensions
                                loc_param(1)-1, loc_param(3)-1, \
                                angle, opts )


(loc_param is the variable containing your XS endpoints).

About your first question, maybe you are missing the 2d interpolation that defines the x-axis?

something like:
    if (xang.gt.20. .and. xang.lt.340) then
      X_plane = wrf_user_intrp2d(lon,xspts,0.,True)
    else
      X_plane = wrf_user_intrp2d(lat,xspts,0.,True)
    end if
    dimsX = dimsizes(X_plane)
    xspan = dimsX(0)-1

     res at tmXBValues              = fspan(0,xspan,2)

Hope that helps,
Evelyn Grell


On Mon, Mar 23, 2015 at 12:21 AM, Matthew Fearon <Matthew.Fearon at dri.edu<mailto:Matthew.Fearon at dri.edu>> wrote:

Dear WRF-help and NCL users,

Has anyone successfully plotted a cross section with wrf data
along a diagonal, say southeast to northwest or vice versa, using
wrf_user_intrp3d? North/south or east/west works fine, but from the
documentation, a diagonal seemed possible using the example below,
although I can't get it working?

Also, the plane is defined (/xstart,xend , ystart,yend/) for sure
rather than (/xstart,ystart , xend,yend/) instead?

thanks,
Matt




load "$NCARG_ROOT/lib/ncarg/nclscripts/csm/gsn_code.ncl"
  load "$NCARG_ROOT/lib/ncarg/nclscripts/wrf/WRFUserARW.ncl"

  a = addfile<https://www.ncl.ucar.edu/Document/Functions/Built-in/addfile.shtml>("wrfout_d01_2000-01-24_12:00:00.nc<http://00.nc>","r")

  time = 1
  tc = wrf_user_getvar<https://www.ncl.ucar.edu/Document/Functions/WRF_arw/wrf_user_getvar.shtml>(a,"tc",time)   ; T [C]
  z  = wrf_user_getvar<https://www.ncl.ucar.edu/Document/Functions/WRF_arw/wrf_user_getvar.shtml>(a,"z",time)    ; z on mass points

  plane = (/  40,81 , 259,81  /) ; approx. START x;y and END x;y point

  ; Extract cross section from point A to point B, as defined in "plane"
  ; And vertically interpolate to height coordinates ("z")
  tc_plane = wrf_user_intrp3d(tc,z,"v",plane,0.,True)

_______________________________________________
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



-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.ucar.edu/pipermail/ncl-talk/attachments/20150323/130c9e45/attachment.html

More information about the ncl-talk mailing list