[ncl-talk] Missing grid lines at zero degree

Arnold.Sullivan at csiro.au Arnold.Sullivan at csiro.au
Wed May 27 20:14:05 MDT 2015


Many thanks all.

Indeed gsnAddCyclic doesn’t work in this case. Mary’s suggestion is brilliant, and it works!

Thanks again,

Arnold

From: Mary Haley [mailto:haley at ucar.edu]
Sent: Wednesday, 27 May 2015 11:31 PM
To: Alan Brammer
Cc: Sullivan, Arnold (O&A, Aspendale); ncl-talk at ucar.edu; Kyle Griffin
Subject: Re: [ncl-talk] Missing grid lines at zero degree

Hi all,

Unfortunately, gsnAddCyclic isn't seen by gsn_coordinates. This is an omission that should be fixed. I'll make a ticket for this.

Meanwhile, you can try the gsn_add_cyclic_point function, which is unadvertised (and hence not guaranteed to live forever). This function increases the input array by 1 point in the longitude dimension and it increases the longitude array associated with it. I think it should work with coordinate arrays and lat2d/lon2d attributes.

  pres                   = True
  pres at gsnCoordsAsLines  = True        ; Default is points
  pres at gsLineThicknessF  = .1
  pres at gsLineColor       = "Gray25"
  pres at gsnCoordsAttach  = True
  gsn_coordinates(wks,plot1,gsn_add_cyclic_point(vMTemp),pres)

I didn't actually try this (because I'm running out the door as I type this), so let me know if it doesn't work.

--Mary


On Wed, May 27, 2015 at 7:20 AM, Alan Brammer <abrammer at albany.edu<mailto:abrammer at albany.edu>> wrote:
Didn't Kyle Griffin reply to this message 2 days ago?
Did his suggestion not work?

"Hi Arnold,

Have you tried also setting

pres at gsnAddCyclic = True

This resource defaults to true in some of the newest versions of NCL, but defaulted to False for a long time and produces issues like this. gsnAddCyclic basically tells NCL to contour between the first (longitude=0) and last (longitude=191 in your case) the same as it does other adjacent points, despite these points not technically being adjacent in the dataset.

If there is still an issues after trying that, let us know and be sure to include the list.


Kyle

----------------------------------------
Kyle S. Griffin
Department of Atmospheric and Oceanic Sciences
University of Wisconsin - Madison
Room 1421
1225 W Dayton St, Madison, WI 53706
Email: ksgriffin2 at wisc.edu<mailto:ksgriffin2 at wisc.edu>
​"​


On 27 May 2015 at 02:48, <Arnold.Sullivan at csiro.au<mailto:Arnold.Sullivan at csiro.au>> wrote:
Hi there,

Somehow I get a gap along the zero degree while plot polar region plus using gsn_coordinates to plot grid lat/lon lines.

Is there any way to fill that gap?

The code that I use to plot the grid lines:
;---Data and grid read in:
  mskFile = addfile("access.n96.newu2.nc<http://access.n96.newu2.nc>","r")
  vTemp = rm_single_dims(mskFile->NEWVAR2(0,0,:,::2))
  vMTemp = vTemp

;---Call function to attach lat/lon lines.
  pres                   = True
  pres at gsnCoordsAsLines  = True        ; Default is points
  pres at gsLineThicknessF  = .1
  pres at gsLineColor       = "Gray25"
  pres at gsnCoordsAttach  = True
  gsn_coordinates(wks,plot1,vMTemp,pres)


The input file information:
;---Ncdump the netcdf file:
dimensions:
        LONGITUDE = 192 ;
        LATITUDE_1 = 144 ;
        HYBRID_HT = 85 ;
variables:
        double LONGITUDE(LONGITUDE) ;
                LONGITUDE:units = "degrees_east" ;
                LONGITUDE:point_spacing = "even" ;
                LONGITUDE:modulo = " " ;
                LONGITUDE:axis = "X" ;
        double LATITUDE_1(LATITUDE_1) ;
                LATITUDE_1:units = "degrees_north" ;
                LATITUDE_1:point_spacing = "even" ;
                LATITUDE_1:axis = "Y" ;
        float NEWVAR2(T, HYBRID_HT, LATITUDE_1, LONGITUDE) ;
                NEWVAR2:missing_value = -1.e+34f ;
                NEWVAR2:_FillValue = -1.e+34f ;
                NEWVAR2:long_name = "U[G=NEW_GRID]" ;
                NEWVAR2:history = "From access.n96.uv" ;


Regards,

Arnold


_______________________________________________
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/20150528/a5e23ce2/attachment.html 


More information about the ncl-talk mailing list