Re: gsn_csm_contour_map problem

 From printVarSummary(tc)

--
Variable: tc
Total Size: 3015874936 bytes   <=== 3GB
753968734 values
Number of Dimensions: 3
Dimensions and sizes: [bottom_top | 34] x [south_north | 4349] x
[west_east | 5099]
--
The variable is 3GB in size. Only the recently
released v6.0.0-beta version of NCL can handle variables > 2GB.
Please see:
http://www.ncl.ucar.edu/current_release.shtml
--
What version of NCL are you using? What OS?
%> ncl -V
%> uname -a
======================================================
The wrf_user_getvar function does not support extracting
a specific level. However, you can use NCL directly
for variables that are on the file.
   tk = f->T(t,10,:,:)  ; perturbation temp; less < 2GB
   tbase =
   tc = (tk+tbase)-273.15
   tc@units = "degC"
On 3/5/11 7:11 AM, sk wrote:
> Hi There,
> the gsn_csm_contour_map reports a strange error for me:
>
> fatal:ContourPlotDraw: CPCLAM/AREDAM - UNABLE TO INSERT NODE INTO MAP
> fatal:ContourPlotDraw: draw error
> warning:WorkstationDeactivate: workstation not active or not opened
>
> a simpler version of the code is:
>
> tc = wrf_user_getvar(a,"tc",t)
> lat = wrf_user_getvar(a,"lat",0)
> lon = wrf_user_getvar(a,"lon",0)
> tc@lat2d=lat
> tc@lon2d=lon
> ;;start plotting
> wks = gsn_open_wks("x11","Temp_1k") ;
>
> res = True
> res@cnFillOn = True
> WRF_map_c(a,res,0)
> plot =gsn_csm_contour_map(wks,tc(10,:,:),res)
>
> and the output from print(tc) :
> Variable: tc
> Type: float
> Total Size: 3015874936 bytes
> 753968734 values
> Number of Dimensions: 3
> Dimensions and sizes: [bottom_top | 34] x [south_north | 4349] x
> [west_east | 5099]
> Coordinates:
> Number Of Attributes: 8
> lon2d : <ARRAY of 22175551 elements>
> lat2d : <ARRAY of 22175551 elements>
> description : Temperature
> FieldType : 104
> MemoryOrder : XYZ
> units : C
> stagger :
> coordinates : XLONG XLAT
> (0,0,0) 18.35291
> (0,0,1) 18.35437
> (0,0,2) 18.35349
> (0,0,3) 18.34885
>
> some idea how can i solve the problem? is it the variable size? the same
> code with smaller variables work fine.
>
> Regards,
> Ico
>
>
>
>
>
> _______________________________________________
> 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