Re: Contouring temperature with GFS dataset.

From: Caroline Serraud <cs_at_nyahnyahspammersnyahnyah>
Date: Thu Jun 10 2010 - 10:02:34 MDT

Hi Mary,

You're totally right, when I comment out the gsnBoxMargin resource, the
script works fine.
In fact, I already use ImageMagick's convert with -trim option to remove
white space that I obtained even with the gsnBoxMargin resource, so
commenting it out doesn't bother me at all.

Thank you very much for your help, I would never have suspected that
specific resource!
Caroline

Mary Haley a écrit :
> Caroline,
>
> The problem appears to be with setting gsnBoxMargin to 0.0.
> This is causing an internal value of 1.0 to be calculated for vpWidthF and
> 0.0 for vpXF which NCL doesn't like for this particular case.
>
> Is there a reason you need the plot to be as big as possible? Until we get a fix for this,
> we might be able to provide you with a work-around if I know what you need to do with
> the image.
>
> For example, I've had good results with making an image larger and trimming all white space
> by using ImageMagick's "convert" utility, along with "-trim" to remove white space, and
> "-geometry" to increase the geometry.
>
> --Mary
>
> On Jun 7, 2010, at 9:35 AM, Mary Haley wrote:
>
>
>> Hi Caroline,
>>
>> This sounds like a possible memory problem.
>>
>> I would like to be able to run your script here .
>>
>> Can you set:
>>
>> res@gsnDebugWriteFileName = "serraud"
>>
>> and rerun your script? This will create "serraud.nc", "serraud.ncl" and "serraud.res".
>> If you can then email me serraud.nc and serraud.ncl I can run it here. I don't need the *.res file.
>>
>> If the data file is too big (> 10 Mb), let me know.
>>
>> Thanks,
>>
>> --Mary
>>
>>
>>
>>
>> On Jun 7, 2010, at 9:03 AM, Caroline Serraud wrote:
>>
>>
>>> Hello everyone,
>>>
>>> I am facing some trouble to plot contour of 2m-temperature from GFS data downloaded from NCEP FTP site (for example, gfs.t18z.pgrb2f06)
>>> NCL crashes when I run my script with default contour fill mode (area fill mode) but is ok with RasterFill mode.
>>>
>>> But with a 0,5° resolution I would really appreciate area mode because raster mode is not that pretty (see example attached)
>>> Did someone already experience plotting filled contour for 2m-temperature from GFS?
>>>
>>> Caroline
>>>
>>> --
>>> Caroline Serraud - Ingénieur Etudes & Développements
>>> ------------------------------------------------------
>>> METEO CONSULT / La Chaine Météo - Groupe Figaro
>>> Département Informatique
>>> Domaine de Marsinval F-78540 Vernouillet, FRANCE
>>> Tél : 01 39 28 1990 - Fax : 01 39 71 85 31
>>> e-mail : cs@meteoconsult.fr
>>> ------------------------------------------------------
>>> Toute la météo sur le Web : http://www.meteoconsult.fr
>>> Toute la météo par téléphone : 3201
>>> ;*************************************************
>>> ; tempe_gfs.ncl
>>> ;************************************************
>>> load "$NCARG_ROOT/lib/ncarg/nclscripts/csm/gsn_code.ncl"
>>> load "$NCARG_ROOT/lib/ncarg/nclscripts/csm/gsn_csm.ncl"
>>> load "$NCARG_ROOT/lib/ncarg/nclscripts/csm/contributed.ncl"
>>> load "$NCARG_ROOT/lib/ncarg/nclscripts/csm/shea_util.ncl"
>>> ;************************************************
>>> begin
>>> if (.not.isvar("v_file")) then
>>> print("Le nom du fichier source n'est pas defini")
>>> exit()
>>> end if
>>> ;************************************************
>>> ; read in netCDF file
>>> ;************************************************
>>> a = addfile(v_file,"r")
>>> ;************************************************
>>> ; read in zonal [u] and meridional [v] winds
>>> ;************************************************
>>> t = a->TMP_P0_L103_GLL0(:,:)
>>> t = t - 273.15
>>> delete(t@units)
>>> delete(t@long_name)
>>>
>>>
>>> i_lat_min = v_lat_min + 0
>>> i_lat_max = v_lat_max + 0
>>> i_lon_min = v_lon_min + 0
>>> i_lon_max = v_lon_max + 0
>>>
>>> ;************************************************
>>> ; create plot
>>> ;************************************************
>>> wks_type = "ps"
>>> wks =gsn_open_wks(wks_type,"tempe")
>>>
>>> gsn_define_colormap(wks,"BlWhRe") ; choose color map
>>>
>>> res = True ; plot mods desired
>>> res@gsnMaximize = True
>>> res@gsnPaperOrientation = "landscape"
>>> res@gsnBoxMargin = 0.0
>>> res@gsnAddCyclic = False ; regional data
>>>
>>> res@lbLabelBarOn = False
>>>
>>> res@tmXBOn = False ; don't draw axis nor axis labels
>>> res@tmXTOn = False
>>> res@tmYLOn = False
>>> res@tmYROn = False
>>>
>>> res@cnLevelSelectionMode = "ManualLevels"
>>> res@cnMinLevelValF = -15.
>>> res@cnMaxLevelValF = 35.
>>> res@cnLevelSpacingF = 1.
>>> res@cnFillOn = True ; turn on color for contours
>>> res@cnLinesOn = False ; turn off contour lines
>>> res@cnLineLabelsOn = False ; turn off contour line labels
>>> res@cnFillColors = ispan(2,102,2)
>>> res@cnInfoLabelOn = False
>>> ; res@cnFillMode = "RasterFill"
>>>
>>> ; zoom in on map
>>>
>>> res@mpDataBaseVersion = "RANGS_GSHHS" ; high resolution coastline
>>> res@mpDataResolution = "FinestResolution"
>>> res@mpFillOn = False
>>> res@mpOutlineOn = False
>>> res@mpPerimOn = False
>>>
>>>
>>> res@mpProjection = "mercator"
>>> res@mpLimitMode = "Corners"
>>> res@mpLeftCornerLatF = i_lat_min
>>> res@mpLeftCornerLonF = i_lon_min
>>> res@mpRightCornerLatF = i_lat_max
>>> res@mpRightCornerLonF = i_lon_max
>>>
>>>
>>> plot=gsn_csm_contour_map(wks,t,res)
>>>
>>> end
>>>
>>>
>>> <1_tmp2m_2010060709.png>_______________________________________________
>>> 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
>>
>
>
>
>

-- 
Caroline Serraud - Ingénieur Etudes & Développements
------------------------------------------------------
METEO CONSULT / La Chaine Météo - Groupe Figaro
Département Informatique
Domaine de Marsinval F-78540 Vernouillet, FRANCE
Tél : 01 39 28 1990 - Fax : 01 39 71 85 31
e-mail : cs@meteoconsult.fr
------------------------------------------------------
Toute la météo sur le Web : http://www.meteoconsult.fr
Toute la météo par téléphone : 3201 
Received on Thu Jun 10 10:02:50 2010

This archive was generated by hypermail 2.1.8 : Fri Jun 11 2010 - 09:33:24 MDT