[ncl-talk] Panel Plot Bottom Row Narrower than the rest

Jatin Kala jatin.kala.jk at gmail.com
Mon Feb 9 23:31:43 MST 2015


Hi,
Don't have the answer, but i think you should definitely turn Off the 
Y-Axis label for the "right-most" plots, i.e, "Latitude (degrees)", and 
also turn the tick-labels off (-90 to 90).
Only need them on the left-most column of plots.
You could also make use of gsn_text_ndc to add text to the panel plot, 
rather than specifying a res at tiXAxisString, but this is Not an ideal 
solution.
Hope this helps.
And i suggest you also use manual levels for you contour plots (0 to 35 
for example), and have only one label-bar at the bottom for the whole 
panel, might make the plot easier to read.
Hope this helps,
Cheers,
Jatin.



On 10/02/15 5:03 PM, Parish, Helen wrote:
> Unfortunately setting resP at gsnPanelScalePlotIndex to 6 or 7 does not 
> seem to work in this case. The result is that the earlier plots in the 
> second column have their y-axis labels cut off, but the last 2 plots 
> still remain the same size. In this particular case, the last 2 plots 
> do not differ in the size of their y-axis text, as in example 
> panel_24.ncl, but in the fact that they actually have x-axis text and 
> labels whilst the earlier plots do not have any x-axis text or labels 
> at all. Perhaps this makes a difference ?.
>
> I am attaching the new plot generated and the modified script.
>
> Any ideas what to do here ?.
>
> Thanks,
> Helen.
>
>
>
>
>
> On Feb 9, 2015, at 8:14 PM, Mary Haley wrote:
>
>> Helen,
>>
>> The issue here appears to be with paneling plots that are different 
>> sizes.  The plots on the bottom are larger than the other plots, 
>> because they contain labels on the bottom axis.
>>
>> If you have plots that are larger than the top leftmost plot, then 
>> you need to tell the gsn_panel routine to base the scaling of each 
>> plot on the plot that's the largest (the default is to base it on the 
>> first plot).
>>
>> Do by by setting resP at gsnPanelScalePlotIndex to 6 or 7 to indicate 
>> that the scaling of each plot should be based on plot # 6 or #7, and 
>> not plot #0.
>>
>> For an example of this, see panel_24.ncl at:
>>
>> http://www.ncl.ucar.edu/Applications/panel.shtml#ex24
>>
>> As a side, your main title is quite large.  You may want to resize it 
>> by setting resP at txFontHeightF to something smaller, like 0.03 or smaller.
>>
>> --Mary
>>
>>
>> On Mon, Feb 9, 2015 at 7:51 PM, Parish, Helen <hparish at epss.ucla.edu 
>> <mailto:hparish at epss.ucla.edu>> wrote:
>>
>>     Does anyone know how to prevent the bottom row of a panel plot
>>     from being narrower than the other rows ?. In this kind of panel
>>     plot only the bottom row has axis labels and x-axis titles.
>>     However, I do not see why this should make the bottom row
>>     narrower. Does anyone know how to correct this ?.
>>
>>     I enclose a copy of the NCL script and the image it produces.
>>
>>     Thanks,
>>     Helen.
>>
>>
>>
>>
>>     _______________________________________________
>>     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/20150210/06725176/attachment-0001.html 
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: image/jpg
Size: 155017 bytes
Desc: not available
Url : http://mailman.ucar.edu/pipermail/ncl-talk/attachments/20150210/06725176/attachment-0002.jpg 
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: image/jpg
Size: 177598 bytes
Desc: not available
Url : http://mailman.ucar.edu/pipermail/ncl-talk/attachments/20150210/06725176/attachment-0003.jpg 


More information about the ncl-talk mailing list