[ncl-talk] panel plot, each with their own colormap

Soma Roy somaroy892 at gmail.com
Thu Aug 9 05:39:15 MDT 2018 

Can you please share your final script for the panel plot settings?

Thanking you,
Soma

On Wed, Aug 8, 2018, 21:25 Soma Roy <somaroy892 at gmail.com> wrote:

> Can you share your script?
>
> Thanks,
> Soma
>
> On Wed, Aug 8, 2018, 21:20 Laura Fowler <laura at ucar.edu> wrote:
>
>> Hi Rashed and Rick:
>>
>> Thanks to Rashed's suggestion, I was finally able to make it work. The
>> last issue I had was that I was not giving the correct cnMinValueF and
>> cnMaxValueF for plot(3). This was the final reason why the very bottom
>> label bar was not correct. I am attaching the final png file (see
>> test2.png).
>>
>> I really would like to thank you for your help and interest for the last
>> couple of days. You have been so helpful.
>>
>> Many thanks,
>> Laura
>>
>>
>>
>>
>>
>> On Tue, Aug 7, 2018 at 1:23 PM, Rashed Mahmood <rashidcomsis at gmail.com>
>> wrote:
>>
>>> Hi Laura,
>>> Could you please share full script or the part that has ALL resources?
>>> It seems that all plots are being created with same contour levels?
>>>
>>> Cheers,
>>> Rashed
>>>
>>> On Tue, Aug 7, 2018 at 12:15 PM, Laura Fowler <laura at ucar.edu> wrote:
>>>
>>>> Hi Rashed:
>>>>
>>>> Many thanks for your suggestion. I really appreciate you taking an
>>>> interest in my very pesky problem! Following your idea helped removed the
>>>> "big" plot in the second row of the panel, but it now shifted the label bar
>>>> for the second row to the third row. If you compare test.png and
>>>> totPREC.dec2015.png, then you will see that the bottom row is the same as
>>>> that for the second row, which is not what I expected (see
>>>> totPREC.dec2015.png where the panels for the top row are the same and not
>>>> centered.)
>>>>
>>>> Great idea though. At this time, I'd tried anything.Thanks.
>>>> Laura
>>>>
>>>>
>>>>
>>>>
>>>>
>>>> On Tue, Aug 7, 2018 at 10:59 AM, Rashed Mahmood <rashidcomsis at gmail.com
>>>> > wrote:
>>>>
>>>>> Hi Laura,
>>>>> I think the problem lies with this line:
>>>>>  pplots = gsn_panel_return(wks,plots(0:1),(/1,2/),resP)
>>>>>
>>>>> Changing the above line with the following should solve your problem:
>>>>>  pplots = gsn_panel_return(wks,plots(1:2),(/1,2/),resP)
>>>>>
>>>>>
>>>>> hope that helps!
>>>>>
>>>>> Cheers,
>>>>> Rashed
>>>>>
>>>>>
>>>>>
>>>>>
>>>>>
>>>>> On Mon, Aug 6, 2018 at 12:17 PM, Laura Fowler <laura at ucar.edu> wrote:
>>>>>
>>>>>> Hi Rich:
>>>>>>
>>>>>> Thanks for your suggestion. I did think of using gsnPanelRowSpec,
>>>>>> but this won't work since I am using the same color bar for the first and
>>>>>> second row but a different color bar for the third row instead of one color
>>>>>> bar for all the rows.That is the reason I decided to follow what is done in
>>>>>> panel_26.ncl.
>>>>>>
>>>>>> It looks like there is some kind of resizing in the second call to
>>>>>> gsn_panel but I have no idea how this happens either. Not sure what to to
>>>>>> try next.
>>>>>>
>>>>>> Thanks,
>>>>>> Laura
>>>>>>
>>>>>>
>>>>>>
>>>>>>
>>>>>> On Mon, Aug 6, 2018 at 10:37 AM, Rick Brownrigg <brownrig at ucar.edu>
>>>>>> wrote:
>>>>>>
>>>>>>> Hi Laura,
>>>>>>>
>>>>>>> I don't have an answer for why the one plot is so much larger than
>>>>>>> the others.
>>>>>>>
>>>>>>> I wonder however, if you might be able to use the following resource
>>>>>>> to specify a layout of (/1, 2, 2/), and let gsn_panel() do the spacing:
>>>>>>>
>>>>>>> *gsnPanelRowSpec* By default, the *gsn_panel*
>>>>>>> <http://ncl.ucar.edu/Document/Graphics/Interfaces/gsn_panel.shtml>
>>>>>>> routine panels its plots according to the *dims* variable which
>>>>>>> indicates rows x columns. If you set *gsnPanelRowSpec*
>>>>>>> <http://ncl.ucar.edu/Document/Graphics/Resources/gsn.shtml#gsnPanelRowSpec>
>>>>>>> to True, however, then you can use the *dims* variable to indicate
>>>>>>> the number of plots you want per row. For example, setting
>>>>>>> *gsnPanelRowSpec*
>>>>>>> <http://ncl.ucar.edu/Document/Graphics/Resources/gsn.shtml#gsnPanelRowSpec>
>>>>>>> to True and *dims* to (/2,3,1/) will put 2 plots in the first row,
>>>>>>> 3 plots in the second row, and 1 plot in the third row.
>>>>>>>
>>>>>>> Default: False
>>>>>>> I don't knwo for certain whether this will work, but might be worth
>>>>>>> a shot.
>>>>>>>
>>>>>>> Rick
>>>>>>>
>>>>>>>
>>>>>>> On Fri, Aug 3, 2018 at 2:36 PM, Laura Fowler <laura at ucar.edu> wrote:
>>>>>>>
>>>>>>>> Hello:
>>>>>>>>
>>>>>>>> I followed panel_26.ncl to create 3 panels plots below each other,
>>>>>>>> each set with their own colormap (see attached file totPREC.dec2015.png).
>>>>>>>> Now, I would like to do the same thing, but with only one TMPA map centered
>>>>>>>> at the top of the figure (see tentative figure test.png).
>>>>>>>>
>>>>>>>>
>>>>>>>> For both the 6 and 5 panel plots, I used:
>>>>>>>>
>>>>>>>> resP                  = True
>>>>>>>>
>>>>>>>> resP at gsnMaximize      = True
>>>>>>>>
>>>>>>>> resP at gsnDraw          = False
>>>>>>>>
>>>>>>>> resP at gsnFrame         = False
>>>>>>>>
>>>>>>>> resP at gsnPanelLabelBar = True
>>>>>>>>
>>>>>>>>
>>>>>>>> pplots = gsn_panel_return(wks,plots(0:1),(/1,2/),resP)
>>>>>>>>
>>>>>>>> bb     = NhlGetBB(pplots)
>>>>>>>>
>>>>>>>> height = max(bb(0:1,0)) - min(bb(0:1,1))
>>>>>>>>
>>>>>>>> print("--- bb(0)  = "+bb(0:1,0))
>>>>>>>>
>>>>>>>> print("--- bb(1)  = "+bb(0:1,1))
>>>>>>>>
>>>>>>>> print("--- height = "+height)
>>>>>>>>
>>>>>>>>
>>>>>>>> ;--- using height value, now we can panel all sets of plots:
>>>>>>>>
>>>>>>>> resP at gsnDraw        = True
>>>>>>>>
>>>>>>>> resP at gsnMaximize    = True
>>>>>>>>
>>>>>>>> resP at gsnPanelTop    = 1.0
>>>>>>>>
>>>>>>>> resP at gsnPanelBottom = resP at gsnPanelTop - height
>>>>>>>>
>>>>>>>>
>>>>>>>> For the 6 panel plot, I used the loop:
>>>>>>>>
>>>>>>>> do n = 0,nplots-1,2
>>>>>>>>
>>>>>>>>    gsn_panel(wks,plots(n:n+1),(/1,2/),resP)
>>>>>>>>
>>>>>>>>
>>>>>>>> ;---Set for the next panel call.
>>>>>>>>
>>>>>>>>    resP at gsnPanelMainString = ""
>>>>>>>>
>>>>>>>>    resP at gsnPanelTop        = resP at gsnPanelBottom - 0.02
>>>>>>>>
>>>>>>>>    resP at gsnPanelBottom     = resP at gsnPanelTop-height
>>>>>>>>
>>>>>>>> end do
>>>>>>>>
>>>>>>>> frame(wks)
>>>>>>>>
>>>>>>>>
>>>>>>>> wheres for the 5 panel plot, I used the loop:
>>>>>>>>
>>>>>>>> do n = 0,nplots-1,2
>>>>>>>>
>>>>>>>>    print("--- n = "+n)
>>>>>>>>
>>>>>>>>    if(n .eq. 0) then
>>>>>>>>
>>>>>>>>       gsn_panel(wks,plots(0:0),(/1,1/),resP)
>>>>>>>>
>>>>>>>>    else
>>>>>>>>
>>>>>>>>       gsn_panel(wks,plots(n-1:n),(/1,2/),resP)
>>>>>>>>
>>>>>>>>    end if
>>>>>>>>
>>>>>>>>    resP at gsnPanelMainString = ""
>>>>>>>>
>>>>>>>>    resP at gsnPanelTop        = resP at gsnPanelBottom - 0.02
>>>>>>>>
>>>>>>>>    resP at gsnPanelBottom     = resP at gsnPanelTop-height
>>>>>>>>
>>>>>>>> end do
>>>>>>>>
>>>>>>>> frame(wks)
>>>>>>>>
>>>>>>>>
>>>>>>>> I do not understand why the second plot of the second column is so
>>>>>>>> much bigger than the one on the left side and the two plots for the third
>>>>>>>> row? Can somebody help me figure this out?
>>>>>>>>
>>>>>>>>
>>>>>>>> Many thanks,
>>>>>>>> Laura
>>>>>>>>
>>>>>>>>
>>>>>>>> --
>>>>>>>>
>>>>>>>> !-------------------------------------------------------------------------------------------------------------
>>>>>>>> Laura D. Fowler
>>>>>>>>
>>>>>>>> Mesoscale and Microscale Meteorology Division (MMM)
>>>>>>>> National Center for Atmospheric Research
>>>>>>>> P.O. Box 3000, Boulder CO 80307-3000
>>>>>>>>
>>>>>>>> e-mail: laura at ucar.edu
>>>>>>>> phone: 303-497-1628
>>>>>>>>
>>>>>>>>
>>>>>>>> !-------------------------------------------------------------------------------------------------------------
>>>>>>>>
>>>>>>>> _______________________________________________
>>>>>>>> ncl-talk mailing list
>>>>>>>> ncl-talk at ucar.edu
>>>>>>>> List instructions, subscriber options, unsubscribe:
>>>>>>>> http://mailman.ucar.edu/mailman/listinfo/ncl-talk
>>>>>>>>
>>>>>>>>
>>>>>>>
>>>>>>
>>>>>>
>>>>>> --
>>>>>>
>>>>>> !-------------------------------------------------------------------------------------------------------------
>>>>>> Laura D. Fowler
>>>>>>
>>>>>> Mesoscale and Microscale Meteorology Division (MMM)
>>>>>> National Center for Atmospheric Research
>>>>>> P.O. Box 3000, Boulder CO 80307-3000
>>>>>>
>>>>>> e-mail: laura at ucar.edu
>>>>>> phone: 303-497-1628
>>>>>>
>>>>>>
>>>>>> !-------------------------------------------------------------------------------------------------------------
>>>>>>
>>>>>> _______________________________________________
>>>>>> ncl-talk mailing list
>>>>>> ncl-talk at ucar.edu
>>>>>> List instructions, subscriber options, unsubscribe:
>>>>>> http://mailman.ucar.edu/mailman/listinfo/ncl-talk
>>>>>>
>>>>>>
>>>>>
>>>>
>>>>
>>>> --
>>>>
>>>> !-------------------------------------------------------------------------------------------------------------
>>>> Laura D. Fowler
>>>>
>>>> Mesoscale and Microscale Meteorology Division (MMM)
>>>> National Center for Atmospheric Research
>>>> P.O. Box 3000, Boulder CO 80307-3000
>>>>
>>>> e-mail: laura at ucar.edu
>>>> phone: 303-497-1628
>>>>
>>>>
>>>> !-------------------------------------------------------------------------------------------------------------
>>>>
>>>
>>>
>>
>>
>> --
>>
>> !-------------------------------------------------------------------------------------------------------------
>> Laura D. Fowler
>>
>> Mesoscale and Microscale Meteorology Division (MMM)
>> National Center for Atmospheric Research
>> P.O. Box 3000, Boulder CO 80307-3000
>>
>> e-mail: laura at ucar.edu
>> phone: 303-497-1628
>>
>>
>> !-------------------------------------------------------------------------------------------------------------
>> _______________________________________________
>> ncl-talk mailing list
>> ncl-talk at ucar.edu
>> 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/20180809/8e66a169/attachment.html>

More information about the ncl-talk mailing list