[ncl-talk] Fwd: Pennants wind barbs thickness

Thu Jul 24 13:41:42 MDT 2014

I would suggest making such an effect in a program like Adobe Illustrator
or similar - it would honestly be an order of magnitude faster to do, or
even to learn if you are unfamiliar with such software. In reality though,
it would probably be easier for you to simply choose another color to plot
in if you are worried about visibility.

The implementation of such a procedure from your end would likely take
several days to accurately script and be terribly inefficient, as the
native NCL code is written in a more efficient language. It would also
still be lacking in many of the basic features that already exist. If you
really want to get started on such a project, I strongly suggest using the
NCL documentation as much as possible (including the search) and looking
over the polygon/polyline example page at
http://ncl.ucar.edu/Applications/polyg.shtml
I cannot possibly guide you beyond this as this is not something I would
recommend anyone try without being very familiar with the intricacies of
NCL or, preferably, the underlying code. That's why it's wonderful to have
such responsive developers helping us out and taking requests such as these
into consideration - even if these requests take a bit of time to fully
implement.

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

On Thu, Jul 24, 2014 at 2:29 PM, Rabah Hachelaf <hachelaf at sca.uqam.ca>
wrote:

> Hi Kyle,
>
> My goal is overlay a green windbarb over a black ones with a higher
> thickness to get a green barb with a black borders like image attached.
> I success to do that but for winds speed over 50KT the problem about
> pennant occur.
>
> Please can you tell me what is steps to do to create a filled polygone and
> rotated it to the angle of each wind barb line.
>
> Thanks.
>
>
>  2014-07-24 15:13 GMT-04:00 Kyle Griffin <ksgriffin2 at wisc.edu>:
>
> As Dave is one of the developers, he does usually know best. However, if
>> it is very important to you to change it and you have a reasonable number
>> on your plot, I would recommend outputting your file as a PS, EPS, or PDF
>> and taking the vector file into a post-processing program such as Adobe
>> Illustrator or similar. There, you can edit the length/width/color/position
>> of the lines as necessary. Beyond that, I doubt this feature is simple to
>> address. If you were going to set about fixing this as a user, you could
>> start writing the code to draw each line individually relative to a base
>> point and use the gsn_add_polyline(...) function or similar to try and
>> construct these on your own. For the pennants, you would likely need
>> gsn_add_polygon(...) to create a filled polygon, once again rotated to the
>> angle of each wind barb line. Of course, this would be duplicating much of
>> the code already written behind-the-scenes in NCL, but it IS possible if
>> you really need a customized implementation in the immediate future.
>>
>>
>> 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
>>
>>
>> On Thu, Jul 24, 2014 at 1:50 PM, Rabah Hachelaf <hachelaf at sca.uqam.ca>
>> wrote:
>>
>>> Hi David,
>>>
>>> Yes i re-post my question maybe some one from NCL users has an other
>>> idea.
>>>
>>> Rabah
>>>
>>>
>>> 2014-07-24 14:11 GMT-04:00 David Brown <dbrown at ucar.edu>:
>>>
>>> Hi Rabah,
>>>> As you know I previously answered this question for you personally
>>>> prior to your joining ncl-talk. For the record I am now answering on
>>>> ncl-talk:
>>>>
>>>> You have diagnosed the problem correctly, and unfortunately there is no
>>>> resource to directly control the size of the pennant part of the wind barb
>>>> glyphs. The size and area of the pennants is fully determined by
>>>> the vcWindBarbTickLengthF and the vcWindBarbTickAngleF. Unfortunately there
>>>> is currently no adjustment made to account for the width of the lines,
>>>> which also has the effect of making the barbs longer than their nominal
>>>> length. No one has ever asked for this kind of flexibility in the past, but
>>>> I can see that it could be useful.
>>>>
>>>> You could try playing with the tick length and tick angle resources.
>>>> They may help give you a combination that looks better. But until we can
>>>> add another option, my suggestion would be to just try to keep the width as
>>>> small as feasible for your purposes.
>>>>  -dave
>>>>
>>>>
>>>>
>>>> On Tue, Jul 22, 2014 at 4:54 PM, Rabah Hachelaf <hachelaf at sca.uqam.ca>
>>>> wrote:
>>>>
>>>>> Hi NCL users,
>>>>>
>>>>> I have problem with plotting wind barb with a relatively big thickness
>>>>>
>>>>> I am using vcWindBarbLineThicknessF command to increase wind barb
>>>>> thickness
>>>>> it's correct for all wind barbs only for higher than 50KT ones where
>>>>> the pennants or small triangle aren't affected by the thickness set.
>>>>> When reading some documentation i noted that
>>>>>
>>>>>   " The pennants are drawn using a filled polygon, while the ticks,
>>>>> the shaft, and the calm circle are all rendered with polylines."
>>>>>
>>>>> So i think this is why pennants aren't affected by the thickness
>>>>> change.
>>>>>
>>>>> Have you some tricks to solve this problem and get the same pennants
>>>>> thickness like the ticks.
>>>>>
>>>>> Attached you can find an example of my case, where wind speed is
>>>>> supposed to be about between 55KT and 60KT
>>>>>
>>>>> Thank you.
>>>>>
>>>>> ------------------------------
>>>>> Cordialement,
>>>>> Best regards,
>>>>> Rabah Hachelaf
>>>>> Msc Studient UQAM
>>>>> Montréal/CANADA
>>>>>      ____
>>>>>     (       )
>>>>>    (        )
>>>>>   (___ __)
>>>>>    /////////
>>>>>
>>>>> _______________________________________________
>>>>> ncl-talk mailing list
>>>>> List instructions, subscriber options, unsubscribe:
>>>>> http://mailman.ucar.edu/mailman/listinfo/ncl-talk
>>>>>
>>>>>
>>>>
>>>
>>>
>>> --
>>> ------------------------------
>>> Cordialement,
>>> Best regards,
>>> Rabah Hachelaf
>>>
>>> _______________________________________________
>>> ncl-talk mailing list
>>> List instructions, subscriber options, unsubscribe:
>>> http://mailman.ucar.edu/mailman/listinfo/ncl-talk
>>>
>>>
>>
>
>
> --
> ------------------------------
> Cordialement,
> Best regards,
> Rabah Hachelaf
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.ucar.edu/pipermail/ncl-talk/attachments/20140724/9104a813/attachment.html 