Re: Wavy contour lines in contour plot

Hi Ekin,
As the cnSmoothing* resources didn't work, I would highly recommend
trying smth9_Wrap. smth9_Wrap allows you to control the level of
smoothing, light->heavy, you can apply the function more than once, and
it is easy to use. See examples 1 and 2 on the documentation page. Adam

On 02/15/2013 06:35 AM, Ekin Akoglu wrote:
> Dear Adam,
>
> Thank you for your suggestions. I have tried using "cnSmoothing"
> resources you have suggested but it did not give satisfactory
> improvements. For a while, Mary and I exchanged a couple of
> off-the-record mails on the issue and it seems that the issue boils
> down to something related to the method used in regridding the data
> before contouring. I will either try your second suggestion on using
> smth9_Wrap function or try to plot the regridded data by MATLAB in NC
> to justify whether it is a problem due to the regridding method used.
> Actually, as you may remember, the MATLAB contour plot had smooth and
> curvy contour lines whereas NCL did not. The problem with MATLAB was
> it did not produce publication quality contour plot no matter how many
> ways I tried. That's why I am trying to plot the data in NCL. Again
> thank you for your suggestions.
>
> Best regards,
>
> Ekin
>
> On Tue, Feb 12, 2013 at 10:25 PM, Adam Phillips <asphilli@ucar.edu> wrote:
>> Hi Ekin,
>> If you are asking about how to make the contours smoother, then you have
>> two options:
>>
>> 1) Let NCL (semi-) automatically smooth the data:
>> Set cnSmoothingOn = True
>> http://www.ncl.ucar.edu/Document/Graphics/Resources/cn.shtml#cnSmoothingOn
>> You may also need to set cnSmoothingTensionF and/or cnSmoothingDistanceF
>> as is stated in the link above.
>>
>> 2) Smooth the data yourself:
>> Before passing the data to gsn_csm_contour, you can smooth it yourself.
>> I would recommend smth9_Wrap:
>> http://www.ncl.ucar.edu/Document/Functions/Contributed/smth9_Wrap.shtml
>>
>> Good luck,
>> Adam
>>
>>
>> On 02/12/2013 01:16 PM, Mary Haley wrote:
>>> Dear Ekin,
>>>
>>> I'm not sure what you mean by "curvy" lines versus "shaky" lines. Do you have
>>> a plot sample from what you think the contours should look like?
>>>
>>> Are your coordinate axes correct? Do the data values in the contour
>>> lines look correct?
>>>
>>> It would help if you could provide your data and your script.
>>>
>>> If you can provide your script and data, then see below for ftp information:
>>>
>>> http://www.ncl.ucar.edu/report_bug.shtml
>>>
>>> --Mary
>>>
>>> On Feb 12, 2013, at 1:00 AM, Ekin Akoglu wrote:
>>>
>>>> Dear all,
>>>>
>>>> I am plotting a contour graph from data and the contour lines which
>>>> are drawn by NCL 6.0.0 looks wavy. I attached the plot to this e-mail.
>>>> Do you have an idea about how to have curvy contour lines instead of
>>>> these shaky drawn lines?
>>>>
>>>> Regards,
>>>>
>>>> Ekin Akoglu
>>>>
>>>> Institute of Marine Sciences
>>>> Middle East Technical University
>>>> P.O. Box 28, 33731
>>>> Erdemli, Mersin
>>>> Turkey
>>>> <pH_ncl-talk.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
>> --
>> ______________________________________________________________
>> Adam Phillips asphilli@ucar.edu
>> NCAR/Climate and Global Dynamics Division (303) 497-1726
>> P.O. Box 3000
>> Boulder, CO 80307-3000 http://www.cgd.ucar.edu/cas/asphilli
>>
>> _______________________________________________
>> ncl-talk mailing list
>> List instructions, subscriber options, unsubscribe:
>> http://mailman.ucar.edu/mailman/listinfo/ncl-talk
>
>

-- 
______________________________________________________________
Adam Phillips                                asphilli@ucar.edu
NCAR/Climate and Global Dynamics Division       (303) 497-1726
P.O. Box 3000				
Boulder, CO 80307-3000    http://www.cgd.ucar.edu/cas/asphilli
_______________________________________________
ncl-talk mailing list
List instructions, subscriber options, unsubscribe:
http://mailman.ucar.edu/mailman/listinfo/ncl-talk