Re: FW: ShadeGtContour Question

From: Adam Phillips <asphilli_at_nyahnyahspammersnyahnyah>
Date: Mon Nov 07 2011 - 10:12:25 MST

Hi Cat,
Mary nailed it: The key with the Shade*Contour or gsn_contour_shade
functions is that they operate starting at a specific contour level,
which is not necessarily the same as the specified value that one gives
the functions.
Again using this example:
contour levels -2, -1, 0, 1, 2, 3, and you call ShadeGtContour(plot,1.5,7)

I would suspect that everything greater than 2 would be shaded, with 2
being the first contour level that would meet the specified criteria of
being greater than 1.5...

And Mary's definitely right on this point: Check that the results of
using these functions is what you would expect them to be. Adam

On 11/07/2011 10:05 AM, Mills, Catrin Marie wrote:
> From Mary Haley:
> ------------------------------------------------------------------------
> From: Mary Haley
> Sent: Sunday, November 06, 2011 1:02 PM
> To: Catrin Mills
> Cc: Dennis Shea; Adam Phillips; mmills2@illinois.edu
> Subject: Re: ShadeGtContour Question
>
> [This emai bounced using your mmills2@illinois.edu address, so I'm
> trying your other one.]
>
> Cat,
>
> I think Adam will have to jump in here, because I'm not familiar with
> how ShadeGtContour works. It seems that
> the shading is based on the contour levels, and not the actual data.
> In other words, If you have contour
> levels -2, -1, 0, 1, 2, 3, and you call ShadeGtContour(plot,1.5,7),
> shading will be applied to *any* contour
> values between 1 and 2, because 1.5 falls between contour levels 1 and
> 2.
>
> The best thing to do when you unsure about the behavior of these kind
> of routines, like I am right now,
> is to create a dummy dataset with very simple values, and try
> different contour levels and shading.
>
> See the attached script, which creates a small array of values equal
> to exactly -5.5, -4.5, ..., -1.5, 0.5, 1.5, 2.5, ... 8.5.
>
> I tried "nice" contour levels of -5, -4, ..., 0, 1, ..., 7, 8 and
> you'll see that the shading appears between contour levels 1 and 2,
> even though this area contains only values equal exactly to 1.5.
>
> --Mary
>
>
> On Nov 5, 2011, at 6:27 PM, Dennis Shea wrote:
>
> >
> >
> > -------- Original Message --------
> > Subject: FW: ShadeGtContour Question
> > Date: Sun, 6 Nov 2011 00:26:17 +0000
> > From: Mills, Catrin Marie <cmmills2@illinois.edu>
> > To: Adam Phillips <asphilli@ucar.edu>, Dennis Shea <shea@ucar.edu>
> >
> >
> >
> > Hi, Dennis and Adam. Sorry to send you this question, but UIllinois
> > recently switched email accounts, and now ncl-talk won't allow me to
> > send requests. I'm trying to figure out what's going on, but until
> > then,
> > if you have a spare moment, could you take a look at the question
> > below?
> >
> > Thanks so much,
> >
> > Cat
> >
> > From: Catrin Mills <cmmills2@illinois.edu <mailto:cmmills2@illinois.edu
> > >>
> > Date: Sat, 5 Nov 2011 19:16:54 -0500
> > To: <ncl-talk@ucar.edu <mailto:ncl-talk@ucar.edu>>
> > Subject: ShadeGtContour Question
> >
> > Hi, I hope all is well.
> >
> > If I want to only contour values greater than 1.5, say, when using
> > ShadeGtContour, if I set resi@cnLevelSpacingF = 1.5 (earlier in my
> > code)
> > and use the function ploti = ShadeGtContour(ploti, 1.5, 17), will that
> > ensure that only values greater than 1.5 are shaded? I just want to
> > make
> > sure since the function description states , "Note that the shading
> > begins at the first contour level less than /value/, and not at the
> > specified /value/."
> >
> > Thanks,
> >
> > Catrin
>
>
>
> _______________________________________________
> 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
Received on Mon Nov 7 10:12:30 2011

This archive was generated by hypermail 2.1.8 : Tue Nov 08 2011 - 15:54:50 MST