[gradsusr] tile and grfill

[Jennifer M Adams]

Thanks for the clarification, Andrew. I think this is a bug — the grfill code path is not doing what it’s supposed to with a contour level of -1. The ’not drawing anything when the color is -1’ is a feature of the new shading algorithm, and it may not have propagated over to the grfill code. I will look into it. I can offter two workarounds.

1. Define a color that is fully transparent and use that instead of -1.

‘set gxout grfill'
‘set rgb 42 0 0 0 0 ‘
'set ccols 41 42'
‘set clevs 0.05’
‘d pval’

2. Leave the -1 in there, but use ‘d maskout(pval,0.5-pval)’ instead of ‘d pval’.


—Jennifer

On Jan 27, 2016, at 9:05 AM, Andrew Friedman <andfried at gmail.com<mailto:andfried at gmail.com>> wrote:

Hi Jeff,

Here's an example of where I have the discrepancy.
I'm looking at a netcdf file with 2 variables, 'slope' and 'pval' (it's a trend of rainfall). I'm trying to plot the slope using grfill, and to apply the pattern in areas where 'pval' is above a certain value.

First, I define a tile as a new color:
'set tile 0 2 6 6 3 0'
'set rgb 41 tile 0'

Then, I shade the slope:
'set gxout shaded' * (or grfill)
'd slope'

Third, I apply the pattern fill to pval:
'set clevs 0.05'
'set ccols 41 -1'
'd pval'

When I use 'set gxout shaded', the pattern fills on top of the existing shading, as in the GrADS website example. However, when I use 'set gxout grfill', the initial fill disappears in areas without pattern fill. I've attached images showing the differences.

Any ideas? I know it's not a major issue, but I often look at coarse spatial resolution datasets, and I like the grfill setting to present where the gridbox boundaries are exactly.

Thanks,
Andrew


On Tue, Jan 26, 2016 at 6:42 PM, Jeff Duda <jeffduda319 at gmail.com<mailto:jeffduda319 at gmail.com>> wrote:
Andrew,
Can you clarify your question please? What exactly do you mean by "tiling"?

The difference between grfill and shaded is that shaded color shades between (smoothed) contours drawn by the software, giving your fields a nicer smoother look, whereas grfill just uses color shades to fill by grid box only, with no contour lines drawn to serve as boundaries for color areas. For very high resolution grids, you won't really see much, if any, difference between a field displayed using grfill and that field displayed using shaded. It only really shows up for coarser grids.

Jeff Duda

On Tue, Jan 26, 2016 at 10:58 AM, Andrew Friedman <andfried at gmail.com<mailto:andfried at gmail.com>> wrote:
Hi GrADS users,

I am trying to tile areas of significance using grfill. However, I have noticed that tiling seems to behave differently with gxout set to grfill v. shaded. When gxout is set to grfill, all values that are not tiled are return to the background. Does anyone know a work-around for this?

Thanks, Andrew

_______________________________________________
gradsusr mailing list
gradsusr at gradsusr.org<mailto:gradsusr at gradsusr.org>
http://gradsusr.org/mailman/listinfo/gradsusr




--
Jeff Duda
Graduate research assistant
University of Oklahoma School of Meteorology
Center for Analysis and Prediction of Storms

_______________________________________________
gradsusr mailing list
gradsusr at gradsusr.org<mailto:gradsusr at gradsusr.org>
http://gradsusr.org/mailman/listinfo/gradsusr


<shaded_example.png><grfill_example.png>_______________________________________________
gradsusr mailing list
gradsusr at gradsusr.org<mailto:gradsusr at gradsusr.org>
http://gradsusr.org/mailman/listinfo/gradsusr

--
Jennifer M. Adams
Center for Ocean-Land-Atmosphere Studies (COLA)
111 Research Hall, Mail Stop 2B3
George Mason University
4400 University Drive
Fairfax, VA 22030


-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://gradsusr.org/pipermail/gradsusr/attachments/20160127/c48eb584/attachment-0001.html