[ncl-talk] Problem: Stripes in contour maps

[David Brown]

The problem with the CellFill plot is that it does some special
temporary resetting of the map limits in order to draw the cells that
intersect with the edges of the plot, but this resetting fails to take
the possibility of "FreeAspect" into account. So you get unfilled
edges along with some cells drawn where the edges would be if it
weren't FreeAspect. Very messy looking.
This should be pretty straightforward to fix I hope.
 -dave

On Fri, Nov 7, 2014 at 7:14 PM, Rick Brownrigg <brownrig at ucar.edu> wrote:
> Hi Werner,
>
> I can certainly reproduce the results. In fact, I get 3 different results
> using CellFill vs. RasterFill vs. AreaFill with that
> res at mpShapeMode="FreeAspect"  resource set.  (CellFill and RasterFill are
> consistent with each other if that mapShapeMode is left at its default, but
> AreaFill still looks dramatically different).
>
> Unfortunately, I do not have a solution or work-around for you. I will file
> a bug report on this.  Is this a show-stopper for what you are trying to do?
>
> Rick
>
>
> On Fri, Nov 7, 2014 at 9:23 AM, Werner Schneider
> <werner.schneider at uni-bonn.de> wrote:
>>
>> Dear NCL team,
>>
>> we have encountered a NCL problem when drawing contour
>> maps. If we use non rectangular grids and the data field
>> covers an area larger than the map limits, clipped data
>> (directly from the map limits) seems to be mirrored into
>> the inner part of the map. This only happens if we use the
>> following resources combinations:
>>
>>
>> res at mpShapeMode       = "FreeAspect"
>>
>> res at cnFillMode            = "CellFill"
>>
>> The specific result depends on the choice of the viewport.
>> In any case no error message is generated.
>>
>> We use NCL version 6.2.0 with Linux 3.11.6-4-desktop
>> x86_64 x86_64 x86_64 GNU/Linux.
>> Please find attached an NCL script demonstrating the
>> problem with simple artificial data and a postscript file
>> with the plot result.
>>
>> Is there any problem in our script (perhaps forgotten
>> additional resources) or maybe a NCL problem?
>>
>> Regards,
>> Werner Schneider
>>
>>
>>
>> ------------------------------------------
>> Werner Schneider
>> Meteorological Institute University Bonn
>> Auf dem Huegel 20
>> 53121 Bonn
>> Germany
>> Phone: +49 (0)228 73 5186
>> E-Mail: werner.schneider at uni-bonn.de
>> ------------------------------------------
>>
>> _______________________________________________
>> 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
>