Hi Todd,
Without seeing your script, I am not sure what your application is. But
one possible problem is
that you may be trying to use the CellFill method for a grid type for
which it is not yet implemented.
If your data is on a non-rectangular grid such as your geodesic grid,
CellFill will not work.
Even though the example sets the 'trGridType' to TriangularMesh, it
does note that it is a bit of a cheat
in this case. It only works because the grid is topologically
rectangular.
Hopefully, it will be possible to get a cell fill method for
non-rectangular grids implemented before too long.
It's definitely doable for grids that are specified via their cell
boundaries, like the geodesic grids.
If, on the other hand, you actually do have a topologically rectangular
grid, it should work. If you want
to send me your script along with some sample data, I'd be happy to
take a look.
-dave
On Oct 19, 2005, at 11:05 AM, Todd Ringler wrote:
>
> From the Mac platform, I am working from the following cell fill
> example:
>
> http://www.ncl.ucar.edu/Applications/Scripts/2dvertcoords_2.ncl
>
> In this example, the data is "irregularly" spaced since pressure
> varies with longitude. To see the data on the native grid and avoid
> interpolation, this example uses CellFill.
>
> I believe I have followed the example correctly. With my data, I can
> get RasterFill to work (but it is jagged) and the data is displayed
> correctly. AreaFill also works. When I use CellFill, I get a solid
> color fill of the entire area that contains any data. It is as if it
> sees the entire area to be plotted as a single cell and fills it
> accordingly. Also, I can not seem to reproduce the coordinate lines as
> shown in the example above.
>
> If anyone has experience with CellFill and knows how to get around
> these issues, please let me know.
>
> Thanks,
> Todd Ringler
>
>
>
> -------------
> Todd Ringler, Ph.D. phone: (970) 491-8339
> Research Scientist III fax: (970) 491-8693
> Department of Atmospheric Sciences
> Colorado State University
> Fort Collins, Colorado 80523-1371
> email: todd@atmos.colostate.edu
> -------------
>
>
> _______________________________________________
> ncl-talk mailing list
> ncl-talk@ucar.edu
> http://mailman.ucar.edu/mailman/listinfo/ncl-talk
_______________________________________________
ncl-talk mailing list
ncl-talk@ucar.edu
http://mailman.ucar.edu/mailman/listinfo/ncl-talk
This archive was generated by hypermail 2b29 : Wed Oct 19 2005 - 18:01:52 MDT