Misleading Masking

When using the NCL mask command to plot a parameter at locations where
parameter M is greater than zero, I find that the displayed area is
smaller than it should be and can even completely omit significant areas
over which M is greater than zero. For example, for a square of 2x2
grid points with positive M values (surrounded by negative values) the
area displayed is only that of a single grid point. Moreover, a region
having the following pattern of grid points with positive values:

       X
       X
      XX
     XX

was completely missed, i.e. nothing at all was displayed for a region
of 6 positive values!

For me this behavior is not useful, since I don't want to miss any
areas with positive values. Is there some way to produce more accurate
masking? The only way I can think of is to not use the mask function
but mimic it manually, i.e. first plot the contours then do a raster
overplot with opaque values where M is negative.

            Jack

_______________________________________________
ncarg-talk mailing list
ncarg-talk_at_ucar.edu
http://mailman.ucar.edu/mailman/listinfo/ncarg-talk
Received on Thu Jun 29 2006 - 22:04:37 MDT