MATLAB: Odd triangles using Matlab’s contour and contourf functions

contourcountorfdelaunay triangulation

Hi,
I make contour maps using MATLAB's contourf function and I've always experienced these odd triangular regions. On the edges of the map, some of the triangles, but not all, appear solid colored when they should clearly have isochrones running through them.
See the map I've made here http://i.imgur.com/ES4Tq.jpg (The solid triangles appear in the top right along the edges of the data)
(Ignore the red vs. black circles. They are the locations of the data points; the red corresponds to points that were interpolated but this has no affect the contouring.)
The above contour map is called using this line of code:
[C,h,cf] =contourf(gca,XI, YI, flipud(plotMatrix), MIN_TIME:dTcont:MAX_TIME);
and those values are as follows:
plotMatrix =
    Columns 1 through 10
         NaN  446.7667  447.7333  448.4000  448.6000       NaN       NaN       NaN  446.8333  446.5333
    448.2667  451.7667  451.3333  450.3667  449.9000  446.5465       NaN  442.8725  442.5158  440.3006
    445.9226  448.1227  444.6667  443.8333  443.3333  442.3333  440.8547  439.4667  440.1333  438.0667
    440.9667  444.8000  443.7000  442.8000  442.2667  441.3910  440.5554  440.5333  440.7667  440.3667
         NaN  441.2437  442.4314  441.7667  441.6333  441.8000  441.3000  441.7000  441.7333  441.7000
         NaN  437.7451  440.9333  440.4333  440.6333  440.8667  441.9000  442.4000  442.8000  443.0333
         NaN  434.4667  439.5667  439.4000  440.4667  440.7667  441.5000  442.0333  442.7333  443.2667
    435.1333  435.9333  437.1000  438.7000  440.3333  440.8000  441.3667  442.0000  442.6219  443.2333
    436.1667  437.0000  437.6333  440.0667  440.3333  440.8000  441.5000  442.1000  442.8000  443.6333
    438.5667  437.9667  438.7000  439.2667  439.9333  440.7333  441.6000  442.4667  443.1333  443.6667
    438.3648  438.8000  439.3333  440.5667  440.6333  441.6333  442.0667  442.5000  443.0333  443.7667
    437.9188  440.5000  441.0667  441.2667  441.3000  441.9667  442.4667  443.2667  443.7000  444.5667
         NaN  440.6333  441.9681  441.8333  442.1667  442.4333  443.5167  444.3874  444.9260  445.4945
         NaN  440.0667  441.8333  442.4333  442.3667  443.8333  444.6212  444.8409       NaN       NaN
         NaN       NaN       NaN  441.5975  443.8667  444.3333       NaN       NaN       NaN       NaN
         NaN       NaN       NaN       NaN       NaN       NaN       NaN       NaN       NaN       NaN
    Columns 11 through 12
    445.8667       NaN
    436.5667  436.9699
    437.2000  437.5000
    439.1333  438.7667
    441.2224  440.5667
    442.8000  442.4333
    443.8004  444.0667
    443.8667  444.3333
    444.1667  444.8000
    444.2667  444.9000
    444.5333  445.1000
    445.2000  446.1000
    446.0412  445.4333
         NaN       NaN
         NaN       NaN
         NaN       NaN
XI =
       0     7    14    21    28    35    42    49    56    63    70    77
       0     7    14    21    28    35    42    49    56    63    70    77
       0     7    14    21    28    35    42    49    56    63    70    77
       0     7    14    21    28    35    42    49    56    63    70    77
       0     7    14    21    28    35    42    49    56    63    70    77
       0     7    14    21    28    35    42    49    56    63    70    77
       0     7    14    21    28    35    42    49    56    63    70    77
       0     7    14    21    28    35    42    49    56    63    70    77
       0     7    14    21    28    35    42    49    56    63    70    77
       0     7    14    21    28    35    42    49    56    63    70    77
       0     7    14    21    28    35    42    49    56    63    70    77
       0     7    14    21    28    35    42    49    56    63    70    77
       0     7    14    21    28    35    42    49    56    63    70    77
       0     7    14    21    28    35    42    49    56    63    70    77
       0     7    14    21    28    35    42    49    56    63    70    77
       0     7    14    21    28    35    42    49    56    63    70    77
YI =
       0     0     0     0     0     0     0     0     0     0     0     0
       7     7     7     7     7     7     7     7     7     7     7     7
      14    14    14    14    14    14    14    14    14    14    14    14
      21    21    21    21    21    21    21    21    21    21    21    21
      28    28    28    28    28    28    28    28    28    28    28    28
      35    35    35    35    35    35    35    35    35    35    35    35
      42    42    42    42    42    42    42    42    42    42    42    42
      49    49    49    49    49    49    49    49    49    49    49    49
      56    56    56    56    56    56    56    56    56    56    56    56
      63    63    63    63    63    63    63    63    63    63    63    63
      70    70    70    70    70    70    70    70    70    70    70    70
      77    77    77    77    77    77    77    77    77    77    77    77
      84    84    84    84    84    84    84    84    84    84    84    84
      91    91    91    91    91    91    91    91    91    91    91    91
      98    98    98    98    98    98    98    98    98    98    98    98
     105   105   105   105   105   105   105   105   105   105   105   105
MIN_TIME =
    434.4667
dTcont =
       1
MAX_TIME =
    451.7667
I'd like to know if these triangles are the result of contour's use of Delaunay triangulation. I've tried to look through the contourf code but I get lost.
Also, any thoughts on how to get rid of these triangles so the maps look better?
Thanks!
Emily

Best Answer

I think your main problem is that there aren't many data, so it is hard to make nice contours out of it. Look at the bit-mapped image:
xi = 0:7:77; yi = 0:7:105;
imagesc(xi,yi,plotMatrix)
colorbar
It's pretty coarse. You can't justify using 16 contours with such data. This plot with a smaller number of contours looks better:
[XI,YI] = meshgrid(xi,yi);
MIN_TIME = 434.4667;
dTcont = 1;
MAX_TIME = 451.7667;
figure
[C,h,cf] =contourf(gca,XI, YI, flipud(plotMatrix),8);
colorbar
Related Question