Hi Harvey,

This is concerning incident # 126.4894 about your questionon TOPOGRID.

There is a known problem (SDBsn57119) with calculating slopes from surfaces created from TOPOGRID. The same problem, bias along the contours, originally exists in ANUDEM program developed by Michael Hutchinson.

Cells falling on contours have a greater weight and cause a slight flattening in the resulting surface at the contour.

This can be easily seen by turning the grid into an INT and creating a histogram of the cell values. You will see that where the contour values fall, there will be spikes in the histogram. This looks much worse than it really is, since this has little affect on the overall drainage characteristics of the DEM. The problem is also documented on pg. 19 of TOPOGRID.

"Contour biasing
There is a slight biasing in the interpolation algorithm that causes input contours to have a stronger affect on the output surface at the contour. This bias can result in a slight `flattening' of the output surface as it crosses the contour. This may result in misleading results when calculating the profile curvature of the output surface."

Arti Patel
apatel@esri.com

Arti, Thanks for your response to my feature report. The contention that the problem "looks much worse than it really is" is not a great comfort to those of us who are very concerned with, nor impressive to those who were getting good profiles from in-house software 14 years ago. Is there an elusive "perfect DEM" that nobody has yet acheived, or should there be different DEMs for different applications? harvey
Hi Harvey,

The 'contour-baising' affect that you are noticing in TOPOGRID is already reported as a bug. TOPOGRID is written to generate a hydrologically correct DEM but obviously has this known limitation currently. You can check out USGS dems also that come in 3 different quality levels ( 1, 2 and 3). There are also other softwares that specialize in creating DEMS for specific applications (check out GIS WORLD & GEO-INFO SYSTEMS magazines).

Arti Patel