Contour Accuracy

[BrianKneller]

There is a  tradeoff between raster grid size and the accuracy / reliability of the contours from a given range contour intervals especially when the contour interval is less than the raster grid spacing. The DEM raster I am using is already quite highly interpolated and I have concerns that some of the contours I am seeing are eroneous or even worse artifacts. Is there any information on the ESRI method of contour caculation or better still a sensitivity analysis/reliability review  of the contours produced?

[DanPatterson_Retired]

I am moving this to the Spatial Analyst  Place since it deals with raster interpolation and its extension to contouring.

To answer your question...No...your contours are only as good as the interpolator...if the interpolator is bad, so will be the contouring.  if you wish to maintain convey a modicum of faith in the results, perhaps, you shouldn't contour or interpolate in the first place since the 'image' will be disjoint from the text which expands upon the value of the results.

[BrianKneller]

Thanks Dan, I wasn't aware of the Spatial Analyst place.  I appreciate your point made but I'm using this toolset on interplanetary data and I am try to glean as much information as I can from the satellite images. A measure of the uncertainties of a contour form and height would be useful. In addition  I endeavour to cross check with infrared or coloured images.