[GRASS5] v.surf.rst - serious errors in DEM

Maciek Sieczka werchowyna at epf.pl
Thu Aug 25 15:55:25 EDT 2005


Helena, All,

I have run v.surf.rst on a dataset which consists of sparse elevation 
contour lines and single elevation points (for hill tops) and several spots 
of very dense elevation points, indicating elevation faults.

On

http://www.biol.uni.wroc.pl/sieczka/udostepnione/vsurfrst.png (160 KB)
and it's 10x exagaration 
http://www.biol.uni.wroc.pl/sieczka/udostepnione/vsurfrst_exag.png (237 KB)

you can see that there is a strange artifact in the interpolated DEM. I mean 
these series of pits between the isoline on the right and the elevation 
points indicating faults to it's left. How come? The elevation of the 
isoline is slightly (decimeters) more than the elevation of the highest 
points in the fault, so I expected a smooth elevation gradient from the 
isoline towards the fault - but there is a ditch instead!

DEM resolution 1m, distance between the isoline and fault about 10-30 
meters.

Such sudden ditches occur everywhere where faults contact the continous 
surface. No matter how far is the closest isoline. Why is v.surf.rst 
introducing such an error?

Maciek 




More information about the grass-dev mailing list