Planform Curvature values most too uniform

2333
3
09-19-2010 11:20 AM
StanaforthHopkins
New Contributor II
Planform curvatures of forest harvest areas are said to partially predict debris slide hazards in forest soils. See attached.

Figures from use of the ESRI toolbox curvature feature show almost the entire forest to have convex planform curves. This does not seem credible to me given the contours, elevations and rugged mountainous areas in the state forest. See attached.

Does anyone know enough about planform curvatures to "validate" my results? Where is a decent explanation and diagrams on the subject?

Thanks,

Stan Hopkins
0 Kudos
3 Replies
StanaforthHopkins
New Contributor II
Disappointed that I haven't heard back from any of the contributors, I poked around until I found a GIS that would give me a more varied view of concave, planar, and convex planforms from an elevation file.

I found Landserf and created a planform of one of their DEM tutorial examples - first in the .bil format and then in the .raw format .  Both created a rich and believable mixture of real (not integer) planform curvature figures.  They both could be added to an ESRI map, but both were rounded to positive integer values (why?).  In one case, the Landserf values were -10 to 10.  In the other -256 to 256 (why?).  They became ESRI values of 0 to 10 and 0 to 256 respectively (why?).

Is planform really that mysterious?

Stan
0 Kudos
curtvprice
MVP Esteemed Contributor

No, in my opinion anyway curvature calculations are not very straightforward as one would think, as there are discretization and numerical effects that can really mess with the numbers.  As I'm sure Dan Patterson will soon point out, you can always write your own focal functions using bumpy arrays if you have a particular flavor of calculation you want to do.

I think it's great you found some other software that seems to do what you want. As for moving the results back to ArcGIS, raster data conversions can be tricky if the header file for the .bil is not exactly right for ArcGIS to properly interpret the data. I suggest messing with the .hdr file options you have been using, consulting this help doc:

BIL, BIP, and BSQ raster files—Help | ArcGIS for Desktop 

0 Kudos
curtvprice
MVP Esteemed Contributor

Sorry to resurrect this thread, I see now it's an old one. More discussion from similar old history here:

https://community.esri.com/message/403827?sr=search&searchId=7ac9d75c-3db0-4b32-8afc-4c75c6e94770&se...

0 Kudos