Problems using path distance tool

Question

Hi all,

I've been trying to use the path distance tool (ArcGIS 10) in order to delineate a 2 hour walking distance around archaeological sites.
I'm trying to model the calculation in a way that will take into account the changing slope in the area.
After consulting the forums and past posts, I'm using the following inputs:
1. a shapefile with the site's location
2. a DEM (floating, 32 bit) of topography, cell size=25m, as a surface raster
3. the same DEM as the vertical factor surface
4. a vertical factor table (created by Nico Tripcevich based on Tobler's hiking function, discussed here http://forums.esri.com/Thread.asp?c=93&f=995&t=138683, among other places).

the results I'm getting do not seem to be reasonable; it seems that the DEM has no effect, as I'm getting a very much circular pattern even when a site is located within a steep valley (in such a case I would expect a more elongated radius, following the valley route).  I've tried running the tool without the vertical factor, and indeed I'm getting a spider-web-like result, with the cost just increasing with distance in straight lines... this again makes me think that for some reason the tool isn't taking the DEM into account.

I've been going almost crazy trying to solve this... any ideas?? I don't know if there's something I'm not doing properly, or if there's a problem with the tool itself...

Thanks
Michal

curtvprice · Answer

I've tried to construct a cost surface based on slope, but I haven't been able to figure out how to do it so it takes into account the direction of travel. I need it to do so because there are different costs to going up and to going down a particular slope degree.Is there any way to calculate slope relative to a starting point other than the path distance tool?

The path distance tool does do this complex analysis I'd maybe go back and check the inputs you are giving the tool in terms of units and such.

I would also give very high-slope areas an additionally high general (nondirectional) cost as travel is difficult on a steep slope no matter what direction you are going.

Units are a huge thing to worry about here because your x y and z units need to be the same (or differences accounted for) and your costs need to be scaled in a way that doesn't overcome the topographic costs pathdistance is modeling.

MichalBirkenfeld · Answer

Hi Curtis, thanks for the advice.
I've tried to construct a cost surface based on slope, but I haven't been able to figure out how to do it so it takes into account the direction of travel. I need it to do so because there are different costs to going up and to going down a particular slope degree.Is there any way to calculate slope relative to a starting point other than the path distance tool?
Thanks again
Michal

curtvprice · Answer

I don't have an answer, but some guidance based on experience:

The place to focus your attention is the cost raster. If you can solve that problem and get cost values that make sense, you will be 95% of the way there.