Path Allocation with a VRMA-VF table

[GregoryBurns]

I've been trying to run the Path Allocation tool with a custom VRMA-VF table to model energy expenditure for travel on foot.  The tool will run if I use the built-in vertical factor functions, but hangs indefinitely (most common in ArcGIS 10) or fails with a meaningless error message (most common in ArcGIS 9.3.1) when I use a custom table.  I assume I have made an error in formatting the table.  The online help (http://help.arcgis.com/en/arcgisdesktop/10.0/help/index.html#//009z000000z9000000.htm) gives an example for building the table, but it is just a copy of the sample table for the horizontal factor.  Does vertical factor also need to run from 0 to 180 degrees?  If so, to what slope do 0, 90, and 180 correspond?

[ErikMarsh]

I have run into the similar problems with custom vertical tables.
The example that lists angles from 0 to 180 might not be the best, in the case of vertical tables.

A great example of custom table that works is ToblerAway.txt, from Tripcevich's excellent tutorial on least cost paths.

The slopes are listed from -90 (downhill) to 90 (uphill). The values are separated by a tab, not a space (as recommended below). Still seems to work. You might try both formats.

The .txt file should use ANSI encoding. By trail and error, I have found that other encodings (e.g., unicode) will result in an error. In this case, ArcMap at least tells you there was a problem with the format of the table, error 010202:

PATHDISTANCE: Format error in the vertical factor parameter file.
Description: The vertical factor parameter file is incorrectly formatted.

The vertical factor file or table is an ASCII file with two columns. The first column identifies the vertical relative moving angle (VRMA) in degrees, and the second, the vertical factor (VF). Each line in the file specifies a point in a vertical factor graph. Two consecutive points produce a line segment in the VRMA-VF coordinate system. The angles must be in input in ascending order.

Solution:

Check the following to ensure the horizontal factor parameter file is in the correct format:
There are two columns.
The values in the first column (VRMA) range from -90 to 90 and are in ascending order.
The values in the columns (VRMA and VF) are separated by a space.
Only one angle and one factor are specified on a line.


I hope this helps! Let us know how it goes. Good luck.
Erik

[NicoTripcevich]

I just tried to use these VF Table values in Arcmap 10.3 and the PathDistance function is currently broken. Consider using the original GRID command line PathDistance or use the new ArcGIS Pro. I just checked and ToblerAway.txt works as expected in ArcGIS Pro 1.0.

best, Nico

[DanielBecker]

Hi, i'm experimenting with this myself.. the whole day again.. i also have crashes with both "versions" of the tobler.txt (with spaces or tabs).
the other thing is that the path distance tool seems to be very picky in regards of the elevation raster (and i couldn't really find anything about that in the documentation)

I have a 10m DEM, i tried .tif or converted it to ESRI GRID (32bit floating point i think), most of the times i get results mentioned in this thread http://forums.arcgis.com/threads/23888-Problem-with-DEM-and-Path-Distance
or it crashes. It only seems to work if i use a very small extent.


That's all quite bad, since i think it would be the perfect tool for my diploma-thesis. 😞

[ErikMarsh]

Hi Daniel -

You're right, Path Distance is picky about the raster - I would follow the advice in the other forum you mentioned. Sometimes when I can't figure what might be wrong, I go through Tripcevich's tutorials again. They have data to download that works for sure (for example, the Vertical Factor table). Other than that I can only recommend trial and error, with low expectations. I try to start new maps with the minimum amount of data, and go carefully step by step to isolate the problem. This also helps in identifying the things that DO work. As for the extent problems (which many have had), I can't offer any specific advice. Good luck!

Erik

[DanielBecker]

Hmm okay did you actually get it to work on a really huge raster? Mine has a cell size of 10m and is about.. well.. more than 20x30km.

[ErikMarsh]

No, I actually haven't run into that issue.... so I've never carefully checked the size of the DEM. But maybe they were never too big to cause problems.

[DanielBecker]

Hmm, well i was a little imprecise there, by extent i of course "meant" the cell-count, i think something around 200.000 pixels may be the threshold. You even see it if the tool starts, when it works, you can see the numbers "counting" in the processing-window. I assume you used elevation-models bigger than that?!

i also tested with a esri-grid 8bit signed int file, no differences and i resampled the raster to a lower resolution with the solution that the working extent grows.. 🙂

the only idea i have right now is to test this on another windows, maybe a non-x64 one.. but i think i should continue this in the other thread, i hope this will get a bit more attention. 🙂

[DanielBecker]

I've been trying to run the Path Allocation tool with a custom VRMA-VF table to model energy expenditure for travel on foot.  The tool will run if I use the built-in vertical factor functions, but hangs indefinitely (most common in ArcGIS 10) or fails with a meaningless error message (most common in ArcGIS 9.3.1) when I use a custom table.  I assume I have made an error in formatting the table.  The online help (http://help.arcgis.com/en/arcgisdesktop/10.0/help/index.html#//009z000000z9000000.htm) gives an example for building the table, but it is just a copy of the sample table for the horizontal factor.  Does vertical factor also need to run from 0 to 180 degrees?  If so, to what slope do 0, 90, and 180 correspond?

Concerning this i just experimented with the table until ArcGis stopped crashing...

I think it doesn't matter if there are Tabs or Spaces between the VRMA and the VF.

What worked was, leaving the smaller intervals below/above -70 and 70 away.. so

Doesn't work:
...
-73 13.11118151
-72 6.668357861
-71 3.633583399
-70 2.099401323
-69 1.275442246
...

Works:

-90 -1
-80 -1
-70 2.099401323
-69 1.275442246

I also tried to change the VFs higher than 3 to -1 without leaving the other VRMAs away.. didn't help either. 😉