I have previously calculated the catchment areas for a series of archaeological sites. After making some minor adjustments to site locations, I now have to repeat the same calculations. The results should be very similar, but they are quite different. I'm using the same table summarizing the Tobler function in the way it is explained elsewhere (e.g. http://mapaspects.org/node/3744/) I'm pretty sure I've been doing the same thing I did last time, only then I used ArcMap 10.2.2 and now am using 10.5.1.
Iv'e tryed different rasters, different versions of Tobler's table, different coordinate systems, adding a slope raster as an additional variable, you name it! If it could be tryed, it was. nothing helped, I never get the same result as before.
To be sure, I ran the Path Distance tool on the old location as well, and again the result seems to be wrong.
In the Image you can see the the problem. the blue dot and polygons are the new location, the red are the old location and the raster is the old location result when I ran it now. The raster and the new (blue) location polygon are very similar, except for the slight shift in position, but I think the red polygon is more accurate.
Could this be the reason for the different results?
How can I say which is true if they are similar?