Hi Hadi,

if I understand correctly, you want to reconstruct the ancient road-network of this region. I know better the Soth-West of Iran, since I was working there some years ago, but I guess for Baluchistan should be the same (for what I know of Baluchistan). In past researches about this topic I worked (within GRASS GIS, in ArcheOS) in Italy me and my collegues were starting from a DTM (Digital Terrain Model) of the landscape on which we were calculating the "friction surface" (automatically removing all the areas with an excessive slope). On this new model we were considering additional costs to cross the landscape, considering that humans vectors are normally anisotropic (so for humans it makes a difference to walk uphill or downhill). These additional costs were of two kinds. The first one is related with "natural resistances" and it is simpler, since it is more static across time (but you have to be careful, considering the 4D transformation of the landscape, such artificial lakes and so on). An example is crossing a river (in this case you can draw a vector buffer on the river and assign a cost to cross it). The second kind is more difficult and has to be calibrated on the historical period. This second type is related with "anthropic resistances" (any kind of cultural barrier, such as linguistic boundaries, presence of bandits or armed gangs and so on...). Anyway you have to do your research, consider any possible "crossing cots" for the landscape (natural or human) and update your friction surface. When I was working on this topic, as I told you, I used GRASS GIS (but now you cand do many things also simply with QGIS). I just tell you that, in that times (more or less 10 years ago) the problem with GRASS were two. 1) it was not yet implement an algorithm to consider also boustrophedonic paths (e.g. to go uphill), so you always have to keep an eye (human controll) over the computer result. 2) for the final least cost path in GRASS we had to use the algorithm used normally for water (if I remember well the command was r.path), so if the track was going inside a river basin, it was no more exiting. I hope I understood well your question and that my answer can be somehow useful. In case I think that me and my collegues wrote someting abot this topic (if I remember well, it was on the annual Italian conference about archaeology and free and open source software, the ArcheoFOSS). In case I can try to find this article, if it could be useful. Have a nice day!

Hi!

As Luca rightly said, there are several possibilities to enhance your friction surface! I would include floodplains/wetlands, flooding zones (high flooding vulnerability), waterlogged soils, erosion-prone areas. You can process data from multispectral satellite imagery (Sentinel 2), SAR (Radar), Corine Riparian Delineations and many other..

I personally do not very much favour LCP but rather calculate movement corridors from accumulated cost surfaces...

Best

Michael

see also:

Irmela Herzog

Martijn van Leusen

and

van Lanen et al

(find more refs of true experts in my paper in Geoarchaeology... )

Luca Bezzi thank you Bezzi. your recommends are very useful for me. actually I'm writing my MA thesis right now and learning more about least-cost path in archaeology especially the slope factor and friction surface. but as I found out this process takes long and needs more attention.

thank you very much

Michael Kempf thank you very much for your guidance Kempf. when I studied the Baluchistan region, found out that today environment and the Bronze age environment is very similar and also the archaeological sites are the same place where today villages are. so what do you think about Ndvi map for evaluating and calculating LCP? How could i use Ndvi for lcp?

as you know lcp evaluates and calculates the least numbers in maps like slope but in Ndvi, I want use the most high number for least-cost path which means the most density vegetation regions are the least cost path and most probable routes.

Hi! If there is land-use continuity (what you need to be sure about.... ususally archaeology is recorded right in the modern villages/settlement spots.. rather due to built-up change and infrastructure then due to the fact that the patterns show 'realistic' archaeological land-use opportunities) you can integrate the NDVI in your friction surface by recalculating the value of the vegetation density:

produce your NDVI map with a remote sensing software (eg multispec, find the guidance for the channel recalculation attached. Download the multicpec software here: https://engineering.purdue.edu/~biehl/MultiSpec/

in your GIS, you can recalculate the values of your raster image to any range you like and finally create no-movement corridors (high values) and potential movement corridors (low values). I am not sure that high vegetation areas would enable movement though...

adding all your raster values to one final raster would result in a friction surface.

in general, you might want to have a look at these books for technical guidance (C/L) and a methodical overview (P):

Conolly/Lake:

Geographical Information Systems in Archaeology (Cambridge Manuals in Archaeology)

Parcak:

Satellite Remote Sensing for Archaeology

Good luck

Michael