Abstract
Recent studying of the Antonine Wall via Viewshed Analysis has presented an interesting pattern. Locations chosen for Roman forts and fortlets offered good observation of the frontier zone, most sites were intervisible each with other and thus fundamental conditions for existence of a visual signal chain on the limes were met.
On the other hand, some problems occurred as well. In order to give potential Roman sentries a full observance of the coastal region, apparently at least one fortlet on the coastline is still missing.
Predictive modelling or Cost Path Analysis could help us to find the most favourable location for this installation. However, some forts were in terms of visibility analysis situated on unfavourable locations.
Despite a fact that visual signal chain could have been maintained even without them, nodal question why they were situated where they were still remains. This leads us back to another very old question: What kind of qualities should a spot chosen for a Roman fort have? While some forts on the Anotnine Wall were clearly situated where they were in order to fit the visibility and intervisibility pattern, some were not.
Could the Cost path Analysis be a key to better understanding the Antonine Wall? Presented paper is principally trying to examine correlation between locations of Iron Age settlements north of the Antonine Wall and positioning of Roman forts via spatial analyses in GIS software.