Reconstructing historical journeys with least-cost analysis: Colonel William Leake in the Mani Peninsula, Greece

Due to the proliferation of Geographic Information Systems (GIS) software and the improvement of graphic user interfaces, least-cost analysis has become one of the most common tools in archaeological research for modelling movement across past landscapes. While more complex techniques and algorithms have been developed to address the theoretical limitations of least-cost analysis, many of these are not readily accessible to typical GIS users, due to either hardware limitations or a lack of in-depth training in the theory and methods of spatial analysis. Seeking a middle ground, we test a number of simple modifications to basic least-cost analysis using the Mani peninsula in southern Greece as a case study. We assess eight procedures for calculating least-cost paths (LCPs), defining cost based on a combination of slope and the location of pre-modern pathways. These procedures are used to generate LCPs for a well-documented journey made by Colonel William Martin Leake in the year 1805. Comparing the LCPs against Leake's detailed notes and "low-cost corridors," we visually assess the reliability of each model and identify likely alternative routes. We conclude that a simple modification to basic least-cost analysis - incorporating a time-based algorithm along with traces of pre-modern footpaths and roads can produce robust, reliable least-cost models in other archaeological contexts where traces of pre-modern pathways can still be detected.